Use of genetic algorithm in new approach to modeling of flood routing

The hydrological models and simpli?ed methods of Saint-venant equations are used extensively in hydrological modeling, in particular for the simulation of the ?ood routing. These models require speci?c and extensive data that usually makes the study of ?ood propagation an arduous practice. We present in this work a new model, based on a transfer function, this function is a function of parametric probability density, having a physical meaning with respect to the propagation of a hydrological signal. The inversion of the model is carried out by an optimization technique called Genetic Algorithm. It consists of evolving a population of parameters based primarily on genetic recombination operators and natural selection to?nd the minimum of an objective function that measures the distance between observed and simulated data. The precision of the simulations of the proposed model is compared with the response of the Hayami model and the applicability of the model is tested on a real case, the N'Fis basin river, located in the High Atlas Occidental, which presents elements that appear favorable to the study of the propagation. The results obtained are very satisfactory and the simulation of the proposed model is very close to the response of the Hayami model.

RECONFIGURABLE PRODUCTION LINE MODELING AND SCHEDULING USING PETRI NETS AND GENETIC ALGORITHM

In response to the production capacity and functionality variations, a genetic algorithm （GA） embedded with deterministic timed Petri nets（DTPN） for reconfigurable production line（RPL） is proposed to solve its scheduling problem. The basic DTPN modules are presented to model the corresponding variable structures in RPL, and then the scheduling model of the whole RPL is constructed. And in the scheduling algorithm, firing sequences of the Petri nets model are used as chromosomes, thus the selection, crossover, and mutation operator do not deal with the elements in the problem space, but the elements of Petri nets model. Accordingly, all the algorithms for GA operations embedded with Petri nets model are proposed. Moreover, the new weighted single-objective optimization based on reconfiguration cost and E/T is used. The results of a DC motor RPL scheduling suggest that the presented DTPN-GA scheduling algorithm has a significant impact on RPL scheduling, and provide obvious improvements over the conventional scheduling method in practice that meets duedate, minimizes reconfiguration cost, and enhances cost effectivity.

A New Modeling Method Based on Genetic Neural Network for Numeral Eddy Current Sensor

In this paper,we present a method used to the numeral eddy current sensor modeling based on genetic neural network to settle its nonlinear problem.The principle and algorithms of genetic neural network are introduced.In this method, the nonlinear model parameters of the numeral eddy current sensor are optimized by genetic neural network (GNN) according to measurement data.So the method remains both the global searching ability of genetic algorithm and the good local searching ability of neural network.The nonlinear model has the advantages of strong robustness,on-line scaling and high precision.The maximum nonlinearity error can be reduced to 0.037% using GNN.However,the maximum nonlinearity error is 0.075% using least square method (LMS).

Modeling of Canonical Switching Cell Converter Using Genetic Algorithm

The working of Canonical switching cell(CSC)converter was studied and its equivalent circuit during ON and OFF states were obtained.State space model of CSC converter in ON and OFF states were developed using the Kirchhoff laws.The state space matrices were used to construct the transfer functions of ON&OFF states.The step response of the converter was simulated using MATLAB.The step response curve was obtained using different values of circuit components(L,C1,C2 and RL)and optimized.The characteristic parameters such as rise time,overshoot,settling time,steady state error and stability were determined using the step response curve.The response curve shows that there is no overshoot;the rise time and settling time are very low as expected for a converter and its stability is very high but the amplitude is very.The circuit was tuned to attain the expected amplitude using PID controller with the help of Genetic algorithm.The excellent results of circuits’characteristic parameters are very useful guideline for constructing such CSC converters for DC-DC conversions.The circuit characteristic parameters are useful in constructing such CSC converters for DCDC conversions in driving solar energy using solar panel.

Modeling and Adaptive Self-Tuning MVC Control of PAM Manipulator Using Online Observer Optimized with Modified Genetic Algorithm

In this paper, the application of modified genetic algorithms (MGA) in the optimization of the ARX Model-based observer of the Pneumatic Artificial Muscle (PAM) manipulator is investigated. The new MGA algorithm is proposed from the genetic algorithm with important additional strategies, and consequently yields a faster convergence and a more accurate search. Firstly, MGA-based identification method is used to identify the parameters of the nonlinear PAM manipulator described by an ARX model in the presence of white noise and this result will be validated by MGA and compared with the simple genetic algorithm (GA) and LMS (Least mean-squares) method. Secondly, the intrinsic features of the hysteresis as well as other nonlinear disturbances existing intuitively in the PAM system are estimated online by a Modified Recursive Least Square (MRLS) method in identification experiment. Finally, a highly efficient self-tuning control algorithm Minimum Variance Control (MVC) is taken for tracking the joint angle position trajectory of this PAM manipulator. Experiment results are included to demonstrate the excellent performance of the MGA algorithm in the NARX model-based MVC control system of the PAM system. These results can be applied to model, identify and control other highly nonlinear systems as well.

