Fault Recovery Research Articles

Power Grid Cloud Resource Status Data Compression Method Based on Deep-Learning

Background: As the "three-type two-net, world-class" strategy proposed, the number of cloud resources in power grid continues to grow, there is a large amount of data to be filed every day. Which are key issues to be addressed, the long-term preservation of data, using back-up data for the operation and maintenance, fault recovery, fault drill and tracking of cloud platform are essential. The traditional compression algorithm faces severe challenges. Method: In this case, this paper proposes the deep-learning method for data compression. First, a more accurate and complete grid cloud resource status data is gathered through data cleaning, correction, and standardization. The preprocessed data is then compressed by SaDE-MSAE. Result: Experiments show that the SaDE-MSAE method can compress data faster. The data compression ratio based on neural network is basically between 45% and 60%, which is relatively stable and stronger than the traditional compression algorithm. Conclusion:: The paper can complete the compressed data quickly and efficiently in a large amount of power data. Improve the speed and accuracy of the algorithm while ensuring that the data is correct and complete, and improve the compression time and efficiency through the neural network. It gives better compression schemes cloud resource data grid.

A Learning-Aided Generic Framework for Fault Detection and Recovery of Inertial Sensors in Automated Driving Systems

A generic framework is proposed for fault detection and recovery (FDR) in automated driving systems (ADSs) with inertial sensors, which is based on multitask one-dimensional convolutional neural networks (MONN). FDR plays an important role for safe and robust vehicle controls in driver-assistance and ADSs. Although model-based fault detection and recovery methods are well established in the literature, they need accurate model description and are often designed under ideal conditions. Data-driven fault diagnosis, however, might overlook apparent physical relations among sensors. In order to take advantage of both approaches, dynamic relations among measured physical quantities and temporal finite differences are introduced in a MONN as additional features. The proposed algorithm achieved more reliable performance and outperformed considered state-of-the-art methods under various testing conditions. Extensive experiments were conducted to empirically show the effectiveness of the proposed generic FDR framework.

Experimental Analysis in Hadoop MapReduce: A Closer Look at Fault Detection and Recovery Techniques.

Hadoop MapReduce reactively detects and recovers faults after they occur based on the static heartbeat detection and the re-execution from scratch techniques. However, these techniques lead to excessive response time penalties and inefficient resource consumption during detection and recovery. Existing fault-tolerance solutions intend to mitigate the limitations without considering critical conditions such as fail-slow faults, the impact of faults at various infrastructure levels and the relationship between the detection and recovery stages. This paper analyses the response time under two main conditions: fail-stop and fail-slow, when they manifest with node, service, and the task at runtime. In addition, we focus on the relationship between the time for detecting and recovering faults. The experimental analysis is conducted on a real Hadoop cluster comprising MapReduce, YARN and HDFS frameworks. Our analysis shows that the recovery of a single fault leads to an average of 67.6% response time penalty. Even though the detection and recovery times are well-turned, data locality and resource availability must also be considered to obtain the optimum tolerance time and the lowest penalties.

Human-Aware Reinforcement Learning for Fault Recovery Using Contextual Gaussian Processes

This work addresses the iterated nonstationary assistant selection problem, in which over the course of repeated interactions on a mission, an autonomous robot experiencing a fault must select a single human from among a group of assistants to restore it to operation. The assistants in our problem have a level of performance that changes as a function of their experience solving the problem. Our approach uses reinforcement learning via a multi-arm bandit formulation to learn about the capabilities of each potential human assistant and decide which human to task. This study, which is built on our past work, evaluates the potential for a Gaussian-process-based machine learning method to effectively model the complex dynamics associated with human learning and forgetting. Application of our method in simulation shows that our method is capable of tracking performance of human-like dynamics for learning and forgetting. Using a novel selection policy called the proficiency window, it is shown that our technique can outperform baseline selection strategies while providing guarantees on human use. Our work offers an effective potential alternative to dedicated human supervisors, with application to any human–robot system where a set of humans is responsible for overseeing autonomous robot operations.

