摘要
抽水蓄能电站在电力系统中具有削峰填谷、调频调相等重要作用,水电站水力机械的事故多发生在其过渡过程中,为方便对过渡过程进行仿真分析,建立其调节系统数学模型并开发仿真平台具有重要的研究和应用价值.针对大型抽水蓄能机组调节系统建模过程中存在的困难,采用分段建模原理建立长引水系统模型,采用RBF神经网络建立水泵水轮机模型,并进行了空载频率扰动和增减负荷等试验分析.仿真结果表明,所建模型在一定程度上反映了实际机组特性.基于Lab Windows/CVI虚拟仪器软件,应用Active X交互技术实现仿真平台与Matlab的通信,开发了所建模型相应的仿真平台,为研究人员和现场运行维护人员进行仿真测试提供了一个友好的人机交互仿真环境.
Pumped storage power stations play an important role in peak load shifting and frequency & phase modulation in power system. Accidents in hydraulic machinery of the hydropower plant often oc- cur in the transition process. For the convenience of transition process simulation and analysis, the es- tablishment of mathematical model of regulating system and development of simulation platform are es- sential or have important value for research and application. Aiming at the difficulty of modeling of re- gulating system of large pumped storage units, piecewise modeling method is proposed for the mathe- matical model of long diversion system and RBF neural network modeling method is proposed for pump turbine respectively in this paper, and frequency disturbance transient under no-load condition and the increase or decrease of load are simulated. Simulation results indicate that the established models reflect the realistic characteristics of the plant to some extent. The simulation platform of the established models is developed based on LabWindows/CVI virtual instrument software, and the ActiveX interactive tech- nology is applied for the communication between simulation platform and Matlab, which provides an ideal environment for researchers and local operators and maintainers to carry out simulation experi-ments, as well as providing reference for the development of similar simulation platform.
出处
《排灌机械工程学报》
EI
北大核心
2014年第12期1045-1050,1056,共7页
Journal of Drainage and Irrigation Machinery Engineering
基金
国家自然科学基金资助项目(51179135)
关键词
抽水蓄能机组
长引水道
神经网络
仿真
虚拟仪器
pumped storage units
long diversion
neural network
simulation
virtual instrument