M.A.L. Badr

Economic Dispatch Using an Enhanced Hopfield Neural Network

Abstract This article introduces some modifications to the conventional Hopfield neural network (HNN) to enhance its performance. A comprehensive study of the effect of the HNN parameters on the solution quality of the economic dispatch problem (EDP), as a case study, has been made. By investigating the describing curves, the best values for the HNN parameters are tuned. To further improve the solution quality, an adaptive correction factor is proposed and introduced to the EDP solution obtained by the HNN. To investigate the effect of the modifications on the solution quality of the EDP, three case studies are selected and solved. Comparisons of results are then made with others to prove the validity and effectiveness of the proposed modifications.

ANN-Based Optimal Energy Control of Induction Motor Drive in Pumping Applications

This paper investigates the opportunity for energy saving in a three-phase induction motor (IM) driving pump load and proposes an improved loss model control (LMC). Compared with other power-loss reduction algorithms for IM, the presented one has the advantages of fast and smooth flux adaptation, high accuracy, and versatile implementation. The performance of LMC depends mainly on the accuracy of modeling the motor drive and losses. In this paper, the derived loss model considers the surplus power loss caused by inverter voltage harmonics using closed-form equations and also includes the magnetic saturation. An artificial neural network controller is synthesized and trained offline to determine the optimal flux level that achieves maximum drive efficiency. The drives voltage and speed control loops are connected via the actual stator frequency, making the scheme comparatively reliable. Simulation and experimental studies are performed on 5.5-kW test motor using the proposed control method. The test results are provided and compared with the fixed flux operation to validate the effectiveness.

A new intelligent optimization technique for distribution systems reconfiguration

The loss minimization problem is one of the most important problems to save the operational cost in distribution systems. Therefore, more efficient approaches are required to handle such a combinatorial problem. This paper presents an efficient meta-heuristic method for reconfiguration of distribution systems. A Tabu Search (TS) algorithm is used to reconfigure distribution systems so that active power losses are globally minimized with turning on / off sectionalizing switches. TS algorithm is introduced with some modifications such as using a tabu list with variable size according to the system size; this should lead to robust algorithm, and prevent cycling. Also, a random multiplicative move is used in the search process to diversify the search toward unexplored regions. A new method to check the radial topology of the system is presented. The proposed method is applied to 32-node and 69-node distribution systems. The obtained results using the proposed TS approach are compared with results obtained using Simulated Annealing (SA) approach and Parallel Tabu Search (PTS) approach in the previous work to examine the performance.

Impact of large scale wind power on power system stability

This paper presents a detailed analysis of the impact of large scale wind power generation on both the dynamic voltage stability and the transient stability of electric power systems. The following problems have been analyzed: different penetration of wind power impact on transient stability and on voltage stability (dynamic voltage stability) following a major fault in the transmission system, determination of the acceptable wind power penetration level in power system without deteriorating both transient stability and voltage stability of the system. Moreover, the effect of SVC on the system stability is studied.

Dynamic Economic Dispatch Using a Hybrid Hopfield Neural Network Quadratic Programming Based Technique

Abstract This article introduces a solution for the dynamic economic dispatch problem using a hybrid technique of the Hopfield neural network and quadratic programming. This hybrid algorithm is based on using the enhanced Hopfield neural network to solve the static part of the problem and the quadratic programming algorithm for solving the dynamic part of the dynamic economic dispatch. This technique guarantees the global optimality of the solution due to its look-ahead capability. The proposed technique is applied to and tested on an example from the literature, and the solution is then compared with that obtained by some other techniques to prove the validity and effectiveness of the proposed algorithm.

A hybrid Hopfield neural network-quadratic programming approach for dynamic economic dispatch problem

This paper introduces a solution of the dynamic economic dispatch (DED) problem using a hybrid approach of Hopfield neural network (HNN) and quadratic programming (QP). The hybrid algorithm is based on using enhanced HNN; to solve the static part of the problem; and the QP algorithm for solving the dynamic part of the DED. This technique guarantees the global optimality of the solution due to its look-ahead capability. The new algorithm is applied and tested to an example from the literature and the solution is then compared with that obtained by some other techniques to prove the superiority and effectiveness of the proposed algorithm.

Gene expression programming for static security assessment of power systems

In this paper, a novel gene expression programming (GEP) algorithm is introduced for power system static security assessment. The GEP algorithms as evolutionary algorithms for pattern classification have recently received attention for classification problems because they can perform global searches. The proposed methodology introduces the GEP for the first time in static security assessment problems. The proposed algorithm is examined using different IEEE standard test systems. Different contingency case studies have been used to test the proposed methodology. The GEP based algorithm formulates the problem as a multi-class classification problem using the one-against-all binarization method. The algorithm classifies the security of the power system into three classes, normal, alert and emergency. Performance of the algorithm is compared with other neural network based algorithm classifiers to show its superiority in static security assessment.

A modified particle swarm Algorithm for distribution systems reconfiguration

This paper introduces the Particle Swarm Optimization (PSO) algorithm to solve the optimal network reconfiguration problem for power loss reduction. The PSO is a relatively new and powerful intelligence evolution method for solving optimization problems. It is a population-based approach. The PSO was inspired from natural behavior of the bees on how they find the location of most flowers. The proposed PSO algorithm is introduced with some modifications such as using an inertia weight that decreases linearly during the simulation. This setting allows the PSO to explore a large area at the start of the simulation. A modification in the number of iterations and the population size is also presented. To verify the effectiveness of the proposed PSO algorithm, comparative studies are conducted on two test distribution systems. The obtained results are compared with those obtained using other approaches in the previous work to examine the performance.

Probabilistic Neural Network Classifier for Static Voltage Security Assessment of Power Systems

Abstract In this article, a probabilistic neural network based classifier is used to evaluate the static voltage security of the power system under two contingency cases. In the first case, the neural network classifies the system security based on the voltage profile of each bus in reference to changes in the generation and load profile in the system contingencies, while in the second case, single and double line outages of the system are used to examine the system security. The performance of the proposed probabilistic neural network is compared with the radial basis function neural network and the back-propagation neural network. The probabilistic neural network achieves superior results in comparison to other types of neural networks in both test contingencies. The proposed methodology is examined using IEEE standard test systems, where the output of the probabilistic neural network in each test contingency classifies the security of the power system into three classes—normal, alert, and emergency.

Implementing distributed generation to mitigate under-frequency load shedding

This paper proposes a strategy for employing distributed generation units to stop the frequency deterioration resulting from severe disturbances in electrical distribution systems rather than relying on load shedding as the obtainable treatment. Also, sufficient comparison is made between both treatment methods. Low order system frequency response model (SFR) is used to estimate the frequency deviation. The model parameters are adjusted when the system is equipped with the distribution generation units. The amounts of load to be shed or the power to be delivered by DGs are compared and the treatment method that presents better dynamic response specifications is recommended.