A new fault diagnosis method of multi-parameter new energy power system

New energy power system possesses characteristics such as nonlinearity, high complexity, multiple parameters, and high dimensionality, which pose obstacles to the efficient and accurate comprehensive diagnosis of system faults. A Fault Diagnosis Method was proposed for multi-parameter new energy power system based on. First, a full digital simulation model of the new energy power system was established by using Simulink simulation. Then, a feature scaling approach based on Multidimensional scaling (MDS) was introduced to reduce system dimensionality. We propose the use of Logistic chaotic mapping to enhance the diversity of Beluga Whale Optimizer (BWO) populations and suggest an elite learning method to improve the optimization precision and efficiency of BWO. Error Correcting Output Coding (ECOC) is used to address the limitations of Support Vector Machine (SVM) multi-classification model, ultimately constructing a fault diagnosis model and conducting instance simulations. Simulation results indicate that the proposed model demonstrates high accuracy, fast speed, and strong overall performance.