The gasoline inline blending process has widely used real-time optimization techniques to achieve optimization objectives,such as minimizing the cost of production.However,the effectiveness of real-time optimization i...The gasoline inline blending process has widely used real-time optimization techniques to achieve optimization objectives,such as minimizing the cost of production.However,the effectiveness of real-time optimization in gasoline blending relies on accurate blending models and is challenged by stochastic disturbances.Thus,we propose a real-time optimization algorithm based on the soft actor-critic(SAC)deep reinforcement learning strategy to optimize gasoline blending without relying on a single blending model and to be robust against disturbances.Our approach constructs the environment using nonlinear blending models and feedstocks with disturbances.The algorithm incorporates the Lagrange multiplier and path constraints in reward design to manage sparse product constraints.Carefully abstracted states facilitate algorithm convergence,and the normalized action vector in each optimization period allows the agent to generalize to some extent across different target production scenarios.Through these well-designed components,the algorithm based on the SAC outperforms real-time optimization methods based on either nonlinear or linear programming.It even demonstrates comparable performance with the time-horizon based real-time optimization method,which requires knowledge of uncertainty models,confirming its capability to handle uncertainty without accurate models.Our simulation illustrates a promising approach to free real-time optimization of the gasoline blending process from uncertainty models that are difficult to acquire in practice.展开更多
Increasing attention has been paid to the efficiency improvement of the induction traction system of high-speed trains due to the high demand for energy saving. In emergency self-propelled mode, however, the dc-link v...Increasing attention has been paid to the efficiency improvement of the induction traction system of high-speed trains due to the high demand for energy saving. In emergency self-propelled mode, however, the dc-link voltage and the traction power of the motor are significantly reduced, resulting in decreased traction efficiency due to the low load and low speed operations. Aiming to tackle this problem, a novel efficiency improved control method is introduced to the emergency mode of high-speed train traction system in this paper. In the proposed method, a total loss model of induction motor considering the behaviors of both iron and copper loss is established. An improved iterative algorithm with decreased computational burden is then introduced, resulting in a fast solving of the optimal flux reference for loss minimization at each control period. In addition, considering the parameter variation problem due to the low load and low speed operations, a parameter estimation method is integrated to improve the controller's robustness. The effectiveness of the proposed method on efficiency improvement at low voltage and low load conditions is demonstrated by simulated and experimental results.展开更多
Deep neural networks(DNNs)have achieved great success in many data processing applications.However,high computational complexity and storage cost make deep learning difficult to be used on resource-constrained devices...Deep neural networks(DNNs)have achieved great success in many data processing applications.However,high computational complexity and storage cost make deep learning difficult to be used on resource-constrained devices,and it is not environmental-friendly with much power cost.In this paper,we focus on low-rank optimization for efficient deep learning techniques.In the space domain,DNNs are compressed by low rank approximation of the network parameters,which directly reduces the storage requirement with a smaller number of network parameters.In the time domain,the network parameters can be trained in a few subspaces,which enables efficient training for fast convergence.The model compression in the spatial domain is summarized into three categories as pre-train,pre-set,and compression-aware methods,respectively.With a series of integrable techniques discussed,such as sparse pruning,quantization,and entropy coding,we can ensemble them in an integration framework with lower computational complexity and storage.In addition to summary of recent technical advances,we have two findings for motivating future works.One is that the effective rank,derived from the Shannon entropy of the normalized singular values,outperforms other conventional sparse measures such as the?_1 norm for network compression.The other is a spatial and temporal balance for tensorized neural networks.For accelerating the training of tensorized neural networks,it is crucial to leverage redundancy for both model compression and subspace training.展开更多
Demand Responsive Transit (DRT) responds to the dynamic users’ requests without any fixed routes and timetablesand determines the stop and the start according to the demands. This study explores the optimization of d...Demand Responsive Transit (DRT) responds to the dynamic users’ requests without any fixed routes and timetablesand determines the stop and the start according to the demands. This study explores the optimization of dynamicvehicle scheduling and real-time route planning in urban public transportation systems, with a focus on busservices. It addresses the limitations of current shared mobility routing algorithms, which are primarily designedfor simpler, single origin/destination scenarios, and do not meet the complex demands of bus transit systems. Theresearch introduces an route planning algorithm designed to dynamically accommodate passenger travel needsand enable real-time route modifications. Unlike traditional methods, this algorithm leverages a queue-based,multi-objective heuristic A∗ approach, offering a solution to the inflexibility and limited coverage of suburbanbus routes. Also, this study conducts a comparative analysis of the proposed algorithm with solutions based onGenetic Algorithm (GA) and Ant Colony Optimization Algorithm (ACO), focusing on calculation time, routelength, passenger waiting time, boarding time, and detour rate. The findings demonstrate that the proposedalgorithmsignificantly enhances route planning speed, achieving an 80–100-fold increase in efficiency over existingmodels, thereby supporting the real-time demands of Demand-Responsive Transportation (DRT) systems. Thestudy concludes that this algorithm not only optimizes route planning in bus transit but also presents a scalablesolution for improving urban mobility.展开更多
