PSI - Issue 59

D. Nosov et al. / Procedia Structural Integrity 59 (2024) 656–663

659

4

D. Nosov et al. / Structural Integrity Procedia 00 (2019) 000 – 000

defects, and if the cooling rate is too high, the material may become brittle and lose impact toughness (Chenyao et al. 2019). To ensure the required strength and impact strength of the welded joint, it is important to carefully control the cooling rate during and after the welding process. This can be achieved by various means, such as adjusting the welding parameters, using preheat or post-weld heat treatment, or using a cooling medium to accelerate the cooling rate. The specific method used will depend on the welding process, the material being welded and the requirements of the application (Mitschang et al. 2013). 3. Results and Discussion A series of experiments was conducted to determine the optimal working conditions of contact welding and heat treatment of welded joints of a band saw made of 65G steel. The following factors affecting the welding result were selected: annealing time (s), heating time (s), temperature coefficient (indicator of the settings of the ASP-1600 welding machine). Indicators that were monitored: angle of destruction; number of cycles to failure; probability of destruction 0-10; relative strength 0-10; type of destruction (seam, ZTI).

а) b) Fig. 1. Band saw blades before mechanical processing of the seam (a) and after (b) Data obtained during the experiment were processed using the STATISTICA program. The results of graphical forecasting are shown below (Fig. 2-6). If it is necessary to accurately determine the parameters (prediction) of heat treatment from two factors, it’s possible to use simplified models according to the formulas given in the second line of each graph, where x and y are the corresponding factors along the coordinate axes. If it is necessary to accurately determine the parameters (prediction) of heat treatment from three factors (according to which the experimental part was carried out), it is possible to use the model according to the general equation (1). v = b1+ b2·v1+b3·v2 + b4·v3 + b5·v1·v2 + b6·v1·v3 +