基于BP算法的多金屬結(jié)核覆蓋率分布預(yù)測模型研究

信息計算科學(xué)論文編號:XXLW048 論文字數(shù):10860,頁數(shù):24

目 錄
摘  要 I
Abstract II
目 錄 III
引  言 1
第一章    基于MATLAB的神經(jīng)網(wǎng)絡(luò)基本理論及設(shè)計 2
1.1  神經(jīng)網(wǎng)絡(luò)基本概念 2
1.1.1  神經(jīng)元模型 2
1.1.2  神經(jīng)網(wǎng)絡(luò)的特性和結(jié)構(gòu) 4
1.2  BP神經(jīng)網(wǎng)絡(luò)原理及設(shè)計 4
1.2.1  BP神經(jīng)網(wǎng)絡(luò)結(jié)構(gòu) 4
1.2.2  BP神經(jīng)網(wǎng)絡(luò)的生成及初始化 5
1.2.3  BP神經(jīng)網(wǎng)絡(luò)的學(xué)習(xí)規(guī)則 6
1.2.4  BP神經(jīng)網(wǎng)絡(luò)的訓(xùn)練和仿真 8
1.2.5  BP神經(jīng)網(wǎng)絡(luò)推廣能力的提高 9
第二章    結(jié)核覆蓋率分布的BP網(wǎng)絡(luò)模型 10
2.1  結(jié)核覆蓋率分布與地質(zhì)因素的相關(guān)性分析 10
2.2  傳遞函數(shù)、訓(xùn)練函數(shù)、學(xué)習(xí)函數(shù)的選擇 10
2.3  Matlab7.0中BP網(wǎng)絡(luò)的訓(xùn)練與仿真實施過程 11
2.3.1  確定隱含層神經(jīng)元數(shù)目 11
2.3.2  BP網(wǎng)絡(luò)的訓(xùn)練過程及結(jié)果分析 12
2.3.3  BP網(wǎng)絡(luò)的仿真過程及結(jié)果分析 14
第三章    多金屬結(jié)核覆蓋率分布的預(yù)測及分析 16
3.1  選取測區(qū)進行預(yù)測 16
3.2  預(yù)測結(jié)果的平面等值線表示 16
第四章    結(jié)  論 17
致  謝 18
參考文獻 19
附  錄 20

摘  要
 多金屬結(jié)核是一種蘊藏量巨大的礦產(chǎn)資源，它廣泛分布于大洋海底，分布很復(fù)雜，受到多種因素的非線性影響，導(dǎo)致多金屬結(jié)核在海底的分布呈現(xiàn)出擬隨機性，因而，多金屬結(jié)核分布問題的研究屬于非線性范疇。本文利用人工神經(jīng)網(wǎng)絡(luò)中的應(yīng)用最為廣泛的BP網(wǎng)絡(luò)，建立控制多金屬結(jié)核分布的地質(zhì)因素經(jīng)度、緯度、水深、坡度與其覆蓋率分布之間的預(yù)測模型。隱含層神經(jīng)元個數(shù)的確定，傳遞函數(shù)、訓(xùn)練函數(shù)、學(xué)習(xí)函數(shù)的選擇，推廣能力的提高，控制了所建網(wǎng)絡(luò)的優(yōu)劣。最終確定了4-5-1的網(wǎng)絡(luò)結(jié)構(gòu)，隱含層中的神經(jīng)元傳遞函數(shù)采用tansig函數(shù)，輸出層神經(jīng)元傳遞函數(shù)采用purelin函數(shù)，訓(xùn)練函數(shù)選擇traindgx函數(shù)，學(xué)習(xí)函數(shù)選擇默認learngdm函數(shù)，提高推廣能力的方法采用“提前停止”法。最終訓(xùn)練成一個較有預(yù)測價值的網(wǎng)絡(luò)，對未探區(qū)進行了預(yù)測。

關(guān)鍵詞：BP神經(jīng)網(wǎng)絡(luò)  隱含層個數(shù) 推廣能力 結(jié)果回歸分析 平面等值線

Abstract
Reserves of polymetallic nodules is a great mineral resources, is widely distributed in the ocean floor, the distribution is complex and nonlinear effects by many factors, resulting in the distribution of polymetallic nodules in the seabed appeared to be random, and therefore, distribution of polymetallic nodules are non-linear aspects of the problem. In this paper, artificial neural networks in the most widely used BP network, the establishment of the geological factors that control the distribution of polymetallic nodules longitude, latitude, depth, slope and its distribution between the coverage prediction model. Number of hidden layer neuron identification, transfer function, training function, the choice of learning function, to promote empowerment, control of the network built by the pros and cons. Finalized the 4-5-1 network structure, hidden layer neurons in the transfer function using tansig function, transfer function of output layer neurons use purelin function, training function selection traindgx function, learning function, choose the default learngdm function, improve the promotion of ability to approach a "stop ahead" method. Final training into a more predictive value of the network, not exploration area is predicted.

Keywords: BP neural network；hidden layer；Regression Analysis；Generalization；Contour