首页|基于模糊逻辑的空调系统广义预测控制参数整定

基于模糊逻辑的空调系统广义预测控制参数整定

扫码查看
针对多输入多输出(multiple-input-multiple-output,MIMO)变风量空调系统(variable air volume,VAV),本文提出了 一种基于模糊逻辑和事件触发的广义预测控制(generalized predictive control,GPC)参数整定方法.针对多变量VAV空调系统,基于输出斜率等新型模糊目标参数及高斯双边形隶属度函数建立模糊预测模型,提高了隶属函数拟合度以更贴切地反映系统当前时刻状态.利用麻雀智能算法(sparrow search algorithm,SSA)构建隶属度函数,对GPC控制器中的柔化因子和加权系数进行在线分段整定,有效提升了系统性能.此外,在参数整定过程中,引入事件触发机制(event-triggered mechanism,ETM),在保证控制性能的同时,避免了不必要的控制器采样与更新,降低了系统在线计算量并减少了能源消耗.最后通过VAV空调系统仿真实验验证,证明了本文提出方法的可行性和有效性.
GPC tuning for air-conditioning systems based on fuzzy logic
For a multiple-input-multiple-output variable air volume(VAV)air-conditioning system,a generalized predictive control(GPC)parameter tuning method based on fuzzy logic and event triggering is proposed,establis-hing a fuzzy prediction model based on the output slope and new fuzzy target parameters and the Gaussian bilateral membership function to improve the membership function and better reflect the current state of the system.Subse-quently,the sparrow search algorithm(SSA)is used to construct the membership function to segmentally adjust the softening factor and weighting coefficient in the GPC controller online,which has effectively improved the system performance.Moreover,in the parameter tuning process,the event trigger mechanism is introduced to avoid unnec-essary controller sampling and update while ensuring control performance.It reduces the amount of system online calculations and energy consumption.Finally,the simulation experiment of the VAV air-conditioning system proves the feasibility and effectiveness of the proposed method.

generalized predictive controlparameter tuningfuzzy logicsoftening factorweighting coefficiente-vent trigger mechanism

贺宁、李尚、许恭博、郝文斌

展开 >

西安建筑科技大学机电工程学院,陕西西安 710055

西安交通大学智能网络与网络安全教育部重点实验室,陕西西安 710049

Monash University,Faculty of Engineering,Victoria 3800,Australia

广义预测控制 参数整定 模糊逻辑 柔化因子 加权系数 事件触发机制

国家自然科学基金中国博士后科学基金面上项目

619032912019M660257

2024

哈尔滨工程大学学报
哈尔滨工程大学

哈尔滨工程大学学报

CSTPCD北大核心
影响因子:0.655
ISSN:1006-7043
年,卷(期):2024.45(5)
  • 22