Calculation of transient temperature of three-core cables based on BP neural network trained by particle swarm optimization algorithm

The conductor temperature is an important state indicator of an operation cable,. But direct measurement of the cable conductor temperature is difficult to realize in practice, which is usually obtained by calculation. Presently, thermal circuit method and numerical calculation are adopted to calculate the conductor temperature, however the accuracy of the two methods are greater affected by physical parameters of the cable itself. Therefore, a method of calculation of three-core cable conductor transient temperature by using BP neural network based on particle swarm optimization is introduced in this paper. A four-layer neural network model is built by using the load current and outer sheath temperature as inputs, using cable conductor temperature as output. Particle swarm optimization algorithm is used to optimize the network first, and then BP algorithm is used to make a second optimization. The research shows that BP neural network based on particle swarm optimization can be used in accurately calculating three-core cable conductor temperature, avoiding of the effecting of its physical parameters, moreover, it also can give a reference on monitoring the state of operating three-core cable.