Large-scale Language Models(LLMs)have achieved significant breakthroughs in Natural Language Processing(NLP),driven by the pre-training and fine-tuning paradigm.While this approach allows models to specialize in speci...Large-scale Language Models(LLMs)have achieved significant breakthroughs in Natural Language Processing(NLP),driven by the pre-training and fine-tuning paradigm.While this approach allows models to specialize in specific tasks with reduced training costs,the substantial memory requirements during fine-tuning present a barrier to broader deployment.Parameter-Efficient Fine-Tuning(PEFT)techniques,such as Low-Rank Adaptation(LoRA),and parameter quantization methods have emerged as solutions to address these challenges by optimizing memory usage and computational efficiency.Among these,QLoRA,which combines PEFT and quantization,has demonstrated notable success in reducing memory footprints during fine-tuning,prompting the development of various QLoRA variants.Despite these advancements,the quantitative impact of key variables on the fine-tuning performance of quantized LLMs remains underexplored.This study presents a comprehensive analysis of these key variables,focusing on their influence across different layer types and depths within LLM architectures.Our investigation uncovers several critical findings:(1)Larger layers,such as MLP layers,can maintain performance despite reductions in adapter rank,while smaller layers,like self-attention layers,aremore sensitive to such changes;(2)The effectiveness of balancing factors depends more on specific values rather than layer type or depth;(3)In quantization-aware fine-tuning,larger layers can effectively utilize smaller adapters,whereas smaller layers struggle to do so.These insights suggest that layer type is a more significant determinant of fine-tuning success than layer depth when optimizing quantized LLMs.Moreover,for the same discount of trainable parameters,reducing the trainable parameters in a larger layer is more effective in preserving fine-tuning accuracy than in a smaller one.This study provides valuable guidance for more efficient fine-tuning strategies and opens avenues for further research into optimizing LLM fine-tuning in resource-constrained environments.展开更多
Poisson's equation is solved numerically by two direct methods, viz. Block Cyclic Reduction (BCR) method and Fourier Method. Qualitative and quantitative comparison between the numerical solutions obtained by two ...Poisson's equation is solved numerically by two direct methods, viz. Block Cyclic Reduction (BCR) method and Fourier Method. Qualitative and quantitative comparison between the numerical solutions obtained by two methods indicates that BCR method is superior to Fourier method in terms of speed and accuracy. Therefore. BCR method is applied to solve (?)2(?)= ζ and (?)2X= D from observed vorticity and divergent values. Thereafter the rotational and divergent components of the horizontal monsoon wind in the lower troposphere are reconstructed and are com pared with the results obtained by Successive Over-Relaxation (SOR) method as this indirect method is generally in more use for obtaining the streamfunction ((?)) and velocity potential (X) fields in NWP models. It is found that the results of BCR method are more reliable than SOR method.展开更多
Knowledge about crop growth processes in relation to N limitation is necessary to optimize N management in farming system. Plant-based diagnostic method, for instance nitrogen nutrition index （NNI） were used to dete...Knowledge about crop growth processes in relation to N limitation is necessary to optimize N management in farming system. Plant-based diagnostic method, for instance nitrogen nutrition index （NNI） were used to determine the crop nitrogen status. This study determines the relationship of NNI with agronomic nitrogen use efficiency （AEN）, tuber yield, radiation use efficiency （RUE） and leaf parameters including leaf area index （LAI）, areal leaf N content （NJ and leaf N concentration （N0. Potatoes were grown in field at three N levels： no N （N 1）, 150 kg N ha^-1 （N2）, 300 kg N ha^-1 （N3）. N deficiency was quantified by NNI and RUE was generally calculated by estimating of the light absorbance on leaf area. NNI was used to evaluate the N effect on tuber yield, RUE, LAI, NAL, and NL. The results showed that NNI was negatively correlated with AEN, N deficiencies （NNI〈 1） which occurred for N 1 and N2 significantly reduced LAI, NL and tuber yield; whereas the N deficiencies had a relative small effect on NAL and RUE. To remove any effect other than N on these parameters, the actual ratio to maximum values were calculated for each developmental linear relationships were obtained between NNI and tuber RUE to NNI. stage of potatoes. When the NNI ranged from 0.4 to 1, positive yield, LAI, NL, while a nonlinear regression fitted the response of展开更多
We propose the maximin efficiency robust test（MERT） for multiple nuisance parameters based on theories about the maximin efficiency robust test for only one nuisance parameter and investigate some theoretical proper...We propose the maximin efficiency robust test（MERT） for multiple nuisance parameters based on theories about the maximin efficiency robust test for only one nuisance parameter and investigate some theoretical properties about this robust test.We explore some theoretical properties about the power of the MERT for multiple nuisance parameters in a specified scenario intuitively further more.We also propose a meaningful example from statistical genetic field to which the MERT for multiple nuisance parameters can be well applied.Extensive simulation studies are conducted to testify the robustness of the MERT for multiple nuisance parameters.展开更多
