基于粒子群算法的神经网络在水资源评价中的应用

为了改善传统的人工神经网络,在训练过程中容易陷入局部最小导致应用于水资源评价时存在对训练样本的拟合精度不高的缺点,采用粒子群算法优化人工神经网络的权值和阈值,然后将其应用于中国12个地区的水资源可持续利用系统评价实例中,并和传统的人工神经网络进行了对照。结果表明,基于粒子群算法的人工神经网络和传统的人工神经网络相比,能较好的提高对训练样本的拟合精度,表明基于粒子群算法的人工神经网络,用于水资源可持续利用系统评价是可行的。     

基于粒子群算法的神经网络在水资源评价中的应用

为了改善传统的人工神经网络,在训练过程中容易陷入局部最小导致应用于水资源评价时存在对训练样本的拟合精度不高的缺点,采用粒子群算法优化人工神经网络的权值和阈值,然后将其应用于中国12个地区的水资源可持续利用系统评价实例中,并和传统的人工神经网络进行了对照。结果表明,基于粒子群算法的人工神经网络和传统的人工神经网络相比,能较好的提高对训练样本的拟合精度,表明基于粒子群算法的人工神经网络,用于水资源可持续利用系统评价是可行的。     

基于粒子群算法的湖泊营养状态指数公式

提出了适用于16个富营养化指标的湖泊营养状态指数公式,并对指数公式中的参数采用粒子群算法进行优化,得到对16项指标都适用的营养状态指数公式.实例分析表明,优化后的湖泊营养状态指数公式计算简单、实用,具有客观性、可比性和通用性.因而,能够对湖泊富营养化进行较好的评价.     

利用混沌粒子群算法确定河流水质模型参数

将混沌寻优思想引入到粒子群优化算法中,提出了混沌粒子群算法,这种方法利用混沌运动的随机性、遍历性和规律性等特性对当前粒子群体中的粒子进行混沌寻优.通过这种处理使得粒子群体的进化速度加快,从而改善了粒子群优化算法摆脱局部极值点的能力,提高了算法的收敛速度和精度.并将混沌粒子群算法应用于求解分析瞬时投放示踪剂情况下的一维河流水团示踪试验数据以及确定河流水质参数的函数优化问题,结果表明,混沌粒子群算法的收敛性能明显优于粒子群优化算法.     

改进的粒子群优化算法

针对粒子群优化算法易陷入局部最优并难以跳出的缺陷,提出了一种改进的算法。算法在采用自适应惯性权重基础上,引入粒子群局部收敛判断机制,对陷入局部最优的粒子采取新的进化模型以增加群体的多样性,进而降低了群体早熟收敛。通过新进化模型的粒子群优化算法对4个基准函数反复试验,仿真结果表明改进的粒子群算法具有更好的收敛精度并且可以有效避免早熟收敛问题,寻找到全局最优解。     

粒子群算法在克里金三维地质建模中的应用

变差函数是克里金法中反映区域化变量空间变化特征的有效数学模型,由其确定的拟合模型参数直接影响插值精度.采用基于线性递减权值的粒子群算法估计变差函数参数,实验对比发现:所提方法比传统的最小二乘法和手工拟合法更精确,采用某油田小层数据进行地质建模,插值得到的数据点与实际离散数据拟合更好,能够体现该地区地质特征,取得较好应用...     

粒子群优化的半监督入侵检测算法

为解决无监督入侵检测算法检测率低,有监督的入侵检测不能有效的检测未知攻击的问题,提出了一种粒子群优化的半监督入侵检测算法,算法对少量的约束信息进行基于密度的扩展获得潜在约束得到聚类模型,以此指导未标记数据聚类,对仍没有确定类别的未标示数据使用粒子群优化的K均值算法进行聚类实现对异常的检测。改进的算法检测率达到83.7%,误报率减少至3.13%,总体效果优于无监督和有监督学习的入侵检测算法。     

基于粒子群算法的GM(1,1)在经济发展预测中的应用

采用粒子群算法对GM(1,1)模型中的参数进行优化求解,然后将得到的模型应用于四川省经济发展预测,并和传统采用最小二乘法求解参数的GM(1,1)的预测结果进行了对照.结果表明,采用粒子群算法优化参数的GM(1,1)模型预测效果明显优于传统采用最小二乘法求解参数的GM(1,1)的模型预测效果.     

Statistical analysis of fracture properties based on particle swarm optimization and Pearson correlation coefficient method

Prediction of reservoir fracture is the key to explore fracture-type reservoir. When a shear-wave propagates in anisotropic media containing fracture,it splits into two polarized shear waves: fast shear wave and slow shear wave. The polarization and time delay of the fast and slow shear wave can be used to predict the azimuth and density of fracture. The current identification method of fracture azimuth and fracture density is cross-correlation method. It is assumed that fast and slow shear waves were symmetrical wavelets after completely separating,and use the most similar characteristics of wavelets to identify fracture azimuth and density,but in the experiment the identification is poor in accuracy. Pearson correlation coefficient method is one of the methods for separating the fast wave and slow wave. This method is faster in calculating speed and better in noise immunity and resolution compared with the traditional cross-correlation method. Pearson correlation coefficient method is a non-linear problem,particle swarm optimization( PSO) is a good nonlinear global optimization method which converges fast and is easy to implement. In this study,PSO is combined with the Pearson correlation coefficient method to achieve identifying fracture property and improve the computational efficiency.     

