Engineering Data Sets Research Articles

TTSNet: Transformer–Temporal Convolutional Network–Self-Attention with Feature Fusion for Prediction of Remaining Useful Life of Aircraft Engines

Accurately predicting the remaining useful life (RUL) is crucial for ensuring the safety and reliability of aircraft engine operation. However, aircraft engines operate in harsh conditions, with the characteristics of high speed, high temperature, and high load, resulting in high-dimensional and noisy data. This makes feature extraction inadequate, leading to low accuracy in the prediction of the RUL of aircraft engines. To address this issue, Transformer-TCN-Self-attention network (TTSNet) with feature fusion, as a parallel three-branch network, is proposed for predicting the RUL of aircraft engines. The model first applies exponential smoothing to smooth the data and suppress noise to the original signal, followed by normalization. Then, it uses a parallel transformer encoder, temporal convolutional network (TCN), and multi-head attention three-branch network to capture both global and local features of the time series. The model further completes feature dimension weight allocation and fusion through a multi-head self-attention mechanism, emphasizing the contribution of different features to the model. Subsequently, it fuses the three parts of features through a linear layer and concatenation. Finally, a fully connected layer is used to establish the mapping relationship between the feature matrix and the RUL label, obtaining the RUL prediction value. The model was validated on the C-MAPSS aircraft engine dataset. Experimental results show that compared to other related RUL models, the RMSE and Score reached 11.02 and 194.6 on dataset FD001, respectively; on dataset FD002, the RMSE and Score reached 13.25 and 874.1, respectively. On dataset FD003, the RMSE and Score reached 11.06 and 200.1 and on dataset FD004, the RMSE and Score reached 18.26 and 1968.5, respectively, demonstrating better performance of RUL prediction.

Persistent Topological Laplacians—A Survey

Persistent topological Laplacians constitute a new class of tools in topological data analysis (TDA). They are motivated by the necessity to address challenges encountered in persistent homology when handling complex data. These Laplacians combine multiscale analysis with topological techniques to characterize the topological and geometrical features of functions and data. Their kernels fully retrieve the topological invariants of corresponding persistent homology, while their non-harmonic spectra provide supplementary information. Persistent topological Laplacians have demonstrated superior performance over persistent homology in the analysis of large-scale protein engineering datasets. In this survey, we offer a pedagogical review of persistent topological Laplacians formulated in various mathematical settings, including simplicial complexes, path complexes, flag complexes, digraphs, hypergraphs, hyperdigraphs, cellular sheaves, and N-chain complexes.

Physics-informed spatio-temporal hybrid neural networks for predicting remaining useful life in aircraft engine

Aircraft engines have complex structures and various operating conditions, which leads to the health monitoring data of aircraft engines’ high coupling in the spatial domain and time invariance in the temporal domain. Existing methods for remaining useful life prediction of aircraft engines are deficient in handling spatial data and spatial–temporal data fusion. To address this problem, a novel method named physics-informed spatio-temporal hybrid neural networks (PI-STHNN) is proposed. In this study, a spatial domain data association graph for aircraft engines is constructed by utilizing a thermal cycle physical model as prior knowledge and combining it with the grey relational grade analysis method. Specifically, the proposed spatio-temporal hybrid neural network framework extracts the temporal features and spatial coupling features of the data in parallel. Then attention mechanism is employed to weigh and fuse all feature representations. Finally, the method is validated on the two aircraft engine datasets, the results show that PI-STHNN outperforms current state-of-the-art methods. Furthermore, our analysis sheds light on the advantages of incorporating physical information into graph neural network models and visually presents the attention weights to elucidate the contributions of time-domain and spatial-domain models within our hybrid approach.

