基于蒙特卡罗模拟的区域建筑冷负荷预测模型

Regional building cooling load prediction model based on Monte Carlo simulation

针对目前区域能源规划时单体建筑的参数不完备、无法用常规负荷计算软件对区域内各单体建筑进行负荷预测等难题,提出基于蒙特卡罗的区域建筑冷负荷预测方法。该方法首先建立适用于区域建筑群的冷负荷预测随机模型,并确定随机模型的风险变量分布特征,然后借助于蒙特卡罗随机模拟方法获得区域内各类建筑以及整个区域建筑群的峰值冷负荷概率分布和全年平均逐时冷负荷。同时利用实际应用场景相关参数模拟区域建筑群的冷负荷特性。仿真结果表明:区域建筑群冷负荷预测随机模型可以有效地计算区域内峰值冷负荷的频数分布和累积概率,在典型应用场景下,区域峰值冷负荷均值为60.9 MW,标准差为5.5。本研究有助于实现决策风险控制和成本控制,并将为区域能源规划提供一种可行的设计思路和方法。

In light of the facts that complete building parameters can not be obtained during the period of regional energy planning, and the conventional software for load calculation cannot accurately simulate the building＇s cooling load, a new method based on Monte Carlo methodology was proposed for forecasting the regional building cooling load. Firstly, the regional building cooling load prediction stochastic model（RBCLPS model） was established, followed by the risk characteristics determination of variable distribution in this model. Besides, the distribution of peak cooling load probability for various buildings（including the whole regional building） as well as the annual average hourly cooling load can also be obtained by using Monte Carlo method. Furthermore, the simulation was performed according to the case study by considering the cooling load characteristics of regional building. The results show that RBCLPS model can calculate the available area of the distribution as well as the cumulative probability of peak cooling load frequency. The average and the standard deviation of regional building peak cooling load are 60.9 MW and 5.5, respectively. This study would be beneficial to achieving decision-making risk control and cost control for regional energy planning and providing a feasible solution for them.