The performance and confidence in fault detection and diagnostic systems can be undermined by data pipelines that feature multiple compounding sources of uncertainty. These issues further inhibit the deployment of data-based analytics in industry, where variable data quality and lack of confidence in model outputs are already barriers to their adoption. The methodology proposed in this paper supports trustworthy data pipeline design and leverages knowledge gained from one fully-observed data pipeline to a similar, under-observed case. The transfer of uncertainties provides insight into uncertainty drivers without repeating the computational or cost overhead of fully redesigning the pipeline. A SHAP-based human-readable explainable AI (XAI) framework was used to rank and explain the impact of each choice in a data pipeline, allowing the decoupling of positive and negative performance drivers to facilitate the successful selection of highly-performing pipelines. This empirical approach is demonstrated in bearing fault classification case studies using well-understood open-source data.

Transfer learning has been highlighted as a promising framework to increase the accuracy of the data-driven model in the case of data sparsity, specifically by leveraging pretrained knowledge to the training of the target model. The objective of this study is to evaluate whether the number of requisite training samples can be reduced with the use of various transfer learning models for predicting, for example, the chemical source terms of the data-driven reduced-order modeling (ROM) that represents the homogeneous ignition of a hydrogen/air mixture. Principal component analysis is applied to reduce the dimensionality of the hydrogen/air mixture in composition space. Artificial neural networks (ANNs) are used to regress the reaction rates of principal components, and subsequently, a system of ordinary differential equations is solved. As the number of training samples decreases in the target task, the ROM fails to predict the ignition evolution of a hydrogen/air mixture. Three transfer learning strategies are then applied to the training of the ANN model with a sparse dataset. The performance of the ROM with a sparse dataset is remarkably enhanced if the training of the ANN model is restricted by a regularization term that controls the degree of knowledge transfer from source to target tasks. To this end, a novel transfer learning method is introduced, Parameter control via Partial Initialization and Regularization (PaPIR), whereby the amount of knowledge transferred is systemically adjusted in terms of the initialization and regularization schemes of the ANN model in the target task.

We propose a physics-constrained convolutional neural network (PC-CNN) to solve two types of inverse problems in partial differential equations (PDEs), which are nonlinear and vary both in space and time. In the first inverse problem, we are given data that is offset by spatially varying systematic error (i.e., the bias, also known as the epistemic uncertainty). The task is to uncover the true state, which is the solution of the PDE, from the biased data. In the second inverse problem, we are given sparse information on the solution of a PDE. The task is to reconstruct the solution in space with high resolution. First, we present the PC-CNN, which constrains the PDE with a time-windowing scheme to handle sequential data. Second, we analyze the performance of the PC-CNN to uncover solutions from biased data. We analyze both linear and nonlinear convection-diffusion equations, and the Navier–Stokes equations, which govern the spatiotemporally chaotic dynamics of turbulent flows. We find that the PC-CNN correctly recovers the true solution for a variety of biases, which are parameterized as non-convex functions. Third, we analyze the performance of the PC-CNN for reconstructing solutions from sparse information for the turbulent flow. We reconstruct the spatiotemporal chaotic solution on a high-resolution grid from only 1% of the information contained in it. For both tasks, we further analyze the Navier–Stokes solutions. We find that the inferred solutions have a physical spectral energy content, whereas traditional methods, such as interpolation, do not. This work opens opportunities for solving inverse problems with partial differential equations.

Urban communities rely on built utility infrastructures as critical lifelines that provide essential services such as water, gas, and power, to sustain modern socioeconomic systems. These infrastructures consist of underground and surface-level assets that are operated and geo-distributed over large regions where continuous monitoring for anomalies is required but challenging to implement. This article addresses the problem of deploying heterogeneous Internet of Things sensors in these networks to support future decision-support tasks, for example, anomaly detection, source identification, and mitigation. We use stormwater as a driving use case; these systems are responsible for drainage and flood control, but act as conduits that can carry contaminants to the receiving waters. Challenges toward effective monitoring include the transient and random nature of the pollution incidents, the scarcity of historical data, the complexity of the system, and technological limitations for real-time monitoring. We design a SemanTics-aware sEnsor Placement framework (STEP) to capture pollution incidents using structural, behavioral, and semantic aspects of the infrastructure. We leverage historical data to inform our system with new, credible instances of potential anomalies. Several key topological and empirical network properties are used in proposing candidate deployments that optimize the balance between multiple objectives. We also explore the quality of anomaly representation in the network through new perspectives, and provide techniques to enhance the realism of the anomalies considered in a network. We evaluate STEP on six real-world stormwater networks in Southern California, USA, which shows its efficacy in monitoring areas of interest over other baseline methods.

Many documents are produced over the years of managing assets, particularly those with long lifespans. However, during this time, the assets may deviate from their original as-designed or as-built state. This presents a significant challenge for tasks that occur in later life phases but require precise knowledge of the asset, such as retrofit, where the assets are equipped with new components. For a third party who is neither the original manufacturer nor the operator, obtaining a comprehensive understanding of the asset can be a tedious process, as this requires going through all available but often fragmented information and documents. While common knowledge regarding the domain or general type of asset can be helpful, it is often based on the experiences of engineers and is, therefore, only implicitly available. This article presents a graph-based information management system that complements traditional PLM systems and helps connect fragments by utilizing generic information about assets. To achieve this, techniques from systems engineering and data science are used. The overarching management platform also includes geometric analyses and operations that can be performed with geometric and product information extracted from STEP files. While the management itself is first described generically, it is also later applied to cabin retrofit in aviation. A mock-up of an Airbus A320 is utilized as the case study to demonstrate further how the platform can provide benefits for retrofitting such long-living assets.

