Radial Distribution Systems state estimation using methods of artificial intelligence networks

An introduction to radial distribution systems and estimating the state and the importance of that estimation in determining the efficiency, reliability and accuracy of those systems, and the use of traditional and smart measuring devices, such as SCADA, in these systems, and the methods used in estimating the voltage and angle value using traditional methods (WLS, WLAV) and the problems and limitations in radial distribution systems. The number of busbars and feeders is large, and the number of measuring devices is small compared to the number of busbars, and the use of many measuring devices leads to a large financial cost. Currently, artificial intelligence methods are used, for example, traditional artificial neural networks, where it is possible to estimate the state of the system with a small number of measuring devices and with acceptable accuracy and efficiency. Many improvements were used in the methods of artificial intelligence networks, but they require a lot of training time. An artificial intelligence network was created for each feeder in order to improve accuracy, efficiency, reliability, and high training speed. In this paper, Functional Link Neural Networks was used, which has proven its effectiveness and accuracy in guessing the state of the system, and to improve accuracy. Because it is more efficient and reliable than previous methods, this method was used for each feeder separately