To determine the fractional vegetation cover (FVC ) and associated driving factors of modeling in... more To determine the fractional vegetation cover (FVC ) and associated driving factors of modeling in mining areas, six types of data were used as driving factors and three methods—multi-linear regression (MLR ), geographically weighted regression (GWR ), and geographically weighted artificial neural network (GWANN )—were adopted in the modeling. The experiments, conducted in Shengli mining areas located in Xilinhot city, China, show that the MLR model without consideration of spatial heterogeneity and spatial non-stationarity performs the worst and that the GWR model presents obvious location differences, since it predefines a linear relationship which is unable to describe FVC for some locations. The GWANN model, improving on these defects, is the most suitable model for the FVC driving process in mining areas; it outperforms the other two models, with root-mean-square error (RMSE ) and mean absolute percentage error (MAPE ) reaching 0.16 and 0.20. It has improvements of approximately...
Proceedings of the AAAI Conference on Artificial Intelligence
In recent years, self-supervised representation learning for skeleton-based action recognition ha... more In recent years, self-supervised representation learning for skeleton-based action recognition has been developed with the advance of contrastive learning methods. The existing contrastive learning methods use normal augmentations to construct similar positive samples, which limits the ability to explore novel movement patterns. In this paper, to make better use of the movement patterns introduced by extreme augmentations, a Contrastive Learning framework utilizing Abundant Information Mining for self-supervised action Representation (AimCLR) is proposed. First, the extreme augmentations and the Energy-based Attention-guided Drop Module (EADM) are proposed to obtain diverse positive samples, which bring novel movement patterns to improve the universality of the learned representations. Second, since directly using extreme augmentations may not be able to boost the performance due to the drastic changes in original identity, the Dual Distributional Divergence Minimization Loss (D3M L...
Proceedings of the AAAI Conference on Artificial Intelligence
Humans and animals learn much better when the examples are not randomly presented but organized i... more Humans and animals learn much better when the examples are not randomly presented but organized in a meaningful order which illustrates gradually more concepts, and gradually more complex ones. Inspired by this curriculum learning mechanism, we propose a reinforced multi-label image classification approach imitating human behavior to label image from easy to complex. This approach allows a reinforcement learning agent to sequentially predict labels by fully exploiting image feature and previously predicted labels. The agent discovers the optimal policies through maximizing the long-term reward which reflects prediction accuracies. Experimental results on PASCAL VOC2007 and 2012 demonstrate the necessity of reinforcement multi-label learning and the algorithm’s effectiveness in real-world multi-label image classification tasks.
Abstract An integrated modelling and operation method for a multienergy system (MES) is proposed ... more Abstract An integrated modelling and operation method for a multienergy system (MES) is proposed in this paper. First, a coupling matrix equation containing the energy flow in the process of energy production, transmission, conversion, storage and consumption is established. Second, the whole MES is selected as the game subject, and each energy subsystem is selected as the game follower. Then, an integrated model of an MES is modelled to quantify the complementary cold-heat-power-gas multi-energy and source-network-load-storage coordinated interactions based on Stackelberg game theory. Third, the day-ahead MES operation scheme is optimized based on the established model and the equilibrium solution is used to realize a reasonable balance of benefits between the whole MES and its energy subsystems. Numerical studies demonstrate that the proposed method increases the operation cost of the whole MES by ¥68.23 (0.917% increase) but reduces the operation cost of the heat and gas subsystems by ¥254.82 (3.29% decrease) and ¥289.4 (3.72%), respectively, with the objective of minimizing operation cost, and improves the whole system exergy efficiency of the power, heat and gas subsystems by 0.572%, 0.548% and 2.076%, respectively, with the objective of maximizing the exergy efficiency. Thus, one can take into account the different benefits among the whole MES and its energy subsystems and provide a multidimensional dispatch scheme for dispatchers, highlighting the potential for MES development.