Experiments in engineering are typically conducted in controlled environments where parameters can be set to any desired value. This assumes that the same applies in a real-world setting, which is often incorrect as many experiments are influenced by uncontrollable environmental conditions such as temperature, humidity, and wind speed. When optimizing such experiments, the focus should be on finding optimal values conditionally on these uncontrollable variables. This article extends Bayesian optimization to the optimization of systems in changing environments that include controllable and uncontrollable parameters. The extension fits a global surrogate model over all controllable and environmental variables but optimizes only the controllable parameters conditional on measurements of the uncontrollable variables. The method is validated on two synthetic test functions, and the effects of the noise level, the number of environmental parameters, the parameter fluctuation, the variability of the uncontrollable parameters, and the effective domain size are investigated. ENVBO, the proposed algorithm from this investigation, is applied to a wind farm simulator with eight controllable and one environmental parameter. ENVBO finds solutions for the entire domain of the environmental variable that outperform results from optimization algorithms that only focus on a fixed environmental value in all but one case while using a fraction of their evaluation budget. This makes the proposed approach very sample-efficient and cost-effective. An off-the-shelf open-source version of ENVBO is available via the NUBO Python package.

This paper proposes to solve the vortex gust mitigation problem on a 2D, thin flat plate using onboard measurements. The objective is to solve the discrete-time optimal control problem of finding the pitch rate sequence that minimizes the lift perturbation, that is, the criterion where is the lift coefficient obtained by the unsteady vortex lattice method. The controller is modeled as an artificial neural network, and it is trained to minimize using deep reinforcement learning (DRL). To be optimal, we show that the controller must take as inputs the locations and circulations of the gust vortices, but these quantities are not directly observable from the onboard sensors. We therefore propose to use a Kalman particle filter (KPF) to estimate the gust vortices online from the onboard measurements. The reconstructed input is then used by the controller to calculate the appropriate pitch rate. We evaluate the performance of this method for gusts composed of one to five vortices. Our results show that (i) controllers deployed with full knowledge of the vortices are able to mitigate efficiently the lift disturbance induced by the gusts, (ii) the KPF performs well in reconstructing gusts composed of less than three vortices, but shows more contrasted results in the reconstruction of gusts composed of more vortices, and (iii) adding a KPF to the controller recovers a significant part of the performance loss due to the unobservable gust vortices.

The condition assessment of underground infrastructure (UI) is critical for maintaining the safety, functionality, and longevity of subsurface assets like tunnels and pipelines. This article reviews various data acquisition techniques, comparing their strengths and limitations in UI condition assessment. In collecting structured data, traditional methods like strain gauge can only obtain relatively low volumes of data due to low sampling frequency, manual data collection, and transmission, whereas more advanced and automatic methods like distributed fiber optic sensing can gather relatively larger volumes of data due to automatic data collection, continuous sampling, or comprehensive monitoring. Upon comparison, unstructured data acquisition methods can provide more detailed visual information that complements structured data. Methods like closed-circuit television and unmanned aerial vehicle produce large volumes of data due to their continuous video recording and high-resolution imaging, posing great challenges to data storage, transmission, and processing, while ground penetration radar and infrared thermography produce smaller volumes of image data that are more manageable. The acquisition of large volumes of UI data is the first step in its condition assessment. To enable more efficient, accurate, and reliable assessment, it is recommended to (1) integrate data analytics like artificial intelligence to automate the analysis and interpretation of collected data, (2) to develop robust big data management platforms capable of handling large volumes of data storage, processing and analysis, (3) to couple different data acquisition technologies to leverage the strengths of each technique, and (4) to continuously improve data acquisition methods to ensure efficient and reliable data acquisition.

In practice, nondestructive testing (NDT) procedures tend to consider experiments (and their respective models) as distinct, conducted in isolation, and associated with independent data. In contrast, this work looks to capture the interdependencies between acoustic emission (AE) experiments (as meta-models) and then use the resulting functions to predict the model hyperparameters for previously unobserved systems. We utilize a Bayesian multilevel approach (similar to deep Gaussian Processes) where a higher-level meta-model captures the inter-task relationships. Our key contribution is how knowledge of the experimental campaign can be encoded between tasks as well as within tasks. We present an example of AE time-of-arrival mapping for source localization, to illustrate how multilevel models naturally lend themselves to representing aggregate systems in engineering. We constrain the meta-model based on domain knowledge, then use the inter-task functions for transfer learning, predicting hyperparameters for models of previously unobserved experiments (for a specific design).

Engineering machines are becoming increasingly complex and possess more control variables, increasing the complexity and versatility of the control systems. Different configurations of the control system, named a policy, can result in similar output behavior but with different resource or component life usage. There is therefore an opportunity to find optimal policies with respect to economic decisions. While many solutions have been proposed to find such economic policy decisions at the asset level, we consider this problem at the fleet level. In this case, the optimal operation of each asset is affected by the state of all other assets in the fleet. Challenges introduced by considering multiple assets include the construction of economic multi-objective optimization criteria, handling rare events such as failures, application of fleet-level constraints, and scalability. The proposed solution presents a framework for economic fleet optimization. The framework is demonstrated for economic criteria relating to resource usage, component lifing, and maintenance scheduling, but is generically extensible. Direct optimization of lifetime distributions is considered in order to avoid the computational burden of discrete event simulation of rare events. Results are provided for a real-world case study targeting the optimal economic operation of a fleet of aerospace gas turbine engines.