Power System Data Research Articles

전압 안정도 향상을 위한 칼만필터 알고리즘 기반의 비선형 P-Q 드룹 커브 추정

An increase in a distributed generation (DG) causes an excessive voltage problem unless there is proper complement action. The direct voltage control, however, might cause control issues of hunting between adjacent DGs. This paper proposes an indirect voltage control method based on a non-linear P-Q droop curve that is estimated using the Kalman-filter algorithm. For effective estimation, PQV curved surface is estimated, from which the P-Q droop curve is extracted with the condition of zero voltage variation. The effectiveness is verified based on a practical microgrid model composed of real island power system data. For the simulation-based verification, power system computer aided design/electromagnetic transient including DC (PSCAD/EMTDCTM) is used.

Promoting citizen science in the energy sector: Generation Solar, an open database of small-scale solar photovoltaic installations

Citizen science is becoming an effective approach in building a new relationship between science and society, in which the desire of citizens to participate actively in knowledge production meets the needs of researchers. A citizen science initiative dealing with the development of photovoltaics (PV) is presented. To generate a “responsible” initiative, the research question has been designed collectively from the beginning, involving diverse actors in order to encourage creativity while addressing their interests and concerns. The result has been called Generation Solar. It aims at co-creating an open database of PV installations including their technical characteristics, and an online map for visualizing them. The initiative responds to a clear scientific demand; an important drawback for researchers working on energy modelling and predictions of production lays precisely in the lack of information about these installations’ locations and characteristics. The initiative invites citizens, companies and public institutions with a PV installation to collaborate by providing such data. Data will follow the format of Open Power System Data in order to be fully exploitable by the scientific community and society. The success of the initiative will rely on the capacity to mobilize citizens and register the largest possible number of installations worldwide.

False Data Injection Attack Detection in Power Systems Based on Cyber-Physical Attack Genes

In the process of the detection of a false data injection attack (FDIA) in power systems, there are problems of complex data features and low detection accuracy. From the perspective of the correlation and redundancy of the essential characteristics of the attack data, a detection method of the FDIA in smart grids based on cyber-physical genes is proposed. Firstly, the principle and characteristics of the FDIA are analyzed, and the concept of the cyber-physical FDIA gene is defined. Considering the non-functional dependency and nonlinear correlation of cyber-physical data in power systems, the optimal attack gene feature set of the maximum mutual information coefficient is selected. Secondly, an unsupervised pre-training encoder is set to extract the cyber-physical attack gene. Combined with the supervised fine-tuning classifier to train and update the network parameters, the FDIA detection model with stacked autoencoder network is constructed. Finally, a self-adaptive cuckoo search algorithm is designed to optimize the model parameters, and a novel attack detection method is proposed. The analysis of case studies shows that the proposed method can effectively improve the detection accuracy and effect of the FDIA on cyber-physical power systems.

Analyzing Information Disclosure in the Chinese Electricity Market

Electricity generation relies heavily on fossil fuels in China, and this has posed great challenges for sustainable development. In 2015, China launched a new series of power reforms, aimed primarily at sustainable development and building a competitive power market system, where an information disclosure system plays an important role. This paper analyzes the effects of sensitive information disclosure, constructs different information disclosure scenarios, and compares the market clearing results under different scenarios. The results show that information transparency is conducive to the promotion of market efficiency. However, some problems, especially collusion, arise and inevitably bring negative impacts to the power market. Therefore, more attention should be paid to the content and quality of data transparency to avoid manipulations in the market. The data of power systems are complex and various, and thus big data applications may be conducive to effective information disclosure and better market regulation. Moreover, disclosure delay will help the electricity market become more transparent by reducing the risk of collusion. Besides, the information’s scope and contents, timing should also be taken into consideration. These findings may provide some references for the sustainable development of the electricity industry and have certain policy implications for policymakers.

