Study On Intelligent Voltage Control System Based On MAS

Power system as one of the largest man-made, how to protect its safety and stable operation has been a theoretical problem. Optimal control of reactive power and voltage is daily operation task of power system, with the development of our smart grid and increasing complex of power grid operation mode, higher technical requirements for optimal control of reactive power and voltage are brought forward. This paper will introduce multi-Agent theory, research organizational structure and specific technical issues of the MAS-based Intelligent control of reactive power and voltage, thus increasing the intelligence of the reactive power and voltage control to meet the needs of power grid operation. This research has some theoretical significance and exigent practical significance.Specific contents are as follows:(1)The physical nature of optimization control of the reactive power and voltage is analyzed and advantages of using hierarchical classification model to achieve optimal control of reactive power and voltage are researched; Based on these, the organizational structure of intelligent voltage control system is proposed in normal operating conditions and emergency operating conditions;the realization of reactive power and voltage control is proposed especially with multi-level Agent, such as organization-level, coordination-level and implementation-level.The basic structure of Agent, multi Agent coordination and communication issues are discussed.(2) The interaction and coordination process of the sub-agent of the partition regional with other agent are given. The electrical distance for partition regional is defined, which is physical meaning more apparent, and a fuzzy clustering method used to partition the regional is given. Method of seeking regional central bus is proposed. Numerical results show that the proposed model and algorithm are reasonable.(3)The interaction and coordination process of the sub-agent of the global optimization control of reactive power and voltage with other agent are researched and given. The global reactive power optimization modeling and solving are researched, the coordination of continuous control variables and discrete control variables with action number constraint is taken into account. The discrete and continuous control variable optimization model are established respectively, crossover iteration solving for them is implemented, until the convergence is appeared, and the genetic algorithm with adaptive weight is proposed to solve optimization model.The numerical results show that the proposed model and algorithm are reasonable.(4) The interaction of the regional coordination-level agent with implementation-level agent and organization-level agent is researched. The regional reactive power and voltage optimization control model using control freedom to coordinate regional security and economic is established, the specific issues of modeling are discussed focus on the control variables, objective function and constraints.The numerical results show that the proposed model and algorithm are reasonable.(5) Aiming at the shortage of response to emergency conditions of multi-agent voltage control in normal operating conditions, organizational structure and collaborative modeling and solving of multi-agent voltage control for emergency condition are studied. Based on sensitivity information, using multi-agent contract negotiation strategy to redefine the control area to respond to emergency conditions, a virtual multi-agency alliance is established, the organizational structure of voltage control for emergency condition is formed. Based on the established organizational structure the task, division of labor and the need alternate information of each control area are confirmed for coordinated established corrective control model, the collaboration modeling and solving flow of voltage control model for emergency conditions are established. The numerical results show that the proposed idea is reasonable.