分户热计量系统室温模糊预测控制

Fuzzy Predictive Control of Room Temperature in a Household Heat Measurement System

研究室温控制问题时,针对分户热计量系统中室温控制具有滞后性、时变性、复杂性,难以建立精确数学模型的特点,为了按需要改变室温,提出一种室温模糊预测控制策略。采用BP网络建立预测模型,对采暖房间升(降)一定温度所需要的时间进行预测,并通过模糊控器对采暖房间温度进行控制,克服了传统室温控制方式不能对升(降)温时间进行预测的缺点,并进行仿真。仿真结果表明,预测模型的时间误差小,预测控制策略具有适应能力、鲁棒性好等特点。

In a household heat measurement system,a precise mathematical model for the control of room temperature can not be easily established with its time-delay,time-varying and complexity.A predictive control strategy for room temperature is proposed in this paper.The BP network is not only used to establish a predictive model for predicting the time of increasing（reducing） certain temperature for a heating room,but also used to control the heating room temperature through the fuzzy controller.The method can overcome the weakness of traditional room temperature control modes which can not be used to predict the time of increasing（reducing） temperature.The simulation results show that the time error of the predictive model is small,and the predictive control strategy has the characteristics of strong robustness and self-adaptive ability.