地采暖木地板释热温度场的BP神经网络预测

详细信息    查看全文 | 推荐本文 |

英文篇名：Prediction of Thermal Released Field by the Wood Flooring for Ground with Heating System Based on BP Network 作者：周世玉 ; 杜光月 ; 褚鑫 ; 刘晓平 ; 周玉成 英文作者：Zhou Shiyu;Du Guangyue;Chu Xin;Liu Xiaoping;Zhou Yucheng;School of Thermal Engineering, Shandong Jianzhu University;School of Information and Electrical Engineering, Shandong Jianzhu University; 关键词：地采暖地板 ; 封闭腔 ; 温度场预测 ; 神经网络 英文关键词：wood flooring for ground with heating system;;closed cavity;;temperature field prediction;;artificial neural network 中文刊名：LYKE 英文刊名：Scientia Silvae Sinicae 机构：山东建筑大学热能工程学院;山东建筑大学信息与电气工程学院; 出版日期：2018-11-15 出版单位：林业科学 年：2018 期：v.54 基金：泰山学者优势特色学科人才团队(2015162);; 山东建筑大学校内博士基金(X18006Z) 语种：中文; 页：LYKE201811023 页数：6 CN：11 ISSN：11-1908/S 分类号：161-166

摘要

【目的】基于测试获取的有限点温度数据,采用BP神经网络对封闭腔内部时间和空间维度的温度场进行预测,为地采暖地板蓄热性能的分析计算提供理论和数据支撑。【方法】基于课题组自行研制的地采暖地板释热温度场测试设备,获取测试腔体内部测点的温度数据,并划分为神经网络训练集和测试集:时间维度上,将每个测点的三维空间坐标和该测点前3个时间节点的温度值作为输入,各个测点第4个时间节点的温度值作为输出,其中,前80组作为训练集,后28组作为测试集;空间维度上,均匀选出总量4/5测温点的数据作为训练集,剩余1/5测温点的数据作为测试集。基于训练集,分别建立时间和空间维度的BP神经网络模型,并由测试集完成对模型的验证。【结果】时间维度上,平均相对误差(MRE)=0.269 2%,最大相对误差(MAE)=5.916 0%,均方误差(MSE)=0.422 4%,拟合度(R~2)=0.998 7;空间维度上, MRE=0.364 2%,MAE=4.781 8%,MSE=0.521 9%,R~2=0.998 5。【结论】BP神经网络方法预测结果可信度较高,可有效获取地采暖地板检测腔体内部连续完整的温度场,为后续地采暖地板蓄热性能的分析计算提供理论和数据支撑。
        【Objective】 Based on the temperature data of the finite points obtained from the test, BP neural network is used to predict and analyze the temperature field in the closed cavity in the time and space dimensions, so as to provide the theoretical and data support for the thermal storage performance analysis of wood floor.【Method】 Based on the equipment for testing the thermal released field by the wood floor which was developed by the author's research group, the temperature data in the closed cavity is acquired and divided into training and testing sets of neural network. In the time dimension, the coordinates and the temperature values of the first three time nodes are taken as inputs and the temperature value of the fourth time node is defined as the output. In this process, the previous 80 sets of temperature data are defined as training set while the next 28 sets of temperature data are defined as testing set. In the space dimension, temperature data of 4/5 of the total sensors are chosen as training set while the remaining 1/5 of the total sensors are defined as testing set. The BP networks of time and space are constructed based on the training set, respectively. Furthermore, the constructed models could be validated according to the testing set.【Result】 In the time dimension, the computed errors and R~2 are as following: mean relative error(MRE)=0.269 2%,maximum relative error(MAE)=5.916 0%,mean square error(MSE)=0.422 4%,fitting degree(R~2)=0.998 7. In the space dimension, MRE=0.364 2%,MAE=4.781 8%,MSE=0.521 9%,R~2=0.998 5.【Conclusion】 The prediction result derived by BP neural network are adequately reliable, demonstrating that this method can effectively obtain the continuous and complete temperature field inside the measurement cavity of the wood floor, thus a new theoretical support is provided for the analysis and calculation of the thermal storage performance of the wood floor.

引文

刘鲭洁,陈桂明,刘小方,等. 2010. BP神经网络权重和阈值初始化方法研究. 西南师范大学学报: 自然科学版,35(6): 137-141.(Liu Q J, Chen G M, Liu X F, et al. 2010. Research on initialization algorithms of weights and biases of BP neural network. Journal of Southwest China Normal University: Natural Science Edition,35(6): 137-141. [in Chinese])
    张 伟, 朱家玲, 苗常海. 2005.低温地板采暖与散热器采暖效果的对比分析. 太阳能学报, 26(3): 304-307.(Zhang W, Zhu J L, Miao C H. 2005.The experiments of radiator heating and floor radiation in heating system. Acta Energiae Solaris Sinica,26(3): 304-307.[in Chinese])
    张晓伟, 由世俊, 张 欢, 等. 2013. 低温辐射地板采暖模拟及实验研究. 太阳能学报, 34(4): 684-688.(Zhang X W,You S J,Zhang H,et al. 2013. Experimental and simulative research on low temperature floor heating. Acta Energiae Solaris Sinica, 34(4): 684-688. [in Chinese])
    张 敏, 晏 刚, 陶 锴. 2010. 内置发热体的封闭方腔自然对流换热数值模拟. 化工学报, 61(6): 1373-1378.(Zhang M,Yan G,Tao K. 2010. Numerical simulation of natural convection in rectangular cavities with a heater of variable dimension. CIESC Journal, 61(6): 1373-1378. [in Chinese])
    Ahamad S I, Balaji C. 2016. Inverse conjugate mixed convection in a vertical substrate with protruding heat sources: a combined experimental and numerical study. Heat & Mass Transfer, 52(6): 1243-1254.
    Atay1lmaz S ?, Demir H, A?ra ?. 2010. Application of artificial neural networks for prediction of natural convection from a heated horizontal cylinder. International Communications in Heat and Mass Transfer, 37(1): 68-73.
    Ben-Nakhi A E, Mahmoud M A, Mahmoud A M. 2008. Inter-model comparison of CFD and neural network analysis of natural convection heat transfer in a partitioned enclosure. Applied Mathematical Modelling, 32(9): 1834-1847.
    Calcagni B, Marsili F, Paroncini M. 2005. Natural convective heat transfer in square enclosures heated from below. Applied Thermal Engineering, 25(16): 2522-2531.
    Das D, Roy M, Basak T. 2017. Studies on natural convection within enclosures of various(non-square)shapes-a review. International Journal of Heat and Mass Transfer, 106: 356-406.
    Mahmoud M A, Ben-Nakhi A E. 2007. Neural networks analysis of free laminar convection heat transfer in a partitioned enclosure. Communications in Nonlinear Science and Numerical Simulation, 12(7): 1265-1276.
    Montana D J, Davis L. 1989. Training feed-forward neutral networks using genetic algorithms. Proceeding of the International Joint Conference on Artificial Intelligence, Los Altos, 762-767.
    Oke S A. 2008. A literature review on artificial intelligence. International Journal of Information and Management Sciences, 19(4): 535-570.