Abstract
The BeiDou-3 navigation satellite system (BDS-3) provides a full-domain high-precision positioning service for the power system to ensure safe and stable operation. However, BDS-3 power system positioning faces certain challenges, such as complex electromagnetic interference and incomplete error elimination. Herein, a back propagation neural network- improved least mean square (BP-LMS) adaptive filtering method is proposed for the BDS-3 full-domain and high-precision power system positioning, which utilizes the loss function to update the weight of the BP hidden layer, computes the pseudo compensation range, and eliminates the impact of electromagnetic interference to enhance the accuracy of power system positioning. Simulation results confirm the superior performance of BP-LMS in positioning accuracy and error elimination. Compared with LMS and normalized least mean square (NLMS), the filtering error of the proposed BP-LMS adaptive filtering method is decreased by 57.14% and 51.38%, respectively.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have