Production optimization has gained increasing attention from the smart oilfield community because it can increase economic benefits and oil recovery substantially.While existing methods could produce high-optimality r...Production optimization has gained increasing attention from the smart oilfield community because it can increase economic benefits and oil recovery substantially.While existing methods could produce high-optimality results,they cannot be applied to real-time optimization for large-scale reservoirs due to high computational demands.In addition,most methods generally assume that the reservoir model is deterministic and ignore the uncertainty of the subsurface environment,making the obtained scheme unreliable for practical deployment.In this work,an efficient and robust method,namely evolutionaryassisted reinforcement learning(EARL),is proposed to achieve real-time production optimization under uncertainty.Specifically,the production optimization problem is modeled as a Markov decision process in which a reinforcement learning agent interacts with the reservoir simulator to train a control policy that maximizes the specified goals.To deal with the problems of brittle convergence properties and lack of efficient exploration strategies of reinforcement learning approaches,a population-based evolutionary algorithm is introduced to assist the training of agents,which provides diverse exploration experiences and promotes stability and robustness due to its inherent redundancy.Compared with prior methods that only optimize a solution for a particular scenario,the proposed approach trains a policy that can adapt to uncertain environments and make real-time decisions to cope with unknown changes.The trained policy,represented by a deep convolutional neural network,can adaptively adjust the well controls based on different reservoir states.Simulation results on two reservoir models show that the proposed approach not only outperforms the RL and EA methods in terms of optimization efficiency but also has strong robustness and real-time decision capacity.展开更多
Real-time interaction with uncertain and dynamic environments is essential for robotic systems to achieve functions such as visual perception,force interaction,spatial obstacle avoidance,and motion planning.To ensure ...Real-time interaction with uncertain and dynamic environments is essential for robotic systems to achieve functions such as visual perception,force interaction,spatial obstacle avoidance,and motion planning.To ensure the reliability and determinism of system execution,a flexible real-time control system architecture and interaction algorithm are required.The ROS framework was designed to improve the reusability of robotic software development by providing a distributed structure,hardware abstraction,message-passing mechanism,and application prototypes.Rich ecosystems for robotic development have been built around ROS1 and ROS2 architectures based on the Linux system.However,because of the fairness scheduling principle of the default Linux system design and the complexity of the kernel,the system does not have real-time computing.To achieve a balance between real-time and non-real-time computing,this paper uses the transmission mechanism of ROS2,combines it with the scheduling mechanism of the Linux operating system,and uses Preempt_RT to enhance the real-time computing of ROS1 and ROS2.The real-time performance evaluation of ROS1 and ROS2 is conducted from multiple perspectives,including throughput,transmission mode,QoS service quality,frequency,number of subscription nodes and EtherCAT master.This paper makes two significant contributions:firstly,it employs Preempt_RT to optimize the native ROS2 system,effectively enhancing the real-time performance of native ROS2 message transmission;secondly,it conducts a comprehensive evaluation of the real-time performance of both native and optimized ROS2 systems.This comparison elucidates the benefits of the optimized ROS2 architecture regarding real-time performance,with results vividly demonstrated through illustrative figures.展开更多
The goal of this research is to look at multi-target optimization of a two-stage helical gearbox in order to determine the best key design elements for reducing gearbox height and enhancing gearbox efficiency.To do th...The goal of this research is to look at multi-target optimization of a two-stage helical gearbox in order to determine the best key design elements for reducing gearbox height and enhancing gearbox efficiency.To do this,the method known as Taguchi and GRA(Grey Relation Analysis)were used in two stages to address the problem.The single-objective optimization problem was addressed first to close the gap between variable levels,and then the multi-objective optimization problem was solved to determine the best primary design variables.The first and second stage CWFWs(Coefficients of Wheel Face Width),ACS(Permissible Contact Stresses),and first stage gear ratio were also calculated.The study’s findings were utilized to identify the best values for five critical design aspects of a two-stage helical gearbox.展开更多
In this article, a novel scalarization technique, called the improved objective-constraint approach, is introduced to find efficient solutions of a given multiobjective programming problem. The presented scalarized pr...In this article, a novel scalarization technique, called the improved objective-constraint approach, is introduced to find efficient solutions of a given multiobjective programming problem. The presented scalarized problem extends the objective-constraint problem. It is demonstrated that how adding variables to the scalarized problem, can lead to find conditions for (weakly, properly) Pareto optimal solutions. Applying the obtained necessary and sufficient conditions, two algorithms for generating the Pareto front approximation of bi-objective and three-objective programming problems are designed. These algorithms are easy to implement and can achieve an even approximation of (weakly, properly) Pareto optimal solutions. These algorithms can be generalized for optimization problems with more than three criterion functions, too. The effectiveness and capability of the algorithms are demonstrated in test problems.展开更多
The Backscatter communication has gained widespread attention from academia and industry in recent years. In this paper, A method of resource allocation and trajectory optimization is proposed for UAV-assisted backsca...The Backscatter communication has gained widespread attention from academia and industry in recent years. In this paper, A method of resource allocation and trajectory optimization is proposed for UAV-assisted backscatter communication based on user trajectory. This paper will establish an optimization problem of jointly optimizing the UAV trajectories, UAV transmission power and BD scheduling based on the large-scale channel state signals estimated in advance of the known user trajectories, taking into account the constraints of BD data and working energy consumption, to maximize the energy efficiency of the system. The problem is a non-convex optimization problem in fractional form, and there is nonlinear coupling between optimization variables.An iterative algorithm is proposed based on Dinkelbach algorithm, block coordinate descent method and continuous convex optimization technology. First, the objective function is converted into a non-fractional programming problem based on Dinkelbach method,and then the block coordinate descent method is used to decompose the original complex problem into three independent sub-problems. Finally, the successive convex approximation method is used to solve the trajectory optimization sub-problem. The simulation results show that the proposed scheme and algorithm have obvious energy efficiency gains compared with the comparison scheme.展开更多
