响应面优化算法在被动式低能耗建筑设计中的应用

被动式低能耗建筑设计过程计算量大,涉及变量多,不同变量间关系错综复杂,传统方法无法获取复杂变量的最优解集,不能保证被动式低能耗建筑设计不同目标函数达到最优。为此,将响应面优化算法应用于被动式低能耗建筑设计中。将最大化舒适度、最小化耗能率作为被动式低能耗建筑设计的目标函数。通过拉丁超立方实验设计方法,获取被动式低能耗建筑设计优化问题的设计空间;在设计空间中随机抽样。在此基础上,通过最小二乘法拟合建立响应面模型。利用遗传方法对拟合响应面模型进行改进和更新处理,直至达到收敛,保证被动式低能耗建筑设计达到低能耗率、高舒适性的目的。实验结果表明,响应面拟合准确性高,将响应面优化算法应用于被动式低能耗建筑设计中,发现其能够有效降低能耗,提高舒适性,和其他方法相比有更好的优化结果。

Application of response surface optimization algorithm in passive low energy building design

the design and optimization of the construction process of passive and low energy consumption of large amount of calculation, involving many variables, different variables between the traditional methods can not obtain optimal perplexing, complex variable solution set, passive and low energy building design in different objective function optimum cannot be guaranteed. To this end, the response surface optimization algorithm is applied to passive low energy consumption architecture design. Maximizing the comfort degree and minimizing the energy dissipation rate are used as the objective function of passive low energy building design. Through Latin hypercube experimental design method, the design space of passive low energy consumption building design optimization problem is obtained. Random sampling is designed in the design space. Based on that, the response surface model is established by least square fitting. The fitting response surface model is improved and updated by genetic algorithm until convergence is achieved, which ensures the purpose of low energy consumption and high comfort for passive low energy consumption building design. The experimental results show that the response surface fitting accuracy is high, and the response surface optimization algorithm is applied to passive low-energy building design, it is found that it can effectively reduce energy consumption and improve comfort, compared with other methods, it has better optimization results.