Medical and Engineering Applications for Estimation and Prediction of a New Competing Risks Model: A Bayesian Approach

The Bayesian approach offers a flexible, interpretable and powerful framework for statistical analysis, making it a valuable tool to help in making optimal decisions under uncertainty. It incorporates prior knowledge or beliefs about the parameters, which can lead to more accurate and informative results. Also, it offers credible intervals as a measure of uncertainty, which are often more interpretable than confidence intervals. Hence, the Bayesian approach is utilized to estimate the parameters, reliability function, hazard rate function and reversed hazard rate function of a new competing risks model. A squared error loss function as a symmetric loss function and a linear exponential loss function as an asymmetric loss function are employed to derive the Bayesian estimators. Credible intervals of the parameters, reliability function, hazard rate function and reversed hazard rate function are obtained. Predicting future observations is important in many fields, from finance and weather forecasting to healthcare and engineering. Thus, two-sample prediction (as a special case of the multi-sample prediction) for future observation is considered. An adaptive Metropolis algorithm is applied to conduct a simulation study to evaluate the performance of the Bayes estimates and predictors. Moreover, two applications of medical and engineering data sets are used to test and validate the theoretical results, ensuring that they are accurate, applicable to real-world scenarios and contribute to the understanding of the world and inform decision-making.

Predictive combination model for CH4 separation and CO2 sequestration with CO2 injection into coal seams: VMD-STA-BiLSTM-ELM hybrid neural network modeling

In the experimental process of separating coal seam gas using the CO2 displacement method, establishing a predictive model for key variables is essential to optimize displacement parameters, increase coal seam gas recovery, and improve CO2 sequestration efficiency. Traditional modeling methods often struggle with the complex nature of industrial data and are susceptible to overfitting due to multicollinearity caused by long-term datasets. This paper presents a hybrid predictive model based on variational mode decomposition (VMD), a spatiotemporal attention mechanism (STA), a bidirectional long short-term memory network (BiLSTM), and an extreme learning machine (ELM). During the offline phase, VMD is used to decompose raw data into intrinsic mode functions (IMFs). The hidden state from the last time step of the STA-BiLSTM is then added to the original data to enrich the features for ELM training. In the online prediction phase, the outputs from the VMD-STA-BiLSTM and VMD-STA-BiLSTM-ELM models are combined using an error reciprocal method to generate the final prediction. The proposed model is validated with experimental datasets from CO2 displacement for coal seam CH4 under various conditions, as well as with N2-ECBM and CO2-ECBM engineering datasets. The results show that the hybrid model surpasses VMD-ELM, STA-BiLSTM-ELM, BiLSTM, STA-BiLSTM, ELM, TCN, and Attention-TCN models in predictive accuracy. Even in multi-step and rolling predictions, the model exhibits minimal impact from cumulative errors, maintaining accurate forecasts with strong generalization and robustness. It effectively captures feature patterns across different datasets and accurately predicts unknown data. The model shows potential for application in diverse scenarios and complex environments, offering reliable support and decision-making for the field application of CO2 displacement in coal seam CH4 separation. It is an effective and promising predictive approach to enhance coal seam gas recovery and CO2 sequestration efficiency.

Bearing fault diagnosis by sparse frequency spiral spectrum driven NAF-LDM under strong noise and small samples

Abstract The complexity of background noise and the scarcity of real fault samples seriously affect the diagnostic accuracy of the model. To address this, a noise-robust two-dimensional feature map, the sparse frequency spiral spectrum (SFSM), based on sparse representation theory, is proposed. A bridge penalty coefficient is applied to the sparse representation model to accurately select impact components, and the fast iterative shrinkage threshold algorithm is used to solve for sparse representation coefficients. Sparse reconstructed signals are obtained by convolving the impact patterns with these coefficients, leading to a sparse reconstruction algorithm with reduced computational complexity. Furthermore, the novel non-linear activation-free blocks (NAF Blocks) are embedded into the latent diffusion model to augment small samples, significantly improving image generation speed and quality. The integration of the Swin transformer for feature extraction and classification further enhances diagnostic performance. The superiority of this method is validated on the XJTU-SY dataset, a bearing experimental platform dataset, and enterprise engineering dataset. Experimental results demonstrate that the structural and generalization advantages of NAF Blocks are crucial for improving image quality and inference speed. The noise suppression capability of the proposed method, facilitated by the SFSM feature processing technique, is confirmed through ablation and noise robustness tests. Finally, the Swin transformer’s excellent feature extraction and classification capabilities for SFSM are verified. The proposed method achieves diagnostic accuracies of 99.10% and 98.7% on the XJTU-SY and experimental platform datasets, respectively.

