This document describes using an artificial neural network (ANN) model to optimize the cost of reinforced concrete beams designed according to ACI 318-08 code requirements. The ANN model considers costs of concrete, reinforcement steel, and formwork. A simply supported beam was designed with variable cross-sectional dimensions to demonstrate the model. Computer models were developed using NEURO SHELL-2 software and results were compared to a classical optimization model in Excel using generalized reduced gradient methods, showing good agreement between the two approaches. The document provides details on the ANN model formulation, including design variables, constraints, and objective function to minimize total cost. An example problem is presented to optimize the design of a simply supported beam.
Applications of Artificial Neural Networks in Civil EngineeringPramey Zode
An artificial brain-like network based on certain mathematical algorithms developed using a numerical computing environment is called as an ‘Artificial Neural Network (ANN)’. Many civil engineering problems which need understanding of physical processes are found to be time consuming and inaccurate to evaluate using conventional approaches. In this regard, many ANNs have been seen as a reliable and practical alternative to solve such problems. Literature review reveals that ANNs have already being used in solving numerous civil engineering problems. This study explains some cases where ANNs have been used and its future scope is also discussed.
Validation of hardness and tensile strength of al 7075 based hybrid composite...IAEME Publication
The document describes research validating the hardness and tensile strength of Al 7075 hybrid composites using artificial neural networks. Experimental results for Al 7075 reinforced with fly ash and E-glass fibers were used to train a neural network model. The neural network predictions matched very closely with experimental values, with a correlation coefficient of 0.99918. This shows artificial neural networks can accurately model and predict the material properties of hybrid composites.
International Journal of Engineering Research and Applications (IJERA) is an open access online peer reviewed international journal that publishes research and review articles in the fields of Computer Science, Neural Networks, Electrical Engineering, Software Engineering, Information Technology, Mechanical Engineering, Chemical Engineering, Plastic Engineering, Food Technology, Textile Engineering, Nano Technology & science, Power Electronics, Electronics & Communication Engineering, Computational mathematics, Image processing, Civil Engineering, Structural Engineering, Environmental Engineering, VLSI Testing & Low Power VLSI Design etc.
The document describes IntelliSuite's integrated MEMS design flow, which allows users to model devices from the schematic level through physical layout and verification, as well as simulate multi-physics processes. IntelliSuite combines top-down and bottom-up design approaches for efficient and accurate modeling. Key capabilities include schematic capture, physical layout, process simulation, and system-level modeling extraction for fast behavioral simulation.
Artificial Neural Network (ANN) is a fast-growing method which has been used in different
industries during recent years. The main idea for creating ANN which is a subset of artificial
intelligence is to provide a simple model of human brain in order to solve complex scientific and
industrial problems. ANNs are high-value and low-cost tools in modelling, simulation, control,
condition monitoring, sensor validation and fault diagnosis of different systems. It have high
flexibility and robustness in modeling, simulating and diagnosing the behavior of rotating machines
even in the presence of inaccurate input data. They can provide high computational speed for
complicated tasks that require rapid response such as real-time processing of several simultaneous
signals. ANNs can also be used to improve efficiency and productivity of energy in rotating
equipment
Implementing Neural Networks Using VLSI for Image Processing (compression)IJERA Editor
Biological systems process the analog signals such as image and sound efficiently. To process the information the way biological systems do we make use of ANN. (Artificial Neural Networks) The focus of this paper is to review the implementation of the neural network architecture using analog components like Gilbert cell multiplier, differential amplifier for neuron activation function and tan sigmoid function circuit using MOS transistor. The neural architecture is trained using Back propagation algorithm for compressing the image. This paper surveys the methods of implementing the neural network using VLSI .Different CMOS technologies are used for implementing the circuits for arithmetic operations (i.e. 180nm, 45nm, 32nm).And the MOS transistors are working in sub threshold region. In this paper a review is made on how the VLSI architecture is used to implement neural networks and trained for compressing the image.
1) The document discusses VLSI architecture and implementation for 3D neural network based image compression. It proposes developing new hardware architectures optimized for area, power, and speed for implementing 3D neural networks for image compression.
2) A block diagram is presented showing the overall process of image acquisition, preprocessing, compression using a 3D neural network, and encoding for transmission.
3) The proposed 3D neural network architecture uses multiple hidden layers with lower dimensions than the input and output layers to perform compression and decompression. The network is trained using backpropagation.
Applications of Artificial Neural Networks in Civil EngineeringPramey Zode
An artificial brain-like network based on certain mathematical algorithms developed using a numerical computing environment is called as an ‘Artificial Neural Network (ANN)’. Many civil engineering problems which need understanding of physical processes are found to be time consuming and inaccurate to evaluate using conventional approaches. In this regard, many ANNs have been seen as a reliable and practical alternative to solve such problems. Literature review reveals that ANNs have already being used in solving numerous civil engineering problems. This study explains some cases where ANNs have been used and its future scope is also discussed.
Validation of hardness and tensile strength of al 7075 based hybrid composite...IAEME Publication
The document describes research validating the hardness and tensile strength of Al 7075 hybrid composites using artificial neural networks. Experimental results for Al 7075 reinforced with fly ash and E-glass fibers were used to train a neural network model. The neural network predictions matched very closely with experimental values, with a correlation coefficient of 0.99918. This shows artificial neural networks can accurately model and predict the material properties of hybrid composites.
International Journal of Engineering Research and Applications (IJERA) is an open access online peer reviewed international journal that publishes research and review articles in the fields of Computer Science, Neural Networks, Electrical Engineering, Software Engineering, Information Technology, Mechanical Engineering, Chemical Engineering, Plastic Engineering, Food Technology, Textile Engineering, Nano Technology & science, Power Electronics, Electronics & Communication Engineering, Computational mathematics, Image processing, Civil Engineering, Structural Engineering, Environmental Engineering, VLSI Testing & Low Power VLSI Design etc.
The document describes IntelliSuite's integrated MEMS design flow, which allows users to model devices from the schematic level through physical layout and verification, as well as simulate multi-physics processes. IntelliSuite combines top-down and bottom-up design approaches for efficient and accurate modeling. Key capabilities include schematic capture, physical layout, process simulation, and system-level modeling extraction for fast behavioral simulation.
Artificial Neural Network (ANN) is a fast-growing method which has been used in different
industries during recent years. The main idea for creating ANN which is a subset of artificial
intelligence is to provide a simple model of human brain in order to solve complex scientific and
industrial problems. ANNs are high-value and low-cost tools in modelling, simulation, control,
condition monitoring, sensor validation and fault diagnosis of different systems. It have high
flexibility and robustness in modeling, simulating and diagnosing the behavior of rotating machines
even in the presence of inaccurate input data. They can provide high computational speed for
complicated tasks that require rapid response such as real-time processing of several simultaneous
signals. ANNs can also be used to improve efficiency and productivity of energy in rotating
equipment
Implementing Neural Networks Using VLSI for Image Processing (compression)IJERA Editor
Biological systems process the analog signals such as image and sound efficiently. To process the information the way biological systems do we make use of ANN. (Artificial Neural Networks) The focus of this paper is to review the implementation of the neural network architecture using analog components like Gilbert cell multiplier, differential amplifier for neuron activation function and tan sigmoid function circuit using MOS transistor. The neural architecture is trained using Back propagation algorithm for compressing the image. This paper surveys the methods of implementing the neural network using VLSI .Different CMOS technologies are used for implementing the circuits for arithmetic operations (i.e. 180nm, 45nm, 32nm).And the MOS transistors are working in sub threshold region. In this paper a review is made on how the VLSI architecture is used to implement neural networks and trained for compressing the image.
1) The document discusses VLSI architecture and implementation for 3D neural network based image compression. It proposes developing new hardware architectures optimized for area, power, and speed for implementing 3D neural networks for image compression.
2) A block diagram is presented showing the overall process of image acquisition, preprocessing, compression using a 3D neural network, and encoding for transmission.
3) The proposed 3D neural network architecture uses multiple hidden layers with lower dimensions than the input and output layers to perform compression and decompression. The network is trained using backpropagation.
From Bio-Intelligence BI to Artificial-Intelligence AI in Engineering and STE...Dr. Fayçal Saffih
This is summary of my research and teaching statement and mission over the course of my career to include all my seminars presented at various universities across the world (see http://bit.ly/Seminars_Map_FaycalS)
It included my journey in embedding intelligence in imaging from the device level, to circuit level and up to the system level.
It also includes my teaching philosophy and proposition by embedding intelligence in modern teaching/education and Blended learning. This method has been published at the 124th conference of American Society for Engineering Education, Columbus, OH, USA on July 2017. You may download the paper here: http://bit.ly/AIM4STEM2017_FaycalS and watch the talk here: http://bit.ly/AIM4STEM_ASEE2017_Talk_FaycalS
You can also watch these slides presented here: http://bit.ly/From-BI-for-AI-Research-Teaching-Vision_FaycalS
Enjoy!
