Development of Equipment and Application of Machine Learning Techniques Using Frequency Response Data for Cap Damage Detection of Porcelain Insulators

Abstract

The most common method for inspection of insulators is to measure the change of electrical characteristics such as electric resistance and partial discharge. However, even if there is no physical damage, these values vary depending on the temperature, humidity, and chloride content of the atmosphere. In this respect, an alternative to such methods can be the impact response test, and a frequency response function (FRF) obtained from the test has been widely used as a tool for damage detection. In this study the FRF was applied to identify the cap damage of porcelain insulators. In addition, to solve the danger of high voltage and poor field accessibility near the insulator, a device with high field applicability was developed to measure FRF from a long distance using an auto impact hammer and Micro Electro Mechanical Systems (MEMS) technology. Even though the FRF is most suitable for inspection of porcelain insulators, dynamic characteristics such as natural frequencies may vary depending on manufacturing errors, installation conditions, etc., which may cause difficulties in damage identification. To overcome this limitation, the machine learning (ML) method was applied in this study to provide a diagnostic method that ensured consistent and accurate judgment. As a result of predicting the normal and the cap damage data using the support vector machine (SVM), bagging, k-nearest neighbor (kNN), and discriminant analysis (DA) methods, the overall F1 score was over 87% and the bagging method achieved the highest accuracy. In this study, the frequency range and dynamic characteristics that are sensitive to the physical damage of the insulator were derived and, based on this, the optimum ML methods with improved equipment could provide analysis with higher accuracy and consistency than general analysis using the FRF.

Determination of Dielectric Losses in a Power Transformer

“…The most common places of partial discharges are paper insulation, electrical insulating liquid, culvert surface, and metal elements of the transformer that are under voltage. Such places may be characterized by insufficient electrical strength, which is the result of the insulation aging processes, or too-high values of electric field strength, which is the result of the occurrence of sharp edges of metal parts [24][25][26][27][28]. Losses caused by partial discharges are very small and their value usually does not exceed 1 W. The value of these losses depends on many factors, such as transformer voltage, temperature, degree of contamination of the surface of bushing insulators, degree of aging of insulation materials, and level of moisture in both paper insulation and insulating liquid.…”

Detection and Localization of Partial Discharge in Connectors of Air Power Lines by Means of Ultrasonic Measurements and Artificial Intelligence Models

“…Li et al [25] used two back propagation neural networks for partial discharge recognition in gas insulated switchgear. Choi et al [26] tested various ML methods like bagging, k-nearest neighbor, support vector machines and linear discriminant analysis to detect cap damage of porcelain insulators using frequency response functions. An increasing number of researches that use deep learning (DL) neural networks for PD classification can be noticed recently.…”

Damage Detection Method of Simply Supported Beams Based on Difference of Nearby Difference Curvature Index

“…The results of numerical example validation indicate that this method has good damage identification performance. Keywords: Building Structure; Continuous Beam; Damage; Deflection; Curvature 1 引言 两跨连续梁结构在实际工程中有着广泛的应用， 例如连续梁桥、架空管道等。对于服役较久的两跨连 续梁结构，需及时发现损伤信息以确保其安全运行。 两跨连续梁上可能出现的损伤情形有多种，但在理论 上一般可以局部刚度的变化来代表损伤 [1][2][3]。在进行 损伤识别时，首先选定某一结构性能指标，然后根据 其在损伤前后的变化来识别损伤信息(刚度变化)。 目前常用的结构性能指标包括静力性能指标与动力性 能指标两大类：静力性能指标包括位移 [4][5][6]、挠度曲 率 [7]、 应变 [8]等， 动力性能指标包括频率 [9]、 振型 [10]、 曲率模态 [11]、应变模态 [12]等。 其中，挠度曲率是值得重点关注的指标之一，尤 其是根据静力挠度数据和中心差分法得到的挠度差分 曲率，可以反映出结构上由损伤出现造成的局部刚度 变化，从而实现损伤识别。同时，挠度差分曲率指标 值的获取过程简单，利用光纤光栅等先进传感器能达 到很高的数据精度。 对于梁结构，基于挠度差分曲率的损伤识别方法 首先应用于简支梁这类静定梁 [7,[13][14][15]…”