Grid Control Research Articles

Reinforcement Learning Techniques in Optimizing Energy Systems

Reinforcement learning (RL) techniques have emerged as powerful tools for optimizing energy systems, offering the potential to enhance efficiency, reliability, and sustainability. This review paper provides a comprehensive examination of the applications of RL in the field of energy system optimization, spanning various domains such as energy management, grid control, and renewable energy integration. Beginning with an overview of RL fundamentals, the paper explores recent advancements in RL algorithms and their adaptation to address the unique challenges of energy system optimization. Case studies and real-world applications demonstrate the efficacy of RL-based approaches in improving energy efficiency, reducing costs, and mitigating environmental impacts. Furthermore, the paper discusses future directions and challenges, including scalability, interpretability, and integration with domain knowledge. By synthesizing the latest research findings and identifying key areas for further investigation, this paper aims to inform and inspire future research endeavors in the intersection of reinforcement learning and energy system optimization.

Design and development of PI controller for DFIG grid integration using neural tuning method ensembled with dense plexus terminals

In a DFIG grid interconnected system, the control of real and reactive power relies on various factors. This paper presents an approach to regulate the flow of real and reactive power using a Neural Tuning Machine (NTM) based on a recurrent neural network. The focus is on controlling the flow of reactive power from the rotor-side converter, which is proportional to the grid-side controller through a coupling voltage. The proposed NTM method leverages neural networks to fine-tune the parameters of the PI controller, optimizing performance for DFIG grid integration. By integrating dense plexus terminals, also known as dense connections, within the neural network, the control system achieves enhanced adaptability, robustness, and nonlinear dynamics, addressing the challenges of the grid. Grid control actions are based on the voltage profile at different bus locations, thereby regulating voltage. This article meticulously examines the analysis in terms of DFIG configuration and highlights the advantages of the neural tuning machine in controlling inner current loop parameters compared to conventional PI controllers. To demonstrate the robustness of the control algorithm, a MATLAB Simulink model is designed, and validation is conducted with three different benchmarking models. All calculations and results presented in the article strictly adhere to IEEE and IEC standards. This research contributes to advancing control methodologies for DFIG grid integration and lays the groundwork for further exploration of neural tuning methods in power system control.

Impact Analysis of High-Altitude Electromagnetic Pulse Coupling Effects on Power Grid Protection Relays

Protection relays are important equipment used for protection, control, and metering functions in the power grid. These relays are used to protect critical and difficult-to-replace equipment, like generators, transformers, and capacitor banks. Once the protection devices are disturbed or damaged, a high risk of power generation interruption occurs. Therefore, it is important and necessary to study the relay’s immunity to electromagnetic pulse (EMP) events. As a preliminary step toward empirical experimentation on actual equipment, this manuscript outlines an economical and efficient methodology for evaluating the impact of an EMP. An impedance measurement strategy was employed to model the equipment, setting the stage for subsequent immunity analyses. These analyses included the pulse current injection (PCI) method, which utilized an injecting probe to introduce the transient, and frequency domain electromagnetic (FEKO) simulation, which integrated electromagnetic coupling effects into the transient simulation. The impedance measurement and simulation results in this paper provide a reliable basis for gauging equipment performance in the face of HEMP threats. The results obtained using the PCI and FEKO simulations demonstrated the performance of different port responses under a high-altitude EMP, indicating the requirement for some protection to ensure the reliable operation of relays.

An enhanced control scheme for multifunctional grid connected PV system using fuzzy and predictive direct power control

This paper presents a combination between a fuzzy logic control (FLC) and a predictive direct power control for multifunctional grid connected photovoltaic (PV) system, to solve the oscillation problem in the DC link voltage of the in-verter caused by the fast irradiation changing. The whole system consists of a PV system which interface a DC-AC inverter, a FLC maximum power point tracking (MPPT) algorithm has been adopted to operate the DC-DC converter at the MPP. The predictive control strategy is applied to the DC-AC inverter with FLC in its voltage control loop to improve the power exchange between the grid and the PV system. Simulation results have been verified through MATLAB/Simulink software for the purpose of giving the effectiveness of the suggested control against existed controllers.

