• This paper proposes a CIGWO-BPNN hybrid prediction model • The model predicts a nonlinear synergy amid individual and overall energy-saving rate • Variations exist in individual energy-saving rates across different climate zones • Different climate zones require tailored energy-saving strategies • Walls and roofs contribute more to energy efficiency than windows for case buildings The overall energy-saving rate of prefabricated buildings is a critical indicator for green building certification. The overall energy-saving rate is not a simple summation of individual energy-saving rates, as complex nonlinear coupling relationships exist among them. This makes it difficult for traditional regression or simple neural network models to precisely characterize these interactions. Prefabricated buildings in four representative Chinese cities—Harbin, Tianjin, Chongqing, Fuzhou—are selected as case studies in this paper. A hybrid CIGWO-BPNN model is proposed that integrates Chaotic Improved Grey Wolf Optimization (CIGWO) with a Back-Propagation Neural Network (BPNN) to analyze the complex nonlinear synergistic relationships between individual energy-saving rates and the overall energy-saving rate. Results indicate that within the specified design ranges, the proposed CIGWO-BPNN hybrid model demonstrates high predictive performance (R²= 0.9467∼0.9851; MSE = 0.0001∼0.0004), with significantly faster convergence speed and no over-fitting compared to conventional BPNN models, providing a more effective tool for precise evaluation of building energy efficiency performance.
Huo et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: