Fault Source Research Articles

3-D Simulations of earthquakes rupture jumps: 1. Homogeneous pre-stress conditions

SUMMARYObservational and modelling studies indicate that earthquake ruptures can jump between fault sections as large as ∼3 and ∼5 km for compressional and extensional offsets, respectively. Here, we compare characteristics of the rupture jump process on parallel but offset fault sections from traditional 3-D dynamic rupture simulations governed by slip weakening friction using the finite element code, FaultMod, to those from quasi-dynamic simulations governed by rate- and state-dependent friction (rate-state friction) using the code RSQSim. These simulations use spatially uniform initial stresses. For a variety of measures the rupture renucleation position on the offset fault, the rate-state friction and slip weakening friction models produce very similar results. The principal difference is the additional occurrence of delayed rupture jumps that arise from the time- and stress-dependent nucleation that is characteristic of rate-state friction. For immediate rupture jumps, models with slip weakening friction span greater offsets than those with rate-state friction. However the jump distances are nearly identical when delayed rupture jumps are included in the comparisons. We propose that delayed rupture jumps are the likely mechanism for adjacent large-earthquake pairs and clusters. Based on the similarity of renucleation positions with both dynamic and quasi-dynamic models, we conclude that the renucleation positions for rupture initiation on the receiver fault (separated by less than ∼3 km from the source fault) are primarily controlled by static stress changes induced by slip on the initiating fault. However, in light of the slightly greater maximum jump distances (>3 km) seen with the dynamic slip weakening friction model, dynamic stress changes from seismic waves play an increasingly important role as offset distances increase.

Fault-based probabilistic seismic hazard analysis in regions with low strain rates and a thick seismogenic layer: a case study from Malawi

SUMMARY Historical and instrumental earthquake catalogs in low strain rate regions are not necessarily indicative of the long-term spatio-temporal distribution of seismicity. This implies that probabilistic seismic hazard analysis (PSHA) should also consider geologic and geodetic data through fault-based seismogenic sources. However, it is not always clear how on-fault magnitude-frequency distributions (MFDs) should be described and, if the seismogenic layer is especially thick, how fault sources should be extrapolated down-dip. We explore these issues in the context of a new PSHA for Malawi, where regional extensional rates are 0.5–2 mm yr−1, the seismogenic layer is 30–40-km thick, the instrumental catalog is ∼60 yr long and fault-based sources were recently collated in the Malawi Seismogenic Source Model. Furthermore, Malawi is one of several countries along the East African Rift where exposure to seismic hazard is growing, but PSHA does not typically consider fault sources. We use stochastic event catalogs to explore different fault source down-dip extents and MFDs. Our PSHA indicates that hazard levels are highest for a Gutenberg–Richter on-fault MFD, even at low probabilities of exceedance (2 per cent in 50 yr), whilst seismic hazard levels are also sensitive to how relatively short (<50 km) fault sources are extrapolated down-dip. For sites close to fault sources (<40 km), seismic hazard levels are doubled compared to previous instrumental-seismicity based PSHA in Malawi. Cumulatively, these results highlight the need for careful fault source modelling in PSHA of low strain rate regions and the need for new fault-based PSHA elsewhere in the East Africa Rift.

Universal source-free domain adaptation method for cross-domain fault diagnosis of machines

Cross-domain machinery fault diagnosis aims to transfer enriched diagnosis knowledge from a labeled source domain to a new unlabeled target domain. Most existing methods assume that the prior information on the fault modes of the target domain is known in advance. However, in engineering practice, prior knowledge of fault modes is rare in a new domain, in which there may be only partial source fault modes or some new fault modes. Furthermore, up to the present, almost all existing cross-domain fault diagnosis methods require the labeled source data during the model training process, which restricts their deployment on certain devices with limited computing resources. To this end, we propose a universal source-free domain adaptation method that can handle cross-domain fault diagnosis scenarios without access to the source data and is free of explicit assumptions about the target fault modes. More specifically, we develop a convolutional network with a Transformer as the attention module to extract discriminative feature information from the source data and then send the model and parameters to the target domain. In target domain training, we first propose a supervised contrastive learning strategy based on source class prototypes, which utilizes high-confident predictions to achieve source-free domain alignment and class alignment. Then, we also introduce a threshold-based entropy max–min loss to further align known class samples in the target domain or reject target outlier samples as an unknown class. Furthermore, we introduce self-supervised learning to further learn feature representations of the target domain to reduce the previous misclassification. A series of experiments on two rotating machine datasets demonstrate the effectiveness and practicability of the proposed method.

