Application of decision–making parametric structures in the analysis of a compound planetary gear

The design process of planetary gears can be divided into two main stages, i.e. conceptual design and construction, that is the preparation of construction documentation. In the first stage, the constructor has a very limited number of project data. In the case of a reduction or multiplier transmission these are usually output parameters (torque and speed), characteristics of the driving and the driven machine, Therefore, the stage of gearbox design must be supported by preliminary calculations, which supports the selection of the correct technical solution, leading to the stage of the structural design preparation. In the method of systematic search, we used an algorithm for generating induction decision trees, based on entropy growth as a method related to machine learning. In the next step, the decision-making parametric structures were employed. In the future, the presented solution will allow further analyses and syntheses such as checking the isomorphism of the proposed solutions, determining the validity of constructions and/or operating parameters of the analysed gears.