Abstract An intraday, multi-objective, hierarchical and coordinated operation scheduling method f... more Abstract An intraday, multi-objective, hierarchical and coordinated operation scheduling method for a multi-energy system (MES), which uses 15-min and 5-min scheduling intervals for different energy subsystems, is proposed. According to the characteristics of MES and the response time of energy conversion equipment, energy subsystems are dispatched on different dispatch intervals instead of unified dispatch intervals to dispatch energy subsystems. In a case study, the prediction error rate for a load is 5% and 2% when the dispatch interval is 15 min and 5 min, respectively; the prediction error rates for wind and solar energy output are 10% and 5%, respectively. Dispatching different subsystems with different intervals according to their characteristics reduces the impact of source and load uncertainties, which improves energy management. Multi-energy power dispatch considers exergy efficiency and the operation cost to improve the utilization energy efficiency. The Tchebycheff method, which considers the fuzzy entropy weight, is employed to balance the objectives between highest exergy efficiency and lowest operation cost. Numerical studies demonstrate that the operation cost of a multi-objective system is ¥500.8877 higher than that of a system with the objective of lowest operation cost. The exergy efficiency is increased by 3.94%, and the operation cost, which is reduced by ¥1706.9606, is 2.13% lower than that for the objective of highest exergy efficiency. Thus, the proposed method can balance the objectives between highest exergy efficiency and lowest operation cost. These findings provide a multi-dimensional dispatch scheme for dispatchers and highlight the development potential of an MES.
N‐substituted bismuth selenide (BSN) material is synthesized by a one‐step solvothermal route and... more N‐substituted bismuth selenide (BSN) material is synthesized by a one‐step solvothermal route and used for the photocatalytic application. As‐prepared BSN consists of ultrathin nanosheets and has a quasi hexagonal morphology. The results show the superior performance of BSN‐2 for degradation of multi‐pollutants and the ciprofloxacin and phenol removal rates reach 88.8 and 83.6%, respectively, under irradiation of simulated solar for 2 h, which is much higher than those over pristine bismuth selenide. The significant improvement of photocatalytic performance is attributed to ultrathin nanosheets structure and N introduction that contribute to the inhibition of photoelectron‐hole pair recombination and the enhancement of light absorption. The possible mechanism is proposed based on characterization and theoretical calculation of the energy band structure. The results indicate that BSN‐2 is a high‐efficient photocatalyst for the removal of multi‐organic pollutants in wastewater and this research provides a simple strategy for reasonable design and synthesis of Bi‐based materials for environmental remediation.
The segment from positions 280 to 283 in BphAEs is located at the entrance of the catalytic pocke... more The segment from positions 280 to 283 in BphAEs is located at the entrance of the catalytic pocket, and it shows variation in conformation. In previous works, results have suggested but never proved that residue Ser283 of BphAE LB400 might play a role in substrate specificity. In the present paper, we found that the Ser283Met substitution significantly increased the specificity of the reaction of BphAE toward biphenyl, 2,3′,4,4′-CB, 2,2′,6,6′-CB, and 2,3′,4,4′,5-CB. Meanwhile, the Ser283Met substitution altered the regiospecificity of BphAE toward 2,2′-dichlorobiphenyl and 2,6-dichlorobiphenyl. Additionally, this substitution extended the range of PCBs metabolized by the mutated BphAE. BphAE S283M and BphAE p4-S283M were clearly improved in oxidizing some of the more highly chlorinated biphenyls (3 to 6 chlorines), which are generally very poorly oxidized by most dioxygenases. We used modeled and docked enzymes to identify some of the structural features that explain the new propert...
Abstract Nowadays, the integrated energy system (IES) has become a hot topic in the field of ener... more Abstract Nowadays, the integrated energy system (IES) has become a hot topic in the field of energy research. In this paper, a bi-level optimal configuration strategy is proposed for a community integrated energy system (CIES), which is based on energy supply-demand responses and robustness adjustable scenarios. First, from the perspective of energy supply and demand, various energy equipment on the community side and aggregated energy loads on the user side are analysed and qualitatively modeled, and a multi energy supply-demand of CIES is established. Next, according to the robustness requirements of IES planning, the scenario sampling, sorting, screening and reduction are performed subsequently to obtain a typical set of robustness adjustable scenarios. On this basis, a bi-level optimal configuration model that coordinates the aggregated configuration and operation is developed to design the CIES. While the upper-level model takes the lowest total annual cost as the goal to configure the quantity and capacity of energy equipment, the lower-level model optimizes the scheduling scheme with the aim of the best operation economy under typical scenarios. Finally, case studies are carried out based on a practical town area, and simulation results show the effectiveness and advantage of the proposed strategy.
In the era of big data, the cost control of enterprises is important, and the cost is still the m... more In the era of big data, the cost control of enterprises is important, and the cost is still the main factor affecting the competitiveness of enterprises. Data information occurs in every part of an enterprise. If these data are collected reasonably and analyzed with professional software system, the effect of enterprise cost control could be quantified. This paper takes Tianbao green food company as an example, studies its cost control situation, summarizes the existing problems in cost control of the company, and analyzes how to use big data to control cost for purchasing, storage, producing and sales.
