Year of Graduation
Development of Neural Network for Reliability Assessment of Electronic Equipment
In this paper, we create a neural network model to assess the reliability of complex electronic equipment, the secondary power supply unit. The reliability of the equipment is shown by its failure rate under different operating conditions. Manufacturers use a constant value of failure rate to assess the reliability of the equipment. At the same time, evaluation of reliability parameters for different operating conditions requires large amounts of data to be stored in tables. A multi-layer perceptron paradigm is used in the model. The method presented in this paper allows to evaluate the failure rate for different operating conditions of electronic equipment with the minimal number of parameters.