方娜 萬暢 余俊杰

摘? 要： 針對水火電力系統(tǒng)短期發(fā)電調(diào)度決策變量多、維數(shù)高、規(guī)模大、非凸、非線性及求解困難等特點，提出改進(jìn)的粒子群算法并進(jìn)行求解。該算法采用反向?qū)W習(xí)策略提高初始解的質(zhì)量，建立參數(shù)自適應(yīng)動態(tài)調(diào)整機(jī)制控制群體進(jìn)化過程，引入混沌局部搜索增強算法局部尋優(yōu)能力。同時，根據(jù)不同類型的約束條件，采用能有效處理多重復(fù)雜約束的方法。仿真結(jié)果驗證了算法和約束處理方法的可行性和有效性，為水火電力系統(tǒng)聯(lián)合調(diào)度問題的求解提供了高效、實用的新方法。

關(guān)鍵詞： 水火電力系統(tǒng); 發(fā)電調(diào)度; 粒子群優(yōu)化算法; 反向?qū)W習(xí)策略; 約束處理; 仿真分析

中圖分類號： TN915?34; TP391.4? ? ? ? ? ? ? ? 文獻(xiàn)標(biāo)識碼： A? ? ? ? ? ? ? ? ? ? ? ?文章編號： 1004?373X（2020）06?0119?05

Short?term hydrothermal power generation optimal scheduling based on improved PSO

FANG Na1，2， WAN Chang1，2， YU Junjie1，2

（1. Hubei Key Laboratory for High?efficiency Utilization of Solar Energy and Operation Control of Energy Storage System， Hubei University of Technology， Wuhan 430068， China; 2. Hubei Power Grid Intelligent Control and Equipment Engineering Technology Research Center， Hubei University of Technology， Wuhan 430068， China）

Abstract： An improved particle swarm optimization （PSO） is proposed to overcome the difficulties of short?term hydrothermal power generation scheduling， such as more decision variables， high dimension， large scale， no convex， nonlinearity and difficult solution. In this algorithm， the backward learning strategy is used to improve quality of initial solutions， the parameter adaptive dynamic adjustment mechanism is established to control the swarm evolution process， and the chaotic local search method is introduced to enhance the algorithm's local optimization capability. At the same time， according to different types of constraint conditions， a method that can effectively deal with multiple complex constraints is adopted. The simulation results verify the feasibility and effectiveness of the proposed algorithm and the constraint handling method， which provides an efficient and practical new method for solving the joint scheduling problem of hydrothermal power system.

Keywords： hydrothermal power system; power generation scheduling; particle swarm optimization; reverse learning strategy; constraints handling; simulation analysis