This paper considers a high energy efficiency dynamic connected(HEDC)structure,which promotes the practicability and reduces the power consumption of hybrid precoding system by lowresolution phase shifters(PSs).Based ...This paper considers a high energy efficiency dynamic connected(HEDC)structure,which promotes the practicability and reduces the power consumption of hybrid precoding system by lowresolution phase shifters(PSs).Based on the proposed structure,a new hybrid precoding algorithm is presented to optimize the energy efficiency,namely,HP-HEDC algorithm.Firstly,via a new defined effective optimal precoding matrix,the problem of optimizing the analog switch precoding matrix is formulated as a sparse representation problem.Thus,the optimal analog switch precoding matrix can be readily obtained by the branch-and-bound method.Then,the digital precoding matrix optimization problem is modeled as a dictionary update problem and solved by the method of optimal direction(MOD).Finally,the diagonal entries of the analog PS precoding matrix are optimized by exhaustive search independently since PS and antenna is one-to-one.Simulation results show that the HEDC structure enjoys low power consumption and satisfactory spectral efficiency.The proposed algorithm presents at least 50%energy efficiency improvement compared with other algorithms when the PS resolution is set as 3-bit.展开更多
This study delves into the intricate relationship between iron(Fe)content in kaolinite and its impact on the adsorption behavior of sodium oleate.The effects of different iron concentrations on adsorption energy,hydro...This study delves into the intricate relationship between iron(Fe)content in kaolinite and its impact on the adsorption behavior of sodium oleate.The effects of different iron concentrations on adsorption energy,hydrogen bond kinetics and adsorption efficiency were studied through simulation and experimental verification.The results show that the presence of iron in the kaolinite structure significantly improves the adsorption capacity of sodium oleate.Kaolinite samples with high iron content have better adsorption properties,lower adsorption energy levels and shorter and stronger hydrogen bonds than pure kaolinite.The optimal concentration of oleic acid ions for achieving maximum adsorption efficiency was identified as 1.2 mmol/L across different kaolinite samples.At this concentration,the adsorption rates and capacities reach their peak,with Fe-enriched kaolinite samples exhibiting notably higher flotation recovery rates.This optimal concentration represents a balance between sufficient oleic acid ion availability for surface interactions and the prevention of self-aggregation phenomena that could hinder adsorption.This study offers promising avenues for optimizing the flotation process in mineral processing applications.展开更多
The task of selecting robotic mechanic assembly technologies (RMAT) is considered as a multi-criteria optimization task, which in this formulation is solved on the set of previously obtained solutions regarding the se...The task of selecting robotic mechanic assembly technologies (RMAT) is considered as a multi-criteria optimization task, which in this formulation is solved on the set of previously obtained solutions regarding the selection of RMAT. The purpose of the paper is to increase the efficiency of technological preparation of robotic mechanical assembly production of machine and instrument engineering due to a new approach to the selection of RMAT using Pareto optimization and the peculiarities of the selection task formulation. The novelty consists in the further development of a science-based approach to solving multi-criteria selection task, based on the first proposed formalisms of the specified process, which reflect the peculiarities of the selection task formulation, its meaningful essence and the content of the Pareto optimization method. The practical value of the research lies in the proposed engineering-acceptable approach to solving applied multi-criteria selection tasks on the example of RMAT selection, which is invariant to the statement of the selection task, the dimension of the task, and its meaningful essence. The methods of discrete optimization, fuzzy multi-criteria selection of alternatives, and the Pareto optimization method were used for the research. The main results of this work consist of the development of formalisms and the demonstration of the efficiency of the proposed approach for the applied task of RMAT selection. The peculiarity of the developed approach is the combination of Pareto optimization, performed on a discrete set of local criteria. Directions for further research are presented.展开更多
Experimentally, the best design gives estimates of the desired effects and contrasts with maximum precision. Efficiency as a discriminating factor enables comparison of designs. The goal of Response Surface Methodolog...Experimentally, the best design gives estimates of the desired effects and contrasts with maximum precision. Efficiency as a discriminating factor enables comparison of designs. The goal of Response Surface Methodology (RSM) is the determination of the best settings of the in-put variables for a maximum (or a minimum) response within a region of interest, R. This calls for fitting a model that adequately represents the mean response since such a model, is then used to locate the optimum. D-, A-, E- and T-Optimal designs of a rotatable design of degree two in four dimensions constructed using balanced incomplete block designs (BIBD) when the number of replications is less than three times the number of pairs of treatments occur together in the design and their relative efficiencies to general designs are presented. D-optimal design had 88 runs after replicating the factorial part twice and the axial part thrice with an optimal variance of 0.6965612 giving an efficiency of 97.7% while for A- and T-optimal designs they are formed with 112 runs each obtained by replicating the factorial part two times and axial part six times. Their optimal variances are 0.05798174 and 1.29828 respectively, with efficiency of 71.8% for A-optimal and 87.5% for T-optimal design. E-optimal design was found to be the most efficient design with an only 32 runs comprising only of the factorial part and with an optimal variance of 0.4182000, attaining an efficiency of approximately 1%. This study proposes the adoption of the E-optimal design in estimating the parameters of a rotatable second-order degree model constructed using BIBD for less costs and time saving.展开更多
Traditional topology optimization methods often suffer from the“dimension curse”problem,wherein the com-putation time increases exponentially with the degrees of freedom in the background grid.Overcoming this challe...Traditional topology optimization methods often suffer from the“dimension curse”problem,wherein the com-putation time increases exponentially with the degrees of freedom in the background grid.Overcoming this challenge,we introduce a real-time topology optimization approach leveraging Conditional Generative Adversarial Networks with Gradient Penalty(CGAN-GP).This innovative method allows for nearly instantaneous prediction of optimized structures.Given a specific boundary condition,the network can produce a unique optimized structure in a one-to-one manner.The process begins by establishing a dataset using simulation data generated through the Solid Isotropic Material with Penalization(SIMP)method.Subsequently,we design a conditional generative adversarial network and train it to generate optimized structures.To further enhance the quality of the optimized structures produced by CGAN-GP,we incorporate Pix2pixGAN.This augmentation results in sharper topologies,yielding structures with enhanced clarity,de-blurring,and edge smoothing.Our proposed method yields a significant reduction in computational time when compared to traditional topology optimization algorithms,all while maintaining an impressive accuracy rate of up to 85%,as demonstrated through numerical examples.展开更多
