Deep learning has been employed to prognostic and health management of automotive and aerospace with promising results. Literature in this area has revealed that most contributions regarding deep learning is largely focused on the model’s architecture. However, contributions regarding improvement of different aspects in deep learning, such as custom loss function for prognostic and health management are scarce. There is therefore an opportunity to improve upon the effectiveness of deep learning for the system’s prognostics and diagnostics without modifying the models’ architecture. To address this gap, the use of two different dynamically weighted loss functions, a newly proposed weighting mechanism and a focal loss function for prognostics and diagnostics task are investigated. A dynamically weighted loss function is expected to modify the learning process by augmenting the loss function with a weight value corresponding to the learning error of each data instance. The objective is to force deep learning models to focus on those instances where larger learning errors occur in order to improve their performance. The two loss functions used are evaluated using four popular deep learning architectures, namely, deep feedforward neural network, one-dimensional convolutional neural network, bidirectional gated recurrent unit and bidirectional long short-term memory on the commercial modular aero-propulsion system simulation data from NASA and air pressure system failure data for Scania trucks. Experimental results show that dynamically-weighted loss functions helps us achieve significant improvement for remaining useful life prediction and fault detection rate over non-weighted loss function predictions.

When machine learning supports decision-making in safety-critical systems, it is important to verify and understand the reasons why a particular output is produced. Although feature importance calculation approaches assist in interpretation, there is a lack of consensus regarding how features’ importance is quantified, which makes the explanations offered for the outcomes mostly unreliable. A possible solution to address the lack of agreement is to combine the results from multiple feature importance quantifiers to reduce the variance in estimates and to improve the quality of explanations. Our hypothesis is that this leads to more robust and trustworthy explanations of the contribution of each feature to machine learning predictions. To test this hypothesis, we propose an extensible model-agnostic framework divided in four main parts: (i) traditional data pre-processing and preparation for predictive machine learning models, (ii) predictive machine learning, (iii) feature importance quantification, and (iv) feature importance decision fusion using an ensemble strategy. Our approach is tested on synthetic data, where the ground truth is known. We compare different fusion approaches and their results for both training and test sets. We also investigate how different characteristics within the datasets affect the quality of the feature importance ensembles studied. The results show that, overall, our feature importance ensemble framework produces 15% less feature importance errors compared with existing methods. Additionally, the results reveal that different levels of noise in the datasets do not affect the feature importance ensembles’ ability to accurately quantify feature importance, whereas the feature importance quantification error increases with the number of features and number of orthogonal informative features. We also discuss the implications of our findings on the quality of explanations provided to safety-critical systems.

There is an increasing need for the use of additive manufacturing (AM) to produce improved critical application engineering components. However, the materials manufactured using AM perform well below their traditionally manufactured counterparts, particularly for creep and fatigue. Research has shown that this difference in performance is due to the complex relationships between AM process parameters which affect the material microstructure and consequently the mechanical performance as well. Therefore, it is necessary to understand the impact of different AM build parameters on the mechanical performance of parts. Machine learning (ML) models are able to find hidden relationships in data using iterative statistical analyses and have the potential to develop process–structure–property–performance relationships for manufacturing processes, including AM. The aim of this work is to apply ML techniques to materials testing data in order to understand the effect of AM process parameters on the creep rate of additively built nickel-based superalloy and to predict the creep rate of the material from these process parameters. In this work, the predictive capabilities of ML and its ability to develop process–structure–property relationships are applied to the creep properties of laser powder bed fused alloy 718. The input data for the ML model included the Laser Powder Bed Fusion (LPBF) build parameters used—build orientation, scan strategy and number of lasers—and geometrical material descriptors which were extracted from optical microscope porosity images using image analysis techniques. The ML model was used to predict the minimum creep rate of the Laser Powder Bed Fused alloy 718 samples, which had been creep tested at $$650\,^\circ $$650 ∘ C and 600 MPa. The ML model was also used to identify the most relevant material descriptors affecting the minimum creep rate of the material (determined by using an ensemble feature importance framework). The creep rate was accurately predicted with a percentage error of $$1.40\%$$1.40 % in the best case. The most important material descriptors were found to be part density, number of pores, build orientation and scan strategy. These findings show the applicability and potential of using ML to determine and predict the mechanical properties of materials fabricated via different manufacturing processes, and to find process–structure–property relationships in AM. This increases the readiness of AM for use in critical applications.

Asymmetric loss functions have been successfully applied to deep learning for image analysis and imbalanced classification. In this paper, we extend the use of particular types of weighted loss functions, namely asymmetric loss functions, to investigate how predictions of engine remaining useful life (RUL) in aerospace are affected. Within prognostics and health management, the main metric used to evaluate deep learning RUL predictions is the scoring function. Our hypothesis is that by using asymmetric loss functions we will improve results for this metric. In order to investigate our hypothesis, we test 4 different asymmetric loss functions, i.e, Mean Square Logarithmic Error-Mean Square Error, Linear-Mean Square Error, Linear-Linear, and Quadratic-Quadratic and evaluate whether and how much they affect different deep learning architectures performance. Results show that the use of asymmetric loss functions improve RUL predictions for the case study investigated.

Designing effective thermal management systems within transmission systems requires simulations to consider the contributions from phenomena such as hydrodynamic lubrication regions. Computational fluid dynamics (CFD) remains computationally expensive for practical cases of hydrodynamic lubrication while the thermo elasto-hydrodynamic lubrication (TEHL) theory has demonstrated good accuracy at a lower computational cost. To account for the effects of hydrodynamic lubrication in high-power transmission systems requires integrating TEHL into a CFD framework such that these methodologies can be interfaced. This study takes an initial step by developing a TEHL solver within OpenFOAM such that the program is prepared to be interfaced with a CFD module in future versions. The OpenFOAM solver includes the Elrod–Adams cavitation model, thermal effects, and elastic deformation of the surfaces, and considers mixing between the recirculating flow and oil feed by applying energy and mass continuity. A sensitivity study of the film mesh is presented to show the solution variation with refinement along the circumferential, axial and radial directions. A validation case is presented of an experimental single axial groove journal bearing which shows good agreement in the pressure and temperature results. The peak pressure in the film is predicted within 12% and the peak temperature in the bush is predicted within 5% when comparing the centerline profiles.