基于PSO优化BP神经网络的水质预测研究

为快速准确地预测河流水质,结合汾河监测数据,使用粒子群算法(PSO)优化BP神经网络模型(PSO-BP)进行水质预测.通过灰色关联度分析确定输入变量,利用PSO算法修正BP网络的初始权值、阈值,优化神经网络结构及算法全局收敛性.采用该模型对汾河主要污染物指标COD、BOD5、氨氮、挥发酚等进行预测和验证.结果表明,与传统的BP神经网络模型相比,PSO-BP模型使最大相对误差从15.43％减小到1.46％,其平均误差由4.00％减小到1.01％,预测均方根误差从5.956×10-3减小到1.605×10-4.因此,基于PSO-BP神经网络模型的预测更加精确,可用于水质预测.     

基于随机森林算法与农户视角的村落级农业面源污染风险分析

当前农业面源污染问题依然严峻,村落级尺度下各地块的污染风险状态精细识别有待进一步研究.本文针对三峡库区典型村落重庆市涪陵区南沱镇睦和村进行研究,结合无人机多光谱技术、农户行为调研、随机森林算法等进行地物识别及泛地块网格划分,通过测算总氮(total nitrogen,TN)、总磷(total phosphorus,TP...     

CO气体激光检测仪中基于FPGA的信号源设计

在激光检测仪的系统设计中,要求信号源四路波形同步异频发生,且有移相、倍频、波形叠加等要求。利用FPGA技术完成满足以上要求的直接数字频率合成器(DDS)的设计。在实现中选用Altera公司的FPGA芯片EP2C8Q208C8为主芯片进行设计,充分利用了该芯片的超大集成性和快速性。在改进的DDS算法结构的基础上,系统的复杂度降低,更趋于模块化,产生的波形频率更准确。     

煤层透气性系数的优化和简化计算方法

煤层透气性系数是用以考察保护层开采效果和设计瓦斯抽采系统的重要参数之一,钻孔径向流量法是求取该系数最常用的算法。利用理论推导和图形分析方法,探讨流量准数、时间准数、煤层透气性系数与参数A,B间的关系。在此基础上,针对科研工作和现场管理的实际情况,提出求解透气性系数的2种新算法及其算法流程,并对时间准数-参数A,B函数关系中的间断点问题进行讨论和分析,解决此区间内求解透气性系数相应计算公式等问题。理论和实例分析表明:算法1在不更改最终计算公式和结果的前提下,省去了繁琐的试算过程,既保证了可靠性也优化了计算流程;算法2以稍减精度的代价对计算流程进行进一步的简化,其结果仍有较高精度。     

基于BP神经网络算法的北京市地铁站应急疏散能力仿真评估模型

地铁站的应急疏散能力是地下铁道安全运营设计中非常重要的部分,对于维持社会稳定,保障人民安全具有重大意义.本文利用BP神经网络计算方法和MATLAB软件,选取可能影响地铁站疏散能力的相关因素作为评价指标,建立了应急疏散指标体系,并以北京北三环区域周边地铁站作为主要研究对象,对北京市地铁站应急疏散能力进行了仿真评估,提出了增加疏散通道、疏散楼梯、安全出口等数量,以提高地铁站应急疏散能力等一些可供参考的建议.     

公路隧道火灾发生位置与人群疏散通道仿真研究

利用EXODUS建立隧道的仿真场景,确定疏散人群仿真参数,并将出口工效OPS作为评价疏散效率的指标,分析不同火灾发生位置对疏散时间的影响。采用K-means算法分别对火灾发生位置和人群疏散通道位置进行聚类,并建立基于两者影响下的出口工效模型,通过对该模型求偏导,得出隧道火灾发生位置和人群疏散通道选择之间的关系。研究结果表明:火灾发生位置越接近人行横通道,疏散时间波动越大,人行横通道和隧道入口的疏散时间随人群疏散区域分界线变化,当火灾发生距人行横通道20~110 m时,变化明显,当距140~350 m时,变化平稳;出口工效OPS总体呈现不稳定的阶梯状变化趋势;火灾发生位置和人群疏散通道的最优位置呈线性递减关系。研究结果可应用于隧道应急指挥中心管理人员制定有效疏散路径和采取诱导分流人群措施。     

Restructured frame‐of‐reference training improves rating accuracy

The use of heuristic judgments is prevalent in organizations and negatively impacts accurate employee assessments. To minimize the negative impact of heuristic judgments (i.e., anchoring and adjustment), we aim to improve rating accuracy by restructuring frame‐of‐reference (FOR) training. We conducted five studies (N = 1,143) using different samples (three including participants with hiring experience), training environments (onsite and online), and rating contexts (evaluations of sales representatives, teachers, contract negotiation specialists, and retail store managers). Across the five studies, the average improvement in rating accuracy was at least twice as large for restructured FOR (vs. control) training as it was for typical FOR (vs. control) training; the difference in rating accuracy between restructured and typical FOR training was statistically significant. Furthermore, minimizing the anchoring effect rather than increasing opportunities for rating adjustments improved rating accuracy (Study 4). Finally, restructured FOR training achieved higher criterion validity (i.e., a higher strength of the association between ratings regarding a target and the target's objective performance) than did typical FOR training (Studies 3 and 5). We discuss implications for improving the effectiveness of diverse training programs and the accuracy of judgments in organizations.     

Hough变换在PCB在线检测系统的应用

在印刷电路板(PCB)视觉检测系统中,图象边缘提取的质量和对瑕疵图象边缘的快速而准确的识别是达到实时检测的关键因素,Hough变换是一种应用非常广泛的图像边缘检测技术,运用快速的Hough变换方法,来完成PCB检测的圆心定位和半径测定,并通过针时PCB在线检测的具体应用情况加以改进并应用,实验分析表明,该算法在PCB在线检测的可行性和优越性.图5,参5.     

实数编码遗传算法的评述

遗传算法是一种基于自然选择和遗传变异等生物进化机制的全局优化搜索算法．使用实数表示基因的实数编码遗传算法常用于求解连续函数优化等问题．本文阐述了实数编码遗传算法研究的有关工作进展，讨论了算法框架及特点，对实数编码遗传算法中常用的选择、交叉和变异算子进行了比较全面的形式化描述，并介绍了其未来研究方向．     

River water quality management model using genetic algorithm

Conventional mathematical programming methods, such as linear programming, non linear programming, dynamic programming and integer programming have been used to solve the cost optimization problem for regional wastewater treatment systems. In this study, a river water quality management model was developed through the integration of a genetic algorithm (GA). This model was applied to a river system contaminated by three determined discharge sources to achieve the water quality goals and wastewater treatment cost optimization in the river basin. The genetic algorithm solution, described the treatment plant efficiency, such that the cost of wastewater treatment for the entire river basin is minimized while the water quality constraints in each reach are satisfied. This study showed that genetic algorithm can be applied for river water quality modeling studies as an alternative to the present methods.