Selective quality controlmetallic functional materials based on cognitive analysis of microships photos

The following problem is considered here: the of selective control of quality of metallic functional materials that must have strictly defined and adjustable physical and chemical properties that make it possible to realize the service characteristics of structures and devices that are unattainable in usage of other materials. It is proposed to check the correspondence of the obtained consumer properties of the created material to the specified values by comparing the microstructure parameters determined from photographs of the sections with the usage of an automated information system. It is the verification of the microstructure that is the distinguishing feature of the production of functional materials. Verification of their properties at the macro level is done only to confirm the results of the analysis of the microstructure in order to increase the reliability of the analysis carried out. The considered decision support system is an information and analytical one, and is intended for informational support of the quality control procedure at the level of recognition and calculation of image parameters from photos of microsections, as well as intellectual support of the procedure for quality control of materials at the level of formation and application of a specialized knowledge base with the possibility to explain the obtained logical conclusions. The parameters of the phase composition (volume fractions of phases) and the grain structure (dimensions and degree of anisotropy of grains) of the material under investigation serve as controlled quantities. It is assumed that the construction of metrics to compare the complex assessment of the phase composition and the complex evaluation of the grain structure is carried out by an expert in material science at the stage of building an intellectual system during the development of a technological process for the production of functional material in the form of rules that are stored in the expert system. Further, the obtained metrics will be used at the enterprises for the production of functional materials to monitor its quality. The decision on the quality of the produced functional material is a multicriterial one, since it is made from the combination of the results of simultaneous comparison of a whole series of properties of the produced material with the criteria formulated by its developers. A demo sample of selective quality control of reinforcing steel with specified strength properties obtained by heat treatment of the material is given. It is shown that by comparing the values of the 3 parameters of the grain-phase structure it is possible to control the value of the hardness of the steel.