基于BP神经网络算法的新一代智能变电站控制障碍分析与定位技术研究

Research on control obstacle analysis and location technology of new generation intelligent substation based on BP neural network

变电站远方操作过程中,控制信号从最初发出开始,在中间诸多环节中,任意因素未满足条件,均会导致控制过程失败,控制障碍发生时,中间环节隐性不可视,缺乏有效的技术手段快速定位。全过程控制信息模型将复杂的控制过程分解为具体的模型,并将模型中的各类因子进行了分类,确定了控制障碍分析的基本原理,为实现障碍定位提供了明确的方法。结合新型电力系统智能变电站控制业务特点,分析104控制报文、MMS控制报文、GOOSE控制报文及相关控制设备、控制过程特点,提出控制障碍分析与定位技术,建立全过程控制分析模型,应用BP神经网络算法训练模型,通过实例验证所提算法可快速定位控制障碍原因,解决了控制障碍发生时,缺乏有效的技术手段快速定位的问题。

In the process of remote operation of the substation,the control signal is sent from the beginning.In many intermediate links,any factor that fails to meet the conditions will lead to the failure of the control process.When the control obstacles occur,the intermediate links are invisible and lack of effective technical means for rapid positioning.The whole process control information model decomposes the complex control process into specific models,classifies various factors in the model,determines the basic principle of control obstacle analysis,and provides a clear method for obstacle positioning.Combined with the control business characteristics of the intelligent substation of the new power system,the characteristics of 104 control message,MMS control message,goose control message and relevant control equipment and control process are analyzed,the control obstacle analysis and positioning technology is proposed,the whole process control analysis model is established,and the BP neural network algorithm is applied to train the model.The effectiveness of the proposed method is verified by an example,and the algorithm can quickly locate the cause of the control obstacle,it solves the problem of lack of effective technical means for rapid positioning when control obstacles occur.