Coronary heart disease is a disease with the highest mortality rates in the world. This makes the development of the diagnostic system as a very interesting topic in the field of biomedical informatics, aiming to detect whether a heart is normal or not. In the literature there are diagnostic system models by combining dimension reduction and data mining techniques. Unfortunately, there are no review papers that discuss and analyze the themes to date. This study reviews articles within the period 2009-2016, with a focus on dimension reduction methods and data mining techniques, validated using a dataset of UCI repository. Methods of dimension reduction use feature selection and feature extraction techniques, while data mining techniques include classification, prediction, clustering, and association rules.
An Investigation towards Effectiveness in Image Enhancement Process in MPSoC IJECEIAES
The document discusses image enhancement techniques for use in multiprocessor systems-on-chip (MPSoC). It reviews existing image enhancement methods implemented on FPGA and identifies limitations like low accuracy. The paper proposes using advanced bus architectures like AMBA/AXI in MPSoC to improve communication between memory and input medical images for enhancement, reducing heterogeneity effects. It summarizes various image enhancement algorithms that could be implemented on reconfigurable FPGA hardware for use in MPSoC, including brightness control, contrast stretching, histogram equalization, and edge detection techniques.
MULTI-LEVEL FEATURE FUSION BASED TRANSFER LEARNING FOR PERSON RE-IDENTIFICATIONgerogepatton
Most of the currently known methods treat person re-identification task as classification problem and used commonly neural networks. However, these methods used only high-level convolutional feature or to express the feature representation of pedestrians. Moreover, the current data sets for person reidentification is relatively small. Under the limitation of the number of training set, deep convolutional networks are difficult to train adequately. Therefore, it is very worthwhile to introduce auxiliary data sets help training. In order to solve this problem, this paper propose a novel method of deep transfer learning, and combines the comparison model with the classification model and multi-level fusion of the
convolution features on the basis of transfer learning. In a multi-layers convolutional network, the characteristics of each layer of network are the dimensionality reduction of the previous layer of results, but the information of multi-level features is not only inclusive, but also has certain complementarity. We can using the information gap of different layers of convolutional neural networks to extract a better feature expression. Finally, the algorithm proposed in this paper is fully tested on four data sets (VIPeR, CUHK01, GRID and PRID450S). The obtained re-identification results prove the effectiveness of the algorithm.
Soft computing is likely to play an important role in science and engineering in the future. The successful applications of soft computing and the rapid growth suggest that the impact of soft computing will be felt increasingly in coming years. Soft Computing encourages the integration of soft computing techniques and tools into both everyday and advanced applications. This Open access peer-reviewed journal serves as a platform that fosters new applications for all scientists and engineers engaged in research and development in this fast growing field.
This document discusses using wireless sensor networks for habitat monitoring. It presents requirements for a system to monitor seabird nesting environments and behaviors. The currently deployed network consists of 32 sensor nodes on an island off the coast of Maine that stream live data online. The application-driven design helps identify important areas for further work like data sampling, communications, network tasks, and health monitoring.
TRANSFER LEARNING WITH CONVOLUTIONAL NEURAL NETWORKS FOR IRIS RECOGNITIONijaia
Iris is one of the common biometrics used for identity authentication. It has the potential to recognize persons with a high degree of assurance. Extracting effective features is the most important stage in the iris recognition system. Different features have been used to perform iris recognition system. A lot of them are based on hand-crafted features designed by biometrics experts. According to the achievement of deep learning in object recognition problems, the features learned by the Convolutional Neural Network (CNN) have gained great attention to be used in the iris recognition system. In this paper, we proposed an effective iris recognition system by using transfer learning with Convolutional Neural Networks. The proposed system is implemented by fine-tuning a pre-trained convolutional neural network (VGG-16) for features extracting and classification. The performance of the iris recognition system is tested on four public databases IITD, iris databases CASIA-Iris-V1, CASIA-Iris-thousand and, CASIA-Iris-Interval. The results show that the proposed system is achieved a very high accuracy rate.
DYNAMIC NETWORK ANOMALY INTRUSION DETECTION USING MODIFIED SOMcscpconf
This document presents a modified Self-Organizing Map (SOM) algorithm for network anomaly intrusion detection. The proposed algorithm allows the neural network to grow dynamically based on a distance threshold, rather than having a fixed architecture. It also uses connection strength to identify neighborhood nodes for weight vector updating. The algorithm was tested on standard intrusion detection datasets and achieved a detection rate of 98% and a false alarm rate of 2%, outperforming a basic SOM approach. The modified SOM addresses limitations of fixed network architecture and random weight initialization in the standard SOM method.
This document presents Jeevn-Net, a new neural network architecture for brain tumor segmentation and overall survival prediction. Jeevn-Net uses a cascaded U-Net structure with two U-Nets and applies auto-encoder regularization. It takes in MRI scans and outputs a segmented tumor image with extracted features. Random forest regression is then used to predict survival based on these features. The network achieves state-of-the-art performance for brain tumor segmentation and survival prediction.
IntelliSuite is a complete design environment for MEMS that provides tools across the entire product development cycle. It includes schematic capture tool Synple, physical design tool Blueprint, process design tools CleanRoom, and multiphysics solvers FastField. IntelliSuite aims to link the entire MEMS organization through a unified platform. It offers a seamless flow from concept to tapeout through integrated tools for simulation, layout, verification, and more. IntelliSuite has established itself as the standard industry tool used by MEMS professionals worldwide.
International Journal of Computational Engineering Research(IJCER)ijceronline
The document discusses image compression using artificial neural networks. It begins with an introduction to image compression and the need for it. Then it reviews various existing neural network approaches for image compression, including backpropagation networks, hierarchical networks, multilayer feedforward networks, and radial basis function networks. It proposes a new approach using a multilayer perceptron with a modified Levenberg-Marquardt training algorithm to improve compression performance. Authentication and protection would be incorporated by exploiting the one-to-one mapping and one-way properties of neural networks. The proposed system is described as compressing images using neural networks trained with a modified LM algorithm to achieve high compression ratios while maintaining image quality.
IRJET - An Robust and Dynamic Fire Detection Method using Convolutional N...IRJET Journal
This document proposes a new fire detection method using convolutional neural networks (CNNs). Specifically, it uses the YOLOv3 object detection algorithm, which can detect objects like fire in images or videos quickly and accurately. The proposed method aims to reduce computational time and costs compared to other CNN-based approaches, while also improving detection accuracy and reducing false alarms. It discusses implementing the method using four main modules: data exploration, pre-processing, feature engineering, and model selection. The workflow involves exploring data, pre-processing images, extracting features, and selecting the YOLOv3 CNN model for fire detection. The goal is to develop a robust and dynamic fire detection system using computer vision techniques to help prevent accidents.
Trends in VLSI circuit in 2020 - International Journal of VLSI design & Commu...VLSICS Design
International Journal of VLSI design & Communication Systems (VLSICS) is a bi monthly open access peer-reviewed journal that publishes articles which contribute new results in all areas of VLSI Design & Communications. The goal of this journal is to bring together researchers and practitioners from academia and industry to focus on advanced VLSI Design & communication concepts and establishing new collaborations in these areas.
Recent articles published in VLSI design & Communication SystemsVLSICS Design
International Journal of VLSI design & Communication Systems (VLSICS) is a bi monthly open access peer-reviewed journal that publishes articles which contribute new results in all areas of VLSI Design & Communications. The goal of this journal is to bring together researchers and practitioners from academia and industry to focus on advanced VLSI Design & communication concepts and establishing new collaborations in these areas.
Authors are solicited to contribute to this journal by submitting articles that illustrate research results, projects, surveying works and industrial experiences that describe significant advances in the VLSI design & Communications.
This paper demonstrates a framework that entails a bottom-up approach to
accelerate research, development, and verification of neuro-inspired sensing
devices for real-life applications. Previous work in neuromorphic
engineering mostly considered application-specific designs which is a strong
limitation for researchers to develop novel applications and emulate the true
behaviour of neuro-inspired systems. Hence to enable the fully parallel
brain-like computations, this paper proposes a methodology where a spiking
neuron model was emulated in software and electronic circuits were then
implemented and characterized. The proposed approach offers a unique
perspective whereby experimental measurements taken from a fabricated
device allowing empirical models to be developed. This technique acts as a
bridge between the theoretical and practical aspects of neuro-inspired
devices. It is shown through software simulations and empirical modelling
that the proposed technique is capable of replicating neural dynamics and
post-synaptic potentials. Retrospectively, the proposed framework offers a
first step towards open-source neuro-inspired hardware for a range of
applications such as healthcare, applied machine learning and the internet of
things (IoT).
A Review of Neural Networks Architectures, Designs, and ApplicationsIRJET Journal
This document provides an overview of various neural network architectures and their applications. It discusses deep neural networks, graph neural networks, fully convolutional neural networks, and recurrent neural networks. Applications mentioned include signal processing, pattern recognition, medical diagnosis, speech recognition, fault detection, and more. The document reviews several recent research papers on neural networks and their use in domains like gravitational wave detection, rolling bearing fault detection, neutrino detection, control systems, and speech segmentation.
