An Improved Genetic Algorithm for Optimum Design of Power Transformers

In order to enhance the ability of global searching for genetic algorithm in power transformers optimization design, some interrelated key technique problems such as encoding method, genetic operators, restrict condition, fitness function for the traditional genetic algorithm are further reformed. An Improved Genetic Algorithm (IGA) is developed. The optimal results of a representative mathematical example show that IGA has high efficiency of global searching. At the same time, a multi-objective algorithm based on IGA is studied in this paper. IGA is applied to the single and multi-objective optimum design of S9 power transformers for the first time. All the achievements in the paper are verified a practical S9-1000/10 kV power transformer. All the optimization results are satisfactory and show that IGA has powerful ability of global searching, excellent solution precision and has a bright application prospect in the fields of power transformers design.     

Optimum Analysis of Damper Braces Based on Genetic Algorithm

The aims at optimization of structure with damper braces is studied in this paper. The sum of damping coefficient of all damper braces is considered as a goal function, and the storey displacements are considered as constraint conditions. The program,which introduces the non-proportional damp matrix, based on genetic algorithm and time history is used to analyze the optimization of a damper braced frame. The results are reasonable and show that the genetic algorithm is an efficient method to optimize the damper braced structure.     

Improved Genetic Algorithm of Float Encoding and Its Application

As an optimal method, Genetic Algorithm has obvious advantages, which is based on the nature selection and genetic transmission mechanisms such as high collateral,stochastic,self-reliance. but when in practical application, it usually has problems of premature convergence and result swing near optimum value.To solve the problem of premature convergence, the method called Monte-Carlo is adopted to prevent the algorithm from local optimal, and to the problem of result swing, the method changing the hunting zone dynamically is proposed to improve the accuracy of the optimal result. Further more, it devises programs to optimize the test functions of two famous optimal methods. The test results indicate that the improved Genetic Algorithm is valid, which can not only avoid local optimal but also improve the accuracy of the optimal result.     

Dynamic Route Guidance System Based on Genetic Algorithm

The study on Dynamic Route Guidance System (DRGS) is an important research in the field of Intelligent Transportation System(ITS),which guide the behaviors of travelers by providing them with optimal route based on real-time traffic information. As a result the travel time can be saved and the traffic congestion can be avoided. The route guidance algorithm can compute the best route between the begin point and destination. The globe near best property and real-time property must be considered , the Genetic Algorithm have qualifications for globe optimal and parallel algorithm. Genetic algorithm(GA),for solving the shortest route,is proposed in this thesis?The ordered real code rule,crossover,mutation are given. The efficiency of the GA is proved through an example.     

Material Recombined System Based on Genetic Algorithm

A new and better mathematical model is presented, which is based on the in-depth analysis of the material recombined system and the sufficient expenses produced during the material recombined. In order to solve the problem, the authors choose the genetic algorithm, put forward a improved coding method, and get satisfactory answer through experiments. Advantages of efficiency and economic value are given.     

Learning Causality Diagram Structure Based on Genetic Algorithm

The Causality Diagram theory,which adopted graphical expression of knowledge and direct causality intensity of causality,overcomes some shortages in Belief Network and has evolved into a mixed causality diagram methodology coped with discrete and continuous variable.But it is difficult that the structure of Causality Diagram given by expert.Because the complexity of causality diagram structure goes up exponentially through the number of the vertex's increasing,it is NP-hard problem to find the most possible structure from a set of data.The authors discuss approaches and present Genetic Algorithm,to find the most possible structure from a set of data.Experiment shows the method is effective.     