A trust evaluation model for secure data aggregation in smart grids infrastructures for smart cities

New technologies and automation systems have changed the traditional smart grid systems into new and integrated intelligent systems. These new smart systems are adopted for energy efficiency, demand and response, management and control, fault recovery, reliability and quality of services. With various benefits, smart grids have vulnerabilities due to open communication systems, and open infrastructures. Smart grids systems are based on real-time services, where privacy and security id one of the major challenge. In order to address these challenges and deal with security and privacy issues, we proposed a Trust Evaluation Model for Smart Grids (TEMSG) for secure data aggregation in smart grids and smart cities. This model tackles privacy and security issues such as data theft, denial of services, data privacy and inside and outside attacks and malware attacks. Machine learning methods are used to gather trust values and then estimate the imprecise information to secure the data aggregation in smart grids. Experiments are conducted to evaluate and analyze the proposed model in terms of detection rate, trustworthiness, and accuracy.

Optimization technique for fault recovery and fast data transmission in MANET

In Mobile Ad Hoc Networks (MANET), during routing, there may be possibility of route recovery delay and overhead. Also there may be issues due to reduced lifetime of link and node. In order to overcome these issues, in this paper, an optimization technique for fault recovery and fast data transmission in MANET is proposed. In order to avoid the faults due to frequent disconnections and node failures, the path with least mobility and high residual energy is selected for transmission. During data transmission, if majority of the neighbours are changed or if the residual energy is low, then the source will choose any shorter alternate route by applying the optimized recovery mechanism. By simulation results, we show that the proposed technique enhances packet delivery ratio and throughput and minimizes end-to-end delay, energy consumption and control overhead.

Research on Distribution Network Fault Recovery Reconfiguration Considering Distributed Generations

In view of the increasing penetration rate of distributed power in the distribution network, the traditional fault recovery reconfiguration method of distribution network needs to be further improved. An economic optimization model of distribution network fault recovery reconfiguration considering different types of distributed generations is proposed. This model takes into account the minimum number of switching operations as target and considers the impact of network active loss on restorative reconfiguration. Integer coding quantum particle swarm optimization algorithm is applied to the fault recovery reconfiguration optimization model and a fault recoverability discriminant method is presented. This method can classify the fault network into recoverable power-lossing area and non-recoverable power-lossing area. Recovering the fault recovery method given by the recoverable power loss area, and restoring the power supply to the unrecoverable power loss area according to the distributed power supply condition connected to the area. The simulation results show that the proposed method can be applied to the fault recovery reconfiguration of distribution network with distributed power accurately and efficiently.

A New Dynamic Simulation Test Method of Intelligent Electricity Metering System

Electricity metering is the core supporting technology for realizing accurate perception of electricity consumption, reasonable dispatching of power grids, rapid market response, diversified charging and new forms of electricity marketing. However, due to the significant differences in the actual metering environment, the operation of the electricity metering and information collection system is unstable, showing many problems such as unsuccessful parameter delivery and low acquisition success rate, which severely restricts the development of smart grid two-way interactive technology and the realization of smart electricity consumption in the last mile. There is an urgent need for a reliable and accurate dynamic simulation and systematic testing method to realize the reproduction and testing of various field faults. In this paper, we propose a new type of dynamic simulation test method for intelligent electricity metering system, which can realize the performance test, fault dynamic simulation, location and recovery of each functional unit of the master station, collection terminal and smart meter. Through experimental testing, our dynamic simulation test methods can realize the rapid recurrence of on-site problems, improve the high reliability and normal operation of the electric energy metering information collection system.