Analysis of fault characteristics and oil and gas enrichment differences in the No. 2 structural belt, Nanpu Sag, Bohai Bay Basin, East China

There are great differences in oil and gas enrichment at different locations in petroliferous basins, and faults are an important factor affecting the differential enrichment of oil and gas. In this paper, taking the No. 2 structure in the Nanpu Sag, Bohai Bay Basin, East China, as an example, based on the analysis of fault geometry and kinematics characteristics, the determination of oil source faults, the analysis of fault trap origin and effectiveness, and the analysis of fault caprock combinations are carried out to study the differences and mechanisms of oil and gas enrichment east and west of the Nanpu No. 2 structure. The results show that the fault-source contact relationship affects the direction of oil and gas migration. The No. 2 structure in the Nanpu Sag has a dual-source hydrocarbon supply. The western structure supplies hydrocarbons from the north and east, with oil and gas migrating along the direction parallel and perpendicular to the fault strike. The east structure supplies hydrocarbons from the south and north, with oil and gas migrating along the direction perpendicular to the fault strike. The effectiveness of fault traps affects the scale of oil and gas accumulation. Most of the traps of the Nanpu No. 2 structure formed before hydrocarbon expulsion, which are all effective traps and are conducive to oil and gas accumulation. The relationship between fault and caprock determines the vertical accumulation layers of oil and gas: The faulted thickness of the east structure is large, the fault is vertically closed, and oil and gas are enriched mainly in the deep layer. The faulted thickness of the west structure is small, and oil and gas are enriched in both deep and shallow layers.

One-Dimensional Residual GANomaly Network-Based Deep Feature Extraction Model for Complex Industrial System Fault Detection

In the era of industrial big data, traditional shallow machine learning-based data analytical technologies can not handle complex industrial system fault detection issue effectively. In regard of this problem, this paper presents a deep learning-based fault detection method, called one-dimensional residual GANomaly (1DRGANomaly). This method builds a semi-supervised deep feature extraction mechanism where the unsupervised generator network is utilized to capture the latent data representation and the semi-supervised discriminator network is adopted to improve the reconstruction quality. Furthermore, one-dimensional convolution operation is applied to mine the local variable relationship and the residual learning blocks are designed for the efficient training. Based on the deep features and the reconstruction errors, the Bayesian inference is applied to construct two overall statistics for monitoring the process status. In order to locate the possible fault source variables, an input importance attribution method is developed by referring to the idea of the Shapley value, which gives a reasonable explanation for the detected fault. Finally, a benchmark industrial process, the Tennessee Eastman system, is applied to perform the method comparison and the application results demonstrate that the 1DRGANomaly method can detect the faults more effectively than the other existing methods.

A novel integrated fault diagnosis method of chemical processes based on deep learning and information propagation hysteresis analysis

BackgroundTo guarantee the stationary operation of chemical plant, the anomaly symptoms of equipment should be discovered as soon as possible. Due to the limited ability of single fault diagnosis method in the increasingly complex industrial process, the integration of different methods has become a hot research direction. MethodsA three-layer fusion fault diagnosis method is proposed in this paper. This method consists of three layers. Time domain features of time series process variables are first extracted by sliding window. Then, the bidirectional long short-term memory network based on time domain feature extraction (T-BiLSTM) is formed to accomplish the online fault identification task. At last, the causalities of fault variables are obtained based on improved signed directed graph model with dynamic time warping (TD-SDG) to locate the fault sources. FindingsThe application to Tennessee Eastman (TE) shows that the average false positive rate (FPR) of T-BiLSTM is 1.98%, and the average fault diagnosis rate (FDR) is 96.6%. TD-SDG model realizes fault inference and location. This novel integrated fault diagnosis method thus can identify fault types and locate fault sources quickly based on graphical scenario object model with a high application capacity.

