Model based tampering for improved process performance application to surface grinding of shafts

In the recent years, industries have been striving to achieve increasingly high levels of quality and performance from manufacturing process. A Model Based Tampering (MBT) methodology has been developed in this research work, by which the performance of many manufacturing operations can be improved significantly. The MBT methodology involves characterizing the dynamics of the process through experimental and data analysis; developing mathematical models to accurately represent the process dynamics; and using these models to dynamically compensate for the effects of process degradation on the performance. This research has focused on the issues pertaining to characterization and model development with specific application to cylindrical grinding of shafts in the aerospace industry.; Most manufacturing processes are nonlinear and stochastic in nature and to accurately represent these processes we need to capture both these behaviors. Nonlinear Stochastic Differential Equations (n-SDEs) can effectively serve this purpose. Earlier research has used simple representations of these processes due to the computation complexities involved in n-SDEs. With the advances in computing power the use of n-SDEs for process modeling and control has become feasible for further research exploration, This research work has developed a method to derive n-SDE models of processes from actual data, This forms the critical step of the overall MBT methodology.; The main issue in aerospace grinding processes is to improve performance, measured in terms of surface finish, during cylindrical grinding of shafts. For example, surface finish is a key determinant of the performance of Air Bearing shafts, which is a critical component of aircraft Environmental Control System (ECS). ECS is one of the core competency products of Honeywell. The MBT methodology was validated using an R&D machine at this company. The validation experiments show that MBT provided an improvement in process performance (i.e., reduction in surface finish variation) by 62% over the no tamper baseline, which is the current practice. This makes MBT a powerful closed loop control technology for the manufacturing community.