GA-BP neural network algorithm-based nonlinear inversion for high density resistivity method
-
摘要: 高密度电法技术在煤矿地质灾害勘探中发挥着重要的作用。近年来,以BP(Backpropagation)神经网络为代表的一类非线性反演方法被广泛运用到高密度电法的反演中。针对BP神经网络方法在高密度电法反演中存在的易陷入局部极小、收敛缓慢、反演精度差等问题,将BP神经网络算法与遗传算法(Genetic Algorithm,简称GA算法)联合演算,实现高密度电法的二维非线性反演。通过典型地电模型对该方法进行验证,结果表明遗传算法能有效优化BP神经网络的权值和阈值,提高了算法的全局寻优性。Abstract: High density resistivity method has played an important role in geological disaster exploration in mining industry. In recent years some non-linear inversion methods represented by BP neural network have been widely used in the two-dimensional inversion of high density resistivity method. Aiming at the shortcomings of BP neural network such as being easy to fall into local minimum, slow convergency and poor inversion accuracy, the proposed method tried to combine the genetic algorithm and BP neural network method to achieve the two-dimensional inversion of high density resistivity method. The results of the classical electric model indicated that the genetic algorithm method can optimize the weights and bias of the BP neural network effectively and improve the performance of global optimization.
-
-
[1] 戴前伟,江沸菠. 基于混沌振荡PSO-BP算法的电阻率层析成像非线性反演[J]. 中国有色金属学报,2013,23(10):2897-2904. DAI Qianwei,JIANG Feibo. Nonlinear inversion for electrical resistivity tomography based on chaotic oscillation PSO-BP algorithm[J]. The Chinese Journal of Nonferrous Metals,2013,23(10):2897-2904.
[2] 穆阿华,周绍磊,刘青志,等. 利用遗传算法改进BP学习算法[J]. 计算机仿真,2005,22(2):150-151. MU Ahua,ZHOU Shaolei,LIU Qingzhi,et al. Using genetic algorithm to improve BP training algorithm[J]. Computer Simulation,2005,22(2):150-151.
[3] 刘春艳,凌建春,寇林元,等. GA-BP神经网络与BP神经网络性能比较[J]. 中国卫生统计,2013(2):173-176. LIU Chunyan,LING Jianchun,KOU Linyuan,et al. Performance comparison between GA-BP neural network and BP neural network[J]. Chinese Journal of Health Sta-tistics,2013(2):173-176.
[4] 温长吉,王生生,于合龙,等. 基于改进蜂群算法优化神经网络的玉米病害图像分割[J]. 农业工程学报,2013,29(13):142-149. WEN Changji,WANG Shengsheng,YU Helong,et al. Image segmentation method for maize diseases based on pulse co upled neural network with modified artificial bee algorithm[J]. Transactions of the Chinese Society of Agricultural Engineering,2013,29(13):142-149.
[5] 尹光志,李铭辉,李文璞,等. 基于改进BP神经网络的煤体瓦斯渗透率预测模型[J]. 煤炭学报,2013,38(7):1179-1184. YIN Guangzhi,LI Minghui,LI Wenpu,et al. Model of coal gas permeability prediction on improved BP neural network[J]. Journal of China Coal Society,2013,38(7):1179-1184.
[6] 潘国荣,谷川. BP算法改进及其在变形数据处理中的应用[J]. 同济大学学报(自然科学版),2008,36(1):118-121. PAN Guorong,GU Chuan. Back Propagetion algorithm impr-ovement and its application to deformation monitoring data processing[J]. Journal of Tongji Univer-sity(Natural Science),2008,36(1):118-121.
[7] 马永杰,云文霞. 遗传算法研究进展[J]. 计算机应用研究,2012,29(4):1201-1206. MA Yongjie,YUN Wenxia. Research progress of genetic algorithms[J]. Application Research of Computers,2012,29(4):1201-1206.
[8] 边霞,米良. 遗传算法理论及其应用研究进展[J]. 计算机应用研究,2010,27(7):2425-2429. BIAN Xia,MI Liang. Development on genetic algorithm theory and its applications[J]. Application Research of Computers,2010,27(7):2425-2429.
[9] 王小平,曹立明. 遗传算法-理论、应用与软件实现[M]. 西安:西安交通大学出版社,2002:1-5. [10] 李敏强. 遗传算法的基本理论与应用[M]. 北京:科学出版社,2002:1-3. [11] 师学明,王家映. 地球物理资料非线性反演方法讲座(四) -遗传算法[J]. 工程地球物理学报,2008,5(2):129-140. SHI Xueming,WANG Jiaying. Lecture on non-linear inverse methods in geophysics(4)-Genetic algorithm method[J]. Chinese Journal of Engineering Geophysics,2008,5(2):129-140.
[12] 江沸菠. 基于神经网络的混合非线性电阻率反演成像[D]. 长沙:中南大学,2014. [13] NEYAMADPOUR A,TAIB S,ABDULLAH Wan W A T. Using artificial neural networks to invert 2D DC resistivity im-aging data for high resistivity contrast regions:A MATLAB application[J]. Computers & Geosciences,2009,35(11):2268-2274.
[14] 张凌云,刘鸿福. ABP法在高密度电阻率法反演中的应用[J].地球物理学报,2011,54(1):227-233. ZHANG Lingyun,LIU Hongfu. The application of ABP method in high density resistivity method inversion[J]. Chinese Journal of Geophysics,2011,54(1):227-233.
计量
- 文章访问数: 67
- HTML全文浏览量: 6
- PDF下载量: 12