Downstream Fault Research Articles

DA-VICReg: a data augmentation-free self-supervised learning approach for diesel engine fault diagnosis

Self-supervised learning (SSL) aims to extract useful representations from unlabeled data by maximizing the agreement between positive pairs. However, traditional SSL relies on carefully designed data augmentation methods to generate positive pairs. When dealing with 1D vibration signals, data augmentation prone to potentially compromise the fault information in the original signals. Therefore, this paper proposes a data augmentation-free SSL framework for diesel engine fault diagnosis called Domain Adaptation Variance Invariance Covariance Regularization (DA-VICReg). The DA-VICReg uses cyclic angular vibrations collected within the same time period as positive pairs and extracts useful features from unlabeled data using a loss function composed of three terms: Variance, Invariance, and Covariance. We found that when positive pairs originate from different operating conditions, such as varying speeds and torques, the model can develop feature extraction capabilities that remain unaffected by changes in operating conditions. In addition, a spatial pyramid pooling layer and a trilinear attention module are used to extract vibration features at different scales and focus on critical spatial locations and channels. Finally, the proposed approach was validated through experiments on two types of diesel engines, and a comparison with prominent SSL methods confirms the superiority of the proposed approach. In engineering practice, this method can utilize a large amount of signals stored in different time periods for self-supervised training and learn useful features for downstream fault diagnosis tasks.

A novel self-supervised representation learning framework based on time-frequency alignment and interaction for mechanical fault diagnosis

Recently, supervised learning methods for intelligent mechanical fault diagnosis have achieved considerable advancements, while they heavily rely on labeled information and neglect massive unlabeled signals. Consequently, some self-supervised research has emerged and employs contrastive learning to mine general knowledge from unlabeled samples. However, these methods emphasize independent feature extraction and lack deep feature interaction, which limits better representation learning. Furthermore, it also restricts the performance improvement of fault diagnosis. To fill this gap, a novel self-supervised representation learning framework based on time-frequency alignment and interaction (TFAI) is proposed to improve the diagnosis reliability under limited labeled data. Comprising a dual encoder and a cross-modal encoder based on Transformers, the proposed TFAI model takes paired time-frequency data as input for self-supervised pretraining. Two self-supervised learning strategies, namely time-frequency alignment and time-frequency interaction, are proposed to enable the model to gain valuable knowledge from unlabeled data. The former leverages a time-frequency contrastive loss to drive the dual encoder to extract profound features from unlabeled data. The latter involves utilizing a time-frequency matching loss to guide the cross-modal encoder to perform deep feature fusion in an unsupervised manner. The well pretrained TFAI model can learn informative representations and can be fine-tuned for specific downstream fault diagnosis tasks to improve the diagnosis accuracy. By conducting three experimental cases on an axial flow pump system and a public gearbox platform system, the effectiveness, generalizability, and domain adaptability of the proposed method are validated.

Facilitating Energy Monitoring and Fault Diagnosis of Pneumatic Cylinders with Exergy and Machine Learning

Pneumatic systems are widely used in industrial production sectors. Increasing penetrations of Intelligent Manufacturing and Green Manufacturing are highlighting the drawbacks of pneumatic technology in terms of particularly low energy efficiency and low-level fault diagnosis intelligence. Here we propose that a combined energy-based maintenance and fault diagnostic approach for pneumatic systems could be a game-changer for pneumatics. In this study, a pneumatic cylinder with internal and external leakages is examined and a typical pneumatic experimental system is built. Exergy is adopted for evaluating the available energy of compressed air. Data-driven machine learning models, SAE + SoftMax neural network model and SAE + SVM model, are developed for fault detection and diagnosis. By comparing different machine learning methods with various pressure, flowrate, and exergy data, it is found that the diagnostic accuracy when using pressure and flowrate data is highly dependent on operating conditions, while the diagnostic accuracy when using exergy data is always high regardless of operating conditions. This indicates the promise of developing an exergy-based maintenance paradigm in pneumatic systems. Besides, with exergy and machine learning, more downstream faults can be detected and diagnosed with fewer upstream sensors. This study is the first attempt to develop an exergy-based maintenance paradigm in pneumatic systems. We hope it could inspire the following investigations in other energy domains.

