基於電性時序波形的光伏陣列故障診斷方法

專利類型

發明

專利國別 (專利申請國家)

中華民國

專利申請案號

109126868

專利證號

I 741727

專利獲證名稱

基於電性時序波形的光伏陣列故障診斷方法

專利所屬機關 (申請機關)

國立臺灣科技大學

獲證日期

2021/10/01

技術說明

光伏陣列輸出的非線性特性以及最大功率點跟蹤演算法，容易影響常規光伏故障診斷方法的實施，導致診斷失敗引起功率損失甚至安全事故。本專利技術針對光伏陣列在故障瞬間時序波形變化規律的基礎上，提出了一種新型基於時序波形的光伏故障診斷方法。首先採集故障發生前後的電壓和電流時序波形，通過標準化操作將標準化後的電壓、電流和功率波形作為輸入信號。接著透過堆疊自動編碼器實現故障特徵提取，並提出一種改進的深度森林對光伏陣列的線-線、開路、遮陰等故障進行診斷。本專利技術所提方法的優點是利用堆疊自動編碼器自動提取出具有較高辨識度的特徵。利用改進的深度森林演算法實現故障特徵的增強和挖掘，特別是所提的改進方法可以降低特徵向量維度的同時，增強各級森林間資訊連通性，提高診斷的準確率。數值模擬和實測資料對方法的有效性進行了進一步驗證，所提方法對模擬和實驗的故障診斷準確率分別達到了98.87%和97.66%，優於傳統softmax、SVM、RF、gcForest、daForest等方法。 The non-linear output power characteristics of photovoltaic (PV) arrays and the implementation of the maximum power point tracking (MPPT) algorithm will affect the precision of conventional PV fault diagnostic methods easily. Due to diagnostic failure, it may lead to power losses and even safety accidents. In this study, the variation characteristics of the sequence waveforms of PV arrays at the moment of failure are investigated and used to develop a newly-designed PV fault diagnostic framework. Firstly, the sequence waveforms of string voltages and currents before and after the fault occurred are collected; the normalized sequence data of voltages, currents, and powers are used as analytic data. Then, the fault feature extraction is realized via a stacked autoencoder (SAE) model. Moreover, an improved multi-grained cascade forest (IgcForest) is proposed to diagnose faults, e.g., line-to-line (L-L) fault, open-circuit (OC) fault, partial-shading of PV arrays, etc. The advantages of the proposed method contain the SAE method to extract features with higher recognition automatically, and the IgcForest to enhance and exploit fault features. Especially, the proposed improvement measures can reduce the feature vector dimension and enhance the information connectivity between forests at all levels for further improving the accuracy of diagnoses. In addition, the validity of the proposed method is verified by numerical simulations and measured data, and the corresponding accuracy of fault diagnoses can reach 98.87% and 97.66%, respectively, which are superior to traditional methods, such as softmax, support vector machines, random forest, gcForest, and daForest.

備註

連絡單位 (專責單位/部門名稱)

技術移轉中心

連絡電話

02-2733-3141#7346

網址