Capacity allocation and energy management strategies for energy storage are critical to the safety and economical operation of microgrids.In this paper,an improved energymanagement strategy based on real-time electric...Capacity allocation and energy management strategies for energy storage are critical to the safety and economical operation of microgrids.In this paper,an improved energymanagement strategy based on real-time electricity price combined with state of charge is proposed to optimize the economic operation of wind and solar microgrids,and the optimal allocation of energy storage capacity is carried out by using this strategy.Firstly,the structure and model of microgrid are analyzed,and the outputmodel of wind power,photovoltaic and energy storage is established.Then,considering the interactive power cost between the microgrid and the main grid and the charge-discharge penalty cost of energy storage,an optimization objective function is established,and an improved energy management strategy is proposed on this basis.Finally,a physicalmodel is built inMATLAB/Simulink for simulation verification,and the energy management strategy is compared and analyzed on sunny and rainy days.The initial configuration cost function of energy storage is added to optimize the allocation of energy storage capacity.The simulation results show that the improved energy management strategy can make the battery charge-discharge response to real-time electricity price and state of charge better than the traditional strategy on sunny or rainy days,reduce the interactive power cost between the microgrid system and the power grid.After analyzing the change of energy storage power with cost,we obtain the best energy storage capacity and energy storage power.展开更多
Finite element(FE) is a powerful tool and has been applied by investigators to real-time hybrid simulations(RTHSs). This study focuses on the computational efficiency, including the computational time and accuracy...Finite element(FE) is a powerful tool and has been applied by investigators to real-time hybrid simulations(RTHSs). This study focuses on the computational efficiency, including the computational time and accuracy, of numerical integrations in solving FE numerical substructure in RTHSs. First, sparse matrix storage schemes are adopted to decrease the computational time of FE numerical substructure. In this way, the task execution time(TET) decreases such that the scale of the numerical substructure model increases. Subsequently, several commonly used explicit numerical integration algorithms, including the central difference method(CDM), the Newmark explicit method, the Chang method and the Gui-λ method, are comprehensively compared to evaluate their computational time in solving FE numerical substructure. CDM is better than the other explicit integration algorithms when the damping matrix is diagonal, while the Gui-λ(λ = 4) method is advantageous when the damping matrix is non-diagonal. Finally, the effect of time delay on the computational accuracy of RTHSs is investigated by simulating structure-foundation systems. Simulation results show that the influences of time delay on the displacement response become obvious with the mass ratio increasing, and delay compensation methods may reduce the relative error of the displacement peak value to less than 5% even under the large time-step and large time delay.展开更多
The set-valued optimization problem with constraints is considered in the sense of super efficiency in locally convex linear topological spaces. Under the assumption of iccone-convexlikeness, by applying the seperatio...The set-valued optimization problem with constraints is considered in the sense of super efficiency in locally convex linear topological spaces. Under the assumption of iccone-convexlikeness, by applying the seperation theorem, Kuhn-Tucker's, Lagrange's and saddle points optimality conditions, the necessary conditions are obtained for the set-valued optimization problem to attain its super efficient solutions. Also, the sufficient conditions for Kuhn-Tucker's, Lagrange's and saddle points optimality conditions are derived.展开更多
By combing the properties of chaos optimization method and genetic algorithm,an adaptive mutative scale chaos genetic algorithm(AMSCGA) was proposed by using one-dimensional iterative chaotic self-map with infinite co...By combing the properties of chaos optimization method and genetic algorithm,an adaptive mutative scale chaos genetic algorithm(AMSCGA) was proposed by using one-dimensional iterative chaotic self-map with infinite collapses within the finite region of [-1,1].Some measures in the optimization algorithm,such as adjusting the searching space of optimized variables continuously by using adaptive mutative scale method and making the most circle time as its control guideline,were taken to ensure its speediness and veracity in seeking the optimization process.The calculation examples about three testing functions reveal that AMSCGA has both high searching speed and high precision.Furthermore,the average truncated generations,the distribution entropy of truncated generations and the ratio of average inertia generations were used to evaluate the optimization efficiency of AMSCGA quantificationally.It is shown that the optimization efficiency of AMSCGA is higher than that of genetic algorithm.展开更多
In this paper,an improved discharging circuit was proposed to quicken the decay of the current in the drive coil in a reluctance accelerator when the armature reaches the center of the coil.The aim of this is to preve...In this paper,an improved discharging circuit was proposed to quicken the decay of the current in the drive coil in a reluctance accelerator when the armature reaches the center of the coil.The aim of this is to prevent the suck-back effect caused by the residual current in drive coil.The method is adding a reverse charging branch with a small capacitor in the traditional pulsed discharging circuit.The results under the traditional circuit and the improved circuit were compared in a simulation.The experiment then verified the simulations and they had good agreement.Simulation and experiment both demonstrated the improved circuit can effectively prevent the suck-back effect and increase the efficiency.At the voltage of 800 V,an efficiency increase of 36.34% was obtained.展开更多
The application and development of a wide-area measurement system(WAMS)has enabled many applications and led to several requirements based on dynamic measurement data.Such data are transmitted as big data information ...The application and development of a wide-area measurement system(WAMS)has enabled many applications and led to several requirements based on dynamic measurement data.Such data are transmitted as big data information flow.To ensure effective transmission of wide-frequency electrical information by the communication protocol of a WAMS,this study performs real-time traffic monitoring and analysis of the data network of a power information system,and establishes corresponding network optimization strategies to solve existing transmission problems.This study utilizes the traffic analysis results obtained using the current real-time dynamic monitoring system to design an optimization strategy,covering the optimization in three progressive levels:the underlying communication protocol,source data,and transmission process.Optimization of the system structure and scheduling optimization of data information are validated to be feasible and practical via tests.展开更多
