空调热舒适度预测及控制算法研究

发布时间：2018-09-07 17:38

【摘要】：随着生活水平的不断提高，人们对生活品质的要求也愈来愈高。在现代生活中，人类的工作、娱乐、生活等大部分时间均处于室内，因此，人们对室内环境品质的需求也越来越高。为顺应人们对舒适、节能、健康的室内环境的追求，，本文对室内环境热舒适度预测建模与室内环境热舒适度控制在空调系统中的应用做了相应研究。针对热舒适度预测是一个复杂的非线性过程，不便于空调的实时控制应用的问题，本文提出一种改进的粒子群算法(PSO)优化反向传播(BP)神经网络的热舒适度预测模型。这一预测模型通过采用PSO算法优化BP神经网络的初始权值和阈值，改善了传统BP算法收敛速度慢及对网络初始值敏感的问题。同时，本文针对标准PSO算法易出现早熟收敛、局部寻优能力弱等缺点，提出了相应改进策略，进一步提高了PSO优化BP神经网络的能力。实验结果表明：基于改进的PSO-BP算法的热舒适度预测模型较传统BP模型和标准PSO-BP模型，具有预测精度高且算法收敛速度快的特点。本文针对室内环境热舒适度控制在空调系统中应用实现的问题，做了控制变量、控制方式、控制算法等分析比较研究，并最终确定以温度与风速作为系统中的控制变量，以热舒适度的直接控制方式结合智能模糊控制算法实现室内环境的热舒适度控制。同时，本文通过对模糊控制器的设计步骤与设计要点的研究，设计了热舒适度模糊控制器，并对空调系统的热舒适度模糊控制器进行了仿真实现。仿真结果表明，本文设计的模糊控制器性能比传统PID控制器更佳，并且采用热舒适度模糊控制的空调系统比传统温度控制的空调系统能进行更好的热舒适度控制，并能提供更舒适度的室内环境。
[Abstract]:With the continuous improvement of living standards, people's requirements for the quality of life are becoming higher and higher. In modern life, people's work, entertainment, life and so on most of the time are in the indoor, therefore, people's demand for indoor environmental quality is also increasing. In order to adapt to people's pursuit of comfortable, energy saving and healthy indoor environment, this paper studies the prediction modeling of indoor thermal comfort and the application of indoor thermal comfort control in air conditioning system. To solve the problem that thermal comfort prediction is a complex nonlinear process and is not convenient for the real-time control of air conditioning, an improved particle swarm optimization algorithm (PSO) is proposed to optimize the thermal comfort prediction model of backpropagation (BP) neural network. By using PSO algorithm to optimize the initial weights and thresholds of BP neural networks, this prediction model improves the problems of slow convergence speed and sensitivity to the initial network values of the traditional BP algorithm. At the same time, aiming at the shortcomings of standard PSO algorithm, such as premature convergence and weak local optimization ability, the corresponding improvement strategy is proposed, which further improves the ability of PSO to optimize BP neural network. The experimental results show that the thermal comfort prediction model based on the improved PSO-BP algorithm is more accurate than the traditional BP model and the standard PSO-BP model. In order to solve the problem of indoor thermal comfort control applied in air conditioning system, this paper analyzes and compares the control variables, control methods and control algorithms, and finally determines the temperature and wind speed as the control variables in the system. The thermal comfort control of indoor environment is realized by direct control of thermal comfort and intelligent fuzzy control algorithm. At the same time, through the study of the design steps and key points of the fuzzy controller, the thermal comfort fuzzy controller is designed, and the simulation of the thermal comfort fuzzy controller of the air conditioning system is carried out. The simulation results show that the performance of the fuzzy controller designed in this paper is better than that of the traditional PID controller, and the thermal comfort fuzzy control system can perform better thermal comfort control than the traditional temperature control system. And can provide a more comfortable indoor environment.
【学位授予单位】：湖南大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP18;TM925.12

【参考文献】