Search

Due to the diversity of power consumption patterns, the false positive rate (FPR) of data-driven electricity theft detection (ETD) methods is too high to meet practical needs, which severely restricts the engineering application of data-based methods. To reduce FPR of ETD methods based on advanced metering infrastructure (AMI), a deep neural network with low FPR (LFPR-DNN) is proposed in this paper. First, a deep model is constructed based on one-dimensional convolution and residual network, which can automatically extract features from consumption data. Then, a two-stage training scheme is used to train the network. In the first stage, the conventional gradient descent algorithm is used to update the network weights. To minimize the impact of data imbalance on detection performance, focal loss is used. Besides, grid search is used to optimize the hyper-parameters of the model. In the second stage, with FPR as the optimization objective, the particle swarm optimization (PSO) algorithm is used to train the network. Finally, the proposed LFPR-DNN is verified by using the open Irish data set. Compared to other state-of-the-art classifiers, LFPR-DNN has the lowest FPR with 0.29% and the highest AUC with 99.42%. The FPR is reduced by an order of magnitude, which verifies the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Economic dispatch (ED) considering valve-point effect, multiple fuel options, prohibited operating zones of generation units is a more accurate model compared to a conventional ED model. It is non-smooth and thus evolutionary algorithms (EAs) are so far the only feasible approaches for the model. In Part II of the paper, a new method, the dimensional steepest decline method (DSD), is proposed for the ED with non-smooth objectives. The DSD is based on the local minimum analysis of the ED problem presented in Part I of the paper. The fuel cost's decline rate between each two adjacent singular points is utilized to find the optimal solutions in a serial sequence. The computational complexity of the DSD is analyzed. The DSD has been applied to solve different types of ED problems on different test systems, including large-scale systems. The simulation results show that DSD can obtain more accurate solutions and consume much less time and its advantage is more obvious on large-scale systems, in comparison with the state-of-art EAs. [ABSTRACT FROM PUBLISHER]

Transient stability and short-term voltage stability have successively attracted the attention of electric power industry. This paper proposes a novel systematic approach for dynamic VAR planning to improve short-term voltage stability level and transient stability level. The dynamic VAR planning problem is formulated as a multi-objective optimization (MOO) model with objectives including investment cost, short-term voltage stability level, and transient stability level. To reduce the complexity of the proposed MOO model, K-means clustering-based severe contingencies selection and global sensitivity analysis-based potential buses selection are employed, leading to a simplified MOO model. The combination of a surrogate modeling technique called support vector regression and the multi-objective evolutionary algorithm (MOEA) are then used to solve the simplified MOO model, considering both the accuracy of models and the optimization computation cost. This combination makes it feasible to perform multiple runs of MOEAs for weakening the effect of the MOEA's randomness to optimal results and offering more diverse Pareto-optimal solutions for decision makers. Simulations are carried on the IEEE 39-bus system and a real power grid of China, illustrating that our methodology is reliable with high efficiency. [ABSTRACT FROM PUBLISHER]

This paper proposes a systematic approach to coordinately design probabilistically-robust wide-area power system stabilizers (WPSSs) for suppressing inter-area oscillations of power systems incorporating wind power. Specifically, the operating point of the system varies stochastically due to wind power integration and each operating point corresponds to a wind power generation scenario in the steady state. Thus, the WPSSs tuned by solving a delicately formulated optimization problem can maximize the occurrence probability of scenarios where the inter-area modes possess the acceptable damping ratios, and strictly constrain their unfavorable impacts. Multiple contingencies are also directly considered. In addition, several advanced techniques are tactfully employed for accurate and efficient evaluation of occurrence probability (objective function) during the optimization so as to ensure the proposed tuning method can deal with the highly nonlinear relationships between the system eigenvalues and the steady-state power outputs of wind farms; a customized differential evolution algorithm is proposed as well to efficiently solve the formulated optimization problem. Simulations and comparisons conducted on two classic test systems with proper modifications show the effectiveness and efficiency of the proposed control design method. [ABSTRACT FROM PUBLISHER]

