混合FLN－Lagrange松驰法用于机组最优投入

A Hybrid FLN and Lagrangian Relaxation Approach to Generator Unit Commitment

本文提出用混合函数链网络与Ｌａｇｒａｎｇｅ松驰法解机组最优投入问题．基于神经网络的监督学习和自适应模式识别概念，ＦＬＮ被用来预测负荷需求与Ｌａｇｒａｎｇｅ乘子之间的关系．为了证实这一方法的有效性，一个具有１６台电机组的实际系统被测试．数值计算结果表明系统发电总成本可获得最少，大大减少了计算时间．

A hybrid method for achieving the generating unit commitment using a functional link network(FLN) is proposed in this paper. Based on the use of supervised learning neural-net technology and the adaptive pattern recognition concept,the developed FLN was used to presume the relationship between power demand pattern and Lagrange multipliers(LMPs).To demonstrate the effectiveness of the proposed approach,a real power generation system with 16 thermal units wasftested.Numerical'r.esults show that the system production cost was minimal and the time taken for processing the unit commitment scheduling in power systems was reduced.