基金supported by National Key Research & Development Program-Intergovernmental International Science and Technology Innovation Cooperation Project (2021YFE0112800)National Natural Science Foundation of China (Key Program: 62136003)+2 种基金National Natural Science Foundation of China (62073142)Fundamental Research Funds for the Central Universities (222202417006)Shanghai Al Lab
文摘The gasoline inline blending process has widely used real-time optimization techniques to achieve optimization objectives,such as minimizing the cost of production.However,the effectiveness of real-time optimization in gasoline blending relies on accurate blending models and is challenged by stochastic disturbances.Thus,we propose a real-time optimization algorithm based on the soft actor-critic(SAC)deep reinforcement learning strategy to optimize gasoline blending without relying on a single blending model and to be robust against disturbances.Our approach constructs the environment using nonlinear blending models and feedstocks with disturbances.The algorithm incorporates the Lagrange multiplier and path constraints in reward design to manage sparse product constraints.Carefully abstracted states facilitate algorithm convergence,and the normalized action vector in each optimization period allows the agent to generalize to some extent across different target production scenarios.Through these well-designed components,the algorithm based on the SAC outperforms real-time optimization methods based on either nonlinear or linear programming.It even demonstrates comparable performance with the time-horizon based real-time optimization method,which requires knowledge of uncertainty models,confirming its capability to handle uncertainty without accurate models.Our simulation illustrates a promising approach to free real-time optimization of the gasoline blending process from uncertainty models that are difficult to acquire in practice.
基金supported in part by the Science Foundation of the Chinese Academy of Railway Sciences under Grant Number:2023QT001。
文摘Increasing attention has been paid to the efficiency improvement of the induction traction system of high-speed trains due to the high demand for energy saving. In emergency self-propelled mode, however, the dc-link voltage and the traction power of the motor are significantly reduced, resulting in decreased traction efficiency due to the low load and low speed operations. Aiming to tackle this problem, a novel efficiency improved control method is introduced to the emergency mode of high-speed train traction system in this paper. In the proposed method, a total loss model of induction motor considering the behaviors of both iron and copper loss is established. An improved iterative algorithm with decreased computational burden is then introduced, resulting in a fast solving of the optimal flux reference for loss minimization at each control period. In addition, considering the parameter variation problem due to the low load and low speed operations, a parameter estimation method is integrated to improve the controller's robustness. The effectiveness of the proposed method on efficiency improvement at low voltage and low load conditions is demonstrated by simulated and experimental results.
基金supported by the National Natural Science Foundation of China(62171088,U19A2052,62020106011)the Medico-Engineering Cooperation Funds from University of Electronic Science and Technology of China(ZYGX2021YGLH215,ZYGX2022YGRH005)。
文摘Deep neural networks(DNNs)have achieved great success in many data processing applications.However,high computational complexity and storage cost make deep learning difficult to be used on resource-constrained devices,and it is not environmental-friendly with much power cost.In this paper,we focus on low-rank optimization for efficient deep learning techniques.In the space domain,DNNs are compressed by low rank approximation of the network parameters,which directly reduces the storage requirement with a smaller number of network parameters.In the time domain,the network parameters can be trained in a few subspaces,which enables efficient training for fast convergence.The model compression in the spatial domain is summarized into three categories as pre-train,pre-set,and compression-aware methods,respectively.With a series of integrable techniques discussed,such as sparse pruning,quantization,and entropy coding,we can ensemble them in an integration framework with lower computational complexity and storage.In addition to summary of recent technical advances,we have two findings for motivating future works.One is that the effective rank,derived from the Shannon entropy of the normalized singular values,outperforms other conventional sparse measures such as the?_1 norm for network compression.The other is a spatial and temporal balance for tensorized neural networks.For accelerating the training of tensorized neural networks,it is crucial to leverage redundancy for both model compression and subspace training.
文摘Demand Responsive Transit (DRT) responds to the dynamic users’ requests without any fixed routes and timetablesand determines the stop and the start according to the demands. This study explores the optimization of dynamicvehicle scheduling and real-time route planning in urban public transportation systems, with a focus on busservices. It addresses the limitations of current shared mobility routing algorithms, which are primarily designedfor simpler, single origin/destination scenarios, and do not meet the complex demands of bus transit systems. Theresearch introduces an route planning algorithm designed to dynamically accommodate passenger travel needsand enable real-time route modifications. Unlike traditional methods, this algorithm leverages a queue-based,multi-objective heuristic A∗ approach, offering a solution to the inflexibility and limited coverage of suburbanbus routes. Also, this study conducts a comparative analysis of the proposed algorithm with solutions based onGenetic Algorithm (GA) and Ant Colony Optimization Algorithm (ACO), focusing on calculation time, routelength, passenger waiting time, boarding time, and detour rate. The findings demonstrate that the proposedalgorithmsignificantly enhances route planning speed, achieving an 80–100-fold increase in efficiency over existingmodels, thereby supporting the real-time demands of Demand-Responsive Transportation (DRT) systems. Thestudy concludes that this algorithm not only optimizes route planning in bus transit but also presents a scalablesolution for improving urban mobility.
基金This work is supported by the National Natural Science Foundation of China under Grant 52274057,52074340 and 51874335the Major Scientific and Technological Projects of CNPC under Grant ZD2019-183-008the Science and Technology Support Plan for Youth Innovation of University in Shandong Province under Grant 2019KJH002,111 Project under Grant B08028.