This paper presents an application of evolutionary particle swarm optimization (EPSO)-based methods to evaluate power system reliability. Population-based (PB) methods appear as competitors to the traditional Monte Carlo simulation (MCS), because they are computationally efficient in estimating a variety of reliability indices. The work reported in this paper demonstrates that EPSO variants can focus the search in the region of the state space where contributions to the formation of a reliability index may be found, instead of conducting a blind sampling of the space. The results obtained with EPSO are compared to MCS and with other PB methods. [ABSTRACT FROM AUTHOR]

This paper introduces a new method of removing thermal overloads and voltage limits in an electric power system by means of the Evolution of Corrective and Preventive control Actions (ECPA). The goal is to find the minimum number of control actions that solve the identified limit violations at minimum cost. A recombination operator based on form theory allows the codification of control actions in a natural and simple way. ECPA has been tested on the IEEE 30-bus and the IEEE 118-bus systems. The limit violations are solved at minimum cost and with fewer control actions on average than alternative methods. [ABSTRACT FROM AUTHOR]

The dynamic economic dispatch problem is a high-dimensional complex constrained optimization problem that determines the optimal generation from a number of generating units by minimizing the fuel cost. Over the last few decades, a number of solution approaches, including evolutionary algorithms, have been developed to solve this problem. However, the performance of evolutionary algorithms is highly dependent on a number of factors, such as the control parameters, diversity of the population, and constraint-handling procedure used. In this paper, a self-adaptive differential evolution and a real-coded genetic algorithm are proposed to solve the dynamic dispatch problem. In the algorithm design, a new heuristic technique is introduced to guide infeasible solutions towards the feasible space. Moreover, a constraint-handling mechanism, a dynamic relaxation for equality constraints, and a diversity mechanism are applied to improve the performance of the algorithms. The effectiveness of the proposed approaches is demonstrated on a number of dynamic economic dispatch problems for a cycle of 24 h. Their simulation results are compared with each other and state-of-the-art algorithms, which reveals that the proposed method has merit in terms of solution quality and reliability. [ABSTRACT FROM AUTHOR]

This paper explores a technique denoted LASCA to solve large scale optimization problems with metaheuristics by reducing the search space dimension with autoassociative neural networks. The technique applies autoencoders as a reversible mapping between the original problem space and a reduced space. A metaheuristic then evolves in the latter, having its objective function assessed in the original space. The technique is illustrated with an application of an Evolutionary Particle Swarm Optimization (EPSO) algorithm to four benchmarking unconstrained optimization functions and to a wind-hydro constrained coordination problem. The new technique allows an improvement in the quality of the solutions attained. [ABSTRACT FROM AUTHOR]

Identifying coherent sub-graphs in networks is important in many applications. In power systems, large systems are divided into areas and zones to aid in planning and control applications. But not every partitioning is equally good for all applications; different applications have different goals, or attributes, against which solutions should be evaluated. This paper presents a hybrid method that combines a conventional graph partitioning algorithm with an evolutionary algorithm to partition a power network to optimize a multi-attribute objective function based on electrical distances, cluster sizes, the number of clusters, and cluster connectedness. Results for the IEEE RTS-96 show that clusters produced by this method can be used to identify buses with dynamically coherent voltage angles, without the need for dynamic simulation. Application of the method to the IEEE 118-bus and a 2383-bus case indicates that when a network is well partitioned into zones, intra-zone transactions have less impact on power flows outside of the zone; i.e., good partitioning reduces loop flows. This property is particularly useful for power system applications where ensuring deliverability is important, such as transmission planning or determination of synchronous reserve zones. [ABSTRACT FROM PUBLISHER]

Robust design methods for power systems must deal with the fixed-structure of controllers and the system large dimension. This limits the applicability of the available design methods. In this paper, the power system stabilizer (PSS) design is posed as a multiobjective optimization problem combining performance and robustness requirements. Robustness is dealt with using concepts of pseudospectra. Evolutionary algorithms are used to solve the multiobjective optimization problem. The method is applied to a small test system, in order to clarify its characteristics and to the New England New York test system, selected for comparison purposes since it is a widely used test system. [ABSTRACT FROM PUBLISHER]