Circulating tumor cells (CTCs), with their close association with cancer metastasis, the most agg... more Circulating tumor cells (CTCs), with their close association with cancer metastasis, the most aggressive feature of solid tumors, represent an important aspect of "liquid biopsy," which provides minimally- or noninvasive approaches for cancer detection and disease status monitoring. CTC analysis has shown the potential clinical applications in several cancer types and has been approved by FDA for clinical use in advanced breast, prostate, and colorectal cancer prognosis. In this chapter, we describe a CTC isolation method using a cell size and deformability-based system, Parsortix, and the immunofluorescence staining method to detect CTCs with both epithelial and mesenchymal features. We also describe a repeated fluorescence in situ hybridization (FISH) approach to detect alterations of multiple genomic regions on the same CTCs after immunofluorescence analysis. This approach allows the study of CTCs as a biomarker for cancer detection, prognosis, and therapeutic response ...
Abstract Production of hydrogen from electrochemical water splitting has been regarded as one of ... more Abstract Production of hydrogen from electrochemical water splitting has been regarded as one of the most economic and sustainable techniques for green fuel production. It is significant and challengeable to develop highly efficient and low cost noble metal-free electrocatalysts. Presently, molybdenum-based electrocatalysts were regarded as potential alternatives for the hydrogen evolution reaction (HER). Here, the well-dispersed and ultrasmall Mo2C nanoparticles (NPs) anchored on 2D carbon nanosheets were synthesized by designing chelate precursor and following pyrolysis, which was proved to be an effective approach for preparing carbon-loaded Mo2C NPs. The as-obtained Mo2C/C material exhibits an outstanding activity and stability in hydrogen evolution reaction (HER). It needs an overpotential of 147 mV to drive 10 mA cm−2 and Tafel slope is 64.2 mV dec−1 in alkaline medium, implying that Mo2C/C material will be a potential noble metal-free electrocatalyst for HER. The design of Mo-chelate precursor is a feasible route to synthesize ultrafine Mo2C and it can provide a reference for synthesizing other nanoparticles and hindering particle coalescence at high preparation temperature.
To determine the fractional vegetation cover (FVC ) and associated driving factors of modeling in... more To determine the fractional vegetation cover (FVC ) and associated driving factors of modeling in mining areas, six types of data were used as driving factors and three methods—multi-linear regression (MLR ), geographically weighted regression (GWR ), and geographically weighted artificial neural network (GWANN )—were adopted in the modeling. The experiments, conducted in Shengli mining areas located in Xilinhot city, China, show that the MLR model without consideration of spatial heterogeneity and spatial non-stationarity performs the worst and that the GWR model presents obvious location differences, since it predefines a linear relationship which is unable to describe FVC for some locations. The GWANN model, improving on these defects, is the most suitable model for the FVC driving process in mining areas; it outperforms the other two models, with root-mean-square error (RMSE ) and mean absolute percentage error (MAPE ) reaching 0.16 and 0.20. It has improvements of approximately...
Proceedings of the AAAI Conference on Artificial Intelligence
In recent years, self-supervised representation learning for skeleton-based action recognition ha... more In recent years, self-supervised representation learning for skeleton-based action recognition has been developed with the advance of contrastive learning methods. The existing contrastive learning methods use normal augmentations to construct similar positive samples, which limits the ability to explore novel movement patterns. In this paper, to make better use of the movement patterns introduced by extreme augmentations, a Contrastive Learning framework utilizing Abundant Information Mining for self-supervised action Representation (AimCLR) is proposed. First, the extreme augmentations and the Energy-based Attention-guided Drop Module (EADM) are proposed to obtain diverse positive samples, which bring novel movement patterns to improve the universality of the learned representations. Second, since directly using extreme augmentations may not be able to boost the performance due to the drastic changes in original identity, the Dual Distributional Divergence Minimization Loss (D3M L...
Proceedings of the AAAI Conference on Artificial Intelligence
Humans and animals learn much better when the examples are not randomly presented but organized i... more Humans and animals learn much better when the examples are not randomly presented but organized in a meaningful order which illustrates gradually more concepts, and gradually more complex ones. Inspired by this curriculum learning mechanism, we propose a reinforced multi-label image classification approach imitating human behavior to label image from easy to complex. This approach allows a reinforcement learning agent to sequentially predict labels by fully exploiting image feature and previously predicted labels. The agent discovers the optimal policies through maximizing the long-term reward which reflects prediction accuracies. Experimental results on PASCAL VOC2007 and 2012 demonstrate the necessity of reinforcement multi-label learning and the algorithm’s effectiveness in real-world multi-label image classification tasks.
