The accurate prediction of photovoltaic(PV)power generation is an important basis for hybrid grid scheduling.With the expansion of the scale of PV power plants and the popularization of distributed PV,this study propo...The accurate prediction of photovoltaic(PV)power generation is an important basis for hybrid grid scheduling.With the expansion of the scale of PV power plants and the popularization of distributed PV,this study proposes a multilayer PV power generation prediction model based on transfer learning to solve the problems of the lack of data on new PV bases and the low accuracy of PV power generation prediction.The proposed model,called DRAM,concatenates a dilated convolutional neural network(DCNN)module with a bidirectional long short-term memory(BiLSTM)module,and integrates an attention mechanism.First,the processed data are input into the DCNN layer,and the dilation convolution mechanism captures the spatial features of the wide sensory field of the input data.Subsequently,the temporal characteristics between the features are extracted in the BiLSTM layer.Finally,an attention mechanism is used to strengthen the key features by assigning weights to efficiently construct the relationship between the features and output variables.In addition,the power prediction accuracy of the new PV sites was improved by transferring the pre-trained model parameters to the new PV site prediction model.In this study,the pre-training of models using data from different source domains and the correlations between these pre-trained models and the target domain were analyzed.展开更多
Remote sensing cross-modal image-text retrieval(RSCIR)can flexibly and subjectively retrieve remote sensing images utilizing query text,which has received more researchers’attention recently.However,with the increasi...Remote sensing cross-modal image-text retrieval(RSCIR)can flexibly and subjectively retrieve remote sensing images utilizing query text,which has received more researchers’attention recently.However,with the increasing volume of visual-language pre-training model parameters,direct transfer learning consumes a substantial amount of computational and storage resources.Moreover,recently proposed parameter-efficient transfer learning methods mainly focus on the reconstruction of channel features,ignoring the spatial features which are vital for modeling key entity relationships.To address these issues,we design an efficient transfer learning framework for RSCIR,which is based on spatial feature efficient reconstruction(SPER).A concise and efficient spatial adapter is introduced to enhance the extraction of spatial relationships.The spatial adapter is able to spatially reconstruct the features in the backbone with few parameters while incorporating the prior information from the channel dimension.We conduct quantitative and qualitative experiments on two different commonly used RSCIR datasets.Compared with traditional methods,our approach achieves an improvement of 3%-11% in sumR metric.Compared with methods finetuning all parameters,our proposed method only trains less than 1% of the parameters,while maintaining an overall performance of about 96%.展开更多
Gear fault diagnosis technologies have received rapid development and been effectively implemented in many engineering applications.However,the various working conditions would degrade the diagnostic performance and m...Gear fault diagnosis technologies have received rapid development and been effectively implemented in many engineering applications.However,the various working conditions would degrade the diagnostic performance and make gear fault diagnosis(GFD)more and more challenging.In this paper,a novel model parameter transfer(NMPT)is proposed to boost the performance of GFD under varying working conditions.Based on the previous transfer strategy that controls empirical risk of source domain,this method further integrates the superiorities of multi-task learning with the idea of transfer learning(TL)to acquire transferable knowledge by minimizing the discrepancies of separating hyperplanes between one specific working condition(target domain)and another(source domain),and then transferring both commonality and specialty parameters over tasks to make use of source domain samples to assist target GFD task when sufficient labeled samples from target domain are unavailable.For NMPT implementation,insufficient target domain features and abundant source domain features with supervised information are fed into NMPT model to train a robust classifier for target GFD task.Related experiments prove that NMPT is expected to be a valuable technology to boost practical GFD performance under various working conditions.The proposed methods provides a transfer learning-based framework to handle the problem of insufficient training samples in target task caused by variable operation conditions.展开更多
