Methods of data mining for modelling of low-pressure heat treatment

Abstract

Purpose: This paper addresses the methods of the modelling of thermal and thermochemical processes used in computer-aided design, optimization and control of processes of thermal and thermochemical treatment in terms of obtaining real-time results of the calculations, which allows for observation of how an item changes during its treatment to respond immediately and to determine the parameters of a corrective process should any irregularities be detected. The main goal of the literature review was to develop a methodology for the design of functional and effective low-pressure processes of thermal and thermochemical treatments using effective calculation methods. Design/methodology/approach: A detailed analysis was conducted regarding the modelling methods with low-pressure carburizing and low-pressure nitriding. Findings: It was found the following criteria of methods selection of heat treatment modelling should be applied: data quality, data quantity, implementation speed, expected relationship complexity, economic and rational factors. Practical implications: Because of its non-equilibrium nature and transient states in the course of the processes computational support is particularly required in low-pressure thermochemical treatments. The primary goal of the simulation is to predict the course of the process and the final properties of the product, what ensures the repeatability of the process results. Originality/value: It was presented a synthetic presentation of modelling methods, in particular methods of artificial intelligence; it was also analysed the possibilities and risks associated with methods.