Distributed transmission technology of node status in distribution network

With the rapid development of distribution network automation and the construction of DC distribution network and AC / DC hybrid distribution network, the network operation mode is more flexible. If the distribution network wants to realize rapid load transfer, such as the connection of power supply, upstream and downstream fault isolation signals and other node states need to be quickly identified. At present, the node status identification of distribution automation is basically completed by the master station. However, the master station has poor real-time transmission of status information, which is generally in seconds or minutes, and can not meet the demand of real-time load transfer. In this paper, a distributed real-time transmission method of AC and DC distribution network node status is constructed for distribution automation system. Based on IEC61850 goose communication protocol, the real-time transmission of node status is realized in distribution automation terminal. In this method, the positive direction of each tie switch is defined, which is independent of power flow direction, and the tie switch is distinguished along the user-defined positive direction When multiple tie switches have the same upstream node, they are each other’s t nodes. Each tie switch needs to be equipped with a distribution automation terminal with goose communication capability. According to different control strategies, the distribution automation terminal can realize millisecond level node state transmission step by step. Node state can be divided into unconditional transmission state and conditional transmission state. The distributed real-time transmission technology of distribution network node state proposed in this paper has been successfully applied to Tongli integrated energy service The center and Suzhou medium and low voltage DC project lay the foundation for the realization of loop closing, loop breaking control, fast fault isolation and power transfer control strategies.

Multivariate SPC methods for controlling manufacturing processes using predictive models – A case study in the automotive sector

The main objective is to develop and test a novel SPC method to ensure the stability of final customer characteristics by controlling the upstream characteristics of manufacturing processes, considering their importance/contribution. The originality of the proposed method lies in the use of predictive models (multivariate regression), whose coefficients are used to weigh the contribution of each upstream characteristic to predict the final characteristics. In the context of continuous improvement environments, the application of different SPC approaches to manufacturing processes were compared. The results showed that the multivariate SPC method based on partial least squares regression was superior to traditional univariate and multivariate SPC methods in terms of the predictive precision to detect downstream faults in customer characteristics. However, the use of multiple linear regression may also be an option, since the identification of what upstream characteristic is causing the out-of-control signal is simpler than that of partial least squares regression, and predictive precision in the case of each of the two methods is comparable in practical terms.

Co-Optimization of Protection Coordination and Power Quality in Microgrids Using Unidirectional Fault Current Limiters

This paper proposes a framework to simultaneously optimize the power quality (PQ) and protection coordination (PC) levels in a distributed energy resource (DER)-enabled microgrid with multiple connection modes. Introduction of DERs in a grid increases the fault currents. Different connection modes for these DERs lead to varying fault levels. Both phenomena jeopardize the PC. A solid-state unidirectional fault current limiter (SSUFCL) is used as an interface between the microgrid as downstream and the utility as upstream networks, which limits the downstream contribution to the upstream fault currents for increasing the PC and PQ margins, while it is disabled in the case of downstream faults to improve the PQ. In this way the complete coordination between upstream and downstream networks will be achieved. To simultaneously optimize the PC and PQ, the time and current settings of the over current relays (OCRs) and the characteristics of the SSUFCL are set. The optimization formulation of the PC problem is extracted. Fuzzy optimization technique is used to make a compromise between PQ and PC after objectives’ fuzzification. Genetic algorithm is used to find the proper characteristics of the SSUFCL as well as the best settings of OCRs.

