摘 要
船舶機(jī)艙是船舶動(dòng)力的源泉，其正常運(yùn)行對(duì)海洋運(yùn)輸?shù)陌踩杂兄e足輕重的意義。而機(jī)艙自動(dòng)化監(jiān)控和故障診斷技術(shù)又是船舶安全性的重要部分。本文以船舶柴油機(jī)作為研究對(duì)象，通過(guò)虛擬儀器、測(cè)試技術(shù)、模糊神經(jīng)網(wǎng)絡(luò)、蟻群算法等理論為基礎(chǔ)，深入的研究了機(jī)艙監(jiān)控的各模塊的設(shè)計(jì)、故障特征提取和診斷方法、以及蟻群算法所優(yōu)化模糊神經(jīng)網(wǎng)絡(luò)系統(tǒng)。
本文是基于LabView的船舶柴油機(jī)故障診斷的研究。針對(duì)傳統(tǒng)的開發(fā)軟件（VB,VC++,C語(yǔ)言等），開發(fā)機(jī)艙監(jiān)控系統(tǒng)開發(fā)周期長(zhǎng)、運(yùn)行速度慢、調(diào)試和維護(hù)困難，以及難以實(shí)現(xiàn)系統(tǒng)冗余等缺點(diǎn)，本文采用了虛擬儀器技術(shù)模塊化思想設(shè)計(jì)了機(jī)艙監(jiān)控及故障診斷系統(tǒng)。實(shí)現(xiàn)機(jī)艙各部件的信號(hào)采集、處理、集中顯示、報(bào)警控制。為實(shí)現(xiàn)船舶柴油機(jī)故障診斷奠定了基礎(chǔ)。
近年來(lái)，故障診斷技術(shù)飛速發(fā)展，在智能故障診斷中，往往不存在簡(jiǎn)單的對(duì)應(yīng)關(guān)系，錯(cuò)綜復(fù)雜。將模糊邏輯和人工神經(jīng)網(wǎng)絡(luò)相結(jié)合形成的模糊神經(jīng)網(wǎng)絡(luò)可以有效地利用模糊的信息，又有自學(xué)習(xí)的特點(diǎn)，能更好地勝任智能故障診斷。本文將最新的蟻群算法和模糊神經(jīng)網(wǎng)絡(luò)故障診斷相結(jié)合，對(duì)參數(shù)進(jìn)行優(yōu)化訓(xùn)練，優(yōu)化網(wǎng)絡(luò)的權(quán)值和閾值，得到了更好地收斂效果和速度。避免了網(wǎng)絡(luò)訓(xùn)練限于局部極值和收斂速度慢的弊端，并且在MATLAB環(huán)境下進(jìn)行了仿真，具有一定的可行性和可靠性。

關(guān)鍵詞 機(jī)艙監(jiān)控；LabView；蟻群算法；故障診斷；模糊神經(jīng)網(wǎng)絡(luò)




Abstract
Marine engine room is the fountainhead of the marine power plant. Its normal operation has the important significance to the ship transportation safety. The engine room automation monitoring and fault diagnosis technologies are primary parts of modern ship work safety.Focusing on diesel engines, this thesis attempts to do a study on modular designs of engine room monitoring, faults feature extraction, fault diagnosis method of diesel engines and Fuzzy Neutral Network optimization by applying principles of Ant Colony Algorithm, Virtual Instrument and test technology.
This paper aims to research the remote fault diagnosis to marine diesel engines using LabView. Moreover, applying modular thinking in Virtual Instrument technology, the thesis designs engine room monitoring and fault diagnosis technologies to overcome the shortcomings of traditional development environments (VB, VC++, C, etc), such as long development cycle, limited running speed and difficulties in debugging, maintaining and redundancy. The data acquisition, processing, central display and alarm-control are also realized, forming solid foundations for the diesel engine remote fault diagnosis.
In recent years, with the swift advance in breakdown diagnosis technology, instead of simple corresponding relationship, complex relationship often exists. Fuzzy Neutral Network (FNN), formed by combining fuzzy logic and artificial neutral network, could effectively apply fuzzy information. With special features of self study, FNN could better handle the task of fault diagnosis. This thesis combines the latest Ant Colony Algorithm and Fuzzy Neutral Network to achieve the optimization training of the parameters and the optimization of network weight and threshold values, leading to better convergence performance and speed. Avoiding shortcomings of local extremum and slow convergence in network training, at the same time, simulating in the environment of MATLAB, the paper demonstrates considerable feasibility and reliability.