Spatial change in differential stress magnitudes around the source fault before intraplate earthquakes

SUMMARYHow are the sizes of the earthquakes determined? To solve this important problem, we analysed the data from a dense temporary seismic observation network installed in the aftershock area of the 2016 Mw 6.2 Central Tottori earthquake, which occurred in an intraplate region in Japan. We compared the stress field estimated from approximately 10 000 accurate focal mechanisms of aftershocks with the calculated post-earthquake stress field and found that the differential stress before the earthquake was very small near both horizontal edges. These results did not depend significantly on the modeled slip distribution and the orientation of the principal stress before the earthquake. Similar results were obtained for the 2000 Mw 6.7 Western Tottori earthquake, which also occurred in the same intraplate region in Japan. These results suggest that the fault size of large intraplate earthquakes can be determined by the region of small differential stress surrounding future earthquake faults.

Digital Twin framework for automated fault source detection and prediction for comfort performance evaluation of existing non-residential Norwegian buildings

Numerous buildings fall short of expectations regarding occupant satisfaction, sustainability, or energy efficiency. In this paper, the performance of buildings in terms of occupant comfort is evaluated using a probabilistic model based on Bayesian networks (BNs). The BN model is founded on an in-depth analysis of satisfaction survey responses and a thorough study of building performance parameters. This study also presents a user-friendly visualization compatible with BIM to simplify data collecting in two case studies from Norway with data from 2019 to 2022. This paper proposes a novel Digital Twin approach for incorporating building information modeling (BIM) with real-time sensor data, occupants’ feedback, a probabilistic model of occupants’ comfort, and HVAC faults detection and prediction that may affect occupants’ comfort. New methods for using BIM as a visualization platform, as well as a predictive maintenance method to detect and anticipate problems in the HVAC system, are also presented. These methods will help decision-makers improve the occupants’ comfort conditions in buildings. However, due to the intricate interaction between numerous equipment and the absence of data integration among FM systems, CMMS, BMS, and BIM data are integrated in this paper into a framework utilizing ontology graphs to generalize the Digital Twin framework so it can be applied to many buildings. The results of this study can aid decision-makers in the facility management sector by offering insight into the aspects that influence occupant comfort, speeding up the process of identifying equipment malfunctions, and pointing toward possible solutions.

Automatic inversion of magnetic anomalies caused by 2D listric fault sources with arbitrary magnetisation

Automatic inversion of magnetic anomalies caused by 2D listric fault sources with arbitrary magnetisation

Hybrid Intelligent Fault Diagnosis Model Based on Improved MPCA-V for Sensors in a Laboratory-Scale Wastewater Treatment Process

The study objective is to propose a hybrid fault diagnosis method for a laboratory-scale sequential batch reactor (SBR) wastewater treatment process based on time-varying covariance and variable-wise unfolded MPCA method (MPCA-V), which can detect the fault batch, determine the fault time simultaneously, and further identify the fault source. To establish and validate the MPCA-V model, 50 normal batches and 55 batches including 7 fault batches were employed separately. Furthermore, the classical MPCA (MPCA-B) model was introduced for comparison. For the three detected fault batches, with the MPCA-V model, not only the fault occurring time and fault source were located and identified by the contribution degree calculation of each variable to the T2 and SPE statistics simultaneously but also the fault detection rate was averaged as 90%, which was much higher than that of MPCA-B (67%). Introducing time dependency and correlation in a laboratory-scale SBR process gives the work practical significance and breakthrough.

