Research On Ice Storage Air Conditioning Load Forecasting And Strategy Optimization Based On Insect Intelligent

There is a large demand for cooling load in public buildings during summer,the ice storage air conditioning system utilizes surplus electricity to store cold at night,and the ice tank jointly auxiliary refrigeration units provides cooling during the day-time.So as to realize the "peak shaving and valley filling" of the air conditioning electricity,and alleviates the power grid pressure during the peak load in summer.The ice storage air conditioning systems are large-scale control systems that are spatially and temporally distributed,the improper control strategies in actual projects have caused problems such as low efficiency,wasted energy,and high operating costs.Therefore,researching real-time cooling load forecasting for the next day of building,optimizing system operation strategy,improving chillers' efficiency,and realizing the optimal combination among units have important guiding value for the realization of energy-saving and economic social benefits of ice storage air conditioning systems.However,across the application of traditional centralized control systems in HVAC,the control network construction is complex,the system upgrading is difficult,and the equipment performance parameters are difficult to obtain,which leads to errors in the model and optimization algorithm.To this end,this paper uses a new type of insect intelligent building technology,a control system based on insect intelligent for building cooling load forecasting and ice storage air conditioning optimization operation strategy is constructed.According to the cooling load forecasting results,the cooling strategy of the chillers and ice tank is optimized.Specifically include:(1)Taking the ice storage air conditioning system of a shopping mall building inXi'an as the research object,the operation energy consumption models of chillers,cooling tower and water pumps are established,as well as the calculation models of the ice tank's cooling storage capacity and melting ice cooling capacity are established,which provide a basis for optimizing control strategy.(2)The thesis studies the cooling load forecasting method based on building space unit,and establishes an improved adaptive learning rate deep belief network cooling load forecasting model to predict the hourly cooling load demand of the next day.According to the division principle of building space units,each space unit independently and concurrently completes the cooling load forecasting of its own controlled units.Based on the spanning tree rules,adjacent units communicate with each other,and finally the total cooling load demand of the entire building is obtained.The experimental results of the first-floor building of the mall show that this improved cooling load forecasting model has achieved better accuracy on a single spatial unit,and compared with the traditional building overall forecasting,the parallel forecasting method of each spatial unit fully excavates the nonlinear dynamic characteristics of the cooling load of commercial buildings,and the prediction result is closer to the actual load.(3)Applying insect intelligent control structure and taking electromechanical equipment as a unit,a distributed multi-objective particle swarm differential evolution combined optimization algorithm is proposed.Taking the energy consumption,operating cost and energy loss as the optimized objectives,and the physical conditions as constraints,the hourly partial load ratio of each chillers and the hourly supplied cooling capacity ratio of ice tank are solved.Compared with the traditional control strategies,the results show that the optimization algorithm is an efficient distributed optimization algorithm with good convergence,high stability,strong robustness,etc.Moreover,the optimization results improve the operating efficiency of chillers,and the load distribution among chillers and ice tank balances the contradiction between energy consumption and operating cost,achieves higher benefits.The above research is based on the insect intelligent building technology,and a cooling load forecasting method based on building space units is proposed,which further improves the accuracy.Furthermore,the multi-objective optimized operationstrategy of ice storage air conditioning based on electromechanical equipment has achieved better energy-saving economic effects.