Accurate basic data are necessary to support performance-based design for achieving carbon peak and carbon neutral targets in the building sector.Meteorological parameters are the prerequisites of building thermal eng...Accurate basic data are necessary to support performance-based design for achieving carbon peak and carbon neutral targets in the building sector.Meteorological parameters are the prerequisites of building thermal engineering design,heating ventilation and air conditioning design,and energy consumption simulations.Focusing on the key issues such as low spatial coverage and the lack of daily or higher time resolution data,daily and hourly models of the surface meteorological data and solar radiation were established and evaluated.Surface meteorological data and solar radiation data were generated for 1019 cities and towns in China from 1988 to 2017.The data were carefully compared,and the accuracy was proved to be high.All the meteorological parameters can be assessed in the building sector via a sharing platform.Then,country-level meteorological parameters were developed for energy-efficient building assessment in China,based on actual meteorological data in the present study.This set of meteorological parameters may facilitate engineering applications as well as allowing the updating and expansion of relevant building energy efficiency standards.The study was supported by the National Science and Technology Major Project of China during the 13th Five-Year Plan Period,named Fundamental parameters on building energy efficiency in China,comprising of 15 top-ranking universities and institutions in China.展开更多
Ethylene cracking process is the core production process in ethylene industry,and is paid more attention to reduce high energy consumption.Because of the interdependent relationships between multi-flow allocation and ...Ethylene cracking process is the core production process in ethylene industry,and is paid more attention to reduce high energy consumption.Because of the interdependent relationships between multi-flow allocation and multi-parameter setting in cracking process,it is difficult to find the overall energy efficiency scheduling for the purpose of saving energy.The traditional scheduling solutions with optimal economic benefit are not applicable for energy efficiency scheduling issue due to the neglecting of recycle and lost energy,as well as critical operation parameters as coil outlet pressure(COP)and dilution ratio.In addition,the scheduling solutions mostly regard each cracking furnace as an elementary unit,regardless of the coordinated operation of internal dual radiation chambers(DRC).Therefore,to improve energy utilization and production operation,a novel energy efficiency scheduling solution for ethylene cracking process is proposed in this paper.Specifically,steam heat recycle and exhaust heat loss are considered in cracking process based on 6 types of extreme learning machine(ELM)based cracking models incorporating DRC operation and three operation parameters as coil outlet temperature(COT),COP,and dilution ratio according to semi-mechanism analysis.Then to provide long-term decision-making basis for energy efficiency scheduling,overall energy efficiency indexes,including overall output per unit net energy input(OONE),output-input ratio per unit net energy input(ORNE),exhaust gas heat loss ratio(EGHL),are designed based on input-output analysis in terms of material and energy flows.Finally,a multiobjective evolutionary algorithm based on decomposition(MOEA/D)is employed to solve the formulated multi-objective mixed-integer nonlinear programming(MOMINLP)model.The validities of the proposed scheduling solution are illustrated through a case study.The scheduling results demonstrate that an optimal balance between multi-flow allocation,multi-parameter setting,and DRC coordinated operation is reached,which achieves 3.37%and 2.63%decreases in net energy input for same product output and conversion ratio,as well as the 1.56%decrease in energy loss ratio.展开更多
The initial efficiency is a very important criterion for carbon anode material of Li-ion battery.The relationship between initial efficiency and structure parameters of carbon anode material of Li-ion battery was inve...The initial efficiency is a very important criterion for carbon anode material of Li-ion battery.The relationship between initial efficiency and structure parameters of carbon anode material of Li-ion battery was investigated by an artificial intelligence approach called Random Forests using D10,D50,D90,BET specific surface area and TP density as inputs,initial efficiency as output.The results give good classification performance with 91%accuracy.The variable importance analysis results show the impact of 5 variables on the initial efficiency descends in the order of D90,TP density,BET specific surface area,D50 and D10;smaller D90 and larger TP density have positive impact on initial efficiency.The contribution of BET specific surface area on classification is only 18.74%,which indicates the shortcoming of BET specific surface area as a widely used parameter for initial efficiency evaluation.展开更多
In the domain of knowledge graph embedding,conventional approaches typically transform entities and relations into continuous vector spaces.However,parameter efficiency becomes increasingly crucial when dealing with l...In the domain of knowledge graph embedding,conventional approaches typically transform entities and relations into continuous vector spaces.However,parameter efficiency becomes increasingly crucial when dealing with large-scale knowledge graphs that contain vast numbers of entities and relations.In particular,resource-intensive embeddings often lead to increased computational costs,and may limit scalability and adaptability in practical environ-ments,such as in low-resource settings or real-world applications.This paper explores an approach to knowledge graph representation learning that leverages small,reserved entities and relation sets for parameter-efficient embedding.We introduce a hierarchical attention network designed to refine and maximize the representational quality of embeddings by selectively focusing on these reserved sets,thereby reducing model complexity.Empirical assessments validate that our model achieves high performance on the benchmark dataset with fewer parameters and smaller embedding dimensions.The ablation studies further highlight the impact and contribution of each component in the proposed hierarchical attention structure.展开更多
