Figure - available from: Electrical Engineering
This content is subject to copyright. Terms and conditions apply.
Source publication
![Variation of hyper-parameter (l,σf2,σn2)\documentclass[12pt]{minimal}...](publication/353341688/figure/fig5/AS:11431281117970055@1675651871001/Variation-of-hyper-parameter-l-sf2-sn2documentclass12ptminimal_Q320.jpg)
Modern power sector requires grid observability under all scenarios for its ideal functioning. This enforces the operator to incorporate state estimation solutions based on a priori measurements to deduce the corresponding operating states of the grid. The key principle for such aforementioned algorithms lies on an occurrence of an over determined...
Similar publications
The ability of government agencies to assign accurate ages of fish is important to fisheries management. Accurate ageing allows for most reliable age-based models to be used to support sustainability and maximize economic benefit. Assigning age relies on validating putative annual marks by evaluating accretional material laid down in patterns in fi...
Citations
... Unexpected events such as unexpected peak load demands, equipment failures, weather fluctuations, and grid instabilities may affect normal energy consumption patterns. It is important to do extensive testing under these conditions in order to assess how effectively deep learning methods react to these unforeseen scenarios [93]. The deep learning models should accurately predict load variations and immediately perform appropriate actions, such as load shedding or enabling backup power sources, which must be evaluated [94]. ...
Electricity is one of the mandatory commodities for mankind today. To address challenges and issues in the transmission of electricity through the traditional grid, the concepts of smart grids and demand response have been developed. In such systems, a large amount of data is generated daily from various sources such as power generation (e.g., wind turbines), transmission and distribution (microgrids and fault detectors), load management (smart meters and smart electric appliances). Thanks to recent advancements in big data and computing technologies, Deep Learning (DL) can be leveraged to learn the patterns from the generated data and predict the demand for electricity and peak hours. Motivated by the advantages of deep learning in smart grids, this paper sets to provide a comprehensive survey on the application of DL for intelligent smart grids and demand response. Firstly, we present the fundamental of DL, smart grids, demand response, and the motivation behind the use of DL. Secondly, we review the state-of-the-art applications of DL in smart grids and demand response, including electric load forecasting, state estimation, energy theft detection, energy sharing and trading. Furthermore, we illustrate the practicality of DL via various use cases and projects. Finally, we highlight the challenges presented in existing research works and highlight important issues and potential directions in the use of DL for smart grids and demand response.
... Many of the existing machine learning techniques if used for power system state estimation demonstrate drawbacks in capturing the complex dynamics and constraints of power systems [1]. Relying solely on statistical patterns in historical data can lead to inaccurate estimates, particularly in scenarios with limited or noisy data. ...
State estimation is the cornerstone of the power system control center since it provides the operating condition of the system in consecutive time intervals. This work investigates the application of physics-informed neural networks (PINNs) for accelerating power systems state estimation in monitoring the operation of power systems. Traditional state estimation techniques often rely on iterative algorithms that can be computationally intensive, particularly for large-scale power systems. In this paper, a novel approach that leverages the inherent physical knowledge of power systems through the integration of PINNs is proposed. By incorporating physical laws as prior knowledge, the proposed method significantly reduces the computational complexity associated with state estimation while maintaining high accuracy. The proposed method achieves up to an 11% increase in accuracy, a 75% reduction in the standard deviation of results, and 30% faster convergence, as demonstrated by comprehensive experiments on the IEEE 14-bus system.
... In works [72][73][74], one can find an overview of solutions containing the use of these algorithms in analyses of the power system operation. Examples of topics considered as part of the use of selected machine learning methods concern the analysis of disturbances in the power system [75], operation, control, planning and diagnostics [76][77][78], as well as forecasting [79][80][81][82][83] and many others. These methods are used not only in power engineering but also in other fields of science and industry, e.g., in aircraft (in aeronautics) [84][85][86]. ...
