7th Workshop on Advanced Control and Diagnosis
The Workshop on Advanced Control and Diagnosis (ACD) organised by the European Institute for Applied Research (IAR) brought together academics and engineers in control engineering and computer science.
Jeffrey R. Wittrock led a 3.8 million dollar plant expansion in 2008 that included new warehousing and manufacturing space meeting EPA and OSHA regulations. He designed an automated high speed paint filling line that reduced costs by 20%. Wittrock launched a business transformation plan in 2008 focusing on safety, technology, and facility upgrades. The plan improved the organization. He achieved energy cost savings over 4 years through automation and strict preventative maintenance. Wittrock reduced labor costs by over $600K annually through process redesign, flow enhancement, and workforce leadership.
The technology integration plan for Cedar Creek Elementary has four goals:
1. Create and communicate a campus vision for technology use by gathering input from teachers and developing a vision aligned with district and campus educational goals.
2. Provide effective leadership for integrating technology into the curriculum by holding family technology events and encouraging digital communication with parents.
3. Improve staff technology skills by reviewing performance data, ensuring understanding of technology terminology, and setting implementation goals.
4. Improve student achievement and teaching effectiveness through targeted technology training and using data to drive instruction with tools like interactive boards, assessment platforms, and classroom response systems.
The document discusses the STaR Chart, which assesses a school's progress toward technology goals and readiness based on a long-range technology plan. It measures areas like teaching and learning, educator preparation, leadership, and infrastructure. Cedar Creek Elementary's 2008-2009 STaR Chart data is presented, showing general upward progress but room for improvement compared to other schools. The mission and goals of integrating 21st century skills through technology are outlined.
Haltom Middle School conducted a STaR Chart Summary to assess the school's technology readiness based on a long-range technology plan. The STaR Chart is a teacher tool that provides data on the school's technology planning, professional development needs, and support for grant applications. The summary shows Haltom Middle School is at the Developing Tech level for Teaching and Learning, Administration and Support, and Infrastructure, but at the Advanced Tech level for Educator Preparation.
The Texas Teacher STaR Chart is a survey that helps teachers evaluate their educational technology skills and progress toward technology goals. It addresses teaching and learning, educator preparation, infrastructure, and administrative support. Teachers can use their results to identify areas for improvement and set personal goals aligned with state, district, and campus targets.
IRJET- Vibration Analysis and Optimization of Housing for ECU in Automobile u...IRJET Journal
The document discusses the vibration analysis and optimization of an electronic control unit (ECU) housing for an automobile. The objectives are to design and optimize the ECU housing using CATIA and finite element analysis (FEA) in ANSYS to minimize vibration. Modal and harmonic analyses are conducted on the original housing and an optimized design with added cross ribs. The natural frequencies increase and acceleration amplitude decreases in the optimized design, improving stiffness and reducing vibration compared to the original housing.
SVC device optimal location for voltage stability enhancement based on a comb...TELKOMNIKA JOURNAL
The increased power system loading combined with the worldwide power industry deregulation requires more reliable and efficient control of the power flow and network stability. Flexible AC transmission systems (FACTS) devices give new opportunities for controlling power and enhancing the usable capacity of the existing transmission lines. This paper presents a combined application of the particle swarm optimization (PSO) and the continuation power flow (CPF) technique to determine the optimal placement of static var compensator (SVC) in order to achieve the static voltage stability margin. The PSO objective function to be maximized is the loading factor to modify the load powers. In this scope, two SVC constraints are considered: the reference voltage in the first case and the total reactance and SVC reactive power in the second case. To test the performance of the proposed method, several simulations were performed on IEEE 30-Bus test systems. The results obtained show the effectiveness of the proposed method to find the optimal placement of the static var compensator and the improvement of the voltage stability.
Star Test Topology for Testing Printed Circuits BoardsIRJET Journal
This document presents a new testing methodology called star test topology (STT) for testing printed circuit boards. STT aims to address limitations of traditional testing methods such as being manual, limited by chip complexity, and requiring expensive test equipment. STT involves developing a shared test access port over the entire PCB and redesigning on-chip design-for-testability circuitry. In STT, devices under test are connected in a star topology with a central test access port acting as a hub. This allows test patterns to be broadcast to devices and results returned, with minimal pins/resources required. The document describes simulating STT using circuit design software and capturing output signals with a logic analyzer.
Jeffrey R. Wittrock led a 3.8 million dollar plant expansion in 2008 that included new warehousing and manufacturing space meeting EPA and OSHA regulations. He designed an automated high speed paint filling line that reduced costs by 20%. Wittrock launched a business transformation plan in 2008 focusing on safety, technology, and facility upgrades. The plan improved the organization. He achieved energy cost savings over 4 years through automation and strict preventative maintenance. Wittrock reduced labor costs by over $600K annually through process redesign, flow enhancement, and workforce leadership.
The technology integration plan for Cedar Creek Elementary has four goals:
1. Create and communicate a campus vision for technology use by gathering input from teachers and developing a vision aligned with district and campus educational goals.
2. Provide effective leadership for integrating technology into the curriculum by holding family technology events and encouraging digital communication with parents.
3. Improve staff technology skills by reviewing performance data, ensuring understanding of technology terminology, and setting implementation goals.
4. Improve student achievement and teaching effectiveness through targeted technology training and using data to drive instruction with tools like interactive boards, assessment platforms, and classroom response systems.
The document discusses the STaR Chart, which assesses a school's progress toward technology goals and readiness based on a long-range technology plan. It measures areas like teaching and learning, educator preparation, leadership, and infrastructure. Cedar Creek Elementary's 2008-2009 STaR Chart data is presented, showing general upward progress but room for improvement compared to other schools. The mission and goals of integrating 21st century skills through technology are outlined.
Haltom Middle School conducted a STaR Chart Summary to assess the school's technology readiness based on a long-range technology plan. The STaR Chart is a teacher tool that provides data on the school's technology planning, professional development needs, and support for grant applications. The summary shows Haltom Middle School is at the Developing Tech level for Teaching and Learning, Administration and Support, and Infrastructure, but at the Advanced Tech level for Educator Preparation.
The Texas Teacher STaR Chart is a survey that helps teachers evaluate their educational technology skills and progress toward technology goals. It addresses teaching and learning, educator preparation, infrastructure, and administrative support. Teachers can use their results to identify areas for improvement and set personal goals aligned with state, district, and campus targets.
IRJET- Vibration Analysis and Optimization of Housing for ECU in Automobile u...IRJET Journal
The document discusses the vibration analysis and optimization of an electronic control unit (ECU) housing for an automobile. The objectives are to design and optimize the ECU housing using CATIA and finite element analysis (FEA) in ANSYS to minimize vibration. Modal and harmonic analyses are conducted on the original housing and an optimized design with added cross ribs. The natural frequencies increase and acceleration amplitude decreases in the optimized design, improving stiffness and reducing vibration compared to the original housing.
SVC device optimal location for voltage stability enhancement based on a comb...TELKOMNIKA JOURNAL
The increased power system loading combined with the worldwide power industry deregulation requires more reliable and efficient control of the power flow and network stability. Flexible AC transmission systems (FACTS) devices give new opportunities for controlling power and enhancing the usable capacity of the existing transmission lines. This paper presents a combined application of the particle swarm optimization (PSO) and the continuation power flow (CPF) technique to determine the optimal placement of static var compensator (SVC) in order to achieve the static voltage stability margin. The PSO objective function to be maximized is the loading factor to modify the load powers. In this scope, two SVC constraints are considered: the reference voltage in the first case and the total reactance and SVC reactive power in the second case. To test the performance of the proposed method, several simulations were performed on IEEE 30-Bus test systems. The results obtained show the effectiveness of the proposed method to find the optimal placement of the static var compensator and the improvement of the voltage stability.
Star Test Topology for Testing Printed Circuits BoardsIRJET Journal
This document presents a new testing methodology called star test topology (STT) for testing printed circuit boards. STT aims to address limitations of traditional testing methods such as being manual, limited by chip complexity, and requiring expensive test equipment. STT involves developing a shared test access port over the entire PCB and redesigning on-chip design-for-testability circuitry. In STT, devices under test are connected in a star topology with a central test access port acting as a hub. This allows test patterns to be broadcast to devices and results returned, with minimal pins/resources required. The document describes simulating STT using circuit design software and capturing output signals with a logic analyzer.
IOT Based Three Phase Transmission Line Fault Detection and ClassificationIRJET Journal
This document describes a proposed IOT-based system for detecting and classifying faults on three-phase transmission lines in real-time. The system uses an Arduino board as the central controller connected to voltage transformers, an LCD display, WiFi module, and relays. When a fault occurs, the Arduino detects the type of fault (e.g. line-to-ground), calculates the exact location, and sends a signal to isolate the faulted section via relays. The information is also transmitted remotely over WiFi and displayed locally on the LCD. The system aims to minimize human effort in fault management and improve reliability of power transmission.
