中圖分類號(hào)： TN102,；TM531 文獻(xiàn)標(biāo)識(shí)碼： A DOI：10.16157/j.issn.0258-7998.201128 中文引用格式： 楊培盛,，付宇，李鴻飛,，等. 基于CNN-LSTM的支撐電容容值軟測(cè)量[J].電子技術(shù)應(yīng)用,，2021，47(9)：16-19. 英文引用格式： Yang Peisheng,，F(xiàn)u Yu,，Li Hongfei，et al. Soft measurement of supporting capacitance based on CNN-LSTM[J]. Application of Electronic Technique,，2021,，47(9)：16-19.
Soft measurement of supporting capacitance based on CNN-LSTM
Yang Peisheng1，F(xiàn)u Yu1,，Li Hongfei2,，Chu Kaiqi2，Wang Mengqian2,，Li Zhengda2
1.Jinan Rail Transit Group Construction Investment Co.,，Ltd.,，Jinan 250014，China,； 2.CRRC Qingdao Sifang Rolling Stock Research Institute Co.,，Ltd.，Qingdao 266033,，China
Abstract： It is of great significance to monitor the aging state of the supporting capacitors in the power converter in real time and to find and replace the defective capacitors in time. In this paper, based on the relevant voltage and current data, through the establishment of data sets, the network model parameters and model training are determined. Finally, the neural network model based on CNN-LSTM is obtained. The accuracy of the neural network model is verified by the data sets under different working conditions. The results show that the model can reliably predict the capacitance value.
Key words : support capacitor,；CNN-LSTM；reliability,；neural network
直流母線支撐電容作為牽引系統(tǒng)的關(guān)鍵部件,，其健康狀態(tài)隨著投入運(yùn)行年限的增加而變差,，直流母線電容失效導(dǎo)致的列車系統(tǒng)停機(jī)甚至損毀給社會(huì)帶來(lái)了巨大的經(jīng)濟(jì)損失[5-6]。因此,，支撐電容的狀態(tài)監(jiān)測(cè)技術(shù)成為了當(dāng)前研究的熱點(diǎn)[7-8],。支撐電容的容值能夠表征其真實(shí)的健康狀態(tài)[9]，本文提出了一種大功率變流器直流母線電容容值的在線監(jiān)測(cè)方法,，利用數(shù)據(jù)訓(xùn)練得到基于卷積神經(jīng)網(wǎng)絡(luò)-長(zhǎng)短期記憶網(wǎng)絡(luò)(Convolutional Neural Networks-Long Short Term Memory,，CNN-LSTM)的神經(jīng)網(wǎng)絡(luò)模型[10]，可以根據(jù)列車系統(tǒng)運(yùn)行過(guò)程中采集到的實(shí)時(shí)運(yùn)行數(shù)據(jù)進(jìn)行支撐電容值的準(zhǔn)確軟測(cè)量,，對(duì)于實(shí)現(xiàn)支撐電容健康狀態(tài)在線監(jiān)測(cè),、提高功率變流器的可靠性具有重要意義。
