Comparative Study Of FA And PSO Algorithms And Application In Coordination Optimization

The supercritical unit has become the main unit of China's power grid because of its energy-saving and high efficiency.Its main task is to meet the load changes,participate in peak regulation and frequency modulation of the power grid by participating in automatic generation control(AGC).Because the supercritical unit is a non-linear multi-input and multi-output strong-coupling controlled object,the traditional PID control is often not easy to meet requirements under deep peak load regulation conditions,and the control effect of the unit in a wide range of variable operating conditions is getting worse.So the introduction of intelligent control strategy to improve the coordinated control quality is quite necessary.In this paper,we mainly study the firefly algorithms(FA),a new kind of intelligent population optimization algorithm,by comparing its performance with mature particle swarm algorithm.Then an improved FA is applied in predictive optimal control for the coordinated system of a supercritical power unit,which is of great significance both in theory and application.Aiming at a 600 MW supercritical unit,this paper analyzes in detail the characteristics of its coordination system and the control mode and control logic.On this basis,the principle of neural network and the modeling method of nonlinear system are studied.A model predictive optimal control(MPOC)method based on BP neural network modeling and chaos scaled firefly optimization algorithm(CSFA)is proposed and applied to supercritical unit coordinated control.The proposed MPOC scheme is programmed with MATLAB software and tested by extensive control simulation experiments in the full-scope simulator of a 600 MW supercritical power generating unit.The simulation results show that the proposed MPOC method can greatly improve the load dynamic response speed of the supercritical power unit,and at the same time keep other key parameters,such as main steam pressure within safety limits.