Bivariate cubic normal distribution for non-Gaussian problems

Probabilistic models play critical role in various engineering fields. Numerous critical issues exist in probabilistic modeling, especially for non-Gaussian correlated random variables. Traditional parameter-based bivariate distribution models are typically developed for specific types of random variables, which limits their flexibility and applicability. In this study, a flexible bivariate distribution model is proposed, in which the joint cumulative distribution function (JCDF) is derived by expressing the probability as the summation of three basic probabilities corresponding to simple functions. These probabilities are computed using a univariate cubic normal distribution, and thus the proposed model is named as bivariate cubic normal (BCN) distribution. The proposed BCN distribution has been applied in modeling several common bivariate distributions and actual engineering datasets. Results show that the BCN distribution accurately fits the JCDFs of both theoretical distributions and practical datasets, offering significant improvement over existing models. Furthermore, the proposed BCN distribution is utilized in seismic reliability assessment and the calculation of the mean recurrence interval and hazard curve of hurricane wind speed and storm size. Results demonstrate that the BCN distribution excels in modeling and matching capabilities, proving its accuracy and effectiveness in civil engineering applications.

Bayesian Approach for Estimating the Parameters of the Time-To-Failure Distribution and Its Percentiles Under Power Degradation Model

The degradation models are often applied on the degradation data for studying time-to-failure distribution. In this study, the Bayesian approach is applied on the power degradation model for estimating the parameters of the time-to-failure distribution and its percentiles. Two different distributions are assumed for the degradation parameter of the model. The degradation parameter is firstly assumed to follow the skew-normal distribution with three jointly independently distributed parameters such that the gamma prior is assumed for the shape parameter, while the scale and the location parameters are assumed uniform. The second distribution assumed for the degradation parameter is the log-logistic distribution with two jointly independent random parameters where the shape parameter is assumed gamma, while the scale parameter is assumed uniform. Based on the Gibbs sampling method carried out under the JAGS platform, the models considered are applied on the simulated data and the NASA turbofan Jet engine dataset and the results found are compared. In modeling the time-to-failure distribution, it is shown that based on the simulated data and real data, the Bayesian approach for the power degradation model with the skew-normal degradation parameter outperformed the Bayesian approach for the power degradation model with the log-logistic degradation parameter.

An improved similarity-based RUL prediction method considering degradation degree of multiple condition monitoring parameters for aero-engines

Abstract The aero-engine is the heart of an airplane. Predicting the remaining useful lifetime (RUL) of an aero-engine bears great significance, not only for improving the reliability and safety of the aero-engine but also for ensuring aircraft safety and performance. However, both issues, namely the selection of uncorrelated parameters for RUL estimation and the lack of a standard theoretical methodology for Health Index (HI) construction, inevitably impact the prediction accuracy. Here, we proposed an improved similarity-based RUL prediction method considering the degradation degree of multiple condition monitoring parameters for aero-engines. This method includes the improved minimum-redundancy maximum-relevancy approach for the quantitative selection of key parameters, and the similarity matching approach which takes into account the degradation degree of multiple parameters instead of constructing HI. The effectiveness of the proposed method was evaluated on turbine engine datasets. Experimental results show that (1) compared with other feature selection methods, the root mean square error (RMSE) of the proposed method is reduced by 12.3%; (2) compared with other RUL prediction methods, the RMSE of the proposed method is smaller than most methods but without a complicated training process. The results demonstrate that the proposed method achieves highly competitive prediction performance. By employing the proposed method, it is possible to significantly reduce the risk of engine failures, thereby improving safety and economic efficiency.

Revisiting Sentiment Analysis for Software Engineering in the Era of Large Language Models

Software development involves collaborative interactions where stakeholders express opinions across various platforms. Recognizing the sentiments conveyed in these interactions is crucial for the effective development and ongoing maintenance of software systems. For software products, analyzing the sentiment of user feedback, e.g., reviews, comments, and forum posts can provide valuable insights into user satisfaction and areas for improvement. This can guide the development of future updates and features. However, accurately identifying sentiments in software engineering datasets remains challenging. This study investigates bigger large language models (bLLMs) in addressing the labeled data shortage that hampers fine-tuned smaller large language models (sLLMs) in software engineering tasks. We conduct a comprehensive empirical study using five established datasets to assess three open-source bLLMs in zero-shot and few-shot scenarios. Additionally, we compare them with fine-tuned sLLMs, using sLLMs to learn contextual embeddings of text from software platforms. Our experimental findings demonstrate that bLLMs exhibit state-of-the-art performance on datasets marked by limited training data and imbalanced distributions. bLLMs can also achieve excellent performance under a zero-shot setting. However, when ample training data is available or the dataset exhibits a more balanced distribution, fine-tuned sLLMs can still achieve superior results.

