Search

Preventive maintenance of rotating machinery, such as motors, is pivotal for cost reduction in manufacturing facilities. Condition-based maintenance is considered the driving force in the health management of industrial machinery, which necessitates the development of efficient, able-to-generalize, and reusable approaches to help manufacturing facilities identify operational conditions. This study proposes an approach for classifying the operational settings of motors based on vibration data. This was accomplished by first creating a merged class of weight distribution and speed percentage ranges to be classified. These two operational settings can be classified to address sustainable manufacturing concerns as they relate to the development of systematic maintenance plans. Ensemble learning was then used as the machine learning algorithm of choice for the classification task. It was found that ensemble learning is capable of producing reasonable classification accuracy results that are superior to those of single-model approaches. In addition, the testing size and window size of the segmented vibration signals have been identified to have a clear effect on classification accuracy that is considered consistent with the logical trend. The proposed approach addresses the need for manufacturing facilities to utilize efficient and accurate algorithms for cost-effective, and sustainable maintenance practices.

Through the design process of automotive composite structures, the designer needs to meet the mechanical characteristics and consider the sensitivity of the design to the environmental factors, which control its safety and feasibility. The current experimental study presents the effect of thermal aging on the crashworthiness of carbon/epoxy composite tubes under lateral crushing. Tubes were subjected to thermal aging at 70 and 100 °C for 14 days. Both empty and foam-filled tubes were studied. In addition, foam specimens were compressed under the same conditions to understand the behavior of the foam-filled tubes. The outcomes demonstrated that the addition of the foam improved the load carrying capacity and energy absorption. The matrix's thermal degradation and debonding were the primary factors in the reduction of crashworthiness. For the foam-filled composites, the reduction in the crashworthiness is lower because of having the foam inside the tube. The average crushing force increases by 14% and 29% for the specimens exposed to 70 and 100 °C for 14 days, respectively, which result in higher energy absorption. The improvement in the foam-filled structures by aging is governed by the higher strength and energy absorbed of the foam after thermal exposure, which supports the use of polyurethane foam as a filler in energy absorption devices.

Novel designs of porous and semi-porous hip stems attempt to alleviate complications such as aseptic loosening, stress shielding, and eventual implant failure. Various designs of hip stems are modeled to simulate biomechanical performance using finite element analysis; however, these models are computationally expensive. Therefore, the machine learning approach is incorporated with simulated data to predict the new biomechanical performance of new designs of hip stems. Six types of algorithms based on machine learning were employed to validate the simulated results of finite element analysis. Afterwards, new designs of semi-porous stems with outer dense layers of 2.5 and 3 mm and porosities of 10–80% were used to predict the stiffness of the stems, stresses in outer dense layers, stresses in porous sections, and factor of safety under physiological loads using machine learning algorithms. It was determined that decision tree regression is the top-performing machine learning algorithm as per the used simulation data in terms of the validation mean absolute percentage error which equals 19.62%. It was also found that ridge regression produces the most consistent test set trend as compared with the original simulated finite element analysis results despite relying on a relatively small data set. These predicted results employing trained algorithms provided the understanding that changing the design parameters of semi-porous stems affects the biomechanical performance without carrying out finite element analysis.