Augmented reality and virtual reality based simulation setups for accelerated life testing

Abstract

Life testing under nominal operating conditions of mechanical parts with a high mean lifetime between failure often consumes a significant amount of time and resources, therefore rendering such procedures is expensive and impractical. As a result, the technology of accelerated life testing (ALT) has been developed to obtain life characteristics of components within a reasonable amount of time. Accelerated life testing involves testing the components at two or more accelerated levels of stress to induce failures quickly. However, the testing involved in the accelerated life testing requires the building of costly setups to perform run to failure tests on components. As these setups are costly, the setups may not be available in most of the institution. Due to the unavailability of these setups, students and researchers may not get hands-on training with the ALT setups. Another challenge associated with these setups is that ALT setups are not remotely accessible, as it requires a physical presence of an operator to perform tests. The objective of this thesis is to address these challenges associated with the building of costly setups for accelerated life testing. For this purpose, a virtual setup can be developed for accelerated life testing instead of costly setups. A simulation algorithm can be incorporated with the virtual setup to simulate the failure behavior of components under the test. Finally, these virtual setups can also be used for the remote control and monitoring of the actual setups in real-time. The virtual setup is developed in such a way that, it provides step by-step instructions regarding the experimental procedure which is to be followed for testing of components. In this way, hands-on training is provided with the virtual setup. Along with virtual setup, a generic simulation algorithm is used for the simulation of failure behavior of the component under test. The algorithm learns from actual failure data and identifies the features like failure trend, inherent noise and, abrupt jumps to generate simulated failure data. The primary purpose of the algorithm is to simulate the stochastic nature of failure involved in the failure behavior of components under test. The data generated by the simulation algorithm is then visualized in the virtual setup for the simulation of the failure behavior of a component in the VR-based setup. For the validation of the concept proposed in this thesis, a virtual setup for an accelerated life testing of shape memory alloy springs is developed. The data generation process of generic simulation is validated by generating simulated failure data for the SMA springs undergoing thermo-mechanical fatigue. Results obtained from the simulation algorithm are compared with actual failure datasets and results obtained are discussed in chapter 5. Results show the similarity between actual datasets and simulated datasets which is illustrated using graphical representations of simulated data with the original datasets along with calculation of similarity indices and performing Anova tests on actual and simulated failure data. For the remote control of the actual setup, an API (Application Programming Interface) is developed. The working procedure of this API is discussed in detail in chapter 4. The procedure for remote control of PPS is discussed for remote control of the setup. For the remote monitoring of the actual setup in real-time, the failure data is needed to be monitored continuously. For this purpose, the web publishing tool of LabView is used along with the WebView plugin of Unity. The detailed procedure for remote monitoring is discussed in chapter 4. The developed LabView application is controlled remotely using the VR-based setup for remote monitoring of the actual setup.