The Research Of Simulation On Solar-earth Source Heat Pump System

Nowadays the practical situation of energy utilization and environment situation has been becoming more and more stern in China, our country has began to make utilization of solar energy, soil source and so on renewable energy source as national prioritize and promoted energy consumption pattern. Solar-earth source heat pump system (SESHPS) is the kind of combined heating source heat pump system that adope the solar energy and soil source as heating source. It is promising because of enengy conservation and environment protection.This paper takes the SESHPS as research object, and sets up heat pump unit model, heat pipe evacuated tubular collector model, vertical U-tube ground heat exchanger model, heat storage tank and indoor thermal load model. According to Lagrange Middle-value theorem, simplify models integrated with BP artificial neural network and RBF artificial neural network for condenser and evaporator are built, as well as set up the predictor-corrector models of condenser and evaporator simplified models based on the BP neural network. Based on the BP artificial neural network and heat pipe evacuated tubular collector theoretical model, the solar collector BP artificial neural network model is built. Genetic algorithm is used to optimize the BP artificial neural network mode of solar collector.Referring to Kelvin's line heat source theory and Kavanaugh's cylindrical source theory of classic constant heat flux, the idea of superposition principle, load step and load aggregation is employed to build a model of vertical U-tube ground heat exchanger. Based on BP artificial neural network and vertical U-tube ground heat exchanger theoretical model, the optimization model of vertical U-tube ground heat exchanger is built.The component models are incorporated to form the SESHPS through energy conservation, mass conservation and the coupling relationships of system units. The feasibility of simulation software is verified by a building's heating engineering in Qingdao which is equipped with SESHPS, and the simulation is applied in the heating condition in winter. The dynamic operation condition and static operation condition of SESHPS are studied and analyzed. The simulation results show the COP_h of heat pump unit in dynamic operation is higher than it in static operation condition.The neural network and genetic algorithm is applied in building SESHPS simulation model. The simulation model is verified by experimental data. The results show that computational accuracy and efficiency of SESHPS simulation model can be improved, as well as the energy consumption of SESHPS in two kinds of operation conditions are analyzed. The conclusion of experiment and numerical calculation analysis will offer reference materials for further popularization and application of this project.