Acoustic Feature Extraction Method of Rotating Machinery Based on the WPE-LCMV

Fault diagnosis plays an important role in the safe and stable operation of rotating machinery, which is conducive to industrial development and economic improvement. However, effective feature extraction of rotating machinery fault diagnosis is difficult in the complex sound field with characteristics of reverberation and multi-dimensional signals. Therefore, this paper proposes a novel acoustic feature extraction method of the rotating machinery based on the Weighted Prediction Error (WPE) integrating the Linear Constrained Minimum Variance (LCMV). The de-reverberation signal is obtained by inputting multi-channel signals into the WPE algorithm using an adaptive optimal parameters selection function with the sound field changes. Then, the incident angle going from the fault source to the center of the microphone array is calculated from the full-band sound field distribution, and the signal is de-noised and fused using the LCMV. Finally, the fault feature frequency is extracted from the fused signal envelope spectrum. The results of fault data analysis of the centrifugal pump test bench show that the Envelope Harmonic Noise Ratio (EHNR) is more than twice that of the original signal after the WPE-LCMV processing. Compared to the Recursive Least Squares and the Resonance Sparse Signal Decomposition (RLS-RSSD) and the parameter optimized Variational Mode Decomposition (VMD), the EHNR has a higher value for all types of faults after applying the WPE-LCMV processing. Furthermore, the proposed method can effectively extract the frequency of bearing faults.

An AI-Empowered Framework for Cross-Layer Softwarized Infrastructure State Assessment

Network softwarization technologies challenge legacy fault management systems. Coordination and dependency among different novel software components for orchestration, switching, virtual machine and container management creates novel monitoring points, besides novel sources of faults, bugs and vulnerabilities. To cope with the high heterogeneity and granularity of software components, we propose a modular network AI framework to detect anomalies, toward closed-loop automation. We design an AI-empowered anomaly detection framework able to assess the running state and the state deviations of a softwarized infrastructure, monitored through different features grouped depending on their layer and connect-compute stack component. Our framework learns the nominal working conditions of the infrastructure, with respect to which anomalies are detected, and characterized tracing back the layer and component root cause; it includes a network state assessment technique that leverages anomalies characterization through their most visible symptoms. We implement and validate the proposed framework through experimental tests on a containerized IP Multimedia Subsystem platform.

Pressure changes of nitrogen blanket transformers under internal short circuit faults

The root cause of the transformer fire is an oil tank crack caused by an internal short circuit fault. The gas layer of the nitrogen blanket transformer helps buffer the pressure change caused by the internal short-circuit fault. Moreover, the pressure change simulation is the basis for the configuration of non-electrical protection. This paper analyzes the relationship between the energy and pressure source of the internal short-circuit fault. Based on finite element simulation, the internal pressure transfer model of the transformer is built, and the pressure of the oil tank is obtained. By comparing with the pressure of conventional transformer, the excellent explosion-proof performance of nitrogen blanket transformer is verified.

Machine learning opportunities to conduct high-fidelity earthquake simulations in multi-scale heterogeneous geology

The 2019 Le Teil earthquake is an illustrative example of a moderate (MW4.9) yet damaging event, occurring at shallow depth (≈1 km) in a region with little to no geophysical data available. Therefore, using a high-fidelity wave propagation code, we performed numerical simulations of the Le Teil earthquake in a highly uncertain framework, investigating several seismic sources and geological set-ups. With respect to the former aspect, a point-source model and an extended kinematic fault model were compared. The latter aspect was investigated by comparing a 1D-layered to a 3D geological model. Those models were enhanced with random fluctuations, in order to obtain three alternative non-stationary random geological fields. The synthetic waveforms obtained from regional geophysical models were globally coherent with the recorded ones. The extended fault source model seemed more realistic than the point-source model. In addition, some geological random fields improved the synthetics’ agreement with the recordings. However, the three random field samplings led to a high variability in induced ground motion responses. Given the computational burden of high-fidelity simulations, we used two dimensionality reduction methods, namely the Principal Component Analysis (PCA) and a deep neural network (3D UNet), to investigate this variability. The methods were applied to a database of 40,000 3D geological random fields. Both the PCA and the 3D UNet condensed the variability of the 3D geological fields into a few components. These were sufficient to reconstruct the original fields with great accuracy. More importantly, the seismic response arising from the propagation throughout the reconstructed fields was in excellent agreement with the response of the original geological fields in more than 75% of the dataset. By building a structured ensemble of complex geological fields from their reduced representation, it may become possible to find a relationship between the reduced representation and the generated ground motion. Thus, our study proves the interest of dimensionality reduction to perform uncertainty analyses in complex geological media.

