| The operating environment of the chilled source system of large public buildings changes frequently,and its actual operating conditions often deviate from the design conditions,resulting in huge energy consumption of the cold source system.The energy consumption of chillers accounts for the largest proportion of the energy consumption of the cold source system,which has always attracted much attention.In the actual operation process,there is randomness in the time-series distribution of air-conditioning system loads.It is difficult to judge the actual demand of the user side load of the chilled water pipe network.It is difficult to assess the adaptability and confidence of energy efficiency of the chiller control strategy under random load constraints.In the running state,the actual energy efficiency ratio is far lower than the design value under the rated load,resulting in low overall efficiency of the cold source system and serious energy waste.Therefore,it is necessary to study the chiller control strategy under random load constraints and the distribution characteristics of user-side load demand of the chilled water pipe network for optimal control of the cold source system.In this paper,in order to study the non-uniform load matching control strategy of the chiller,the thermal characteristic model of multiple chillers in parallel operation is constructed,and the parallel operation characteristic of chillers under variable load is studied.The static control strategy of chiller based on energy efficiency benchmark is proposed;Load Markov chain prediction model,using Bayesian decision theory to study the impact of randomness of load timing distribution on the selection probability of chillers in parallel,as well as Markov chain prediction duration changes and small probability events in the load event group on chiller control strategies The effect of the dynamic control strategy of the chiller under random load constraints is proposed.The results show that under the self-balanced cooling capacity,there is an energy efficiency benchmark when multiple chillers are operated in parallel,and the chiller group can be operated in accordance with the control strategy corresponding to the energy efficiency benchmark to maximize the energy efficiency ratio;the building load predicted by the Markov chain is It is feasible that there is a balance between the predicted time and the calculation time of the chiller control strategy.According to this balance point,the chiller control strategy is determined to minimize the additional energy consumption risk of the chiller group.According to the energy efficiency expectations,the chiller control strategy can be operated for a long time.In the time span,the energy-saving benefit is maximized,and the control strategy’s energy-saving confidence is improved.In this example,the average energysaving rate of the optimized control strategy is 9.7%,and the energy-saving confidence is increased by 56.7%.In order to study the user-side load demand distribution characteristics of the chilled water pipe network,a user-side thermal characteristic model of the chilled water pipe network was constructed to study the effect of the user’s random behavior on the terminal load demand distribution characteristics under different total air conditioning system loads,and the terminal load distribution randomness was analyzed.The influence of the total load of the airconditioning system has explored the load monitoring method of the air-conditioning system of large public buildings.The results show that under different total air-conditioning system loads,the effect of user random behavior on terminal load distribution increases with the increase of the system total load,and the effect of terminal load changes on the system total load decreases with the increase of the system total load;It is feasible to monitor the load change of the airconditioning system by the frequency of terminal load distribution.The average error in this example is 12.4%. |
PDF Full Text Request