Distribution Network Structure Planning Based on Improved Immune Genetic Algorithm

Distribution Network Structure planning is a complex combinatorial optimization problem,which is difficult to solve properly by using traditional optimization methods.In order to solve this problem,Improved Immune Genetic Algorithm is introduced to the distribution network optimal planning. Improved Immune Genetic Algorithm draws into the immune diversity and antibody's density mechanism to maintain the individual's diversity and remains evolution algorithm's global stochastic searching ability,so it can promote diversity and the whole optimal-searching ability of genetic algorithm.The optimal module takes the minimum annual cost as its object,and the capacity and voltage drop of feeder and the radiation of distribution network as its restrictions.According to the require of radiation of distribution network,the spanning tree of the alternative network is taken as the initial solution to speed up the calculation.And the branch-exchange method is used in designing crossover operator and mutation operator to avoid the radiation checking and enhance the optimizing ability.This algorithm has been illustrated effectively by examples,at the same time,the calculation example demonstrates that,the algorithm has higher calculation speed than the traditional immune genetic algorithm.     

Hybrid Genetic Algorithm Based on Novel Adaptive Penalty Function

The authors introduce a sort of novel adaptive penalty gene, transform the constrained problem into unconstrained problems. An solution is given for this unconstrained problem with genetic algorithm, and then it is used as initial values for the constrained variable metric method to get precise solution. The numerical experiments illustrate that this hybrid genetic algorithm is more efficient than the genetic algorithm, and at most situations globally optimal solution can be gotten.     

Distribution Network Structure Planning Based on Multi-population Immune Genetic Algorithm

Distribution Network Structure planning is a complex combinatorial optimization problem, which is difficult to solve properly by using traditional optimization methods. The authors put forward Multiple Population Immune Genetic Algorithm (MPIGA)for optimal planning of distribution network structure, and do optimal search to different aspects of optimization goals. During the genetic evolution process, biologic immune mechanism is introduced to do some immune operator operation on chromosomes of each population, which can interact mutually by the shift of excellent units. By this way, it can effectively prevent population retrogression, promote diversity and the whole optimal searching ability of genetic algorithm. In order to minimize network annual expenditure, a mathematic model is established. The optimal solution is obtained by this algorithm, which has been illustrated effectively by specific examples at the same time.     

About the Synthesis Method of Planar Multilink Mechanisms Based on Genetic Algorithm

This paper discusses the method of synthesizing planar multilink mechanisms based on GA. The pivotal technology is coding, creating population, account fitness, genetic operation and defining population size. This method does not require initial mechanisms and can search for plural appropriate mechanisms simultaneously. It is efficient about the non-linear problem. As an example, configuration of 4-bar planar mechanisms is decided in a practical application. It can be used the synthesis of 6-bar or more bar planar mechanisms as well.     

Multiple Load Flow Solutions in Electric Power Systems Based on Genetic Algorithm

Methods for satisfying the power balance requirement and the voltage magnitude costraint are developed and incorporated into the genetic algorithm method to form a constrained genetic algorithm for solving the load flow problem.The robustness of the load flow algorithm is enhanced by the dynamic population.the technique for accelerating the convergence of the optimisation process and the network node sequencing procedure described in the paper.The efficiency and feasibility of the developed CGALF algorithm have been tested using KK 11 node system under light load and heavy load conditions.     

Application of RBF Algorithm Based on Generalized Genetic Optimization in Prediction Model for Molten Iron Desulfuration

Desulfuration process is a very sophisticate reaction which is not only diverse but also non-line. A RBF algorithm based on generalized genetic optimization is proposed after studying the standard genetic and RBF algorithm. The authors also introduce its application in prediction Model for molten Iron Desulfuration. The algorithm perfectly resolve the problem of random selection of RBF cluster center number. Furthermore, it also reduces the time which GA uses. Comparison between the simulation results of RBF and RBF algorithm Based on GGA optimization further proves the efficiency and precision of its application in Prediction Model for Molten Iron Desulfuration. Finally the result of the test shows that after adopting the algorithm, the end-point hitting ratio can reach eighty-five percent. This indicate the algorithm has the engineering practicability.     