Machine cross-domain remaining useful life prediction via contrastive adversarial variational recurrent method

The performance degradation process of the machine is non-stationary which varies with the operating environment and is reflected as a temporal shift phenomenon. Models or methods that assume the test and training sets have the same distribution will no longer be suitable for solving problems where domain shifts exist. This brings new challenges for accurately evaluating the machine’s remaining useful life (RUL). This paper studies a contrastive adversarial variational recurrent method for machine RUL prediction under different service conditions. In this new approach, the variational recurrent networks are developed to extract distribution features and latent spaces of raw data, meanwhile, the adversarial strategies are applied to enable the learning of domain-invariant features to make the model achieve cross-domain task processing. To supervise the learning process of the model to learn more mutual information between the extracted features and the raw input data, a contrastive loss is also introduced in the proposed method. Sufficient experiments were conducted to verify the feasibility and superiority of the suggested approach, including 12 sets of cross-scenario tests on the turbofan engine dataset C-MAPSS from NASA. Experimental findings confirm that the proposed method performs satisfactorily and competitively even compared to current state-of-the-art works.

AEEFCSR: an adaptive ensemble empirical feed-forward cascade stochastic resonance system for weak signal detection

Abstract A novel adaptive ensemble empirical feed-forward cascade stochastic resonance (AEEFCSR) method is proposed in this study for the challenges of detecting target signals from intense background noise. At first, we create an unsaturated piecewise self-adaptive variable-stable potential function to overcome the limitations of traditional potential functions. Subsequently, based on the foundation of a feed-forward cascaded stochastic resonance method, a novel weighted function and system architecture is created, which effectively addresses the issue of low-frequency noise enrichment through ensemble empirical mode decomposition. Lastly, inspired by the spider wasp algorithm and nutcracker optimization algorithm, the spider wasp nutcracker optimization algorithm is proposed to optimize the system parameters and overcome the problem of relying on manual experience. In this paper, to evaluate its performance, the output signal-to-noise ratio (SNR), spectral sub-peak difference, and time-domain recovery capability are used as evaluation metrics. The AEEFCSR method is demonstrated through theoretical analysis. To further illustrate the performance of the AEEFCSR method, Validate the adoption of multiple engineering datasets. The results show that compared with the compared algorithms, the output SNR of the AEEFCSR method is at least 6.2801 dB higher, the spectral subpeak difference is more than 0.25 higher, and the time-domain recovery effect is more excellent. In summary, the AEEFCSR method has great potential for weak signal detection in complex environments.

Extended Association Rule Mining and Its Application to Software Engineering Data Sets

Association rule mining is a highly effective approach to data analysis for datasets of varying sizes, accommodating diverse feature values. Nevertheless, deriving practical rules from datasets with numerical variables presents a challenge, as these variables must be discretized beforehand. Quantitative association rule mining addresses this issue, allowing the extraction of valuable rules. This paper introduces an extension to quantitative association rules, incorporating a two-variable function in their consequent part. The use of correlation functions, statistical test functions, and error functions is also introduced. We illustrate the utility of this extension through three case studies employing software engineering datasets. In case study 1, we successfully pinpointed the conditions that result in either a high or low correlation between effort and software size, offering valuable insights for software project managers. In case study 2, we effectively identified the conditions that lead to a high or low correlation between the number of bugs and source lines of code, aiding in the formulation of software test planning strategies. In case study 3, we applied our approach to the two-step software effort estimation process, uncovering the conditions most likely to yield low effort estimation errors.

