基于改进遗传算法优化的混凝土坝坝基扬压力统计模型研究
|
摘要
混凝土坝坝基扬压力是外界复合环境因素综合作用的结果,如何建立准确反应大坝实际渗流性态的统计模型对于掌握大坝安全性态具有十分重要的意义.本文基于外界环境量对混凝土坝坝基扬压力的影响过程分析,将前期库水位及降雨对坝基扬压力的非线性影响过程视为"正态分布"过程,并采用改进遗传算法对相应滞后天数和影响天数进行智能寻优,以提高坝基扬压力统计模型准确性.最后,以某混凝土坝为例,采用本文方法对其坝基扬压力实测进行拟合和预测,通过与传统统计模型对比,表明本文方法具有更高的拟合和预测精度,具有一定的科学和实践应用价值.
The variation of uplift pressure on concrete dam foundation is effected by the compound external environmental factors.How to establish statistical model which accurately reflects the real dam seepage state is quite important to grasp the safety state of a dam.Based on analyzing the influential process of ambient variables,the paper considers the nonlinear influential process as"Normal"process and adopts the improved genetic algorithm to optimize the lagged days and the influential days to raise the accuracy of the statistical model.Finally,taking a certain concrete dam for example,the recorded series of the uplift pressure is fitted and predicted with the proposed method.By comparing with the traditional model,it is verified that the proposed method possesses higher fitting and prediction accuracy and is of certain scientific and practical value.
The variation of uplift pressure on concrete dam foundation is effected by the compound external environmental factors.How to establish statistical model which accurately reflects the real dam seepage state is quite important to grasp the safety state of a dam.Based on analyzing the influential process of ambient variables,the paper considers the nonlinear influential process as"Normal"process and adopts the improved genetic algorithm to optimize the lagged days and the influential days to raise the accuracy of the statistical model.Finally,taking a certain concrete dam for example,the recorded series of the uplift pressure is fitted and predicted with the proposed method.By comparing with the traditional model,it is verified that the proposed method possesses higher fitting and prediction accuracy and is of certain scientific and practical value.
引文
[1]顾冲时,吴中如.大坝与坝基安全监控理论和方法及其应用[M].南京:河海大学出版社,2006.
[2]姜谙男,梁冰.基于PSO-SVM的大坝渗流监测时间序列非线性预报模型[J].水利学报,2006,37(3):331-335.
[3]田伟,魏光辉,高强.基于主成分分析与BP神经网络模型的大坝渗流监测资料分析[J].大坝与安全,2009(5):29-31.
[4]李波,顾冲时,李智录,等.基于偏最小二乘回归和最小二乘支持向量机的大坝渗流监控模型[J].水利学报,2008,39(12):1390-1394.
[5]梅一韬,仲云飞.基于熵权的大坝渗流性态模糊可拓评价模型[J].水电能源科学,2011,29(8):58-61.
[6]刘耀儒,杨强,覃振朝.基于统计模型的裂隙岩体渗流场的并行数值模拟[J].岩石力学与工程学报,2008(4)::736-742.
[7]洪云,郑东键.古田溪一级大坝14号坝段以左的渗压确定性模型[J].水电自动化与大坝监测,1996(3):30-31.
[8]顾冲时,李波,徐宝松,等.碾压混凝土坝渗漏安全监控模型建模方法研究[J].河海大学学报:自然科学版,2008,36(5):579-585.
[9]胡江,郑鹏翔.基于滞后效应和防洪调度的大坝渗流预测模型研究[J].三峡大学学报:自然科学版,2008,30(6):16-19.
[10]毕奇龙,刘星,徐波,等.因子分析法在绕坝渗流监测资料分析中的应用[J].水电能源科学,2010,28(5):62-65.
[11]杨建文,李志鹏,喻哲欣.基于改进遗传算法的泵站优化运行[J].中国水利,2015(8):33-35.
[12]张尚然,汤亚芳,林俐.基于改进遗传算法的无功优化方法的研究[J].中小企业管理与科技,2015(7):162-163.
[13]邵改革,卢小平,杜耀刚,等.基于优化模糊遗传算法的地理信息个性化推送方法[J].测绘工程,2015,24(7):20-23.
[14]夏俊,凌培亮,虞丽娟,等.基于量子遗传算法的无线传感网络路由优化[J].同济大学学报:自然科学版,2015,43(7):1097-1103.
[15]沙林秀,贺昱曜,陈延伟.一种变步长双链量子遗传算法[J].计算机工程与应用,2012(20):59-63.
[16]李延芳,王玉洁,赵花城.水口水电站大坝原型观测设计[J].水力发电,1994(6):23-25.
[2]姜谙男,梁冰.基于PSO-SVM的大坝渗流监测时间序列非线性预报模型[J].水利学报,2006,37(3):331-335.
[3]田伟,魏光辉,高强.基于主成分分析与BP神经网络模型的大坝渗流监测资料分析[J].大坝与安全,2009(5):29-31.
[4]李波,顾冲时,李智录,等.基于偏最小二乘回归和最小二乘支持向量机的大坝渗流监控模型[J].水利学报,2008,39(12):1390-1394.
[5]梅一韬,仲云飞.基于熵权的大坝渗流性态模糊可拓评价模型[J].水电能源科学,2011,29(8):58-61.
[6]刘耀儒,杨强,覃振朝.基于统计模型的裂隙岩体渗流场的并行数值模拟[J].岩石力学与工程学报,2008(4)::736-742.
[7]洪云,郑东键.古田溪一级大坝14号坝段以左的渗压确定性模型[J].水电自动化与大坝监测,1996(3):30-31.
[8]顾冲时,李波,徐宝松,等.碾压混凝土坝渗漏安全监控模型建模方法研究[J].河海大学学报:自然科学版,2008,36(5):579-585.
[9]胡江,郑鹏翔.基于滞后效应和防洪调度的大坝渗流预测模型研究[J].三峡大学学报:自然科学版,2008,30(6):16-19.
[10]毕奇龙,刘星,徐波,等.因子分析法在绕坝渗流监测资料分析中的应用[J].水电能源科学,2010,28(5):62-65.
[11]杨建文,李志鹏,喻哲欣.基于改进遗传算法的泵站优化运行[J].中国水利,2015(8):33-35.
[12]张尚然,汤亚芳,林俐.基于改进遗传算法的无功优化方法的研究[J].中小企业管理与科技,2015(7):162-163.
[13]邵改革,卢小平,杜耀刚,等.基于优化模糊遗传算法的地理信息个性化推送方法[J].测绘工程,2015,24(7):20-23.
[14]夏俊,凌培亮,虞丽娟,等.基于量子遗传算法的无线传感网络路由优化[J].同济大学学报:自然科学版,2015,43(7):1097-1103.
[15]沙林秀,贺昱曜,陈延伟.一种变步长双链量子遗传算法[J].计算机工程与应用,2012(20):59-63.
[16]李延芳,王玉洁,赵花城.水口水电站大坝原型观测设计[J].水力发电,1994(6):23-25.