Predicting Insulation Resistance of Enamelled Wire using Neural Network and Curve Fit Methods Under Thermal Aging

Abstract

Health monitoring has gained a massive interest in power systems engineering, as it has the advantage to reduce operating costs, improve reliability of power supply and provide a better service to customers. This paper presents surrogate methods to predict the electrical insulation lifetime using the neural network approach and three curve fitting models. These can be used for the health monitoring of insulating systems in electrical equipment, such as motors, generators and transformers. The curve fit models and the supervised back propagation neural network are employed to predict the insulation resistance trend of enameled copper wires, when stressed with a temperature of 290 ˚C. After selecting a suitable end of life criterion, the specimens' mean time-to-failure is estimated, and the performance of each of the analyzed models is apprised through a comparison with the standard method for thermal life evaluation of enameled wires. Amongst all, the best prediction accuracy is achieved by a Backpropagation neural network approach, which gives an error of just 3.29% when compared with the conventional life evaluation method, whereas, the error is above 10% for all the three investigated curve fit models.

Lifetime Estimation of Enameled Wires Under Accelerated Thermal Aging Using Curve Fitting Methods

“…These inputs are propagated through a neuron which takes a linear combination of the inputs and transforms them through the sigmoid activation function. The weighted value from a bias node is represented by θ with an output value of 1 [14], [16]. An appropriate activation function (i.e.…”

“…The exponential function, (6), is generally used when the rate of change of a parameter is proportional to the initial amount of the quantity. If the coefficient B is negative, the function exponentially falls, whereas, if coefficient B is positive, then y represents exponential growth [14].…”

“…dissipation factor (Tanδ), Insulation Resistance (IR), Insulation Capacitance (IC), tensile strength, elongation, breakdown voltage etc. ), through different NN architectures such as Radial Basis Function Network (RBFN), Backpropagation (BP), Long Short-Term Memory (LSTM), and curve fitting methods [5], [6], [10]- [14]. In [10], the tensile strength and elongation of XLPE cables were predicted with a RBFN using the Random Optimisation method as its training algorithm.…”

“…This method has shown to result in an overall time-savings of about 1560 hours which is about 68% for the presented case-study. In addition, Gulrukh Turabee et al [14] have introduced a comparative study of three CF approaches at 290 • C aging temperature by predicting IR, under thermal stress, which is further explored in this paper.…”

Thermal Lifetime Evaluation of Electrical Machines Using Neural Network

“…Transmission of input signals takes places from input layer in turn through the hidden layer nodes, and reaches the final output nodes at the last. Hence, the output of next layer is only influenced by the nodes of previous layer [22,23]. Having advantage of a reliability and simple structure, this neural network is an outstanding tool for modelling of complex systems.…”