A multi-agent based control scheme for accelerator pre-injector and transport line for enhancement of accelerator operations

A B ST R A C T Reliable accelerator operation requires control system with higher level of automation, flexibility, robustness, and optimisation. In this paper a multi-agent system based control scheme is presented for optimal control of accelerator system that improves the plant performance in wide-range of operations. The multi-agent based control schemes for accelerators have been reported in literature. But the scheme proposed in this paper differs significantly form the existing schemes. In this work the agent architecture is formulated based on the control requirements of pre-injector accelerator subsystem (Microtron in particular) and transport line of synchrotron radiation sources. The scheme consists of two software agents at supervisory level that work in an autonomous manner for the optimized control of dynamic system. The Microtron agent architecture augments model assisted adaptive controller for realizing feedback control action at lower layer and goal based logic controller with pre-structure model identifier along with the pattern recognizer at supervisory layer. The TL-1 agent has a model-based, goal-based modular architecture and optimizes the TL-1 control using differential evolution based algorithm. The simulation results of applying this scheme to model of Microtron and Transport Line-1 of INDUS complex shows that this approach is very effective in optimizing the Microtron and TL-1 tuning.

[1]  Nicholas R. Jennings,et al.  Using ARCHONTM to develop real-world DAI applications for electricity transportation management and particle accelerator control , 2007 .

[2]  George F. Luger,et al.  An Intelligent Control Architecture for Accelerator Beamline Tuning , 1997, AAAI/IAAI.

[3]  Sukumar Kamalasadan.  A Novel Multi-Agent Controller for Dynamic Systems based on Supervisory Loop Approach , 2007, Eng. Lett..

[4]  Sukumar Kamalasadan.  An Organizational Coordinated Control Paradigm for Complex Systems based on Intelligent Agent Supervisory Loops , 2007, 2007 International Joint Conference on Neural Networks.

[5]  R. Svensson.  Applying Intelligent System Concepts to Automatic Beamline Alignment , 1997 .

[6]  F Christoph Iselin,et al.  The MAD program (methodical accelerator design) : version 8.10 ; user's reference manual , 1993 .

[7]  Sheng Gehao,et al.  Optimal Coordination For Multi-Agent Based Secondary Voltage Control In Power System , 2005, 2005 IEEE/PES Transmission & Distribution Conference & Exposition: Asia and Pacific.

[8]  George F. Luger,et al.  Framework for a general purpose, intelligent control system for particle accelerators , 1995, Proceedings Particle Accelerator Conference.

[9]  George F. Luger,et al.  Developing a general purpose intelligent control system for particle accelerators , 1999, J. Intell. Fuzzy Syst..

[10]  K.Y. Lee,et al.  A multi-agent system-based intelligent identification system for power plant control and fault-diagnosis , 2006, 2006 IEEE Power Engineering Society General Meeting.

[11]  Carl R. Stern,et al.  Designing a portable architecture for intelligent particle accelerator control , 1997, Proceedings of the 1997 Particle Accelerator Conference (Cat. No.97CH36167).

[12]  Kwang Y. Lee,et al.  A Multi-Agent System-Based Intelligent Heuristic Optimal Control System for A Large-Scale Power Plant , 2006, 2006 IEEE International Conference on Evolutionary Computation.

[13]  C. Stern,et al.  A Control System for Accelerator Tuning Combining Adaptive Plan Execution with Online Learning , 2022 .

[14]  B.N. Hichem,et al.  A multi-agent predictive control approach based on fuzzy supervisory loop for fast dynamic systems , 2008, 2008 5th International Multi-Conference on Systems, Signals and Devices.

[15]  J. Niederer,et al.  The MAD program , 1983, Proceedings of the 1989 IEEE Particle Accelerator Conference, . 'Accelerator Science and Technology.

[16]  Kwang Y. Lee,et al.  Implementation of a multi-agent system for optimized multiobjective power plant control , 2010, North American Power Symposium 2010.

[17]  George F. Luger,et al.  Teleo-reactive Control for Accelerator Beamline Tuning , 2000, Artificial Intelligence and Soft Computing.

[18]  Yinong Zhang,et al.  Cement rotary kiln control system realized by PROFIBUS based on multi-agent , 2011, 2011 Eighth International Conference on Fuzzy Systems and Knowledge Discovery (FSKD).

[19]  Kenwood H. Hall,et al.  Distributed multi-agent architecture for automation systems , 2004, Expert Syst. Appl..