人工神经网络模型在地学研究中的应用进展

doi:10.11867/j.issn.1001-8166.2003.01.0068

近年来，随着人工神经网络（ANNs）自身技术的不断完善，应用ANNs模型成功解决各类地学问题的案例大量出现。通过对其发展历程进行分析发现，20世纪80年代末国际地学分析中已开始融入ANNs技术，国内则滞后 1～2年。在地学分析中使用的各类人工神经网络类型中，BP模型应用最广，占到85％以上。在10余年的应用过程中，虽然地学的各个分支学科都移植了一种或数种ANNs模型作为其分析工具，但水文、地质、大气、遥感等领域应用较为广泛。传统地学定量分析中的单变量或多变量预测成为人工神经网络地学模型的主要应用客体。同时，诸如模式识别和过程模拟等也是ANNs模型求解的对象。目前，随着建模经验和知识的积累，地学ANNs模型的发展呈现出多种技术综合集成的态势，遗传算法、小波转换、模拟退火算法以及模糊逻辑等方法与ANNs模型融合，成为解决地学分析中非线性问题的利器。

关键词: 人工神经网络（ANNs）, 地学模拟

    Neural networks are increasingly popular in geosciences due to big progress in neural network modelling techniques and imperative demands in geosciences. Without assuming parametric relationship, artificial neural networks have the ability to learn patterns of relationships in data from being shown a given set of inputs including combinations of descriptive and quantitative data, generalize or abstract results from imperfect data, and be insensitive to minor variations in input such as noise in the data, missing data, or a few incorrect values. When the neural network is trained appropriately, it generalizes therelationship so that it can be applied to other new data sets. The theoretical basis of this technique is the universal approximation theorem, which states that a multi-layer feed-forward neural network, such as the radial-basis function or perceptron neural network, is capable of performing any nonlinear input-output mapping.
    In this paper, the applications of ANNs in various branches of geosciences have been examined here. It is found that ANNs are robust tools and alternative approaches for modeling many of the nonlinear processes in geosciences. After appropriate training, they are able to generate satisfactory results for solving many problems such as prediction, classification, pattern recognition and optimization. By counting the 349 ANNs-based papers in geosciences during the 1997—2000, three predominant subjects are hydrology, geology and atmospheric science, andprediction is major modelling purpose.
    After reviewing the state of the art of geoscience's ANN modelling, author thinks that integrated ANNs modelling framework must be developed in order to dealwith more complex nonlinear processes. Such integrated framework consists of non-parametric techniques such as neural network, fuzzy logic, genetic algorithm,simulated annealing algorithm, fractal theory, Cellular automata and wavelet transform etc. It is helpful for selecting appropriate input and output neurons and designing more efficient networks to profoundly understand the linear or nonlinear processes being modeled in geosciences. Important aspects such as physical interpretation of ANN architecture, optimal training data set, adaptive learning, and generalization must be explored further. The merits and limitations of ANN applications have been discussed, and potential research avenues have been explored briefly.

Key words: Artificial neural network(ANNs), Geosciences modelling.

中图分类号:

TP183

[1] Shi Chenggang,Liu Xila. Application of neural network to earthquake engineering[J]. Earthquake Engineering Vibration,1991,15(2):39-47.[石成钢,刘西拉. 人工神经网络在震害预测中的应用[J].地震工程与工程振动,1991,15(2):39-47．]

[2] Mccann D W. A neural network short-term load forecast of significant thunderstorms[J]. Weather Forecasting,1992,7:525-535

[3] Wu Xiao. Application of artificial neural network to rainstorm forecast[A]. In:Intelligent Control and Intelligent Automation[C].Beijing:Science Press,1993.[吴晓. 人工神经网络在暴雨预测中的应用[A]. 智能控制与智能自动化[C]. 北京：科学出版社,1993．]

[4] Macher G L,Fuller J D. Back-propagation in hydrological time series of forecasting[A]. In:Symposium of International Conference on Stochastic and Statistical Method in Hydrology and Environmental Engineering[C]. Ontario,Canada:1993.229-242．

