Research On Optimal Operation Strategy Of Ice Storage Air Conditioning System Based On ABC-BP Neural Network Load Forecasting Model

As China’s urbanization process progresses,the share of building energy consumption in China’s total energy consumption has increased significantly.By 2020,building energy consumption is expected to surpass industrial energy consumption and become the first area of social energy consumption.As the main energy consumption of building energy consumption,the energy consumption of air conditioning system is very important for the sustainable development of energy in China.At the same time,due to the high peak power load of the HVAC system,the peak load period is similar to the peak power consumption period of the city,etc.,it may cause insufficient power supply during the peak period of the urban power grid.As an air-conditioning energy-saving operation scheme capable of“peak-cutting and valley-filling”capacity,ice-storage air-conditioning system has been widely used in engineering in recent years.However,due to the complexity of the system,it is difficult for the conventional operation adjustment methods to take full advantage of the system,which severely limits its economic and social benefits.Therefore,this paper has taken a regional energy station in Chongqing as the research object to study the optimal operation strategy of ice storage air conditioning system.First,the factors affecting the load of the energy station are analyzed from two aspects:outdoor natural factors and indoor human factors.After fully considering the influence of the microclimate and other factors on the air conditioning load of the building,eight items such as outdoor dry-bulb temperature,relative humidity,solar radiation intensity,outdoor wind speed,wind direction,atmospheric pressure,precipitation,load day type,etc.are determined as input items of the load prediction model.Establish the traditional BP neural network load prediction model and use hourly load of energy station from 1st July 2018 to 31st August 2018 as a sample for training and validating the prediction model.The prediction results show that the relative error is 10%-30%,the final mean square error value is 0.00679,and the optimal scatter R~2value is 0.9635.The load prediction model established by BP neural network has poor prediction accuracy and degree of fit.Poor,network prediction stability is insufficient.In addition,the artificial bee colony algorithm was used to optimize the BP neural network to establish the ABC-BP neural network load prediction model.After verification with the same training sample,it was found that the relative error of the prediction of the ABC-BP prediction model is less than 5%,the MSE value is 0.0018296,which is close to the target value of 0.001,and the optimal R~2 value is 0.9927,which proves that the ABC-BP load prediction model effectively optimizes BP Neural network prediction models have disadvantages such as difficulty in determining the initial network parameters and being easily trapped in local minimums.Compared with BP neural networks,they have higher prediction accuracy,performance convergence speed,and fitting degree.Secondly,analyze the mathematical models of the energy consumption of the refrigeration mainframe and the transmission and distribution system,and establish an optimal mathematical model for the ice storage air conditioning system at the 1st Energy Station with daily operating costs as the objective function.;use Matlab to solve the model and give 100%,75%,50%And 25%and other optimized operating strategies for different design load days,and finally summarized and summarized to get the electricity price policy-oriented operating mode.Finally,the economic and social benefits of the optimized operational strategy are examined.The calculation shows that the optimized operating strategy can save operating costs of 5.093 million yuan,with a saving rate of about 11.2%;the annual peak electricity transfer rate is 40%,and the low-peak electricity utilization rate is 28.6%,indicating that it has significant economic and social benefits.At the same time,according to the analysis of the calculation results of different load design days,the conclusion is that as the system load factor decreases,the optimized operational strategy of the ice storage air conditioning system has substantial economic and social benefits.For projects with more partial load conditions,the ice storage air conditioner The system optimized operation strategy has better application value.