VALIDATION OF HARDNESS AND TENSILE STRENGTH OF AL 7075 BASED HYBRID COMPOSITE...IAEME Publication
than those of usual composites. Based on the extensive literature review, it is concluded that majority of investigations were carried out on Aluminium alloy based composite materials involving Silicon Carbide and Alumina as
reinforcements. The investigations using fly ash and E-glass have been carried out with matrix aluminium alloys other
than Al 7075 alloy. In the present investigation a new class of hybrid composite, Al 7075 alloy reinforced with fly ash
particulates, E-glass short fibers has been formed and experimental results have been validated by using Artificial Neural Networks (ANNs). The ANN predictions were in very good agreement with experimental results, with correlation
coefficient 0.99918.
This document reviews object detection techniques using convolutional neural networks (CNNs). It begins with introducing object detection and CNNs. It then discusses the problem of object detection in computer vision and the need for more precise and accurate detection systems. The majority of the document reviews eight previous works that developed algorithms to improve object detection systems, including R-CNN and approaches using K-SVD, deep equilibrium models, non-local networks, transformers, and selective kernel networks. It evaluates these approaches and their abilities to achieve high detection rates while requiring fewer computations or model parameters. The document provides an overview of recent research aiming to advance CNN-based object detection.
An overview on application of machine learning techniques in optical networksKhaleda Ali
This document provides an overview of machine learning techniques applied to optical networks. It discusses how optical networks have become more complex with the introduction of technologies like coherent transmission and elastic optical networks. This increased complexity motivates the use of machine learning to analyze network data and make decisions. The document surveys existing work on machine learning applications in optical communications and networking. It aims to introduce researchers to this field and propose new research directions to further the application of machine learning to optical networks.
Many intellectual property (IP) modules are present in contemporary system on chips (SoCs). This could provide an issue with interconnection among different IP modules, which would limit the system's ability to scale. Traditional bus-based SoC architectures have a connectivity bottleneck, and network on chip (NoC) has evolved as an embedded switching network to address this issue. The interconnections between various cores or IP modules on a chip have a significant impact on communication and chip performance in terms of power, area latency and throughput. Also, designing a reliable fault tolerant NoC became a significant concern. In fault tolerant NoC it becomes critical to identify faulty node and dynamically reroute the packets keeping minimum latency. This study provides an insight into a domain of NoC, with intention of understanding fault tolerant approach based on the XY routing algorithm for 4×4 mesh architecture. The fault tolerant NoC design is synthesized on field programmable gate array (FPGA).
A SURVEY OF NEURAL NETWORK HARDWARE ACCELERATORS IN MACHINE LEARNING mlaij
The use of Machine Learning in Artificial Intelligence is the inspiration that shaped technology as it is today. Machine Learning has the power to greatly simplify our lives. Improvement in speech recognition and language understanding help the community interact more naturally with technology. The popularity of machine learning opens up the opportunities for optimizing the design of computing platforms using welldefined hardware accelerators. In the upcoming few years, cameras will be utilised as sensors for several applications. For ease of use and privacy restrictions, the requested image processing should be limited to a local embedded computer platform and with a high accuracy. Furthermore, less energy should be consumed. Dedicated acceleration of Convolutional Neural Networks can achieve these targets with high flexibility to perform multiple vision tasks. However, due to the exponential growth in technology constraints (especially in terms of energy) which could lead to heterogeneous multicores, and increasing number of defects, the strategy of defect-tolerant accelerators for heterogeneous multi-cores may become a main micro-architecture research issue. The up to date accelerators used still face some performance issues such as memory limitations, bandwidth, speed etc. This literature summarizes (in terms of a survey) recent work of accelerators including their advantages and disadvantages to make it easier for developers with neural network interests to further improve what has already been established.
From Bio-Intelligence BI to Artificial-Intelligence AI in Engineering and STE...Dr. Fayçal Saffih
This is summary of my research and teaching statement and mission over the course of my career to include all my seminars presented at various universities across the world (see http://bit.ly/Seminars_Map_FaycalS)
It included my journey in embedding intelligence in imaging from the device level, to circuit level and up to the system level.
It also includes my teaching philosophy and proposition by embedding intelligence in modern teaching/education and Blended learning. This method has been published at the 124th conference of American Society for Engineering Education, Columbus, OH, USA on July 2017. You may download the paper here: http://bit.ly/AIM4STEM2017_FaycalS and watch the talk here: http://bit.ly/AIM4STEM_ASEE2017_Talk_FaycalS
You can also watch these slides presented here: http://bit.ly/From-BI-for-AI-Research-Teaching-Vision_FaycalS
Enjoy!
Coronary heart disease is a disease with the highest mortality rates in the world. This makes the development of the diagnostic system as a very interesting topic in the field of biomedical informatics, aiming to detect whether a heart is normal or not. In the literature there are diagnostic system models by combining dimension reduction and data mining techniques. Unfortunately, there are no review papers that discuss and analyze the themes to date. This study reviews articles within the period 2009-2016, with a focus on dimension reduction methods and data mining techniques, validated using a dataset of UCI repository. Methods of dimension reduction use feature selection and feature extraction techniques, while data mining techniques include classification, prediction, clustering, and association rules.
An Investigation towards Effectiveness in Image Enhancement Process in MPSoC IJECEIAES
The document discusses image enhancement techniques for use in multiprocessor systems-on-chip (MPSoC). It reviews existing image enhancement methods implemented on FPGA and identifies limitations like low accuracy. The paper proposes using advanced bus architectures like AMBA/AXI in MPSoC to improve communication between memory and input medical images for enhancement, reducing heterogeneity effects. It summarizes various image enhancement algorithms that could be implemented on reconfigurable FPGA hardware for use in MPSoC, including brightness control, contrast stretching, histogram equalization, and edge detection techniques.
MULTI-LEVEL FEATURE FUSION BASED TRANSFER LEARNING FOR PERSON RE-IDENTIFICATIONgerogepatton
Most of the currently known methods treat person re-identification task as classification problem and used commonly neural networks. However, these methods used only high-level convolutional feature or to express the feature representation of pedestrians. Moreover, the current data sets for person reidentification is relatively small. Under the limitation of the number of training set, deep convolutional networks are difficult to train adequately. Therefore, it is very worthwhile to introduce auxiliary data sets help training. In order to solve this problem, this paper propose a novel method of deep transfer learning, and combines the comparison model with the classification model and multi-level fusion of the
convolution features on the basis of transfer learning. In a multi-layers convolutional network, the characteristics of each layer of network are the dimensionality reduction of the previous layer of results, but the information of multi-level features is not only inclusive, but also has certain complementarity. We can using the information gap of different layers of convolutional neural networks to extract a better feature expression. Finally, the algorithm proposed in this paper is fully tested on four data sets (VIPeR, CUHK01, GRID and PRID450S). The obtained re-identification results prove the effectiveness of the algorithm.
Soft computing is likely to play an important role in science and engineering in the future. The successful applications of soft computing and the rapid growth suggest that the impact of soft computing will be felt increasingly in coming years. Soft Computing encourages the integration of soft computing techniques and tools into both everyday and advanced applications. This Open access peer-reviewed journal serves as a platform that fosters new applications for all scientists and engineers engaged in research and development in this fast growing field.
This document discusses using wireless sensor networks for habitat monitoring. It presents requirements for a system to monitor seabird nesting environments and behaviors. The currently deployed network consists of 32 sensor nodes on an island off the coast of Maine that stream live data online. The application-driven design helps identify important areas for further work like data sampling, communications, network tasks, and health monitoring.
TRANSFER LEARNING WITH CONVOLUTIONAL NEURAL NETWORKS FOR IRIS RECOGNITIONijaia
Iris is one of the common biometrics used for identity authentication. It has the potential to recognize persons with a high degree of assurance. Extracting effective features is the most important stage in the iris recognition system. Different features have been used to perform iris recognition system. A lot of them are based on hand-crafted features designed by biometrics experts. According to the achievement of deep learning in object recognition problems, the features learned by the Convolutional Neural Network (CNN) have gained great attention to be used in the iris recognition system. In this paper, we proposed an effective iris recognition system by using transfer learning with Convolutional Neural Networks. The proposed system is implemented by fine-tuning a pre-trained convolutional neural network (VGG-16) for features extracting and classification. The performance of the iris recognition system is tested on four public databases IITD, iris databases CASIA-Iris-V1, CASIA-Iris-thousand and, CASIA-Iris-Interval. The results show that the proposed system is achieved a very high accuracy rate.
DYNAMIC NETWORK ANOMALY INTRUSION DETECTION USING MODIFIED SOMcscpconf
This document presents a modified Self-Organizing Map (SOM) algorithm for network anomaly intrusion detection. The proposed algorithm allows the neural network to grow dynamically based on a distance threshold, rather than having a fixed architecture. It also uses connection strength to identify neighborhood nodes for weight vector updating. The algorithm was tested on standard intrusion detection datasets and achieved a detection rate of 98% and a false alarm rate of 2%, outperforming a basic SOM approach. The modified SOM addresses limitations of fixed network architecture and random weight initialization in the standard SOM method.
This document presents Jeevn-Net, a new neural network architecture for brain tumor segmentation and overall survival prediction. Jeevn-Net uses a cascaded U-Net structure with two U-Nets and applies auto-encoder regularization. It takes in MRI scans and outputs a segmented tumor image with extracted features. Random forest regression is then used to predict survival based on these features. The network achieves state-of-the-art performance for brain tumor segmentation and survival prediction.
