The Study And Application Of The Ice Storage Air Conditioning Control System

When it comes to21th century, the lack of energy becomes an important issue for it challenges living and development of human beings. People have to seek for the solutions to this problem. In this case, the proposal of this dissertation is to study and analysis the application of ice storage air conditioning in building energy conservation.The consuming of air conditioning system takes a high percentage of power consumption in summer. In hot weather, the living electricity consumption increased rapidly in some regions which led to the shortage of peak power supply. The ice storage air conditioning system is an effective way of solving this problem. It can make ice to store cooling during valley time blocks and release the cooling at peak blocks. According to this method, the cool storage system is capable of peak-load shifting and saving energy further.The management objective of ice storage air conditioning system is to achieve optimal control. Therefore an accurate forecasting of power load is an initial condition of cooperation between ice tank and refrigerator. The main purpose of this dissertation is to make a prediction of the power load of ice storage air conditioning system. The method of artificial neural network is introduced in this study to find the ideal solution of using ice storage air conditioning system in large space.A research on training methods of traditional BP neural network and quantum neural network is done at first. Then the forecast simulation model is built based on a network construction designed for the characteristics of the huge effect of population in gym on power load. The daily power load predictions of ice storage air conditioning system are predicted separately, and then there was a comparison between them. At the same time, artificial fish swarm algorithm was introduced into quantum neural network to improve convergence.After that, a power load prediction of this air conditioning system, which based on an original BP neural network and quantum neural network, was made with Matlab simulation software.In the last place, according to the simulation result, we consider that the research of ice storage system is about a large space stadium, in which the load changes largely and the time is concentrated relatively. So we adopt the method of load balanced and the ice storage tank fist to control. And on this basis, we complete the control plan of the ice storage tank and use WebAccess to design the control interface. The remote monitoring has been actualized finally.