Vibration Analysis of a Roller Bearing Condition Used in a Tangential Threshing Drum of a Combine Harvester for the Smooth and Continuous Performance of Agricultural Crop Harvesting

Testing the reliability of the threshing unit is difficult and thus often neglected before the harvesting season, which can result in breakdown maintenance during peak harvesting time in difficult-to-access areas for sensor mounting. In this paper, the vibration analysis of the threshing condition of the combine harvester was performed by introducing the bracket for inaccessible locations. The Adash A4900 Vibrio M analyzer (Adash spol. s.r.o., Ostrava, Czech Republic) was used for a vibration signal measurement and the DDS Adash software was used for signal processing. The demodulated fast Fourier transform (FFT) root mean square (RMS) (500 Hz–16 kHz) method was used to evaluate the bearing condition and DDS Adash Fault Source Identification Tool (FASIT) technology was used to evaluate other mechanical conditions such as the looseness, misalignment, and unbalance of the threshing unit of the Massey Ferguson series of combine harvesters. Modal and random vibration analyses were performed on the bracket and components and compared to prevent the resonance phenomenon using the Ansys Software (Ansys, Inc., Canonsburg, PA, USA). The constrained modal analysis of the threshing drum was performed to observe the deformation. Decent results were obtained from the method used. The method was again validated by the tangential threshing test bench and successfully determined the bearing fault condition. The method used is an uncomplicated and effective way of performing the bearing analysis of the tangential unit of the combine harvester.

Improving urban seismic risk estimates for Bishkek, Kyrgyzstan, through incorporating recently gained geological knowledge of hazards

Many cities are built on or near active faults, which pose seismic hazard and risk to the urban population. This risk is exacerbated by city expansion, which may obscure signs of active faulting. Here, we estimate the risk to Bishkek city, Kyrgyzstan, due to realistic earthquake scenarios based on historic earthquakes in the region and an improved knowledge of the active fault sources. We use previous literature and fault mapping, combined with new high-resolution digital elevation models to identify and characterise faults that pose a risk to Bishkek. We then estimate the hazard (ground shaking), damage to residential buildings and distribution of losses (economical cost and fatalities) using the Global Earthquake Model OpenQuake engine. We model historical events and hypothetical events on a variety of faults that could plausibly host significant earthquakes. This includes proximal, recognised, faults as well as a fault under folding in the north of the city that we identify using satellite DEMs. We find that potential earthquakes on faults nearest to Bishkek—Issyk Ata, Shamsi Tunduk, Chonkurchak and the northern fault—would cause the most damage to the city. An Mw 7.5 earthquake on the Issyk Ata fault could potentially cause 7900 ± 2600 completely damaged buildings, a further 16,400 ± 2000 damaged buildings and 2400 ± 1500 fatalities. It is vital to properly identify, characterise and model active faults near cities to reduce uncertainty as modelling the northern fault as a Mw 6.5 instead of Mw 6.0 would result in 37% more completely damaged buildings and 48% more fatalities.

Associated Fault Diagnosis of Power Supply Systems Based on Graph Matching: A Knowledge and Data Fusion Approach

With the rapid development of more-electric and all-electric aircraft, the role of power supply systems in aircraft is becoming increasingly prominent. However, due to the complex coupling within the power supply system, a fault in one component often leads to parameter abnormalities in multiple components within the system, which are termed associated faults. Compared with conventional faults, the diagnosis of associated faults is difficult because the fault source is hard to trace and the fault mode is difficult to identify accurately. To this end, this paper proposes a graph-matching approach for the associated fault diagnosis of power supply systems based on a deep residual shrinkage network. The core of the proposed approach involves supplementing the incomplete prior fault knowledge with monitoring data to obtain a complete cluster of associated fault graphs. The association graph model of the power supply system is first constructed based on a topology with characteristic signal propagation and the associated measurements of typical components. Furthermore, fault propagation paths are backtracked based on the Warshall algorithm, and abnormal components are set to update and enhance the association relationship, establishing a complete cluster of typical associated fault mode graphs and realizing the organic combination and structured storage of knowledge and data. Finally, a deep residual shrinkage network is used to diagnose the associated faults via graph matching between the current state graph and the historical graph cluster. The comparative experiments conducted on the simulation model of an aircraft power supply system demonstrate that the proposed method can achieve high-precision associated fault diagnosis, even under circumstances where there are an insufficient number of samples and missing parameters.