Automatic Fault Detection System with IOT BasedYogeshIJTSRD
The fault location is an important part for any transmission line and distribution system. The location of fault is difficult task sometimes it takes lot of times needed for the exact location of the fault. The exact fault location can help the service man to overcome the fault free system in very less time. In this paper we are able to detect the fault range in easy way using the ESP module and the message is transferred on the mobile. This project is cost effective and reliable. Fast fault detection provide the protection of equipment before any significant damage. Er. Sanjeev Kumar | Mohd Mehraj Khan | Nadeem | Shailesh Kumar Yadav | Harsh Gupta "Automatic Fault Detection System with IOT Based" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-5 | Issue-5 , August 2021, URL: https://www.ijtsrd.com/papers/ijtsrd43806.pdf Paper URL: https://www.ijtsrd.com/engineering/electrical-engineering/43806/automatic-fault-detection-system-with-iot-based/er-sanjeev-kumar
IRJET- A Research on Vibration Analysis & Optimization of Housing for ECU in ...IRJET Journal
This document discusses a research project that aims to analyze and optimize the housing for an electronic control unit (ECU) in an automobile using finite element analysis (FEA) and an FFT analyzer. The researchers first design an ECU housing model using CAD software. They then perform modal and harmonic analysis using FEA to investigate the housing's mode shapes and response at different frequencies. Experimental modal analysis is also conducted using an accelerometer, impact hammer, and FFT analyzer. The FEA results are compared to the experimental results. The objective is to minimize vertical vibrations in the ECU housing by adding stiffening ribs and optimizing the design to withstand vibrations from the engine.
Recently, research has picked up a fervent pace in the area of fault diagnosis of electrical vehicle. Like failures of a position sensor, a voltage sensor, and current sensors. Three-phase induction motors are the “workhorses” of industry and are the most widely used electrical machines. This paper presents a scheme for Fault Detection and Isolation (FDI). The proposed approach is a sensor-based technique using the mains current measurement. Current sensors are widespread in power converters control and in electrical drives. Thus, to ensure continuous operation with reconfiguration control, a fast sensor fault detection and isolation is required. In this paper, a new and fast faulty current sensor detection and isolation is presented. It is derived from intelligent techniques. The main interest of field programmable gate array is the extremely fast computation capabilities. That allows a fast residual generation when a sensor fault occurs. Using of Xilinx System Generator in Matlab / Simulink allows the real-time simulation and implemented on a field programmable gate array chip without any VHSIC Hardware Description Language coding. The sensor fault detection and isolation algorithm was implemented targeting a Virtex5. Simulation results are given to demonstrate the efficiency of this FDI approach.
Incorporation of IoT in Assembly Line Monitoring SystemIRJET Journal
This document proposes incorporating IoT into assembly line monitoring systems to make them more efficient. It suggests using master and slave electronic trackers, where slaves at each assembly station update the master at the end of the line via the internet. This allows real-time data transfer without costly wired connections. The master display would show assembly progress and statistics for supervisors to remotely monitor production. The proposed system could reduce costs for many assembly lines compared to traditional hard-wired monitoring systems.
This document summarizes a research paper that proposes using a genetic algorithm-optimized fuzzy controller to improve power quality issues like voltage sags and total harmonic distortion using a static synchronous compensator (STATCOM). The paper describes how a genetic algorithm can be used to optimize the parameters of a fuzzy logic controller for the STATCOM in order to minimize voltage sag and total harmonic distortion more effectively than conventional controllers. Simulation results demonstrate that the proposed genetic algorithm technique improves sag compensation and reduces harmonic distortion in distribution systems during fault conditions.
IRJET- Overhead Line Protection with Automatic Switch by using PLC AutomationIRJET Journal
This document describes a system to automatically protect overhead transmission lines from overload faults using a PLC-controlled air break switch. The system monitors current on the line using a sensor and can open the switch if overload is detected. This allows faulty sections to be isolated without interrupting the whole line. The automatic switch provides remote operation and is more reliable than conventional manual switches. It aims to quickly detect and resolve overload problems through automation using a PLC to control the switch based on current readings.
IRJET- A Review on Wireless Sensor System of Fault Detection of Motor ArraysIRJET Journal
This document reviews methods for fault detection in motor arrays using wireless sensor systems. It discusses induction motors, which are widely used in industry. Early fault detection is important for motor protection and reducing downtimes. The document surveys various fault detection methods, including motor current signature analysis, artificial neural networks applied to bearing fault detection using vibration data, and reflectometry techniques for detecting ground faults in photovoltaic arrays. It examines prior research applying methods such as dynamic space parity and artificial neural networks to detect faults in DC motors and induction motor bearings. Overall, the document aims to discuss health monitoring and fault detection techniques for induction motors.
IRJET- Modelling and Condition Monitoring of 3훟 Induction Motor using Fuzzy L...IRJET Journal
This document presents a MATLAB model and simulation of a 3-phase induction motor for condition monitoring using fuzzy logic. The model monitors for common faults like stator winding issues, voltage imbalances, and phase faults. Fuzzy logic membership functions and rules are used to assess the motor's health based on stator current values. Simulations show the motor operating normally and with introduced faults like turn-to-turn shorts and broken windings. In fault cases, the stator current becomes unbalanced and the fuzzy logic output indicates the motor health deteriorating from good to damaged or seriously damaged states.
Design for Testability in Timely Testing of Vlsi CircuitsIJERA Editor
Even though a circuit is designed error-free, manufactured circuits may not function correctly. Since the manufacturing process is not perfect, some defects such as short-circuits, open-circuits, open interconnections, pin shorts, etc., may be introduced. Points out that the cost of detecting a faulty component increases ten times at each step between prepackage component test and system warranty repair. It is important to identify a faulty component as early in the manufacturing process as possible. Therefore, testing has become a very important aspect of any VLSI manufacturing system.Two main issues related to test and security domain are scan-based attacks and misuse of JTAG interface. Design for testability presents effective and timely testing of VLSI circuits. The project is to test the circuits after design and then reduce the area, power, delay and security of misuse. BIST architecture is used to test the circuits effectively compared to scan based testing. In built-in self-test (BIST), on-chip circuitry is added to generate test vectors or analyze output responses or both. BIST is usually performed using pseudorandom pattern generators (PRPGs). Among the advantages of pseudorandom BIST are: (1) the low cost compared to testing from automatic test equipment (ATE). (2) The speed of the test, which is much faster than when it is applied from ATE. (3) The applicability of the test while the circuit is in the field, and (4) the potential for high quality of test.
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.
Black Box Model based Self Healing Solution for Stuck at Faults in Digital Ci...IJECEIAES
The paper proposes a design strategy to retain the true nature of the output in the event of occurrence of stuck at faults at the interconnect levels of digital circuits. The procedure endeavours to design a combinational architecture which includes attributes to identify stuck at faults present in the intermediate lines and involves a healing mechanism to redress the same. The simulated fault injection procedure introduces both single as well as multiple stuck-at faults at the interconnect levels of a two level combinational circuit in accordance with the directives of a control signal. The inherent heal facility attached to the formulation enables to reach out the fault free output even in the presence of faults. The Modelsim based simulation results obtained for the Circuit Under Test [CUT] implemented using a Read Only Memory [ROM], proclaim the ability of the system to survive itself from the influence of faults. The comparison made with the traditional Triple Modular Redundancy [TMR] exhibits the superiority of the scheme in terms of fault coverage and area overhead.
FUZZY LOGIC APPROACH FOR FAULT DIAGNOSIS OF THREE PHASE TRANSMISSION LINEJournal For Research
This document summarizes a journal article that proposes using fuzzy logic to diagnose faults on three-phase transmission lines. It begins with an abstract of the journal article, which describes using fuzzy logic as an intelligent technique to quickly and accurately identify the type of fault that occurs on a transmission system. It then provides background on transmission line faults, fault types, and challenges with transmission line protection. The document outlines the proposed fuzzy logic approach, including defining fault types as fuzzy sets and developing if-then rules to relate transmission line voltages and currents to faults. Simulation results are presented showing the fuzzy logic approach can identify different fault types based on the current responses. The conclusion is that the proposed fuzzy logic method allows for fast and reliable fault detection on transmission
IRJET-Condition Monitoring based Control using Piezo Sensor for Rotating Elec...IRJET Journal
This document discusses condition monitoring of rotating electrical motors using piezoelectric sensors. It presents a simulation model developed to detect problems in motors based on vibration analysis. If vibration exceeds unsatisfactory or unacceptable thresholds, the system will display alerts on a computer screen to indicate defective parts. Fast Fourier Transform (FFT) analysis and Motor Current Signature Analysis (MCSA) are used to diagnose faults in induction motors. The document focuses on developing this condition monitoring system to protect motors from unexpected shutdowns and increase lifetime through early problem detection without requiring human observation.