Nonlinear Model Predictive Control for Droop-Based Grid Forming Converters Providing Fast Frequency Support

Nonlinear Model Predictive Control for Droop-Based Grid Forming Converters Providing Fast Frequency Support

Mission Profile-Tailored Design and Control of an Interconnected Hybrid Grid Connecting Converter Architecture

Mission Profile-Tailored Design and Control of an Interconnected Hybrid Grid Connecting Converter Architecture

Fault Detection, Classification and Localization Along the Power Grid Line Using Optimized Machine Learning Algorithms

Distributed energy generation increases the need for smart grid monitoring, protection, and control. Localization, classification, and fault detection are essential for addressing any problems immediately and resuming the smart grid as soon as possible. Simultaneously, the capacity to swiftly identify smart grid issues utilizing sensor data and easily accessible frequency and voltage data from PMU devices is a prerequisite of this task. Therefore, this paper proposes new methods using fuzzy logic and adaptive fuzzy neural networks as well as machine learning and meta-heuristic algorithms. First, line voltage is used by a fuzzy thresholding method to estimate when a transmission line defect would develop in less than 1.2 clock cycles. Next, features taken from frequency signals in the real-time interval are utilized to classify the type of error using machine learning systems (decision tree algorithm and random forest algorithm) optimized with wild horse meta-heuristic algorithm. To locate the precise problem location, we finally use a neural fuzzy inference system that is capable of adapting to new data. We employ a simulated power transmission system in MATLAB to test our proposed solutions. Mean square error (MSE) and confusion matrix are used to assess the efficiency of a classifier or detector. For the decision tree algorithm method, the detector attained an acceptable MSE of 2.34e−4 and accuracy of 98.1%, and for the random forest algorithm method, an acceptable MSE of 3.54e−6 and accuracy of 100%. Furthermore, the placement error was less than 153.6 m in any direction along the line.

Smart distribution network voltage estimation using PMU technology considering zero injection constraints.

To properly control the network of the power system and ensure its protection, Phasor measurement units (PMUs) must be used to monitor the network's operation. PMUs can provide synchronized real-time measurements. These measurements can be used for state estimation, fault detection and diagnosis, and other grid control applications. Conventional state estimation methods use weighting factors to balance the different types of measurements, and zero injection measurements can lead to large weighting factors that can introduce computational errors. The offered methods are designed to ensure that these zero injection criteria can be strictly satisfied while calculating the voltage profile and observability of the various distribution networks without sacrificing computing efficiency. The proposed method's viability is assessed using standard IEEE distribution networks. MATLAB coding is used to simulate the case analyses. Overall, the study provides a valuable contribution to the field of power distribution system monitoring and control by simplifying the process of determining the optimal locations for PMUs in a distribution network and assessing the impact of ZI buses on the voltage profile of the system.

Electric vehicle participation for optimized load frequency control in a multi-area restructured power system

ABSTRACT Electric vehicles (EVs) can be employed intelligently in Vehicle-to-Grid (V2G) and Grid-to-Vehicle (G2V) modes for Load Frequency Control (LFC) in the power grid. Due to the rapid variation of charging or discharging power, EVs can perform frequency regulation tasks and serve as distributed energy storage devices. In this paper, a linearized model of an interconnected multi-area power system is formulated in the restructured environment to include Electric Vehicle Aggregators (EVA). The DISCO Participation Matrix (DPM) is used to model Poolco and bilateral contracts in restructured multi-area power systems along with the participation of EVs. The faster dynamic response of EVs can effectively control the deviations in area frequencies imposed by the DISCOs load demand. The gains of the integral controller employed in each area are tuned by Real Coded Genetic Algorithm (RGA) by minimizing the integral square error function for area frequency deviations. The efficacy of power support from the EVs for load frequency control is validated by simulating different Poolco and bilateral transactions. It is shown that the EVs have a significant impact on obtaining optimal transient response for area frequencies, tie-line powers and power generation of conventional thermal generating units for any type of transaction in a restructured environment.