In this paper,we propose a parameterization transfer algorithm for planar domains bounded by B-spline curves,where the shapes of the planar domains are similar.The domain geometries are considered to be similar if the...In this paper,we propose a parameterization transfer algorithm for planar domains bounded by B-spline curves,where the shapes of the planar domains are similar.The domain geometries are considered to be similar if their simplified skeletons have the same structures.One domain we call source domain,and it is parameterized using multi-patch B-spline surfaces.The resulting parameterization is C1 continuous in the regular region and G1 continuous around singular points regardless of whether the parameterization of the source domain is C1/G1 continuous or not.In this algorithm,boundary control points of the source domain are extracted from its parameterization as sequential points,and we establish a correspondence between sequential boundary control points of the source domain and the target boundary through discrete sampling and fitting.Transfer of the parametrization satisfies C1/G1 continuity under discrete harmonic mapping with continuous constraints.The new algorithm has a lower calculation cost than a decomposition-based parameterization with a high-quality parameterization result.We demonstrate that the result of the parameterization transfer in this paper can be applied in isogeometric analysis.Moreover,because of the consistency of the parameterization for the two models,this method can be applied in many other geometry processing algorithms,such as morphing and deformation.展开更多
Soil capacity to support life and to produce economic goods and services is strongly linked to the maintenance of good soil physical quality(SPQ). In this study, the SPQ of citrus orchards was assessed under three dif...Soil capacity to support life and to produce economic goods and services is strongly linked to the maintenance of good soil physical quality(SPQ). In this study, the SPQ of citrus orchards was assessed under three different soil managements, namely no-tillage using herbicides, tillage under chemical farming, and no-tillage under organic farming. Commonly used indicators, such as soil bulk density,organic carbon content, and structural stability index, were considered in conjunction with capacitive indicators estimated by the Beerkan estimation of soil transfer parameter(BEST) method. The measurements taken at the L'Alcoleja Experimental Station in Spain yielded optimal values for soil bulk density and organic carbon content in 100% and 70% of cases for organic farming. The values of structural stability index indicated that the soil was stable in 90% of cases. Differences between the soil management practices were particularly clear in terms of plant-available water capacity and saturated hydraulic conductivity. Under organic farming, the soil had the greatest ability to store and provide water to plant roots, and to quickly drain excess water and facilitate root proliferation.Management practices adopted under organic farming(such as vegetation cover between the trees, chipping after pruning, and spreading the chips on the soil surface) improved the SPQ. Conversely, the conventional management strategies unequivocally led to soil degradation owing to the loss of organic matter, soil compaction, and reduced structural stability. The results in this study show that organic farming has a clear positive impact on the SPQ, suggesting that tillage and herbicide treatments should be avoided.展开更多
Osteosarcoma is a type of malignant bone tumor that is reported across the globe.Recent advancements in Machine Learning(ML)and Deep Learning(DL)models enable the detection and classification of malignancies in biomed...Osteosarcoma is a type of malignant bone tumor that is reported across the globe.Recent advancements in Machine Learning(ML)and Deep Learning(DL)models enable the detection and classification of malignancies in biomedical images.In this regard,the current study introduces a new Biomedical Osteosarcoma Image Classification using Elephant Herd Optimization and Deep Transfer Learning(BOIC-EHODTL)model.The presented BOIC-EHODTL model examines the biomedical images to diagnose distinct kinds of osteosarcoma.At the initial stage,Gabor Filter(GF)is applied as a pre-processing technique to get rid of the noise from images.In addition,Adam optimizer with MixNet model is also employed as a feature extraction technique to generate feature vectors.Then,EHOalgorithm is utilized along with Adaptive Neuro-Fuzzy Classifier(ANFC)model for recognition and categorization of osteosarcoma.EHO algorithm is utilized to fine-tune the parameters involved in ANFC model which in turn helps in accomplishing improved classification results.The design of EHO with ANFC model for classification of osteosarcoma is the novelty of current study.In order to demonstrate the improved performance of BOIC-EHODTL model,a comprehensive comparison was conducted between the proposed and existing models upon benchmark dataset and the results confirmed the better performance of BOIC-EHODTL model over recent methodologies.展开更多