文摘Production optimization has gained increasing attention from the smart oilfield community because it can increase economic benefits and oil recovery substantially.While existing methods could produce high-optimality results,they cannot be applied to real-time optimization for large-scale reservoirs due to high computational demands.In addition,most methods generally assume that the reservoir model is deterministic and ignore the uncertainty of the subsurface environment,making the obtained scheme unreliable for practical deployment.In this work,an efficient and robust method,namely evolutionaryassisted reinforcement learning(EARL),is proposed to achieve real-time production optimization under uncertainty.Specifically,the production optimization problem is modeled as a Markov decision process in which a reinforcement learning agent interacts with the reservoir simulator to train a control policy that maximizes the specified goals.To deal with the problems of brittle convergence properties and lack of efficient exploration strategies of reinforcement learning approaches,a population-based evolutionary algorithm is introduced to assist the training of agents,which provides diverse exploration experiences and promotes stability and robustness due to its inherent redundancy.Compared with prior methods that only optimize a solution for a particular scenario,the proposed approach trains a policy that can adapt to uncertain environments and make real-time decisions to cope with unknown changes.The trained policy,represented by a deep convolutional neural network,can adaptively adjust the well controls based on different reservoir states.Simulation results on two reservoir models show that the proposed approach not only outperforms the RL and EA methods in terms of optimization efficiency but also has strong robustness and real-time decision capacity.
基金Supported by National Key Research and Development Program of China(Grant No.2019YFB1309900)Institute for Guo Qiang,Tsinghua University of China(Grant No.2019GQG0007).
文摘Real-time interaction with uncertain and dynamic environments is essential for robotic systems to achieve functions such as visual perception,force interaction,spatial obstacle avoidance,and motion planning.To ensure the reliability and determinism of system execution,a flexible real-time control system architecture and interaction algorithm are required.The ROS framework was designed to improve the reusability of robotic software development by providing a distributed structure,hardware abstraction,message-passing mechanism,and application prototypes.Rich ecosystems for robotic development have been built around ROS1 and ROS2 architectures based on the Linux system.However,because of the fairness scheduling principle of the default Linux system design and the complexity of the kernel,the system does not have real-time computing.To achieve a balance between real-time and non-real-time computing,this paper uses the transmission mechanism of ROS2,combines it with the scheduling mechanism of the Linux operating system,and uses Preempt_RT to enhance the real-time computing of ROS1 and ROS2.The real-time performance evaluation of ROS1 and ROS2 is conducted from multiple perspectives,including throughput,transmission mode,QoS service quality,frequency,number of subscription nodes and EtherCAT master.This paper makes two significant contributions:firstly,it employs Preempt_RT to optimize the native ROS2 system,effectively enhancing the real-time performance of native ROS2 message transmission;secondly,it conducts a comprehensive evaluation of the real-time performance of both native and optimized ROS2 systems.This comparison elucidates the benefits of the optimized ROS2 architecture regarding real-time performance,with results vividly demonstrated through illustrative figures.
文摘The goal of this research is to look at multi-target optimization of a two-stage helical gearbox in order to determine the best key design elements for reducing gearbox height and enhancing gearbox efficiency.To do this,the method known as Taguchi and GRA(Grey Relation Analysis)were used in two stages to address the problem.The single-objective optimization problem was addressed first to close the gap between variable levels,and then the multi-objective optimization problem was solved to determine the best primary design variables.The first and second stage CWFWs(Coefficients of Wheel Face Width),ACS(Permissible Contact Stresses),and first stage gear ratio were also calculated.The study’s findings were utilized to identify the best values for five critical design aspects of a two-stage helical gearbox.
文摘In this article, a novel scalarization technique, called the improved objective-constraint approach, is introduced to find efficient solutions of a given multiobjective programming problem. The presented scalarized problem extends the objective-constraint problem. It is demonstrated that how adding variables to the scalarized problem, can lead to find conditions for (weakly, properly) Pareto optimal solutions. Applying the obtained necessary and sufficient conditions, two algorithms for generating the Pareto front approximation of bi-objective and three-objective programming problems are designed. These algorithms are easy to implement and can achieve an even approximation of (weakly, properly) Pareto optimal solutions. These algorithms can be generalized for optimization problems with more than three criterion functions, too. The effectiveness and capability of the algorithms are demonstrated in test problems.
文摘The Backscatter communication has gained widespread attention from academia and industry in recent years. In this paper, A method of resource allocation and trajectory optimization is proposed for UAV-assisted backscatter communication based on user trajectory. This paper will establish an optimization problem of jointly optimizing the UAV trajectories, UAV transmission power and BD scheduling based on the large-scale channel state signals estimated in advance of the known user trajectories, taking into account the constraints of BD data and working energy consumption, to maximize the energy efficiency of the system. The problem is a non-convex optimization problem in fractional form, and there is nonlinear coupling between optimization variables.An iterative algorithm is proposed based on Dinkelbach algorithm, block coordinate descent method and continuous convex optimization technology. First, the objective function is converted into a non-fractional programming problem based on Dinkelbach method,and then the block coordinate descent method is used to decompose the original complex problem into three independent sub-problems. Finally, the successive convex approximation method is used to solve the trajectory optimization sub-problem. The simulation results show that the proposed scheme and algorithm have obvious energy efficiency gains compared with the comparison scheme.