Design and implementation of a reduced switch seventeen-level multilevel inverter for grid integration of battery storage system

Multilevel inverters (MLI) have a variety of applications in battery storage system, renewable energy integration, electric vehicle (EV), drives for industry and in various flexible AC transmission system (FACTS). Applications of MLI in battery storage system have greatly increased over the past few decades. Conventional MLIs used for grid integration of battery energy storage is having large number of components and it requires transformer for grid integration. The large number of components and transformer increases the cost, size, weight and losses of the MLI, thus reduces the efficiency of the system. Keeping this in view, the goal of the proposed work is to design a novel reduced switch seventeen-level MLI which is having a lower number of components and do not require transformer for grid integration. This modular nature of proposed MLI is used to increase power handling capacity without introducing additional converters. For this, first proposed reduced switch seventeen-level MLI has been compared with other existing seventeen level topologies which in order to show the effectiveness of proposed topology. After analysis its effectiveness, the proposed MLI has been designed and then closed loop d-q current controller for grid connected mode is analyzed. The different curves are plotted to justify the stability of proposed inverter. The control of output voltage is also included using modified sine pulse width modulation (SPWM) method. The proposed MLI has been implemented in MATLAB/Simulink and validated using real time hardware-in-loop (HIL) emulator.

Grid interlocking fault prevention and control method considering safety and network loss

Grid interlocking fault prevention and control method considering safety and network loss

Active Energy Meters Tested in Realistic Non-Sinusoidal Conditions Recorded on the Field and Reproduced in Laboratory

Metrological characterisation of static energy meters under realistic low power quality conditions is a basic requirement for proper grid control and fair energy billing. The paper reports about a new proposed methodology, where the meters are tested under conditions directly recorded at installation sites. The waveforms of voltages and currents are sampled using a portable instrument; they are reproduced in laboratory conditions with a phantom power generator, with a bandwidth covering up to the 40th harmonic. The recording site is a photovoltaic energy production facility, having a a nominal power of 50 kW, at the coupling section to the grid. These waveforms were then reproduced in the laboratory, and tested on different models of single- and three-phase commercial static energy meters; the models chosen represent both energy meters used by energy providers at the point of common-coupling, and also meters typically used for in-line monitoring by end users. The quantity of interest is the reading error of the measured energy, when tested with the conditions reproduced from the on-field measurements, in comparison with a reference meter. All tested energy meter models comply with the present international documentary standards, which require tests under low power quality conditions; nevertheless, there are models that show unacceptable errors (up to 25%) in the measurement of active energy when tested with the on-field recorded waveforms. This suggests that the standardised testing waveforms might, in some cases, be not fully representative of the actual conditions encountered in the field.

Distributed Energy Resources Management System (DERMS) and Its Coordination with Transmission System: A Review and Co-Simulation

Ever-increasing penetration of distributed energy resources (DERs) in the power grids, alongside their numerous benefits, brings new challenges that call for enhanced solutions in the field of control and management of power grids. The majority of the available research have considered either distribution or transmission grids in their studies. In this paper, a comprehensive review of the effects of DERs on the distribution and transmission grids is performed. The focus of this paper is on hierarchical management methods in order to categorize different approaches and highlight the gaps. Moreover, a review is conducted in the field of the newly introduced distributed energy resources management system (DERMS) concept. A DERMS can facilitate the hierarchical energy management procedure due to its functionalities and broad capabilities. Hence, its implementation in energy management and its impact on the power grid will be assessed with the aid of a co-simulation platform that considers both transmission and distribution grids.

