Detecting Electronic Circuitboard Thermal Abnormalities with Dynamic Hypernumbers

Abstract

This paper introduces the algebra of data acquisition using dynamic hypernumbers used in predicting thermal abnormalities with a rapid forecasting algorithm. Algebraic rules are defined for monitoring data from multiple sources. Each data source is considered a dynamic receptive synchronized hypernumber for building the resulting dynamic hypernumber. The algebraic dynamic hypernumber
definition is used to present a low-complexity algorithm for identifying the thermal state of an electronic circuit board. The algorithm is implemented by translating the pixel colors from an infrared camera into a dynamic hypernumber temperature index. The dynamic hypernumber algorithm output is used to dynamically predict the thermal process. This prediction allows the detection of thermal abnormalities. It is shown that the method of building a dynamic hypernumber from the set of the circuit board temperatures has less electronic and computational complexity in comparison with looking for elevated temperature areas. Considering dynamic hypernumbers are acquired from the stochastic process, regression polynomials are used with best criteria fit theory to find the continuous mathematical expression of such numbers. The solution of non-linear operator equations for computing the coefficients of the polynomial is obtained using the constructive hypernumber.