本文選題：關(guān)聯(lián)規(guī)則 + 艦船�。� 參考：《艦船科學(xué)技術(shù)》2017年24期

【摘要】：為了提高艦船故障檢測能力,需要進(jìn)行艦船故障數(shù)據(jù)的實(shí)時(shí)挖掘和分類分析,提出一種基于關(guān)聯(lián)規(guī)則的艦船故障數(shù)據(jù)的定位挖掘方法。采用電磁探測器、水聲換能器、聲吶裝置、聲學(xué)傳感器等設(shè)備進(jìn)行不同工況下艦船數(shù)據(jù)采集,包括艦船輻射噪聲、機(jī)械振動等數(shù)據(jù),對采集的數(shù)據(jù)進(jìn)行高維特征融合處理,提取艦船故障數(shù)據(jù)的關(guān)聯(lián)規(guī)則特征量,對提取的特征量采用K均值算法進(jìn)行聚類分析,并通過BP神經(jīng)網(wǎng)絡(luò)分類器實(shí)現(xiàn)艦船故障數(shù)據(jù)的分類識別,實(shí)現(xiàn)艦船故障數(shù)據(jù)定位挖掘。仿真結(jié)果表明,采用該方法進(jìn)行艦船故障數(shù)據(jù)挖掘的準(zhǔn)確性較好,對故障的定位能力較強(qiáng),提高了艦船實(shí)時(shí)故障診斷能力。
[Abstract]:In order to improve the capability of ship fault detection, it is necessary to mine and classify ship fault data in real time, and a location mining method of ship fault data based on association rules is proposed. Using electromagnetic detector, underwater transducer, sonar device, acoustic sensor and other equipment to collect ship data under different working conditions, including ship radiation noise, mechanical vibration and other data, the data collected are processed with high dimensional feature fusion. The feature quantity of association rules is extracted from ship fault data. K-means algorithm is used to cluster the extracted feature quantity. The classification and recognition of ship fault data are realized by BP neural network classifier, and the location mining of ship fault data is realized. The simulation results show that the method is accurate and has a strong ability to locate faults and improve the ability of real-time fault diagnosis.
【作者單位】： 欽州學(xué)院;
【基金】：基金項(xiàng)目 欽州市“互聯(lián)網(wǎng)+先進(jìn)制造”工程技術(shù)研究中心(欽科發(fā)[2016]138號)
【分類號】：TP311.13;U674.707

【相似文獻(xiàn)】

相關(guān)期刊論文 前2條

1 叢林虎;徐廷學(xué);卜祝濤;楊繼坤;;基于故障數(shù)據(jù)的導(dǎo)彈突發(fā)故障預(yù)測方法[J];戰(zhàn)術(shù)導(dǎo)彈技術(shù);2014年02期

2 史全林,周源泉;一組軟件故障數(shù)據(jù)的分析[J];導(dǎo)彈與航天運(yùn)載技術(shù);1999年06期

，

本文編號：1942237