Electric Transmission Line Insulator Flashover Predictor System

9523267卡拉迪绝缘子污染案例在沿海和工业区附近频繁闪络。这个问题发生在输电线路和变电站中。目前，还没有可靠的方法来监测污染积累和预测即将发生的闪络。该项目的目标是开发一种传感器系统，该系统可以监控污染的积累，并在存在闪络危险时向操作员发出警报。在亚利桑那州立大学进行了初步的实验室研究，并使用快速傅立叶变换(FFT)分析了放电电流特征。视觉表现与当前签名同时被记录下来。研究发现，表面放电具有明显的发展阶段。虽然电流的幅值不断变化，但谐波含量(信号)始终表明放电状态。观察到，接近闪络改变了谐波含量。这些发现表明，可以通过对泄漏电流的特征分析来预测闪络。其概念是在每个区域选择一个裸露的绝缘子，并监测通过绝缘子的泄漏电流。泄漏电流将被数字化，并将执行特征分析。签名的变化表示正在接近闪络。首先，将在实验室测试不同的特征分析方法，以确定哪种方法能产生可靠的闪络指示。在该方法被识别和测试后，将构建一个传感器系统来持续执行该特征分析。预计该系统将包括一个目前的宽带(20 Hz-100 kHz)传感器、一个微处理器和一个发射机。泄漏电流将持续数字化，并将50-100个循环放入临时存储。将分析存储的数据，如果未发生签名更改，则丢弃该数据。签名更改将触发发送器，从而向操作员发出警报。将使用不同级别的ALA RM。***

9523267 Karady Insulator contamination cases frequent flashovers near coastal and industrial areas. This problem occurs in transmission lines as well as in substations. Currently, there are no reliable methods available for monitoring pollution build-up and predicting imminent flashover. The aim of this project is to develop a sensor system that monitors pollution build-up and that would alert operators when there is danger of flashover. Preliminary laboratory studies were performed at ASU and discharge current signatures were analyzed using the Fast Fourier Transform (FFT). Visual manifestations were recorded simultaneously with the current signature. It was found that surface discharges have distinct stages of development. Although the amplitude of current changes continuously, the harmonic content (signature) consistently indicated the status of discharge. It was observed that approaching flashover modifies the harmonic content. These findings suggest that flashover can be predicted by signature analysis of the leakage current. The concept is to select an exposed insulator in each area and monitor the leakage current through the insulator. The leakage current will be digitized and signature analysis will be performed. A change in signature indicates approaching flashover. First, different signature analysis methods will be tested in a laboratory to determine which one produces reliable flashover indication. After the method is identified and tested, a sensor system will be built to perform this signature analysis continuously. It is foreseen that the system will include a broad-band (20Hz-100kHz) currently sensor, a microprocessor and a transmitter. Leakage current will be continuously digitized and 50-100 cycles will be put in temporary storage. The stored data will be analyzed and if no signature change occurs, the data will be discarded. A signature change will trigger a transmitter that will alert an operator. Different levels of ala rm will be used. ***