Reserve services are an essential part of a standard market design and are under development in major electricity markets. This paper presents the development of differential evolution algorithm for optimal allocation of energy and spinning reserve. The optimization is performed in an integrated market with biddable spinning reserve and sectional price offer curves while considering all security and power systems constraints in steady state and system credible contingencies. The effectiveness of the proposed method is examined by application of the proposed algorithm on the IEEE 30-bus test system. Some comparisons are made with analytical and evolutionary methods in various cases. [ABSTRACT FROM PUBLISHER]

In this paper, fuzzy inductive reasoning (FIR) is applied to the problem of short-term load forecasting (STLF) in power systems for a day in advance. The FIR model learns both past and future relations from the load and the temperature. The proposed optimization model uses an evolutionary algorithm based on a local random controlled search-simulated rebounding algorithm (SRA)-to choose the inputs to the FIR model. Using an optimization method to determine linear and nonlinear relationships between the variables, a parsimonious set of input variables can be identified improving the accuracy of the forecast. The input variables are updated when a new load pattern is happened or when relative errors are unacceptable. With this update is achieved, a better monitoring of the load trend due to the process is not strictly stationary. The FIR and SRA methodology is applied to the Ecuadorian power system as an application example. Results and comparisons with other STLF methodologies (autoregressive integrated moving average, artificial neural networks, and adaptive neuro-fuzzy inference system) are shown, and conclusions are derived. [ABSTRACT FROM AUTHOR]

This paper proposes a methodology to achieve integrated planning and projects for secondary distribution circuits. The planning model is formulated as a mixed integer nonlinear programming problem (MINLP). In order to resolve this problem, a tabu search (TS) algorithm is used, with a neighborhood structure developed to explore the physical characteristics of specific geographies included in the planning and expansion of secondary networks, thus obtaining effective solutions as well as low operating costs and investments. The project stage of secondary circuits consists of calculating the mechanical efforts to determine the support structures of the primary and secondary distribution systems and determining the types of structures that should be used in the system according to topological and electrical parameters of the network and, therefore, accurately assessing the costs involved in the construction and/or reform of secondary systems. A constructive heuristic based on information of the electrical and topological conditions between the medium voltage and low voltage systems is used to connect the primary systems and secondary circuits. The results obtained from planning and design simulations of a real secondary system of electric energy distribution are presented. [ABSTRACT FROM AUTHOR]

The application of evolutionary computation methods in search and optimization has been growing over the past few decades. As a promising approach in metaheuristic optimization algorithms, differential evolution (DE) has been attracting increasing attention for wide engineering applications including power engineering. Different from conventional evolutionary algorithms using predefined probability distribution function for mutation process, differential evolution exploits the differences of randomly sampled pairs of objective vectors for its mutation process. Consequently the variation between vectors will outfit the objective functions topographical information toward the optimization process, and therefore provides efficient global optimization capability. However, although DE is shown to be precise, fast as well as robust, its search efficiency will be impaired during solution process with fast descending diversity of population. In this paper, detailed numerical studies are carried out to propose the characterization of the performance of several DE mutation methods with and without fitness sharing scheme. All the approaches using the proposed modified DE are presented on an example in power system planning. [ABSTRACT FROM AUTHOR]

This paper describes the use of an interactive evolutionary algorithm for the identification and selection of direct load control actions in electrical distribution networks. The evolutionary algorithm accommodates a progressive articulation of the decision maker's preferences by changing aspiration or reservation levels used in the fitness assessment of the individuals in the population (load control strategies). Genetic operators have revealed as an adequate way to supply the evolutionary process with relevant information about the search results. Besides contributing to reduce the scope of the search, and thus the computer effort, this also enables the identification of solutions more in accordance with the decision maker's evolving preferences. [ABSTRACT FROM AUTHOR]

