Adaptive Neuro-Fuzzy Inference System for Thermal Field Evaluation of Underground Cable System  

作　　者：Mamdooh S. AI-Saud 

机构地区：[1]Electrical Engineering Department, King Saud University, Riyadh 11421, Saudi Arabia

出　　处：《Journal of Energy and Power Engineering》2012年第10期1643-1650,共8页能源与动力工程（美国大卫英文）

摘　　要：The influence of thermal circuit parameters on a buried underground cable is investigated using an ANFIS （adaptive neuro-fuzzy inference system）. Finite element solution of the heat conduction equation is used, combined with artificial intelligence methods. The cable temperature depends on several parameters, such as the ambient temperature, the currents flowing through the conductor and the resistivity of the surrounding soil. In this paper, ANFIS is used to simulate the problem of the thermal field of underground cables under various parameters variation and climatic conditions. The developed model was trained using data generated from FEM （finite element method） for different configurations （training set） of the thermal field problem. After training, the system is tested for several scenarios, differing significantly from the training cases. It is shown that the proposed method is very time efficient and accurate in calculating the thermal fields compared to the relatively time consuming finite element method; thus ANFIS provides a potential computationally efficient and inexpensive predictive tool for more effective thermal design of underground cable systems.

关 键 词：Underground cables AMPACITY thermal analysis finite element method adaptive neuro-fuzzy inference system. 

分 类 号：TP273.4[自动化与计算机技术—检测技术与自动化装置] TN943.6[自动化与计算机技术—控制科学与工程]