Abstract An integrated modelling and operation method for a multienergy system (MES) is proposed ... more Abstract An integrated modelling and operation method for a multienergy system (MES) is proposed in this paper. First, a coupling matrix equation containing the energy flow in the process of energy production, transmission, conversion, storage and consumption is established. Second, the whole MES is selected as the game subject, and each energy subsystem is selected as the game follower. Then, an integrated model of an MES is modelled to quantify the complementary cold-heat-power-gas multi-energy and source-network-load-storage coordinated interactions based on Stackelberg game theory. Third, the day-ahead MES operation scheme is optimized based on the established model and the equilibrium solution is used to realize a reasonable balance of benefits between the whole MES and its energy subsystems. Numerical studies demonstrate that the proposed method increases the operation cost of the whole MES by ¥68.23 (0.917% increase) but reduces the operation cost of the heat and gas subsystems by ¥254.82 (3.29% decrease) and ¥289.4 (3.72%), respectively, with the objective of minimizing operation cost, and improves the whole system exergy efficiency of the power, heat and gas subsystems by 0.572%, 0.548% and 2.076%, respectively, with the objective of maximizing the exergy efficiency. Thus, one can take into account the different benefits among the whole MES and its energy subsystems and provide a multidimensional dispatch scheme for dispatchers, highlighting the potential for MES development.
Abstract An intraday, multi-objective, hierarchical and coordinated operation scheduling method f... more Abstract An intraday, multi-objective, hierarchical and coordinated operation scheduling method for a multi-energy system (MES), which uses 15-min and 5-min scheduling intervals for different energy subsystems, is proposed. According to the characteristics of MES and the response time of energy conversion equipment, energy subsystems are dispatched on different dispatch intervals instead of unified dispatch intervals to dispatch energy subsystems. In a case study, the prediction error rate for a load is 5% and 2% when the dispatch interval is 15 min and 5 min, respectively; the prediction error rates for wind and solar energy output are 10% and 5%, respectively. Dispatching different subsystems with different intervals according to their characteristics reduces the impact of source and load uncertainties, which improves energy management. Multi-energy power dispatch considers exergy efficiency and the operation cost to improve the utilization energy efficiency. The Tchebycheff method, which considers the fuzzy entropy weight, is employed to balance the objectives between highest exergy efficiency and lowest operation cost. Numerical studies demonstrate that the operation cost of a multi-objective system is ¥500.8877 higher than that of a system with the objective of lowest operation cost. The exergy efficiency is increased by 3.94%, and the operation cost, which is reduced by ¥1706.9606, is 2.13% lower than that for the objective of highest exergy efficiency. Thus, the proposed method can balance the objectives between highest exergy efficiency and lowest operation cost. These findings provide a multi-dimensional dispatch scheme for dispatchers and highlight the development potential of an MES.
N‐substituted bismuth selenide (BSN) material is synthesized by a one‐step solvothermal route and... more N‐substituted bismuth selenide (BSN) material is synthesized by a one‐step solvothermal route and used for the photocatalytic application. As‐prepared BSN consists of ultrathin nanosheets and has a quasi hexagonal morphology. The results show the superior performance of BSN‐2 for degradation of multi‐pollutants and the ciprofloxacin and phenol removal rates reach 88.8 and 83.6%, respectively, under irradiation of simulated solar for 2 h, which is much higher than those over pristine bismuth selenide. The significant improvement of photocatalytic performance is attributed to ultrathin nanosheets structure and N introduction that contribute to the inhibition of photoelectron‐hole pair recombination and the enhancement of light absorption. The possible mechanism is proposed based on characterization and theoretical calculation of the energy band structure. The results indicate that BSN‐2 is a high‐efficient photocatalyst for the removal of multi‐organic pollutants in wastewater and this research provides a simple strategy for reasonable design and synthesis of Bi‐based materials for environmental remediation.