An X-shaped-switching-network high-step-up converter for grid integration of renewable energy sources

Power converters play a pivotal role in forming modern renewable energy sources integrated power grids. Among all grid-forming converters, high step-up converters with great nonlinear dynamic responses and fault recovery abilities have been sought after and developed in recent decades. This study proposes a novel X-shaped switching network (X-net)-based high-step-up converter. Specifically, an X-shaped switching network is designed to realize energy conversion of the converter, which, different from traditional impedance networks, only comprises switches and diodes. With this radically new X-shaped network, the proposed X-net converter is able to realise additional voltage pumping and low capacitor stress, hence improving dynamic responses and system reliability. In this study, the theory, analysis and simulation and prototype experiments of the proposed X-net converter with additional voltage pumping are presented in detail. Extensive comparisons between the X-net converter and its traditional counterparts are also presented which verifies the superiority of the proposition. The proposed X-net based high step-up converter bears significant potential as future grid-forming converters.

Adaptive fault recovery strategy of LCC‐MMC based hybrid HVDC

Hybrid HVDC consisting of line commutated converter (LCC) on rectifier side and modular multilevel converter (MMC) on inverter side is the favourable future trend of HVDC projects utilising the advantages of both converter technologies. However, fault clearing and adaptive restart capabilities of such hybrid HVDC are not much investigated in the literature. In this paper fault clearing of hybrid HVDC transmission line is investigated using mechanical DC breaker at MMC terminal and forced firing angle retardation at LCC terminal. Then, the characteristics of the line voltages are analysed during the fault blocking states of converters to clear the transient fault. Characteristics of the line voltages of LLC rectifier have shown prominent difference under permanent and temporary fault states, and also dependent on the severity of fault. Therefore, based on line voltage characteristics, a restart strategy is proposed for hybrid HVDC which is adaptive to the fault characteristics. Several PSCAD/EMTDC based simulation studies are conducted to validate the proposed fault recovery strategy. Simulation results validate the effectiveness of the proposed strategy for hybrid HVDC under DC line faults.

Current Sag and Mismatch Based Earth Fault Location for Distribution Network with Renewable Energy Resources

Fault diagnosis is increasingly becoming more challenging in distribution network due to complicated network topology and coupling disturbances introduced into the systems. This requires a higher level of rapidity and accuracy of fault diagnosis, location and recovery. This paper proposes an earth fault location algorithm for distribution network using current analysis. The fault information can be obtained by those measuring equipment that have already installed in critical nodes and generations. Then, the mechanism model is proposed to illustrate the equilibrium relationship of current flow in distribution network with renewable energy sources. The criterions named current sag and current mismatch are also proposed in this paper for fast fault area location and accurate fault node location. Assessment of this method is carried out using a distribution network model with renewable energy sources in PSCAD. The experimental results confirm the effectiveness and superiority of the proposed method in distribution network fault diagnosis over conventional methods.

Fault-Tolerant Control of Programmable Logic Controller-Based Production Systems With Deep Reinforcement Learning

Abstract Fault-tolerant control policies that automatically restart programable logic controller-based automated production system during fault recovery can increase system availability. This article provides a proof of concept that such policies can be synthesized with deep reinforcement learning. The authors specifically focus on systems with multiple end-effectors that are actuated in only one or two axes, commonly used for assembly and logistics tasks. Due to the large number of actuators in multi-end-effector systems and the limited possibilities to track workpieces in a single coordinate system, these systems are especially challenging to learn. This article demonstrates that a hierarchical multi-agent deep reinforcement learning approach together with a separate coordinate prediction module per agent can overcome these challenges. The evaluation of the suggested approach on the simulation of a small laboratory demonstrator shows that it is capable of restarting the system and completing open tasks as part of fault recovery.

Quality-related monitoring of papermaking wastewater treatment processes using dynamic multiblock partial least squares

Environmental problems have attracted much attention in recent years, especially for papermaking wastewater discharge. To reduce the loss of effluence discharge violation, quality-related multivariate statistical methods have been successfully applied to achieve a robust wastewater treatment system. In this work, a new dynamic multiblock partial least squares (DMBPLS) is proposed to extract the time-varying information in a large-scale papermaking wastewater treatment process. By introducing augmented matrices to input and output data, the proposed method not only handles the dynamic characteristic of data and reduces the time delay of fault detection, but enhances the interpretability of model. In addition, the DMBPLS provides a capability of fault location, which has certain guiding significance for fault recovery. In comparison with other models, the DMBPLS has a superior fault detection result. Specifically, the maximum fault detection rate of the DMBPLS is improved by 35.93% and 12.5% for bias and drifting faults, respectively, in comparison with partial least squares (PLS).

