In this study, a xylanase-produeing Aspergillus niger strain, NS-1, was screened and isolated from agricultural and forestry wastes. Based on single-fac- tor experiments, the effects of different carbon sources, compo...In this study, a xylanase-produeing Aspergillus niger strain, NS-1, was screened and isolated from agricultural and forestry wastes. Based on single-fac- tor experiments, the effects of different carbon sources, composite carbon sources, nitrogen sources, calcium carbonate concentrations, initial pH and surfactants on xylanase production by A. niger NS-1 were investigated. The results indicated that the most appropriate carbon source was corncobs ; the best composite carbon source was corncobs ＋ xylan, which was conducive to xylanase secretion; the most suitable nitrogen source was ammonium sulfate. Xylanase activity reached the highest in the medium added with 1.5% calcium carbonate and SDS as a surfactant with an initial pH of 5.0. This study provided the basis for the production of high-activity xylanase.展开更多
Based on exact penalty function, a new neural network for solving the L1-norm optimization problem is proposed. In comparison with Kennedy and Chua’s network(1988), it has better properties.Based on Bandler’s fault ...Based on exact penalty function, a new neural network for solving the L1-norm optimization problem is proposed. In comparison with Kennedy and Chua’s network(1988), it has better properties.Based on Bandler’s fault location method(1982), a new nonlinearly constrained L1-norm problem is developed. It can be solved with less computing time through only one optimization processing. The proposed neural network can be used to solve the analog diagnosis L1 problem. The validity of the proposed neural networks and the fault location L1 method are illustrated by extensive computer simulations.展开更多
A new trust region algorithm for solving convex LC 1 optimization problem is presented.It is proved that the algorithm is globally convergent and the rate of convergence is superlinear under some reasonable assum...A new trust region algorithm for solving convex LC 1 optimization problem is presented.It is proved that the algorithm is globally convergent and the rate of convergence is superlinear under some reasonable assumptions.展开更多
In this paper, a new trust region algorithm for unconstrained LC1 optimization problems is given. Compare with those existing trust regiion methods, this algorithm has a different feature: it obtains a stepsize at eac...In this paper, a new trust region algorithm for unconstrained LC1 optimization problems is given. Compare with those existing trust regiion methods, this algorithm has a different feature: it obtains a stepsize at each iteration not by soloving a quadratic subproblem with a trust region bound, but by solving a system of linear equations. Thus it reduces computational complexity and improves computation efficiency. It is proven that this algorithm is globally convergent and locally superlinear under some conditions.展开更多
In this paper, a new trust region algorithm for nonlinear equality constrained LC1 optimization problems is given. It obtains a search direction at each iteration not by solving a quadratic programming subprobiem with...In this paper, a new trust region algorithm for nonlinear equality constrained LC1 optimization problems is given. It obtains a search direction at each iteration not by solving a quadratic programming subprobiem with a trust region bound, but by solving a system of linear equations. Since the computational complexity of a QP-Problem is in general much larger than that of a system of linear equations, this method proposed in this paper may reduce the computational complexity and hence improve computational efficiency. Furthermore, it is proved under appropriate assumptions that this algorithm is globally and super-linearly convergent to a solution of the original problem. Some numerical examples are reported, showing the proposed algorithm can be beneficial from a computational point of view.展开更多
This paper presents a novelmulticlass systemdesigned to detect pleural effusion and pulmonary edema on chest Xray images,addressing the critical need for early detection in healthcare.A new comprehensive dataset was f...This paper presents a novelmulticlass systemdesigned to detect pleural effusion and pulmonary edema on chest Xray images,addressing the critical need for early detection in healthcare.A new comprehensive dataset was formed by combining 28,309 samples from the ChestX-ray14,PadChest,and CheXpert databases,with 10,287,6022,and 12,000 samples representing Pleural Effusion,Pulmonary Edema,and Normal cases,respectively.Consequently,the preprocessing step involves applying the Contrast Limited Adaptive Histogram Equalization(CLAHE)method to boost the local contrast of the X-ray