Construction On Multi-objective Optimization Model For Ultra-low Energy Steppe Houses In West Inner Mongolia Based On MABC-BPNN

In recent years,the country cares more about people’s life in countryside and the environment they live,herdsmen have also made some concrete demands for their own living environment.Nowadays,most grassland dwellings were built decades ago,and their architectural construction is very rough.Due to the lack of good design,the insulation performance of the houses is relatively poor,and the herdsmen’s use of clean energy is also a blank,leading to the current situation.The houses of some farmers and herdsmen consume a lot of energy,and the indoor thermal comfort is not up to the standard in winter.Based on the multi-objective optimization model of ultra-low energy grassland dwellings in western Inner Mongolia based on MABC-BPNN,this paper optimizes the design of ultra-low energy grassland dwellings in western Inner Mongolia,and provides a basis and reference for the construction of local ultra-low energy dwellings.In this regard,this article will start with three optimization goals based on local specific requirements and local unique climate conditions,and use building heating energy consumption,indoor thermal comfort,and engineering costs to constrain post-influencing factors to find the best solution that are more appropriate for local area.The article has applied the modified artificial bee colony algorithm(Modified Artificial Bee Colony algorithm,MABC)which improved by the artificial bee colony algorithm(Artificial Bee Colony algorithm,ABC)and BP neural network(Error Back Propagation Neural Networks,BPNN).To begin with,the BP neural network is used to construct a predictive model,and then a multiobjective optimization model is constructed,so as to provide some ideas for the construction and transformation of grassland dwellings in western Inner Mongolia.At the same time,It also has a certain guiding significance for the transform,rebuilding and the optimize of energysaving for dwellings of grassland in western Inner Mongolia.After analyzing the previous literature,this article first propose to construct a multiobjective optimization model.In the process of constructing models while dealing with these actual problems,the correlation analysis module of the SPSS software was used to relevantly analysis the problems that advanced by this article.Finally,through multiple screenings,7influencing factors were selected and then they were used as variables of the multi-objective optimization model.All these influencing factors were used for optimizing three optimization goals.Firstly,before simulating a large number of data to obtaining training samples and test samples,these 7 influencing factors are constrained by national standards,and then a BPNN prediction model will be constructed by these training data and test data.After that the error of the prediction model will be analyzed,which found to be 0.00535,which means that the prediction model would be available.Secondly,the MABC-BPNN multi-objective optimization model would be constructed.Because of the faultiness of the ABC algorithm,the ABC algorithm need to be optimized.After that,the fitness function of the MABC algorithm was set as the BPNN prediction model,thanks for the combination of these two algorithms,a multi-objective optimization model for energy conservation of peoples’ buildings in the grasslands of western Inner Mongolia was finally constructed.Finally,an optimal solution will be obtained from a Pareto optimal solution set,which came out by this multi-objective optimization model.The energy consumption of the optimized solution is greatly reduced by 80.98%,and the indoor thermal comfort PMV is also not small.The degree of improvement is 1.843,and the final cost is only increased by 10.1%.From the analysis of the results,the multi-objective optimization model meets the requirements.At the end of the article,the MABC-BPNN model and the ABC-BPNN model are respectively optimized and designed for the same comparative building.The final result turns out that the ABC algorithm after improving has a better performance on the convergence of the MABC-BPNN model,in the meantime,the quality of the optimal solution set which came out from MABC-BPNN model was also very high.In summary,by constructing a multi-objective optimization model for ultra-low-energy grassland dwellings in western Inner Mongolia based on MABC-BPNN,the model basically meets the design needs of local ultra-low-energy dwellings,and it achieves the balance of these three optimization goals,the heating energy consumption,indoor thermal comfort,and engineering cost.It also provides some kind of guidelines for the structure and design of the multi-objective optimization of ultra-low energy consumption residential buildings in western Inner Mongolia.