In order to study the effects of geometric parameters of the rudder on the hydrodynamic performance of the propeller-rudder system,the surface panel method is used to build the numerical model of the steady interactio...In order to study the effects of geometric parameters of the rudder on the hydrodynamic performance of the propeller-rudder system,the surface panel method is used to build the numerical model of the steady interaction between the propeller and rudder to analyze the relevant factors.The interaction between the propeller and rudder is considered through the induced velocities,which are circumferentially averaged,so the unsteady problem is translated to steady state.An iterative calculation method is used until the hydrodynamic performance converges.Firstly,the hydrodynamic performance of the chosen propeller-rudder system is calculated,and the comparison between the calculated results and the experimental data indicates that the calculation program is reliable.Then,the variable parameters of rudder are investigated,and the calculation results show that the propeller-rudder spacing has a negative relationship with the efficiency of the propeller-rudder system,and the rudder span has an optimal match range with the propeller diameter.Futhermore,the rudder chord and thickness both have a positive correlation with the hydrodynamic performance of the propeller-rudder system.展开更多
A detailed analysis of operational process and principle of ammonia-recovery system in the modified equipment of flax fiber,which will be applied to parameters optimizing of the ammoniarecovery system as a foundationa...A detailed analysis of operational process and principle of ammonia-recovery system in the modified equipment of flax fiber,which will be applied to parameters optimizing of the ammoniarecovery system as a foundational principle,is presented. According to the principle,an ammonia compressor,whose working conditions are based on key operational parameters of the whole ammoniarecovery system, is the mainly energy-consumption part of ammonia-recovery system in the modified equipment of flax fiber. A generally mathematical model based on work efficiency of an ammonia compressor is founded,which is available to rate effective work and energy consumption of the ammonia compressor. The optimum operation-efficiency of the ammonia compressor is chosen as the goal to analyze and calculate the key operational parameters of the ammonia-recovery system. In the above analyzing and calculating,a mathematical model on ammonia flowing from the reactor to the register 1 is developed,in order to provide further understanding of the principle of an ammonia-recovery system. At the meantime,the ammonia flow regime in the pipeline and the process of ammonia inflation and deflation from the reactor to the register 1 are taken separately into account in the model. An iterative method is for obtaining parametric solutions of the mathematical model on ammonia flowing from the reactor to the register 1 and the key operational parameters of the ammoniarecovery system. A parametric analysis is put forward to complete showing the ammonia velocity or the state of the reactor and the register 1. The key optimized parameters will be achieved in term of the minimum efficiency after comparing the work efficiencies of an ammonia compressor at different working conditions.展开更多
Laboratory scale model of DMMBF （dual mixed media biofilter） were designed and installed in AI-Mustansiriya University Environmental Hydraulic Lab. Experiments were conducted using two mixed layers through PVR colum...Laboratory scale model of DMMBF （dual mixed media biofilter） were designed and installed in AI-Mustansiriya University Environmental Hydraulic Lab. Experiments were conducted using two mixed layers through PVR column--2.2 m height and 300 mm diameter. The first mixed media filter of depth 640mm mixed of sand, rice husk and granular activated carbon. The percentage volume mix is 1：1：1. While the other mixed media of depth 740 mm, consisting of coal, crash porcelinaite, rock and granite with equally percentage volume. Fifty samples were collected during the experiments, which was spread over a period of forty two weeks. The obtained results indicate that when the flow loading raised from 0.15 L/min to 2.7 L/rain, the removal efficiency of BOD decreased 8%-11%, and the removal efficiency of COD deceased 3%-4%, while the removal efficiency of turbidity increased with the decreasing of hydraulic loading. The results showed that the removal efficiency of turbidity is more than 95% at the lower discharge （0.15 L/min）. Therefore, infiltration should be conservatively designed using low loading rates.展开更多
