Prediction Method of Dissolved Gas Volume Fraction in Transformer Oil Based on OVMD-HWOA-KELM Model

To address the problems that it is difficult to accurately predict the volatility and stochasticity of dissolved gas sequences in transformer oil, this paper proposes a combined prediction model based on optimal variational mode decomposition (OVMD), hybrid whale optimization algorithm (HWOA), and kernel extreme learning machine (KELM). Firstly, OVMD is applied to obtain the optimal decomposition parameters and decompose the original sequence into a series of relatively smooth components. Secondly, the HWOA algorithm is proposed by incorporating chaotic mapping, nonlinear convergence parameters, adaptive weight factors and improved arithmetic optimization algorithm in the whale population, and the superiority of the HWOA algorithm is verified by using the test function. Then, the KELM prediction model is constructed for each component separately, and the key parameters of KELM are optimized by using HWOA. Finally, the prediction results of each component are superimposed and reconstructed to obtain the final prediction results. The case study shows that the decision coefficients of the model proposed in this paper for the prediction of normal and abnormal transformer cases can be up to 97.7% and 93.46%, respectively. Compared with the existing methods, the model in this paper has better accuracy and adaptability, and it can provide favorable technical supports for the operation and maintenance management of power transformers. © 2024 Science Press. All rights reserved.