摘要
研究变风量空调负荷准确预测问题,由于变风量空调系统可根据负荷的需求动态改变送风量,变风量空调大系统稳态优化控制策略可以使系统节能高效运行,为准确预测负荷,优化系统,首先分析了影响负荷预测的主要因素,对温度、相对湿度的预测模型进行改进,提出了一种自适应扰动粒子群算法的BP神经网络的空调负荷预测模型,加快粒子群算法的收敛速度,提高了空调负荷的预测精度。通过仿真比较,验证了模型在空调负荷预测中的有效性。
The variable air volume air conditioning systems have been widely used with its dynamically air volume change according to the load requirements. The steady state optimization control strategy of the variable air volume air conditioning systems can realize high efficient and energy-saving operation. The accuracy of the load prediction is the basis of system optimization. In this paper, the influencing factors of the load prediction were analyzed, and the pre- diction model of temperature, relative humidity was improved to increasing the prediction accuracy. A load prediction model based on BP neural network combined with adaptive disturbance particle swarm optimization algorithm was pro- posed to speed up the convergence rate of the particle swarm algorithm and improve the prediction precision of the air conditioning load. The results of the simulation and comparison show that the model is effective in the air-condition- ing load prediction.
出处
《计算机仿真》
CSCD
北大核心
2014年第1期391-394,共4页
Computer Simulation
基金
住房和城乡建设部科学技术项目(2012-K1-35)
陕西省教育厅自然科学专项基金项目(11JK0906)
关键词
变风量空调系统
负荷预测
模型
粒子群算法
仿真
Variable air volume air conditioning systems
Load prediction
Model
Particle swarm optimization(PSO)
Simulation