【摘要】：探索了BP(Back Propagation)神經(jīng)網(wǎng)絡(luò)在IGBT模塊結(jié)溫預(yù)測中的應(yīng)用。首先從物理機理角度分析了IGBT模塊中飽和壓降、集電極電流與溫度的內(nèi)在關(guān)系。通過IGBT結(jié)溫測試試驗,獲取了三者的關(guān)系圖,并建立了基于飽和壓降的溫敏參數(shù)法結(jié)溫預(yù)測模型。最后建立了BP神經(jīng)網(wǎng)絡(luò)結(jié)溫預(yù)測模型,并與溫敏參數(shù)法進行了對比分析,證明該方法計算速度更快,誤差降低。
[Abstract]:The application of BP (back Propagation) neural network in IGBT module junction temperature prediction is explored. Firstly, the relationship between saturation voltage drop, collector current and temperature in IGBT module is analyzed from the point of view of physical mechanism. Through IGBT junction temperature test, the relationship diagram of the three is obtained, and the temperature prediction model of temperature sensitive parameter method based on saturation pressure drop is established. Finally, the prediction model of junction temperature based on BP neural network is established, and compared with the temperature sensitive parameter method, it is proved that the calculation speed of this method is faster and the error is reduced.
【作者單位】： 河北工業(yè)大學(xué)電氣工程學(xué)院;天津電力公司城東供電公司;
【基金】：國家自然科技基金資助項目(51377044) 河北省科技支撐計劃基金資助項目(13214303D,14214503D)
【分類號】：TN322.8;TP183

【參考文獻】

相關(guān)期刊論文 前4條

1 梅霜;李志剛;姚芳;張亞玲;;IGBT模塊電氣參數(shù)測試及分析[J];西北大學(xué)學(xué)報(自然科學(xué)版);2016年02期

2 張亞玲;李志剛;姚芳;黃歡;;一種IGBT熱阻測試方法的研究[J];河北工業(yè)大學(xué)學(xué)報;2016年01期

3 陳明;胡安;唐勇;汪波;;絕緣柵雙極型晶體管傳熱模型建模分析[J];高電壓技術(shù);2011年02期

4 楊U，

本文編號：2123891