基于准稳态过程的混流泵启动性能优化

在ANSYS Workbench的基础上结合OptiSLang优化设计软件,对混流泵启动过程进行优化.利用准稳态理论和全流域模型计算数据,建立了混流泵准稳态启动过程的水力性能响应面替代模型.通过NSGA遗传算法优化内部流动状态,降低启动过程中的涡致能量损失,从而提升混流泵的水力性能.将原始混流泵与优化后混流泵的外特性、叶片压力分布进行对比,并分析混流泵内部涡结构,验证了所提出的准稳态启动过程水力性能优化方案的可行性.研究结果表明:叶片进口安放角α1,α4,α5,叶片出口安放角β1,β2,β3,β4,β5,叶片包角φ和叶片厚度系数θ对混流泵加权平均扬程和加权平均效率的影响程度较大;对比不同启动时刻下混流泵的叶片压力分布,发现该方法能够有效提升混流泵启动中后期的瞬态扬程和启动中期的水力效率.

陕西省月用水量预测方法研究

基于国家水资源管理信息系统的月用水量数据分析,选用ARIMA模型、BP神经网络模型以及经过遗传算法(GA)优化的BP神经网络模型(GA-BP神经网络模型)进行月用水量模拟。在构建BP神经网络模型过程中,通过多源社会经济数据的整合与分析,采用平均影响值算法(MIV)和皮尔逊相关系数联合方法筛选月用水量的关键影响因子。研究结果表明,三种模型在陕西省月用水量预测中均表现出较高的精度,其中GA-BP神经网络模型的预测精度最高。为进一步验证影响因子对模拟结果的影响,采用不同方法筛选影响因子作为GA-BP神经网络模型的输入,模拟结果表明,MIV和皮尔逊相关系数联合方法提高了影响因子的选取精度,能够有效提升GA-BP神经网络模型的模拟性能。

微通道内纳米流体传热流动特性

为提高微通道散热器的传热效率,需要对微通道进行结构优化设计。以热阻Rt和泵功Pp为目标函数,在Re=100的条件下,采用多目标遗传算法对文丘里管微通道的结构参数,如通道深度、收缩角度、喉颈宽度和扩散角度进行优化,通过遗传迭代计算得到Pareto优化解集,利用k-means聚类法对优化解集进行比较分析,通过强化传热因子η对各聚类点综合性能进行评价,得到最优的微通道结构。采用数值模拟方法,研究优化后的微通道结构的流动与传热特性。结果表明:当去离子水中加入纳米颗粒后微通道内的压降具有小幅度上升,但其流动阻力在相同Reynolds数的条件下并没有发生较大的变化。在文丘里管微通道喉部位置会产生喉部效应,强化纳米颗粒与微通道中流动工质的融合。熵产分析表明,传热熵随着Reynolds数的增大而减小,摩擦熵随着Reynolds数的增大而增大,不过总熵值中主要是传热熵占据主导地位。纳米流体随着体积分数的增加不可逆损失均小于去离子水。

基于基因算法的数据中心冷源系统能耗建模与优化

针对数据中心冷源系统,采用数据机理双驱动的方法对冷源系统中的冷水机组、水泵及冷却塔能耗进行建模,提出了基于基因算法的数据中心冷源系统能耗优化方法,并以重庆市某数据中心制冷系统为研究案例进行了分析。计算结果显示,通过使用该建模优化方法,相比优化前凭工人经验调节的运行方法,冷源系统的能耗平均减少约8.5%。

BP神经网络在离心压缩机叶轮优化中的应用

为了提高离心式压缩机叶轮设计效率并降低计算资源消耗,针对遗传算法优化中计算量大、效率低的问题,提出基于改进粒子群优化算法(IPSO)优化BP神经网络的方法。通过少量计算流体动力学(CFD)仿真样本,训练BP神经网络建立效率与叶轮参数的映射关系,结合IPSO优化其参数,同时利用遗传算法(GA)确定叶轮的最佳性能参数。研究表明,改进的IPSO算法通过增强粒子群的动态适应性和全局搜索能力,提高了BP神经网络的预测精度和优化效率。优化后的叶轮等熵效率提高1.34%,多变效率提高1.04%,流量增加10.4%。该方法显著提升了离心式压缩机叶轮的设计效率和性能,为复杂流体机械的优化设计提供了新思路。