samples,then resizing the images to 380×380 dimensions,followed by using the data augmentation technique.The classification task employs a deep learning model based on the EfficientNet-V1-B4 architecture and is trained using the AdamW optimizer.The proposed multiclass system achieved an accuracy(ACC)of 98.3%,recall of 98.3%,precision of 98.7%,and F1-score of 98.7%.Moreover,the robustness of the model was revealed by the Receiver Operating Characteristic(ROC)analysis,which demonstrated an Area Under the Curve(AUC)of 1.00 for edema and normal cases and 0.99 for effusion.The experimental results demonstrate the superiority of the proposedmulti-class system,which has the potential to assist clinicians in timely and accurate diagnosis,leading to improved patient outcomes.Notably,ablation-CAM visualization at the last convolutional layer portrayed further enhanced diagnostic capabilities with heat maps on X-ray images,which will aid clinicians in interpreting and localizing abnormalities more effectively.展开更多
A methodology for performance optimization of torque converters is put forward based on the one-dimensional (1D) flow model. It is found that the inaccuracy of 1D flow model for predicting hydraulic performance at the...A methodology for performance optimization of torque converters is put forward based on the one-dimensional (1D) flow model. It is found that the inaccuracy of 1D flow model for predicting hydraulic performance at the low speed ratio is mainly caused by the separation phenomenon at the stator cascade which is induced by large flow impinging at the pressure side of the stator blades. A semi-empirical separation model is presented and incorporated to the original 1D flow model. It is illustrated that the improved model is able to predict the circumferential velocity components accurately, which can be applied to performance optimization. Then, the Pareto front is obtained by using the genetic algorithm (GA) in order to inspect the coupled relationship among stalling impeller torque capacity, stalling torque ratio and efficiency. The efficiency is maximized on the premise that a target stalling impeller torque capacity and torque ratio are achieved. Finally, the optimized result is verified by the computational fluid dynamics(CFD) simulation, which indicates that the maximal efficiency is increased by 0.96%.展开更多
In this paper we present a filter-trust-region algorithm for solving LC1 unconstrained optimization problems which uses the second Dini upper directional derivative. We establish the global convergence of the algorith...In this paper we present a filter-trust-region algorithm for solving LC1 unconstrained optimization problems which uses the second Dini upper directional derivative. We establish the global convergence of the algorithm under reasonable assumptions.展开更多
In recent years,wearable devices-based Human Activity Recognition(HAR)models have received significant attention.Previously developed HAR models use hand-crafted features to recognize human activities,leading to the e...In recent years,wearable devices-based Human Activity Recognition(HAR)models have received significant attention.Previously developed HAR models use hand-crafted features to recognize human activities,leading to the extraction of basic features.The images captured by wearable sensors contain advanced features,allowing them to be analyzed by deep learning algorithms to enhance the detection and recognition of human actions.Poor lighting and limited sensor capabilities can impact data quality,making the recognition of human actions a challenging task.The unimodal-based HAR approaches are not suitable in a real-time environment.Therefore,an updated HAR model is developed using multiple types of data and an advanced deep-learning approach.Firstly,the required signals and sensor data are accumulated from the standard databases.From these signals,the wave features are retrieved.Then the extracted wave features and sensor data are given as the input to recognize the human activity.An Adaptive Hybrid Deep Attentive Network(AHDAN)is developed by incorporating a“1D Convolutional Neural Network(1DCNN)”with a“Gated Recurrent Unit(GRU)”for the human activity recognition process.Additionally,the Enhanced Archerfish Hunting Optimizer(EAHO)is suggested to fine-tune the network parameters for enhancing the recognition process.An experimental evaluation is performed on various deep learning networks and heuristic algorithms to confirm the effectiveness of the proposed HAR model.The EAHO-based HAR model outperforms traditional deep learning networks with an accuracy of 95.36,95.25 for recall,95.48 for specificity,and 95.47 for precision,respectively.The result proved that the developed model is effective in recognizing human action by taking less time.Additionally,it reduces the computation complexity and overfitting issue through using an optimization approach.展开更多