In rural power systems, the automatic voltage regulators (AVRs) help to reduce energy losses and to improve the energy quality of electric utilities, compensating the voltage drops through distribution lines. In order to help electric companies in the decision-making process, this paper presents a method to define the optimal location of a set of AVRS in electric distribution networks. The optimization process is treated as a multiobjective problem considering the total power losses and the voltage drops in the system as the objectives to be optimized. A novel technique called micro genetic algorithm (pGA) is used to solve the multiobjective problem. This technique is capable of finding, in a very efficient way, the Pareto optimal solutions, giving the decision maker a set of possible (trade-off) solutions from which to choose. [ABSTRACT FROM AUTHOR]

Evolutionary algorithms are heuristic methods that have yielded promising results for solving nonlinear, nondifferentiable, and multi-modal optimization problems in the power systems area. The differential evolution (DE) algorithm is an evolutionary algorithm that uses a rather greedy and less stochastic approach to problem solving than do classical evolutionary algorithms, such as genetic algorithms, evolutionary programming, and evolution strategies. DE also incorporates an efficient way of self-adapting mutation using small populations. The potentialities of DE are its simple structure, easy use, convergence property, quality of solution, and robustness. This paper proposes a new approach for solving economic load dispatch problems with valve-point effect. The proposed method combines the DE algorithm with the generator of chaos sequences and sequential quadratic programming (SQP) technique to optimize the performance of economic dispatch problems. The DE with chaos sequences is the global optimizer, and the SQP is used to fine-tune the DE run in a sequential manner. The combined methodology and its variants are validated for two test systems consisting of 13 and 40 thermal units whose incremental fuel cost function takes into account the valve-point loading effects. The proposed combined method outperforms other state-of-the-art algorithms in solving load dispatch problems with the valve-point effect. [ABSTRACT FROM AUTHOR]

Distribution system problems, such as planning, loss minimization, and energy restoration, usually involve network reconfiguration procedures. The determination of an optimal network configuration is, in general, a combinatorial optimization problem. Several Evolutionary Algorithms (EAs) have been proposed to deal with this complex problem. Encouraging results have been achieved by using such approaches. However, the running time may be very high or even prohibitive in applications of EAs to large-scale networks. This limitation may be critical for problems requiring online solutions. The performance obtained by EAs for network reconfiguration is drastically affected by the adopted computational tree representation. Inadequate representations may drastically reduce the algorithm performance. Thus, the employed representation for chromosome encoding and the corresponding operators are very important for the performance achieved. An efficient data structure for tree representation may significantly increase the performance of evolutionary-based approaches for network reconfiguration problems. The present paper proposes a tree encoding and two genetic operators to improve the EA performance for network reconfiguration problems. The corresponding EA approach was applied to reconfigure large-scale systems. The performance achieved suggests that the proposed methodology can provide an efficient alternative for reconfiguration problems. [ABSTRACT FROM AUTHOR]

This paper presents a new multiobjective evolutionary algorithm for Environmental/Economic power Dispatch (EED) problem. The EED problem is formulated as a nonlinear constrained multiobjective optimization problem. A new Strength Pareto Evolutionary Algorithm (SPEA) based approach is proposed to handle the EED as a true multiobjective optimization problem with competing and noncommensurable objectives. The proposed approach employs a diversity-preserving mechanism to overcome the premature convergence and search bias problems. A hierarchical clustering algorithm is also imposed to provide the decision maker with a representative and manageable Pareto-optimal sell Moreover, fuzzy set theory is employed to extract the best compromise nondominated solution. Several optimization runs of the proposed approach have been carried out on a standard test system. The results demonstrate the capabilities of the proposed approach to generate well-distributed Pareto-optimal solutions of the multiobjective EED problem in one single run. The comparison with the classical techniques demonstrates the superiority of the proposed approach and confirms its potential to solve the multiobjective EED problem. In addition, the extension of the proposed approach to include more objectives is a straightforward process. [ABSTRACT FROM AUTHOR]