论文部分内容阅读
中央空调的能耗是整个建筑的主体功耗。本文通过对已有的历史数据进行分析,挖掘数据内部关联,建立相应的数学模型达到降低中央空调能耗的效果。首先在进行数据挖掘前,对原始数据进行预处理,建立相应的目标。目标一,对变量分布和相互关系进行可视化,并做相关性分析,找出相关性强的几组变量。通过高斯混合模型对设备状态信息进行聚类,用线性回归进行拟合,建立功率与转速的多元线性回归关系。目标二,以降低系统总耗电量作为优化目标,将设备安全性与达到实际制冷效果作为约束条件,优化目标主要使用目标一中的高斯混合模型和多元线性回归进行衡量,遗传算法可以寻找全局次优解,据此得出优化参数数值。本控制策略降低了系统的总耗电量,使得耗电量为原先的90%左右。
Central air conditioning energy consumption is the main building power consumption. In this paper, we analyze the existing historical data to find out the internal correlation of data and establish the corresponding mathematical model to reduce the energy consumption of central air conditioning. First, prior to data mining, the original data pretreatment, the establishment of the corresponding goals. Goal 1. Visualize the distribution and interrelationships of variables and do correlation analysis to identify several groups of variables that are highly relevant. The Gaussian mixture model is used to cluster the equipment state information and fitted with linear regression to establish the multiple linear regression relationship between power and speed. The second goal is to reduce the total power consumption of the system as the optimization goal, and to take the safety of the equipment and the actual cooling effect as the constraints, the Gaussian mixture model and the multiple linear regression in the main objective 1 of the optimization goal, and the genetic algorithm can find the global The second best solution, derived from the optimal parameter values. The control strategy reduces the total system power consumption, making the power consumption is about 90% of the original.