The gestation and occurrence of strong earthquakes are closely related to fault activity, which is not only revealed by abundant experimentation and seismism but also proved by modern seismology. On the Chinese mainla...The gestation and occurrence of strong earthquakes are closely related to fault activity, which is not only revealed by abundant experimentation and seismism but also proved by modern seismology. On the Chinese mainland, the relation between earthquake activity and active faults is one of the bases for partitioning potential seismic sources, analyzing the seismotectoulcs and estimating location of strong earthquakes.Due to the nonuniformity of earth media, instability of observation systems and disturbance of the environment, etc, the variety of observational data is complicated, that is, there is no absolutely ＂normal＂ or ＂abnormal＂, and seismic anomalies can be divided into many mutually exdusive＂ abnormal states＂. In different conditions of combined time-spacestrength, determining seismic anomalies by different monomial forecast methods and its efficiency could be different due to the uncertainty of a precursor itself or complexity of the relationship between a precursor and earthquake gestation. It is very difficult to discover and dispose of this difference in actual application in a ＂two-state＂ model. But in a ＂multi-state＂ model, the difference can be easily reflected and the optimal combination of forecasting parameters for a forecast method can also be determined easily. Based on the ＂multi-state＂ precursory model and the optimization method for parameters of earthquake forecast model under the condition of optimal forecast efficiency, the relationship of the spatial location of earthquake with M ≥ 6.0 and active faults in three seismic belts are analyzed. The results demonstrate that in the Hetao Seismic Belt, seismicity is mostly concentrated in the range of 20 km along the fault, the optimization model can forecast the location of potential earthquakes of M ≥ 6.0 near the faults with a relatively high accuracy and the reliability is 0.5 ; while in the Qilian Mt. Seismic Belt, the reliability only reaches 0.14 when we use the model to estimate earthquakes within 30 km range along the faults. The ＂multi-state＂ precursory model, the efficiency-evaluating model and the parameter selection of individual earthquake forecast model based on optimal efficiency are of certain revelatory and practicable meanings for developing knowledge about precursors, investigating the laws of earthquake preparation and searching for optimal forecasting methods.展开更多
Solid oxide fuel cell combined with heat and power(SOFC-CHP)system is a distributed power generation system with low pollution and high efficiency.In this paper,a 10 kW SOFC-CHP system model using syngas was built in ...Solid oxide fuel cell combined with heat and power(SOFC-CHP)system is a distributed power generation system with low pollution and high efficiency.In this paper,a 10 kW SOFC-CHP system model using syngas was built in Aspen plus.Key operating parameters,such as steam to fuel ratio,stack temperature,reformer temperature,air flow rate,and air preheating temperature,were analyzed.Optimization was conducted based on the simulation results.Results suggest that higher steam to fuel ratio is beneficial to the electrical efficiency,but it might decrease the gross system efficiency.Higher stack and reformer temperatures contribute to the electrical efficiency,and the optimal operating temperatures of stack and reformer when considering the stack degradation are 750℃and 700℃,respectively.The air preheating temperature barely affects the electrical efficiency but affects the thermal efficiency and the gross system efficiency,the recommended value is around 600℃under the reference condition.展开更多
In the modern era of manufacturing, it is important to optimize every design parameter in product development stage to reduce cost, material usage and to achieve the desired efficacy level. There are various models wh...In the modern era of manufacturing, it is important to optimize every design parameter in product development stage to reduce cost, material usage and to achieve the desired efficacy level. There are various models which serve those purposes, for instance, Design of Experiment (DoE) is used to check the parameters after adopting optimization tactics which results in reduced cost or saving operating time. In this regard, this research aims to construct a DoE model on a portable workstation to optimize its design parameters. The methodology of DOE would be a 2 level 3 factors full factorial DOE which is conducted to determine the optimal value for three design parameters (factors) which are material density, the length of the table and the length of the table stand in terms of the response which is the required time of fold ability function of the portable workstation. Based upon the evaluated interactions between the parameters, the optimized parameters are chosen for responses. Here, the resultant design parameters are at their lowest level, so the goal of time efficiency in fold ability function is achieved. This similar sort of DoE can be implemented in the furniture and other manufacturing industries who wish to optimize their material usage as well as increase efficiency and reduce cycle time.展开更多
In designing a canal system, a major problem is to decide what conveyance parameter to apply in the calculations. Since basic knowledge on this subject is lacking, it is usually taken from literatures. Most of the irr...In designing a canal system, a major problem is to decide what conveyance parameter to apply in the calculations. Since basic knowledge on this subject is lacking, it is usually taken from literatures. Most of the irrigation projects in Ethiopia are found to work below their expectation. One of the main reasons is the conveyance parameters variation from the expected (design) value which ultimately affects the envisioned conveyance efficiency. To evaluate this variation, Fentale irrigation scheme was used as case study. The conveyance efficiency used at the design stage was 80%, which was within Food and Agricultural Organization of the united nations recommendations;while the field survey value was 17%. Such huge variation was due to the fact that the assigned conveyance parameter values (roughness coefficient, hydraulic radius and bed slope) no longer represent the current situation of the scheme. Such variation has resulted in increase in the depth and top width of the water surface which further resulted in 13% and 3% increase in wetted perimeter and top width of the canal, respectively. Thus this study suggests that conveyance parameters shall be derived from history of existing irrigation schemes in a country, rather than adopting it from standard literatures. As such construction quality, maintenance activities and technological transfer activities in a country shall be seen in deciding the conveyance parameters. The study also suggests that the ever increasing water shortage in an irrigation project could be managed by proper maintenance of the entire irrigation system.展开更多
