Two methods are suggested of solving the inverse problem of thermal control. The first method is based on an approximation of the dependences of the useful signal on the thickness and depth of occurrence of flaws; in the unidimensional problem it ensures a defectometric error of 5-30%. The second method is based on the exact solution of the inverse problem by means of its regularization, and it is characterized by an error of 0.3-10%.
|Number of pages||4|
|Journal||The Soviet journal of nondestructive testing|
|Publication status||Published - Dec 1989|
ASJC Scopus subject areas