IRJET- A Simple Approach to Identify Power System Transmission Line Faults us...IRJET Journal
This paper presents a Probabilistic Neural Network (PNN) approach for identifying and classifying faults on power transmission lines. The PNN is trained on voltage waveform data simulated using Electromagnetic Transient Program (EMTP) software for different fault types and locations on a 150km transmission line. Only two sets of simulated data are used to train the PNN, requiring less computation than other methods that preprocess data. The trained PNN is able to accurately identify and classify fault types based on the voltage waveform, which helps ensure reliable power transmission by isolating only faulty lines or phases.
IRJET- Fuzzy Logic based Fault Detection in Induction Machines using CloudIRJET Journal
This document presents a system for online condition monitoring of induction motors using fuzzy logic and cloud computing. It involves two phases:
1) Developing a simulation model in MATLAB/Simulink to monitor motor parameters and detect faults.
2) Implementing an online condition monitoring system using a Raspberry Pi, sensors to measure motor vibrations, temperature etc., and a cloud platform to remotely store and access data in real-time.
Fuzzy logic is used to analyze sensor data and detect faults based on defined rules and membership functions. The system aims to monitor key parameters of induction motors and identify failures to enable safe and efficient operation in industrial applications.
Design Verification and Test Vector Minimization Using Heuristic Method of a ...ijcisjournal
The reduction in feature size increases the probability of manufacturing defect in the IC will result in a
faulty chip. A very small defect can easily result in a faulty transistor or interconnecting wire when the
feature size is less. Testing is required to guarantee fault-free products, regardless of whether the product
is a VLSI device or an electronic system. Simulation is used to verify the correctness of the design. To test n
input circuit we required 2n
test vectors. As the number inputs of a circuit are more, the exponential growth
of the required number of vectors takes much time to test the circuit. It is necessary to find testing methods
to reduce test vectors . So here designed an heuristic approach to test the ripple carry adder. Modelsim and
Xilinx tools are used to verify and synthesize the design.
Underground Cable Fault Detection Using IOTIRJET Journal
This document discusses a system to detect faults in underground cable lines using IoT. It proposes using a microprocessor, LCD display, fault sensing circuit module, LoRa module, and power supply to detect the location and type of fault (single line to ground, double line to ground, or three phase faults). The system measures voltage changes across series resistors when a short circuit occurs to determine the fault location. It can display the fault location and phase on the LCD and transmit the data over WiFi. The document reviews literature on condition monitoring of underground cables, current transformer saturation effects, and comparing optical and magnetic current transformers.
Noise analysis & qrs detection in ecg signalsHarshal Ladhe
The document discusses removing noise from ECG signals using adaptive filtering techniques. It focuses on using an LMS algorithm to remove powerline interference at 50 Hz from ECG signals. The LMS algorithm is tested with different filter tap lengths and step sizes to determine the optimal parameters for noise cancellation. Additional filtering using notch filters is also explored to remove harmonics and high frequency noise. The results show that the LMS algorithm effectively removes powerline interference from ECG signals.
IRJET- Power Quality Improvement by using Three Phase Adaptive Filter Control...IRJET Journal
This document discusses a proposed adaptive filter control system for improving power quality in a microgrid. The system contains a combination of solar PV and diesel generation connected through a voltage source converter. An adaptive filter is designed to reduce harmonic distortion levels in microgrid currents and voltages within specified limits. The adaptive filter removes harmonics from load current caused by nonlinear loads, making the current smooth and sinusoidal and reducing the total harmonic distortion according to IEEE standards. Simulation results show the adaptive filtering technique is able to reduce the total harmonic distortion to within standard levels after a fault occurs.
Driving Business Innovation: Latest Generative AI Advancements & Success StorySafe Software
Are you ready to revolutionize how you handle data? Join us for a webinar where we’ll bring you up to speed with the latest advancements in Generative AI technology and discover how leveraging FME with tools from giants like Google Gemini, Amazon, and Microsoft OpenAI can supercharge your workflow efficiency.
During the hour, we’ll take you through:
Guest Speaker Segment with Hannah Barrington: Dive into the world of dynamic real estate marketing with Hannah, the Marketing Manager at Workspace Group. Hear firsthand how their team generates engaging descriptions for thousands of office units by integrating diverse data sources—from PDF floorplans to web pages—using FME transformers, like OpenAIVisionConnector and AnthropicVisionConnector. This use case will show you how GenAI can streamline content creation for marketing across the board.
Ollama Use Case: Learn how Scenario Specialist Dmitri Bagh has utilized Ollama within FME to input data, create custom models, and enhance security protocols. This segment will include demos to illustrate the full capabilities of FME in AI-driven processes.
Custom AI Models: Discover how to leverage FME to build personalized AI models using your data. Whether it’s populating a model with local data for added security or integrating public AI tools, find out how FME facilitates a versatile and secure approach to AI.
We’ll wrap up with a live Q&A session where you can engage with our experts on your specific use cases, and learn more about optimizing your data workflows with AI.
This webinar is ideal for professionals seeking to harness the power of AI within their data management systems while ensuring high levels of customization and security. Whether you're a novice or an expert, gain actionable insights and strategies to elevate your data processes. Join us to see how FME and AI can revolutionize how you work with data!
Generating privacy-protected synthetic data using Secludy and MilvusZilliz
During this demo, the founders of Secludy will demonstrate how their system utilizes Milvus to store and manipulate embeddings for generating privacy-protected synthetic data. Their approach not only maintains the confidentiality of the original data but also enhances the utility and scalability of LLMs under privacy constraints. Attendees, including machine learning engineers, data scientists, and data managers, will witness first-hand how Secludy's integration with Milvus empowers organizations to harness the power of LLMs securely and efficiently.
IOT Based Three Phase Transmission Line Fault Detection and ClassificationIRJET Journal
This document describes a proposed IOT-based system for detecting and classifying faults on three-phase transmission lines in real-time. The system uses an Arduino board as the central controller connected to voltage transformers, an LCD display, WiFi module, and relays. When a fault occurs, the Arduino detects the type of fault (e.g. line-to-ground), calculates the exact location, and sends a signal to isolate the faulted section via relays. The information is also transmitted remotely over WiFi and displayed locally on the LCD. The system aims to minimize human effort in fault management and improve reliability of power transmission.
Automatic Fault Detection System with IOT BasedYogeshIJTSRD
The fault location is an important part for any transmission line and distribution system. The location of fault is difficult task sometimes it takes lot of times needed for the exact location of the fault. The exact fault location can help the service man to overcome the fault free system in very less time. In this paper we are able to detect the fault range in easy way using the ESP module and the message is transferred on the mobile. This project is cost effective and reliable. Fast fault detection provide the protection of equipment before any significant damage. Er. Sanjeev Kumar | Mohd Mehraj Khan | Nadeem | Shailesh Kumar Yadav | Harsh Gupta "Automatic Fault Detection System with IOT Based" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-5 | Issue-5 , August 2021, URL: https://www.ijtsrd.com/papers/ijtsrd43806.pdf Paper URL: https://www.ijtsrd.com/engineering/electrical-engineering/43806/automatic-fault-detection-system-with-iot-based/er-sanjeev-kumar
IRJET- A Research on Vibration Analysis & Optimization of Housing for ECU in ...IRJET Journal
This document discusses a research project that aims to analyze and optimize the housing for an electronic control unit (ECU) in an automobile using finite element analysis (FEA) and an FFT analyzer. The researchers first design an ECU housing model using CAD software. They then perform modal and harmonic analysis using FEA to investigate the housing's mode shapes and response at different frequencies. Experimental modal analysis is also conducted using an accelerometer, impact hammer, and FFT analyzer. The FEA results are compared to the experimental results. The objective is to minimize vertical vibrations in the ECU housing by adding stiffening ribs and optimizing the design to withstand vibrations from the engine.
Recently, research has picked up a fervent pace in the area of fault diagnosis of electrical vehicle. Like failures of a position sensor, a voltage sensor, and current sensors. Three-phase induction motors are the “workhorses” of industry and are the most widely used electrical machines. This paper presents a scheme for Fault Detection and Isolation (FDI). The proposed approach is a sensor-based technique using the mains current measurement. Current sensors are widespread in power converters control and in electrical drives. Thus, to ensure continuous operation with reconfiguration control, a fast sensor fault detection and isolation is required. In this paper, a new and fast faulty current sensor detection and isolation is presented. It is derived from intelligent techniques. The main interest of field programmable gate array is the extremely fast computation capabilities. That allows a fast residual generation when a sensor fault occurs. Using of Xilinx System Generator in Matlab / Simulink allows the real-time simulation and implemented on a field programmable gate array chip without any VHSIC Hardware Description Language coding. The sensor fault detection and isolation algorithm was implemented targeting a Virtex5. Simulation results are given to demonstrate the efficiency of this FDI approach.
Incorporation of IoT in Assembly Line Monitoring SystemIRJET Journal
This document proposes incorporating IoT into assembly line monitoring systems to make them more efficient. It suggests using master and slave electronic trackers, where slaves at each assembly station update the master at the end of the line via the internet. This allows real-time data transfer without costly wired connections. The master display would show assembly progress and statistics for supervisors to remotely monitor production. The proposed system could reduce costs for many assembly lines compared to traditional hard-wired monitoring systems.