IntelliSuite is a complete design environment for MEMS that provides tools across the entire product development cycle. It includes schematic capture tool Synple, physical design tool Blueprint, process design tools CleanRoom, and multiphysics solvers FastField. IntelliSuite aims to link the entire MEMS organization through a unified platform. It offers a seamless flow from concept to tapeout through integrated tools for simulation, layout, verification, and more. IntelliSuite has established itself as the standard industry tool used by MEMS professionals worldwide.
International Journal of Computational Engineering Research(IJCER)ijceronline
The document discusses image compression using artificial neural networks. It begins with an introduction to image compression and the need for it. Then it reviews various existing neural network approaches for image compression, including backpropagation networks, hierarchical networks, multilayer feedforward networks, and radial basis function networks. It proposes a new approach using a multilayer perceptron with a modified Levenberg-Marquardt training algorithm to improve compression performance. Authentication and protection would be incorporated by exploiting the one-to-one mapping and one-way properties of neural networks. The proposed system is described as compressing images using neural networks trained with a modified LM algorithm to achieve high compression ratios while maintaining image quality.
IRJET - An Robust and Dynamic Fire Detection Method using Convolutional N...IRJET Journal
This document proposes a new fire detection method using convolutional neural networks (CNNs). Specifically, it uses the YOLOv3 object detection algorithm, which can detect objects like fire in images or videos quickly and accurately. The proposed method aims to reduce computational time and costs compared to other CNN-based approaches, while also improving detection accuracy and reducing false alarms. It discusses implementing the method using four main modules: data exploration, pre-processing, feature engineering, and model selection. The workflow involves exploring data, pre-processing images, extracting features, and selecting the YOLOv3 CNN model for fire detection. The goal is to develop a robust and dynamic fire detection system using computer vision techniques to help prevent accidents.
Trends in VLSI circuit in 2020 - International Journal of VLSI design & Commu...VLSICS Design
International Journal of VLSI design & Communication Systems (VLSICS) is a bi monthly open access peer-reviewed journal that publishes articles which contribute new results in all areas of VLSI Design & Communications. The goal of this journal is to bring together researchers and practitioners from academia and industry to focus on advanced VLSI Design & communication concepts and establishing new collaborations in these areas.
Recent articles published in VLSI design & Communication SystemsVLSICS Design
International Journal of VLSI design & Communication Systems (VLSICS) is a bi monthly open access peer-reviewed journal that publishes articles which contribute new results in all areas of VLSI Design & Communications. The goal of this journal is to bring together researchers and practitioners from academia and industry to focus on advanced VLSI Design & communication concepts and establishing new collaborations in these areas.
Authors are solicited to contribute to this journal by submitting articles that illustrate research results, projects, surveying works and industrial experiences that describe significant advances in the VLSI design & Communications.
This paper demonstrates a framework that entails a bottom-up approach to
accelerate research, development, and verification of neuro-inspired sensing
devices for real-life applications. Previous work in neuromorphic
engineering mostly considered application-specific designs which is a strong
limitation for researchers to develop novel applications and emulate the true
behaviour of neuro-inspired systems. Hence to enable the fully parallel
brain-like computations, this paper proposes a methodology where a spiking
neuron model was emulated in software and electronic circuits were then
implemented and characterized. The proposed approach offers a unique
perspective whereby experimental measurements taken from a fabricated
device allowing empirical models to be developed. This technique acts as a
bridge between the theoretical and practical aspects of neuro-inspired
devices. It is shown through software simulations and empirical modelling
that the proposed technique is capable of replicating neural dynamics and
post-synaptic potentials. Retrospectively, the proposed framework offers a
first step towards open-source neuro-inspired hardware for a range of
applications such as healthcare, applied machine learning and the internet of
things (IoT).
A Review of Neural Networks Architectures, Designs, and ApplicationsIRJET Journal
This document provides an overview of various neural network architectures and their applications. It discusses deep neural networks, graph neural networks, fully convolutional neural networks, and recurrent neural networks. Applications mentioned include signal processing, pattern recognition, medical diagnosis, speech recognition, fault detection, and more. The document reviews several recent research papers on neural networks and their use in domains like gravitational wave detection, rolling bearing fault detection, neutrino detection, control systems, and speech segmentation.
VALIDATION OF HARDNESS AND TENSILE STRENGTH OF AL 7075 BASED HYBRID COMPOSITE...IAEME Publication
than those of usual composites. Based on the extensive literature review, it is concluded that majority of investigations were carried out on Aluminium alloy based composite materials involving Silicon Carbide and Alumina as
reinforcements. The investigations using fly ash and E-glass have been carried out with matrix aluminium alloys other
than Al 7075 alloy. In the present investigation a new class of hybrid composite, Al 7075 alloy reinforced with fly ash
particulates, E-glass short fibers has been formed and experimental results have been validated by using Artificial Neural Networks (ANNs). The ANN predictions were in very good agreement with experimental results, with correlation
coefficient 0.99918.
This document reviews object detection techniques using convolutional neural networks (CNNs). It begins with introducing object detection and CNNs. It then discusses the problem of object detection in computer vision and the need for more precise and accurate detection systems. The majority of the document reviews eight previous works that developed algorithms to improve object detection systems, including R-CNN and approaches using K-SVD, deep equilibrium models, non-local networks, transformers, and selective kernel networks. It evaluates these approaches and their abilities to achieve high detection rates while requiring fewer computations or model parameters. The document provides an overview of recent research aiming to advance CNN-based object detection.
An overview on application of machine learning techniques in optical networksKhaleda Ali
This document provides an overview of machine learning techniques applied to optical networks. It discusses how optical networks have become more complex with the introduction of technologies like coherent transmission and elastic optical networks. This increased complexity motivates the use of machine learning to analyze network data and make decisions. The document surveys existing work on machine learning applications in optical communications and networking. It aims to introduce researchers to this field and propose new research directions to further the application of machine learning to optical networks.
Many intellectual property (IP) modules are present in contemporary system on chips (SoCs). This could provide an issue with interconnection among different IP modules, which would limit the system's ability to scale. Traditional bus-based SoC architectures have a connectivity bottleneck, and network on chip (NoC) has evolved as an embedded switching network to address this issue. The interconnections between various cores or IP modules on a chip have a significant impact on communication and chip performance in terms of power, area latency and throughput. Also, designing a reliable fault tolerant NoC became a significant concern. In fault tolerant NoC it becomes critical to identify faulty node and dynamically reroute the packets keeping minimum latency. This study provides an insight into a domain of NoC, with intention of understanding fault tolerant approach based on the XY routing algorithm for 4×4 mesh architecture. The fault tolerant NoC design is synthesized on field programmable gate array (FPGA).
A SURVEY OF NEURAL NETWORK HARDWARE ACCELERATORS IN MACHINE LEARNING mlaij
The use of Machine Learning in Artificial Intelligence is the inspiration that shaped technology as it is today. Machine Learning has the power to greatly simplify our lives. Improvement in speech recognition and language understanding help the community interact more naturally with technology. The popularity of machine learning opens up the opportunities for optimizing the design of computing platforms using welldefined hardware accelerators. In the upcoming few years, cameras will be utilised as sensors for several applications. For ease of use and privacy restrictions, the requested image processing should be limited to a local embedded computer platform and with a high accuracy. Furthermore, less energy should be consumed. Dedicated acceleration of Convolutional Neural Networks can achieve these targets with high flexibility to perform multiple vision tasks. However, due to the exponential growth in technology constraints (especially in terms of energy) which could lead to heterogeneous multicores, and increasing number of defects, the strategy of defect-tolerant accelerators for heterogeneous multi-cores may become a main micro-architecture research issue. The up to date accelerators used still face some performance issues such as memory limitations, bandwidth, speed etc. This literature summarizes (in terms of a survey) recent work of accelerators including their advantages and disadvantages to make it easier for developers with neural network interests to further improve what has already been established.
Design of Tele command SOC-IP by AES Cryptographic Method Using VHDLdbpublications
The goal of this project is to implement the (AES) encryption system using Verilog. To do this, several separate sections of the algorithm will be coded to work together towards the end goal of performing the correct encryption routines. A telecommand is a command sent to control a remote system or systems i.e not directly connected (e.g. via wires) to the place from which the telecommand is sent. The telecommand word is derived from tele = remote (Greek), and command = to entrust/order (Latin). Systems that need remote measurement and reporting of information of interest to the system designer or operator, require the counterpart of telecommand, telemetry. For a telecommand (TC) to be effective, it must be compiled into a pre-arranged format (which may follow a standard structure), modulated onto a carrier wave which is then transmitted with adequate power to the remote system. The remote system will then demodulates the digital signal from the carrier, decode the telecommand, and execute it.
This document provides an overview of artificial neural networks (ANN). It discusses that ANN is a tool in artificial intelligence that mimics the human brain and is useful for modeling nonlinear relationships in data. The document outlines the historical development of ANN and describes multilayer perceptron models, which contain input, output, and hidden layers of neurons connected by weighted links. ANN is well-suited for tasks like pattern recognition. It learns by presenting examples to adjust weights and reduce error through backward propagation.