A Data-Driven Approach for Online Inter-Area Oscillatory Stability Assessment of Power Systems Based on Random Bits Forest Considering Feature Redundancy

To utilize the rapidly refreshed operating data of power systems fully and effectively, an integrated scheme for inter-area oscillatory stability assessment (OSA) is proposed in this paper using a compositive feature selection unit and random bits forest (RBF) algorithm. This scheme consists of offline, update, and online stages, and it can provide fast and accurate estimation of the oscillatory stability margin (OSM) by using the real-time system operating data. In this scheme, a compositive feature selection unit is specially designed to realize efficient feature selection, which can significantly reduce the data dimensionality, effectively alleviate feature redundancy, and provide accurate correlation information to system operators. Then, the feature set consisting of the selected pivotal features is used for the RBF training to build the mapping relationships between the OSM and the system operating variables. Moreover, to enhance the robustness of the scheme in the face of variable operating conditions, an update stage is developed. The effectiveness of the integrated scheme is verified on the IEEE 39-bus system and a larger 1648-bus system. Tests of estimation accuracy, data processing speed, and the impact of missing data and noise data on this scheme are implemented. Comparisons with other methods reveal the superiority of the integrated scheme. In addition, the robustness of the scheme to variations in system topology, distribution among generators and loads, and peak and minimum load is studied.

An Online Network Model-Free Wide-Area Voltage Control Method Using PMUs

This paper proposes a novel online measurement-based Wide-Area Voltage Control (WAVC) method using Phasor Measurement Unit (PMU) data in power systems with Flexible AC Transmission System (FACTS) devices. As opposed to previous WAVC methods, the proposed WAVC does not require any model knowledge or the participation of all buses and considers both active and reactive power perturbations. Specifically, the proposed WAVC method exploits the regression theorem of the Ornstein-Uhlenbeck process to estimate the sensitivity matrices through PMU data online, which are further used to design and apply the voltage regulation by updating the reference points of FACTS devices. Numerical results on the IEEE 39-Bus and IEEE 68-Bus systems demonstrate that the proposed model-free WAVC can provide effective voltage control in various network topologies, different combinations of voltage-controlled and voltage-uncontrolled buses, under measurement noise, and in case of missing PMUs. Particularly, the proposed WAVC algorithm may outperform the model-based WAVC when an undetected topology change happens.

Dynamic lightning protection method of electric power systems based on the large data characteristics

In this paper, a new active dynamic lightning protection method is proposed based on the large data characteristics of electric power. This method mainly includes two parts: Part one, Neo4j framework model which is used to analyze large data of power system and dynamic regulation of power system, and Python software which is used to compare and analyze different framework models; Part two, the comparison between dynamic lightning and conventional protection methods. The results show that Neo4j traversal speed is 87.5% and 89.1% faster than Hadoop and Spark respectively, clustering effect is 12.5% and 17.8% higher than Hadoop and Spark respectively. As a result, Neo4j framework model is more suitable for the characteristics of large data in power system. After the lightning accident, the power-off time of dynamic lightning protection system is reduced by about 53.1%, and the recovery time of the system also decreased by about 42.8%. In the dynamic regulation of power system, the output of power supply is reduced by 35.1 MW and the load is cut out by 15.8 MW, which greatly reduce the impact of lightning strike on power supply and important load.

Dynamic State Estimation of New Energy Power Systems Considering Multi-Level False Data Identification Based on LSTM-CNN

With the increase of new energy integration, it is difficult to identify the measured data and false data in power system when they are mixed into cyber network. If false data with error information is utilized in the power system state estimation, the accuracy of state estimation will be reduced. The inaccurate estimation results will lead to wrong control decisions by operators. This paper proposes an improved dynamic state estimation method based on multi-level false data identification. This method uses innovation vector for the first-level identification, long-short term memory neural network for the second-level temporal identification, and convolution neural network for the third-level spatial identification. Through the identification, the mutation data are distinguished as fluctuant real data and false data. The identification results can provide precise operation information for power system, dynamically correct the filtering direction of state estimation and improve the accuracy of state estimation. The method is verified by IEEE-57 power system with actual operating data. The results show that the improved method can not only resist false data injection attacks, but also maintain high estimation accuracy in new energy power systems with strong data volatility.

