Adaptive Identification Algorithms
Design and Investigation of Adaptive Identification Algorithms for Closed-Loop Process Control Systems
Tech Area / Field
- MAN-MPS/Manufacturing, Planning, Processing and Control/Manufacturing Technology
- INF-SOF/Software/Information and Communications
- MAN-COM/CAD and CAM/Manufacturing Technology
- MAN-PDM/Plant Design and Maintenance/Manufacturing Technology
3 Approved without Funding
Research Institute of Industrial Automation, Georgia, Rustavi
- University of Southern California, USA, CA, Los-Angeles\nDuke University, USA, NH, Durham
Project summaryTo enhance the quality of production in many industries (metallurgy, chemistry, petrochemistry, etc.), the new control techniques have to be used.
Nowadays, wide applications for complex processes control obtain Adaptive Control Systems with Identifier in Closed-Loop (ASI). ASI posses a number of advantages giving them high efficiency and competitive ability in control systems market.
The efficiency of adaptive systems is to a great extent determined by identification algorithms. Now the large efforts are spent for developing the new identification techniques in USA, Europe, Japan. This work is done also in CIS countries.
In proposed project, different identification algorithms for control and disturbance channels will be used.
Currently, several ASI are successfully used for relatively long period, e.g., for control of tube rolling mills in metallurgy. However, their capabilities were constrained by existing hardware properties. The fast growth of modern hardware and system software possibilities gives the opportunity to implement more sophisticated and therefore more efficient control algorithms.
The aim of the project is to develop adaptive identification algorithms with improved characteristics, first of all:
- Higher forecasting accuracy both for immeasurable and controllable variables, what will result in production quality improving (e.g., decreasing of output variance), energy and materials saving.
- Wide opportunities for algorithms applications on industrial IBM-compatible computers so ensuring that respective ASI are flexible, easy to install, handle, upgrade, modify, etc.
The principal problem is to develop identification algorithms for closed-loop systems, as, firstly, most of existing open-loop algorithms may become unstable and loose control quality in closed system, and, secondly, large number of processes can be identified only in closed loop.
Moreover, it is planned to analyze the applicability limits for improved algorithms using catastrophe theory, fractal techniques, etc. The question of forbidden control zones forecasting will be investigated, giving the means to diminish defective articles production.
As majority of real plants are nonlinear, the suggested project concentrates on new ideas and techniques use in identification algorithms to make them applicable for nonlinear systems and so enhance the opportunities of implementation.
The main objectives of project are as follows:
- Design of modified adaptive identification algorithms for time-variable linear and nonlinear plants.
- Determination of non-linearity classes, for which designed algorithms are most effective.
- Computer simulation and experimental testing of algorithms.
- Recommendations for application of algorithms to different linear and nonlinear plants.
- Applications coding for industrial ASI implementation.
The algorithms to be designed will be able to ensure the basic ASI properties for complex processes control.
As the project result, based on system identification, catastrophe theory and relative disciplines, identification algorithms, easy to implement in industrial ASI will be developed. So the research under suggested project will have essential impact on industrial ASI and other process control systems implementation.
The principal participants of proposed project are highly experienced specialists in control systems theory and applications. For a long time they are actively working in research and application in system identification, adaptive control, catastrophe theory, reliability theory.
The project leader and members have designed, manufactured and implemented ASI for process control on a number of plants former Soviet Union states.
The suggested project corresponds the ISTC targets, as in it scientists and specialists previously experienced in weapon technologies are engaged, and the results will be used for peaceful and commercial purposes.
The collaborators from USA will review and comment project technical reports, consult the project members in the field of choice of simulation methods, hardware and system software for computer simulation, assist in evaluation of project accomplishment.
The International Science and Technology Center (ISTC) is an intergovernmental organization connecting scientists from Kazakhstan, Armenia, Tajikistan, Kyrgyzstan, and Georgia with their peers and research organizations in the EU, Japan, Republic of Korea, Norway and the United States.
ISTC facilitates international science projects and assists the global scientific and business community to source and engage with CIS and Georgian institutes that develop or possess an excellence of scientific know-how.