[5] GovindaraJu R S. Artificial neural networks in hydrology I:Preliminary concepts[J]. Journal of Hydrologic Engineering,2000,5(2):115-123．

[6] GovindaraJu R S. Artificial neural networks in hydrology II:Hydrologic applications[J]. Journal of Hydrologic Engineering,2000,5(2):124-137．

[7] Gardner M W,Dorling S R. Artificial neural networks (multilayer perceptron)—A review of applications in the atmospheric sciences[J]. Atmospheric Environment,1998,32(14/15): 2 627-2 636．

[8] Dawson C W,Wilby R L. A comparison of artificial neural networks used for river flow forecasting[J]. Hydrology and Earth System Sciences,1999,3(4): 529-540．

[9] Snell S S,Gopal S,Kaufman R K. Spatial interpolation of surface air temperatures using artificial neural networks: Evaluating their use for downscaling GCMs[J]. Journal of Climate,2000,13(5):886-895．

[10] Brown W M,Gedeon T D,Groves D I,et al. A new method fro mineral prospectivity mapping[J]. Australian Journal of Earth Sciences,2000,47(4):757-770．

[11] Minasny B,McBratney A B. Evaluation and development of hydraulic conductivity pedotransfer function for Australian soil[J]. Australian Journal of Soil Research,2000,38(4):905-926．

[12] Spellman G. An application of artificial neural networks to the prediction of surface ozone concentrations in the United Kingdom[J]. Applied-Geography,1999,19(2):123-136．

[13] Zhang B,GovindaraJu R S. Prediction of watershed runoff using Bayesian concepts and modular neural networks[J]. Water Resources Research,2000,36(3): 753-762．

[14] Imrie C E,Durucan S,Korre A. River flow prediction using artificial networks:Generalization beyond the calibration range[J]. Journal of Hydrology,2000,233(1/4):138-153．

[15] Maier H R,Dandy G C. Empirical comparison of various methods for training feed-forward neural networks for salinity forecasting[J]. Water Resources Research,1999,35(8):2 591-2 596．

[16] Anmala J,Zhang B,GovindaraJu R S. Comparison of ANNs and empirical approaches for predicting watershed runoff[J]. Journal of Water Resources Planning and Management,2000,126(3): 156-166．

[17] Abuelgasim A A,Ross W D,Gopal S,et al. Change detection using adaptive fuzzy neural networks:Environmental damage assessment after the Gulf War[J]. Remote Sensing of Environment,1999,70(2):208-223．

[18] Zhu Chengshan,Pan YingJie. A fuzzy neural network method for predicting the runoff into Ouyanghai reservoir[J]. International Journal of Hydroelectric Energy,2000,18(3):16-18.[朱承山,潘英杰.欧阳海水库径流量预报的模糊神经网络方法[J].水电能源科学, 2000,18(3):16-18．]

[19] Cao Jie,Xie Yingqi. One kind of associative memory by neural network[J]. Journal of Tropical Meteorology,1997,13(1):82-87.[曹杰,谢应齐. 一种基于混沌理论的联想记忆神经网络模型[J].热带气象学报,1997,13(1):82-87．]

[20] Nunnari G,Nucifora A F M,Randieri C. The application of neural techniques to the modelling of time-series of atmospheric pollution data[J]. Ecological Modelling,1998,111(2/3):187-205．

[21] Jin Long,Luo Ying,Miao Qilong,et al. Forecast model of farmland soil moisture by artificial neural networks[J]. Acta Fedologica Sinica,1998,35(1):25-32.[金龙,罗莹,缪启龙,等. 农田土壤湿度的人工神经网络预报模型研究[J].土壤学报,1998,35(1):25-32．]

[22] Hong Wei,Wu Chengzhen. Study on soil loss prediction by artificial network in southeast FuJian[J]. Journal of Soil Erosion and Soil and Water Conservation,1997,3(3):52-57.[洪伟,吴承祯. 闽东南土壤流失人工神经网络预报研究[J]. 土壤侵蚀与水土保持学报,1997,3(3):52-57.]