The segment from positions 280 to 283 in BphAEs is located at the entrance of the catalytic pocke... more The segment from positions 280 to 283 in BphAEs is located at the entrance of the catalytic pocket, and it shows variation in conformation. In previous works, results have suggested but never proved that residue Ser283 of BphAE LB400 might play a role in substrate specificity. In the present paper, we found that the Ser283Met substitution significantly increased the specificity of the reaction of BphAE toward biphenyl, 2,3′,4,4′-CB, 2,2′,6,6′-CB, and 2,3′,4,4′,5-CB. Meanwhile, the Ser283Met substitution altered the regiospecificity of BphAE toward 2,2′-dichlorobiphenyl and 2,6-dichlorobiphenyl. Additionally, this substitution extended the range of PCBs metabolized by the mutated BphAE. BphAE S283M and BphAE p4-S283M were clearly improved in oxidizing some of the more highly chlorinated biphenyls (3 to 6 chlorines), which are generally very poorly oxidized by most dioxygenases. We used modeled and docked enzymes to identify some of the structural features that explain the new propert...
Abstract Nowadays, the integrated energy system (IES) has become a hot topic in the field of ener... more Abstract Nowadays, the integrated energy system (IES) has become a hot topic in the field of energy research. In this paper, a bi-level optimal configuration strategy is proposed for a community integrated energy system (CIES), which is based on energy supply-demand responses and robustness adjustable scenarios. First, from the perspective of energy supply and demand, various energy equipment on the community side and aggregated energy loads on the user side are analysed and qualitatively modeled, and a multi energy supply-demand of CIES is established. Next, according to the robustness requirements of IES planning, the scenario sampling, sorting, screening and reduction are performed subsequently to obtain a typical set of robustness adjustable scenarios. On this basis, a bi-level optimal configuration model that coordinates the aggregated configuration and operation is developed to design the CIES. While the upper-level model takes the lowest total annual cost as the goal to configure the quantity and capacity of energy equipment, the lower-level model optimizes the scheduling scheme with the aim of the best operation economy under typical scenarios. Finally, case studies are carried out based on a practical town area, and simulation results show the effectiveness and advantage of the proposed strategy.
In the era of big data, the cost control of enterprises is important, and the cost is still the m... more In the era of big data, the cost control of enterprises is important, and the cost is still the main factor affecting the competitiveness of enterprises. Data information occurs in every part of an enterprise. If these data are collected reasonably and analyzed with professional software system, the effect of enterprise cost control could be quantified. This paper takes Tianbao green food company as an example, studies its cost control situation, summarizes the existing problems in cost control of the company, and analyzes how to use big data to control cost for purchasing, storage, producing and sales.
Circulating tumor cells (CTCs), with their close association with cancer metastasis, the most agg... more Circulating tumor cells (CTCs), with their close association with cancer metastasis, the most aggressive feature of solid tumors, represent an important aspect of "liquid biopsy," which provides minimally- or noninvasive approaches for cancer detection and disease status monitoring. CTC analysis has shown the potential clinical applications in several cancer types and has been approved by FDA for clinical use in advanced breast, prostate, and colorectal cancer prognosis. In this chapter, we describe a CTC isolation method using a cell size and deformability-based system, Parsortix, and the immunofluorescence staining method to detect CTCs with both epithelial and mesenchymal features. We also describe a repeated fluorescence in situ hybridization (FISH) approach to detect alterations of multiple genomic regions on the same CTCs after immunofluorescence analysis. This approach allows the study of CTCs as a biomarker for cancer detection, prognosis, and therapeutic response ...
Abstract Production of hydrogen from electrochemical water splitting has been regarded as one of ... more Abstract Production of hydrogen from electrochemical water splitting has been regarded as one of the most economic and sustainable techniques for green fuel production. It is significant and challengeable to develop highly efficient and low cost noble metal-free electrocatalysts. Presently, molybdenum-based electrocatalysts were regarded as potential alternatives for the hydrogen evolution reaction (HER). Here, the well-dispersed and ultrasmall Mo2C nanoparticles (NPs) anchored on 2D carbon nanosheets were synthesized by designing chelate precursor and following pyrolysis, which was proved to be an effective approach for preparing carbon-loaded Mo2C NPs. The as-obtained Mo2C/C material exhibits an outstanding activity and stability in hydrogen evolution reaction (HER). It needs an overpotential of 147 mV to drive 10 mA cm−2 and Tafel slope is 64.2 mV dec−1 in alkaline medium, implying that Mo2C/C material will be a potential noble metal-free electrocatalyst for HER. The design of Mo-chelate precursor is a feasible route to synthesize ultrafine Mo2C and it can provide a reference for synthesizing other nanoparticles and hindering particle coalescence at high preparation temperature.
Uploads
Papers by tianyu guo