The document discusses recent trends in using artificial neural networks (ANNs) for analog circuit design. It provides an introduction to analog design flows and challenges. ANNs can help automate analog circuit sizing and placement tasks. The document reviews literature on using ANNs for tasks like sizing circuits for different technologies and optimizing parameters. In conclusion, ANNs show promise for reducing analog design time and efforts by training on previous simulation data to optimize circuits.
This document discusses modeling DC servo motors using artificial neural networks (ANNs). It contains the following key points:
1. ANNs are a computational intelligence technique that can be used to model nonlinear systems like DC servo motors by learning from input-output data. ANNs have an interconnected structure that allows them to learn complex relationships.
2. A DC servo motor has electrical and mechanical components that can be modeled, including resistance, inductance, inertia, damping, input voltage, back EMF, and angular position/speed. Nonlinear effects like saturation and dead zones also need to be accounted for in the model.
3. The paper presents a motor model developed using ANN techniques to mimic the behavior of
A Review: Integrating SUGAR simulating tool and MEMS sensorIJERA Editor
Micro-Electro-Mechanical Systems (MEMS) are forming the basis for a rapidly growing industry and fields of
research, many MEMS designers still rely on calculations due to a lack of efficient computer-aided design
(CAD) tools that can assist with the initial stages of design exploration. This paper review about the techniques
to the design of MEMS and sugar simulating tool. SUGAR simulating tool utilizes past successful MEMS
designs and sub-assemblies as building blocks stored in an indexed library, allowing reuse and modification of
previous successful designs to help deal with the complexities of a new design tool. Reasoning tools find cases
in the library with solved problems similar to the current design problem in order to propose promising
conceptual designs. The paper recommends strategies for integrating the MEMS Design with evolutionary
computation of SST
Things like growing volumes and varieties of available data, cheaper and more powerful computational processing, data storage and large-value predictions that can guide better decisions and smart actions in real time without human intervention are playing critical role in this age. All of these require models that can automatically analyse large complex data and deliver quick accurate results – even on a very large scale. Machine learning plays a significant role in developing these models. The applications of machine learning range from speech and object recognition to analysis and prediction of finance markets. Artificial Neural Network is one of the important algorithms of machine learning that is inspired by the structure and functional aspects of the biological neural networks. In this paper, we discuss the purpose, representation and classification methods for developing hardware for machine learning with the main focus on neural networks. This paper also presents the requirements, design issues and optimization techniques for building hardware architecture of neural networks.
Things like growing volumes and varieties of available data, cheaper and more powerful computational processing, data storage and large-value predictions that can guide better decisions and smart actions inreal time without human intervention are playing critical role in this age. All of these require models thatcan automatically analyse large complex data and deliver quick accurate results – even on a very largescale. Machine learning plays a significant role in developing these models. The applications of machinelearning range from speech and object recognition to analysis and prediction of finance markets. Artificial Neural Network is one of the important algorithms of machine learning that is inspired by the structure and functional aspects of the biological neural networks. In this paper, we discuss the purpose, representationand classification methods for developing hardware for machine learning with the main focus on neuralnetworks. This paper also presents the requirements, design issues and optimization techniques for buildinghardware architecture of neural networks.
APPLYING GENETIC ALGORITHM TO SOLVE PARTITIONING AND MAPPING PROBLEM FOR MESH...ijcsit
This paper presents a genetic based approach to the partitioning and mapping of multicore SoC cores over a NoC system that uses mesh topology. The proposed algorithm performs the partitioning and mapping by reducing communication cost and minimizing power consumption by placing those intercommunicated cores as close as possible together. A program developed in C++ in which the provided specification of the multicore MPSoC system captures all data dependencies before any start of the design process. Experimental results of several multimedia benchmarks demonstrates that the genetic-based approach able to find different satisfied implementations to the problem of partitioning and mapping of MPSoC cores over mesh-based NoC system that satisfies design goals.
This document summarizes a research paper that proposes applying a genetic algorithm to solve the partitioning and mapping problem for mesh network-on-chip (NoC) systems. The genetic algorithm aims to reduce communication costs and power consumption by placing intercommunicating cores close together on the mesh topology. Experimental results on multimedia benchmarks show the genetic approach finds different solutions that satisfy design goals of partitioning and mapping multicore system-on-chip cores onto a mesh-based NoC.
Presented at the Intel Global IoT DevFest (Oct 2017)
- Real-world use cases: healthcare, building management, retail, smart cities, transportation
- Time-series analysis
- AI / ML overview & applications
A novel and innovative method for designing of rf mems deviceseSAT Journals
Abstract
The design complexity of the RF MEMS devices is increasing with fast rate, which require more accurate designing and simulation techniques. With upcoming technologies accurate simulation capability is a necessity for designing smaller chips. When devices are designed at nano scale, it presents a number of unique challenges. As the scale of the individual device decreases and the complexity of the physical structure increases, the nature of the device characteristics depart from those obtained from many of the classically held modeling concepts. Furthermore, the difficulty encountered in performing measurements on these devices means we have to put more emphasis on the results obtained from theoretical characteristics. Modeling also allows new device structures to be rigorously investigated prior to fabrication. This paper reports a novel and innovative method of design and simulation of MEMS based inductor by the method of co-simulation.
Keywords: Micro Electro Mechanical System, Radio Frequency, Co-Simulation, COMSOL Multiphysics, SOLIDWORKS, Computer Aided Design, Integrated Circuit
A novel and innovative method for designing of rf mems deviceseSAT Journals
Abstract
The design complexity of the RF MEMS devices is increasing with fast rate, which require more accurate designing and simulation techniques. With upcoming technologies accurate simulation capability is a necessity for designing smaller chips. When devices are designed at nano scale, it presents a number of unique challenges. As the scale of the individual device decreases and the complexity of the physical structure increases, the nature of the device characteristics depart from those obtained from many of the classically held modeling concepts. Furthermore, the difficulty encountered in performing measurements on these devices means we have to put more emphasis on the results obtained from theoretical characteristics. Modeling also allows new device structures to be rigorously investigated prior to fabrication. This paper reports a novel and innovative method of design and simulation of MEMS based inductor by the method of co-simulation.
Keywords: Micro Electro Mechanical System, Radio Frequency, Co-Simulation, COMSOL Multiphysics, SOLIDWORKS, Computer Aided Design, Integrated Circuit
Similar to International Journal of Engineering Inventions (IJEI) (20)
This document discusses the impact of data mining on business intelligence. It begins by defining business intelligence as using new technologies to quickly respond to changes in the business environment. Data mining is an important part of the business intelligence lifecycle, which includes determining requirements, collecting and analyzing data, generating reports, and measuring performance. Data mining allows businesses to access real-time, accurate data from multiple sources to improve decision making. Using business intelligence and data mining techniques can help businesses become more efficient and make better decisions to increase profits and customer satisfaction. The expected results of applying business intelligence include improved decision making through accurate, timely information to support organizational goals and strategic plans.
This document presents a novel technique for solving the transcendental equations of selective harmonics elimination pulse width modulation (SHEPWM) inverters based on the secant method. The proposed algorithm uses the secant method to simplify the numerical solution of the nonlinear equations and solve them faster compared to other methods. Simulation results validate that the proposed method accurately estimates the switching angles to eliminate specific harmonics from the output voltage waveform and achieves near sinusoidal output current for various modulation indices and numbers of harmonics eliminated.
This document summarizes a research paper that designed and implemented a dual tone multi-frequency (DTMF) based GSM-controlled car security system. The system uses a DTMF decoder and GSM module to allow a car to be remotely controlled and secured from a mobile phone. It works by sending DTMF tones from the phone through calls to the GSM module in the car. The decoder interprets the tones and a microcontroller executes commands to disable the ignition or control other devices. The system was created to improve car security and accessibility through remote monitoring and control with DTMF and GSM technology.
This document presents an algorithm for imperceptibly embedding a DNA-encoded watermark into a color image for authentication purposes. It applies a multi-resolution discrete wavelet transform to decompose the image. The watermark, encoded into DNA nucleotides, is then embedded into the third-level wavelet coefficients through a quantization process. Specifically, the watermark nucleotides are complemented and used to quantize coefficients in the middle frequency band, modifying the coefficients. The watermarked image is reconstructed through inverse wavelet transform. Extraction reverses these steps to recover the watermark without the original image. The algorithm aims to balance imperceptibility and robustness through this wavelet-based, blind watermarking scheme.
1) The document analyzes the dynamic saturation point of a deep-water channel in Shanghai port based on actual traffic data and a ship domain model.
2) A dynamic channel transit capacity model is established that considers factors like channel width, ship density, speed, and reductions due to traffic conditions.
3) Based on AIS data from the channel, the average traffic flow is calculated to be 15.7 ships per hour, resulting in a dynamic saturation of 32.5%, or 43.3% accounting for uneven day/night traffic volumes.
