国产欧美精品一区二区,中文字幕专区在线亚洲,国产精品美女网站在线观看,艾秋果冻传媒2021精品,在线免费一区二区,久久久久久青草大香综合精品,日韩美aaa特级毛片,欧美成人精品午夜免费影视

Building energy consumption data has non-stationary and nonlinear characteristics. A single prediction model is difficult to predict accurately, and a new hybrid model for short-term prediction of building energy consumption is proposed. Complementary ensemble empirical mode decomposition (CEEMD) is utilized to decompose volatility energy data into a set of intrinsic mode functions and a residual sequence. Based on reverse learning, differential evolution algorithm and control parameters, the Whale Optimization Algorithm (WOA) is upgraded to effectively solve the problem of premature convergence and local optimality, and the upgraded whale optimization algorithm (UWOA) is proposed. UWOA is utilized to optimize the weights and thresholds of the Elman neural network. The optimized Elman neural network predicts and integrates the intrinsic mode functions and the residual sequence, and then the energy prediction is obtained. The CEEMD-UWOA-Elman hybrid model is used to predict the energy consumption of a large public building in Shanghai. The results show that the hybrid model has a good prediction effect.