Welcome to visit Scientia Silvae Sinicae,Today is

Scientia Silvae Sinicae ›› 2018, Vol. 54 ›› Issue (11): 53-58.doi: 10.11707/j.1001-7488.20181108

Previous Articles     Next Articles

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

Chu Xin1, Zhou Shiyu2, Liu Dawei1, Du Guangyue1, Cao Zhengbin1, Liu Xiaoping1, Zhou Yucheng1   

  1. 1. School of Information and Electrical Engineering, Shandong Jianzhu University Jinan 250101;
    2. School of Thermal Engineering, Shandong Jianzhu University Jinan 250101
  • Received:2018-04-02 Revised:2018-06-15 Online:2018-11-25 Published:2018-12-04

Abstract: [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.

Key words: floor with geothermal system, closed chamber, temperature field, ANFIS

CLC Number: