To improve the inference efficiency of convolutional neural networks(CNN),the existing neural networks mainly adopt heuristic and dynamic programming algorithms to realize parallel scheduling among operators.Heuristic...To improve the inference efficiency of convolutional neural networks(CNN),the existing neural networks mainly adopt heuristic and dynamic programming algorithms to realize parallel scheduling among operators.Heuristic scheduling algorithms can generate local optima easily,while the dynamic programming algorithm has a long convergence time for complex structural models.This paper mainly studies the parallel scheduling between operators and proposes an inter-operator parallelism schedule(IOPS)scheduling algorithm that guarantees the minimum similar execution delay.Firstly,a graph partitioning algorithm based on the largest block is designed to split the neural network model into multiple subgraphs.Then,the operators that meet the conditions is replaced according to the defined operator replacement rules.Finally,the optimal scheduling method based on backtracking is used to schedule the computational graph.Network models such as Inception-v3,ResNet-50,and RandWire are selected for testing.The experimental results show that the algorithm designed in this paper can achieve a 1.6×speedup compared with the existing sequential execution methods.展开更多
Battery pack capacity estimation under real-world operating conditions is important for battery performance optimization and health management,contributing to the reliability and longevity of batterypowered systems.Ho...Battery pack capacity estimation under real-world operating conditions is important for battery performance optimization and health management,contributing to the reliability and longevity of batterypowered systems.However,complex operating conditions,coupling cell-to-cell inconsistency,and limited labeled data pose great challenges to accurate and robust battery pack capacity estimation.To address these issues,this paper proposes a hierarchical data-driven framework aimed at enhancing the training of machine learning models with fewer labeled data.Unlike traditional data-driven methods that lack interpretability,the hierarchical data-driven framework unveils the“mechanism”of the black box inside the data-driven framework by splitting the final estimation target into cell-level and pack-level intermediate targets.A generalized feature matrix is devised without requiring all cell voltages,significantly reducing the computational cost and memory resources.The generated intermediate target labels and the corresponding features are hierarchically employed to enhance the training of two machine learning models,effectively alleviating the difficulty of learning the relationship from all features due to fewer labeled data and addressing the dilemma of requiring extensive labeled data for accurate estimation.Using only 10%of degradation data,the proposed framework outperforms the state-of-the-art battery pack capacity estimation methods,achieving mean absolute percentage errors of 0.608%,0.601%,and 1.128%for three battery packs whose degradation load profiles represent real-world operating conditions.Its high accuracy,adaptability,and robustness indicate the potential in different application scenarios,which is promising for reducing laborious and expensive aging experiments at the pack level and facilitating the development of battery technology.展开更多
Sewage source heat pump unit operates under partial load most of the time, and study on the law of coefficient of performance (COP) of the unit varying with load ratio can provide basis for the heat pump units running...Sewage source heat pump unit operates under partial load most of the time, and study on the law of coefficient of performance (COP) of the unit varying with load ratio can provide basis for the heat pump units running in high efficiency. A mathematical model determining COP, evaporation temperature and condensation temperature of a single unit was proposed. Under the condition of uniform load distribution, the model was established according to different ways of bearing partial load with the same type multiple units but different parallel operation models, and the operation characteristics of units were analyzed as well. Results show that the single screw water-source heat pump can maintain high COP only at 60%—100% load ratio, COP decreases sharply with the decrease of load ratio, and the units with parallel operation are controlled by the load ratio of a single unit according to the reduction of total load which can keep the average COP at high level within a wide load range.展开更多
A series-parallel hydraulic hybrid system applied to public buses is put forward,and parameters of key components are analyzed and determined.Energy management strategy based on logic threshold is designed which is ai...A series-parallel hydraulic hybrid system applied to public buses is put forward,and parameters of key components are analyzed and determined.Energy management strategy based on logic threshold is designed which is aimed at efficient operation of the overall system considering the operational characteristic of the components and taking the curves of engine,hydraulic pump/motor and hydraulic pump as the main design basis;regenerative control strategy which makes regenerative brake system and frictional brake system work harmoniously is designed to raise recovery rate of regenerative brake energy.System dynamic modeling and simulation results show that the energy control strategy designed here is able to adapt system to changes of working condition and switch the operating mode reasonably.The regenerative braking control strategy is effective in raising the utilization of energy and improving fuel economy.展开更多