Here we propose a new method of NDVI difference analysis and NDVI difference correction for multi-sensors to detect NDVI.This method integrate PROSPECT model,SAIL model and MODTRAN atmospheric radioactive transfer mod...Here we propose a new method of NDVI difference analysis and NDVI difference correction for multi-sensors to detect NDVI.This method integrate PROSPECT model,SAIL model and MODTRAN atmospheric radioactive transfer model to simulate the remote sensing signals of different satellites for both NDVI difference analysis and correction without using real satellite images.The effects of both the sensors' spectral responses and atmospheric condition are simulated,and the differences among NDVI values derived from thirty different satellites are analyzed quantitatively.Focusing on the conversions of NDVI values among different satellites,through linear regression analysis,we estimate the transfer parameters between any two different satellite NDVI values,and present the lookup tables of transfer parameters under the atmospheric conditions of three surface visibility range values (10,23 and 50 km).The proposed method is useful for NDVI applications and analyses for multi-sensors.展开更多
Protein subcellular localization prediction is im- portant for studying the function of proteins. Recently, as significant progress has been witnessed in the field of mi- croscopic imaging, automatically determining t...Protein subcellular localization prediction is im- portant for studying the function of proteins. Recently, as significant progress has been witnessed in the field of mi- croscopic imaging, automatically determining the subcellular localization of proteins from bio-images is becoming a new research hotspot. One of the central themes in this field is to determine what features are suitable for describing the pro- tein images. Existing feature extraction methods are usually hand-crafted designed, by which only one layer of features will be extracted, which may not be sufficient to represent the complex protein images. To this end, we propose a deep model based descriptor （DMD） to extract the high-level fea- tures from protein images. Specifically, in order to make the extracted features more generic, we firstly trained a convolu- tion neural network （i.e., AlexNe0 by using a natural image set with millions of labels, and then used the partial parame- ter transfer strategy to fine-tnne the parameters from natural images to protein images. After that, we applied the Lasso model to select the most distinguishing features from the last fully connected layer of the CNN （Convolution Neural Net- work）, and used these selected features for final classifica- tions. Experimental results on a protein image dataset vali- date the efficacy of our method.展开更多
基金Science and Technology Project of State Grid Ningxia Electric Power Co.,Ltd Research on Distributed Photovoltaic Fine Power Prediction Technology for Day-Ahead Scheduling,5229NX230007.
文摘The accurate prediction of photovoltaic(PV)power generation is an important basis for hybrid grid scheduling.With the expansion of the scale of PV power plants and the popularization of distributed PV,this study proposes a multilayer PV power generation prediction model based on transfer learning to solve the problems of the lack of data on new PV bases and the low accuracy of PV power generation prediction.The proposed model,called DRAM,concatenates a dilated convolutional neural network(DCNN)module with a bidirectional long short-term memory(BiLSTM)module,and integrates an attention mechanism.First,the processed data are input into the DCNN layer,and the dilation convolution mechanism captures the spatial features of the wide sensory field of the input data.Subsequently,the temporal characteristics between the features are extracted in the BiLSTM layer.Finally,an attention mechanism is used to strengthen the key features by assigning weights to efficiently construct the relationship between the features and output variables.In addition,the power prediction accuracy of the new PV sites was improved by transferring the pre-trained model parameters to the new PV site prediction model.In this study,the pre-training of models using data from different source domains and the correlations between these pre-trained models and the target domain were analyzed.
基金supported by the National Key R&D Program of China(No.2022ZD0118402)。
文摘Remote sensing cross-modal image-text retrieval(RSCIR)can flexibly and subjectively retrieve remote sensing images utilizing query text,which has received more researchers’attention recently.However,with the increasing volume of visual-language pre-training model parameters,direct transfer learning consumes a substantial amount of computational and storage resources.Moreover,recently proposed parameter-efficient transfer learning methods mainly focus on the reconstruction of channel features,ignoring the spatial features which are vital for modeling key entity relationships.To address these issues,we design an efficient transfer learning framework for RSCIR,which is based on spatial feature efficient reconstruction(SPER).A concise and efficient spatial adapter is introduced to enhance the extraction of spatial relationships.The spatial adapter is able to spatially reconstruct the features in the backbone with few parameters while incorporating the prior information from the channel dimension.We conduct quantitative and qualitative experiments on two different commonly used RSCIR datasets.Compared with traditional methods,our approach achieves an improvement of 3%-11% in sumR metric.Compared with methods finetuning all parameters,our proposed method only trains less than 1% of the parameters,while maintaining an overall performance of about 96%.
