The Research On Decomposition-Coordination Optimization Control Of Variable Air Volume Central Air Conditioning System

The Application of VAV central air conditioning system is becoming more and more popular in intelligent buildings.VAV central air conditioning system has attracted more and more attention due to the advantages of its structure design,high energy efficiency,and flexible control,etc.In practical applications,different climate,building structure and load distribution could affect the operation of air conditioning system.On the other hand,because of the complexity of VAV central air conditioning system,it could deviate from the optimum conditions to affect the stability and energy saving of the system if it is controlled inappropriately.Therefore,it's an important research topic for HVAC system to keep the system operating steady with appropriate control strategies,achieve the global optimization by using decomposition-coordination algorithm based on the thermal comfort and indoor air quality,and finally realize high efficient operation and energy saving.The VAV central air conditioning system is composed of several subsystems,and the aim of energy saving is to achieve the global energy efficiency on the basis of local optimization for subsystems.In the research,the VAV air conditioning system is decomposed on the basis of large-scale system decomposition-coordination theory according to different functions of subsystems.The steady-state model for the VAV air conditioning system is constructed by using experiment data,and the corresponding constraint conditions are found out according to the characteristics and relationships between each subsystem.The stable operation of each subsystem in air-conditioning system is the foundation for the global system operation.The robust predictive control algorithm and robust-PID control algorithm for local unit were designed in consideration of the difficulties of accurate model construction for control objects and multi-interference.These two control methods have the advantages of the traditional control and robust control,and it can improve the stability of local control system under the condition of the object model uncertainty.The linear matrix inequality method has been used for the algorithm to solve the problem and the robust predictive control algorithm was improved by using offline and online methods.In this research,the robust control algorithm is used for each subsystem in VAV air conditioning system,and the experimental results verify the effectiveness of the control algorithm.Accurate prediction of air conditioning load is the basis for the optimization of global system.In this paper,the traditional prediction methods for temperature and humidity are modified to improve the prediction accuracy.An improved neural network prediction algorithm based on particle swarm optimization is proposed in this paper,and is used for air conditioning load prediction to provide scientific basis for global optimization.Based on the construction of energy consumption model for subsystems,the paper presents a global optimization model of the VAV system,which combined with thermal comfort and indoor air quality requirements.The improved Interaction Prediction Method was designed to coordinate the global system.The set values of subsystems can be obtained by optimization,and subsystems can be controlled by local controllers according to the searched optimized values in each large-scale system steady state optimization cycle,optimization cycle in order to achieve the purpose of global optimization and energy consumption control.The experimental results show that,using the strategy of large-scale system steady state optimization can be a good solution to the central air-conditioning control and optimization of the global system in summer conditions.The results also show that the global optimized control method can ensure the system operate stably and reach the purpose of energy saving and high efficiency.