State of Health Prediction in Electric Vehicle Batteries Using a Deep Learning Model

Accurately estimating the state of health (SOH) of lithium-ion batteries plays a significant role in the safe operation of electric vehicles. Deep learning (DL)-based approaches for estimating state of health (SOH) have consistently been the focus of study in recent years. In the current era of electric mobility, the utilization of lithium-ion batteries (LIBs) has evolved into a necessity for energy storage. Ensuring the safe operation of EVs requires a precise assessment of the state-of-health (SOH) of LIBs. To estimate battery SOH accurately, this paper employs a deep learning (DL) algorithm to enhance the estimation accuracy of SOH to obtain accurate SOH measurements. This research introduces the Diffusion Convolutional Recurrent Neural Network (DCRNN) with a Support Vector Machine-Recursive Feature Elimination (SVM-RFE) algorithm (DCRNN + SVM-RFE) for enhancing classification and feature selection performance. The data gathered from the dataset were pre-processed using the min–max normalization method. The Center for Advanced Life Cycle Engineering (CALCE) dataset from the University of Maryland was employed to train and evaluate the model. The SVM-RFE algorithm was used for feature selection of pre-processed data. DCRNN algorithm was used for the classification process to enhance prediction precision. The DCRNN + SVM-RFE model’s performance was calculated using Mean Absolute Percentage Error (MAPE), Mean Squared Error (MAE), Mean Squared Error (MSE), and Root MSE (RMSE) metric values. The proposed model generates accurate results for SOH prediction; all RMSEs are within 0.02%, MAEs are within 0.015%, MSEs were within 0.032%, and MAPEs are within 0.41%. The mean values of RMSE, MSE, MAE, and MAPE were 0.014, 0.026, 0.011, and 0.32, respectively. Experiments confirmed that the DCRNN + SVM-RFE model has the highest accuracy among those that predict SOH.

Prediction of Remaining Useful Life for Lithium‐Ion Batteries Using Complete Ensemble Empirical Mode Decomposition with Adaptive Noise for Feature Analysis, and Bidirectional Long Short‐Term Memory Coupled with a Gaussian Process Regression Model

Accurately predicting the remaining useful life (RUL) of lithium‐ion batteries is a challenging task, with significant implications for managing battery usage risks and ensuring equipment stability. However, the phenomenon of capacity regeneration and the lack of confidence interval expression result in imprecise predictions. To tackle these challenges, this article proposes a novel method for predicting RUL by optimizing health features (HFs) and integrating multiple models. First, multiple HFs are collected from the charging curves, and the fusion HF is optimized by kernel principal component analysis. To eliminate local fluctuations caused by capacity regeneration effects, the complete ensemble empirical mode decomposition with adaptive noise is employed to decompose the fusion HF. Second, to address the issue of lacking confidence interval expression, a hybrid model is proposed by integrating bidirectional long short‐term memory neural network with Gaussian process regression for effectively capturing the lithium‐ion battery capacity‐declining trend and accurately predicting the RUL. Finally, the proposed model's effectiveness is validated by comparing it with several other models using National Aeronautics and Space Administration and Center for Advanced Life Cycle Engineering datasets. The results indicate that this model achieves a root mean square error of 0.0023 and a mean absolute error of 0.0058, demonstrating significant improvements in predictive accuracy for RUL with high reliability.

Optimized quantum LSTM using modified electric Eel foraging optimization for real-world intelligence engineering systems

The integration of metaheuristics with machine learning methodologies presents significant advantages, particularly in optimization and computational intelligence. This amalgamation leverages the global search capabilities of metaheuristics alongside the pattern recognition and predictive prowess of machine learning, facilitating enhanced convergence rates and solution quality in complex problem spaces. The Quantum Long Short-Term Memory (QLSTM) emerges as a highly efficient deep learning model tailored to tackle such intricate engineering problems. The QLSTM's architecture, comprising data encoding, variational, and quantum measurement layers, facilitates the effective encoding and processing of civil engineering data, leading to heightened prediction accuracy. However, the task of determining optimal values for QLSTM parameters presents challenges due to its NP-problem nature and time-consuming characteristics. To address this, we propose an alternative technique to optimize the QLSTM based on a modified Electric Eel Foraging Optimization (MEEFO). The MEEFO is a modified version of the original EEFO that applies triangular mutation operators to boost the search capability of the traditional EEFO. Thus, the MEEFO optimizes the QLSTM and boosts its prediction performance. To validate the efficacy of our proposed method, we conduct comprehensive experiments utilizing five real-world engineering datasets related to construction and structure engineering. The evaluation outcomes unequivocally demonstrate that the MMEFO significantly enhances the performance of the QLSTM.