基于KCC模型-ALE算法的充填体动力响应数值模拟研究

胶结充填体作为人工矿柱常受到邻近矿房开采的爆破扰动,严重影响其稳定性。基于KCC(Karagozian and Case Concrete)本构模型和ALE(Arbitrary Lagrangian-Eulerian)流固耦合算法,采用LS-DYNA软件开展胶结充填体受邻近矿房爆破扰动的数值模拟研究,考虑不同边孔间距(0.6、1.2、1.8、2.4 m)及养护龄期(14、21、28 d)影响,揭示充填体中爆破波传播规律,探究爆破扰动对胶结充填体的动力响应及其失稳破坏情况。结果表明:邻近采场爆破对胶结充填体的破坏损伤主要集中在边界处,为提高矿石回收率并保证充填体安全,建议采用边孔间距1.2 m方案;矿山爆破开采中充填体至少需要养护21 d;低龄期充填体矿柱顶底部区域是整个矿柱中的薄弱部分,易发生失稳,应予以重点监测。

GA-BP模型在HSS模型参数取值中的应用

小应变硬化土(HSS)模型可以有效反映土的压缩硬化特性和小应变特性,非常适合黄土基坑的数值模拟计算。但是,HSS模型包含了11个硬化土(HS)模型参数和2个小应变参数,而这2个小应变参数往往需要采用试验方法确定,获取过程复杂。为了探讨小应变参数的预测方法,采用经过遗传算法优化的BP神经网络模型,即GA-BP神经网络模型,首先根据预设的小应变参数水平经过数值模拟计算得到49组位移数据,然后将得到的数据用于GA-BP神经网络的训练,待GA-BP神经网络的预测误差达到要求之后,再使用实际的位移数据反演得到小应变参数,最后基于预测得到的小应变参数进行数值模拟。结果显示,GA-BP神经网络模型预测的小应变参数在基坑围护结构最大水平位移和地表最大沉降计算方面表现良好,可以应用于实际工程。

Evaluation of volcanic reservoirs with the "QAPM mineral model" using a genetic algorithm

Gas-bearing volcanic reservoirs have been found in the deep Songliao Basin, China. Choosing proper interpretation parameters for log evaluation is difficult due to complicated mineral compositions and variable mineral contents. Based on the QAPF classification scheme given by IUGS, we propose a method to determine the mineral contents of volcanic rocks using log data and a genetic algorithm. According to the QAPF scheme, minerals in volcanic rocks are divided into five groups： Q（quartz）, A （Alkaline feldspar）, P （plagioclase）, M （mafic） and F （feldspathoid）. We propose a model called QAPM including porosity for the volumetric analysis of reservoirs. The log response equations for density, apparent neutron porosity, transit time, gamma ray and volume photoelectrical cross section index were first established with the mineral parameters obtained from the Schlumberger handbook of log mineral parameters. Then the volumes of the four minerals in the matrix were calculated using the genetic algorithm （GA）. The calculated porosity, based on the interpretation parameters, can be compared with core porosity, and the rock names given in the paper based on QAPF classification according to the four mineral contents are compatible with those from the chemical analysis of the core samples.

SOI MOSFET Model Parameter Extraction via a Compound Genetic Algorithm

We improve the genetic algorithm by combining it with a simulated annealing algorithm. The improved algorithm is used to extract model parameters of SOI MOSFETs, which are fabricated with standard 1.2μm CMOS/SOI technology developed by the Institute of Microelectronics of the Chinese Academy of Sciences. The simulation results using this model are in excellent agreement with experimental results. The precision is improved noticeably compared to commercial software. This method requires neither a deeper understanding of SOl MOSFETs model nor more complex computations than conventional algorithms used by commercial software. Comprehensive verification shows that this model is applicable to a very large range of device sizes.

Dynamic finite element model updating using meta-model and genetic algorithm

Current dynamic finite element model updating methods are not efficient or restricted to the problem of local optima. To circumvent these, a novel updating method which integrates the meta-model and the genetic algorithm is proposed. Experimental design technique is used to determine the best sampling points for the estimation of polynomial coefficients given the order and the number of independent variables. Finite element analyses are performed to generate the sampling data. Regression analysis is then used to estimate the response surface model to approximate the functional relationship between response features and design parameters on the entire design space. In the fitness evaluation of the genetic algorithm, the response surface model is used to substitute the finite element model to output features with given design parameters for the computation of fitness for the individual. Finally, the global optima that corresponds to the updated design parameter is acquired after several generations of evolution. In the application example, finite element analysis and modal testing are performed on a real chassis model. The finite element model is updated using the proposed method. After updating, root-mean-square error of modal frequencies is smaller than 2%. Furthermore, prediction ability of the updated model is validated using the testing results of the modified structure. The root-mean-square error of the prediction errors is smaller than 2%.