水质评价的污染损害指数公式

在选择了各污染物的“参照值”情况下,用污染物的实测学相对值代替水质污染损害率公式中的浓度监测值,公式中的参数可认为与污染物的特性无关。采用遗传算法优化中的参数,得到了多种水质污染物都能适用的水质污染损害率和污染损害指数公式。提出用对比加权对污染损害分指数进行赋权的新方法,计算水质综合污染损害指数。该方法用于水质综合评价的物理意义明确、计算简单、使用方便,具有普适性和可比性。     

适用于多指标的水环境质量评价的普适指数公式

在适当设定水质指标的“基准”值情况下,用相对于“基准”值的指标相对值替换对数形式的指数公式中的水质指标值,采用遗传算法优化不同指标的指数公式中的参数,得到对多项指标通用的水质普适指数公式;并用折衷型激励功效函数将单项指标的指数加权为综合指数。公式的正确性已被用于多处水质评价检验所证实。该公式形式简单,易于计算,使用方便;具有客观性、可比性和通用性。     

Underground water quality model inversion of genetic algorithm

The underground water qualitymodel with non-linear inversion problem is illposed, and boils down to solving the minimum of nonlinear function. Genetic algorithms are adopted in a number of individuals of groups by iterative search to find the optimal solution of the problem, the encoding strings as its operational objective, and achieving the iterative calculations by the genetic operators. It is an effective method of inverse problems of groundwater, with incomparable advantages and practical significances.     

Underground water quality model inversion of genetic algorithm

The underground water quality model with non-linear inversion problem is ill-posed, and boils down to solving the minimum of nonlinear function. Genetic algorithms are adopted in a number of individuals of groups by iterative search to find the optimal solution of the problem, the encoding strings as its operational objective, and achieving the iterative calculations by the genetic operators. It is an effective method of inverse problems of groundwater, with incomparable advantages and practical significances.     

Application of multivariate model based on three simulated sensors for water quality variables estimation in Shitoukoumen Reservoir,Jilin Province,China

This study applied a multivariate model based on three simulated sensors to estimating water quality variables in Shitoukoumen Reservoir, Changchun City, Jilin Province, China, including concentration of total suspended matter, concentration of chlorophyll-a and non-pigment matter absorption. Two field campaigns for spectra measurements with a total of 40 samples were carried out on June 13 and September 23, 2008. The in-situ spectra were recalculated to the spectral bands and sensitivities of the instruments applied in this paper, i.e. Landsat TM, Alos and P6, by using the average method. And the recalculated spectra were used for estimating water quality variables by the single model and multivariate model. The results show that the multivariate model is superior to the single model as the multivariate model takes the combined effects of water components into consideration and can estimate water quality variables simultaneously. According to R ² and RMSE, Alos is superior to other sensors for water quality variables estimation although the precision of non-pigment matter absorption inversion performed the second.     

Assessment of point and nonpoint sources pollution in Songhua River Basin,Northeast China by using revised water quality model

Individual participation of pollutants in the pollution load should be estimated even if roughly for the appropriate environmental management of a river basin. It is difficult to identify the sources and to quantify the load, especially in modeling nonpoint source. In this study a revised model was established by integrating point and nonpoint sources into one-dimensional Streeter-Phelps (S-P) model on the basis of real-time hydrologic data and surface water quality monitoring data in the Jilin Reach of the Songhua River Basin. Chemical oxygen demand (COD) and ammonia nitrogen (NH_3-N) loads were estimated. Results showed that COD loads of point source and nonpoint source were 134 958 t/yr and 86 209 t/yr, accounting for 61.02% and 38.98% of total loads, respectively. NH_3-N loads of point source and nonpoint source were 16 739 t/yr and 14 272 t/yr, accounting for 53.98% and 46.02%, respectively. Point source pollution was stronger than nonpoint source pollution in the study area at present. The water quality of upstream was better than that of downstream of the rivers and cities. It is indispensable to treat industrial wastewater and municipal sewage out of point sources, to adopt the best management practices to control diffuse pollutants from agricultural land and urban surface runoff in improving water quality of the Songhua River Basin. The revised S-P model can be successfully used to identify pollution source and quantify point source and nonpoint source loads by calibrating and validating.     

基于GA优化的灾情评估普适公式

为建立适用于多种灾情的评估指数公式,运用遗传算法,优化公式中的参数.该公式对于分级指标项数没有限制,使用范围广泛.应用该公式对1990年我国部分省市发生的9次自然灾害进行等级划分,并与其它方法的评估结果进行比较,验证了公式的正确性.     

禁忌搜索优化的投影寻踪及其在水资源评价中的应用

在深入分析区域水资源可持续利用评价系统的基础上，建立了区域水资源可持续利用能力评价指标体系，并将基于禁忌搜索优化的投影寻踪技术（TSPP）应用于水资源可持续利用评价。通过对淮河片区水资源可持续利用能力的初步研究，表明了此模型能完整而系统地反映区域水资源可持续利用能力，并且直观、简便，在各种实际系统评价中具有应用价值。