神经网络智能算法在发电机主绝缘状态评估领域的应用

Application of Neural Network Intelligent Algorithm in Assessment of Generator Groundwall Insulation

随着人工智能技术的不断发展,传统的电力装备制造业迎来了巨大的再升级空间。电力装备的智能制造、智能检测及智能控制等方面已经成为我国工业革新向多维度拓展的关键环节。同时智能电网的快速发展也为电力设备智能化提供了新的机遇。本文介绍了神经网络作为人工智能技术的核心算法在电力装备领域的应用及其在国内外现状和前景,并给出了神经网络的基本学习算法和拓扑结构,着重分析了不同类型神经网络的数学模型和特点。最后通过介绍径向基神经网络,给出了该算法在发电机主绝缘检测、诊断及大时间尺度下在神经网络寿命评估中的应用。

Traditional power equipment manufacturing is facing a broad space for re-upgrading with the continuous development of artificial intelligence technology.The intelligent manufacturing,intelligent detection and intelligent control and so on of power equipment have become the key link for the expansion of industrial innovation in China.At the same time,the internalization of power equipment is provided new opportunities with the rapid development of smart grids.In this paper,the development potential of neural network as the core algorithm of artificial intelligence technology in the field of power equipment are introduced in detail.The basic learning algorithm and topology of neural network are given.The mathematical models of different types of neural networks are analyzed.And the application of radial basis function neural network in the detection,diagnosis and lifetime assessment of generator stator bars is presented.