Tailored heat treatment through a digitalized process chain

Important results from the project

Problems in the heat treatment industry are often very complex where all prehistory, all steps and all parameters affect the result. The purpose of the project is to take a holistic approach to the process chain and apply today´s available technologies for digitalization. The effect goal for the project is increased quality, smoother process, shorter process time and reduced scrapping for heat treatment processes through digitalization. Goal fulfillment has been good and the potential to achieve the impact goal during implementation is considered to be very good.

Expected long term effects

The project has shown that by controlling the process by the measured CO content, higher quality and smoother process can be obtained. Case depths can be predicted with an error that is of the same order of magnitude as that in case depth measurements. The results show that today´s processes are quite optimized. The processes are controlled with the help of feedback loops and this means that correlations are cancelled out, which limits the variable space for machine learning. The processes, on the other hand, proved to be suitable for physical simulations linked with measured process data.

Approach and implementation

The project has used the methodology for digitization that was developed in a previous project. New gas analysis equipment has been installed. The companies have then logged their processes carefully, including times, temperatures and gas compositions. This in combination with measured hardening results (hardness, microstructure, case depth, etc.) has constituted the datasets that are then cleaned and pre-treated before the modeling of the process (causal relationships, machine learning, simulation, statistical model, logical model, etc).