This document summarizes a research paper that proposes using a genetic algorithm-optimized fuzzy controller to improve power quality issues like voltage sags and total harmonic distortion using a static synchronous compensator (STATCOM). The paper describes how a genetic algorithm can be used to optimize the parameters of a fuzzy logic controller for the STATCOM in order to minimize voltage sag and total harmonic distortion more effectively than conventional controllers. Simulation results demonstrate that the proposed genetic algorithm technique improves sag compensation and reduces harmonic distortion in distribution systems during fault conditions.
IRJET- Overhead Line Protection with Automatic Switch by using PLC AutomationIRJET Journal
This document describes a system to automatically protect overhead transmission lines from overload faults using a PLC-controlled air break switch. The system monitors current on the line using a sensor and can open the switch if overload is detected. This allows faulty sections to be isolated without interrupting the whole line. The automatic switch provides remote operation and is more reliable than conventional manual switches. It aims to quickly detect and resolve overload problems through automation using a PLC to control the switch based on current readings.
IRJET- A Review on Wireless Sensor System of Fault Detection of Motor ArraysIRJET Journal
This document reviews methods for fault detection in motor arrays using wireless sensor systems. It discusses induction motors, which are widely used in industry. Early fault detection is important for motor protection and reducing downtimes. The document surveys various fault detection methods, including motor current signature analysis, artificial neural networks applied to bearing fault detection using vibration data, and reflectometry techniques for detecting ground faults in photovoltaic arrays. It examines prior research applying methods such as dynamic space parity and artificial neural networks to detect faults in DC motors and induction motor bearings. Overall, the document aims to discuss health monitoring and fault detection techniques for induction motors.
IRJET- Modelling and Condition Monitoring of 3훟 Induction Motor using Fuzzy L...IRJET Journal
This document presents a MATLAB model and simulation of a 3-phase induction motor for condition monitoring using fuzzy logic. The model monitors for common faults like stator winding issues, voltage imbalances, and phase faults. Fuzzy logic membership functions and rules are used to assess the motor's health based on stator current values. Simulations show the motor operating normally and with introduced faults like turn-to-turn shorts and broken windings. In fault cases, the stator current becomes unbalanced and the fuzzy logic output indicates the motor health deteriorating from good to damaged or seriously damaged states.
Design for Testability in Timely Testing of Vlsi CircuitsIJERA Editor
Even though a circuit is designed error-free, manufactured circuits may not function correctly. Since the manufacturing process is not perfect, some defects such as short-circuits, open-circuits, open interconnections, pin shorts, etc., may be introduced. Points out that the cost of detecting a faulty component increases ten times at each step between prepackage component test and system warranty repair. It is important to identify a faulty component as early in the manufacturing process as possible. Therefore, testing has become a very important aspect of any VLSI manufacturing system.Two main issues related to test and security domain are scan-based attacks and misuse of JTAG interface. Design for testability presents effective and timely testing of VLSI circuits. The project is to test the circuits after design and then reduce the area, power, delay and security of misuse. BIST architecture is used to test the circuits effectively compared to scan based testing. In built-in self-test (BIST), on-chip circuitry is added to generate test vectors or analyze output responses or both. BIST is usually performed using pseudorandom pattern generators (PRPGs). Among the advantages of pseudorandom BIST are: (1) the low cost compared to testing from automatic test equipment (ATE). (2) The speed of the test, which is much faster than when it is applied from ATE. (3) The applicability of the test while the circuit is in the field, and (4) the potential for high quality of test.
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.
Black Box Model based Self Healing Solution for Stuck at Faults in Digital Ci...IJECEIAES
The paper proposes a design strategy to retain the true nature of the output in the event of occurrence of stuck at faults at the interconnect levels of digital circuits. The procedure endeavours to design a combinational architecture which includes attributes to identify stuck at faults present in the intermediate lines and involves a healing mechanism to redress the same. The simulated fault injection procedure introduces both single as well as multiple stuck-at faults at the interconnect levels of a two level combinational circuit in accordance with the directives of a control signal. The inherent heal facility attached to the formulation enables to reach out the fault free output even in the presence of faults. The Modelsim based simulation results obtained for the Circuit Under Test [CUT] implemented using a Read Only Memory [ROM], proclaim the ability of the system to survive itself from the influence of faults. The comparison made with the traditional Triple Modular Redundancy [TMR] exhibits the superiority of the scheme in terms of fault coverage and area overhead.
FUZZY LOGIC APPROACH FOR FAULT DIAGNOSIS OF THREE PHASE TRANSMISSION LINEJournal For Research
This document summarizes a journal article that proposes using fuzzy logic to diagnose faults on three-phase transmission lines. It begins with an abstract of the journal article, which describes using fuzzy logic as an intelligent technique to quickly and accurately identify the type of fault that occurs on a transmission system. It then provides background on transmission line faults, fault types, and challenges with transmission line protection. The document outlines the proposed fuzzy logic approach, including defining fault types as fuzzy sets and developing if-then rules to relate transmission line voltages and currents to faults. Simulation results are presented showing the fuzzy logic approach can identify different fault types based on the current responses. The conclusion is that the proposed fuzzy logic method allows for fast and reliable fault detection on transmission
IRJET-Condition Monitoring based Control using Piezo Sensor for Rotating Elec...IRJET Journal
This document discusses condition monitoring of rotating electrical motors using piezoelectric sensors. It presents a simulation model developed to detect problems in motors based on vibration analysis. If vibration exceeds unsatisfactory or unacceptable thresholds, the system will display alerts on a computer screen to indicate defective parts. Fast Fourier Transform (FFT) analysis and Motor Current Signature Analysis (MCSA) are used to diagnose faults in induction motors. The document focuses on developing this condition monitoring system to protect motors from unexpected shutdowns and increase lifetime through early problem detection without requiring human observation.
IRJET- A Simple Approach to Identify Power System Transmission Line Faults us...IRJET Journal
This paper presents a Probabilistic Neural Network (PNN) approach for identifying and classifying faults on power transmission lines. The PNN is trained on voltage waveform data simulated using Electromagnetic Transient Program (EMTP) software for different fault types and locations on a 150km transmission line. Only two sets of simulated data are used to train the PNN, requiring less computation than other methods that preprocess data. The trained PNN is able to accurately identify and classify fault types based on the voltage waveform, which helps ensure reliable power transmission by isolating only faulty lines or phases.
IRJET- Fuzzy Logic based Fault Detection in Induction Machines using CloudIRJET Journal
This document presents a system for online condition monitoring of induction motors using fuzzy logic and cloud computing. It involves two phases:
1) Developing a simulation model in MATLAB/Simulink to monitor motor parameters and detect faults.
2) Implementing an online condition monitoring system using a Raspberry Pi, sensors to measure motor vibrations, temperature etc., and a cloud platform to remotely store and access data in real-time.
Fuzzy logic is used to analyze sensor data and detect faults based on defined rules and membership functions. The system aims to monitor key parameters of induction motors and identify failures to enable safe and efficient operation in industrial applications.
Design Verification and Test Vector Minimization Using Heuristic Method of a ...ijcisjournal
The reduction in feature size increases the probability of manufacturing defect in the IC will result in a
faulty chip. A very small defect can easily result in a faulty transistor or interconnecting wire when the
feature size is less. Testing is required to guarantee fault-free products, regardless of whether the product
is a VLSI device or an electronic system. Simulation is used to verify the correctness of the design. To test n
input circuit we required 2n
test vectors. As the number inputs of a circuit are more, the exponential growth
of the required number of vectors takes much time to test the circuit. It is necessary to find testing methods
to reduce test vectors . So here designed an heuristic approach to test the ripple carry adder. Modelsim and
Xilinx tools are used to verify and synthesize the design.
Underground Cable Fault Detection Using IOTIRJET Journal
This document discusses a system to detect faults in underground cable lines using IoT. It proposes using a microprocessor, LCD display, fault sensing circuit module, LoRa module, and power supply to detect the location and type of fault (single line to ground, double line to ground, or three phase faults). The system measures voltage changes across series resistors when a short circuit occurs to determine the fault location. It can display the fault location and phase on the LCD and transmit the data over WiFi. The document reviews literature on condition monitoring of underground cables, current transformer saturation effects, and comparing optical and magnetic current transformers.
Noise analysis & qrs detection in ecg signalsHarshal Ladhe
The document discusses removing noise from ECG signals using adaptive filtering techniques. It focuses on using an LMS algorithm to remove powerline interference at 50 Hz from ECG signals. The LMS algorithm is tested with different filter tap lengths and step sizes to determine the optimal parameters for noise cancellation. Additional filtering using notch filters is also explored to remove harmonics and high frequency noise. The results show that the LMS algorithm effectively removes powerline interference from ECG signals.
