Distribution Network Reactive Power Planning Based on Improved Evolutionary Strategy Theory

2013 ◽  
Vol 732-733 ◽  
pp. 1023-1028
Author(s):  
Si Qing Sheng ◽  
Xing Li ◽  
Yang Lu
Keyword(s):  
Reactive Power ◽  
Distribution Network ◽  
Load Condition ◽  
Optimal Solution ◽  
Evolution Strategy ◽  
Evolutionary Strategy ◽  
Initial Population ◽  
Fitness Value ◽  
Cauchy Mutation ◽  
Reactive Power Planning

In this paper a distribution network reactive power planning mathematical model was established, taking the minimized sum of electrical energy loss at the different load operation modes and the investment for reactive power compensation equipments as objective function to solve the planning question respectively and taking the transformer tap as equality constraint. The evolution strategy is improved, The Euclidean distance is introduced into the formation of the initial population, and the initial population under the max load operation mode is based on the optimal solution of the min load condition. The Cauchy mutation and variation coefficient are introduced into the evolution strategy method. By means of improvement of fitness to ensure diversity of population in early and accuracy of the fitness value.

scholarly journals Design of Gas Cyclone Using Hybrid Particle Swarm Optimization Algorithm

2021 ◽  
Vol 11 (20) ◽  
pp. 9772
Author(s):  
Xueli Shen ◽  
Daniel C. Ihenacho
Keyword(s):  
Particle Swarm Optimization ◽  
Research Effort ◽  
Particle Swarm ◽  
Optimal Solution ◽  
Design Parameters ◽  
Initial Population ◽  
Swarm Optimization ◽  
Hybrid Particle Swarm Optimization ◽  
Fitness Value ◽  
Sensitivity Approach

The method of searching for an optimal solution inspired by nature is referred to as particle swarm optimization. Differential evolution is a simple but effective EA for global optimization since it has demonstrated strong convergence qualities and is relatively straightforward to comprehend. The primary concerns of design engineers are that the traditional technique used in the design process of a gas cyclone utilizes complex mathematical formulas and a sensitivity approach to obtain relevant optimal design parameters. The motivation of this research effort is based on the desire to simplify complex mathematical models and the sensitivity approach for gas cyclone design with the use of an objective function, which is of the minimization type. The process makes use of the initial population generated by the DE algorithm, and the stopping criterion of DE is set as the fitness value. When the fitness value is not less than the current global best, the DE population is taken over by PSO. For each iteration, the new velocity and position are updated in every generation until the optimal solution is achieved. When using PSO independently, the adoption of a hybridised particle swarm optimization method for the design of an optimum gas cyclone produced better results, with an overall efficiency of 0.70, and with a low cost at the rate of 230 cost/second.

Research on Reactive Power Optimization Based on Genetic Algorithms in Distribution Network

2014 ◽  
Vol 644-650 ◽  
pp. 2476-2478
Author(s):  
Lin Yuan Wang
Keyword(s):  
Genetic Algorithms ◽  
Mutation Rate ◽  
Exponential Function ◽  
Reactive Power ◽  
Power Optimization ◽  
Distribution Network ◽  
Adaptive Mutation ◽  
Initial Population ◽  
Reactive Power Optimization ◽  
Slow Convergence

The reactive power optimization is formulated based on genetic algorithms in distribution net work. SGA has defects of slow convergence and being prone to immature convergence. In order to eliminate the defects, an improved GA is proposed in this thesis. CIP scheme is presented, which can guarantee diversity of the population by designing the initial population to obtain all the values within the definition area. A parameter called individual distributing degree is defined to describe how individuals are distributed in the definition area. Adaptive mutation rate is defined as an exponential function of the retained generations of the Elitism, and it is in inverse proportion to individual distribution degree. It accelerates the convergent process.

A Natural Evolution Based Numerical Optimisation Framework to Develop and Enhance Airfoil-Slat Arrangement

2019 ◽  
Author(s):  
Sushrut Kumar ◽  
Priyam Gupta ◽  
Raj Kumar Singh
Keyword(s):  
Genetic Algorithm ◽  
Fitness Function ◽  
Fuel Efficiency ◽  
Optimal Solution ◽  
Leading Edge ◽  
Initial Population ◽  
Invasive Weed ◽  
Fitness Value ◽  
Leading Edge Slats ◽  
Dynamics Simulations

Abstract Leading Edge Slats are popularly being put into practice due to their capability to provide a significant increase in the lift generated by the wing airfoil and decrease in the stall. Consequently, their optimum design is critical for increased fuel efficiency and minimized environmental impact. This paper attempts to develop and optimize the Leading-Edge Slat geometry and its orientation with respect to airfoil using Genetic Algorithm. The class of Genetic Algorithm implemented was Invasive Weed Optimization as it showed significant potential in converging design to an optimal solution. For the study, Clark Y was taken as test airfoil. Slats being aerodynamic devices require smooth contoured surfaces without any sharp deformities and accordingly Bézier airfoil parameterization method was used. The design process was initiated by producing an initial population of various profiles (chromosomes). These chromosomes are composed of genes which define and control the shape and orientation of the slat. Control points, Airfoil-Slat offset and relative chord angle were taken as genes for the framework and different profiles were acquired by randomly modifying the genes within a decided design space. To compare individual chromosomes and to evaluate their feasibility, the fitness function was determined using Computational Fluid Dynamics simulations conducted on OpenFOAM. The lift force at a constant angle of attack (AOA) was taken as fitness value. It was assigned to each chromosome and the process was then repeated in a loop for different profiles and the fittest wing slat arrangement was obtained which had an increase in CL by 78% and the stall angle improved to 22°. The framework was found capable of optimizing multi-element airfoil arrangements.

