摘要
末端智能通断控制技术是通过设定温度发出通断信号实现热用户末端散热设备的管路阀门启闭的控制方式。末端控制不仅影响用户室内温度,还影响热力站水力工况。本文以天津市某安装通断阀的小区为例,提出了变压差的控制方式。首先,基于K-means算法和阀门开启比例α变化规律辨识调节时间;其次,由箱型图和概率分析确定阻力数目标值s_(m),由K-means算法辨识流量目标值G_(2,m);之后,由s_(m)和G_(2,m)计算压差设定值;最后,基于预测热负荷调节供水温度,由BC值推测典型用户室内温度。结果表明,变压差控制有2种调节模式,α较稳定时一天给定一个压差,α变化较大时分别在0:00、8:00和16:00进行调节;该方案可使流量变化更稳定;水泵可节能3.7%;典型用户室内温度标准差降低0.02~0.052℃。
The terminal intelligent on-off control technology is a method that realizes the opening and closing of the pipeline valve of the heat user’s heat-dissipating equipment by sending an on-off signal based on the set temperature.Terminal control not only affects the indoor temperature at user side,but also affects the hydraulic conditions of the heating station.This paper took a community in Tianjin with on-off valves as an example,and proposed a method of changing the differential pressure.The adjustment time was first identified based on the K-means algorithm and the proportion of valve opening time,α.Then,the target value of the resistance number,s_(m),was determined by box plot and probability analysis,and the target value of flow rate,G_(2,m),was identified by the K-means algorithm.The differential pressure setpoint was further calculated by s_(m)and G_(2,m).Finally,the water supply temperature was adjusted based on the predicted heat load,and the indoor temperature of the typical users was predicted by the BC value.The results show that the variable differential pressure control method has two adjustment modes.Ifαis relatively stable,one differential pressure is given for each day,otherwise,it is adjusted at 0:00,8:00,and 16:00 respectively.The method can make flow rate change more stable,the pump can save 3.7%of the energy,and the standard deviation of the indoor temperature of typical users is reduced by 0.02~0.052℃.
作者
马文菁
郭晓杰
曹姗姗
孙春华
齐承英
MA Wenjing;GUO Xiaojie;CAO Shanshan;SUN Chunhua;QI Chengying(School of Energy and Environmental Engineering,Hebei University of Technology,Tianjin 300401,China)
出处
《建筑科学》
CSCD
北大核心
2023年第2期215-224,共10页
Building Science
基金
天津市科技计划项目“基于机器学习的供热系统源-网-荷协同调控策略研究”(21YDTPJC00900)
河北省高等学校科学技术研究项目资助“基于多元时序数据挖掘的供热系统用能模式识别与诊断研究”(QN2021212)。
关键词
通断控制
水力稳定性
K-MEANS
变压差
热负荷
BC
on-off control
hydraulic stability
K-means
variable differential pressure
heat load
BC