The Luangwa Rift Active Fault Database and fault reactivation along the southwestern branch of the East African Rift

Abstract. Seismic hazard assessment in slow straining regions is challenging because earthquake catalogues only record events from approximately the last 100 years, whereas earthquake recurrence times on individual faults can exceed 1000 years. Systematic mapping of active faults allows fault sources to be used within probabilistic seismic hazard assessment, which overcomes the problems of short-term earthquake records. We use Shuttle Radar Topography Mission (SRTM) data to analyse surface deformation in the Luangwa Rift in Zambia and develop the Luangwa Rift Active Fault Database (LRAFD). The LRAFD is an open-source geospatial database containing active fault traces and their attributes and is freely available at https://doi.org/10.5281/zenodo.6513691. We identified 18 faults that display evidence for Quaternary activity, and empirical relationships suggest that these faults could cause earthquakes up to Mw 8.1, which would exceed the magnitude of historically recorded events in southern Africa. On the four most prominent faults, the median height of Quaternary fault scarps varies between 12.9 ± 0.4 and 19.2 ± 0.9 m, which suggests they were formed by multiple earthquakes. Deformation is focused on the edges of the Luangwa Rift: the most prominent Quaternary fault scarps occur along the 207 km long Chipola and 142 km long Molaza faults, which are the rift border faults and the longest faults in the region. We associate the scarp on the Molaza Fault with possible surface ruptures from two 20th century earthquakes. Thus, the LRAFD reveals new insights into active faulting in southern Africa and presents a framework for evaluating future seismic hazard.

A weight recognition method for movable objects in sealed cavity based on supervised learning

Weight information has important reference value for the analysis of the fault source of measurement and industry. Whenthe object in the sealing cavity collides with the inner wall, it will produce undetectable weak sound signal containing weight information. Thus this paper presents a novel method to recognize the weight information of movable object in a sealed cavity. In this study, we compared and analyzed a variety of sound features, and a method combining qualitative and quantitative analysis is proposed to recognize the feature datasets that retained by feature selection. The support vector machine (SVM) algorithm was first adopted to determine the approximate range of object weight. Thereafter, the Multilayer Perceptron (MLP) based regression model was constructed to recognize the exact weight. Finally, weight detection experiment of movable particles from the sealed electronic components is utilized to verify the effectiveness of the proposed method.

Earthquake Threats in Ranau – From The Sources of Mensaban and Mesilou Fault

Unlike the majority of Malaysia’s other states, Sabah is characterised by frequent seismological activity; on average, an earthquake of moderate magnitude occurs roughly every 20 years, originating primarily from one of two major sources: either a local source or a regional source. Sabah has seen an upsurge in low to moderate seismic activity in recent years as a result of the fault activities. Between 1900 and lately, magnitudes ranging from MW 2.9 to 6.0 were recorded. While big magnitude earthquakes are relatively uncommon, the area has previously been struck by disastrous earthquakes. Fortunately, the region is demarcated by active Quaternary fault networks. As a result, the area’s seismicity is calculated using line sources corresponding to these faults. Two major fault systems are suspected of being the cause of such activity: the Mensaban fault zone and the Mesilou fault zone, as well as several additional nearby faults. This article explains the process for doing a probabilistic seismic hazard analysis (PSHA) while taking into consideration the peak ground acceleration (PGA) on bedrock in the Ranau region for a 10% and 2% chance of exceedance. The PGA estimate values for Ranau are between 0.08g to 0.16g for PGA 10%, while for PGA 2% was between 0.15g to 0.28g. The PGA value is divided into the following five earthquake hazard classes based on a quarterly geostatistical analysis method. These are “very low,” “low,” “medium,” “high,” and “very high” classes. In general, this quarterly classification is a good way to see how much local seismic activity there is, based on the PGA value. Which can be seen in 10% probability, 19.88 percent of the study area was in “very low,” 20.19 percent was low, 19.98 percent medium, and 19.92 percent high class. 2% probability, it was found that 19.74 percent of the total area was very low class, 20.06 percent low class, 19.79 percent middle class, and 19.82 percent high class. Only 20.6 percent of the area was “very high”.