Control on Abnormal Data Overflow of Distribution Network Management Platform

A large amount of system operation and equipment status data in current power systems are recorded and stored in text form, and text analysis has become an important basis for power system operation analysis. However, the current research still lacks relevant experience in applying it to the disposal of abnormal data spills in power grids. Therefore, the text-oriented data mining is first introduced in the paper to explain its basic concepts and implementation processes. Then, combining text grid-related data abnormal spillover requirements, text analysis and decision-making methods are proposed. Finally, considering actual development and application situation, the implementation benefits are analyzed. The experiment proves that the data abnormal overflow control technology studied in the paper can effectively guarantee the normal operation of the system.

Development of a methodology for data validation in power systems using different types of measurements

In recent years, the technology of synchrophasor measurements has been introduced in power systems around the world. When managing a power system and predicting its operation conditions, the important task is to check the validity of power system data. At the same time, the traditional types of measurements, such as digital fault recorders and telemetry devices are still widely used. It is known that redundancy of measurements contributes to a more accurate solution of the data validation problem. It is useful to create a method for data validation in power systems, which could involve various types of measurements in order to increase the redundancy and, hence, the overall accuracy of measurements. This study presents some validity criteria that use the idea described above. The results of testing the proposed methodology on the substation model in Matlab software package are presented and discussed.

Grid-Graph Signal Processing (Grid-GSP): A Graph Signal Processing Framework for the Power Grid

The underlying theme of this paper is to explore the various facets of power systems data through the lens of graph signal processing (GSP), laying down the foundations of the Grid-GSP framework. Grid-GSP provides an interpretation for the spatio-temporal properties of voltage phasor measurements, by showing how the well-known power systems modeling supports a generative low-pass graph filter model for the state variables, namely the voltage phasors. Using the model we formalize the empirical observation that voltage phasor measurement data lie in a low-dimensional subspace and tie their spatio-temporal structure to generator voltage dynamics. The Grid-GSP generative model is then successfully employed to investigate the problems pertaining to the grid of data sampling and interpolation, network inference, detection of anomalies and data compression. Numerical results on a large synthetic grid that mimics the real-grid of the state of Texas, ACTIVSg2000, and on real-world measurements from ISO-New England verify the efficacy of applying Grid-GSP methods to electric grid data.

Short-Term Load Forecasting Based on Adabelief Optimized Temporal Convolutional Network and Gated Recurrent Unit Hybrid Neural Network

To fully mine the relationship between temporal features in load data, improve the accuracy and efficiency of short-term load forecasting and overcome the difficulties caused by load nonlinearity and volatility in accurate load forecasting. In this paper, a hybrid neural network short-term load forecasting model based on temporal convolutional network (TCN) and gated recurrent unit (GRU) is proposed. Firstly, the correlation between meteorological features and load is measured with the distance correlation coefficient, and the fixed-length sliding time window method is used to reconstruct the features. Next, temporal convolutional network is adopted to extract the hidden historical information and time relationship including meteorological features, electricity price, etc., and a better-performing gated recurrent unit is utilized for perdition. Furthermore, the state-of-the-art AdaBelief optimizer and Attention mechanism are utilized to enhance the prediction accuracy and efficiency. The effectiveness and superiority of the proposed model are verified by load and weather data from Spain and PJM power system data. Short-term load forecasting results in different periods and comprehensive comparisons with the performance of different models show that the proposed model can provide accurate load forecasting results rather quickly. The highlights of this paper are that temporal convolutional network and gated recurrent unit are combined for load forecasting for the first time, and the forecasting performance is improved by the novel optimizer AdaBelief and feature selection based on distance correlation coefficient.

