基于蚁群优化ANFIS模型的建筑室温状态和能耗预测  

PREDICTION OF BUILIDING ROOM TEMPERATURE STATE AND ENERGY CONSUMPTION BASED ON ANFIS WITH ANT COLONY OPTIMIZATION

作　　者：徐超 于忠清 李劲华 Xu Chao;Yu Zhongqing;Li Jinhua(School of Data Science and Software Engineering,Qingdao University,Qingdao 223800,Shandong,China)

机构地区：[1]青岛大学数据科学与软件工程学院,山东青岛223800

出　　处：《计算机应用与软件》2023年第6期63-69,共7页Computer Applications and Software

摘　　要：建筑采暖、通风和空调(HVAC)系统占据了超过一半的建筑能耗,系统的运行状态和能耗预测是节约建筑能耗、确保热舒适性的关键。提出一种基于蚁群优化算法(ACO)优化的自适应神经网络模糊推理系统(ANFIS),对暖通空调中空气处理单元(AHU)的状态和能耗进行建模和预测。通过蚁群优化算法和最小二乘法对ANFIS网络训练过程中前提参数和结论参数的寻优,进一步提高ANFIS方法对于HVAC等非线性系统建模的速度和精度。与随机森林(RF)、支持向量机(SVM)、BP神经网络和一般ANFIS等模型进行比较,验证了该方法具有更好的预测效果。Building heating,ventilation,and air conditioning(HVAC)systems account for more than half of building energy consumption.The operation state and energy consumption prediction of the system is the key to save building energy consumption and ensure thermal comfort.This paper proposes an adaptive neuro fuzzy inference system(ANFIS)based on ant colony optimization(ACO)to model and predict the state and energy consumption of air handling unit(AHU)in HVAC system.ACO and least square method were used to optimize the premise parameters and conclusion parameters in the training process of the ANFIS,which further improved the speed and accuracy of the ANFIS method for the modeling of HVAC and other nonlinear systems.Compared with random forest(RF),support vector machine(SVM),BP neural network,and general ANFIS,the proposed method was proved to have better prediction effect.

关 键 词：建筑能耗 暖通空调 自适应神经网络模糊推理系统 蚁群优化算法 非线性系统建模 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]