Optimal Control Strategy Study Of Heating Room Temperature Based On Model Prediction

Under the trend of diversification of heating demand,room temperature control is one of the ways to realize heating and energy saving.At present,users have not realized on-demand heating in a large scope in China.The application of envelope insulation and heat metering device has promoted the reduction of heating energy consumption,but compared with developed countries,the energy consumption of heating is still high.The control of indoor temperature has the characteristic of hysteresis,traditional control methods cannot meet the requirements of high precision and high comfort.Using model predictive control(MPC)to regulate the heating room temperature can not only predict the future temperature according to the current state,improve the indoor thermal comfort,but also reduce heating energy consumption.Firstly,this paper verifies the accuracy of IDA-ICE simulation of dynamic heat load through experiments,the room temperature,inner surface temperature of external wall,radiator power and meteorological parameters of a laboratory room in Dalian were measured and simulated synchronously in the software IDA-ICE under fixed and variable conditions during the heating period.The results show that the mean bias error of the measured and simulated indoor temperature and the inner surface temperature of the south outer wall are all within-5%,and the coefficient of variation of the root mean square error are within 10%,which verifies the accuracy of the IDA-ICE energy consumption simulation software and the reliability of the model.Secondly,establish the laboratory RC thermal dynamic model based on the conservation of energy,and transform it into a discrete state space model for simulation in Simulink.Establish the power model of variable frequency water pump and build a MPC simulation platform in Simulink to control the room temperature in the laboratory,the changes of room temperature,flow rate,heat dissipation and frequency conversion pump power during the control process are obtained.The results show that the room temperature can change along the set track under model predictive control,and the fluctuation range of the upper and lower ranges does not exceed 1°C.On this basis,a room temperature PID control simulation model is built in Simulink and compared with the MPC control results.Through calculation,the error indicators of room temperature and set value under PID control and MPC control are 0.03107 and 0.01618 respectively,and the energy consumption of variable frequency water pump are 4.87 k Wh and4.2k Wh respectively.The room temperature thermal comfort under MPC control is better and the flow fluctuation is small,and the energy consumption of variable frequency water pump is saved by about 13.8%.Finally,taking a residential building in Dalian as the research object,the room temperature change law is obtained through IDA-ICE simulation,the number of room models is simplified according to the symmetry of the building,and the MPC strategy is applied to the building.The results show that when the room temperature is set to aconstant value and is affected by outdoor disturbance variables,for different rooms of the same user,the response time of north facing bedroom and south facing bedroom with gable wall is longer.For the same type of rooms with different users on the same floor,the response time of the rooms with side users is longer.For the same type of rooms on different floors,the response of user rooms on the top floor is longer.Adjusting the household flow rate through the variable frequency water pump reduces throttling loss and heat waste,compared with not using model predictive control,heating energy consumption was reduced by 24.9%,which has a better application prospect in actual room temperature control.