US 12,068,717 B2
Photovoltaic array fault diagnosis method based on composite information
Fang Deng, Beijing (CN); Zelang Liang, Beijing (CN); Ning Ding, Beijing (CN); Xinyu Fan, Beijing (CN); Xin Gao, Beijing (CN); Yeyun Cai, Beijing (CN); and Jie Chen, Beijing (CN)
Assigned to BEIJING INSTITUTE OF TECHNOLOGY, Beijing (CN)
Filed by BEIJING INSTITUTE OF TECHNOLOGY, Beijing (CN)
Filed on Nov. 9, 2020, as Appl. No. 17/093,244.
Application 17/093,244 is a continuation of application No. PCT/CN2019/000095, filed on May 7, 2019.
Claims priority of application No. 201810439521.5 (CN), filed on May 9, 2018.
Prior Publication US 2021/0135625 A1, May 6, 2021
Int. Cl. H02S 50/10 (2014.01); G06F 18/21 (2023.01); G06F 18/2135 (2023.01); G06F 18/214 (2023.01); G06F 18/241 (2023.01); G06F 18/25 (2023.01); G06N 3/04 (2023.01); G06N 3/084 (2023.01); G06T 7/00 (2017.01); G06V 10/764 (2022.01); G06V 10/774 (2022.01); G06V 10/776 (2022.01); G06V 10/80 (2022.01); G06V 10/82 (2022.01)
CPC H02S 50/10 (2014.12) [G06F 18/2135 (2023.01); G06F 18/214 (2023.01); G06F 18/217 (2023.01); G06F 18/241 (2023.01); G06F 18/25 (2023.01); G06N 3/04 (2013.01); G06N 3/084 (2013.01); G06T 7/0002 (2013.01); G06V 10/764 (2022.01); G06V 10/774 (2022.01); G06V 10/776 (2022.01); G06V 10/809 (2022.01); G06V 10/82 (2022.01); G06T 2207/10024 (2013.01); G06T 2207/10048 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 5 Claims
OG exemplary drawing
 
1. A fault diagnosis method for a photovoltaic array, comprising:
collecting and preprocessing composite information data of a status of the photovoltaic array, the composite information data including image data of the status of the photovoltaic array and text data of the status of the photovoltaic;
training a fault classification model of a deep convolutional neural network using the image data of the status of the photovoltaic array, thereby obtaining an image fault classification model after the training, and training a fault classification model based on a support vector machine (SVM) using the text data of the status of the photovoltaic array, thereby obtaining a text fault classification model after the training; and
fusing the image fault classification model and the text fault classification model by a logistic regression algorithm to obtain the fusion model, and training the fusion model using the composite information data of the status of the photovoltaic array, thereby obtaining a photovoltaic array fault diagnosis model,
wherein the status of the photovoltaic array includes a normal working state, a hot spot fault state, an open circuit fault state and a short circuit fault state; wherein a label is set for each of the normal working state, the hot spot fault state, the open circuit fault state and the short circuit fault state;
wherein the image data of the status of the photovoltaic array include the photovoltaic array infrared images and the labels for the normal working state, hot spot fault state, open circuit fault state and short circuit fault state of the photovoltaic array; and
wherein the text data of the status of the photovoltaic array include the open-circuit voltage, short-circuit current, maximum high-power point voltage, maximum high-power point current, ambient light intensity, temperature, and the labels for the normal working state, hot spot fault state, open circuit fault state and short circuit fault state of the photovoltaic array.