Modern power engineering is struggling with various problems that have not been observed be-fore or have occurred very rarely. The main cause of these problems results from the increasing number of connected distributed electricity sources, mainly renewable energy sources (RESs). Therefore, energy generation is becoming more and more diverse, both in terms of technology and location. Grids that have so far worked as receiving networks change their original function and become generation networks. The directions of power flow have changed. In the case of dis-tribution networks, this is manifested by power flows towards transformer stations and further to the network with a higher voltage level. As a result of a large number of RESs, their total share in the total generation increases. This has a significant impact on various aspects of the operation of the power system. Voltage profiles, branch loads, power flows and directions of power flows be-tween areas change. As a result of the random nature of RES generation, there are problems with the quality of electricity, source stability issues, branch overloading, voltage exceedances and power balance. The occurrence of various types of problems requires the use of more and more advanced methods to solve them. This review paper, which is an introduction to the Special Issue Advanced Optimisation and Forecasting Methods in Power Engineering, describes and justifies the need to reach for effective and available mathematical and IT methods that are necessary to deal with the existing threats appearing in the operation of modern power systems. It indicates exemplary, current problems and advanced methods to solve them. This article is an introduction and justifi-cation for the use of advanced calculation methods and algorithms. Engineering intuition and ex-perience are often not enough due to the size and complexity of power grid operation. Therefore, it becomes necessary to use methods based on artificial intelligence and other advanced solutions that will facilitate and support decision making in practice.
... This method helps to capture the uncertainty in power system operation and ensures a diverse set of data points for model training. By analyzing the statistical properties of the generated samples, machine learning models can be trained to make accurate predictions of system behavior under different operating conditions [34]- [36]. By using a combination of these database generation methods, it is possible to create a diverse and representative dataset for training machine learning models in data-driven power system security assessment [37]. ...
Boosting the complexity of the electricity network, penetration of renewable resources, and modernization of power systems has resulted in an increase in the complexity of the power systems security assessment (PSSA). In this context, to decrease the vulnerability of the systems to multiple instability threats and security issues while ensuring the safe operation of the power systems, providing effective online security assessment methods capable of monitoring the systems’ security under varying conditions is vital. However, although the traditional methods have demonstrated efficient PSSA performance, intelligent data-driven approaches have effectively overcome the traditional approaches by delivering impressive and rapid PSSA performance. Artificial intelligence (AI) -based techniques are required to guarantee the efficient, optimal, and safe security assessment. The usage of AI is emphasized due to its computational speed for online performance and its flexibility for providing corrective actions for insecure operating conditions to achieve a seamless transition in power systems. In this review, various available data-driven methods in power system security are comprehensively reviewed into two primary classifications: static and dynamic security assessment. The evaluated study aims to highlight the merits and demerits of developed techniques as well as their limitations to provide decision-making assistant for future investigations.
... The model is applied or further trained on the comprehensive training data set based on the error value until a minimum satisfactory error is obtained [22], [23], [24]. Machine learning methods allow computers to work independently, to learn, perceive, evolve, and adapt [25]. 1. Supervised Machine learning means learning by monitoring each step from the input to the known final output value. ...
... ECF (Environmental Complexity Factor) is one of the factors affecting the size of the project expressed in Use Case points. It is calculated according to the following formulas (24), (25): ...
... The prediction meter, which measures the proportion of projects with errors below a given standard, enables to verify the accuracy and dependability of the used models. The prediction value is constantly 100% for all three proposed criteria, PRED (25), PRED (30), and PRED (50), and for all three experimental phases (training, testing, and validation), when utilizing both of the suggested structures, as shown in Table 64. This steady 100% prediction rate highlights the models' ability to accurately and successfully estimate effort and cost using the UCP model, across a variety of criteria and experimentation phases. ...
The swift advancement of computer technology leads to an even faster development and application of artificial intelligence, its various techniques and methods through the construction of various models for different purpose. Today, software is applied in basically all spheres of life; people use it and do not think about its life cycle. It starts from the simplest mobile applications to complex software used to perform tasks that completely replace humans. This book will show some of the fundamental techniques and methods of artificial intelligence used in everyday life and business. They mainly differ concerning the aspect of the application, but the most significant number of models combine various techniques and methods depending on the problem to be solved. In addition to explaining and applying the essential aspects of division and their basic techniques, examples of different combined approaches will be given. The central part of the book will devoted to applying artificial intelligence tools and techniques in software engineering, more precisely, in software estimation. In the context of estimating effort and costs associated with software project development and implementation, this book will introduce various approaches and methodologies, including artificial neural networks, Taguchi's method, software quality metrics, and other relevant criteria. Furthermore, it can be concluded that three new, improved, specially constructed models can give much better results in estimation and thus significantly improve this area of software engineering. Finally, this book is intended for students who want to expand their knowledge and skills in the field of artificial intelligence application in software engineering and software project management, as well as for all other experts in these areas
... Such schemes incorporate advanced deep learning structures which work as multilabel classifiers and are capable of detecting the locations of intrusions of attack with high accuracy. With a rapid advancement in machine learning approaches, an effective state forecasting scheme has been showcased [39,40]. Such trained state forecasting models demonstrate a minimal root mean squared error (RMSE), mean squared error (MSE) and mean absolute error (MAE) index. ...
