Study Of Natural Ventilation Of Office Space In Cold Region Based On MPC During Transition Season

With the development of architectural design,low energy consumption and high comfort have become one of the development directions of architectural design.At present,the energy consumption of buildings is still very large,and the energy consumption of public buildings is the main one.The mechanical ventilation and air conditioning system of public buildings brings huge energy consumption,and the performance in comfort adjustment is not enough to satisfy everyone.At present,in the exploration of low energy consumption in office space,domestic research on office buildings in severe cold areas is generally concentrated on winter insulation and cold protection,and there is a relative lack of research on transitional season and summer indoor natural ventilation.In fact,compared with Guangzhou,Shanghai,Beijing and other places,the percentage of natural ventilation and energy saving in the cold season is the highest in the transitional season and summer,and the air temperature and humidity in these two seasons are compared with those in summer,hot winter,hot summer and warm winter.It is milder and more comfortable,compared to the design with natural ventilation.Existing research shows that natural ventilation is a means to effectively improve indoor thermal comfort.Based on the above situation,this paper proposes to introduce natural ventilation in the transitional season to solve the problem of high energy consumption and low comfort,establish a neural network prediction model,and propose a set of model prediction control strategies to carry out effective natural ventilation and improve the indoor thermal environment.To improve thermal comfort.This paper presents a method for establishing an indoor temperature prediction model based on LSTM algorithm.The indoor and outdoor temperatures and the indoor humidity sequence at the current and historical times are used as inputs to predict the indoor temperature change at the next moment.A one-month field survey was then conducted at the No.2 Teaching Building of the Heilongjiang Provincial Business School in Shuangcheng District,Harbin,and the temperature-predicted model of the room temperature was established using the measured data.By comparing with the actual value,it is proved that the prediction result is more accurate,and the fitting degree reaches 98.7%.Based on the above indoor temperature prediction model,this paper proposes a natural ventilation prediction control strategy for office buildings in severe cold regions.Through typical research,a typical office unit site was built,and the above natural ventilation prediction control strategy was verified in the site.The results show that compared with the fixed point temperature control method,the indoor room temperature fluctuation range under the natural ventilation prediction control is smaller,the time in the comfort range is longer,and the room interior thermal comfort is better.By establishing an indoor temperature prediction model and applying the natural ventilation prediction control strategy,the indoor thermal comfort of the office building transition season in the cold region is improved,and the energy consumption brought by the air conditioning system is reduced.