This paper presents an investigation on a voltage and frequency controller (VFC), which functions as an improved elec-tronic load controller (IELC) for parallel operated isolated asynchronous generators (IAGs) in an a...This paper presents an investigation on a voltage and frequency controller (VFC), which functions as an improved elec-tronic load controller (IELC) for parallel operated isolated asynchronous generators (IAGs) in an autonomous hydro power generation system. In such type of hydro scheme whole generating system is isolated from the grid and supply electricity to the remote communities. The single point operation of these generators is realized, in such a manner that excitation capacitors, speeds, load, voltage, currents of generators remain constant under various operating conditions. The proposed controller consists of a 3-leg IGBT (Insulated Gate Bipolar Transistor) based voltage source converter (VSC) and a DC chopper with an auxiliary load at the DC bus of the VSC. The IELC controls the reactive and active powers simultaneously for controlling the voltage and frequency under varying consumer loads. Along with voltage and frequency control through single point operation of IAGs driven by uncontrolled pico hydro turbines, the IELC meets the power quality standard an IEEE-519 and it keeps the total harmonic distortion (THD) of the terminal voltage and currents within the limit of 5%. Here the proposed electrical system along with its controller is modeled in MATLAB along with Simulink and PSB (Power System Block-set) toolboxes. Simulation results are presented to demonstrate the capability of proposed controller for an isolated generating system.展开更多
We developed a general framework for parallel all-optical logic operations with independent phase control of arbitrary orthogonal polarization state enabled by a single-layer metasurface.A pair of orthogonal circular ...We developed a general framework for parallel all-optical logic operations with independent phase control of arbitrary orthogonal polarization state enabled by a single-layer metasurface.A pair of orthogonal circular polarized bases are used as two input channels of the logic operator,and their four combinations perfectly match various binary input states.Correspondingly,distinct phase profiles are encoded into the metasurface,which enables parallel operation of the two logic gates by way of polarization switching.It allows for an efficient and compact way to implement multi-channel multiplexed logic gate operations with the capability of fast optical computing at the chip scale.展开更多
基金Supported by the National Key Research and Development Project of China(No.2020AAA0104603)the National Natural Science Foundation of China(No.61834005,61772417)the Shaanxi Province Key R&D Plan(No.2021GY-029).
文摘To improve the inference efficiency of convolutional neural networks(CNN),the existing neural networks mainly adopt heuristic and dynamic programming algorithms to realize parallel scheduling among operators.Heuristic scheduling algorithms can generate local optima easily,while the dynamic programming algorithm has a long convergence time for complex structural models.This paper mainly studies the parallel scheduling between operators and proposes an inter-operator parallelism schedule(IOPS)scheduling algorithm that guarantees the minimum similar execution delay.Firstly,a graph partitioning algorithm based on the largest block is designed to split the neural network model into multiple subgraphs.Then,the operators that meet the conditions is replaced according to the defined operator replacement rules.Finally,the optimal scheduling method based on backtracking is used to schedule the computational graph.Network models such as Inception-v3,ResNet-50,and RandWire are selected for testing.The experimental results show that the algorithm designed in this paper can achieve a 1.6×speedup compared with the existing sequential execution methods.
基金supported by the National Outstanding Youth Science Fund Project of National Natural Science Foundation of China[Grant No.52222708]the Natural Science Foundation of Beijing Municipality[Grant No.3212033]。
文摘Battery pack capacity estimation under real-world operating conditions is important for battery performance optimization and health management,contributing to the reliability and longevity of batterypowered systems.However,complex operating conditions,coupling cell-to-cell inconsistency,and limited labeled data pose great challenges to accurate and robust battery pack capacity estimation.To address these issues,this paper proposes a hierarchical data-driven framework aimed at enhancing the training of machine learning models with fewer labeled data.Unlike traditional data-driven methods that lack interpretability,the hierarchical data-driven framework unveils the“mechanism”of the black box inside the data-driven framework by splitting the final estimation target into cell-level and pack-level intermediate targets.A generalized feature matrix is devised without requiring all cell voltages,significantly reducing the computational cost and memory resources.The generated intermediate target labels and the corresponding features are hierarchically employed to enhance the training of two machine learning models,effectively alleviating the difficulty of learning the relationship from all features due to fewer labeled data and addressing the dilemma of requiring extensive labeled data for accurate estimation.Using only 10%of degradation data,the proposed framework outperforms the state-of-the-art battery pack capacity estimation methods,achieving mean absolute percentage errors of 0.608%,0.601%,and 1.128%for three battery packs whose degradation load profiles represent real-world operating conditions.Its high accuracy,adaptability,and robustness indicate the potential in different application scenarios,which is promising for reducing laborious and expensive aging experiments at the pack level and facilitating the development of battery technology.
