Raw Time Research Articles

Automated seismic event detection considering faulty data interference using deep learning and Bayesian fusion

AbstractStructural health monitoring (SHM) aims to assess civil infrastructures' performance and ensure safety. Automated detection of in situ events of interest, such as earthquakes, from extensive continuous monitoring data, is important to ensure the timeliness of subsequent data analysis. To overcome the poor timeliness of manual identification and the inconsistency of sensors, this paper proposes an automated seismic event detection procedure with interpretability and robustness. The sensor‐wise raw time series is transformed into image data, enhancing the separability of classification while endowing with visual understandability. Vision Transformers (ViTs) and Residual Networks (ResNets) aided by a heat map–based visual interpretation technique are used for image classification. Multitype faulty data that could disturb the seismic event detection are considered in the classification. Then, divergent results from multiple sensors are fused by Bayesian fusion, outputting a consistent seismic detection result. A real‐world monitoring data set of four seismic responses of a pair of long‐span bridges is used for method validation. At the classification stage, ResNet 34 achieved the best accuracy of over 90% with minimal training cost. After Bayesian fusion, globally consistent and accurate seismic detection results can be obtained using a ResNet or ViT. The proposed approach effectively localizes seismic events within multisource, multifault monitoring data, achieving automated and consistent seismic event detection.

A novel TCN-GRU based open set method for unknown damage diagnosis

Abstract In the aerospace and high-speed rail industries, CFRP has seen widespread application. CFRP plates and connectors in operation are often subjected to impacts that can cause damage. The unpredictable nature of the impacts introduces uncertainties in both the location and extent of the damage, posing significant challenges to traditional supervised learning models, which often struggle with missed detections or misclassifications when identifying unknown damages. To address the issue, a deep learning model based on TCN-GRU is proposed. TCN extracts features from the raw time domain signals, and GRU selectively retains the significant features and completes sequence modeling. A center loss function is incorporated into the fully connected layer to improve the effects of intra-class aggregation and inter-class separation. An unknown detection module is introduced to realize the identification and classification of unknown damages based on a predefined threshold. The experimental results indicate that the proposed method can achieve effective unknown damage diagnosis in the open set case. This study provides a feasible solution for open set unknown damage diagnosis in CFRP plates and connectors.

Exploring the Sustainable Utilization of Deep Eutectic Solvents for Chitin Isolation from Diverse Sources.

Deep eutectic solvents (DES) represent an innovative and environmentally friendly approach for chitin isolation. Chitin is a natural nitrogenous polysaccharide, characterized by its abundance of amino and hydroxyl groups. The hydrogen bond network in DES can disrupt the crystalline structure of chitin, facilitating its isolation from bioresources by dissolving or degrading other components. DES are known for their low cost, natural chemical constituents, and recyclability. Natural deep eutectic solvents (NADES), a subclass of DES made from natural compounds, offer higher biocompatibility, biodegradability, and the lowest biotoxicity, making them highly promising for the production of eco-friendly chitin products. This review summarized studies on chitin isolation by DES, including reviews of biomass resources, isolation conditions (raw materials, DES compositions, solid-liquid ratios, temperature, and time), and the physicochemical properties of chitin products. Consequently, we have concluded that tailoring an appropriate DES-based process on the specific composition of the raw material can notably improve isolation efficiency. Acidic DES are particularly effective for extracting chitin from materials with high mineral content, such as crustacean bio-waste; for instance, the choline chloride-lactic acid DES achieved purity levels comparable to those of commercial chemical methods. By contrast, alkaline DES are better suited for chitin isolation from protein-rich sources, such as squid pens. DES facilitate calcium carbonate removal through H+ ion release and leverage unique hydrogen bonding interactions for efficient deproteination. Among these, potassium carbonate-glycerol DES have demonstrated optimal efficacy. Nonetheless, further comprehensive research is essential to evaluate the environmental impact, economic feasibility, and safety of DES application in chitin production.

