Research On Multi-agent Simulation Platform For AGC Based On JADE

With the development of grid-scale, the current grid has become a complexinterconnected system contains several control areas. An important indicator of thequality of grid power supply is the frequency. Qualified grid frequency is necessaryfor safe and stable operation. The main means of frequency control is the use ofautomatic generation control. Therefore, the effect of automatic generation control iscrucial to power quality. Traditional control algorithm based on fixed gain controlleris difficult to meet the increasingly complex power system control performancerequirements because of the power system operating point changing with the day,month, quarter, and year. System frequency and tie-line power exchange would not bestable operation when a region’s controller is not zero its control error in a certainperiod. This is because the intelligent algorithm has better adaptability than thetraditional control algorithms. Therefore intelligent control algorithms is introduced tosolve the AGC controller design in order to solve the multi-regional interconnectedsystem collaboration control problem. However, at home and abroad, there is nounified standard of AGC intelligent simulation platform to realize interconnectioncontrol area coordination control and assess the effect of intelligence controlalgorithm more effective.In order to deal with automatic generation control (AGC) under strong randomgrid environment considering emerging renewable energy sources, a comprehensivenew multi-agent simulation platform based on java agent development toolkit (JADE)was proposed. The main contents include the following:（1）This paper presents the multi-agent systems, and multi-agent systemsapplications in power systems, and automatic generation control are outlined.（2）In the third chapter, this paper introduces the basic concepts of multi-agentsimulation platform of JADE, and the agent’s behavior and communication in JADEhas carried on the detailed introduction.（3）In the fourth chapter, this paper presents a multi-agent simulation platformbased on JADE. This platform consists of three main parts: detailed and overall process dynamic load frequency response model, real time data transporting moduleand control algorithm module.（4）In the fifth chapter, this paper analyses the simulation results and the controleffect of the platform is good. This platform provides more fair and reliableevaluation of future research in control coordination algorithm of complexinterconnected multi-regional. It provides a research basis for the algorithm designand improving the algorithm performance. Meanwhile, the establishment of theplatform also provides a practical application towards the AGC control means infuture complex multi-regional interconnected system with strong random.