基金Supported by National Natural Science Foundation of China (No. 50578048)"Heating, Gas, Ventilation and Air Conditioning" Key Laboratory Open Subject in Beijing (No. KF200710)the Postdoctoral Researcher Science Foundation of China (No. 20090450986)
文摘Sewage source heat pump unit operates under partial load most of the time, and study on the law of coefficient of performance (COP) of the unit varying with load ratio can provide basis for the heat pump units running in high efficiency. A mathematical model determining COP, evaporation temperature and condensation temperature of a single unit was proposed. Under the condition of uniform load distribution, the model was established according to different ways of bearing partial load with the same type multiple units but different parallel operation models, and the operation characteristics of units were analyzed as well. Results show that the single screw water-source heat pump can maintain high COP only at 60%—100% load ratio, COP decreases sharply with the decrease of load ratio, and the units with parallel operation are controlled by the load ratio of a single unit according to the reduction of total load which can keep the average COP at high level within a wide load range.
基金Supported by the National Natural Science Foundation of China(No.50875054)Weihai Science and Technology Development Plan Project(No.2012DXGJ13)
文摘A series-parallel hydraulic hybrid system applied to public buses is put forward,and parameters of key components are analyzed and determined.Energy management strategy based on logic threshold is designed which is aimed at efficient operation of the overall system considering the operational characteristic of the components and taking the curves of engine,hydraulic pump/motor and hydraulic pump as the main design basis;regenerative control strategy which makes regenerative brake system and frictional brake system work harmoniously is designed to raise recovery rate of regenerative brake energy.System dynamic modeling and simulation results show that the energy control strategy designed here is able to adapt system to changes of working condition and switch the operating mode reasonably.The regenerative braking control strategy is effective in raising the utilization of energy and improving fuel economy.
文摘This paper presents an investigation on a voltage and frequency controller (VFC), which functions as an improved elec-tronic load controller (IELC) for parallel operated isolated asynchronous generators (IAGs) in an autonomous hydro power generation system. In such type of hydro scheme whole generating system is isolated from the grid and supply electricity to the remote communities. The single point operation of these generators is realized, in such a manner that excitation capacitors, speeds, load, voltage, currents of generators remain constant under various operating conditions. The proposed controller consists of a 3-leg IGBT (Insulated Gate Bipolar Transistor) based voltage source converter (VSC) and a DC chopper with an auxiliary load at the DC bus of the VSC. The IELC controls the reactive and active powers simultaneously for controlling the voltage and frequency under varying consumer loads. Along with voltage and frequency control through single point operation of IAGs driven by uncontrolled pico hydro turbines, the IELC meets the power quality standard an IEEE-519 and it keeps the total harmonic distortion (THD) of the terminal voltage and currents within the limit of 5%. Here the proposed electrical system along with its controller is modeled in MATLAB along with Simulink and PSB (Power System Block-set) toolboxes. Simulation results are presented to demonstrate the capability of proposed controller for an isolated generating system.
基金supported by the National Key Research and Development Program of China(No.2019YFE0107400)the National Natural Science Foundation of China(No.52005147).
文摘We developed a general framework for parallel all-optical logic operations with independent phase control of arbitrary orthogonal polarization state enabled by a single-layer metasurface.A pair of orthogonal circular polarized bases are used as two input channels of the logic operator,and their four combinations perfectly match various binary input states.Correspondingly,distinct phase profiles are encoded into the metasurface,which enables parallel operation of the two logic gates by way of polarization switching.It allows for an efficient and compact way to implement multi-channel multiplexed logic gate operations with the capability of fast optical computing at the chip scale.