[23] Li Xianbin,Ding Jing. Combining neural network forecasts on wavelet-transformed series[J]. Journal of Hydraulic Engineering,1999,2:1-4. [李贤彬,丁晶. 基于子波变换序列的人工神经网络组合预测[J]. 水利学报,1999,2:1-4.]

[24] Berberoglu S,Lloyd C D,Atkinson P M,et al. The integration of spectral and textural information using neural networks for-land cover mapping in the Mediterranean[J]. Computers and Geosciences,2000,26(4):385-396

[25] Tzeng Y C,Chen K S. A fuzzy neural network to SAR image classification[J]. IEEE Transactions on Geoscience and Remote Sensing,1998,36(1):301-307．

[26] Ito Y,Omatu S. Polarimetric SAR data classification using competitive neural networks[J]. International Journal of Remote Sensing,1998,19(14):2 665-2 684．

[27] Bruzzone L,Prieto D F. An incremental-learning neural network for the classification of remote-sensing images[J].Pattern Recognition Letters,1999,20(11/13 speciss):1 241-1 248．

[28] Liang Yitong,Hu Jianglin. Application of neural network technique to classification of NOAA satellite image[J]. Journal of Wuhan Technical University of Surveying and Mapping,2000,25(2):149-152．[梁益同,胡江林. NOAA卫星影像神经网络分类方法探讨[J]. 武汉测绘科技大学学报,2000,25(2):149-152．]

[29] Xiong Zhen,Zheng Lanfen,Tong Qingxi. Hierarchical neural network classification algorithm [J]. Acta Geodaetica et Cartographica Sinica,2000,29(3):229-234．[熊祯,郑兰芬,童庆禧.分层神经网络算法[J].测绘学报,2000,29(3):229-234．]

[30] Morlini I. Radial basis function networks with partially classified data[J]. Ecological Modelling,1999,120:109-118．

[31] Luo Xianxiang,Deng Wei. The application of self-organizing neural network models based on the theory of ART in water resources classification[J]. Journal of Changchun University of Science and Technology,2001,31(1):54-57.[罗先香,邓伟. 基于ART理论的自组织神经网络模型在水资源分类中的应用[J]]. 长春科技大学学报,2001,31(1):54-57．]

[32] Foody G M. Applications of the self-organising feature map neural network in community data analysis[J]. Ecological Modelling,1999,120:97-107．

[33] Li Shuangcheng. Plant species response to climatic change:Theory,model and application—A case study in the warm temperate area of North China[D]. Beijing:Institute of Geographic Sciences and Natural Resources Research,CAS.2000.[李双成. 植物种响应气候变化研究的理论、模型及区域性应用实践——以中国华北暖温带东部森林区为例[D]. 北京：中国科学院地理科学与资源研究所,2000．]

[34] Reich S L,Gomez D R,Dawidowski L E. Artificial neural network for the identification of unknown air pollution sources[J]. Atmospheric Environment,1999,33(18):3 045-3 052．

[35] Deadman P J,Gimblett H R. Applying neural networks to vegetation management plan development[J]. AI Applications,1997,11(3): 107-112．

[36] Malmgren B A,Winter A. Climate zonation in Puerto Rico based on principal components analysis and an artificial neural network[J]. Journal of Climate,1999,12(4): 977-985．

[37] Chang H C,Kopaska-Merkel D C,Chen H C,et al. Lithofacies identification using multiple adaptive resonance theory neural networks and group decision expert system[J]. Computers and Geosciences,2000,26(5):591-601．

[38] Luk K C,Ball J E,Sharma A. A study of optimal model lag and spatial inputs to artificial neural network for rainfall forecasting[J]. Journal of Hydrology,2000,227(1/4):56-65．

[39] Fang ZhiJiang,Qu Zheng. Application of Bp neural network in environmental recognition model[J].Coal Geology and Exploration,2000,28(2):13-15.[方志江,曲政. BP网络在沉积环境自动识别中的应用[J].煤田地质与勘探. 2000,28(2):13-15．]

[40] Findlay C S,Zheng L. Estimating ecosystem risks using cross-validated multiple regression and cross-validated holographic neural networks[J]. Ecological-Modelling,1999,119(1): 57-72．

[41] Yang Jianqiang,Luo Xianxiang. Models of self-organizing neural network for the evaluation of watershed water resources abundance[J]. Geography and Territorial Research,2000,16(2):45-47.[杨健强,罗先香.流域水资源丰富度评价的自组织神经网络[J].地理学与国土研究,2000,16(2):45-47.]

