Development of predictive mathematical models of the quality of radio electronic equipment based on the results of autonomous tests

The article discusses the methodology for developing a predictive model (forecasting operator) of the quality of onboard equipment using the extrapolation method. It is shown that the most efficient information about the quality and reliability of the equipment can be obtained from the results of autonomous tests. The choice of the test object was made. A microprocessor temperature controller was chosen as the object of autonomous tests. The transition resistance between the electrical circuits of the microprocessor temperature controller was chosen as a predicted parameter. The results of the training experiment are presented. To construct the forecasting operator, quasi-deterministic models of linear, logarithmic, exponential, and parabolic forms are used. When developing the models, the normalization of predicted parameter by mathematical expectation was used. The choice of predictive models was based on the criteria of minimum average variance, calculated at the test time points, and the minimum values of the probability of erroneous decisions and consumer risk. The research of the developed operator was carried out, probabilistic characteristics of its efficiency are obtained.