Detection of breast cancer using machine learning on time-series diffuse optical transillumination data.

Optical mammography as a promising tool for cancer diagnosis has largely fallen behind expectations. Modern machine learning (ML) methods offer ways to improve cancer detection in diffuse optical transmission data. We aim to quantitatively evaluate the classification of cancer-positive versus cancer-negative patients using ML methods on raw transmission time series data from bilateral breast scans during subjects' rest. We use a support vector machine (SVM) with hyperparameter optimization and cross-validation to systematically explore a range of data preprocessing and feature-generation strategies. We also apply an automated ML (AutoML) framework to validate our findings. We use receiver operating characteristics and the corresponding area under the curve (AUC) to quantify classification performance. For the sample group available ( , 18 cancer patients), we demonstrate an AUC score of up to 93.3% for SVM classification and up to 95.0% for the AutoML classifier. ML offers a viable strategy for clinically relevant breast cancer diagnosis using diffuse-optical transmission measurements. The diagnostic performance of ML on raw data can outperform traditional statistical biomarkers derived from reconstructed image time series. To achieve clinically relevant performance, our ML approach requires simultaneous bilateral scanning of the breasts with spatially dense channel coverage.

A simple but tough-to-beat baseline for fMRI time-series classification

Current neuroimaging studies frequently use complex machine learning models to classify human fMRI data, distinguishing healthy and disordered brains, often to validate new methods or enhance prediction accuracy. Yet, where prediction accuracy is a concern, our results suggest that precision in prediction does not always require such sophistication. When a classifier as simple as logistic regression is applied to feature-engineered fMRI data, it can match or even outperform more sophisticated recent models. Classification of the raw time series fMRI data generally benefits from complex parameter-rich models. However, this complexity often pushes them into the class of black-box models. Yet, we found that a relatively simple model can consistently outperform much more complex classifiers in both accuracy and speed. This model applies the same multi-layer perceptron repeatedly across time and averages the results. Thus, the complexity and black-box nature of the parameter rich models, often perceived as a necessary trade-off for higher performance, do not invariably yield superior results on fMRI.Given the success of straightforward approaches, we challenge the merit of research that concentrates solely on complex model development driven by classification. Instead, we advocate for increased focus on designing models that prioritize the explainability of fMRI data or pursue applicable objectives beyond mere classification accuracy, unless they significantly outperform logistic regression or our proposed model. To validate our claim, we explore possible reasons for the superior performance of our straightforward model by examining the innate characteristics of fMRI time series data. Our findings suggest that the sequential information hidden in the temporal order may be far less important for the accurate fMRI classification than the stand-alone pieces of information scattered across the frames of the time series.

An in-air synthetic aperture sonar dataset of target scattering in environments of varying complexity

This paper describes a synthetic aperture sonar (SAS) dataset collected in-air consisting of four types of targets in four environments of different complexity. The in-air laboratory based experiments produced data with a level of fidelity and ground truth accuracy that is not easily attainable in data collected underwater. The range of complexity, high level of data fidelity, and accurate ground truth provides a rich dataset with acoustic features on multiple scales. It can be used to develop new signal-processing and image reconstruction algorithms, as well as machine learning models for object detection and classification. It may also find application in model verification and validation for acoustic simulators. The dataset consists of raw acoustic time series returns, associated environmental conditions, hardware configuration, array motion, as well as the reconstructed imagery.

Temporal-Frequency Attention Focusing for Time Series Extrinsic Regression via Auxiliary Task.