Virtual Power Grid Flux-Oriented Vector Control for Three-Phase Voltage-Type PWM Rectifier

More considerable power distortions are produced by the non-reversible diode-bridge rectifiers in the power system’s AC-DC conversion technology. In addition, bridge-rectifiers worsen energy quality and decrease power factor. Three-Phase Pulse Width Modulation (3phase-PWM) rectifiers have seen rapid growth in use due to recent breakthroughs in power device technology. Its benefits include a unity power factor, no harmonic distortion, and pulsating direct currents. An actual power regulation technique based on a novel virtual power grid controller is suggested to cheaply regulate a 3phase-PWM converter with good performance. The study introduces a new sensorless control scheme to improve the 3phase-PWM-performance rectifier. The proposed control technique employs the virtual power grid flux-oriented vector control with a Second-Order Generic Integral with Instantaneous Phase Lock Loop (SOGI-IPLL) to overcome the problems inherent in relying only on low-pass filter estimates. The virtual power grid flux value may then determine the phase difference. The IPLL takes the phase angle as an input and is utilized to construct the power-oriented vector control. The proposed power grid flux-oriented control strategy combines the VFidea with SOGI-IPLL to circumvent the restrictions imposed by relying only on an integrator of low-pass filters (LPF) to provide an estimate. A refined Virtual flux (VF) estimator with direct power control can improve management at a cheaper cost and more efficiently. The modeling findings demonstrate that the DC voltage input and power grid throughput of the 3phase-PWM rectifier can be successfully regulated in both the rectification and inversion states, allowing for the effective functioning of the 3phase-PWM rectifier.

Power System Frequency Dynamics Modeling, State Estimation, and Control using Neural Ordinary Differential Equations (NODEs) and Soft Actor-Critic (SAC) Machine Learning Approaches

With the global energy transition of the electric power system, grid control, supervision, and protection is becoming more challenging. With the increasing integration of renewable energy sources (RES), the system dynamics are changing, causing traditional power system dynamic modeling with swing equation-based modeling approaches to fail. Additionally, the converter-dominated power grid is decreasing the system inertia, making the power system more fragile to the frequency swings. This paper first investigates and compares the application of a model-based Kalman filter state estimation approach with (i) a model-free machine learning approach --- neural ordinary differential equations (NODEs) --- and (ii) a data-driven system identification (SysId) approach to model and infer critical state values of the power system frequency dynamics. Then a model predictive control (MPC) framework is compared to a model-free Soft Actor-Critic (SAC) reinforcement learning (RL) control algorithm in providing efficient fast frequency response (FFR) to the power system frequency dynamics. The approaches are compared in terms of their performance goals as well as their per-timestep computational efficiency. The comparative study for state estimation shows that for the model-free requirement, both NODEs and SysId can provide accurate state estimates; however, with increasing model complexity, NODEs can be a better choice for model identification. Similarly, the results from the FFR comparative study show that the SAC RL-based FFR, once trained, outperforms MPC with better control signals and faster computation time, making the SAC RL-based FFR better option for providing FFR to the power system.

Toward Fair Power Grid Control: A Hierarchical Multiobjective Reinforcement Learning Approach

Toward Fair Power Grid Control: A Hierarchical Multiobjective Reinforcement Learning Approach

Prediction of Electricity Generation Using Onshore Wind and Solar Energy in Germany

Renewable energy production is one of the most important strategies to reduce the emission of greenhouse gases. However, wind and solar energy especially depend on time-varying properties of the environment, such as weather. Hence, for the control and stabilization of electricity grids, the accurate forecasting of energy production from renewable energy sources is essential. This study provides an empirical comparison of the forecasting accuracy of electricity generation from renewable energy sources by different deep learning methods, including five different Transformer-based forecasting models based on weather data. The models are compared with the long short-term memory (LSTM) and Autoregressive Integrated Moving Average (ARIMA) models as a baseline. The accuracy of these models is evaluated across diverse forecast periods, and the impact of utilizing selected weather data versus all available data on predictive performance is investigated. Distinct performance patterns emerge among the Transformer-based models, with Autoformer and FEDformer exhibiting suboptimal results for this task, especially when utilizing a comprehensive set of weather parameters. In contrast, the Informer model demonstrates superior predictive capabilities for onshore wind power and photovoltaic (PV) power production. The Informer model consistently performs well in predicting both onshore wind and PV energy. Notably, the LSTM model outperforms all other models across various categories. This research emphasizes the significance of selectively using weather parameters for improved performance compared to employing all parameters and a time reference. We show that the suitability and performance of a prediction model can vary significantly, depending on the specific forecasting task and the data that are provided to the model.

