| In the energy consumption of large public buildings,the energy consumption ratio of central air conditioning system has become a key issue in the field of intelligent building research.In view of this phenomenon,adopting scientific and effective technical means is the core of realizing efficient,safe and economical intelligent building.In the central air conditioning system,the energy consumption of cold source system accounts for a large proportion.Therefore,setting more reasonable system operating parameters and adopting appropriate optimization strategies to save energy and reduce emission play a positive role in reducing the energy consumption of the whole air conditioning system and building,and realizing load prediction quickly and accurately can provide important guidance for the energy saving optimization of cold source system.In addition,the airconditioning system in the peak period of high demand for refrigeration,electricity demand,resulting in a rapid rise in power load caused a huge pressure on the power grid.Therefore,under demand response,the optimization strategy of cold source system is formulated based on load prediction to reduce energy consumption and enhance the stability of power grid.Taking the central air conditioning system of an office building in Xi ’an High-tech Zone as the research object,the building load simulation platform and cold source system simulation platform based on TRNSYS were first built,and the required load and system operation data were obtained through simulation.Then,Sparrow Search Algorithm(SSA)is proposed to optimize General Regression Neural Network(GRNN)prediction model.According to the diversity of factors affecting building load,The correlation analysis method was used to select the main load influence factors from the numerous input variables,and the load prediction model was established based on the simulation data.Meanwhile,aiming at the problem that SSA would fall into the local optimal,Random Walk(RW)was introduced to disturb sparrow,so as to improve the global search ability and avoid falling into the local optimal.The combined prediction model of RW-SSAGRNN was constructed.By comparing and analyzing the improved model with the unimproved model and the model of the same combination prediction type,it is found that the proposed RW-SSA-GRNN combination prediction model has better prediction accuracy.Compared with BP,SSA-BP,RW-SSA-BP,SVM,SSA-SVM and RW-SSASVM,The prediction accuracy of this method is improved by 9.73%,6.51%,3.03%,7.49%,3.44% and 0.82%,respectively,which verifies the rationality and effectiveness of this method.Secondly,the energy consumption model and thermal comfort index model of related equipment of cold source system are established.The nonlinear curve fitting function in Origin software is used to identify and fit unknown parameters in the model.The fitting degree of energy consumption model of chiller,chilled water pump and cooling water pump is 0.9264,0.9671 and 0.9895,respectively.The identified performance parameters have high reliability and high fitting degree,and can be used in practice.Then,the energy-saving characteristics of the chiller unit were analyzed,the inlet temperature of cooling water and the supply temperature of chilled water were quantitatively analyzed,and the influence of indoor temperature setting on the air conditioning system was analyzed based on the global temperature regulation strategy,so as to prepare for the subsequent energy-saving optimization.Finally,starting from the set value of chilled water supply temperature at the cold source side and the set value of indoor temperature,the two-layer optimal control strategy of air conditioning system participating in demand response is proposed.In one layer,based on the energy consumption model of each device in the air-conditioning cold source system,the optimization objective function of energy consumption of the cold source system is established.After determining the constraint conditions,the optimization problem of total energy consumption is solved,so as to obtain the optimal chilled water supply temperature and optimized energy consumption at all times,so as to reduce energy consumption and make the chillers in a state of efficient operation.In the other layer,the strategy of precooling in advance is adopted and the indoor temperature setting value of the air conditioning system is set.Under the premise of ensuring the thermal comfort of human body,the energy consumption in peak hours is reduced or transferred.The energy consumption and peak load reduction of air conditioning system under the proposed strategy are studied.The results show that energy consumption can be reduced by adopting the energy-saving optimization strategy of cold source system,and the highest energy-saving rate can reach 26.95%.On the premise of meeting the thermal comfort demand of human body,when adopting the indoor temperature setting optimization strategy,the energy saving effect can be achieved under the optimized working conditions,and the power consumption in the peak period can be effectively transferred to the normal period.The transfer amount of electricity load in the peak period is about 22.7%.When the proposed two-layer optimal control strategy is adopted,the energy consumption is the least,and the energy saving rate can reach 35.82%. |