Time series extrinsic regression (TSER) aims at predicting numeric values based on the knowledge of the entire time series. The key to solving the TSER problem is to extract and use the most representative and contributed information from raw time series. To build a regression model that focuses on those information suitable for the extrinsic regression characteristic, there are two major issues to be addressed. That is, how to quantify the contributions of those information extracted from raw time series and then how to focus the attention of the regression model on those critical information to improve the model's regression performance. In this article, a multitask learning framework called temporal-frequency auxiliary task (TFAT) is designed to solve the mentioned problems. To explore the integral information from the time and frequency domains, we decompose the raw time series into multiscale subseries in various frequencies via a deep wavelet decomposition network. To address the first problem, the transformer encoder with the multihead self-attention mechanism is integrated in our TFAT framework to quantify the contribution of temporal-frequency information. To address the second problem, an auxiliary task in a manner of self-supervised learning is proposed to reconstruct the critical temporal-frequency features so as to focusing the regression model's attention on those essential information for facilitating TSER performance. We estimated three kinds of attention distribution on those temporal-frequency features to perform auxiliary task. To evaluate the performances of our method under various application scenarios, the experiments are carried out on the 12 datasets of the TSER problem. Also, ablation studies are used to examine the effectiveness of our method.

Multivariate time-series clustering of cardiopulmonary exercise test data for cardiovascular risk stratification

Abstract Aims Current clinical practice assesses cardiorespiratory fitness (CRF) by evaluating summary indexes of cardiopulmonary exercise tests (CPET). Yet, the raw time series recordings may hold additional information with clinical relevance. In this study, we investigated if the interpretation of raw CPET data could identify distinct CRF phenogroups and improve risk stratification. Methods In 1399 participants (mean age, 56.4 years; 37.7% women), who underwent maximal CPET by cycle ergometer, we collected baseline clinical characteristics and information on incident cardiovascular (CV) events (n=297) on average 4.3 years later. We employed dynamic time warping combined with k-medoids to identify phenogroups from raw CPET time-series. To evaluate the clinical significance of the derived phenogroups we compared clinical characteristics and incidence of CV events, while an external population cohort (n=171) was used to further validate the trained model. Results The optimal number of clusters was 5. We observed significant differences in age, use of medication, spirometry indexes and disease prevalence across all clusters. Cluster 5 was associated with the worst CV profile with higher use of antihypertensive medication and history of CV disease while cluster 1 represented the most favourable risk profile. After adjusting for traditional clinical covariables, clusters 4 (HR: 1.30; 95%CI: 1.02-1.65; p=0.033) and 5 (HR: 1.36; 95%CI: 1.08-1.72; p=0.0088) had significantly higher risk for incident CV events compared to clusters 1 and 2. Comparing clinical characteristics across clusters in the external validation cohort revealed similar patterns. Conclusion – Employing unsupervised machine learning on time series CPET recordings, we identified clinically meaningful CRF phenogroups. An integrative CRF profiling might facilitate CV risk stratification and consequently risk management.

Anomaly detection for bridge health monitoring data based on multiple encoded images and convolutional neural network

In bridge health monitoring systems, abnormal monitoring data inevitably appear due to the complex operational conditions, and these anomalous data will seriously affect the accuracy and reliability of the results of bridge health assessment. Hence, it is vital to detect the abnormal data before further processing. However, the pattern diversity and sample scarcity usually make the accuracy of data anomaly detection unsatisfactory. A novel anomaly detection method for bridge health monitoring data based on encoded images and convolutional neural network (CNN) is proposed in this paper. First, the raw time series are encoded into images to highlight the intrinsic features in the anomaly data, and multiple encoding techniques are adopted to represent the data in different perspectives. Then, the CNN is utilized to establish the mapping between the encoded images and the anomaly patterns for achieving the data anomaly detection. At last, the proposed method is verified with the acceleration data measured from an actual large span cable-stayed bridge. It shows that the proposed method can significantly improve the performance of data anomaly detection while the results are hardly affected by the small sample sizes, and a detection accuracy of over 95% is achieved based on the multiple encoded images.

