•A core obstacle that prevents the direct comparison of power quality disturbances classification techniques is the lack of a standard database that can be used as a benchmark.•This work proposes an ...open-source software which enables the creation of synthetic power quality disturbances and is designed specifically for comparison of power quality disturbances classifiers.•The software produces several types of standard disturbances from the literature, with varying repetitions and random parameters of the labeled disturbances.•The software includes two reference of deep-learning classifiers which can be used by the community as benchmarks for the development of new and better power quality disturbances classification algorithms.
In recent years power quality monitoring tools are becoming a necessity, and many studies focus on detection and classification of Power Quality Disturbances (PQD)s. However, presently a core obstacle that prevents the direct comparison of such classification techniques is the lack of a standard database that can be used as a benchmark. In this light, we propose here an open-source software which enables the creation of synthetic power quality disturbances, and is designed specifically for comparison of PQD classifiers. The software produces several types of standard disturbances from the literature, with varying repetitions and random parameters of the labeled disturbances, and includes two reference classifiers that are based on deep-learning techniques. Due to the good performance of these classifiers, we suggest that they can be used by the community as benchmarks for the development of new and better PQD classification algorithms. The developed code is available online, and is free to use.
Despite widespread adoption and outstanding performance, machine learning models are considered as “black boxes”, since it is very difficult to understand how such models operate in practice. ...Therefore, in the power systems field, which requires a high level of accountability, it is hard for experts to trust and justify decisions and recommendations made by these models. Meanwhile, in the last couple of years, Explainable Artificial Intelligence (XAI) techniques have been developed to improve the explainability of machine learning models, such that their output can be better understood. In this light, it is the purpose of this paper to highlight the potential of using XAI for power system applications. We first present the common challenges of using XAI in such applications and then review and analyze the recent works on this topic, and the on-going trends in the research community. We hope that this paper will trigger fruitful discussions and encourage further research on this important emerging topic.
•This paper highlights the potential of using XAI for energy and power systems applications.•The challenges and limitations of adopting and implementing XAI techniques in the field of energy and power systems are being covered.•A review of the recent works on the topic of XAI in the energy domain and an analyze of the on-going trends are presented.•Opportunities and future research directions are identified based on applications that use ML but have not yet considered XAI.
The paper addresses the problem of exponential stability of linear control systems defined on nonuniform discrete domains. The main goal of this research is to find necessary and sufficient stability ...conditions in terms of the coefficients of the characteristic polynomial, associated with a system, and to study the geometry of stability regions, determined by the derived conditions in the coefficient space. As the first step in this direction, this paper presents such conditions for a special case of second-order systems defined on a discrete time scale with two asymptotic graininesses. The geometry of the corresponding regions in the coefficient space is illustrated for different values of the asymptotic graininesses.
The paper proves that the order of the minimal state-space realisation of a set of irreducible implicit input--output (i/o) difference equations is equal to the degree of the Dieudonné determinant of ...a certain polynomial matrix. This matrix is defined by the generic linearisation of the i/o equations, and the map that incorporates the knowledge of the i/o equations in the polynomial coefficients. A simple algorithm to compute the degree of the Dieudonné determinant is given. Several examples demonstrate the theoretical result.
Non-Intrusive Load Monitoring (NILM) techniques estimate the consumption of individual appliances in a household or facility, based on readings of a centralized meter. Lately, deep learning ...techniques have demonstrated outstanding performance for NILM predictions. Nevertheless, a possible problem is that users and consumers may find it hard to trust the results of such algorithms if they do not fully understand the reasons for their outputs. In this light, this work presents a method that explains and justifies the outputs of NILM convolutional neural network (CNN) classifiers, using Explainable Artificial Intelligence (XAI) techniques. The method operates as follows: a CNN model for NILM is used to estimate which appliances are activated in the system based on the total power consumption. Then, a XAI technique uses this model and its outputs to explain and justify the prediction of this model. Thereby, the NILM CNN classifier outputs are both accurate and are more interpretable, allowing users to make informed and trustworthy decisions. These ideas are demonstrated on the REDD dataset using a convolutional neural network classifier and two state-of-the-art XAI techniques.
This article studies the realisability property of discrete-time bilinear and quadratic input-output (i/o) equations in the classical state-space form. Constraints on the parameters of the i/o model ...are suggested that lead to realisable models. Using new formulae for computing basis vectors of certain vector spaces of differential one-forms, we present in this article the complete list of the third- and fourth-order realisable i/o bilinear models, and a new realisable subclass of an arbitrary order is suggested. Moreover, we provide the sufficient conditions of the second- and third-order realisable i/o quadratic models, respectively. All the developed theory and algorithms are illustrated by means of several examples.
•The paper addresses the challenge of allocating storage units to minimize frequency deviations.•The Cross-Entropy (CE) method is used for finding the optimal locations of multiple storage units.•The ...CE method is used since the high-dimensional search space does not affect its run time.•An enhanced CE method is developed for cases when locations of potential transients are known.•The enhanced CE method convergences faster and accurately to the optimal solution.
The paper addresses the well-known challenge of allocating storage units within the grid to minimize frequency deviations during transients. The current research is mainly motivated by a sequence of recent works that focus on location of inertia and storage units, considering the limitations of the transmission network, and its spatial effects. One possible drawback of these algorithms is their numerical complexity. To address this gap, this paper studies the storage distribution problem using the general framework of combinatorial optimization, and proposes algorithms to solve it. Among them, one solution is based on exhaustive search, while the others strongly rely on the cross-entropy method. We show that the latter is especially suitable for this application, since it can efficiently solve high-dimensional combinatorial optimization problems. Since the CE method may converge to sub-optimal solutions if the number of storage units increases significantly, this paper also proposes an enhanced CE method which solves the distribution problem considering the assumption that the locations of potential transients are known. In this regard, a key result is that the enhanced CE method convergences faster to the optimal solution and significantly reduces the number of computations.
Modern and complex dynamical modeling requires effective and precise state estimation techniques to make the system modes available for control actions without expensive measurement tools. By ...proposing a reliable Kalman-type estimation algorithm in this study, we aim to synchronize a chaotic fractional order singular (FOS) system which introduces some important information related to memory effects and interconnection properties. Then, the proposed algorithm is used to a secure communication framework, and an image-based cryptography algorithm is provided, in which the main image is well-masked by chaotic FOS signals and is successfully decrypted from the estimated chaotic states in receiver side. Furthermore, the performance of the suggested encryption scheme are demonstrated using histograms, pixel correlation, and key space analyses.