IRJET- Power Quality Improvement by using Three Phase Adaptive Filter Control...IRJET Journal
This document discusses a proposed adaptive filter control system for improving power quality in a microgrid. The system contains a combination of solar PV and diesel generation connected through a voltage source converter. An adaptive filter is designed to reduce harmonic distortion levels in microgrid currents and voltages within specified limits. The adaptive filter removes harmonics from load current caused by nonlinear loads, making the current smooth and sinusoidal and reducing the total harmonic distortion according to IEEE standards. Simulation results show the adaptive filtering technique is able to reduce the total harmonic distortion to within standard levels after a fault occurs.
Driving Business Innovation: Latest Generative AI Advancements & Success StorySafe Software
Are you ready to revolutionize how you handle data? Join us for a webinar where we’ll bring you up to speed with the latest advancements in Generative AI technology and discover how leveraging FME with tools from giants like Google Gemini, Amazon, and Microsoft OpenAI can supercharge your workflow efficiency.
During the hour, we’ll take you through:
Guest Speaker Segment with Hannah Barrington: Dive into the world of dynamic real estate marketing with Hannah, the Marketing Manager at Workspace Group. Hear firsthand how their team generates engaging descriptions for thousands of office units by integrating diverse data sources—from PDF floorplans to web pages—using FME transformers, like OpenAIVisionConnector and AnthropicVisionConnector. This use case will show you how GenAI can streamline content creation for marketing across the board.
Ollama Use Case: Learn how Scenario Specialist Dmitri Bagh has utilized Ollama within FME to input data, create custom models, and enhance security protocols. This segment will include demos to illustrate the full capabilities of FME in AI-driven processes.
Custom AI Models: Discover how to leverage FME to build personalized AI models using your data. Whether it’s populating a model with local data for added security or integrating public AI tools, find out how FME facilitates a versatile and secure approach to AI.
We’ll wrap up with a live Q&A session where you can engage with our experts on your specific use cases, and learn more about optimizing your data workflows with AI.
This webinar is ideal for professionals seeking to harness the power of AI within their data management systems while ensuring high levels of customization and security. Whether you're a novice or an expert, gain actionable insights and strategies to elevate your data processes. Join us to see how FME and AI can revolutionize how you work with data!
Generating privacy-protected synthetic data using Secludy and MilvusZilliz
During this demo, the founders of Secludy will demonstrate how their system utilizes Milvus to store and manipulate embeddings for generating privacy-protected synthetic data. Their approach not only maintains the confidentiality of the original data but also enhances the utility and scalability of LLMs under privacy constraints. Attendees, including machine learning engineers, data scientists, and data managers, will witness first-hand how Secludy's integration with Milvus empowers organizations to harness the power of LLMs securely and efficiently.
GraphRAG for Life Science to increase LLM accuracyTomaz Bratanic
GraphRAG for life science domain, where you retriever information from biomedical knowledge graphs using LLMs to increase the accuracy and performance of generated answers
Taking AI to the Next Level in Manufacturing.pdfssuserfac0301
Read Taking AI to the Next Level in Manufacturing to gain insights on AI adoption in the manufacturing industry, such as:
1. How quickly AI is being implemented in manufacturing.
2. Which barriers stand in the way of AI adoption.
3. How data quality and governance form the backbone of AI.
4. Organizational processes and structures that may inhibit effective AI adoption.
6. Ideas and approaches to help build your organization's AI strategy.
Main news related to the CCS TSI 2023 (2023/1695)Jakub Marek
An English 🇬🇧 translation of a presentation to the speech I gave about the main changes brought by CCS TSI 2023 at the biggest Czech conference on Communications and signalling systems on Railways, which was held in Clarion Hotel Olomouc from 7th to 9th November 2023 (konferenceszt.cz). Attended by around 500 participants and 200 on-line followers.
The original Czech 🇨🇿 version of the presentation can be found here: https://www.slideshare.net/slideshow/hlavni-novinky-souvisejici-s-ccs-tsi-2023-2023-1695/269688092 .
The videorecording (in Czech) from the presentation is available here: https://youtu.be/WzjJWm4IyPk?si=SImb06tuXGb30BEH .
Fueling AI with Great Data with Airbyte WebinarZilliz
This talk will focus on how to collect data from a variety of sources, leveraging this data for RAG and other GenAI use cases, and finally charting your course to productionalization.
Trusted Execution Environment for Decentralized Process MiningLucaBarbaro3
Presentation of the paper "Trusted Execution Environment for Decentralized Process Mining" given during the CAiSE 2024 Conference in Cyprus on June 7, 2024.
Unlock the Future of Search with MongoDB Atlas_ Vector Search Unleashed.pdfMalak Abu Hammad
Discover how MongoDB Atlas and vector search technology can revolutionize your application's search capabilities. This comprehensive presentation covers:
* What is Vector Search?
* Importance and benefits of vector search
* Practical use cases across various industries
* Step-by-step implementation guide
* Live demos with code snippets
* Enhancing LLM capabilities with vector search
* Best practices and optimization strategies
Perfect for developers, AI enthusiasts, and tech leaders. Learn how to leverage MongoDB Atlas to deliver highly relevant, context-aware search results, transforming your data retrieval process. Stay ahead in tech innovation and maximize the potential of your applications.
#MongoDB #VectorSearch #AI #SemanticSearch #TechInnovation #DataScience #LLM #MachineLearning #SearchTechnology
A Comprehensive Guide to DeFi Development Services in 2024Intelisync
DeFi represents a paradigm shift in the financial industry. Instead of relying on traditional, centralized institutions like banks, DeFi leverages blockchain technology to create a decentralized network of financial services. This means that financial transactions can occur directly between parties, without intermediaries, using smart contracts on platforms like Ethereum.
In 2024, we are witnessing an explosion of new DeFi projects and protocols, each pushing the boundaries of what’s possible in finance.
In summary, DeFi in 2024 is not just a trend; it’s a revolution that democratizes finance, enhances security and transparency, and fosters continuous innovation. As we proceed through this presentation, we'll explore the various components and services of DeFi in detail, shedding light on how they are transforming the financial landscape.
At Intelisync, we specialize in providing comprehensive DeFi development services tailored to meet the unique needs of our clients. From smart contract development to dApp creation and security audits, we ensure that your DeFi project is built with innovation, security, and scalability in mind. Trust Intelisync to guide you through the intricate landscape of decentralized finance and unlock the full potential of blockchain technology.
Ready to take your DeFi project to the next level? Partner with Intelisync for expert DeFi development services today!
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slackshyamraj55
Discover the seamless integration of RPA (Robotic Process Automation), COMPOSER, and APM with AWS IDP enhanced with Slack notifications. Explore how these technologies converge to streamline workflows, optimize performance, and ensure secure access, all while leveraging the power of AWS IDP and real-time communication via Slack notifications.
Skybuffer AI: Advanced Conversational and Generative AI Solution on SAP Busin...Tatiana Kojar
Skybuffer AI, built on the robust SAP Business Technology Platform (SAP BTP), is the latest and most advanced version of our AI development, reaffirming our commitment to delivering top-tier AI solutions. Skybuffer AI harnesses all the innovative capabilities of the SAP BTP in the AI domain, from Conversational AI to cutting-edge Generative AI and Retrieval-Augmented Generation (RAG). It also helps SAP customers safeguard their investments into SAP Conversational AI and ensure a seamless, one-click transition to SAP Business AI.
With Skybuffer AI, various AI models can be integrated into a single communication channel such as Microsoft Teams. This integration empowers business users with insights drawn from SAP backend systems, enterprise documents, and the expansive knowledge of Generative AI. And the best part of it is that it is all managed through our intuitive no-code Action Server interface, requiring no extensive coding knowledge and making the advanced AI accessible to more users.
leewayhertz.com-AI in predictive maintenance Use cases technologies benefits ...alexjohnson7307
Predictive maintenance is a proactive approach that anticipates equipment failures before they happen. At the forefront of this innovative strategy is Artificial Intelligence (AI), which brings unprecedented precision and efficiency. AI in predictive maintenance is transforming industries by reducing downtime, minimizing costs, and enhancing productivity.
Programming Foundation Models with DSPy - Meetup SlidesZilliz
Prompting language models is hard, while programming language models is easy. In this talk, I will discuss the state-of-the-art framework DSPy for programming foundation models with its powerful optimizers and runtime constraint system.
Letter and Document Automation for Bonterra Impact Management (fka Social Sol...Jeffrey Haguewood
Sidekick Solutions uses Bonterra Impact Management (fka Social Solutions Apricot) and automation solutions to integrate data for business workflows.
We believe integration and automation are essential to user experience and the promise of efficient work through technology. Automation is the critical ingredient to realizing that full vision. We develop integration products and services for Bonterra Case Management software to support the deployment of automations for a variety of use cases.
This video focuses on automated letter generation for Bonterra Impact Management using Google Workspace or Microsoft 365.
Interested in deploying letter generation automations for Bonterra Impact Management? Contact us at sales@sidekicksolutionsllc.com to discuss next steps.
Letter and Document Automation for Bonterra Impact Management (fka Social Sol...
