Abstract
The cutting force and cutting temperature have a significant impact on the service life and durability of gear skiving cutters. Due to unreasonable design, the existing process parameters lead to dramatically nonuniform cutting force and cutting temperature, which aggravates the rapid wear of gear skiving cutters. To address this issue, this paper first establishes a finite element model of skiving the internal circular arc tooth in pin wheel housing, and the simulation model is simplified to improve computation efficiency. Next, the impact of single process parameter on cutting force and cutting temperature is analyzed by controlling variable. Then, an orthogonal experiment is designed and the method of range analysis is employed to evaluate the significance of each process parameter. Furthermore, a prediction model of cutting force and cutting temperature is established using a neural network optimized by genetic algorithm. This prediction model allows for the construction of a multi-objective optimization model for the process parameters. By solving this model, the optimal combination of process parameters within the given ranges can be obtained to achieve reasonable and balanced cutting force and cutting temperature.