风电场有功出力和风电机组载荷协调优化方法

This project focuses on the wake effect of wind turbine and its sequel on the active power and load of wind turbine. By coordinating the operation point of each wind turbine, this project optimize the wind speed distributiorn of the wind farm, and make the active power output of wind farm tracing the reference of power grid, at the same time, minimize the load of wind turbines. Focusing on the unresolved problems of the real-time and robustness of control, this project develope a solution by integrating high performance parallel calculation, multi-agent technology and advanced control theory. The major research contents of this subject as follows.(1) Aerodynamic principles and models of active power of the wind farm and load of wind turbine and their coordination.(2)Distributed model predictive control principles of different operation conditions of wind farm. One of these operation conditions is all of the wind turbines in the wind farm in their working order. Other operation conditions include some wind turbines have to stop,or active power outputs of some wind turbines are limited at a lower value, or load peak margins of some turbines are charged.(3)Compatible mechanisms of control principles of different operation conditions based on multi-agent technology and social rational negotiation model. So as to guarantee the robustness of control.(4) Control equations fast solving mothed. The planed solving principle includes two steps. First, we transfer the origin equations to equivalent equations, which suitting for machine solving. Next, using high performance computing platform speed up equivalent equations solving. In this way, guarantee real-time of control. Especially, we develop a semi-physical experimental system based on the research results and the existed physical wind turbine test bed. More over, we analyze and verify research results on this semi-physical experimental system. These work can provide a theoretical and technical supports for grid-connected wind farm control, and promote the key common technologies in computer applications.

项目以尾流效应对风电机组有功与载荷的影响为研究背景，以协调各机组工作点为手段来优化风电场尾流分布，实现风电场有功满足电网要求，风电机组载荷最小。针对控制鲁棒性和实时性难题，探索综合运用多代理技术、高性能并行计算技术、先进控制理论的问题解决机理与方法，基本满足理论和应用需要。研究计及尾流的有功、载荷协调优化动力学原理及模型；研究风电场所有机组正常，部分机组停机、降功率或载荷容限改变等工况下有功与载荷协调的分布式模型预测控制机理，研究基于多代理技术和社会理性协商模型的风电场不同工况控制机理相容机制，保证控制的鲁棒性；研究基于极大值原理的控制方程等效变换方法、并行计算模型和任务分解策略，实现基于高性能计算平台的控制问题快速求解，保证控制的实时性。并将研究结果集成入风电机组测试系统进行分析、对比和验证。成果可为大规模并网风电场控制管理提供理论和技术支持，还可丰富和发展计算机应用领域的关键共性技术。

项目以尾流效应对风电机组有功与载荷的影响为研究背景，以协调各机组工作点为手段来优化风电场尾流分布，实现风电场有功满足电网要求，风电机组载荷最小。针对控制鲁棒性和实时性难题，探索综合运用多代理技术、高性能并行计算技术、先进控制理论的问题解决机理与方法，基本满足理论和应用需要。研究计及尾流的有功、载荷协调优化动力学原理及模型；研究风电场所有机组正常，部分机组停机、降功率或载荷容限改变等工况下有功与载荷协调的分布式模型预测控制机理，研究基于多代理技术和社会理性协商模型的风电场不同工况控制机理相容机制，保证控制的鲁棒性；研究基于极大值原理的控制方程等效变换方法、并行计算模型和任务分解策略，实现基于高性能计算平台的控制问题快速求解，保证控制的实时性。并将研究结果集成入风电机组测试系统进行分析、对比和验证。成果可为大规模并网风电场控制管理提供理论和技术支持，还可丰富和发展计算机应用领域的关键共性技术。