The mixed l1/H2 optimization problem for MIMO (multiple input-multiple output) discrete-time systems is considered. This problem is formulated as minimizing the l1-norm of a closed-loop transfer matrix while maintaini...The mixed l1/H2 optimization problem for MIMO (multiple input-multiple output) discrete-time systems is considered. This problem is formulated as minimizing the l1-norm of a closed-loop transfer matrix while maintaining the H2-norm of another closed-loop transfer matrix at prescribed level. The continuity property of the optimal value in respect to changes in the H2-norm constraint is studied. The existence of the optimal solutions of mixed l1/H2 problem is proved. Because the solution of the mixed l1/H2 problem is based on the scaled-Q method, it avoids the zero interpolation difficulties. The convergent upper and lower bounds can be obtained by solving a sequence of finite dimensional nonlinear programming for which many efficient numerical optimization algorithms exist.展开更多
[Objectives]The extraction conditions of flavonoids from flowers of Edgeworthia chrysantha Lind1.were optimized.[Methods]With ethanol as an extraction agent,firstly by single-factor tests,the best levels of ethanol co...[Objectives]The extraction conditions of flavonoids from flowers of Edgeworthia chrysantha Lind1.were optimized.[Methods]With ethanol as an extraction agent,firstly by single-factor tests,the best levels of ethanol concentration,material-to-liquid ratio,extraction temperature and extraction time were determined.Then the optimal extraction conditions were determined through the quadratic orthogonal regression rotatable design.[Results]The optimal extraction conditions for flavonoids from E.chrysantha were∶material-to-liquid ratio 1∶15.4,extraction temperature 75℃,extraction time 120 min and ethanol concentration 75%,with which the highest yield of flavonoids was 6.02 mg/g.[Conclusions]This study provides a theoretical basis for the further development and utilization of E.chrysantha flowers.展开更多
This paper present a simulation study of an evolutionary algorithms, Particle Swarm Optimization PSO algorithm to optimize likelihood function of ARMA（1, 1） model, where maximizing likelihood function is equivalent ...This paper present a simulation study of an evolutionary algorithms, Particle Swarm Optimization PSO algorithm to optimize likelihood function of ARMA（1, 1） model, where maximizing likelihood function is equivalent to maximizing its logarithm, so the objective function ＇obj.fun＇ is maximizing log-likelihood function. Monte Carlo method adapted for implementing and designing the experiments of this simulation. This study including a comparison among three versions of PSO algorithm “Constriction coefficient CCPSO, Inertia weight IWPSO, and Fully Informed FIPSO”, the experiments designed by setting different values of model parameters al, bs sample size n, moreover the parameters of PSO algorithms. MSE used as test statistic to measure the efficiency PSO to estimate model. The results show the ability of PSO to estimate ARMA＇ s parameters, and the minimum values of MSE getting for COPSO.展开更多
基金Supported by Project of Taizhou Science and Technology Bureau(TS019)
文摘In this study, a xylanase-produeing Aspergillus niger strain, NS-1, was screened and isolated from agricultural and forestry wastes. Based on single-fac- tor experiments, the effects of different carbon sources, composite carbon sources, nitrogen sources, calcium carbonate concentrations, initial pH and surfactants on xylanase production by A. niger NS-1 were investigated. The results indicated that the most appropriate carbon source was corncobs ; the best composite carbon source was corncobs ＋ xylan, which was conducive to xylanase secretion; the most suitable nitrogen source was ammonium sulfate. Xylanase activity reached the highest in the medium added with 1.5% calcium carbonate and SDS as a surfactant with an initial pH of 5.0. This study provided the basis for the production of high-activity xylanase.
基金Supported by Doctoral Special Fund of State Education Commissionthe National Natural Science Foundation of China,Grant No.59477001 and No.59707002
文摘Based on exact penalty function, a new neural network for solving the L1-norm optimization problem is proposed. In comparison with Kennedy and Chua’s network(1988), it has better properties.Based on Bandler’s fault location method(1982), a new nonlinearly constrained L1-norm problem is developed. It can be solved with less computing time through only one optimization processing. The proposed neural network can be used to solve the analog diagnosis L1 problem. The validity of the proposed neural networks and the fault location L1 method are illustrated by extensive computer simulations.