This paper presents methods for computing a second-order sensitivity matrix and the Hessian matrix of eigenvalues and eigenvectors of multiple parameter structures. Second-order perturbations of eigenvalues and eigenv...This paper presents methods for computing a second-order sensitivity matrix and the Hessian matrix of eigenvalues and eigenvectors of multiple parameter structures. Second-order perturbations of eigenvalues and eigenvectors are transformed into multiple parameter forms,and the second-order perturbation sensitivity matrices of eigenvalues and eigenvectors are developed.With these formulations,the efficient methods based on the second-order Taylor expansion and second-order perturbation are obtained to estimate changes of eigenvalues and eigenvectors when the design parameters are changed. The presented method avoids direct differential operation,and thus reduces difficulty for computing the second-order sensitivity matrices of eigenpairs.A numerical example is given to demonstrate application and accuracy of the proposed method.展开更多
A numerical model of thermoelectric module （TEM） is created by academic analysis,and the impacts of the resistance ratio and thermoelement size on the output power and thermoelectric efficiency of the TEM are analyz...A numerical model of thermoelectric module （TEM） is created by academic analysis,and the impacts of the resistance ratio and thermoelement size on the output power and thermoelectric efficiency of the TEM are analyzed by the MATLAB numerical calculation.The numerical model is validated by the ANSYS thermal,electrical,and structural coupling simulation.The effects of the variable physical property parameters and contact effect on the output power and thermoelectric efficiency are evaluated,and the concept of aspect ratio optimal domain is proposed,which provides a new design approach for the TEM.展开更多
The simulation of the brake disc temperature is an important tool in the development of passenger cars.Nowadays thermal models of brake discs are real-time applications,running on electronic control units(ECUs)of cars...The simulation of the brake disc temperature is an important tool in the development of passenger cars.Nowadays thermal models of brake discs are real-time applications,running on electronic control units(ECUs)of cars to improve the vehicle safety in several ways.These models are often working with full empirical methods,leading to large deviations between calculation and measurement.To meet the requirements of automotive safety integrity levels(ASILs),these thermal models cannot rely on the state of the art ambient air temperature sensors,which causes unacceptable deviations.Focusing on numerical efficient thermal simulations,a new approach of a semi-analytical thermal network for simulating the brake disc temperature with minimal effort is proposed.The thermal network is based on lumped parameters,using two thermal capacity nodes and a constant ambient temperature.Using semi-analytical correlations,the model can be adapted to different geometries and car lines effortlessly.The empirical parameters of the model result only from two standardized tests.These parameters are used to evaluate the estimation accuracy in real driving situations.Additionally,the adaptability is tested for two different car lines and four brake disc dimensions.These tests are initially performed with unchanged parameters and afterwards with refitted parameters.The model shows a good estimation for the tested load cases.Compared to the state of the art,the proposed model is less accurate than complex finite element method(FEM)models and computational fluid dynamic(CFD)approaches,but shows a higher accuracy and better adaptability than other lumped parameter models with comparable numerical effort.Hence,possible applications can be dimensioning the brake system in the development process of new car lines or a real-time simulation on the latest ECU in the vehicle.展开更多
In this article, research was conducted to improve Linter machines that remove short fibers remaining in ginned cotton seeds at cotton ginneries. The study examined the effect of changing the dimensions of the brush d...In this article, research was conducted to improve Linter machines that remove short fibers remaining in ginned cotton seeds at cotton ginneries. The study examined the effect of changing the dimensions of the brush drum, guide and mesh surface in the cleaning device proposed for the linting machine on the movement of the peg and the cleaning efficiency, and the highest level of efficiency in separating impurities from the peg was determined. During the study, the main factors influencing the effective operation of the improved linting machine were identified, the limits of their values were determined, and studies were carried out using the mathematical modeling method. As a result, at the values of the given coefficients, efficient operation of the improved linting machine was observed, that is, the lint cleaning efficiency reached 55.1%.展开更多
基金supported by the National Key R&D Program of China(No.2021YFB0301200)National Natural Science Foundation of China(No.62025208).