基金Supported by National Natural Science Foundation of China(Grant No.51835009).
文摘Gear fault diagnosis technologies have received rapid development and been effectively implemented in many engineering applications.However,the various working conditions would degrade the diagnostic performance and make gear fault diagnosis(GFD)more and more challenging.In this paper,a novel model parameter transfer(NMPT)is proposed to boost the performance of GFD under varying working conditions.Based on the previous transfer strategy that controls empirical risk of source domain,this method further integrates the superiorities of multi-task learning with the idea of transfer learning(TL)to acquire transferable knowledge by minimizing the discrepancies of separating hyperplanes between one specific working condition(target domain)and another(source domain),and then transferring both commonality and specialty parameters over tasks to make use of source domain samples to assist target GFD task when sufficient labeled samples from target domain are unavailable.For NMPT implementation,insufficient target domain features and abundant source domain features with supervised information are fed into NMPT model to train a robust classifier for target GFD task.Related experiments prove that NMPT is expected to be a valuable technology to boost practical GFD performance under various working conditions.The proposed methods provides a transfer learning-based framework to handle the problem of insufficient training samples in target task caused by variable operation conditions.
基金supported by the National Natural Science Foundation of China(Grant Nos.62072148 and U22A2033)the National Key R&D Program of China(Grant Nos.2022YFB3303000 and 2020YFB1709402)+2 种基金the Zhejiang Provincial Science and Technology Program in China(Grant No.2021C01108)the NSFC-Zhejiang Joint Fund for the Integration of Industrialization and Informatization(Grant No.U1909210)the Fundamental Research Funds for the Provincial Universities of Zhejiang(Grant No.490 GK219909299001-028).
文摘In this paper,we propose a parameterization transfer algorithm for planar domains bounded by B-spline curves,where the shapes of the planar domains are similar.The domain geometries are considered to be similar if their simplified skeletons have the same structures.One domain we call source domain,and it is parameterized using multi-patch B-spline surfaces.The resulting parameterization is C1 continuous in the regular region and G1 continuous around singular points regardless of whether the parameterization of the source domain is C1/G1 continuous or not.In this algorithm,boundary control points of the source domain are extracted from its parameterization as sequential points,and we establish a correspondence between sequential boundary control points of the source domain and the target boundary through discrete sampling and fitting.Transfer of the parametrization satisfies C1/G1 continuity under discrete harmonic mapping with continuous constraints.The new algorithm has a lower calculation cost than a decomposition-based parameterization with a high-quality parameterization result.We demonstrate that the result of the parameterization transfer in this paper can be applied in isogeometric analysis.Moreover,because of the consistency of the parameterization for the two models,this method can be applied in many other geometry processing algorithms,such as morphing and deformation.
基金supported by the RECARE Project from the European Union Seventh Framework Programme (FP7/2007-2013) (No. 603498)COST actions 1306
文摘Soil capacity to support life and to produce economic goods and services is strongly linked to the maintenance of good soil physical quality(SPQ). In this study, the SPQ of citrus orchards was assessed under three different soil managements, namely no-tillage using herbicides, tillage under chemical farming, and no-tillage under organic farming. Commonly used indicators, such as soil bulk density,organic carbon content, and structural stability index, were considered in conjunction with capacitive indicators estimated by the Beerkan estimation of soil transfer parameter(BEST) method. The measurements taken at the L'Alcoleja Experimental Station in Spain yielded optimal values for soil bulk density and organic carbon content in 100% and 70% of cases for organic farming. The values of structural stability index indicated that the soil was stable in 90% of cases. Differences between the soil management practices were particularly clear in terms of plant-available water capacity and saturated hydraulic conductivity. Under organic farming, the soil had the greatest ability to store and provide water to plant roots, and to quickly drain excess water and facilitate root proliferation.Management practices adopted under organic farming(such as vegetation cover between the trees, chipping after pruning, and spreading the chips on the soil surface) improved the SPQ. Conversely, the conventional management strategies unequivocally led to soil degradation owing to the loss of organic matter, soil compaction, and reduced structural stability. The results in this study show that organic farming has a clear positive impact on the SPQ, suggesting that tillage and herbicide treatments should be avoided.