基金supported by the National Natural Science Foundation of China(Grant No.61971117)the Natural Science Foundation of Hebei Province(Grant No.F2020501007)the S&T Program of Hebei(No.22377717D)。
文摘This paper considers a high energy efficiency dynamic connected(HEDC)structure,which promotes the practicability and reduces the power consumption of hybrid precoding system by lowresolution phase shifters(PSs).Based on the proposed structure,a new hybrid precoding algorithm is presented to optimize the energy efficiency,namely,HP-HEDC algorithm.Firstly,via a new defined effective optimal precoding matrix,the problem of optimizing the analog switch precoding matrix is formulated as a sparse representation problem.Thus,the optimal analog switch precoding matrix can be readily obtained by the branch-and-bound method.Then,the digital precoding matrix optimization problem is modeled as a dictionary update problem and solved by the method of optimal direction(MOD).Finally,the diagonal entries of the analog PS precoding matrix are optimized by exhaustive search independently since PS and antenna is one-to-one.Simulation results show that the HEDC structure enjoys low power consumption and satisfactory spectral efficiency.The proposed algorithm presents at least 50%energy efficiency improvement compared with other algorithms when the PS resolution is set as 3-bit.
基金supported by the Natural Science Foundation of China(No.52174232)the Project was supported by Open Research Grant of Joint National-Local Engineering Research Centre for Safe and Precise Coal Mining(Nos.EC2022003 and EC2023005)+1 种基金Anhui University of Science and Technology 2023 Graduate Student Innovation Fund(No.2023cx2106)Open Research Grant of Anhui Engineering Research Center for Coal Clean Processing and Carbon Emission Reduction(No.CCCE-2023003).
文摘This study delves into the intricate relationship between iron(Fe)content in kaolinite and its impact on the adsorption behavior of sodium oleate.The effects of different iron concentrations on adsorption energy,hydrogen bond kinetics and adsorption efficiency were studied through simulation and experimental verification.The results show that the presence of iron in the kaolinite structure significantly improves the adsorption capacity of sodium oleate.Kaolinite samples with high iron content have better adsorption properties,lower adsorption energy levels and shorter and stronger hydrogen bonds than pure kaolinite.The optimal concentration of oleic acid ions for achieving maximum adsorption efficiency was identified as 1.2 mmol/L across different kaolinite samples.At this concentration,the adsorption rates and capacities reach their peak,with Fe-enriched kaolinite samples exhibiting notably higher flotation recovery rates.This optimal concentration represents a balance between sufficient oleic acid ion availability for surface interactions and the prevention of self-aggregation phenomena that could hinder adsorption.This study offers promising avenues for optimizing the flotation process in mineral processing applications.
文摘The task of selecting robotic mechanic assembly technologies (RMAT) is considered as a multi-criteria optimization task, which in this formulation is solved on the set of previously obtained solutions regarding the selection of RMAT. The purpose of the paper is to increase the efficiency of technological preparation of robotic mechanical assembly production of machine and instrument engineering due to a new approach to the selection of RMAT using Pareto optimization and the peculiarities of the selection task formulation. The novelty consists in the further development of a science-based approach to solving multi-criteria selection task, based on the first proposed formalisms of the specified process, which reflect the peculiarities of the selection task formulation, its meaningful essence and the content of the Pareto optimization method. The practical value of the research lies in the proposed engineering-acceptable approach to solving applied multi-criteria selection tasks on the example of RMAT selection, which is invariant to the statement of the selection task, the dimension of the task, and its meaningful essence. The methods of discrete optimization, fuzzy multi-criteria selection of alternatives, and the Pareto optimization method were used for the research. The main results of this work consist of the development of formalisms and the demonstration of the efficiency of the proposed approach for the applied task of RMAT selection. The peculiarity of the developed approach is the combination of Pareto optimization, performed on a discrete set of local criteria. Directions for further research are presented.
文摘Experimentally, the best design gives estimates of the desired effects and contrasts with maximum precision. Efficiency as a discriminating factor enables comparison of designs. The goal of Response Surface Methodology (RSM) is the determination of the best settings of the in-put variables for a maximum (or a minimum) response within a region of interest, R. This calls for fitting a model that adequately represents the mean response since such a model, is then used to locate the optimum. D-, A-, E- and T-Optimal designs of a rotatable design of degree two in four dimensions constructed using balanced incomplete block designs (BIBD) when the number of replications is less than three times the number of pairs of treatments occur together in the design and their relative efficiencies to general designs are presented. D-optimal design had 88 runs after replicating the factorial part twice and the axial part thrice with an optimal variance of 0.6965612 giving an efficiency of 97.7% while for A- and T-optimal designs they are formed with 112 runs each obtained by replicating the factorial part two times and axial part six times. Their optimal variances are 0.05798174 and 1.29828 respectively, with efficiency of 71.8% for A-optimal and 87.5% for T-optimal design. E-optimal design was found to be the most efficient design with an only 32 runs comprising only of the factorial part and with an optimal variance of 0.4182000, attaining an efficiency of approximately 1%. This study proposes the adoption of the E-optimal design in estimating the parameters of a rotatable second-order degree model constructed using BIBD for less costs and time saving.
基金supported by the National Key Research and Development Projects (Grant Nos.2021YFB3300601,2021YFB3300603,2021YFB3300604)Fundamental Research Funds for the Central Universities (No.DUT22QN241).
文摘Traditional topology optimization methods often suffer from the“dimension curse”problem,wherein the com-putation time increases exponentially with the degrees of freedom in the background grid.Overcoming this challenge,we introduce a real-time topology optimization approach leveraging Conditional Generative Adversarial Networks with Gradient Penalty(CGAN-GP).This innovative method allows for nearly instantaneous prediction of optimized structures.Given a specific boundary condition,the network can produce a unique optimized structure in a one-to-one manner.The process begins by establishing a dataset using simulation data generated through the Solid Isotropic Material with Penalization(SIMP)method.Subsequently,we design a conditional generative adversarial network and train it to generate optimized structures.To further enhance the quality of the optimized structures produced by CGAN-GP,we incorporate Pix2pixGAN.This augmentation results in sharper topologies,yielding structures with enhanced clarity,de-blurring,and edge smoothing.Our proposed method yields a significant reduction in computational time when compared to traditional topology optimization algorithms,all while maintaining an impressive accuracy rate of up to 85%,as demonstrated through numerical examples.
