A cross-layer approach to energy management in manufacturing
•
0 likes•179 views
This document describes a proposed cross-layer approach to energy management in manufacturing ecosystems. The approach involves developing an Ecosystem Oriented Energy Management architecture to holistically evaluate and optimize energy consumption across interconnected factories, buildings, equipment and utilities. The architecture was implemented in an industrial testbed containing a manufacturing line with 10 cells, robots, conveyors and sensors. Production and building data is exposed via web services and analyzed using complex event processing, rules engines and visualization tools to provide energy reports and key performance indicators. The approach aims to overcome traditional fragmentation between factory and building automation for more efficient enterprise-wide energy management.
Recommended
Recommended
More Related Content
What's hot
What's hot (7)
Viewers also liked
Viewers also liked (20)
Similar to A cross-layer approach to energy management in manufacturing
Similar to A cross-layer approach to energy management in manufacturing (20)
Recently uploaded
Recently uploaded (20)
A cross-layer approach to energy management in manufacturing
- 1. A cross-layer approach to energy management in manufacturing •Date: October, 2012 •Linked to: RTD at FAST Contact information Tampere University of Technology, FAST Laboratory, P.O. Box 600, FIN-33101 Tampere, Finland Email: fast@tut.fi www.tut.fi/fast Conference: 2012 IEEE International Conference on Systems, Man, and Cybernetics Title of the paper: A cross-layer approach to energy management in manufacturing Authors: Anna Florea Dr. Corina Postelnicu Bin Zhang Prof. Jose L. Martinez Lastra, Dr.Sc. If you would like to receive a reprint of the original paper, please contact us
- 2. Ecosystem Oriented Energy Management: An Implementation Anna Florea Dr. Corina Postelnicu Bin Zhang Prof. Jose L. Martinez Lastra, Dr.Sc. Tampere University of Technology, Finland
- 3. Outline Background and motivation. Proposed Architecture Ecosystem Oriented Architecture for Energy Management Implementation Summary
- 4. Background. Ecosystem Dimensions • Geographical distribution • Dynamic supply chain • Enterprise layers • Manufacturing equipment Interaction patterns • Factories • Devices • Collaboration • Competition long-term strategic efficiency objectives short-term optimisation Businessrelevanttargets Reports
- 5. 6.6.2013 5 Background. Implications for energy management Saving opportunities: Dynamic Utility contracts Smarter Building Automation Services: Holistic evaluation of consumption Status of production process as input control variable High quality production Equipment Utilities BuildingsPersonnel Logistics Contributors to energy consumption
- 6. Architecture proposed Functions: • Date retrieval and processing • Reasoning • Reporting • Visualization Functions: • Date retrieval and processing • Reasoning • Reporting • Visualization
- 7. Implementation Industrial testbed • 10 manufacturing cells • robot (except the buffer cell) and a • conveyor system • 729 different layouts of mobile phones • Frame • Keyboard • Screen types • Production line information exposed as web services. • Data concerning the building environment is captured via wireless sensors and routed to a server through edge routers.
- 8. Implementation 6.6.2013 8 • WS • Data acquisition • Reporting • KPI • CEP • Esper Rule Engine • Visualization • web browser • Reporting • Standard web technologies
- 9. Summary • Factories and buildings are interdependent and should be approached as a whole, to efficiently manage associated energy consumption. • Ecosystem Oriented Energy Management Systems are needed to overcome the fragmentation traditionally residing in factory and building automation. • Future research will focus on optimization methods that take into consideration all options for energy saving and transfer.
- 10. Thank you! FAST @ YouTube FAST @ Facebook FAST @ Slideshare
Editor's Notes
- Manufacturing enterprise together with its interfaces to the world became nowadays a complex ecosystems.It is composed of various geographically distributed component [production sites, offices, RTD units] connected via logisticsThe connections of the enterprise in the supply chain network have to be dynamically maintained to address changes in the network structure and be able to chose best optionsDifferent layers of the enterprise are supposed to work in synergy, yet objectives and time scale vary from layer to layer.Legacy equipment exists side by side with advanced toolsNumerous interactions occur among factories and devices, which sometimes are part of collaboration towards a common goal and other time is competition in resource constrained environment.
- In context of energy performance, it can be seen that overall consumption is influenced by buildings, equipment, logistic operations and personnel.It becomes crucial to obtain holistic view on the enterprise in order to assure energy efficient performance of the enterprise.Without ecosystem oriented view on the problem, some saving opportunities may remain hiddenRerouting the solar energy acquired by the building to the MES system because e.g. Energy prices in a certain time range are high and the energy required to heat/illuminate the building need not be so much because humans do not reside in the area at that time slot of the dayWe propose an integrated energy management architecture to fulfill this taskHaving in mind SOA related developments, researchers at Loughborough [9] have highlighted two main directions to explore further for a new generation of Digital Business Ecosystems for SMEs: visualization & reconfiguration, and integration of Manufacturing Execution Systems with the higher and lower levels (so that businesses would be able to interrogate on production data and influence production metrics). Aalst and Dustdar [16] highlight the need to take into consideration the situations of concept drift, i.e. when the process is changing while being analyzed (e.g. changing energy prices in the described scenarios).
- Frame 1:Four functionsTheir implementationRE for cross-layer integrationFrame 2:Integrated view: four functions at ERP layerDAE collecting relevant data from BA and FA systems, computation of Key Performance Indicators (KPI) and storing this data in the corresponding (BAS or FAS) database. DSS fwd process chaining evaluates the acquired data and calculated KPI values with respect to the objectives defined in the Knowledge Base, stored in dedicated database, and suggests actions for performance improvement. SYNC check is done whether DSS output may propagate down to BAS and FAS. eEMS Report Engine to build and provide consolidated reports after accessing the database to obtain the required information. The RE functionality can be exposed as web service to the users.Reasoning in eEMS is to be implemented by means of a CEP engine, having the database as input and internal Knowledge Base, stored in dedicated database of CEP engine, for reasoning. The CEP engine analyzes energy consumption patterns, passes the refined information to other ERP systems, and updates the Knowledge Base of mEMS if needed.
- Production line located at the premises of the FAST Lab, Tampere University of Technology, FinlandProduction line information (e.g. status of robots and products, quality inspection results, energy consumption of conveyors, robots, and controllers) is exposed as web services. Data concerning the building environment (temperature, relative humidity and ambient light) is captured via wireless sensors and routed to a server through edge routers.
- The architecture was implemented as a set of web apps deployed on Apache TomcatTwo WS endpoints are present in the system: one serves as interface for FAS data sources and other provides access to reporting engine for third party apps.Efficiency KPIVisualized in browser
- Approaches to modelling and knowledge representation are out of the scope of present paper.