The document summarizes research on the use of earth air tunnels and wind towers as passive solar techniques. Key findings include:
- Earth air tunnels circulate air through underground pipes to take advantage of the stable temperature 4 meters below ground for cooling in summer and heating in winter. Testing showed the technique can reduce ambient temperatures by up to 14 degrees Celsius.
- Wind towers circulate air through tall shafts to cool air entering buildings at night and provide downward airflow of cooled air during the day.
- Experimental testing of an earth air tunnel system over multiple months found maximum temperature reductions of 33% in spring and minimum reductions of 15% in summer.
The document compares the mechanical and physical properties of low density polyethylene (LDPE) thin films and sheets reinforced with graphene nanoparticles. LDPE/graphene thin films were produced via solution casting, while sheets were made by compression molding. Testing showed that the thin films had enhanced tensile strength, lower melt flow index, and higher thermal stability compared to sheets. The tensile strength of thin films increased by up to 160% with 1% graphene, while sheets increased by 70%. Melt flow index decreased more for thin films, indicating higher viscosity. Thin films also showed greater improvement in glass transition temperature. These results demonstrate that processing technique affects the properties of LDPE/graphene nanocomposites.
The document describes improvements made to a friction testing machine. A stepper motor and PLC control system were added to automatically vary the load on friction pairs, replacing the manual method. Tests using the improved machine found that the friction coefficient decreases as the load increases, and that abrasive and adhesive wear increased with higher loads. The improved machine allows more accurate and convenient testing of friction pairs under varying load conditions.
This document summarizes a research article that investigates the steady, two-dimensional Falkner-Skan boundary layer flow over a stationary wedge with momentum and thermal slip boundary conditions. The flow considers a temperature-dependent thermal conductivity in the presence of a porous medium and viscous dissipation. Governing partial differential equations are non-dimensionalized and transformed into ordinary differential equations using similarity transformations. The equations are highly nonlinear and cannot be solved analytically, so a numerical solver is used. Numerical results are presented for the skin friction coefficient, local Nusselt number, velocity and temperature profiles for varying parameters like the Falkner-Skan parameter and Eckert number.
An improvised white board compass was designed and developed to enhance the teaching of geometrical construction concepts in basic technology courses. The compass allows teachers to visually demonstrate geometric concepts and constructions on a white board in an engaging, hands-on manner. It supports constructivist learning principles by enabling students to observe and emulate the teacher. The design process utilized design and development research methodology to test educational theories and validate the practical application of the compass. The improvised compass was found to effectively engage students and improve their performance in learning geometric constructions.
The document describes the design of an energy meter that calculates energy using a one second logic for improved accuracy. The meter samples voltage and current values using an ADC synchronized to the line frequency via PLL. It calculates active and reactive power by averaging the sampled values over each second. The accumulated active power for each second is multiplied by one second to calculate energy, which is accumulated and converted to kWh. Test results showed the meter achieved an error of 0.3%, within the acceptable limit for class 1 meters. Considering energy over longer durations like one second helps reduce percentage error in the calculation.
This document presents a two-stage method for solving fuzzy transportation problems where the costs, supplies, and demands are represented by symmetric trapezoidal fuzzy numbers. In the first stage, the problem is solved to satisfy minimum demand requirements. Remaining supplies are then distributed in the second stage to further minimize costs. A numerical example demonstrates using robust ranking techniques to convert the fuzzy problem into a crisp one, which is then solved using a zero suffix method. The total optimal costs from both stages provide the solution to the original fuzzy transportation problem.
1) The document proposes using an Adaptive Neuro-Fuzzy Inference System (ANFIS) controller for a Distributed Power Flow Controller (DPFC) to improve voltage regulation and power quality in a transmission system.
2) A DPFC is placed at a load bus in an IEEE 4 bus system and its performance is compared using a PI controller and ANFIS controller.
3) Simulation results show the ANFIS controller provides faster convergence and better voltage profile maintenance during voltage sags and swells compared to the PI controller.
The document describes an improved particle swarm optimization algorithm to solve vehicle routing problems. It introduces concepts of leptons and hadrons to particles in the algorithm. Leptons interact weakly based on individual and neighborhood best positions, while hadrons (local best particles) undergo strong interactions by colliding with the global best particle. When stagnation occurs, particle decay is used to increase diversity. Simulations show the improved algorithm avoids premature convergence and finds better solutions compared to the basic particle swarm optimization.
This document presents a method for analyzing photoplethysmographic (PPG) signals using correlative analysis. The method involves calculating the autocorrelation function of the PPG signal, extracting the envelope of the autocorrelation function using a low pass filter, and approximating the envelope by determining attenuation coefficients. Ten PPG signals were collected from volunteers and analyzed using this method. The attenuation coefficients were found to have similar values around 0.46, providing a potentially useful parameter for medical diagnosis.
This document describes the simulation and design of a process to recover monoethylene glycol (MEG) from effluent waste streams of a petrochemical company in Iran. Aspen Plus simulation software was used to model the process, which involves separating water, salts, and various glycols (MEG, DEG, TEG, TTEG) using a series of distillation columns. Sensitivity analyses were performed to optimize column parameters such as pressure, reflux ratio, and boilup ratio. The results showed that MEG, DEG, TEG, and TTEG could be recovered at rates of 5.01, 2.039, 0.062, and 0.089 kg/hr, respectively.
This document presents a numerical analysis of fluid flow and heat transfer characteristics of ventilated disc brake rotors using computational fluid dynamics (CFD). Two types of rotor configurations are considered: circular pillared (CP) and diamond pillared radial vane (DP). A 20° sector of each rotor is modeled and meshed. Governing equations for mass, momentum, and energy are solved using ANSYS CFX. Boundary conditions include 900K and 1500K isothermal rotor walls for different speeds. Results show the DP rotor has 70% higher mass flow and 24% higher heat dissipation than the CP rotor. Velocity and pressure distributions are more uniform for the DP rotor at higher speeds, ensuring more uniform cooling. The
This document describes the design and testing of an automated cocoa drying house prototype in Trinidad and Tobago. The prototype included automated features like a retractable roof, automatic heaters, and remote control. It aims to address issues with the traditional manual sun drying process, which is time-consuming and relies on human monitoring of changing weather conditions. Initial testing with farmers showed interest in the automated system as a potential solution.
This document presents the design of a telemedical system for remote monitoring of cardiac insufficiency. The system includes an electrocardiography (ECG) device that collects and digitizes ECG signals. The ECG signals undergo digital signal processing including autocorrelation analysis. Graphical interfaces allow patients and doctors to view ECG data and attenuation coefficients derived from autocorrelation analysis. Data is transmitted between parties using TCP/IP protocol. The system aims to facilitate remote monitoring of cardiac patients to reduce hospitalizations through early detection of health changes.
The document summarizes a polygon oscillating piston engine invention. The engine uses multiple pistons arranged around the sides of a polygon within cylinders. As the pistons oscillate, they compress and combust air-fuel mixtures to produce power. This design achieves a very high power-to-weight ratio of up to 2 hp per pound. Engineering analysis and design of a prototype 6-sided engine is presented, showing it can produce 168 hp from a 353 cubic feet per minute air flow at 12,960 rpm. The invention overcomes issues with prior oscillating piston designs by keeping the pistons moving in straight lines within cylinders using conventional piston rings.
More from International Journal of Engineering Inventions www.ijeijournal.com (20)
International Journal of Engineering Inventions (IJEI)
1. International Journal of Engineering Inventions
ISSN: 2278-7461, www.ijeijournal.com
Volume 1, Issue 8 (October2012) PP: 07-13
DESIGN OPTIMIZATION OF REINFORCED CONCRETE
BEAMS USING ARTIFICIAL NEURAL NETWORK
Sara A. Babiker1, Fathelrahman. M. Adam2, Abdelrahman E. Mohamed3
1
Sudanese Police, SUDAN
2
Nile Valley University, SUDAN
3
Sudan University of Science and Technology, SUDAN
Abstract:–– This paper presents an Artificial Neural Networks (ANN) model for the cost optimization of simply supported
beams designed according to the requirements of the ACI 318-08 code. The model formulation includes the cost of concrete,
the cost of reinforcement and the cost of formwork.
A simply supported beam was designed adopting variable cross sections, in order to demonstrate the model capabilities in
optimizing the beam design. Computer models have been developed for the structural design optimization of reinforced
concrete simple beams using NEURO SHELL-2 software. The results obtained were compared with the results obtained by
using the classical optimization model, developed in the well known Excel software spreadsheet which uses the generalized
reduced gradient (GRG). The results obtained using the two modes are in good agreement
Keywords: Reinforced Concrete Beam, Cost Optimization, Artificial Neural Networks, Generalized Reduced Gradient.
Eng. Sara A. Babiker is civil engineer in Sudanese Police. Her research interest includes design of reinforced concrete
elements and cost optimization using artificial neural network technique.
Dr Fathelrahman M. Adam is Assistant Professor of Civil Engineering at Nile Valley University, Sudan. His research
interests are analysis of shells and plates using finite element method, optimization design of reinforced elements, formwork
design.
Dr Abdelrahaman E. Mohamed is Associate Professor of Civil Engineering at Sudan University of Science and
Technology, Sudan. His research interests are Finite Element Analysis of R.C. Structures, laminated Shells and plates, shear
walls subjected to lateral and Dynamic Loads, space frame structures.