Scale-Aware Neural Architecture Search for Multivariate Time Series Forecasting

Multivariate time series (MTS) forecasting has attracted much attention in many intelligent applications. It is not a trivial task, as we need to consider both intra-variable dependencies and inter-variable dependencies. However, existing works are designed for specific scenarios, and require much domain knowledge and expert efforts, which is difficult to transfer between different scenarios. In this paper, we propose a scale-aware neural architecture search framework for MTS forecasting (SNAS4MTF). A multi-scale decomposition module transforms raw time series into multi-scale sub-series, which can preserve multi-scale temporal patterns. An adaptive graph learning module infers the different inter-variable dependencies under different time scales without any prior knowledge. For MTS forecasting, a search space is designed to capture both intra-variable dependencies and inter-variable dependencies at each time scale. The multi-scale decomposition, adaptive graph learning, and neural architecture search modules are jointly learned in an end-to-end framework. Extensive experiments on two real-world datasets demonstrate that SNAS4MTF achieves a promising performance compared with the state-of-the-art methods.

Comparison of CAMS and CMAQ analyses of surface-level PM2.5 and O3 over the conterminous United States (CONUS)

To reduce economic and health impacts from poor air quality (AQ) in the U.S., the National Air Quality Forecasting Capability (NAQFC) at the National Oceanic and Atmospheric Administration (NOAA) produces forecasts of surface-level ozone (O3), fine particulate matter (PM2.5), and other pollutants so that advance notice and warning can be issued to help individuals and communities limit their exposure. The NAQFC uses the U.S. Environmental Protection Agency (EPA) Community Multiscale Air Quality (CMAQ) model for operational forecasts. This study is a first step in proposing a potential upgrade to the current operational NAQFC bias-correction system, by examining potential candidates for a gridded analysis (“truth”) dataset.In this paper, we compare the performance of the “analysis” time series over the period of August 2020–December 2021 at EPA AirNow stations for both PM2.5 and O3 from raw Copernicus Atmosphere Monitoring Service (CAMS) reanalyses, raw CAMS near real-time forecasts, raw near real-time CMAQ forecasts, bias-corrected CAMS forecasts, and bias-corrected CMAQ forecasts (CMAQ FC BC). This 17-month period spans two wildfire seasons, to assess model “analysis” performance in high-end AQ events. In addition to determining the best-performing gridded product, this process allows us to benchmark the performance of CMAQ forecasts against other global datasets (CAMS reanalysis and forecasts). For both PM2.5 and O3, the bias correction algorithm employed here greatly improved upon the raw model time series, and CMAQ FC BC was the best-performing model “analysis” time series, having the lowest RMSE, smallest bias error, and largest critical success index at multiple thresholds.

Limit hypersurface state-of-the-art damage assessment approach for a galloping energy harvester, accounting for memory effects

Energy harvesters (i.e., EH) constituting nowadays an essential part of renewable energy engineering; hence, apart from numerical modeling, experimental studies are necessary, constituting reliable input for durable design and structural safety assessments. In the current study, galloping EH’s performance was analyzed, utilizing extensive laboratory wind-tunnel tests, conducted under realistic windspeed conditions. The novel structural multivariate risks assessment methodology, presented in the current study is particularly feasible for multi-dimensional nonlinear EH dynamic systems, that have been either directly Monte Carlo (i.e., MC) numerically simulated or physically measured over a representative temporal lapse, providing piecewise ergodic time series. As shown in this analysis, the suggested multivariate methodology accurately enables accurate predictions of the EH dynamic system’s failure/hazard or damage risks, based on laboratory-measured EH system’s dynamics. Furthermore, nonlinear inter-correlations between various systems’ critical components are not always easily handled by classic risk assessment techniques, when dealing with a high-dimensional system’s raw time series. The primary goal of this study was validation and benchmarking the novel risk assessment methodology, which can extract pertinent information, contained within the EH system’s dynamics, based on lab-recorded time histories. In conclusion, the novel hypersurface methodology presented in this study is generic, providing additional capability to accurately, yet efficiently predict damage/failure risks for a variety of nonlinear EH multidimensional systems. Relatively narrow confidence bands have been reported for the forecasted damage and failure levels, indicating both the robustness of the experimental setup, as well as practical design virtues of the advocated Gaidai hypersurface risks assessment methodology. Note that the presented methodology being mathematically exact does not rely on pre-assumptions and yet it is of general purpose.

