基于CPS统计信息自学习机理的AGC自适应控制

CPS Statistic Information Self-learning Methodology Based Adaptive Automatic Generation Control

提出一种利用控制性能标准(control performance standard,CPS)的统计信息进行自动学习,从而为调度端自动发电控制(automatic generation control,AGC)的比例积分控制参数进行自动调整的自适应控制策略。CPS标准的提出一方面是为提高区域电网对整个电网频率的支援作用,另一方面其统计特性放宽对区域控制偏差控制的要求。该自适应控制策略充分利用CPS的统计特性,对电网结构、参数和运行方式具有良好的适应性,其自学习机制简单实用,易于在现有大多数电网调度AGC系统上实现。以广东电网为对象的实例研究显示,该AGC自适应控制可在保证CPS考核合格率的同时有效减轻发电厂的调节压力。

The paper proposed a novel adaptive control method for power automatic generation control （AGC） systems by using the statistic control performance standard （CPS） information. The purposes of CPS are to enhance the frequency dispatch support effect from power control areas and relax the requirement of area control error （ACE） regulation. The proposed method takes advantage of CPS statistic information to create a group of self-learning rules, which can automatically tune proportional integral （PI） control parameters of AGC system. The adaptive control method is demonstrated effective and easy to be implemented in AGC systems of most power dispatch centers. The case study shows that the pressure of AG-C plants is released while the CPS compliance is ensured after using the proposed control method.