储水式电热水器的剩余洗浴时间预报系统

Residual Bath Time Forecast System of Storage Water Heater

本文根据实际经验与理论分析.提出剩余洗浴时间硬件系统,并在剩余洗浴时间预报硬件系统,采集大量的数据,基于BP神经网络建立剩余洗浴时间预报模型,预报误差在5 min以内。在此基础上,结合非出水阶段特点,提出模糊自学习用户习惯的算法,预报非出水阶段的剩余洗浴时间,并通过数据验证其有效性;最后,针对洗浴不断电的工况下剩余洗浴时间的计算,推导并发现其等比数列规律,最终化简计算模型,得到的简化模型与实际数据相吻合。实验研究表明,本文提出的一种剩余洗浴时间预报系统,针对用户使用的不同工况,能在电热水器中实现较为全面和准确地预报电热水器剩余洗浴时间,非出水阶段预报剩余洗浴时间误差最大6~7 min,出水阶段误差在3 min以内。

This paper proposed a hardware system of residual bath time based on practical experience and theoretical analysis.In addition,a large amount of data was collected in the residual bath time forecast hardware system,and the residual bath time prediction model was established based on BP neural network,and the forecast error was within 5 minutes.On the basis of this,combining with the characteristics of non-effluent stage,this paper puts forward the algorithm of fuzzy self-learning user habit,predicts the remaining bath time in the non-effluent stage,and verifies its validity through data.Finally,according to the calculation of residual bath time under the condition of continuous electric bath,the law of geometric sequence is derived and found,which simplified calculation model,and the simplified model is consistent with the actual data.Research shows that,the residual bath time forecasting system can forecast the residual bath time comprehensively and accurately according to users different operating modes when using the electric water heater,that is the maximum error of residual bath time in the non effluent stage is 6~7 minutes,and the error in the effluent stage is less than 3 minutes.