DATASET OF CURVES OF YOCASOL G

Mismatching and partial shading identification in photovoltaic arrays by an artificial neural net-work ensemble Michel Piliougine ([e-mail]) and Giovanni Spagnuolo ([e-mail])<br> <br> UNIVERSITA DEGLI STUDI DI SALERNO<br> DIEM – Via Giovanni Paolo II 132. 84084 Fisciano (SA), Italy <br> Date: May 5th, 2021<br> Version: 1.1 Filename: dataset.mat<br> Description: This file contains the dataset related to the article titled<br> Detection of partial shading and some faults in PV arrays by an ensemble of multilayer perceptrons This file has been stored using the native Matlab format for its workspace (*.mat) Now, the variables stored in the file are described: Variables:<br> (*) full_curvesOk_simulated<br> Cell array with 5000 full IV curves simulated using the SDM, assuming uniform conditions<br> Each IV curve is a 300 x 2 matrix (double), with column 1= voltage (V), column 2 = current (A)<br> (all the points of the simulated IV curve, not only those around MPP) <br> <br> From this variable, the following splitting has been made:<br> TRAINING =&gt; from index 1 to 3500<br> VALIDATION =&gt; from index 3501 to 4000<br> TEST =&gt; from index 4001 to 5000 (referred as TEST#1 in the paper) (*) full_curvesSh_simulated<br> Cell array with 5000 full IV curves simulated using the SDM, assuming FAULTY conditions (steeped curve)<br> Each IV curve is a 300 x 2 matrix (double), with column 1= voltage (V), column 2 = current (A)<br> (all the points of the simulated IV curve, not only those around MPP) <br> <br> From this variable, the following splitting has been made:<br> TRAINING =&gt; from index 1 to 3500<br> VALIDATION =&gt; from index 3501 to 4000<br> TEST =&gt; from index 4001 to 5000 (referred as TEST#1 in the paper)<br> <br> <br> (*) full_curvesOk_experimental<br> Cell array with 1000 full IV curves measured of a real PV module, assigned to be measured under<br> normal operating conditions, and assumed to be classified as HEALTHY<br> <br> Each IV curve is a n x 2 matrix (double), with column 1= voltage (V), column 2 = current (A)<br> (all the points of the simulated IV curve, not only those around MPP)<br> where n is the number of experimental points of each IV curve, although it is around 100, but it could<br> be different for each measured IV curve.<br> <br> All these curves has been used for testing<br> TEST =&gt; from index 1 to 1000 (referred as TEST#2 in the paper) (*) full_curvesSh_experimental<br> Cell array with 1000 full IV curves measured of a real PV module, assigned to be measured under<br> mismatched operating conditions, and assumed to be classified as FAULTY<br> <br> Each IV curve is a n x 2 matrix (double), with column 1= voltage (V), column 2 = current (A)<br> (all the points of the simulated IV curve, not only those around MPP)<br> where n is the number of experimental points of each IV curve, although it is around 100, but it could<br> be different for each measured IV curve.<br> <br> All these curves has been used for testing<br> TEST =&gt; from index 1 to 1000 (referred as TEST#2 in the paper)<br> <br>基于人工神经网络集成的光伏（Photovoltaic, PV）阵列失配与局部遮荫识别 作者：Michel Piliougine（邮箱：[e-mail]）与Giovanni Spagnuolo（邮箱：[e-mail]） 萨勒诺大学（UNIVERSITA DEGLI STUDI DI SALERNO） DIEM系——乔瓦尼·保罗二世大道132号，84084菲西亚诺（萨勒诺省），意大利 日期：2021年5月5日 版本：1.1 文件名：dataset.mat 本文件包含与题为《基于多层感知器（Multilayer Perceptron, MLP）集成检测光伏阵列局部遮荫与部分故障》的学术论文相关的数据集。本文件采用Matlab工作区原生格式（*.mat）存储。 以下对文件中存储的变量进行说明： 1. full_curvesOk_simulated 该元胞数组包含5000条基于单二极管模型（Single Diode Model, SDM）、在均匀光照工况下模拟得到的完整IV曲线。每条IV曲线为300×2的双精度矩阵，其中第一列为电压（单位：V），第二列为电流（单位：A），包含模拟IV曲线的全部采样点，而非仅最大功率点（Maximum Power Point, MPP）附近的采样点。 该变量的数据集拆分规则如下： - 训练集：索引1至3500 - 验证集：索引3501至4000 - 测试集：索引4001至5000（即论文中提及的TEST#1） 2. full_curvesSh_simulated 该元胞数组包含5000条基于单二极管模型（SDM）、在故障工况下模拟得到的完整IV曲线（曲线呈陡降型）。每条IV曲线为300×2的双精度矩阵，第一列为电压（V），第二列为电流（A），包含全部采样点而非仅最大功率点附近的采样点。 数据集拆分规则同上： - 训练集：索引1至3500 - 验证集：索引3501至4000 - 测试集：索引4001至5000（论文中TEST#1） 3. full_curvesOk_experimental 该元胞数组包含1000条实际光伏组件在正常运行工况下测得的完整IV曲线，归类为健康样本。 每条IV曲线为n×2的双精度矩阵，第一列为电压（V），第二列为电流（A），包含实测IV曲线的全部采样点，而非仅最大功率点附近的采样点。其中n为每条IV曲线的实验采样点数，通常约为100，但不同实测曲线的采样点数可能存在差异。 所有曲线均用于测试，测试集索引为1至1000（论文中TEST#2） 4. full_curvesSh_experimental 该元胞数组包含1000条实际光伏组件在失配运行工况下测得的完整IV曲线，归类为故障样本。 每条IV曲线为n×2的双精度矩阵，第一列为电压（V），第二列为电流（A），包含实测IV曲线的全部采样点，而非仅最大功率点附近的采样点。其中n为每条IV曲线的实验采样点数，通常约为100，但不同实测曲线的采样点数可能存在差异。 所有曲线均用于测试，测试集索引为1至1000（论文中TEST#2）