Virtual impedance control for a regional dynamic voltage restorer

ABSTRACTA dynamic voltage restorer (DVR) can be installed in a middle-voltage (MV) power grid, to concurrently protect a cluster of sensitive loads from voltage sags. To further improve its efficiency and reduce the difficulty in its implementation, a novel control strategy for operating such a DVR as a virtual impedance in series with sensitive loads is proposed in this paper. In addition to its usual function of compensating for voltage sags, such a DVR can also operate as a virtual inductance, to function as a fault current limiter (FCL) during a downstream fault, or a virtual capacitance, to function as a series compensator (SC) to compensate the voltage loss along the feeder line during heavy load. Based on a dual-loop control design, strategies for operating a DVR as a series virtual inductance and a virtual capacitance are proposed, and methods for tuning the parameter values and a stability analysis of the whole system are presented. The feasibility and effectiveness of the proposed method are verified by simulations using the PSCAD software, and experimental results obtained using a prototype DVR are presented.

A Single-phase Grounding Fault Judgment Method Based on Steady-state Current Characteristics in Distribution

In order to solve the problem of single-phase grounding fault judgment in non-solid-earthed distribution network, the power flow in non-solid-earthed distribution is analyzed. The steady-state current value in the upstream fault phase will be abruptly changed, and the current in the downstream fault phase will keep the load current constant. The steady-state current values in other non-faulty phases and non-faulty lines do not change abruptly. A single-phase grounding fault judgment method based on steady-state current characteristics is proposed. The power supply path real-time current matrix, the power supply path history current matrix, the power supply path mutation current matrix and the power supply path mutation current logical matrix are established. The problem of the single-phase grounding fault judgment is transformed into the matrix operation problem. The detailed calculation formulas are given. The simulation results show that the method proposed in this paper has a good adaptability to the permanent grounding fault and it is worth popularizing.

A new multifunctional DVR for compensation of voltage sag

Today, power quality is considered as a significant application. i) Highly effectual variable speed drive, ii) sophisticated electronic equipment together with iii) power electronic controller, is utilized to enhance power quality (PQ). Voltage sags are the prevalent PQ disturbances on the distribution system. It occurs on account of the faults developed in the electrical network or by the working of a huge induction motor. It is resolved by utilizing the conventional power device termed DVR (Dynamic Voltage Restorer). It is of small size and also low cost. It proffers quick dynamic response to the disturbance. A multifunctional DVR is utilized here to augment the PQ with the aid of P+ Resonant and Posicast controllers to remove the steady-state error (SSE) and enhance the transient response. Simulation outcome displays that the DVR enables to restrain the emergency scenarios of the distribution systems. The prevailing limitation reinstate the PCC (i.e. point of common coupling) (i.e. the bus to which every feeders are linked) voltage and Defend the DVR itself. The DVR performs as virtual impedance with the chief target of shielding the PCC voltage all through the downstream fault without affecting real power injection in the DVR which is the significant innovation discussed here.

Fault Current Interruption by the Dynamic Voltage Restorer

This paper introduces and evaluates an auxiliary control strategy for downstream fault current interruption in a radial distribution line by means of a dynamic voltage restorer (DVR). The proposed controller supplements the voltage-sag compensation control of the DVR. It does not require phase-locked loop and independently controls the magnitude and phase angle of the injected voltage for each phase. Fast least error squares digital filters are used to estimate the magnitude and phase of the measured voltages and effectively reduce the impacts of noise, harmonics, and disturbances on the estimated phasor parameters, and this enables effective fault current interrupting even under arcing fault conditions. The results of the simulation studies performed in the PSCAD/EMTDC software environment indicate that the proposed control scheme: 1) can limit the fault current to less than the nominal load current and restore the point of common coupling voltage within 10 ms; 2) can interrupt the fault current in less than two cycles; 3) limits the dc-link voltage rise and, thus, has no restrictions on the duration of fault current interruption; 4) performs satisfactorily even under arcing fault conditions; and 5) can interrupt the fault current under low dc-link voltage conditions.