[42] Hu Mingxing,Guo Dazhi,Guo Lingxiang. Application of fuzzy association memory neural network in land quality evaluation[J]. Journal of NanJing Agricultural University,1999,22(1):104-107.[胡明星,郭达志,郭玲香.模糊联想记忆神经网络在土地质量评价中的应用[J].南京农业大学学报,1999,22(1):104-107．]

[43] Liu Guodong,Huang Chuanyou,Ding Jing. The models of artificial neural networks for comprehensive assessment of water quality[J]. China Environmental Science,1998,18(6):514-517.[刘国东,黄川友,丁晶. 水质综合评价的人工神经网络模型[J].中国环境科学,1998,18(6):514-517.]

[44] Lu Wenxi,Zhu Tingcheng. Artificial neural network evaluation of lake eutrophication[J]. Chinese Journal of Applied Ecology,1998,9(6):645-650 [卢文喜,祝廷成. 应用人工神经网络评价湖泊的富营养化[J].应用生态学报,1998,9(6):645-650．]

[45] Gao Xuemin,Chen Jingsheng,Wang Lixin. Appling BP neural network to study water quality of the Yangtze river[J]. Research of Environmental Sciences,2001,14(1):49-52．[高学民,陈静生,王立新. BP网络应用于长江水质研究[J]. 环境科学研究,2001,14(1):49-52．]

[46] Wen C G,Lee C S. A neural network approach to multiobJective optimization for water quality management in a river basin[J]. Water Resources Research,1998, 34(3):427-436．

[47] Chen Shouyu,Nie Xiangtian,Zhu Wenbin,et al. A model of fuzzy optimization neural networks and its application[J]. Advances in Water Science,1999,10(1):69-74.[陈守煜,聂相田,朱文彬,等. 模糊优选神经网络模型及其应用[J]. 水科学进展,1999,10(1):69-74.]

[48] Hong Wei, Wu Chengzhen, He DongJin. A study on the model of forest resources management based on the artificial neural network[J]. Journal of Natural Resources, 1998,13(1):69-72.[洪伟,吴承祯,何东进. 基于人工神经网络的森林资源管理模型研究[J].自然资源学报,1998,13(1):69-72.]

[49] Li Jiping. Building the optimal model of tree species structure by Hopfield network[J]. Journal of Central South Forestry University,1998,18(1):80-83.[李际平.用Hopfield网络模型解决林种树种结构优化问题[J]. 中南林学院学报,1998,18(1):80-83．]

[50] Sahimi M. Fractal-wavelet neural-network approach to characterization and upscaling of fractured reservoirs[J]. Computers and Geosciences,2000,26(8):877-905．

[51] Liu Wei,Li Jinping,Xiong Jianhui,et al. Application of wavelet neural network to the data compression of infrared spectrum[J]. Chinese Science Bulletin,42(2):824-826. [刘伟,李金屏,熊建辉,等. 小波神经网络在红外光谱数据压缩中的应用[J]. 科学通报,42(2):824-826.]

[52] Zhang Zhixing,Sun Chunzai,Mizutani E. Neuro-fuzzy and soft computing[M]. Xi’an:Xi’an Traffic University Press,2000.[张智星,孙春在,水谷英二. 神经—模糊和软计算[M]. 西安：西安交通大学出版社,2000．]

No related articles found!

阅读次数

全文

摘要

APPLICATIONS OF ARTIFICIAL NEURAL NETWORKS TO GEOSCIENCES:REVIEW AND PROSPECT