Combining the genetic algorithms with artificial neural networks for optimization of board allocating

This paper introduced the Genetic Algorithms (GAs) and Artificial Neural Networks (ANNs), which have been widely used in optimization of allocating. The combination way of the two optimizing algorithms was used in board allocating of furniture production. In the experiment, the rectangular flake board of 3650 mm 1850 mm was used as raw material to allocate 100 sets of Table Bucked. The utilizing rate of the board reached 94.14 % and the calculating time was only 35 s. The experiment result proofed that the method by using the GA for optimizing the weights of the ANN can raise the utilizing rate of the board and can shorten the time of the design. At the same time, this method can simultaneously searched in many directions, thus greatly in-creasing the probability of finding a global optimum.

Small-Signal Equivalent Circuit Modeling of a Photodetector Chip

A small-signal equivalent circuit model and the ted. The equivalent lumped circuit, which takes the main extraction techniques for photodetector chips are presen- factors that limit a photodetector＇s RF performance into consideration,is first determined based on the device＇s physical structure. The photodetector＇s S parameters are then on-wafer measured, and the measured raw data are processed with further calibration. A genetic algorithm is used to fit the measured data, thereby allowing us to calculate each parameter value of the model. Experimental resuits show that the modeled parameters are well matched to the measurements in a frequency range from 130MHz to 20GHz, and the proposed method is proved feasible. This model can give an exact description of the photodetector chip＇s high frequency performance,which enables an effective circuit-level prediction for photodetector and optoelectronic integrated circuits.

Study on the tradeoff between interpretability and precision in fuzzy modeling

An approach to identifying fuzzy models considering both interpretability and precision was proposed. Firstly, interpretability issues about fuzzy models were analyzed. Then, a heuristic strategy was used to select input variables by increasing the number of input variables, and the Gustafson-Kessel fuzzy clustering algorithm, combined with the least square method, was used to identify the fuzzy model. Subsequently, an interpretability measure was described by the product of the number of input variables and the number of rules, while precision was weighted by root mean square error, and the selection objective function concerning interpretability and precision was defined. Given the maximum and minimum number of input variables and rules, a set of fuzzy models was constructed. Finally, the optimal fuzzy model was selected by the objective function, and was optimized by a genetic algorithm to achieve a good tradeoff between interpretability and precision. The performance of the proposed method was illustrated by the well-known Box-Jenkins gas furnace benchmark; the results demonstrate its validity.

Reconstruction of the linac photon spectrum based on prior knowledge and the genetic algorithm

In order to derive the linac photon spectrum accurately both the prior constrained model and the genetic algorithm GA are employed using the measured percentage depth dose PDD data and the Monte Carlo simulated monoenergetic PDDs where two steps are involved.First the spectrum is modeled as a prior analytical function with two parameters αand Ep optimized with the GA.Secondly the linac photon spectrum is modeled as a discretization constrained model optimized with the GA. The solved analytical function in the first step is used to generate initial solutions for the GA’s first run in this step.The method is applied to the Varian iX linear accelerator to derive the energy spectra of its 6 and 15 MV photon beams.The experimental results show that both the reconstructed spectrums and the derived PDDs with the proposed method are in good agreement with those calculated using the Monte Carlo simulation.

APPROXIMATION TECHNIQUES FOR APPLICATION OF GENETIC ALGORITHMS TO STRUCTURAL OPTIMIZATION

Although the genetic algorithm (GA) has very powerful robustness and fitness, it needs a large size of population and a large number of iterations to reach the optimum result. Especially when GA is used in complex structural optimization problems, if the structural reanalysis technique is not adopted, the more the number of finite element analysis (FEA) is, the more the consuming time is. In the conventional structural optimization the number of FEA can be reduced by the structural reanalysis technique based on the approximation techniques and sensitivity analysis. With these techniques, this paper provides a new approximation model-segment approximation model, adopted for the GA application. This segment approximation model can decrease the number of FEA and increase the convergence rate of GA. So it can apparently decrease the computation time of GA. Two examples demonstrate the availability of the new segment approximation model.