With the rapid adoption of renewables within the conventional power grid, the need of real-time monitoring is inevitable. State estimation algorithms play a significant role in defining the current operating scenario of the grid. False data injection attack (FDIA) has posed a serious threat to such kind of estimation strategies as adopted by modern grid operators by injecting malicious data within the obtained measurements. Real-time detection of such class of attacks enhances grid resiliency along with ensuring a secured grid operation. This work presents a novel real-time FDIA identification scheme using a deep learning based state forecasting model followed with a novel intrusion detection technique using the error covariance matrix. The proposed deep learning architecture with its optimum class of hyper-parameters demonstrates a scalable, real-time, effective state forecasting approach with minimal error margin. The developed intrusion detection algorithm defined on the basis of the error covariance matrix furnishes an effective real-time attack detection scheme within the obtained measurements with high accuracy. The aforementioned propositions are validated on the standard IEEE 14-bus test bench.
... Based on the structure and protection policies of the MMC-HVDC system, the faults of the MMC-HVDC system are divided into three parts: the internal fault, AC system fault, and DC line fault [20][21][22][23]. The conventional AC side fault removal method is proposed in [24], ignoring the internal operation mechanism of the sub-module. ...
With the development of power energy technology, flexible high voltage direct current (HVDC) systems with high control degree of freedom flexibility, power supply to passive systems, small footprint, and other advantages stand out in the field of long-distance large-capacity transmission engineering. HVDC transmission technology based on a modular multilevel converter has been widely used in power grids due to its advantages such as large transmission capacity, less harmonic content, low switching loss, and wide application field. In the modular multilevel converter (MMC)-based HVDC system, the protection strategy of converter station internal faults is directly related to the reliability and security of the power transmission system. Starting from the MMC topological structure, this paper establishes the MMC mathematical model in a synchronous rotation coordinate system by combining the working state of sub-modules and the relationship between each variable of the upper and lower bridge arms of each phase of the MMC. It provides a theoretical basis for the design of the MMC-HVDC control system. The causes of the AC system faults and the internal faults of the converter station in the MMC-HVDC system are analyzed, and the sub-module faults and bridge arm reactor faults in the converter station are studied. The sub-module redundancy protection and bridge arm overcurrent protection strategies are designed for the faults, and the correctness of the scheme is verified by Matlab/Simulink.
The norm in the design of epidemic agent-based models (ABM) for communication networks is to represent only the parameters of propagation and control for a particular malware infection. However, nothing is said about the transmission of data packets, i.e., the routing protocols in such models. Most likely, this trend originated from the design of ABMs for disease propagation in biological/social networks, where the concept of packet transmission is absent. The inherent assumption is that ordinary ABMs are enough to highlight infection strategies of viruses/worms. Therefore, the study aims to complement a hypothetical epidemic agent computational model for wireless sensor network by including actual routing protocols such as flooding and gossiping using NetLogo version 5.3.1—a popular programmable multi-agent language. Coding was done using the agent-oriented programming approach, and the simulation experiments for the two protocols were performed. Implementing these data transmission strategies strengthens the possibilities of reifying the actual communication networks, thereby advocating their use in subsequent ABMs.
The application of machine learning (ML) to power and energy systems (PES) is being researched at an astounding rate, resulting in a significant number of recent additions to the literature. As the infrastructure of electric power systems evolves, so does interest in deploying ML techniques to PES. However, despite growing interest, the limited number of reported real-world applications suggests that the gap between research and practice is yet to be fully bridged. To help highlight areas where this gap could be narrowed, this article discusses the challenges and opportunities in developing and adapting ML techniques for modern electric power systems, with a particular focus on power distribution systems. These systems play a crucial role in transforming the electric power sector and accommodating emerging distributed technologies to mitigate the impacts of climate change and accelerate the transition to a sustainable energy future. The objective of this article is not to provide an exhaustive overview of the state-of-the-art in the literature, but rather to make the topic accessible to readers with an engineering or computer science background and an interest in the field of ML for PES, thereby encouraging cross-disciplinary research in this rapidly developing field. To this end, the article discusses the ways in which ML can contribute to addressing the evolving operational challenges facing power distribution systems and identifies relevant application areas that exemplify the potential for ML to make near-term contributions. At the same time, key considerations for the practical implementation of ML in power distribution systems are discussed, along with suggestions for several potential future directions.