Bidirectional consistency with temporal-aware for semi-supervised time series classification

Semi-supervised learning (SSL) has achieved significant success due to its capacity to alleviate annotation dependencies. Most existing SSL methods utilize pseudo-labeling to propagate useful supervised information for training unlabeled data. However, these methods ignore learning temporal representations, making it challenging to obtain a well-separable feature space for modeling explicit class boundaries. In this work, we propose a semi-supervised Time Series classification framework via Bidirectional Consistency with Temporal-aware (TS-BCT), which regularizes the feature space distribution by learning temporal representations through pseudo-label-guided contrastive learning. Specifically, TS-BCT utilizes time-specific augmentation to transform the entire raw time series into two distinct views, avoiding sampling bias. The pseudo-labels for each view, generated through confidence estimation in the feature space, are then employed to propagate class-related information into unlabeled samples. Subsequently, we introduce a temporal-aware contrastive learning module that learns discriminative temporal-invariant representations. Finally, we design a bidirectional consistency strategy by incorporating pseudo-labels from two distinct views into temporal-aware contrastive learning to construct a class-related contrastive pattern. This strategy enables the model to learn well-separated feature spaces, making class boundaries more discriminative. Extensive experimental results on real-world datasets demonstrate the effectiveness of TS-BCT compared to baselines.

Adapting the percentage intensity method to assess accelerations and decelerations in football: moving beyond absolute and arbitrary thresholds.

We adapted the percentage intensity approach to monitor accelerations and decelerations allowing players' individualisation. Forty-two players were monitored during four microcycles via global navigation satellite system devices. Raw velocity and time data were collected to calculate acceleration and deceleration magnitudes according to specific starting speed intervals, and the efforts intensities were established as very low (<25% of the maximal effort), low (25-50%), moderate (50-75%) and high (>75%). Linear regressions and Pearson correlation (r) analysed the relationship between maximal efforts and starting speeds; additionally, mean paired differences compared efforts magnitudes between subsequent starting speed intervals. Most very low intensity accelerations (86%) and decelerations (79%) started from <5 km.h-1. Correlation between maximal efforts and starting speeds were r = -0.97 (p < .001) for acceleration, and r = -0.94 (p < .01) for deceleration. Maximal acceleration decreased as starting speed increases (very large effect sizes), but deceleration is less starting speed dependent (unclear to large effect sizes). This adaptation allows practitioners to individualise accelerations and decelerations classification during real-life scenarios, leading to a more precise training prescription. The very low intensity interval could be excluded to consider only relevant efforts. Maximal acceleration should be collected for each starting speed interval because accelerations are starting speed dependents.

A comparative study of principal component analysis and kernel principal component analysis for photogrammetric shape-based turbine blade damage analysis

Photogrammetry is popular as a non-contact full-field data acquisition technique for machine condition monitoring purposes. Two popular implementations thereof are Digital Image Correlation (DIC) and 3D Point Tracking (3DPT). Since these two approaches require surface preparation in the form of applying speckle patterns or discrete markers, their use is negated in several industrial applications, such as the online condition monitoring and/or assessment of horizontal axis wind turbines. A shape-based photogrammetric approach that focuses on analysing changes in the form of a rotating blade’s boundary contour is a viable alternative that does not require any surface preparation. 3D shapes of a blade at a particular instance can be extracted from a pair of stereoscopic images of the blade. Shape Principal Component Descriptors (SPCDs) determined from applying Principal Component Analysis (PCA) to Fourier coefficients of chain-code shape signatures of the blade contour can be determined. These can be regarded as indicators of the form of the shape, and as the blade rotates, variations in the SPCDs can be investigated to better understand the dynamics of the blades. This paper focuses on the application and performance evaluation of data reduction and classification techniques for a novel shape-based photogrammetry condition monitoring technique. Investigations are conducted to show that applying data reduction methods to raw time domain SPCDs can result in successful classification of differently damaged blades. Post-processing strategies are developed to ensure a more robust shape based condition monitoring tool for analysing turbine blades. Kernel Principal Component Analysis (KPCA) is implemented and it is shown that it out-performs the conventional PCA in terms of classifying different blade damage modes. The feasibility of using multi-domain statistical features as feature vectors to which PCA or KPCA is applied for classification purposes is also investigated. Results indicating how well differently damaged blades can be distinguished are provided, and it is clearly illustrated that multi-domain statistical features are more robust to noise contamination in the signals compared to using the raw SPCDs data.