Key words: Engine room monitoring, LabView, Ant Colony Algorithm, Breakdown diagnosis, Fuzzy Neutral Network




目 錄
摘 要 I
Abstract III
第1章 緒 論 1
1.1 引言 1
1.2 故障診斷技術(shù)的研究現(xiàn)狀與發(fā)展趨勢(shì) 1
1.3 蟻群算法的研究與發(fā)展 3
1.4 論文結(jié)構(gòu)安排和主要研究?jī)?nèi)容 3
第2章 系統(tǒng)的開發(fā)環(huán)境——虛擬儀器 5
2.1 引言 5
2.2 虛擬儀器的概念 5
2.3 虛擬儀器的系統(tǒng)構(gòu)成 5
2.3.1 虛擬儀器的硬件構(gòu)成 6
2.3.1.1.GPIB系統(tǒng) 6
2.3.1.2數(shù)據(jù)采集系統(tǒng)DAQ 7
2.3.1.3 PXI系統(tǒng) 7
2.3.1.4 VXI系統(tǒng) 7
2.3.1.5 PC端口-LPT并行口式、USB口式和1394口式系統(tǒng) 7
2.3.2 虛擬儀器系統(tǒng)軟件構(gòu)成 8
2.4 虛擬儀器的優(yōu)勢(shì)與發(fā)展趨勢(shì) 9
2.4.1 虛擬儀器的特點(diǎn)與優(yōu)勢(shì) 9
2.4.2 虛擬儀器的發(fā)展趨勢(shì) 11
2.5 圖形化編程語(yǔ)言LabView 11
2.5.1 LabView概述 11
2.5.2 LabView組成與特點(diǎn) 11
2.5.3 LabView的優(yōu)勢(shì) 12
2.6 本章小結(jié) 13
第3章 船舶機(jī)艙監(jiān)控系統(tǒng)功能模塊設(shè)計(jì) 14
3.1 機(jī)艙監(jiān)控的設(shè)計(jì)思想與方案 14
3.1.1 系統(tǒng)設(shè)計(jì)思想 14
3.1.2 系統(tǒng)設(shè)計(jì)方案 15
3.2 系統(tǒng)硬件設(shè)計(jì) 17
3.3 系統(tǒng)軟件設(shè)計(jì) 20
3.3.1 軟件總體設(shè)計(jì) 20
3.3.2 軟件設(shè)計(jì)算法分析 21
3.3.3 軟件主程序流程 24
3.3.4 基于LabView的各功能模塊設(shè)計(jì) 25
3.3.4.1 數(shù)據(jù)采集程序設(shè)計(jì) 26
3.3.4.2 菜單設(shè)置程序設(shè)計(jì) 29
3.3.4.3 實(shí)時(shí)監(jiān)控程序設(shè)計(jì) 30
3.3.4.4 監(jiān)測(cè)波形程序設(shè)計(jì) 30
3.3.4.5 歷史數(shù)據(jù)波形 31
3.3.4.6 報(bào)警程序設(shè)計(jì) 32
3.3.4.7 參數(shù)配置程序設(shè)計(jì) 33
3.3.4.8 數(shù)據(jù)管理程序設(shè)計(jì) 34
3.4 本章小結(jié) 35
第4章 基于模糊神經(jīng)網(wǎng)絡(luò)的船舶柴油機(jī)故障診斷 36
4.1船舶柴油機(jī)故障診斷概述 36
4.1.1船舶柴油機(jī)的故障原因及性質(zhì) 36
4.1.2船舶柴油機(jī)故障主要模式 36
4.1.3船舶柴油機(jī)特征參量 37
4.2 模糊神經(jīng)網(wǎng)絡(luò) 38
4.2.1 模糊理論 38
4.2.1.1 模糊集合與隸屬度 38
4.2.1.2 模糊邏輯和模糊條件推理 40
4.2.2 神經(jīng)網(wǎng)絡(luò)的基本概念 41
4.2.3模糊神經(jīng)網(wǎng)絡(luò) 42
4.3 基于FNN 的船舶柴油機(jī)智能故障診斷 45
4.3..