A New Resilience Metric to Compare System of Systems Architecture

By combining independent systems (e.g., financial systems, power infrastructure, transportation networks), systems of systems (SoSs) can provide additional services. The ability to withstand faults, minimize service disruption, and efficiently recover (resilience) is a key performance characteristic. Current resilience metrics, however, require improvement. Each SoSs is approached with different assumptions about fault magnitude and recovery strategy. Thus, engineers currently have difficulty answering the basic question: Between alternative designs A and B, which is more resilient? SoS resilience metrics should reflect that the system interfaces are the most potent leverage point, have a standardized fault and recovery strategy, incorporate SoS dynamics, and apply to both natural and artificial systems of systems. In response, we develop the system of systems resilience metric (SoSRM). SoSRM methodology uses the definition of SoS to identify appropriate fault locations and graph theory to define fault and recovery duration. SoSRM is applied to two case studies to demonstrate application. First, an intertidal oyster reef ecosystem's system dynamics model is presented to demonstrate SoSRM methodology. Next, a military SoS demonstrates the application of SoSRM to a stochastic multi-agent system. Validation simulations show that SoSRM reflects expected fault impact and that methodological choices for fault duration do not drive SoSRM results. This article presents a new design tool capable of comparing the resilience between two SoS architectures.

Characterization and Prediction of Performance Loss and MTTR During Fault Recovery on Scale-Out Storage Using DOE & RSM: A Case Study with Ceph

Recognizing the impact from cluster recovery operations on performance and mean time to recovery (MTTR) is essential to maintain service and availability objectives. Testing the impact of recovery operation can reveal a cause-and-effect relationship between recovery parameters and responses of performance and MTTR during the recovery process. This study introduces a combination of systematic methodologies of design of experiments and response surface methodologies to effectively and efficiently find out main and factor-to-factor interaction effects toward the responses. Two Ceph clusters using different storage device technologies, HDD and SSD respectively, were used to characterize the impact of recovery operation on performance and MTTR. The combination of quadratic and linear effects from both Ceph clusters were determined and reported. With 28 tests, MTTR and performance models were developed for each Ceph cluster based on those quadratic and linear effects. These models demonstrate good prediction on performance and MTTR when recovery parameters are adjusted. Using design of experiment and response surface not only allow cause and effect analysis, but also provide the potential inefficient parameter that causes performance loss during recovery. This not only introduces a new method to study cause-and-effect in MTTR but serves as the indicator to areas for improvement for more efficient recovery operation.

Fault Feature Selection for Distribution Network

For the connection of DGs in distribution network, the fault power flow is different from that in normal operation. Further, the size of the fault current is limited by the electronic components and greatly reduces. Therefore, fault detection, the protections and their coordination become very complex. Fault detection technology helps to achieve fault isolation and recovery, and plays an important role in distribution network control and operation. This paper proposes a data-driven fault feature selection method for distribution network. This method collects various electrical quantities during normal and short-circuit faults of the distribution network as a feature library, and uses the support vector machine-recursive feature elimination method for feature selection to remove redundant features. The optimized fault features can be used to fault detection for distribution network with DGs.

