Research On Model Control Strategy Of DH Network Energy System Based On Model Predictive Control

The regulation of the current regional energy system is characterized by large scale,uncertainty,and low control accuracy.Therefore,the conventional feedback-based regulation strategy not only causes the system power supply and demand load to be mismatched in time series,but also causes high energy consumption of the energy system.The model predictive control strategy can solve the problem of low system control precision and unstable control parameters due to its predictive control,rolling optimization and feedback correction.Therefore,this paper proposes a regional energy system regulation model based on model predictive control,which is of great significance for reducing the energy consumption of regional energy systems,ensuring the stability of control parameter output,and overcoming the large hysteresis in the process of energy system regulation.This paper takes the energy system of Tianjin University as the research object,establishes the energy system model predictive control regulation model through the joint operation of TRNSYS and MATLAB,and simulates the energy system operation optimization strategy.With the lowest energy consumption of the system as the optimization goal,under the input signal of building load and the corresponding constraints,the genetic algorithm is used to optimize the two control parameters of the water system outlet temperature and the effluent flow,and finally the two optimal ones will be obtained.The control parameters are returned to the mathematical model as the set value to continuously perform rolling optimization and feedback correction to achieve predictive control,ensuring the stability of the control parameter output while reducing the time delay and overshoot in the control process.Considering that the energy system has many service buildings,large load value,uncertainty of building load,etc.,predictive load is used instead of simulated building load to input control the energy system and solve the problems of time delay and load mismatch in the energy system control from the front end.In addition,the media of the energy system needs to be transmitted to the buildings through the regional pipe network,and a huge flow time delay occurs during the flow of the medium.In this paper,the CFD simulation method is used to determine the flow time delay of each building due to media transmission.The delay time is loaded into the established control model time delay module to complete the feedforward input to the predicted load,and on the other hand,overcome the influence of the flow time delay on the energy system control.The TRNSYS simulation platform can verify the proposed model predictive control strategy.Under the model predictive control strategy,the system energy consumption of the two operating conditions of summer cooling and winter heating is simulated.The results show that compared with the actual control strategy,the model predictive control strategy can not only improve the load in time series,but also reduce the energy consumption of the system.In summer and winter,the energy consumption can be reduced by 9.56% and 7.75% respectively.Compared with the proposed contrast condition,the load feedforward fuzzy control strategy also increases the energy efficiency in summer and winter by 5% and 2.5%,respectively.