Transmission of Frequency Balance Instructions and Secure Data Sharing Based on Chaos Encryption in Smart Grid-Based Energy Systems Applications

Data communication security between power systems, which has become a chronic problem in the classical network, is characterized by the application of information communication technology (ICT) in modern network models. Nowadays, the electricity network is spread over very large areas, and the load taking, and load shedding instructions are transmitted one-way and unsafe in the current structure. However, utilities in the smart grid are developing new methodologies to secure the communication infrastructure. This article presents a design that will make the current grid model more reliable and provide a secure data communication in accordance with the modern grid infrastructure. This design provides auxiliary communication codes such as chaotic codes embedded in the communication instructions between power systems, which are generation balance instructions sent to the power plants by the load dispatch center to achieve frequency balance. Used as a case study, one-day instructions are transmitted and encoded four times a day, which is the amount of capacity that only the receiver and transmitter can understand with two-sided encryption. The necessary reserve capacity, chaotic encryption and suitability for real-time applications were evaluated using experimentally obtained power system data for frequency control, which is very important in terms of sustainability in energy and economy against cyber-attacks, and very accurate results were obtained.

Graph computing based security constrained unit commitment in hydro-thermal power systems incorporating pumped hydro storage

This paper proposes a graph computing based mixed integer programming (MIP) framework for solving the security constrained unit commitment (SCUC) problem in hydro-thermal power systems incorporating pumped hydro storage (PHS). The proposed graph computing-based MIP framework considers the economic operations of thermal units, cascade hydropower stations and PHS stations, as well as their technical impacts towards the network security. First, the hydro-thermal power system data and unit information are stored in a graph structure with nodes and edges, which enables nodal and hierarchical parallel computing for the unit commitment (UC) solution calculation and network security analysis. A MIP model is then formulated to solve the SCUC problem with the mathematical models of thermal units, cascade hydropower stations and PHS stations. In addition, two optimization approaches including convex hull reformulation (CHR) and special ordered set (SOS) methods are introduced for speeding up the MIP calculation procedure. To ensure the system stability under the derived UC solution, a parallelized graph power flow (PGPF) algorithm is proposed for the hydro-thermal power system network security analysis. Finally, case studies of the IEEE 118-bus system and a practical 2749-bus hydro-thermal power system are introduced to demonstrate the feasibility and validity of the proposed graph computing-based MIP framework.

Detection of Cyber-Attacks of Power Systems Through Benford’s Law

This letter proposes an application of the Benford’s law for the detection of cyber attacks in power system state estimators. Benford’s law, also known as 1st-digit law, states an unexpected property of the distribution of the first digit of certain sets of data, and has been found to apply to a surprisingly wide range of data domains. The first novel contribution of the letter is to show that the Benford’s law applies to power system data as well. A relevant property of this law is its high sensitivity to manipulations and, in fact, it is often utilized to detect frauds. Based on this feature, the second contribution of the letter is to utilize the Benford’s law to detect malicious data introduced by hackers in the supervisory control and data acquisition (SCADA) system of a transmission network. Tests based on power system models ranging from 9 to 21,177 buses show promising results.

Highlighting and overcoming data barriers: creating open data for retrospective analysis of US electric power systems by consolidating publicly available sources

Studies of power system operation commonly draw from two key databases produced by the US Environmental Protection Agency: the Acid Rain Program’s Continuous Emission Monitoring System (CEMS) data, and the Emissions and Generation Resource Integrated Database (eGRID). Separate reporting requirements and heterogeneity in data aggregation between these two databases creates a barrier to systematic spatial and temporal retrospective power system analysis. This work describes the inherent challenges to this undertaking and documents a method for reconciling the two seemingly disparate data sources. While fundamental differences in data reporting and aggregation prevent us from achieving full coverage, this work represents an important initial step to aligning these two repositories of US power system data. We demonstrate the value of this linkage by computing relative unit-level, hourly utilization metrics for most thermal power plants in the US. Analysis of these metrics across time illustrates thermal generator cycling trends in California between 2011 and 2017. These unit-level results indicate that combined cycle units within California increased their part-load generation by 15% and resultant CO2 emissions by 17% and decreased their start/stop frequency over time by 8.5% and resultant emissions by 47%. Open cycle gas turbines overall increased their generation- number of start/stop cycles by 97% and resultant emissions by 85%, part-load generation by 120% and resultant emissions by 100%, and full load generation by 40% and resultant emissions 18%. We also observe a temporal shift in thermal generation from morning hours to evening in California.