Analysing Smart Power Grid against different Cyber Attacks on SCADA System

The integration of information and communications technology (ICT) with traditional power grid is transforming the power grid into a more reliable and smarter cyber physical system. A supervisory control and data acquisition (SCADA) system is used to implement the cyber layer for power grid, which is responsible for real-time monitoring and control of power grid and power delivery network. A SCADA system combined with other devices creates smart power grid environment, but it also makes power grid vulnerable against different cyber-attacks as it requires connection with various kinds of open networks. Smart power grids are a extremely critical infrastructure and attack of any kind can affect socio economical condition of the region. In this paper we present a non-complex co-simulation environment for analysing smart power grid and also includes two different cyber-attack scenarios. Index Terms- co-simulation; smart power grid; SCADA system; cyber-attack scenarios.

Dual Converters Coupled to Thermal Grid for Simultaneous Control on Regulation and Compensation in a Hybrid AC-DC Network

In conventional, radial, AC distribution system, the unbalances in load voltages & source currents at distributor junction (DJ) may affect the quality of power. Mostly, a single converter is employed to address any one unbalance at an instant and the power transfer during critical conditions is not met out effectively. To overcome this, an effective, coordinated operation of dual converters (C1&C2) in a hybrid AC-DC network coupled to thermal grid (Electro-thermal network) for meeting the heating demand of a community, is enunciated in this paper. The DC micro grid (DCM) is energized by renewable for serving the community’s heating and public electrical demands. Irrespective of source voltage & load conditions, the C1&C2 aim at voltage regulation & load compensation at DJ, respectively. Also, the control exhibits around 3% THD which is well within the standard levels. During any source outage/less power production, DCM supplies power to loads through inversion. An uninterrupted, quality power to loads at DJ by the dual converters, irrespective of source voltage & load variations along with a conservative approach through the thermal grid, claims the prime merit of the setup. Various unbalance cases including voltage sag & swell, no source & excess DCM power cases are examined through a supervisory control algorithm for analyzing the effectiveness of the control & system flexibility. The effective performance of the above is verified in real time using Hardware-in-Loop approach through DS1104 (dSPACE Controller) and OP4510 (Opal-Real Time) Controllers.

An Intelligent Classification Framework for Complex PQDs Using Optimized KS-Transform and Multiple Fusion CNN

Intelligent classification of multiple power quality disturbances (PQDs) is a top priority in pollution control of the power grid. However, the large-scale application of renewable energy introduces lots of nonlinear and impact loads, which makes the PQDs more complex and challenges the effectiveness of conventional detection frameworks. In this paper, a novel framework based on optimized Kaiser window-based S-transform (OKST) and multiple fusion convolutional neural network (MFCNN) is proposed to identify multiple complex PQDs. First, the OKST is used for the time-frequency positioning of PQDs, where an improved control function is proposed to meet different detection requirements of time-frequency. Additionally, the parameters of the control function are adjusted automatically using maximum energy concentration. Then, the MFCNN based on residual networks (ResNets) is further proposed to extract and classify these time-frequency features automatically. In MFCNN, feature information is fused using different convolution kernels at a 2-dimensional level, which can effectively reduce information loss and improve classification performance. The network model is set up using the Pytorch platform, and the dataset containing 28 types of PQDs and 2 types of nonlinearly mixed PQDs is built to test our framework. The result shows that the proposed OKST-MFCNN obtains an average accuracy of 99.38% under the 20 dB noise level, which is more accurate and robust than some advanced PQDs detection frameworks. Moreover, the accuracy of 97.94% is achieved with satisfactory real-time performance in hardware platform experiments, proving its superior identification performance for complex PQDs.