.png){kind=icon}
.png){kind=icon}
Return
Research on building energy-saving based on GA-BP coupled improved multi-objective whale optimization algorithm
.png){kind=icon}
PRE
AI
DOI：10.1016/j.enbuild.2024.115141.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
High building energy consumption, significant greenhouse gas emissions, and high investment are current issues in building renovation. To address this issue, an optimization model based on the improved multi-objective whale optimization algorithm (IMOWOA) coupled with GA-BP is proposed. Firstly, an energy consumption simulation model for buildings was established, and the simulated data was trained using GA-BP. Next, three improvement strategies were introduced: optimal Latin hypercube sampling, cosine control factor, and archimedes spiral. Enhance the search efficiency and convergence time of the IMOWOA algorithm. To test the optimization performance of the IMOWOA algorithm, it was evaluated using DTLZ and UF series test functions. Finally, the Entropy Weight-TOPSIS method was utilized to further screen the optimization results, thereby determining the best renovation strategy. The results show that the R2 of GA-BP is close to 1, indicating good predictive performance. The GA-BP training results can serve as the fitness function for the IMOWOA algorithm. The optimal renovation strategy results in a building energy consumption of 49.98 kW center dot h /m2, an environmental benefit of & YEN;1.67 x 105, and a net present value of & YEN;1.21 x 105. It is evident that the proposed method can achieve a trade-off renovation strategy among multiple conflicting objectives. The feasibility of combining GA-BP with the IMOWOA algorithm for multi-objective building energy-saving renovation was verified, which provides a theoretical reference for future similar studies.
Keywords:
Building renovation
Improved Multi-objective whale optimization algorithm
Renovation strategy
Environmental benefit
Net present value