文摘Large-scale Language Models(LLMs)have achieved significant breakthroughs in Natural Language Processing(NLP),driven by the pre-training and fine-tuning paradigm.While this approach allows models to specialize in specific tasks with reduced training costs,the substantial memory requirements during fine-tuning present a barrier to broader deployment.Parameter-Efficient Fine-Tuning(PEFT)techniques,such as Low-Rank Adaptation(LoRA),and parameter quantization methods have emerged as solutions to address these challenges by optimizing memory usage and computational efficiency.Among these,QLoRA,which combines PEFT and quantization,has demonstrated notable success in reducing memory footprints during fine-tuning,prompting the development of various QLoRA variants.Despite these advancements,the quantitative impact of key variables on the fine-tuning performance of quantized LLMs remains underexplored.This study presents a comprehensive analysis of these key variables,focusing on their influence across different layer types and depths within LLM architectures.Our investigation uncovers several critical findings:(1)Larger layers,such as MLP layers,can maintain performance despite reductions in adapter rank,while smaller layers,like self-attention layers,aremore sensitive to such changes;(2)The effectiveness of balancing factors depends more on specific values rather than layer type or depth;(3)In quantization-aware fine-tuning,larger layers can effectively utilize smaller adapters,whereas smaller layers struggle to do so.These insights suggest that layer type is a more significant determinant of fine-tuning success than layer depth when optimizing quantized LLMs.Moreover,for the same discount of trainable parameters,reducing the trainable parameters in a larger layer is more effective in preserving fine-tuning accuracy than in a smaller one.This study provides valuable guidance for more efficient fine-tuning strategies and opens avenues for further research into optimizing LLM fine-tuning in resource-constrained environments.
文摘Poisson's equation is solved numerically by two direct methods, viz. Block Cyclic Reduction (BCR) method and Fourier Method. Qualitative and quantitative comparison between the numerical solutions obtained by two methods indicates that BCR method is superior to Fourier method in terms of speed and accuracy. Therefore. BCR method is applied to solve (?)2(?)= ζ and (?)2X= D from observed vorticity and divergent values. Thereafter the rotational and divergent components of the horizontal monsoon wind in the lower troposphere are reconstructed and are com pared with the results obtained by Successive Over-Relaxation (SOR) method as this indirect method is generally in more use for obtaining the streamfunction ((?)) and velocity potential (X) fields in NWP models. It is found that the results of BCR method are more reliable than SOR method.
基金supported by the National Key Technology R&D Program (2011BAD12B03)
文摘Knowledge about crop growth processes in relation to N limitation is necessary to optimize N management in farming system. Plant-based diagnostic method, for instance nitrogen nutrition index （NNI） were used to determine the crop nitrogen status. This study determines the relationship of NNI with agronomic nitrogen use efficiency （AEN）, tuber yield, radiation use efficiency （RUE） and leaf parameters including leaf area index （LAI）, areal leaf N content （NJ and leaf N concentration （N0. Potatoes were grown in field at three N levels： no N （N 1）, 150 kg N ha^-1 （N2）, 300 kg N ha^-1 （N3）. N deficiency was quantified by NNI and RUE was generally calculated by estimating of the light absorbance on leaf area. NNI was used to evaluate the N effect on tuber yield, RUE, LAI, NAL, and NL. The results showed that NNI was negatively correlated with AEN, N deficiencies （NNI〈 1） which occurred for N 1 and N2 significantly reduced LAI, NL and tuber yield; whereas the N deficiencies had a relative small effect on NAL and RUE. To remove any effect other than N on these parameters, the actual ratio to maximum values were calculated for each developmental linear relationships were obtained between NNI and tuber RUE to NNI. stage of potatoes. When the NNI ranged from 0.4 to 1, positive yield, LAI, NL, while a nonlinear regression fitted the response of
基金supported by the Natural Science Foundation of China(11401240,11471135)the self-determined research funds of CCNU from the colleges’basic research of MOE(CCNU15A05038,CCNU15ZD011)
文摘We propose the maximin efficiency robust test（MERT） for multiple nuisance parameters based on theories about the maximin efficiency robust test for only one nuisance parameter and investigate some theoretical properties about this robust test.We explore some theoretical properties about the power of the MERT for multiple nuisance parameters in a specified scenario intuitively further more.We also propose a meaningful example from statistical genetic field to which the MERT for multiple nuisance parameters can be well applied.Extensive simulation studies are conducted to testify the robustness of the MERT for multiple nuisance parameters.