Substation Equipment Fault Identification Based on Infrared Image Analysis

As a key part of the power system, the substations undertake important work. With the development and construction of smart grids, the status data, image monitoring data, and environmental meteorological data of power systems are gradually being integrated and shared on a unified platform. Traditional models based on theoretical analysis are difficult to deal with multi-dimensional, massive data set information. Under this background, starting from the inherent law of the data itself, the use of machine learning methods combined with the infrared image data of the equipment can achieve intelligent identification and early warning of substation equipment failures. This paper first uses the convolutional neural network to identify the substation equipment in the picture, and then combines the infrared image to perform image registration. Finally, the deep belief network is used to determine whether the device is in wrong condition. The overall substation equipment fault identification is tested on real data, and the experimental results show that the proposed method has high accuracy.

Design and Implementation of a Data-Driven Approach to Visualizing Power Quality

Numerous underlying causes of power-quality (PQ) disturbances have enhanced the application of situational awareness to power systems. This application provides an optimal overall response for contingencies. With measurement data acquired by a multi-source PQ monitoring system, we propose an interactive visualization tool for PQ disturbance data based on a geographic information system (GIS). This tool demonstrates the spatio–temporal distribution of the PQ disturbance events and the cross-correlation between PQ records and environmental factors, leveraging Getis statistics and random matrix theory. A methodology based on entity matching is also introduced to analyze the underlying causes of PQ disturbance events. Based on real-world data obtained from an actual power system, offline and online PQ data visualization scenarios are provided to verify the effectiveness and robustness of the proposed framework.

Soft decision-making based on decision-theoretic rough set and Takagi-Sugeno fuzzy model with application to the autonomous fault diagnosis of satellite power system

The satellite power system is one of the core systems to ensure the normal on-orbit operation of satellite, and is also a representative of typical complex nonlinear systems. Autonomous prognostics and health management (A-PHM) is an inevitable trend in the future development of satellite, and autonomous fault diagnosis is a key part of A-PHM. Therefore, it is very necessary to carry out the research on autonomous fault diagnosis for satellite power system to improve the capability of satellite to perform its on-orbit tasks independently. Therefore, we put forward a feasible framework of soft decision-making to cope with this issue, mainly including attribute reduction, attribute weight assignment, rule extraction, and rule matching. Specifically, a neighborhood decision-theoretic rough set model (named DNDTRS) is first designed with the help of a data-driven loss function matrix for attribute reduction and weight assignment. Subsequently, the classification probability generated by DNDTRS is fed to a rule extraction model developed by the Takagi-Sugeno (T-S) fuzzy theory. Finally, a soft decision-making mechanism is proposed to execute the output after rule matching. In the experimental part, the proposed methodology is verified by the benchmark datasets and the fault data of satellite power system. The experimental results demonstrate the promised performance of our methodology.

Distributed Machine Learning on Dynamic Power System Data Features to Improve Resiliency for the Purpose of Self-Healing

Numerous online methods for post-fault restoration have been tested on different types of systems. Modern power systems are usually operated at design limits and therefore more prone to post-fault instability. However, traditional online methods often struggle to accurately identify events from time series data, as pattern-recognition in a stochastic post-fault dynamic scenario requires fast and accurate fault identification in order to safely restore the system. One of the most prominent methods of pattern-recognition is machine learning. However, machine learning alone is neither sufficient nor accurate enough for making decisions with time series data. This article analyses the application of feature selection to assist a machine learning algorithm to make better decisions in order to restore a multi-machine network which has become islanded due to faults. Within an islanded multi-machine system the number of attributes significantly increases, which makes application of machine learning algorithms even more erroneous. This article contributes by proposing a distributed offline-online architecture. The proposal explores the potential of introducing relevant features from a reduced time series data set, in order to accurately identify dynamic events occurring in different islands simultaneously. The identification of events helps the decision making process more accurate.