论文部分内容阅读
本文首次提出一种基于人工神经网络无功预测和优化决策相结合的变电站电压和无功综合控制策略。它是根据电压发生变化的原因和变化趋势确定综合控制策略,该策略的有效性在于预测指导,确保控制效果的最优,可大大减少变压器分接头的调节次数。仿真测试证明了预期的效果,并要投入现场运用。在该系统中还构造了控制决策神经网络模型和训练样本,用以实现多种综合因素控制,即可综合考虑电压、无功和功率因数等因素的并列控制时的相对权重,完成组合优化控制策略的灵活性。
This paper presents for the first time a comprehensive control strategy of substation voltage and reactive power based on the combination of artificial neural network (RBV) reactive power prediction and optimization decision. It is based on the voltage change causes and trends to determine a comprehensive control strategy, the effectiveness of the strategy is to predict the guidance to ensure the optimal control effect, which can greatly reduce the number of times the transformer tap adjustment. Simulation tests proved the expected results, and put into field use. In this system, a control decision neural network model and a training sample are also constructed to control a variety of comprehensive factors. The relative weights of the parallel control with voltage, reactive power and power factor can be synthetically considered to complete the combined optimal control Strategic flexibility.