Research On Optimaztion Design Method Of Building Envelope Based On Energy-saving And Comfort

Due to the development of social and economic level and the improvement of people’s requirements for indoor environmental comfort,the total energy consumption related to buildings continues to rise.The performance of building envelope has a significant impact on building energy consumption.Therefore,how to propose energysaving design of building envelope on the premise of meeting indoor comfort requirement is one of the key problems to be solved.The traditional energy-saving optimization design of building envelope is conducted by establishing a complex heat transfer model,calculating building energy consumption by changing the thermal parameters of building envelope,and selecting the one with less energy consumption as the optimal design.The optimization design model based on traditional methods has the defects of low optimization efficiency,unable to solve the optimal solution and lack of diverse evaluation indexes.Therefore,an improved optimization model is raised in this paper,design principles and strategies that should be followed by the energy-saving optimization of building envelope while ensuring comfort are explored,as well as the technical measures that should be adopted.Based on the vigorous development of new technologies such as machine learning,this research mainly studied how to introduce these new technical methods into building envelope design process,and investigate the improvement that need to be made.In this way,an overall system which can quickly and efficiently carry out the optimization design of building envelope was established.The specific research contents of this dissertation are as follows:(1)The traditional building envelope energy-saving and comfort optimization design methods were studied.A typical office building model was investigated as an example.Combined with multi-objective optimization,the optimization design model was established.The reference building model was established as the optimization benchmark,Annual Energy Demand(AED)was used as the energy-saving evaluation index,and Annual Uncomfortable Hours(UCH)was used as the comfort evaluation index.Energy load and comfort level was calculated by Energy Plus simulation.The limitations of the optimization design model based on traditional methods were studied.(2)The adaptability of sensitivity analysis method in the research of building energy-saving and comfort was studied.Through theoretical analysis and calculation verification,a suitable Morris ranking score-SRC model was proposed,and the influences of 23 thermal parameters on energy and comfort were investigated.The model can identify the design parameters that have a great contribution on both energy saving and comfort.Main sensitive factors and corresponding optimization direction can be given by this model,thus reduce the scale of the optimization model and improve the efficiency of building envelope design.(3)Artificial Neural Network(ANN)was used to build a new energy and comfort prediction model as the alternative of traditional large-scale time-consuming and inefficient Energy Plus simulation tool.The applicability and accuracy of basic BP neural network for predicting AED and UCH were studied.The model showed good adaptability but relatively low accuracy.The Bayesian Regularization algorithm was applied to improve the accuracy of the neural network,and a Bayesian Regularization Neural Network(BRNN)model was established.The model gave very precise AED and UCH prediction for the test samples,and the calculation speed was greatly improved compared with the traditional Energy Plus simulation model.(4)The improvement strategies that can be adopted in the optimization algorithm were studied,and the limitations of efficiency and parameter range in the traditional forward type algorithm were analyzed.Non-dominated Sorting Genetic Algorithm(NSGA-II)was applied to building envelope design,and an optimization model combining sensitivity analysis with BRNN and NSGA-II was established in order to estimate the applicability of the algorithm for reverse design and muti-objective optimization.By applying the model to solve different example problems,a series of Pareto front solution sets of thermal parameters were obtained.Each solution can intuitively and accurately provide designers with a both energy-saving and comfort design.At the same time,the accuracy and reliability of the BRNN-NSGA-II coupling model for reverse optimization was verified.The model can comprehensively improve the efficiency of optimization design.(5)The adaptability of the model established in this paper in different climate regions was studied,and the Pareto front solution under different meteorological conditions of five thermal climate regions in China were calculated respectively.The principle of collaborative optimization of thermal parameters in different climate regions were explored.This paper not only provides a new method for building envelope design,but also improves the efficiency of accurate envelope optimization design under certain energy-saving and comfort objectives.It can be applied to the decision-making process of architectural design.The specific results of cases studied in this paper can also provide theoretical reference and technical guidance for the design of building envelope.