Research On Distributed Predictive Control For Multi-area Power System

With the rapid development of economy and electric power,multi-area interconnection and networking has become a new trend of power grid development.On the one hand,multi-region interconnection power system has the advantages of improving the reliability of power supply and making full use of resources,on the other hand,due to the state coupling between each subsystem of multi-region system,which increases the complexity of the whole system,this problem also puts forward higher requirements for automatic generation control.With the continuous improvement of communication technology,the distributed control strategy can effectively coordinate the overall control performance of power system.The influence of the inevitable network induced delay and packet loss on the power system adopting distributed control strategy cannot be ignored,which poses a new challenge to the automatic generation control of multi-area interconnected power system.In this paper,the automatic generation control strategy of multi-area interconnected power system is studied.1.According to the mechanism of each regional link,a single regional model is established for the multi-regional interconnection power network consisting of thermal power units,hydropower units and photovoltaic units.Secondly,considering the uncertainty of photovoltaic power generation,the output power of photovoltaic power set is taken as part of the thermal power unit output compensation,and further unified modeling of photovoltaic power set and thermal power unit is carried out.Finally,based on the power system tie line model,a unified model of multi-area interconnected power system with photovoltaic units is established.2.Aiming at the subsystem coupling problem of multi-area power system,the load frequency control problem based on distributed predictive control strategy is studied.The distributed control strategy is used to decompose the complex and centralized optimization problem into regional controllers,and the influence of the nonlinear constraint of generator change rate is considered.The simulation results show the effectiveness and feasibility of the algorithm through the simulation of four-region interconnection power system model.3.In order to reduce the influence of network delay,packet loss and other problems on the application of distributed control strategy,an event-driven distributed control method was designed on the basis of distributed predictive control strategy.By using the event-driven control rule given by the system frequency deviation state,the optimal performance function of distributed predictive control with the threshold of event-driven control strategy was designed,and an event-driven distributed control algorithm was proposed.Compared with distributed algorithm and distributed control algorithm,simulation analysis proves that this algorithm can save communication resources and reduce the possibility of network delay and packet loss under the premise of ensuring system control performance.