基于神经网络的避雷器带电状态监测

Live state monitoring of lightning arrester based on neural network

目前研究的避雷器带电状态监测方法监测过程中抗干扰能力较差,导致监测准确率难以得到保障。为了解决上述问题,利用神经网络研究了一种新的避雷器带电状态监测方法。该方法建立了神经元,将不同的神经元节点连接,利用神经元分析避雷器电流状态和电压状态,从输入层、输出层和隐藏层三个层次进行分析,选取共模电感、TVS管和共模电容,确定不同频率下的避雷器带电状态,并输出监测结果。实验结果表明,基于神经网络的避雷器带电状态监测具有很强的抗干扰能力,可以确保监测过程中监测的准确率,为避雷器维护人员的安全巡检提供了保障,具有一定可行性。

At present,the lightning arrester live state monitoring method has poor anti-interference ability in the monitoring process,which makes the monitoring accuracy difficult to be guaranteed.In order to solve the above problems,a new monitoring method of live state of arrester by using neural network is studied.The neuron is established and connected with different neuron nodes.The current state and voltage state of the arrester are analyzed by neuron.The common mode inductor,TVS tube and common mode capacitor are selected from the input layer,output layer and hidden layer to determine the live state of the arrester at different frequencies and output the monitoring results.The experimental results show that the lightning arrester live state monitoring based on neural network has a strong anti-interference ability,which can ensure the accuracy of monitoring in the monitoring process,and provide a guarantee for the safety inspection of lightning arrester maintenance personnel.