Integrating Feature Selection with Machine Learning for Accurate Reservoir Landslide Displacement Prediction

Accurate prediction of reservoir landslide displacements is crucial for early warning and hazard prevention. Current machine learning (ML) paradigms for predicting landslide displacement demonstrate superior performance, while often relying on various feature engineering techniques, such as decomposing into different temporal lags and feature selection. This study investigates the impact of various feature selection techniques on the performance of ML algorithms for landslide displacement prediction. The Shuping and Baishuihe landslides in China’s Three Gorges Reservoir Area are used to comprehensively benchmark four prevalent ML algorithms. Both static ML models, including backpropagation neural network (BPNN), support vector machine (SVM), and dynamic models, such as long short-term memory (LSTM), and gated recurrent unit (GRU), are included. Each ML model is evaluated under three feature engineering techniques: raw multivariate time series, and feature selection under maximal information coefficient-partial autocorrelation function (MIC-PACF), or grey relational analysis-PACF (GRA-PACF). The results demonstrate that appropriate feature selection methods could significantly improve the performance of static ML models. In contrast, dynamic models effectively leverage inherent capabilities in capturing temporal dynamics within raw multivariate time series, seeing marginal gains with extensive feature engineering compared to no feature selection strategy. The optimal feature selection approach varies based on the ML model and specific landslide, highlighting the importance of case-specific assessments. The findings in this study offer guidance on integrating feature selection techniques with different machine learning models to maximize the robustness and generalizability of data-driven landslide displacement prediction frameworks.

Anomaly detection in hydrological network time series via multiresolution analysis

The territorial extension of Brazil imposes logistic challenges on hydrologic monitoring and increases the chances of systematic observational errors, gauge displacements, downsampling during observation, and low density of streamflow samples on high water stage. Consequently, time series would exhibit artifacts such as gaps, high-frequency anomalies, inhomogeneity, and trends. An innovative anomaly detection system based on multiresolution analysis (MADE) was developed and used to detect synthetic anomalies in raw water stage time series. MADE uses both time and frequency as contextual attributes to identify anomalies. The balance between temporal and frequency resolution allows for capturing low-frequency coarse information and high-frequency details. On MADE, Generalized Extreme Studentized Deviate (GESD) detected short time anomalies on water stage wavelet faster modes (2-, 4-, and 8-days periods). For periods greater than 16 days, Wavelet Coherence and Multiple Wavelet Coherence detected similar water stage time series anomalies. MADE outperformed a Long Short-Term Memory model, a time regression neural network k-Nearest Neighbor model and an Empirical Data Analytics model coupled with clustering techniques on detecting both fast and slow time-scales synthetic anomalies introduced in a São Francisco River water stage time series, using time and frequency domain to pursuit point, collective and contextual anomalies.

Transfer learning with cross-domain adaptation of vibro-acoustic signals for electric motor mechanical fault detection and diagnosis.