遗传算法优化人工神经网络模型在日光温室湿度预报中的应用

提出一种基于遗传算法优化BP神经网络的方法预测日光温室湿度环境因子。实测日光温室内影响空气湿度的环境因子组成数据样本作为神经网络的输入,采用基于实数编码的遗传算法替代随机设定神经网络的初始权阈值,然后通过改进的BP算法在由遗传算法确定的搜索空间中对网络进行精确训练。模型预报值和实测值基于1：1线的决定系数R2和预测平均相对误差MSE分别为0.9857和3．1％。结果表明,遗传算法优化BP神经网络预报模型收敛速度快、预测精度高。可为日光温室的湿度环境调控制提供理论依据和决策支持。     

Wavelet Neural Network Equalizer Using Error Feedback Structure and Algorithm

The inter-symbol interference in the nonlinear time-varying channel is a serious problem in the wireless communication. In order to overcome it, wavelet neural network equalizer using error feedback is employed to cut the auto-correlation of the error signal. Exploiting the decent time-frequency localization of the wavelet analysis, as well as the self-training feature of the neural network, a quicker convergent nalysis and computer simulation confirm the effectiveness of the equalizer and a lower BER are attained. Theoretical aalgorithm. It at The wavelet neural network equalizer based on error feedback advances the communication in the nonlinear time-varying channel.     

Analysis on Application in Profile Optimizing Based on Genetic Algorithms

In order to achieving the Automation of profile optimization and offering more reasonable schemes,the author proposes a kind of computer method based on hereditary algorithm,which is using mathematic methods to simplify the profile lines as a list of points and using cut-fill gross as aim's function and using restriced conditions to control the profile lines,the authors optimize the list of points of changing slopes,the results of the method are more accurate than the artificial design and can provide many kinds of schemes within short time.In the actual projects we can properly increase the condition restraints to design more high-quality lines.If GIS platform is added to it,the profile can be optimized more effectively and more intuitionisticly.     

Optimizing Algorithm for Disassembly Sequence Based on Tabu Search and Genetic Algorithms

In order to optimize disassembly sequence about wornout or malfunctioning products,firstly,it is proposed to build Interference-Freeness Matrix for describing the structure of assembly.Secondly,computing model of automatic generating and optimizing disassembly sequence of assembly is proposed based on Genetic Algorithms.Then,after inputing some disassembly sequences and other controlling parameters,the program can search optimizing disassembly(sequences) valid in geometry.Minimal reorientation number of times during disassembling assembly is assigned as optimizing objective.At last,because the neighborhood may converge too fast and limit the search to a local optimum prematurely during the process of Genetic Algorithms(GAs),the authors combine the strengths of GAs and Tabu search and presented the detailed flow chart of the hybrid approach.More robust search behavior can possibly be obtained by incorporating the(Tabu's) intensification and diversification strategies into GAs.The details of the hybrid approach and a case study are presented here.Much engineering examples is tested to demonstrate the approach.The results given show that the valid disassembly sequences obtained are superior to those derived from GAs alone in fitness value,number and distribution.     

基于遗传算法结合偏最小二乘的潮土碱解氮高光谱特征及含量估测

提取土壤碱解氮特征光谱是利用高光谱数据进行其含量估测的关键。对山东省典型潮土土壤样本测试高光谱并进行变换;采用遗传算法（GA）结合偏最小二乘法（PLS）,在筛选潮土碱解氮含量特征谱区的基础上,构建潮土碱解氮含量偏最小二乘（PLS）回归估测模型;优选最佳模型并与相关分析、逐步回归分析和单纯偏最小二乘回归分析的模型进行比较。结果表明：潮土碱解氮特征波段为449~469nm,988~1001nm,1065~1078nm,1716~1736nm,1912~1925nm,2213~2233nm,2262~2275nm;基于各输入光谱特征谱区构建的估测模型决定系数R²均较高,其中基于反射率一阶导数光谱筛选的特征谱区,构建的模型精度最高,数据点（147个）为原始全谱的7.17%,建模R²达到0.97,均方根误差RMSE为4.78mg/kg,验证R²为0.95,RMSE为5.49mg/kg,对潮土碱解氮含量具有较好的预测准确性;在光谱变换形式中,反射率的一阶导数表现最佳;方法比较显示采用遗传算法结合偏最小二乘（GA-PLS）获得较高预测精度的同时,可简化模型。说明遗传算法结合偏最小二乘法（GA-PLS）,可有效筛选土壤碱解氮的特征波段,减少模型参与变量,提高估测精度。