基金Supported by the National Natural Science Foundation of P.R.China(1 9971 0 0 2 ) and the Subject ofBeijing Educational Committ
文摘A new trust region algorithm for solving convex LC 1 optimization problem is presented.It is proved that the algorithm is globally convergent and the rate of convergence is superlinear under some reasonable assumptions.
文摘In this paper, a new trust region algorithm for unconstrained LC1 optimization problems is given. Compare with those existing trust regiion methods, this algorithm has a different feature: it obtains a stepsize at each iteration not by soloving a quadratic subproblem with a trust region bound, but by solving a system of linear equations. Thus it reduces computational complexity and improves computation efficiency. It is proven that this algorithm is globally convergent and locally superlinear under some conditions.
文摘In this paper, a new trust region algorithm for nonlinear equality constrained LC1 optimization problems is given. It obtains a search direction at each iteration not by solving a quadratic programming subprobiem with a trust region bound, but by solving a system of linear equations. Since the computational complexity of a QP-Problem is in general much larger than that of a system of linear equations, this method proposed in this paper may reduce the computational complexity and hence improve computational efficiency. Furthermore, it is proved under appropriate assumptions that this algorithm is globally and super-linearly convergent to a solution of the original problem. Some numerical examples are reported, showing the proposed algorithm can be beneficial from a computational point of view.
文摘This paper presents a novelmulticlass systemdesigned to detect pleural effusion and pulmonary edema on chest Xray images,addressing the critical need for early detection in healthcare.A new comprehensive dataset was formed by combining 28,309 samples from the ChestX-ray14,PadChest,and CheXpert databases,with 10,287,6022,and 12,000 samples representing Pleural Effusion,Pulmonary Edema,and Normal cases,respectively.Consequently,the preprocessing step involves applying the Contrast Limited Adaptive Histogram Equalization(CLAHE)method to boost the local contrast of the X-ray samples,then resizing the images to 380×380 dimensions,followed by using the data augmentation technique.The classification task employs a deep learning model based on the EfficientNet-V1-B4 architecture and is trained using the AdamW optimizer.The proposed multiclass system achieved an accuracy(ACC)of 98.3%,recall of 98.3%,precision of 98.7%,and F1-score of 98.7%.Moreover,the robustness of the model was revealed by the Receiver Operating Characteristic(ROC)analysis,which demonstrated an Area Under the Curve(AUC)of 1.00 for edema and normal cases and 0.99 for effusion.The experimental results demonstrate the superiority of the proposedmulti-class system,which has the potential to assist clinicians in timely and accurate diagnosis,leading to improved patient outcomes.Notably,ablation-CAM visualization at the last convolutional layer portrayed further enhanced diagnostic capabilities with heat maps on X-ray images,which will aid clinicians in interpreting and localizing abnormalities more effectively.
基金National Natural Science Foundation of China(No. 51175379)
文摘A methodology for performance optimization of torque converters is put forward based on the one-dimensional (1D) flow model. It is found that the inaccuracy of 1D flow model for predicting hydraulic performance at the low speed ratio is mainly caused by the separation phenomenon at the stator cascade which is induced by large flow impinging at the pressure side of the stator blades. A semi-empirical separation model is presented and incorporated to the original 1D flow model. It is illustrated that the improved model is able to predict the circumferential velocity components accurately, which can be applied to performance optimization. Then, the Pareto front is obtained by using the genetic algorithm (GA) in order to inspect the coupled relationship among stalling impeller torque capacity, stalling torque ratio and efficiency. The efficiency is maximized on the premise that a target stalling impeller torque capacity and torque ratio are achieved. Finally, the optimized result is verified by the computational fluid dynamics(CFD) simulation, which indicates that the maximal efficiency is increased by 0.96%.
基金Supported by CERG: CityU 101005 of the Government of Hong Kong SAR, Chinathe National Natural ScienceFoundation of China, the Specialized Research Fund of Doctoral Program of Higher Education of China (Grant No.20040319003)the Natural Science Fund of Jiangsu Province of China (Grant No. BK2006214)
文摘In this paper we present a filter-trust-region algorithm for solving LC1 unconstrained optimization problems which uses the second Dini upper directional derivative. We establish the global convergence of the algorithm under reasonable assumptions.