Study on Dynamic Characteristics of UHVDC System under Hierarchical Infeed Mode with STATCOM

This paper focuses on the investigation of the dynamic characteristics of Ultra High Voltage Direct Current (UHVDC) system Under Hierarchical Infeed Mode (UHVDC-HIM) with Static Synchronous Compensator (STATCOM). The STATCOM at the inverter side of UHVDC-HIM is developed in PSCAD/EMTDC, and their power flow equations are derived. Based on these equations, the control schemes for rectifier and inverter side of UHVDC-HIM and STATCOM are properly designed. Several dynamic characteristics indices such as Commutation Failure Immunity Index (CFII), Commutation Failure Probability Index (CFPI), fault recovery time and Temporary Overvoltage (TOV) are evaluated in detail. The impact of different sizes of STATCOM and Short Circuit Ratio (SCR) of AC sources on dynamic characteristics of UHVDC-HIM system during single-phase and three-phase faults are examined. The analysis shows that higher SCR values and greater capacity of STATCOM can make the UHVDC-HIM system less vulnerable to Commutation Failure (CF), considerably enhance the fault recovery time and effectively reduce the TOV.

Design and development of Cu‐Al‐Mn‐Ni shape memory alloy coated optical fibre sensor for condition‐based monitoring of physical systems

Online fault detection, isolation and recovery using smart sensors play an important role in intelligent manufacturing system. Fibre optic sensors are very interesting for condition monitoring applications due to the advantage of this technology. Here, the experimental demonstration of Cu-based shape memory alloy (SMA) coated optical fibre for temperature-based sensing applications is reported. The benefit of Cu-based SMA coated optical fibre over conventional metallic coating has been evaluated in the study. For consistent coating, an in situ fixture with a rotary drive setup has been designed and developed. Thermo optic test bench has been developed to study the actuation characteristics of the SMA coated optical fibre for varying current and voltage. Experiments were performed to investigate the light intensity in the SMA coated optical fibre at different actuation conditions. The displacement that takes place in the optical fibre due to the external temperature stimuli will create proportional intensity and wavelength shifts. The maximum average displacement of 4.9 mm has been achieved for Cu-Al-Mn-Ni coated optical fibre. Results show variation in the optical fibre signal due to heating and cooling of the fibre from the applied electrical stimulus on Cu-based SMA coating in the form of austenite to martensite transformation.

Quantifying the Contribution of Dynamic Reactive Power Compensators on System Strength at LCC-HVdc Converter Terminals

This paper presents improved indices to evaluate the contribution of reactive power compensators at HVdc converter terminals. The presence of such compensators impacts converter performance, such as the maximum available power (MAP) limit, susceptance to commutation failure, fault recovery time and temporary over voltage. Traditionally, the Effective Short Circuit Ratio (ESCR) has been used to indicate system strength. However, this index cannot be directly applied when power-electronics based converters are used to provide the reactive power support. The approach of this paper is to use an improved form of the ‘Apparent increase in Short Circuit Ratio’ (AISCR) index to quantify the impact of the compensator on the HVdc system. This index is evaluated by comparing the performance with the reactive power device with a system without the device, but having the same performance. The paper shows that a single AISCR index cannot quantify all behaviors and suggest modifications to AISCR for each of several HVdc phenomena. In this paper two types of compensators are considered – the Synchronous Compensator and the Static Synchronous Compensator (STATCOM).

Optimal Design of Excitation Systems of Synchronous Condensers for HVDC Systems in Power Grid Environment Based on Variable Universe Fuzzy Adaptive PID Controller

In a gesture to optimize the reactive power properties of synchronous condensers and improve capabilities of condensers to support the voltages of AC systems, the outer loop control of the reactive power of condensers and the outer loop control of the voltages of AC systems are introduced into the conventional design of the main excitation systems of condensers in high voltage direct current (HVDC) systems in this study. Moreover, variable universe fuzzy adaptive PID controller is proposed to serve excitation systems of condensers to improve the response ability of condensers in power grid faults and the recovery speed of condensers after power grid faults, and to reduce overvoltage of power grids generated by condensers in the process of fault recovery. To verify its experimental validation, the model of ±100 kV HVDC system containing a condenser is set up in MATLAB/Simulink. The verified results reveal that the optimal design strategy of excitation systems of synchronous condensers in the proposal can ensure rapid regulation of the voltages of AC systems arising from condensers, and reduce overcompensation of the voltages and reactive power of AC systems arising from condensers when faults recover.