I. INTRODUCTION
A designer’s goal is to develop an “optimal solution” for the structural design under consideration. An optimal
solution normally implies the most economic structure without impairing the functional purposes the structure is supposed to
serve (Rafiq, 1995) [1]. The total cost of the concrete structure is the sum of the costs of its constituent materials; these
constituent materials are at least: concrete, reinforcement steel and formwork, (Sarma and Adeli, 1998) [2], [3]. There are
some characteristics of reinforced concrete (RC) structures which make their design optimization distinctly different from
other structures. The cost of RC structures is influenced by several cost items. In the design optimization of RC structures the
cross-sectional dimensions of elements and detailing of reinforcement e.g. size and number of steel bars, need to be
determined. Consequently, the number of design parameters, which need to be optimized for a RC structure can be larger
than that for a steel structure. Also cracking and durability requirements are two characteristic properties of RC structures;
these increase the number of design constraints of the optimization problem of RC structures. (Sahab, 2002) [4]
The existence of optimization methods can be traced to the days of Newton, Lagrange, Bernoulli, Euler, Lagrange
and Weirstrass [5]. Despite of these early contributions, very little progress was made until the middle of the twentieth
century, when high-speed digital computers made implementation of the optimization procedures possible and stimulated
further research on new methods. Spectacular advances followed, producing a massive literature on optimization techniques.
This advancement also resulted in the emergence of several well defined new areas in optimization theory. The development
of the simplex method by Dantzig in 1963 [6] for linear programming problems and the annunciation of the principle of
optimality in 1957 [7] by Bellman for dynamic programming problems paved the way for development of the methods of
constrained optimization. Work by Kuhn and Tucker on the necessary and sufficiency conditions for the optimal solution of
programming problems laid the foundations for a great deal of later research in nonlinear programming. The contributions of
Zoutendijk and Rosen to nonlinear programming during the early 1960 [8] have been significant. Although no single
technique has been found to be universally applicable for nonlinear programming problems, work of Carroll and Fiacco and
McCormick allowed many difficult problems to be solved by using the well-known techniques of unconstrained
optimization. Geometric programming was developed in 1967 [9] by Duffin, Zener, and Peterson. Gomory did pioneering
work in integer programming.
An artificial neural network is a system based on the operation of biological neural networks, in other words, is an
emulation of biological neural system. Why would the implementation of artificial neural networks be necessary? Although
computing these days is truly advanced, there are certain tasks that a program made for a common microprocessor is unable
to perform; even so a software implementation of a neural network can be made with their advantages and Disadvantages.
ISSN: 2278-7461 www.ijeijournal.com P a g e |7
2. DESIGN OPTIMIZATION OF REINFORCED…
Another aspect of the artificial neural networks is that there are different architectures, which consequently require
different types of algorithms, but despite being an apparently complex system, a neural network is relatively simple.
Artificial Neural Networks (ANN) is among the newest signal-processing technologies in the engineer's toolbox.. In
engineering, neural networks serve two important functions: as pattern classifiers and as nonlinear adaptive filters. An
Artificial Neural Network is an adaptive, most often nonlinear system that learns to perform a function (an input/output map)
from training data. Adaptive means that the system parameters are changed during operation, normally called the training
phase. After the training phase the Artificial Neural Network parameters are fixed and the system is deployed to solve the
problem at hand (the testing phase). The Artificial Neural Network is built with a systematic step-by-step procedure to
optimize a performance criterion or to follow some implicit internal constraint, which is commonly referred to as the
learning rule . The input/output training data are fundamental in neural network technology, because they convey the
necessary information to "discover" the optimal operating point. The nonlinear nature of the neural network processing
elements (PEs) provides the system with lots of flexibility to achieve practically any desired input/output map, i.e., some
Artificial Neural Networks are universal mappers.
An input is presented to the neural network and a corresponding desired or target response set at the output (when
this is the case the training is called supervised). An error is composed from the difference between the desired response and
the system output. This error information is fed back to the system and adjusts the system parameters in a systematic fashion
(the learning rule). The process is repeated until the performance is acceptable. It is clear from this description that the
performance hinges heavily on the data. If one does not have data that cover a significant portion of the operating conditions
or if they are noisy, then neural network technology is probably not the right solution. On the other hand, if there is plenty of
data and the problem is poorly understood to derive an approximate model, then neural network technology is a good choice.
This operating procedure should be contrasted with the traditional engineering design, made of exhaustive subsystem
specifications and intercommunication protocols. In artificial neural networks, the designer chooses the network topology,
the performance function, the learning rule, and the criterion to stop the training phase, but the system automatically adjusts
the parameters. So, it is difficult to bring prior information into the design, and when the system does not work properly it is
also hard to incrementally refine the solution. But ANN-based solutions are extremely efficient in terms of development time
and resources, and in many difficult problems artificial neural networks provide performance that is difficult to match with
other technologies. Denker states that "artificial neural networks are the second best way to implement a solution" motivated
by the simplicity of their design and because of their universality, only shadowed by the traditional design obtained by
studying the physics of the problem. At present, artificial neural networks are emerging as the technology of choice for many
applications such as pattern recognition, prediction, system identification, and control.
II. THE MODEL OF OPTIMAL DESIGN
In an optimization problem some of the parameters can be considered as preassigned or fixed parameters and
others are considered as design variables. The design variables are determined in such a way that the value of an objective
function, which is often the ost of the structure, becomes minimum. Some restrictions, called design constraints, may limit
the acceptable values of the design variables. A simply supported rectangular RC beam model is studied in this paper.
Fixed Parameters
The present model is designed to consider all fixed parameters that may have an impact on the cost optimization of
simple beams. These include the characteristic strength, modulus of elasticity, unit weight of concrete and reinforcement and
the intensity of the dead and live loads. In addition it is assumed that the total cost of concrete and reinforcement is
proportional to volume and weight of each material, respectively.
Consequently, the total cost of a structure is calculated using fixed parameters to calculate the cost of unit volume
of concrete and unit weight of reinforcement.
The Design Variables
An important first step in the formulation of an optimization problem is to identify the design variables. Design
variables should be independent of each other. If one of the design variables can be expressed in terms of others then that
variable can be eliminated from the model.
Any structure can described by a set of quantities some of these quantites are pre-assigned because the designer is
not free to change them or because it may be known from experience that certain value for these quantites produce a good
result this type of quantities is called pre-assigned parameters the other type is the design variables .the design variables
represent some or all of the following properties of the structure.
The design variables which were considered in this RC beam model are listed below:
b = Beam width (integer values)
d = Effective beam depth (real values)
nb = Number of flexural bars (integer values)
db = Diameter of flexural bar (integer values)
The Design Variables’ bounds
The variables bounds result from different issues such as the provisions of the code under consideration, the
aesthetic of the structural elements in the building, the practical issues and the availability of some sizes of the material at the
local market. The following Equations are the bounds considered for the model.
Effective depth:
ISSN: 2278-7461 www.ijeijournal.com P a g e |8
3. DESIGN OPTIMIZATION OF REINFORCED…
dmin = hmin-db/2 - Sc - ds , dmax = hmax- db/2 - Sc - ds (1)
Effective width:
b ≥ bmin, b ≤ bmax (2)
Wher:e bmin and bmax are chosen according to architectural and practical considerations.
Bar diameter:
db ≥ dbmin , db ≤ dbmax (3)
Where: dbmin and dbmax are chosen according to range of reinforcement available at market.
Number of bars:
nb ≥ nbmin , nb ≤ nbmax (4)
Where: nbmin and nbmax are chosen according to practical considerations.
Where:
hmin = L (with refer to the Table 9.5a at the ACI - code)
16
hmax is chosen according to architectural considerations.
Sc is the concrete cover.
ds is the diameter of stirrups.
b (in meter) [ 0.25, 0.35 ] i.e 0.25 ≤ b ≤ 0.35 m
d (in meter) [0.059 ,1.184 ]
nb [ 4 ,12 ]
db (in mm) [12 ,24 ]
The Constraints
In many practical problems, the design variables cannot be chosen arbitrarily; rather, they have to satisfy certain
specified functional and other requirements. The restrictions that must be satisfied to produce an acceptable design which
collectively called design constraints. The design constraints which are considered in this optimization model are listed
below:
Design for flexure
Mu ≤ Ф Mn (5)
Minimum spacing between flexural bars
S ≥ Smin (6)
Smin = max of (db, ¾ of max agg. Size, 2.5) cm (7)
Maximum spacing between bars (cracking control)
S ≤ S max (8)
280
S max min 300 * * fy , 380 *280 / 0.67 * fy 2.5 *50 cm ACI code (9)
0.67
Maximum and minimum reinforcement ratios
Ast Ast
m ax m in (10)
bd bd
Design for shear
ISSN: 2278-7461 www.ijeijournal.com P a g e |9
4. DESIGN OPTIMIZATION OF REINFORCED…
VS 2
3
f 'c * b * d kN ACI code (11)
Deflection control
Md 3Mcr Ms 3Mcr (12)
(∆i) ℓ ≤ limit by ACI-code. as mentioned in table (4.1) in section (4.1.16)
2*(∆i) d + ℓ + (∆i)ℓ ≤ limit by ACI-code, as mentioned in table (4.1) in section (4.1.16)
Where:
Md = Bending moment under service dead loads only.
