Combining the active control of gear vibration with condition monitoring

This thesis is concerned with the combination of active vibration control and condition monitoring of meshing gears. These are related to each other because they use the same vibration signals from sensors and implemented in the same gearing system. Compared with previous work done on the active control of gear vibration in which most control strategies were applied at structural interfaces, for examples at bearing or supporting structures connected to the gearbox, the method suggested in this thesis appears to be a more direct control approach to reduce the gear vibration. The effects of active vibration control on the condition monitoring of gearing systems are also investigated.A SDOF model of a pair of meshing gears is first established to provide a simple model that captures the fundamental dynamics of a pair of meshing gears. Some up-to-date techniques for condition monitoring of gearing systems are reviewed and summarised. The required secondary force to suppress the gear vibration is determined from the simple SDOF model, and also the effects of the applied secondary force on the original system are discussed. Essential components for the active control of gear vibration, such as the configurations of sensors and actuators fixed on the rotating gear and the adaptive control algorithm are investigated. This is followed by some simulations to examine the feasibility of the proposed method to combine the active control and condition monitoring systems. The experimental work carried out validating the models is presented and the results obtained on a pair of spur gears confirm the predictions.Experimental results show the advantages of directly measuring the vibration signals from the rotating gear for condition monitoring of meshing gears, and also the advantages of controlling the gear vibration by applying the secondary force directly to the rotating gear. Moreover, simulations show that the active vibration control in this combined method is effective and works in a way similar to the adaptive noise cancellation in reducing some unwanted 'noise' which helps to detect a fault seeded on one pinion tooth.