基金a phased achievement of Gansu Province’s Major Science and Technology Project(W22KJ2722005)“Research on Optimal Configuration and Operation Strategy of Energy Storage under“New Energy+Energy Storage”Mode”.
文摘Capacity allocation and energy management strategies for energy storage are critical to the safety and economical operation of microgrids.In this paper,an improved energymanagement strategy based on real-time electricity price combined with state of charge is proposed to optimize the economic operation of wind and solar microgrids,and the optimal allocation of energy storage capacity is carried out by using this strategy.Firstly,the structure and model of microgrid are analyzed,and the outputmodel of wind power,photovoltaic and energy storage is established.Then,considering the interactive power cost between the microgrid and the main grid and the charge-discharge penalty cost of energy storage,an optimization objective function is established,and an improved energy management strategy is proposed on this basis.Finally,a physicalmodel is built inMATLAB/Simulink for simulation verification,and the energy management strategy is compared and analyzed on sunny and rainy days.The initial configuration cost function of energy storage is added to optimize the allocation of energy storage capacity.The simulation results show that the improved energy management strategy can make the battery charge-discharge response to real-time electricity price and state of charge better than the traditional strategy on sunny or rainy days,reduce the interactive power cost between the microgrid system and the power grid.After analyzing the change of energy storage power with cost,we obtain the best energy storage capacity and energy storage power.
基金National Natural Science Foundation of China under Grant Nos.51639006 and 51725901
文摘Finite element(FE) is a powerful tool and has been applied by investigators to real-time hybrid simulations(RTHSs). This study focuses on the computational efficiency, including the computational time and accuracy, of numerical integrations in solving FE numerical substructure in RTHSs. First, sparse matrix storage schemes are adopted to decrease the computational time of FE numerical substructure. In this way, the task execution time(TET) decreases such that the scale of the numerical substructure model increases. Subsequently, several commonly used explicit numerical integration algorithms, including the central difference method(CDM), the Newmark explicit method, the Chang method and the Gui-λ method, are comprehensively compared to evaluate their computational time in solving FE numerical substructure. CDM is better than the other explicit integration algorithms when the damping matrix is diagonal, while the Gui-λ(λ = 4) method is advantageous when the damping matrix is non-diagonal. Finally, the effect of time delay on the computational accuracy of RTHSs is investigated by simulating structure-foundation systems. Simulation results show that the influences of time delay on the displacement response become obvious with the mass ratio increasing, and delay compensation methods may reduce the relative error of the displacement peak value to less than 5% even under the large time-step and large time delay.
基金Supported by the National Natural Science Foundation of China (10461007)the Science and Technology Foundation of the Education Department of Jiangxi Province (GJJ09069)
文摘The set-valued optimization problem with constraints is considered in the sense of super efficiency in locally convex linear topological spaces. Under the assumption of iccone-convexlikeness, by applying the seperation theorem, Kuhn-Tucker's, Lagrange's and saddle points optimality conditions, the necessary conditions are obtained for the set-valued optimization problem to attain its super efficient solutions. Also, the sufficient conditions for Kuhn-Tucker's, Lagrange's and saddle points optimality conditions are derived.
基金Project(60874114) supported by the National Natural Science Foundation of China
文摘By combing the properties of chaos optimization method and genetic algorithm,an adaptive mutative scale chaos genetic algorithm(AMSCGA) was proposed by using one-dimensional iterative chaotic self-map with infinite collapses within the finite region of [-1,1].Some measures in the optimization algorithm,such as adjusting the searching space of optimized variables continuously by using adaptive mutative scale method and making the most circle time as its control guideline,were taken to ensure its speediness and veracity in seeking the optimization process.The calculation examples about three testing functions reveal that AMSCGA has both high searching speed and high precision.Furthermore,the average truncated generations,the distribution entropy of truncated generations and the ratio of average inertia generations were used to evaluate the optimization efficiency of AMSCGA quantificationally.It is shown that the optimization efficiency of AMSCGA is higher than that of genetic algorithm.
基金This work was supported by the Fundamental Research Funds for the Central Universities[Grant number 2019XJ01].
文摘In this paper,an improved discharging circuit was proposed to quicken the decay of the current in the drive coil in a reluctance accelerator when the armature reaches the center of the coil.The aim of this is to prevent the suck-back effect caused by the residual current in drive coil.The method is adding a reverse charging branch with a small capacitor in the traditional pulsed discharging circuit.The results under the traditional circuit and the improved circuit were compared in a simulation.The experiment then verified the simulations and they had good agreement.Simulation and experiment both demonstrated the improved circuit can effectively prevent the suck-back effect and increase the efficiency.At the voltage of 800 V,an efficiency increase of 36.34% was obtained.
文摘The application and development of a wide-area measurement system(WAMS)has enabled many applications and led to several requirements based on dynamic measurement data.Such data are transmitted as big data information flow.To ensure effective transmission of wide-frequency electrical information by the communication protocol of a WAMS,this study performs real-time traffic monitoring and analysis of the data network of a power information system,and establishes corresponding network optimization strategies to solve existing transmission problems.This study utilizes the traffic analysis results obtained using the current real-time dynamic monitoring system to design an optimization strategy,covering the optimization in three progressive levels:the underlying communication protocol,source data,and transmission process.Optimization of the system structure and scheduling optimization of data information are validated to be feasible and practical via tests.