Robust control design for multi-functional DVR implementation in distribution systems using quantitative feedback theory

One of the most effective tools to prevent the voltage amplitude from a sudden change in distribution systems is the dynamic voltage restorer (DVR). In this paper, a novel voltage control loop using quantitative feedback theory (QFT) is proposed for the emergency control of distribution systems by means of DVR. The proposed multi-functional DVR control strategy is applied to a three-phase fault which is the main cause of the worst voltage sag. The simulation results clearly show that this control strategy would be able to effectively regulate the voltage, especially when the parameters of the system change. The ability of the proposed DVR to limit the downstream fault current has been also examined. This current limitation will restore the Point of Common Coupling (PCC) (the bus to which all feeders are attached) voltage and also protect the DVR itself.

Fault Current Contribution of Medium Voltage Inverter and Doubly-Fed Induction-Machine-Based Flywheel Energy Storage System

One of the main concerns with inverter-based distributed generation (IBDG) is its negligible fault current contribution compared to synchronous generators. IBDG hardly affects fault current level, which shadows conventional protection schemes especially with a high penetration of IBDGs at high power levels and/or in island operation mode. This paper proposes a solution for IBDG in a medium voltage distribution network that allows active and adjustable IBDG fault current contribution without violating the nonoverload nature of the inverters. This is achieved by introducing a flywheel energy storage system based on a doubly-fed induction machine (DFIM) in parallel with the IBDG. Normally, the flywheel system is dedicated to power leveling; however, during faults, the flywheel DFIM storage system (FW-DFIM) has the ability to supply an exponentially decaying current to the grid. The parallel combination of the IBDG and FW-DFIM as a distributed generation (DG) unit is capable of providing a response similar to that of synchronous generators during a fault, but with additional control capability. The simulated system shows the effectiveness of the DG unit for downstream faults where it is capable of providing sufficient fault current to trigger the distribution network's protection devices when operating in island mode.

A Redundancy Design Schema of Distributed Real-Time Database Applied in ISCS

Reliability is one of the most important properties of integrated supervisory and control system (ISCS) in metro. Redundancy technology, a fault tolerant mechanism, can significantly improve ISCS reliability. This paper introduces a redundancy design schema and its implementation in distributed real-time database, which is the kernel part of ISCS, Including upstream and downstream data redundancy processing technology, fault detection and redundancy switch technology. The result shows that this schema is feasible and reasonable.

Sensitive load voltage compensation against voltage sags/swells and harmonics in the grid voltage and limit downstream fault currents using DVR

In the past decades, the power quality (PQ) has become a critical issue in highly automated industries and sensitive load centers. The voltage quality is the most important part of the PQ for customers. The voltage disturbances in the form of voltage sags, swells and harmonics can cause huge financial losses. In the past few years, the power electronic solutions have been proposed to avoid these problems. The dynamic voltage restorer (DVR) is one of the solutions. This paper proposes a new topology for single-phase DVR which is able to compensate multiple voltage disturbances and limit the downstream fault currents with a simple topology. The proposed DVR is based on the single-phase direct ac/ac converter which is energized from the main grid. As a result, there is no need for energy storage elements and dc link capacitors. The proposed topology can deal with high voltage with low voltage rating switches. Therefore, the switches can be IGBTs without needing an auxiliary forced commutation circuit as in thyristors which are high voltage switches. The performance of the proposed topology is validated by digital simulations and a laboratory scale prototype implementation.

A New Technique to Detect Faults in De-Energized Distribution Feeders—Part II: Symmetrical Fault Detection

To ensure safe re-energizing of an overhead distribution feeder after it is de-energized for an extended period, a novel fault detection technique by controlling a thyristor-based device is proposed in a companion paper. The device connected in parallel with a breaker or recloser can inject electrical pulses with adjustable strength for the downstream fault detection in a de-energized system. The proposed method can effectively detect different kinds of asymmetrical faults based on the unbalanced fault currents. However, the unbalanced current-based fault detection scheme is not effective for three-phase symmetrical faults detection. Furthermore, a stalled motor or a shunt-connected capacitor bank in the downstream may also behaves like a short-circuit. Therefore, a fault detection algorithm based on the analysis of the harmonic impedance of the de-energized system is developed in this paper. This method is very effective for the symmetrical fault detection and for distinguishing a stalled motor and capacitor bank from a fault. Extensive lab test results are provided in the paper to verify the effectiveness of the proposed method.