文摘In recent years,wearable devices-based Human Activity Recognition(HAR)models have received significant attention.Previously developed HAR models use hand-crafted features to recognize human activities,leading to the extraction of basic features.The images captured by wearable sensors contain advanced features,allowing them to be analyzed by deep learning algorithms to enhance the detection and recognition of human actions.Poor lighting and limited sensor capabilities can impact data quality,making the recognition of human actions a challenging task.The unimodal-based HAR approaches are not suitable in a real-time environment.Therefore,an updated HAR model is developed using multiple types of data and an advanced deep-learning approach.Firstly,the required signals and sensor data are accumulated from the standard databases.From these signals,the wave features are retrieved.Then the extracted wave features and sensor data are given as the input to recognize the human activity.An Adaptive Hybrid Deep Attentive Network(AHDAN)is developed by incorporating a“1D Convolutional Neural Network(1DCNN)”with a“Gated Recurrent Unit(GRU)”for the human activity recognition process.Additionally,the Enhanced Archerfish Hunting Optimizer(EAHO)is suggested to fine-tune the network parameters for enhancing the recognition process.An experimental evaluation is performed on various deep learning networks and heuristic algorithms to confirm the effectiveness of the proposed HAR model.The EAHO-based HAR model outperforms traditional deep learning networks with an accuracy of 95.36,95.25 for recall,95.48 for specificity,and 95.47 for precision,respectively.The result proved that the developed model is effective in recognizing human action by taking less time.Additionally,it reduces the computation complexity and overfitting issue through using an optimization approach.
基金This project was supported by the National Nature Science Foundation of China (60374009)Nature Science Foundation of Guangdong Province of China (990795).
文摘The mixed l1/H2 optimization problem for MIMO (multiple input-multiple output) discrete-time systems is considered. This problem is formulated as minimizing the l1-norm of a closed-loop transfer matrix while maintaining the H2-norm of another closed-loop transfer matrix at prescribed level. The continuity property of the optimal value in respect to changes in the H2-norm constraint is studied. The existence of the optimal solutions of mixed l1/H2 problem is proved. Because the solution of the mixed l1/H2 problem is based on the scaled-Q method, it avoids the zero interpolation difficulties. The convergent upper and lower bounds can be obtained by solving a sequence of finite dimensional nonlinear programming for which many efficient numerical optimization algorithms exist.
基金Supported by Science and Technology Project of Lishui City(No.2017ZDYF08).
文摘[Objectives]The extraction conditions of flavonoids from flowers of Edgeworthia chrysantha Lind1.were optimized.[Methods]With ethanol as an extraction agent,firstly by single-factor tests,the best levels of ethanol concentration,material-to-liquid ratio,extraction temperature and extraction time were determined.Then the optimal extraction conditions were determined through the quadratic orthogonal regression rotatable design.[Results]The optimal extraction conditions for flavonoids from E.chrysantha were∶material-to-liquid ratio 1∶15.4,extraction temperature 75℃,extraction time 120 min and ethanol concentration 75%,with which the highest yield of flavonoids was 6.02 mg/g.[Conclusions]This study provides a theoretical basis for the further development and utilization of E.chrysantha flowers.
文摘This paper present a simulation study of an evolutionary algorithms, Particle Swarm Optimization PSO algorithm to optimize likelihood function of ARMA（1, 1） model, where maximizing likelihood function is equivalent to maximizing its logarithm, so the objective function ＇obj.fun＇ is maximizing log-likelihood function. Monte Carlo method adapted for implementing and designing the experiments of this simulation. This study including a comparison among three versions of PSO algorithm “Constriction coefficient CCPSO, Inertia weight IWPSO, and Fully Informed FIPSO”, the experiments designed by setting different values of model parameters al, bs sample size n, moreover the parameters of PSO algorithms. MSE used as test statistic to measure the efficiency PSO to estimate model. The results show the ability of PSO to estimate ARMA＇ s parameters, and the minimum values of MSE getting for COPSO.