基金Project(2018YFC0704500)supported by the National Science and Technology Major Project of China during the 13th Five-Year Plan Period。
文摘Accurate basic data are necessary to support performance-based design for achieving carbon peak and carbon neutral targets in the building sector.Meteorological parameters are the prerequisites of building thermal engineering design,heating ventilation and air conditioning design,and energy consumption simulations.Focusing on the key issues such as low spatial coverage and the lack of daily or higher time resolution data,daily and hourly models of the surface meteorological data and solar radiation were established and evaluated.Surface meteorological data and solar radiation data were generated for 1019 cities and towns in China from 1988 to 2017.The data were carefully compared,and the accuracy was proved to be high.All the meteorological parameters can be assessed in the building sector via a sharing platform.Then,country-level meteorological parameters were developed for energy-efficient building assessment in China,based on actual meteorological data in the present study.This set of meteorological parameters may facilitate engineering applications as well as allowing the updating and expansion of relevant building energy efficiency standards.The study was supported by the National Science and Technology Major Project of China during the 13th Five-Year Plan Period,named Fundamental parameters on building energy efficiency in China,comprising of 15 top-ranking universities and institutions in China.
基金supported by the High-tech Research and Development Program of China(2014AA041802)。
文摘Ethylene cracking process is the core production process in ethylene industry,and is paid more attention to reduce high energy consumption.Because of the interdependent relationships between multi-flow allocation and multi-parameter setting in cracking process,it is difficult to find the overall energy efficiency scheduling for the purpose of saving energy.The traditional scheduling solutions with optimal economic benefit are not applicable for energy efficiency scheduling issue due to the neglecting of recycle and lost energy,as well as critical operation parameters as coil outlet pressure(COP)and dilution ratio.In addition,the scheduling solutions mostly regard each cracking furnace as an elementary unit,regardless of the coordinated operation of internal dual radiation chambers(DRC).Therefore,to improve energy utilization and production operation,a novel energy efficiency scheduling solution for ethylene cracking process is proposed in this paper.Specifically,steam heat recycle and exhaust heat loss are considered in cracking process based on 6 types of extreme learning machine(ELM)based cracking models incorporating DRC operation and three operation parameters as coil outlet temperature(COT),COP,and dilution ratio according to semi-mechanism analysis.Then to provide long-term decision-making basis for energy efficiency scheduling,overall energy efficiency indexes,including overall output per unit net energy input(OONE),output-input ratio per unit net energy input(ORNE),exhaust gas heat loss ratio(EGHL),are designed based on input-output analysis in terms of material and energy flows.Finally,a multiobjective evolutionary algorithm based on decomposition(MOEA/D)is employed to solve the formulated multi-objective mixed-integer nonlinear programming(MOMINLP)model.The validities of the proposed scheduling solution are illustrated through a case study.The scheduling results demonstrate that an optimal balance between multi-flow allocation,multi-parameter setting,and DRC coordinated operation is reached,which achieves 3.37%and 2.63%decreases in net energy input for same product output and conversion ratio,as well as the 1.56%decrease in energy loss ratio.
基金Project(2001AA501433)supported by the National High-Tech Research and Development Program of China
文摘The initial efficiency is a very important criterion for carbon anode material of Li-ion battery.The relationship between initial efficiency and structure parameters of carbon anode material of Li-ion battery was investigated by an artificial intelligence approach called Random Forests using D10,D50,D90,BET specific surface area and TP density as inputs,initial efficiency as output.The results give good classification performance with 91%accuracy.The variable importance analysis results show the impact of 5 variables on the initial efficiency descends in the order of D90,TP density,BET specific surface area,D50 and D10;smaller D90 and larger TP density have positive impact on initial efficiency.The contribution of BET specific surface area on classification is only 18.74%,which indicates the shortcoming of BET specific surface area as a widely used parameter for initial efficiency evaluation.
基金supported by the National Science and Technology Council(NSTC),Taiwan,under Grants Numbers 112-2622-E-029-009 and 112-2221-E-029-019.
文摘In the domain of knowledge graph embedding,conventional approaches typically transform entities and relations into continuous vector spaces.However,parameter efficiency becomes increasingly crucial when dealing with large-scale knowledge graphs that contain vast numbers of entities and relations.In particular,resource-intensive embeddings often lead to increased computational costs,and may limit scalability and adaptability in practical environ-ments,such as in low-resource settings or real-world applications.This paper explores an approach to knowledge graph representation learning that leverages small,reserved entities and relation sets for parameter-efficient embedding.We introduce a hierarchical attention network designed to refine and maximize the representational quality of embeddings by selectively focusing on these reserved sets,thereby reducing model complexity.Empirical assessments validate that our model achieves high performance on the benchmark dataset with fewer parameters and smaller embedding dimensions.The ablation studies further highlight the impact and contribution of each component in the proposed hierarchical attention structure.