A New Approach to Multifunctional Dynamic Voltage Restorer Implementation for Emergency Control in Distribution Systems

The dynamic voltage restorer (DVR) is one of the modern devices used in distribution systems to protect consumers against sudden changes in voltage amplitude. In this paper, emergency control in distribution systems is discussed by using the proposed multifunctional DVR control strategy. Also, the multiloop controller using the Posicast and P+Resonant controllers is proposed in order to improve the transient response and eliminate the steady-state error in DVR response, respectively. The proposed algorithm is applied to some disturbances in load voltage caused by induction motors starting, and a three-phase short circuit fault. Also, the capability of the proposed DVR has been tested to limit the downstream fault current. The current limitation will restore the point of common coupling (PCC) (the bus to which all feeders under study are connected) voltage and protect the DVR itself. The innovation here is that the DVR acts as a virtual impedance with the main aim of protecting the PCC voltage during downstream fault without any problem in real power injection into the DVR. Simulation results show the capability of the DVR to control the emergency conditions of the distribution systems.

Fault Current Limiting Function of Dynamic Voltage Restorer Utilizing Signals from Existing Protective Relays

AbstractDynamic voltage restorer (DVR) is a series custom compensator used in power distribution network; however, due to the fact that it is connected in series with the distribution line, DVR would suffer from downstream faults. To limit the flow of large fault currents and protect DVR itself as well, a fault current limiting function is proposed in the DVR control strategy. Fault current limiting function of DVR will be activated by the protection system and then DVR will start injecting a series voltage to the line in such a way as to limit fault current to an appropriate level (in accordance with required sensitivity of the protection systems). The contribution of this proposal is the use of signals from existing feeder protection system to activate the DVR. This will simplify the structure of DVR because the extra build‐in fault detection module is not required. Moreover, it will ensure proper coordination between DVR and protection systems. Copyright © 2010 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

A Dual-Functional Medium Voltage Level DVR to Limit Downstream Fault Currents

The dynamic voltage restorer (DVR) is a modern custom power device used in power distribution networks to protect consumers from sudden sags (and swells) in grid voltage. Implemented at medium voltage level, the DVR can be used to protect a group of medium voltage or low voltage consumers. However, the DVR will therefore be tasked to mitigate even more faults involving downstream loads. Large fault currents would flow through the DVR during a downstream fault before the opening of a circuit breaker. This will cause the voltage at point of common coupling (PCC) to drop, which would affect the loads on the other parallel feeders connected to PCC. Furthermore, if not controlled properly, the DVR might also contribute to this PCC voltage sag in the process of compensating the missing voltage, thus further worsening the fault situation. To limit the flow of large line currents, and therefore restore the PCC voltage as well as protect the DVR system components, a downstream fault limiting function is proposed and integrated in the DVR operation. A flux-charge-model feedback algorithm is implemented so that the DVR would act as a large virtual inductance in series with the distribution feeder in fault situations. Controlling the DVR as a virtual inductor would also ensure zero real power absorption during the DVR compensation and thus minimize the stress in the dc link. Finally, the proposed fault current limiting algorithm has been tested in Matlab/Simulink simulation and experimentally on a medium voltage level laboratory DVR system.

Practical Low Voltage Equipment Ground Fault Protection for Solidly Grounded Systems with Wye-Connected Source Transformers

A low voltage equipment ground fault protection system for solidly grounded radial systems is described. The protection provides primary protection for source substation buses. It is capable of providing selective operation with downstream ground fault trip devices or phase-overcurrent devices set to detect ground faults. It also provides source transformer protection for phase-to-ground through faults to supplement transformer primary phase-to-phase and three-phase through-fault overcurrent protection.