基金The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University for funding this work through Large Groups Project under grant number(42/43)Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2022R151)Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.The authors would like to thank the Deanship of Scientific Research at Umm Al-Qura University for supporting this work by Grant Code:(22UQU4340237DSR16).
文摘Osteosarcoma is a type of malignant bone tumor that is reported across the globe.Recent advancements in Machine Learning(ML)and Deep Learning(DL)models enable the detection and classification of malignancies in biomedical images.In this regard,the current study introduces a new Biomedical Osteosarcoma Image Classification using Elephant Herd Optimization and Deep Transfer Learning(BOIC-EHODTL)model.The presented BOIC-EHODTL model examines the biomedical images to diagnose distinct kinds of osteosarcoma.At the initial stage,Gabor Filter(GF)is applied as a pre-processing technique to get rid of the noise from images.In addition,Adam optimizer with MixNet model is also employed as a feature extraction technique to generate feature vectors.Then,EHOalgorithm is utilized along with Adaptive Neuro-Fuzzy Classifier(ANFC)model for recognition and categorization of osteosarcoma.EHO algorithm is utilized to fine-tune the parameters involved in ANFC model which in turn helps in accomplishing improved classification results.The design of EHO with ANFC model for classification of osteosarcoma is the novelty of current study.In order to demonstrate the improved performance of BOIC-EHODTL model,a comprehensive comparison was conducted between the proposed and existing models upon benchmark dataset and the results confirmed the better performance of BOIC-EHODTL model over recent methodologies.
基金supported by National Basic Research Program of China (Grant No. 2010CB950803)National Natural Science Foundation of China (Grant No. 40971227)projects funded by the Inter-governmental Scientific and Technological Cooperation in Hungary
文摘Here we propose a new method of NDVI difference analysis and NDVI difference correction for multi-sensors to detect NDVI.This method integrate PROSPECT model,SAIL model and MODTRAN atmospheric radioactive transfer model to simulate the remote sensing signals of different satellites for both NDVI difference analysis and correction without using real satellite images.The effects of both the sensors' spectral responses and atmospheric condition are simulated,and the differences among NDVI values derived from thirty different satellites are analyzed quantitatively.Focusing on the conversions of NDVI values among different satellites,through linear regression analysis,we estimate the transfer parameters between any two different satellite NDVI values,and present the lookup tables of transfer parameters under the atmospheric conditions of three surface visibility range values (10,23 and 50 km).The proposed method is useful for NDVI applications and analyses for multi-sensors.
基金This work was supported in part by the National Nat- ural Science Foundation of China （Grant Nos. 61422204, 61473149 and 61671288）, Jiangsu Natural Science Foundation for Distinguished Young Scholar （BK20130034）, and Science and Technology Commission of Shang- hai Municipality （16JC1404300）.
文摘Protein subcellular localization prediction is im- portant for studying the function of proteins. Recently, as significant progress has been witnessed in the field of mi- croscopic imaging, automatically determining the subcellular localization of proteins from bio-images is becoming a new research hotspot. One of the central themes in this field is to determine what features are suitable for describing the pro- tein images. Existing feature extraction methods are usually hand-crafted designed, by which only one layer of features will be extracted, which may not be sufficient to represent the complex protein images. To this end, we propose a deep model based descriptor （DMD） to extract the high-level fea- tures from protein images. Specifically, in order to make the extracted features more generic, we firstly trained a convolu- tion neural network （i.e., AlexNe0 by using a natural image set with millions of labels, and then used the partial parame- ter transfer strategy to fine-tnne the parameters from natural images to protein images. After that, we applied the Lasso model to select the most distinguishing features from the last fully connected layer of the CNN （Convolution Neural Net- work）, and used these selected features for final classifica- tions. Experimental results on a protein image dataset vali- date the efficacy of our method.
