Prediction the Thermal of Compression Dead-end and Jumper Terminal for Electrical Equipment Based on High Voltage Transmission Lines, Case Study of Electricity Du Cambodia (EDC)

Abstract

The high voltage overhead transmission line system is one of the methods of delivering electric power at a high voltage of Electricity Du Cambodia (EDC) is the largest company in Cambodia’s location, which transmitted from one source to another distribution, especially over a long distance to the customers. In an electrical power system, compression dead end and jumper terminal is one of the most important pieces of equipment and play important roles in the transmission system high voltage as well. Moreover, a bad or loose connection can be made a hot spot for generating the thermal that affecting for transmitting power, for the survey thermal energy of electrical equipment that they can to observed and quantified as thermography by applying infrared imaging and measuring the camera techniques. This method of detecting heat output is known as infrared thermography (IRT). Infrared thermography is not only a cost-effective technology but also powerful diagnostic equipment for improving system efficiency and dependability, power quality, and worker safety, and preventing outages, expensive equipment failure, and line losses. Using the Neural Network (NN) in the comparison to the Linear Regression (LR) model. This paper regarded schedule working actual base on the history of transmission line reports of using IRT which was noted by EDC, and the input parameters were the weather factors, wind speed, duration of operation, and load on the line. The prediction accuracy of these two methodologies was calculated with the use of the mean absolute percentage error (MAPE), Mean Square Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE). Based on the act result of the two models, the Neural Network (NN) model is better performance accuracy.