Contact-based vibration measurement techniques involving accelerometers are typically utilized for machine health monitoring and manufacturing end-of-line quality control. Acoustic signals, encompassing sound pressure and particle velocity, carry crucial information about the operational status of underlying equipment and can be utilized for detecting various machine faults. Utilizing one-dimensional Convolution Neural Networks (1D-CNN) based on deep learning for processing raw acoustic time signals has proven to be a valuable approach in detecting machinery faults, offering an effective alternative to using traditional accelerometer signals. However, the success of deep learning methods is heavily dependent on the quantity and availability of data for robust network training. When transitioning to new non-contact type sensor technology, manufacturers may face challenges due to insufficient data for training deep learning models. In such cases, leveraging existing historical accelerometer-based data becomes crucial. Transfer learning emerges as a promising solution, allowing deep learning models trained on a source domain (SD) to be applied to a separate but related target domain (TD). This paper hence introduces a domain adaptation methodology by implementing transfer learning, where the SD 1D-CNN network is trained on accelerometer signals, then the trained model weights are transferred to the TD 1D-CNN network that can process acoustic signals.

Analyzing variation of water inflow to inland lakes under climate change: Integrating deep learning and time series data mining

The alarming depletion of global inland lakes in recent decades makes it essential to predict water inflow from rivers to lakes (WIRL) trend and unveil the dominant influencing driver, particularly in the context of climate change. The raw time series data contains multiple components (i.e., long-term trend, seasonal periodicity, and random noise), which makes it challenging for traditional machine/deep learning techniques to effectively capture long-term trend information. In this study, a novel FactorConvSTLnet (FCS) method is developed through integrating STL decomposition, convolutional neural networks (CNN), and factorial analysis into a general framework. FCS is more robust in long-term WIRL trend prediction through separating trend information as a modeling predictor, as well as unveiling predominant drivers. FCS is applied to typical inland lakes (the Aral Sea and the Lake Balkhash) in Central Asia, and results indicate that FCS (Nash-Sutcliffe efficiency = 0.88, root mean squared error = 67m³/s, mean relative error = 10%) outperforms the traditional CNN. Some main findings are: (i) during 1960–1990, reservoir water storage (WSR) was the dominant driver for the two lakes, respectively contributing to 71% and 49%; during 1991–2014 and 2015–2099, evaporation (EVAP) would be the dominant driver, with the contribution of 30% and 47%; (ii) climate change would shift the dominant driver from human activities to natural factors, where EVAP and surface snow amount (SNW) have an increasing influence on WIRL; (iii) compared to SSP1-2.6, the SNW contribution would decrease by 26% under SSP5-8.5, while the EVAP contribution would increase by 9%. The findings reveal the main drivers of shrinkage of the inland lakes and provide the scientific basis for promoting regional ecological sustainability.

A multi-stage sub-structural damage localization approach using multi-label radial basis function neural network and auto-regressive model parameters

This paper proposes a Structural Health Monitoring (SHM) approach to detect localized damage by employing a Multi-Label Radial Basis Function Neural Network (ML-RBFNN). The proposed research methodology aims to identify structural damage more efficiently. This involves dividing the main structure into smaller sub-structures known as zones instead of employing a global search for damage across the entire structure. Each zone is further divided into smaller groups, including sub-zones, nodes, or elements. To reduce the complexities associated with feature extraction, auto-regressive (AR) model parameters extracted from raw time series are considered damage-sensitive features. Considering the simultaneous effects of multiple sensors can increase the number of inputs to the networks, making it challenging to reduce the complexity of the dataset. To address this problem, the principal component analysis (PCA) method is utilized to decrease the dimension of feature space. The effectiveness of the proposed approach is evaluated through a comprehensive comparative analysis against conventional single-stage damage detection methods, with a focus on computational costs and accuracy. Furthermore, the noise-sensitivity characteristics of the proposed method are explored, providing insights into their robustness under various conditions. The validation of the approach is carried out through numerical modeling using the ASCE benchmark structure and experimental data collected from the Qatar University Grandstand Simulator. The application of ML-RBFNN with AR parameters as features yielded a remarkable accuracy exceeding 98% in diagnosing damage occurrence and location for both multi-stage and single-stage SHM methods. Moreover, our proposed approach demonstrates enhanced computational efficiency compared to the single-stage method.