仿生型算法和人工智能方法在发电厂群优洗化运行中的应用研究-水利水电工程专业毕业论文.docx

ABSTRACTTitle：0PERATlON ABSTRACT Title：0PERATlON 0F POWER PLANT SWARM OPTlMIZATION BASED ON BIONlCS AND ARTIFICIAL INTELLIGENCE METHODS Major：Water Resources and Hydro-electric Engineering Date：】坠!：辽2娑ll Name：Biao SHI Signature：臣：蚴5也●■■●●■●■■_■■■__●●●●- Supervisor：Prof．Yuxia LI Signature：丝笋 Abstract To increase the energy conversion efficiency of the electric power industry,reduce the pollution emissions，and realize power plant safe，economic and e伍cient operation under electricity market environment is urgently need to investigate the major project of Our country electric power industry．Tllis paper comprehensive investigated the key technologies such as forecasting method，mathematical model solution algorithm and optimized operation strategy as the electrical power system load forecast，market clearing price forecast the change rule，the reservoir runoff forecast and power plant(s)optimizes operation the objects． The paper were carried out the following aspects research：investigated the short-term load variation and effected factors under power market environment．proposed criteria for the load and price similarity classification，obtained main factors affecting short-term load，proposed the 7 variants of new particle swarm biomimetic algorithms．build 4 kinds of short．term load forecasting model and the intelligent prediction model associated with the neural network theory,developed variable weight forecasting model method which calculated adaptive variable．metre particle swarm algorithm，solved the problem of conventional forecasting methods low accuracy． By analyzing the market clear-out price various rules under the influence of multiple factors，the result show that electrical power system have characteristic of cyclical price changes，similarities，differences，mutation and random，that load iS a direct mutation factor affecting the short．term electricity as well as the temperature．humidity,rainfall and wind are the major factors．The paper proposed adaptive particle sw