Nonlinear photovoltaic(PV)output is greatly affected by the nonuniform distribution of daily irradiance,preventing conventional protection devices from reliably detecting faults.Smart fault diagnosis and good maintena...Nonlinear photovoltaic(PV)output is greatly affected by the nonuniform distribution of daily irradiance,preventing conventional protection devices from reliably detecting faults.Smart fault diagnosis and good maintenance systems are essential for optimizing the overall productivity of a PV system and improving its life cycle.Hence,a multiscale smart fault diagnosis model for improved PV system maintenance strategies is proposed.This study focuses on diagnosing permanent faults(open-circuit faults,ground faults,and line-line faults)and temporary faults(partial shading)in PV arrays,using the random forest algorithm to conduct time-series analysis of waveform length and autoregression(RF-WLAR)as the main features,with 10-fold cross-validation using Matlab/Simulink.The actual irradiance data at 5.86°N and 102.03°E were used as inputs to produce simulated data that closely matched the on-site PV output data.Fault data from the maintenance database of a 2 MW PV power plant in Pasir Mas Kelantan,Malaysia,were used for field testing to verify the developed model.The RF-WLAR model achieved an average fault-type classification accuracy of 98%,with 100%accuracy in classifying partial shading and line-line faults.展开更多
Lidar (light detection and ranging) remote sensing is a breakthrough of active remote sensing technology in recent years. It has shown enormous potential for forest parameters retrieval. Lidar remote sensing has the u...Lidar (light detection and ranging) remote sensing is a breakthrough of active remote sensing technology in recent years. It has shown enormous potential for forest parameters retrieval. Lidar remote sensing has the unique advantage of providing horizontal and vertical information at high accuracies. Especially it can be used to measure forest height directly with unprecedented accuracy. Large footprint lidar has demonstrated its great potential for accurate estimation of many forest parameters. The geoscience laser altimeter system (GLAS) instrument aboard the ice, cloud and land elevation satellite (ICEsat) has acquired a large amount of data including topography and vegetation height information. Although GLAS’ primary mission is the topographic mapping of the ice sheets of greenland and antarctica, it has potential use over land, especially for vegetation height extraction. These data provide an unprecedented vegetation height data set over large area. After a general discussion of GLAS waveform pre_processing, the waveform length extraction method has been developed. Then the waveform length from GLAS Laser 2a data in the northeast China was calculated. The waveform length map was analyzed together with land cover map from Landsat ETM+. The waveform length shows good accordant with land cover types from Landsat ETM+ data. As for forest area, the waveform length map contains much more information about forest height information, which can be used to inverse other forest parameters quantitatively together with other remote sensing data.展开更多
基金Supported by the Universiti Malaysia Pahang (UMP)for the Financial Support Received under Project Number RDU223001 and PGRS2003189.
文摘Nonlinear photovoltaic(PV)output is greatly affected by the nonuniform distribution of daily irradiance,preventing conventional protection devices from reliably detecting faults.Smart fault diagnosis and good maintenance systems are essential for optimizing the overall productivity of a PV system and improving its life cycle.Hence,a multiscale smart fault diagnosis model for improved PV system maintenance strategies is proposed.This study focuses on diagnosing permanent faults(open-circuit faults,ground faults,and line-line faults)and temporary faults(partial shading)in PV arrays,using the random forest algorithm to conduct time-series analysis of waveform length and autoregression(RF-WLAR)as the main features,with 10-fold cross-validation using Matlab/Simulink.The actual irradiance data at 5.86°N and 102.03°E were used as inputs to produce simulated data that closely matched the on-site PV output data.Fault data from the maintenance database of a 2 MW PV power plant in Pasir Mas Kelantan,Malaysia,were used for field testing to verify the developed model.The RF-WLAR model achieved an average fault-type classification accuracy of 98%,with 100%accuracy in classifying partial shading and line-line faults.
文摘Lidar (light detection and ranging) remote sensing is a breakthrough of active remote sensing technology in recent years. It has shown enormous potential for forest parameters retrieval. Lidar remote sensing has the unique advantage of providing horizontal and vertical information at high accuracies. Especially it can be used to measure forest height directly with unprecedented accuracy. Large footprint lidar has demonstrated its great potential for accurate estimation of many forest parameters. The geoscience laser altimeter system (GLAS) instrument aboard the ice, cloud and land elevation satellite (ICEsat) has acquired a large amount of data including topography and vegetation height information. Although GLAS’ primary mission is the topographic mapping of the ice sheets of greenland and antarctica, it has potential use over land, especially for vegetation height extraction. These data provide an unprecedented vegetation height data set over large area. After a general discussion of GLAS waveform pre_processing, the waveform length extraction method has been developed. Then the waveform length from GLAS Laser 2a data in the northeast China was calculated. The waveform length map was analyzed together with land cover map from Landsat ETM+. The waveform length shows good accordant with land cover types from Landsat ETM+ data. As for forest area, the waveform length map contains much more information about forest height information, which can be used to inverse other forest parameters quantitatively together with other remote sensing data.