46 Acd 2009
1. FaultBuster: data driven fault detection
and diagnosis for industrial systems
Nicola Bergantino ∗ Fabio Caponetti ∗∗,∗,∗∗∗ Sauro Longhi ∗∗
∗
Integra Software, Jesi, Italy (e-mail:n.bergantino@integrasoftware.it)
∗∗
Universit` Politecnica delle Marche, DIIGA, Ancona, Italy (e-mail:
a
f.caponetti@diiga.univpm.it, sauro.longhi@univpm.it)
∗∗∗
Technical University of Denmark, DTU-Electrical Engineering,
Kgs.Lyngby, Denmark
AbstractEfficient and reliable monitoring systems are mandatory to assure the required security
standards in industrial complexes. This paper describes the recent developments of FaultBuster,
a purely data-driven diagnostic system. It is designed so to be easily scalable to different monitor
tasks. Multivariate statistical models based on principal components are used to detect abnormal
situations. Tailored to alarms, a probabilistic inference engine process the fault evidences to
output the most probable diagnosis. Results from the DX 09 Diagnostic Challenge shown strong
detection properties, whereas the need of further investigations in the diagnostic system.
Keywords: Statistical Process Control, Fault detection, Artificial intelligence, Diagnostic
inference
1. INTRODUCTION statistical model to detect deviant situations. Tailored to
detection is the diagnosis. Based on the knowledge learnt
from past observations and diagnosis examples it gives the
Alarms are essential in every process, system and indus-
set of the probable faults that may be occurring.
trial complex. Unplanned shut-downs are one of the most
serious sources of production loss. Even if triggered by FaultBuster has been participant of the DX 09 Diagnostic
minor problems, intervents may require the temporary Challenge Competition (http://dx-competition.org),
process stop. Detect and diagnose quickly and precisely demonstrating low rates of missed/false alarms whereas
the root cause of a developing abnormal situation is a key some problems in the diagnostic part. The competition
to keep systems working as smoothly as possible. winner, ProDiagnose used a probabilistic approach, ac-
complishing the diagnostic task with Bayesian Network
Plant monitoring represents also a way to fullfill envi-
models compiled to Arithmetic Circuits (Ricks and Meng-
ronmental regulations on emissions. Governments impose
shoel, 2009).
fines on violations of environmental protection regulations,
which erode profits and the social image. In this paper it is described the fault detection and
diagnosis algorithm implemented in FaultBuster and the
The other and probably the most important reason for
DX 09 Competition results are presented as benchmark.
plant monitoring is related to safety. According to the Ab-
normal Situation Management Consortium, petrochemical
plants on average suffer a major incident every three years,
which usually cause human casualties. These incidents oc-
cur not usually because of major design flaws or equipment 2. ADAPT TESTBED
malfunctions, but rather simple mistakes. In all these cases
a prevention system (comprising means of detection and Researchers at NASA Ames Research Centre have devel-
diagnosis, logic/control equipment and independent means oped the Advanced Diagnostics and Prognostics Testbed
of control) would probably have prevented the deploying (ADAPT). It allows performance assessment of diagnostic
of these situations. algorithms in a standardised testbed and repeatable failure
This paper presents the first developments of FaultBuster, scenarios. The hardware of the testbed is an Electrical
an industrial fault detection and diagnosis system. It is Power System (EPS) of a space exploration vehicle and
built to extract as much information as possible from the consists of three major modules: a power generation unit,
data flowing from and to the monitored plant without a power storage unit and a power distribution unit (See
embedding specific process knowledge. Fig.1). The system has hybrid dynamics where mode tran-
sitions are commanded or triggered by events. the installed
FaultBuster is composed by co-operating modules organ- sensors provide data sample only at the rate of 2Hz, which
ised in the Integra Agent Framework (IAF), an agent cannot capture the dynamics of some ADAPT subsystems
architecture discussed by Caponetti et al. (2009). Modules that operate at much higher frequencies. ADAPT has been
can be exchanged or modified independently to let the used as basis for the Diagnostic Challenge 2009 to which
system be easily scalable. FaultBuster may tune on-line a all the data used to produce the results discussed refers to.
2. Load Bank 1 LT
500
Battery Cabinet TE
TE 500
133 LGT400
ESH
170 TE
LGT401 501
TE LGT402 L1A
128 TE
EY170 502
ISH ESH ESH ISH ISH ESH
E135 E140 E161 E165 E167
136 141A 160A 162 166 171
INV ST
BAT1 FAN415 515 L1B
CB136
IT140
EY141 EY160 IT161
CB162 1 ST
CB166
IT167 XT167
EY171
ESH
165
ESH 172
TE
129 144A
FAN480 L1C
120V AC >>
EY172
ESH
EY144 173
LGT481 L1D
EY173
ESH
E142 174
FT
PMP425 525 L1E
EY174
ESH
175
TE
LGT411 511 L1F
EY175
TE ISH ESH
E181
228 180 183
ISH ESH DC482
E235
236
E240
241A L1G
CB180
EY183
BAT2 CB236 E242
IT181 ESH
184
IT240
EY241
ESH 24V DC >> L1H
TE
229 244A EY184
EY244
Load Bank 2
ESH
ADAPT-Lite 270
FT
PMP420 520 L2A
EY270
ESH ISH ISH ESH
E261 E265 E267
260A 262 266 271
INV LGT410
TE
510 L2B
EY260 IT261
CB262 2 ST
CB266
IT267 XT267
EY271
ESH
265
272
TE FAN483 L2C
328 120V AC >>
ISH ESH EY272
E335 E340 ESH
336 341A 273 LGT484
L2D
BAT3 CB336 LT
IT340
EY341 EY273 505
ESH
274 TE
TE ESH
505
329 344A
LGT405 L2E
TE
EY274 LGT406 506
ESH
EY344 275 LGT407
TE
507
Sensor Symbols ISH
EY275
ESH
E281 ST
Circuit Breaker 280 283 FAN416 516 L2F
E Voltage ISH ST Speed
Position Feedback
Relay Position CB280
EY283
L2G
ESH IT Current TE Temperature ESH
Feedback IT281
284
FT Flow LT Light XT Phase Angle 24V DC >> DC485
L2H
EY284
Figure 1. The ADAPT Tier 2 EPS. ADAPT Tier 1 (Lite) is a subset of Tier 2. (Courtesy of Tolga Kurtoglu)
3. FAULTBUSTER statistics of the data projected through the model. Once
an alarm is issued the contribution of each sensor to the
FaultBuster is engineered to fullfill the requirements of abnormal situation is determined. In this way, the vector
a fast, accurate, reliable and reconfigurable diagnostic of contribution rates represents the fault pattern that an
system. Opposite and tight constraints lead to implement inference engine based on Markov logic networks has to
a tradeoff solution. interpret. The inference output is the set of most probable
faults. Fig. 2 shows the concept scheme of the system.
The system was born to supervise tightly coupled, complex
industrial systems. Quite often poor process knowledge is 3.1 Statistical model
available, whereas huge archives of data may be accessible.
This is the case when small-medium enterprises wants FaultBuster can be bootstrapped on a pre built model
to improve their throughput and quality by monitoring or configured to fit on-line a PCA model. For industrial
already running machinery. systems, with slow dynamics, a pre-built model adapted
on-line would be the best option. Due to the high number
Statistical process monitoring techniques have been heav-
of working modes which ADAPT may present and the
ily researched in the last few years. Multivariate statistical
limited amount of available training examples available,
methods based on Principal Component Analysis (PCA),
a pre-built model resulted to be too conservative.
partial Least Squares, and Independent Component Anal-
ysis have been used and extended with success in various ADAPT works fault-less for 30s after a boot. The first
applications (Qin, 2009; Liu et al., 2009; Du et al., 2007; observations collected on-line compose a training dataset
Al Ghazzawi and Lennox, 2008; Liu et al., 2005; Bakshi, constructed in a way that columns represent the monitored
1998). variables (m) and each row an observation (n).
FaultBuster processes all the observations using a statis- X ∈ A(n×m)
tical reference model and an adaptive detection scheme. Because of the different magnitude of variables, the dataset
Detection is based on residuals built from multivariate is standardised to null mean and unit variance. Boolean
3. PCA Least
Model Squares
^ ^ ^
X S R
MEWMA Kalman Filter
X PCA Detection
Projection ~ ~ Logic Alarms
X ~
S R
MEWMA Kalman Filter
Least Back
Squares Propagation
Fault signature
Markov Logic
Networks Diagnosis
Figure 2. Concept scheme of FaultBuster.
observations from ADAPT are fed to the model by adding Two statistical distance measures are commonly computed
a Gaussian noise with small standard deviation to the true to generate residual signals, the Hotelling’s T 2 and the
value. squared prediction error (SP E)
−1/2
The dataset can be decomposed as: T 2 (x) =||Dλk P T x||2 , (4)
ˆ
X = X + E, (1) 2
SP E(x) =||˜|| = x (I − P P )x.
x T T
(5)
The matrices X ˆ and E represent the modeled and the Where
−1/2
Dλk
=
−1/2
diag(λi )
with λi=1,...,l equal to the
residual variation of X. first l eigenvalues of the correlation matrix R.
