基于ANFIS的木质采暖地板温度场预测分析  

Prediction and Analysis of Temperature Field for Wood Flooring with Geothermal System Based on ANFIS

作　　者：褚鑫 周世玉 刘大伟 杜光月 曹正彬 刘晓平 周玉成[1] Chu Xin;Zhou Shiyu;Liu Dawei;Du Guangyue;Cao Zhengbin;Liu Xiaoping;Zhou Yucheng(School of Information and Electrical Engineering,Shandong Jianzhu University Jinan 250101;School of Thermal Engineering,Shandong Jianzhu University Jinan 250101)

机构地区：[1]山东建筑大学信息与电气工程学院,济南250101 [2]山东建筑大学热能工程学院,济南250101

出　　处：《林业科学》2018年第11期53-58,共6页Scientia Silvae Sinicae

基　　金：泰山学者优势特色学科人才团队(2015162)

摘　　要：【目的】提出基于自适应神经模糊推理系统(ANFIS)的木质地板蓄热性能温度场预测算法,为木质地板蓄热性能分析提供数据支撑。【方法】以地采暖地板蓄热性能分析仪密闭绝热圆柱腔为研究对象,在腔体内部布设由150个温度传感器组成的6层阵列,将腔体内部分为150个子空间。首先,将腔体内温度传感器序号和时间作为系统输入,传感器阵列采集的温度值作为系统输出,构建ANFIS温度场模型。然后,将所选用的训练数据输入模型,调整相应参数,完成封闭腔温度场时间模型训练。最后,将其他未参与训练数据输入到已训练好的模型中,得到预测值,并通过相应计算公式验证该方法对木质采暖地板温度场预测分析的适用性。【结果】温度场预测值拟合度达0.988以上,拟合误差也控制在较低水平,其中均方误差低于0.19%、最大相对误差低于1.22%、平均相对误差低于0.36%。【结论】基于ANFIS的封闭腔温度场预测模型能够完整表达出试验仪器腔体的温度场特征,且在建模的简化度、泛化能力和稳健性上均具有较好表现,能够用已被训练过的系统较好地对传感器任意时间点进行温度预测。【Objective】 A temperature field prediction algorithm based on ANFIS is proposed to provide data support for the performance analysis of wood floor thermal storage.【Method】 In this paper, the airtight cylindrical cavity of the self-developed floor heat storage performance analyzer is taken as the research object. There is a six-layer array of 150 temperature sensors inside the cavity. The inner space of the cavity is divided into 150 subspaces. Firstly, the temperature field model of the adaptive neuro-fuzzy inference system（ANFIS）is constructed by taking the temperature sensor number and time as the input of the system. Meanwhile, the temperature value collected by the sensor array is regard as the output. Then, the selected training data are brought into the model proposed. The corresponding parameters are adjusted to complete the training of the time model of the temperature field in the closed cavity. Finally, the other data not involved in the training are inputted into the trained model. The predicted values are obtained and the corresponding calculation formulas are used to prove that this method is suitable for the prediction and analysis of the temperature field of wood flooring with geothermal system.【Result】 The fitting degree of the predicted temperature field obtained from the experimental data is more than 0.988.The fitting error is also controlled at a lower level, in which the mean square error is less than 0.19%, the maximum relative error is below 1.22%, and the average relative error is less than 0.36%.【Conclusion】 Temperature field prediction model based on ANFIS can fully display the characteristics of temperature field in the chamber of the test instrument, and has good performance in simplification, generalization and robustness. It can predict the temperature of the sensor at any time point with the trained system.

关 键 词：地采暖地板 封闭腔 温度场 ANFIS 

分 类 号：S772[农业科学—森林工程] TU111[农业科学—林学]