Multi-objective energy saving optimization of residential buildings based on MABC-BP

In order to better satisfy the energy saving requirements of residential buildings in China, this paper proposes a model based on MABC-BP, which performs multi-objective optimization of residential building energy saving to further improve the energy-saving effect. Firstly, in view of the limitations of the traditional ABC algorithm, it was proposed to utilize the particle swarm algorithm to improve the artificial bee colony algorithm, so as to improve the optimal search performance of the traditional bee colony. Then, the parameters of back propagation (BP) neural network were optimized by the improved artificial bee colony algorithm. Finally, using the parameters of a building’s exterior wall, roof and ground as a base, the BP neural network was trained, and the trained BP neural network was utilized to realize the multi-objective prediction of residential building energy saving. Simulation results show that the BP neural network with optimized parameters has good prediction results with an average error value of 0.005. Compared with the original BP neural network, the optimized BP network has better convergence and solution. Compared with the ABC-BP algorithm model, the optimized BP algorithm model achieves better optimization results when optimizing the optimization target selected by the experiment. The above simulation results verify the superiority of the design and the necessity of optimization in this paper, which has certain application value.