首页|暖通空调调温自适应强化学习下的节能控制方法设计

暖通空调调温自适应强化学习下的节能控制方法设计

扫码查看
当前基于规则的暖通空调节能控制方法往往是静态的,其控制规则的制定依赖工程师或者设备管理员的经验,而强化学习的节能控制方法不依赖工程师的经验并能进行自适应学习.因此,对于缺乏历史数据的旧建筑而言,基于强化学习的控制方法更具有研究价值.因此,提出暖通空调调温自适应强化学习下的节能控制方法.通过引入等效热参数建模法,建立热泵温度调节过程中的热力学模型,从而分析暖通空调能耗控制的问题.分析电力消耗过程中的供应环节,结合峰谷电价、激励补贴、用户适宜温度三方面因素,实现用户需求响应条件下的理想暖通空调温度变化量计算.基于负荷响应公平性考量,结合理想温度变化量与相对温度距离计算方法,设计强化学习算法,规划暖通空调主机温度调节顺序,实现对暖通空调热泵的控制,以实现自适应学习下的节能控制.实验结果表明:所提方法能够在低能耗前提下实现室内温度的精准控制,对暖通空调技术的发展具有重要意义.
Design of energy saving control method based on adaptive reinforcement learning for HVAC temperature regulation
Current rule-based HVAC energy-saving control methods are often static,and the formulation of control rules depends on the experience of engineers or equipment administrators,while reinforcement learning energy-saving control methods do not rely on the experience of engineers and can be adaptive learning.Therefore,for the old buildings lacking historical data,the control method based on reinforcement learning has more research value.Therefore,an energy saving control method based on adaptive reinforcement learning for HVAC temperature regulation is proposed.By introducing equivalent thermal parameter modeling method,the thermody-namic model of heat pump temperature regulation process is established,so as to analyze the problem of HVAC energy consumption control.Based on the analysis of the supply link in the power consumption process,the ideal HVAC temperature change calculation under the condition of user demand response is realized by combining the three factors of peak-valley electricity price,incentive sub-sidy and user's appropriate temperature.Based on the fairness of load response,combined with the calculation method of ideal temper-ature variation and relative temperature distance,the reinforcement learning algorithm is designed to plan the temperature adjustment sequence of HVAC host,and realize the control of HVAC heat pump,so as to realize the energy-saving control under adaptive learn-ing.The experimental results show that the proposed method can achieve accurate control of indoor temperature under the premise of low energy consumption,which is of great significance for the development of HVAC technology.

equivalent thermal parameter modelingload cluster responseuser demand responserelative temperature dis-tancepriority rankingreinforcement learning

李程、刘长寅

展开 >

广西工业职业技术学院建筑工程学院,南宁 530001

吉林省水利电力职业学院新能源学院,长春 130000

等效热参数建模 负荷集群响应 用户需求响应 相对温度距离 优先度排序 强化学习

2024

自动化与仪器仪表
重庆工业自动化仪表研究所,重庆市自动化与仪器仪表学会

自动化与仪器仪表

CSTPCD
影响因子:0.327
ISSN:1001-9227
年,卷(期):2024.(11)