On the accuracy of numerical methods for solving the Volterra equations of the kind in the inverse heat conduction problems

The article is devoted to the study of the accuracy of methods for solving the measurement challenge that arises when determining the temperature inside an object subjected to the influence of an external control thermal effect. The approach to the construction of a numerical solution of the measurement problem associated with the problem of determining temperature is based on reducing the initial problem to solving an integral equation that are characterized the direct dependence of temperature on the measured values. The integral equation is obtained using the direct and inverse Laplace transforms with the involvement of the regularizing approach and the regularization teqhiqie. The resulting integral equation is the Volterra equations of the first kind of convolution type with a specific kernel. In this paper, we investigate the accuracy of numerical methods for solving an integral equation with a specific kernel from the point of view of the mechanisms for the implementation of machine arithmetic. Computing method schemas are based on the product integration method, squaring the middle rectangles. The article also presents the results of a study of the error of the computational scheme of the order-optimal method based on the application of Fourier transforms and the projection regularization method. The method is used to directly solve the original problem without redusing it to an integral model and allows one to obtain numerical solutions with guaranteed accuracy. In order to obtain experimental estimates of the accuracy of numerical methods and a comparative analysis of the machine accuracy of the integral approximation methods and the order-optimal method, a computational experiment was carry out. The experimental results indicate that it is possible to obtain the numerical solutions of the measurement challenge with a high level of accuracy.