ˆ
X =T P T ,
Both statistics can be evaluated against fixed thresholds
ˆˆ
E =T P T , designed on the average run length (ARL). Due to the
where T ∈ n×d
and P ∈ m×1
are the score and loading hybrid nature of ADAPT, poor results have been obtained
matrices. utilising fixed thresholds. As done previously by Wang and
Tsung (2008), FaultBuster tries to improve the detection
The decomposition of X is such that the matrix [P P T ] performances by using a predictor on the PCA subspaces
is orthonormal and [T T T ] is orthogonal. The columns of issuing an alarm only after violation of an adaptive thresh-
P are eigenvectors of the correlation matrix R associated old.
ˆ
to the largest l eigenvalues. The columns of P are the
A multivariate exponentially weighted moving average
remaining eigenvectors of R. The correlation matrix is
(MEWMA) is used in each subspace (xp may be x or x) to
ˆ ˜
evaluated on the scaled dataset as
overcome the T 2 chart limitations (Montgomery, 2005).
1
R= XX T Zi = αxp + (1 − α)Zi−1 , (6)
n−1 i
The PCA model partitions the measurement space (m where 0 < α ≤ 1 and Z0 = 0. The control chart is
dimensional) into two orthogonal subspaces. One spanned T 2 (Zi ) = Zi Σ−1 Zi ,
T
Zi (7)
by the first l principal components, in which the normal where the covariance matrix is
data variations should occur, and a residual space where α
abnormal situations and noise fall. The interested reader ΣZi = 1 − (1 − α)2i Σ, (8)
2−α
may refer to Jackson (1991).
where Σ is a diagonal matrix containing the eigenvalues of
Once a command is issued, ADAPT is supposed to change R corresponding to the subspace considered.
working mode. Several solution may be appliable to con-
The MEWMA signals are monitored by a set of Kalman
tinue the monitoring, e.g. a new PCA model may be
filters on the four signal features in Tab. 1. To maintain
trained. To minimise the work load and supervise fast
dynamical systems like ADAPT, FaultBuster implements Table 1. Features used for residual generation.
an adaptive detection solution.
Feature Description
To detect faults in discrete dynamic, FaultBuster has to be Si ˆ ˆ ˜ ˆ
Value of T 2 (Zi ) or T 2 (Zi )
extended with a discrete observer by embedding knowledge ∆Si = Si − Si−1 Approximated first derivative
on the expected state after commands or triggers. ∆2 Si = ∆Si − ∆Si−1 Approximated second derivative
f = ∆Si ∗ ∆2 Si Relation between derivatives
3.2 Fault detection
a simple implementation each feature has its own Kalman
filter. The models are based on a linear regression updated
Each observation x is decomposed by the PCA model into:
on-line by least square minimisation each nls samples,
x = P P T x,
ˆ (2) allowing to handle eventual non linearity in the statistics.
the projection on the principal component subspace The detection of both abrupt and progressive faults de-
(PCS), and pends on a correct design of nls .
x = (I − P P T )x,
˜ (3) A 3σ control chart is used to monitor the Kalman pre-
the projection on the residual subspace. diction. The control variance which defines the Upper and
4. 7
x 10 T2(Zi)
6
Alarm
5 Normal
100 120 140 160 180 200
∆ T2(Zi)
4
Alarm
3 Normal
∆ T2(Zi)
100 120 140 160 180 200
∆2 T2(Zi)
2
Alarm
Normal
1 100 120 140 160 180 200
f(Zi)
0 Alarm
Normal
−1 100 120 140 160 180 200
100 120 140 160 180 200
Alarm combination
Figure 3. Exp 675 pb t2, ∆T 2 (Zi ) 3σ control chart. The Alarm
Kalman prediction is the solid line while the UCL
and LCL control limits are denoted as dotted lines.
Vertical dashed-dotted lines mark the faults.
Lower Control Limit (UCL, LCL) is the estimated Kalman Normal
variance. A detection is issued if all the control charts of 100 120 140 160 180 200
Time [s]
a subspace register a violation of the same control limit.
Fig. 3 shows the control chart for ∆Si in Exp 675 pb t2. Figure 4. Exp 675 pb t2 PC subspace alarm signals. Scalar
The Inverter 2 fails off and afterward the position sensor detector would be inadequate while their combination
ESH273 fails stuck closed. This experiment shows the lead to 2 true positives and a false alarm in t = 100.
ability of FaultBuster to detect faults on components not
directly observable (Inverter) and multiple faults by the The hidden node j is responsible for some fraction of
combination of the single feature detectors (Fig. 4). the error Err in each of the output nodes to which it
connects (Russell and Norvig, 2003). Thus, the values
To manage intermittent faults, after each alarm a refer- of Err are divided according to the strength of the
ence model has to be stored to establish when the fault connection between the hidden node and the output node
disappears. The detection capability of progressive faults and propagated back to provide the ∆j values for the
with slow dynamics has to be investigated, since some hidden layer. The propagation rule for the hidden nodes j
limitations expected due to the adaptivity of the detector. is,
As response to command, the detection is inhibited for ∆j = (xf lt − xref )
j j Wj,i Erri . (9)
im samples. New feature models are fitted on the data, i
avoiding the computation of a new PCA model. Where Erri is the ith component of the error vector
Err = (xTlt W −xT W ). The term (xf lt −xref ) is intended
f ref j j
3.3 Observation identification as the first derivative of the not known activation function.
The resulting vector ∆ is interpreted as the contribution
After an alarm the anomaly source have tos be identified. of each sensor to the abnormal situation.
Several approaches have been proposed to boost the iden-
tification capabilities of the commonly used contribution To filter false alarms, ∆ is pruned using a threshold.
plots (Mnassri et al., 2008). Since the detection is based Once fitted a PCA model the backpropagation is tested
on MEWMA signals and on some non-Gaussian sensors, using normal observations as examples and targets. The
FaultBuster explores and implements an alternative solu- maximum ∆j among all the sensors is stored in ∆max .
tion. It represent the maximum expected contribution rate for
a sensor in nominal conditions. At run-time, each ∆j
The PCA model is seen as a Multi-Layer Perceptron is zeroed if less than ∆max . If this results into a null
network (MLP) where: the output stage represents the vector the alarm is classified as false and removed. Fig.
PCA projection, the hidden the observations, and the 5 shows how the alarms issued by the detectors are
input stage is not used. By retropropagating the quadratic filtered, removing the false alarm characterised by zero
error between the mean of the last nm normal measures contributions and leaving the true alarms.
and the abnormal, the contribution rate of each sensor to
the abnormality is estimated. An analogous solution for the PCA has been discussed by
Chiang et al. (2001), where each component of the weight
The corresponding MLP realises the mapping Y = X T W matrix has been scaled by the corresponding variance.
where W = P if the alarm comes from the PC subspace ti f lt
ˆ
(W = P otherwise). Letting xref given by the mean of ∆j = 2 Wi,j (xj − µj )
σi
the last nm fault-less observations, and xf lt be the faulty i
observation, the quadratic error is: 2
Where ti is ist column of the score matrix T , σi the related
1 T 2 variance and µj is the off-line learnt mean value of the
E = Err2 = x W − xT W . sensor j. FaultBuster does not use the PCA model to
2 f lt ref
5. !command(c)=>!FA
!command(c)^oocss(+s,+v)=>fault(+x)
!command(c)^oocsb(+s)=>fault(+x)
The inference on the general KB leads to the probability
distribution among the fault classes in Tab. 3. Knowing
the most likely fault class, the class-specific KBs can be
interrogated (Tab. 4). For example, in ADAPT system,
the knowledge base relative to a large fan looks like:
//Variable declaration
FaultModeLargeFan = {OverSpeed,
Figure 5. Exp 675 pb t2, sensor identification. The alarm UnderSpeed, FailedOff}
for t = 100 is discarded since ∆j < ∆max ∀j. [..]
//Predicate declaration
directly estimate the contribution rate, since a new PCA FaultLargeFan(LargeFan, FaultModeLargeFan)
model is not trained after each mode switch. [..]
//Rules
3.4 Fault diagnosis !command(z)^oocss(+s,+v)=>FaultLargeFan(+x,+f)
!command(z)^oocsb(+s)=>FaultLargeFan(+x,+f)
PCA can efficiently isolate faults on the monitored vari- Table 4. Exp 675 pb t2, Markov logic inference.
ables, but as also described recently by Mnassri et al.
