Overhead Transmission Line Temperature Estimation Based on Artificial Neural Networks

To ensure the power system operates optimally and economically, precise evaluations of component physical limitations are necessary. Hence, to evaluate these constrains, electrical grid operators usually adopt well-known methods based on mathematical model which better describe the physics behaviour of the system. An effective approach to estimate these limits is here presented employing Artificial Neural Network (ANN). With traditional methods, these estimations could be inaccurate due to many factors. Therefore, in this work an ANN based method to estimate all the working temperatures of overhead transmission lines has been presented. The estimation of the proposed ANN based method is compared to the CIGRE physical model. The case study results, based in real data, clearly show the great applicability and the improved accuracy of this proposed ANN based method.