Learning the hard-to-learn: Active learning for imbalanced datasets in data-centric tunnel engineering

Data-driven methodologies hold the potential to revolutionize mechanized tunneling by informing decision-making. However, imbalanced data categories are prevalent within the realm of mechanized tunneling. Data-driven models often struggle to accurately predict data from minor sample categories, which, despite their scarcity, are crucial in practice. To address this bottleneck, this study introduces a committee-based active learning strategy to tackle the imbalanced sample identification problem. This strategy involves constructing multiple committee models to quantify the uncertainty in surrogate predictions. These specimens exhibiting high forecast uncertainty will be resampled with replacement from the validation data to iteratively augment the training data, thereby enhancing the model’s capability to handle imbalanced datasets. This study validates the strategy by applying it to two real tunneling engineering scenarios. The first case is to predict the grade of surrounding rock and the other one involves forecasting muck clogging in mechanised tunneling. The results demonstrate that the proposed strategy significantly improves the prediction accuracy of minority categories. Further, the study reveals that both the “entropy method” and “weight method” can effectively quantify the learning difficulty of samples. The same strategy can be equally applied to other fields of engineering and science.

Exploring the utility of the Enhanced Vegetation Index as rainfall and agricultural proxy in a Caribbean case study event

Highly fragile small island states experience disproportionate climate impacts given their limited capacity to implement cost‐effective tools for detecting emerging signals of drying conditions and monitoring systems for sensitive sectors such as agriculture, especially for uncertain, ‘creeping’ events such as droughts. Despite the existence of open‐source Google Earth Engine datasets, untapped potential remains for their full deployment in disaster management infrastructure. Given this gap, this paper explores the utility of the Enhanced Vegetation Index (EVI) for detecting spatio‐temporal variations of Mid‐Summer Drought (MSD) impacts on vegetation in the small island of Jamaica, with emphasis on major historical drought events. Geospatial analyses of EVI datasets from the Terra Moderate‐Resolution Imaging Spectroradiometer (MODIS) between 2000−2015 archived by the United States Geological Survey (USGS), were computed, and validated by station‐based precipitation and production data for selected parishes for historical case study MSD events. Results revealed highly asymmetrical drought impacts, with Jamaica's agriculturally intense Southern coastline displaying the most stressed vegetation (EVI < 0.5). North‐Western and North‐Eastern regions had the healthiest vegetation during the MSD (EVI > 0.6). A ‘fair’ to ‘moderate’ concurrent correlation was found between EVI and precipitation (R > 0.6), with lower correlations vis‐a‐vis agricultural production (R = 0.2–0.4). The results provide evidence of EVI's utility as a drought monitoring tool in a small island context.

Classical and Bayesian techniques for modelling engineering dataset using new generalized probability distribution with mathematical features

Classical and Bayesian techniques for modelling engineering dataset using new generalized probability distribution with mathematical features

A novel interactive prognosis framework with nonlinear Wiener process and multi-sensor fusion for remaining useful life prediction

Accurate remaining useful life (RUL) prediction plays a vital role in increasing the system operation safety and reducing maintenance costs. In industrial applications, there is usually a large amount of multi-sensor data generated. Therefore, how to construct an appropriate health index (HI) based on multi-sensor signals is very important for the RUL prediction. However, existing works treat sensor selection, HI construction, and degradation modeling independently as unrelated parts, which may result in the combination of sensors selected not constituting an optimal HI or the constructed HI not matching the degradation model. In addition, most existing works treat prior units as a whole to obtain a unique set of sensor combinations and fusion coefficients, which cannot reflect unit-to-unit heterogeneity, thus affecting the accuracy of RUL prediction. Therefore, a novel interactive feedback framework is established to construct HI, where the sensor selection, fusion coefficient calculation, and nonlinear Wiener process degradation modeling are incorporated into the feedback. Furthermore, an adaptive weight selection method based on particle swarm optimization and leave-one-out cross-validation (PSO-LV) is proposed to adjust the fusion coefficients in real-time. Then, the RUL is estimated by updating model parameters online, detecting degradation trends, and deriving the probability density function (PDF) of the RUL. Finally, two examples of engine datasets are provided to verify the effectiveness of the proposed method.