Reactive Power Optimization for Distribution Network with Distributed Generators Based on Semi-Definite Programming

2014 ◽  
Vol 1070-1072 ◽  
pp. 809-814
Author(s):  
Lei Dong ◽  
Ai Zhong Tian ◽  
Tian Jiao Pu ◽  
Zheng Fan ◽  
Ting Yu
Keyword(s):  
Mathematical Model ◽  
Reactive Power ◽  
Power Optimization ◽  
Distribution Network ◽  
Optimal Solution ◽  
Mixed Integer ◽  
Reactive Power Optimization ◽  
Global Optimal Solution ◽  
Distributed Generators ◽  
Global Optimal

Reactive power optimization for distribution network with distributed generators is a complicated nonconvex nonlinear mixed integer programming problem. This paper built a mathematical model of reactive power optimization for distribution network and a new method to solve this problem was proposed based on semi-definite programming. The original mathematical model was transformed and relaxed into a convex SDP model, to guarantee the global optimal solution within the polynomial times. Then the model was extended to a mixed integer semi-definite programming model with discrete variables when considering discrete compensation equipment such as capacitor banks. Global optimal solution of this model can be obtained by cutting plane method and branch and bound method. Numerical tests on the modified IEEE 33-bus system show this method is exact and can be solved efficiently.

Optimization of Reactive Power Compensation Rate in 20kV Cable Distribution Network

2014 ◽  
Vol 1008-1009 ◽  
pp. 391-398
Author(s):  
Xiao Lang Lin ◽  
Ze Xing Chen ◽  
Yu Yao Yang ◽  
Jun Xiong Zou ◽  
Zheng Min Zuo ◽  
...  
Keyword(s):  
Reactive Power ◽  
Distribution Network ◽  
Distribution Networks ◽  
Reactive Power Compensation ◽  
Statistical Characteristics ◽  
Analysis Model ◽  
Point Model ◽  
Reactive Power Planning ◽  
Cable Lines ◽  
Compensation Rate

A method based on modeling approach that from points to face and boundary conditioned parameter is put forward in this paper. Starting from the Point Model with statistical characteristics, the characteristic parameters which influence the configuration rate of reactive power compensation (RPC) are adopted as boundary conditions. Expanding from Point Model to Analysis Model, different features of actual 20kV cable distribution feeders can be covered. Then, with the optimization research on RPC of urban cable lines in 20 kV distribution networks, the recommended range of RPC rate in different transformers is worked out, which can be extended to the application of 20 kV urban distribution network. The simulation shows that the proposed method can optimize the reactive power configuration in 20 kV distribution networks without the complicated computation of optimal reactive power planning.

An Intelligent Approach for Reactive Power Planning in Distribution Network with Wind Power

2013 ◽  
Vol 344 ◽  
pp. 279-284
Author(s):  
Ling Yun Wang ◽  
Jie Pan ◽  
Ling Lu
Keyword(s):  
Wind Power ◽  
Wind Farm ◽  
Reactive Power ◽  
Distribution Network ◽  
Power Losses ◽  
Power Requirement ◽  
Real Power ◽  
Control Objective ◽  
Reactive Power Planning ◽  
Intelligent Approach

An intelligent approach for reactive power planning in distribution network with wind power is presented. A genetic optimization algorithm is applied in order to optimize the reactive power dispatch. The reactive power requirement for wind farm is considered as a restriction so that the active power control objective can be achieved while the real power losses are minimized. Finally, the proposed method is tested in the standard IEEE 30 node system with five wind turbines as a simulated wind farm, while considering the capability limits of wind turbine. The numerical results show that the proposed intelligent genetic algorithm can reduce real power losses effectively and improve the node voltage stability.

scholarly journals Accelerating reinforcement learning with a Directional-Gaussian-Smoothing evolution strategy

2021 ◽  
Vol 0 (0) ◽  
pp. 0
Author(s):  
Jiaxin Zhang ◽  
Hoang Tran ◽  
Guannan Zhang
Keyword(s):  
Monte Carlo ◽  
Reinforcement Learning ◽  
Large Scale ◽  
Optimal Solution ◽  
Evolution Strategy ◽  
Evolutionary Strategy ◽  
Gradient Estimates ◽  
Great Promise ◽  
Gaussian Smoothing ◽  
Reward Function

<p style='text-indent:20px;'>The objective of reinforcement learning (RL) is to find an optimal strategy for solving a dynamical control problem. Evolution strategy (ES) has been shown great promise in many challenging reinforcement learning (RL) tasks, where the underlying dynamical system is only accessible as a black box such that adjoint methods cannot be used. However, existing ES methods have two limitations that hinder its applicability in RL. First, most existing methods rely on Monte Carlo based gradient estimators to generate search directions. Due to low accuracy of Monte Carlo estimators, the RL training suffers from slow convergence and requires more iterations to reach the optimal solution. Second, the landscape of the reward function can be deceptive and may contain many local maxima, causing ES algorithms to prematurely converge and be unable to explore other parts of the parameter space with potentially greater rewards. In this work, we employ a Directional Gaussian Smoothing Evolutionary Strategy (DGS-ES) to accelerate RL training, which is well-suited to address these two challenges with its ability to (i) provide gradient estimates with high accuracy, and (ii) find nonlocal search direction which lays stress on large-scale variation of the reward function and disregards local fluctuation. Through several benchmark RL tasks demonstrated herein, we show that the DGS-ES method is highly scalable, possesses superior wall-clock time, and achieves competitive reward scores to other popular policy gradient and ES approaches.</p>

Sign in / Sign up

Export Citation Format

Share Document