Ms = Bending moment under service dead and live loads.
Mcr = cracking bending moment.
(∆i)ℓ = immediate deflection due to live load only.
2* (∆i) d + ℓ = long term deflection due to service dead an live loads.
The Objective Function
The objective function aim to finding an acceptable or adequate design that merely satisfies the functional and
other requirements of the problem. there will be more than one acceptable design, and the purpose of optimization is to
choose the best one of the many acceptable designs available. Thus a criterion has to be chosen for comparing the different
alternative acceptable designs and for selecting the best one. The criterion with respect to which the design is
optimized,when expressed as a function of the design variables, is known as the criterion or meritor objective function. The
choice of objective function is governed by the nature of problem. The objective function for minimization is generally taken
as weight in aircraft and aerospace structural design problems. In civil engineering structural designs, the objective is usually
taken as the minimization of cost. in structural design,the minimum weight design may not correspond to minimum stress
design, and the minimum stress design, again, may not correspond to maximum frequency design. Thus the selection of the
objective function can be one of the most important decisions in the whole optimum design process.
The objective function for the simply supported reinforced concrete beam model is:
MIN COST = Cc[(Ac - As) L] + Cc [ As L)] + Cw [2(b + h)] SDG (13)
Where:
Cc = Cost of concrete per cubic meter.
Cs = Cost of reinforcement steel per ton.
Cw = Cost of concrete formwork along the vertical and horizontal surface per square meter.
Ac = Area of concrete cross section.
As = Area of longitudinal reinforcement.
L = Span of the beam
b = width of the beam.
h = depth of the beam.
The model was run to optimize the design of a rectangular cross section for this beam and loadings while satisfying the
provisions of the ACI 318-08 Code.
III. EXAMPLE PROBLEMS
Neural Network For Optimum Design Of Beams
Design a least-cost reinforced concrete of beam simply supported span of 4 m supporting a uniform dead load of
1.5 kN/m and uniform live load of 1 kN/m. The concrete strength (f’c) is 28 MPa and the steel yield strength (fy) is 420 MPa.
The cost of concrete per cubic meter (Cc) is 1200 SDG; the cost of normal steel bars per ton (Cs) is 4750 SDG and the Cost
of concrete formwork along the vertical and horizontal surface per meter is 45 SDG.
The developed database for the optimum design of beams according to the requirement of ACI code, which are
based on the equations above, were used to train a neural network. The design input to the problem includes the values
mentioned above and listed in the Table 1 shown below. With a comprehensive set of examples about 73 examples have
been obtained for different values of width (b) ranging between 0.3 m and 0.35 m with increment of 0.0014 m with aided by
using spreadsheet for that. For each set, the depth of beam, reinforcement required and cost are obtained. Out of these, the
examples have been used for training. The outputs are presented in Figures 1 and 2. Figure 1 presents the relation between
the overall depth and the area of steel and Figure 2 presents the relation between the overall depth and the minimum cost.
The main output is the optimization cost which it found equal to the amount of 635.81 SDG.
The Classical Optimization Spreadsheet Models
In addition to the developed neural network model, another model is developed for the design optimization of RC
simple beams using EXCEL software spreadsheets.
The Generalized Reduced Gradient (GRG) method is considered one of the classical optimization methods. This
method was chosen among other classical methods since it is already programmed in the EXCEL SOLVER. Therefore the
user builds the design model and then uses the SOLVER toolbox to run the optimization process. According to the user
interface and the input data required for the model, the model output the optimization cost value of 633.45 SDG.
ISSN: 2278-7461 www.ijeijournal.com P a g e | 10
5. DESIGN OPTIMIZATION OF REINFORCED…
Table 1 Part of the developed optimization for the RC simple beam
INPUT
The Parameter The value The unit
Span Length 4 M
Uniform distributed dead load (wd) 1.5 kN/m
Uniform distributed live load (wL) 1.0 kN/m
f'c 28 MPa
fy 420 MPa
max. agg. Size 25 Mm
cost of concrete per cubic meter( Cc) 1200 SDG
cost of steel per ton (Cs) 4750 SDG
cost of formwork per meter (Cw) 45 SDG
DESIGN VARIABLE
Width (b) in effective depth in Diameter db in
no of bars
m m mm
0.3 0.184 16 x4 6
minimum values of design variables 0.25 0.059 12 4
maximum values of design variables 0.35 1.184 24 12
Optimization Cost (ANN Optimization) 635.81 SDG
Optimization Cost (Classical Optimization) 633.45 SDG
0.012
B = 0.35
0.01 B = 0.30
B = 0.25
Area of Steel
0.008
0.006
0.004
0.002
0
0.120 0.320 0.520 0.720 0.920 1.120 1.320
Overall Depth (H)
Figure 1 shown the relation between overall depth and area of steel for the RC simple beam
ISSN: 2278-7461 www.ijeijournal.com P a g e | 11
6. DESIGN OPTIMIZATION OF REINFORCED…
0.35
5004 0.30
0.25
4004
Min Cost
3004
2004
1004
4
0.120 0.320 0.520 0.720 0.920 1.120 1.320
Total Depth (H)
Figure 2 shown the relation between overall depth and min cost for the RC simple beam with variable depth.
Comparison between the Developed ANN and Classical Optimization Models
In order to check the validity of the developed ANN optimization models, the results of these models were
compared with those of the classical optimization models. The optimum solution for the cost of RC simple beam according
to the data listed in Table (1), and by using the ANN model was found 635.81 SDG and by using the spreadsheet classical
model it is 633.45 SDG with percentage difference of 0.37 % which verified that the results obtained by using ANN model
is acceptable.
IV. CONCLUSIONS
The paper presents a design optimization model for simply supported concrete beams using Artificial Neural
Networks (ANN) as an optimization technique. The main conclusions drawn can be summarized as follows:
ANN optimization model based on NEURO SHELL 2 software was developed for the design optimization of
reinforced concrete simple beams.
A design optimization model based on EXCEL spreadsheets was developed for the same reason above to verify
the work. This model applies the Generalized Reduced Gradient (GRG) optimization method as one of the
classical optimization techniques.
The results obtained from the ANN optimization technique showed good agreement with the one obtained by the
GRG technique with a percentage difference of 0.37 %.
The optimum solution presented satisfies the provisions of the code and minimizes the cost of the structure. This
may be of great value to practicing engineers.
The results presented in Figures 1 and 2, can be used for quickly finding the area of steel and the cost respectively
according to the values of overall depth ranging from 0.12 m to 1.3 m coincide with values of width ranging
between 0.25 and 0.35 m provided that the design provisions presented are adopted.
REFERENCES
1. RAFIG, M Y, Genetic algorithms in optimum design, capacity check and final detailing of reinforced concrete columns,
School of Civil and Engineering University of Plymouth, Plymouth, 1995
2. SARMA A, K C AND ADELI H, Cost optimization of concrete structures, Journal of Structural Engineering, ASCE,
Vol. 124, No. 5, 1998, pp. 570-578.
3. SARMA B, K C AND ADELI H Cost optimization of steel structures, Engineering Optimization, Vol. 32, 2000, pp.
777-802.
4. SAHAB A, AND M G, Cost optimization of reinforced concrete flat slab buildings, PhD thesis, University of Bradford,
UK., 2002.
5. SAHAB A, AND M G, ASHOUR A F AND TOROPOV V V, Cost optimization of reinforced concrete flat slab
buildings, Engineering Structures, Vol. 27, 2004, pp. 313–322
6. CHAKRABARTY A AND B K, Models for optimal design of reinforced concrete beams, Computers and Structures,
Vol. 42, No. 3, 1992, pp. 447-451.
ISSN: 2278-7461 www.ijeijournal.com P a g e | 12
7. DESIGN OPTIMIZATION OF REINFORCED…
7. DAMU A, KARIHALLO A, KARIHALOO, B L AND ROZANY G I N, Minimum cost design of reinforced concrete
beams using continuum-type optimality criteria, Structural Optimization, Vol. 7, No. 1/2, 1994, pp. 91-102.
8. Al-SALLOUM Y A AND SIDDQI G H, Cost-optimum design of concrete beams, ACI Structural Journal, Vol. 91, No.
6, 1994, pp. 647-655.
9. SENOUI A B AND ABDUL-SALAM M A, Prediction of Reinforced Concrete Beam Depth Using Neural Networks,
Engineering Journal of the University of Qatar, Vol. 11, 1998, pp. 117-132.
10. ACI COMMITTEE 318, BUILING CODE REQUIREMENTS FOR STRUCTURAL CONCRETE, ACI 318-08,
American Concrete Institute, 2008.
11. STEPHEN L, NELSON AND DAVID B, MBA’s Guide to Microsoft Excel 2002, Maguiness Redmond Technology
Press, 2001.
ISSN: 2278-7461 www.ijeijournal.com P a g e | 13