Measurement of Localized Dynamic Gear Meshing Forces Based on Distributed Root Stress

The acquisition of gear dynamic meshing force is helpful to study the dynamic response of gear system and can also guide the strength design and vibration reduction design of gear transmission system. At present, the acquisition of gear dynamic meshing force is mainly based on the calculation or simulation of physical model, and there is a lack of effective experimental measurement methods in practical application. In view of the shortcomings of the traditional strain/stress signal method, a local dynamic meshing force measurement method based on distributed stress is proposed, which overcomes the shortcomings of perforated wiring, calculation of load distribution coefficient, and nonlinearity. The mapping relationship between meshing force, meshing position, and single-tooth root stress is changed to the mapping relationship between meshing force and distributed tooth root stress. First, an experimental platform was built to collect distributed root stresses under different torques under quasistatic conditions, which was used to train radial basis function neural network (RBFNN) models. Finally, the trained model is used to measure the dynamic meshing force under any working condition. The results show that the measured dynamic meshing force can effectively reflect the average meshing force level, and the dynamic behavior is similar to the dynamic simulation results.