(2008) has limitations to diagnose problems in components Predicate Probability
non directly observable. FaultBuster tries to fill the gap by General KB
using a diagnostic module based on probabilistic reasoning Fault(BasicLoad) 0.00
and first order logic. Markov logic combines the two by Fault(Battery) 0.00
Fault(BooleanSensor) 0.00
attaching weights to first-order formulae and viewing them
Fault(CircuitBreaker) 0.00
as templates for features of Markov networks (Richardson Fault(CommandableCircuitBreaker) 0.04
and Domingos, 2006). Weights are tuned by learning and Fault(Inverter) 0.94
the implementation in FaultBuster is based on Alchemy. Fault(LargeFan) 0.00
Fault(LightBulb) 0.00
ADAPT and normal industrial systems are in general
Fault(Relay) 0.03
complex. To manage the number of components and Fault(ContinuousSensor) 0.57
failure modes, the inference is done hierarchically. Using Fault(WaterPump) 0.00
the evidences a general Knowledge Base (KB) outputs FA 0.00
the component class, i.e. pump, relay, voltage sensor. By Inverter KB
inference on specific component class KBs, the probable FaultInverter(INV1, FailedOff) 0.00
faulty component is individuated with its failure mode. FaultInverter(INV2, FailedOff ) 0.96
The predicates defined in the KBs are reported in Tab. 2. No specific knowledge about the physical system intercon-
Each predicate may make use of the variables in Tab. 3. nection has been modeled. This gives generalisation power
Them describe the facts that the KB is able to interpret. at the cost of depending on the amount and quality of
Evidences are generated from a fault pattern as an ordered examples used to train the initial KBs. The diagnostic
sequence of discrete contributions in the form of oocss and performances are expected to increase with the amount
oocb. The discrete value is obtained by quantisation of the of information modeled in the single KBs and with the
contribution space (See values in Tab.3). number of faults that occour in the monitored plant. For
industrial applications the system can be taught to classify
Table 2. Knowledge base predicates new fault patterns by the plant operators. Planned exten-
Predicate Description sion is in the way to use Bond Graph Models to generate
FA False alarm interconnection rules to boost the diagnostic performance
oocsb(boolsens) Boolean variable fault contribution by evaluating possible failure chains.
oocss(contsens,value) Continuous variable fault contribution
command(cmd) Command sent 4. RESULTS
fault(faultClass) Fault diagnosis
To demonstrate FaultBuster in action the metrics in (Kur-
The KB for the component classes is composed by simple toglu et al., 2009) has been evaluated in 233 ADAPT
first order logic rules not specific to the monitored system. scenarios and summarised in Tab.5. The data consists
with either nominal, single, double or triple fault, with
If a command has been executed recently and an alarm
various relay and circuit breaker open/close (Kurtoglu
has been issued than it likely to be a false alarm. This
et al., 2009). Each scenario starts with ADAPT unpow-
rule allows the system to move from one operating point
to another avoiding nuisance alarms. ered. Two configurations has been tested: fully adaptive,
namely FBuster, and with a bootstrap model, FBusterM .
command(+c)^oocss(+s,+v)=>FA
If no commands has been sent, the presence of a fault have The fully adaptive solution shown very low false posi-
to be investigated, hence a false alarm is unlikely to be. tive/negative rate and a fair detection accuracy. This is
6. Table 3. Knowledge base variables
Variable Description
boolsens ESH141A, ESH144A, ESH160A, ESH170, ESH171, ESH172, ESH173, ESH174, ESH175, ESH183, ESH184,
ESH241A, ESH244A, ESH260A, ESH270, ESH271, ESH272, ESH273, ESH274, ESH275, ESH283, ESH284,
ESH341A, ESH344A, ISH136, ISH162, ISH166, ISH180, ISH236, ISH262, ISH266, ISH280, ISH336
contsens E135, E140, E142, E161, E165, E167, E181, E235, E240, E242, E261, E265, E267, E281, E335, E340, FT520,
FT525, IT140, IT161, IT167, IT181, IT240, IT261, IT267, IT281, IT340, LT500, LT505, ST165, ST265, ST515,
ST516, TE128, TE129, TE133, TE228, TE229, TE328, TE329, TE500, TE501, TE502, TE505, TE506, TE507,
TE510, TE511, XT167, XT267
cmd EY136 OP, EY236 OP, EY336 OP, EY141 CL, EY144 CL, EY160 CL, EY170 CL, EY171 CL, EY172 CL,
EY173 CL, EY174 CL, EY175 CL, EY183 CL, EY184 CL, EY241 CL, EY244 CL, EY260 CL, EY270 CL,
EY271 CL, EY272 CL, EY273 CL, EY274 CL, EY275 CL, EY283 CL, EY284 CL, EY341 CL, EY344 CL
faultClass BasicLoad, Battery, BooleanSensor, CircuitBreaker, CommandableCircuitBreaker, Inverter, LargeFan, LightBulb,
Relay, ContinuousSensor, WaterPump
value Big, Medium, Small
Table 5. DX Competition results metrics Caponetti, F., Bergantino, N., and Longhi, S. (2009). Fault
tolerant software: a multi-agent system solution. In 7th
Metric FBuster FBusterM ProDiagnose
IFAC Symposium on Fault Detection, Supervision and
Detection Accuracy 74% 83.1% 88.33%
False Positives Rate 2.53% 15.56% 7.32%
Safety of Technical Processes.
False Negatives Rate 38.96% 17.53% 13.92% Chiang, L., Russell, E., and Braatz, R. (2001). Fault De-
Classification Errors 236 217 76 tection and Diagnosis in Industrial Systems. Springer.
Mean Time To Detect (ms) 14553.3 17789.9 5873 Du, Z., Jin, X., and Wu, L. (2007). Fault detection and
Mean Time To Isolate (ms) 48893.5 54104.6 11988 diagnosis based on improved pca with jaa method in vav
systems. Building and Environment, 42(9), 3221–3232.
related to its capacity to better tune a reference PCA Jackson, J.E. (1991). A User’s Guide to Principal Com-
model scaled on the actual working status. False negative ponents. Wiley.
and detection rate are influenced strongly by the inability Kurtoglu, T., Narasimhan, S., Poll, S., Garcia, D., and
to detect fault in the first 120 samples. This limitation has Wright, S. (2009). Benchmarking diagnostic algorithms
been removed in FBusterM by providing an off-line model. on an electrical power system testbed. In International
Results confirm the increase of detection rate by a lever- Conference on Prognostics and Heath Management.
age of the false and missed alarms. Isolation performance Liu, J., Lim, K.W., Srinivasan, R., and Doan, X.T. (2005).
confirm the need to introduce process knowledge to boost On-line process monitoring and fault isolation using pca.
the diagnosis. Proceedings of the 2005 IEEE International Symposium
on, Mediterrean Conference on Control and Automation
5. CONCLUSION Intelligent Control, 2005., 658–661.
Liu, X., Kruger, U., Littler, T., Xie, L., and Wang,
FautlBuster was born to diagnose complex industrial sys- S. (2009). Moving window kernel pca for adaptive
tem where limited or no process knowledge were available. monitoring of nonlinear processes. Chemometrics and
FaultBuster combined the performances of a statistical Intelligent Laboratory Systems, 96(2), 132–143.
model based detector and of a probabilistic first order Mnassri, B., El Adel, E., and Ouladsine, M. (2008). Fault
logic inference engine. The system demonstrated good localization using principal component analysis based
detection capabilities in the DX 09 Diagnostic Challenge. on a new contribution to the squared prediction error.
Both detection and diagnosis modules have been based 2008 16th Mediterranean Conference on Control and
on knowledge directly extracted from example data to Automation, 65–70.
explore the capabilities of a pure data-driven system. The Montgomery, D.C. (2005). Introduction to Statistical
detection module needs minor refinements, whereas to control. Wiley.
better diagnose, the KBs have to embed process knowledge Qin, J. (2009). Data-driven fault detection and diagnosis
or have to be trained on larger example sets. FaultBuster for complex industrial processes. In 7th IFAC Sym-
demonstrated to be computationally lightweight since the posium on Fault Detection, Supervision and Safety of
inference was executed only after confirmed alarms. Technical Processes.
Richardson, M. and Domingos, P. (2006). Markov logic
ACKNOWLEDGEMENTS networks. Machine Learning, 62, 107–136.
Ricks, B.W. and Mengshoel, O.J. (2009). The diagnostic
Giovanni Mazzuto and Fabio Virgulti are acknowledged challenge competition: Probabilistic techniques for fault
for having contributed to the solutions in FaultBuster. diagnosis in electrical power systems. In Proc. of the
20th International Workshop on Principles of Diagnosis
REFERENCES (DX-09). Stockholm, Sweden.
Al Ghazzawi, A. and Lennox, B. (2008). Monitoring a Russell, S. and Norvig, P. (2003). Artificial intelligence: a
complex refining process using multivariate statistics. modern approach. Prentice Hall.
Control Engineering Practice, 16(3), 294–307. Wang, K. and Tsung, F. (2008). An adaptive t2 chart
Bakshi, B.R. (1998). Multiscale pca with application for monitoring dynamic systems. Journal of Quality
to multivariate statistical process monitoring. AIChE Technology, 40(1), 109–123.
Journal, 44(7), 1596–1610.