基金Supported by the China Postdoctoral Science Foundation(Grant No.2012M512133)the National Natural Science Foundation of China(Grant NO.41176074)the Fundamental Research Funds for the Central University(Grant No.T013513015)
文摘In order to study the effects of geometric parameters of the rudder on the hydrodynamic performance of the propeller-rudder system,the surface panel method is used to build the numerical model of the steady interaction between the propeller and rudder to analyze the relevant factors.The interaction between the propeller and rudder is considered through the induced velocities,which are circumferentially averaged,so the unsteady problem is translated to steady state.An iterative calculation method is used until the hydrodynamic performance converges.Firstly,the hydrodynamic performance of the chosen propeller-rudder system is calculated,and the comparison between the calculated results and the experimental data indicates that the calculation program is reliable.Then,the variable parameters of rudder are investigated,and the calculation results show that the propeller-rudder spacing has a negative relationship with the efficiency of the propeller-rudder system,and the rudder span has an optimal match range with the propeller diameter.Futhermore,the rudder chord and thickness both have a positive correlation with the hydrodynamic performance of the propeller-rudder system.
基金National Science and Technology Support Program,China(No.2012BAF13B03)Program of Shanghai Subject Chief Scientist,China(No.12XD1420300)
文摘A detailed analysis of operational process and principle of ammonia-recovery system in the modified equipment of flax fiber,which will be applied to parameters optimizing of the ammoniarecovery system as a foundational principle,is presented. According to the principle,an ammonia compressor,whose working conditions are based on key operational parameters of the whole ammoniarecovery system, is the mainly energy-consumption part of ammonia-recovery system in the modified equipment of flax fiber. A generally mathematical model based on work efficiency of an ammonia compressor is founded,which is available to rate effective work and energy consumption of the ammonia compressor. The optimum operation-efficiency of the ammonia compressor is chosen as the goal to analyze and calculate the key operational parameters of the ammonia-recovery system. In the above analyzing and calculating,a mathematical model on ammonia flowing from the reactor to the register 1 is developed,in order to provide further understanding of the principle of an ammonia-recovery system. At the meantime,the ammonia flow regime in the pipeline and the process of ammonia inflation and deflation from the reactor to the register 1 are taken separately into account in the model. An iterative method is for obtaining parametric solutions of the mathematical model on ammonia flowing from the reactor to the register 1 and the key operational parameters of the ammoniarecovery system. A parametric analysis is put forward to complete showing the ammonia velocity or the state of the reactor and the register 1. The key optimized parameters will be achieved in term of the minimum efficiency after comparing the work efficiencies of an ammonia compressor at different working conditions.
文摘Laboratory scale model of DMMBF （dual mixed media biofilter） were designed and installed in AI-Mustansiriya University Environmental Hydraulic Lab. Experiments were conducted using two mixed layers through PVR column--2.2 m height and 300 mm diameter. The first mixed media filter of depth 640mm mixed of sand, rice husk and granular activated carbon. The percentage volume mix is 1：1：1. While the other mixed media of depth 740 mm, consisting of coal, crash porcelinaite, rock and granite with equally percentage volume. Fifty samples were collected during the experiments, which was spread over a period of forty two weeks. The obtained results indicate that when the flow loading raised from 0.15 L/min to 2.7 L/rain, the removal efficiency of BOD decreased 8%-11%, and the removal efficiency of COD deceased 3%-4%, while the removal efficiency of turbidity increased with the decreasing of hydraulic loading. The results showed that the removal efficiency of turbidity is more than 95% at the lower discharge （0.15 L/min）. Therefore, infiltration should be conservatively designed using low loading rates.
基金This project was sponsored by the Joint Earthquake Science Foundation of CEA(Grant No.103075 and No.104016)
文摘The gestation and occurrence of strong earthquakes are closely related to fault activity, which is not only revealed by abundant experimentation and seismism but also proved by modern seismology. On the Chinese mainland, the relation between earthquake activity and active faults is one of the bases for partitioning potential seismic sources, analyzing the seismotectoulcs and estimating location of strong earthquakes.Due to the nonuniformity of earth media, instability of observation systems and disturbance of the environment, etc, the variety of observational data is complicated, that is, there is no absolutely ＂normal＂ or ＂abnormal＂, and seismic anomalies can be divided into many mutually exdusive＂ abnormal states＂. In different conditions of combined time-spacestrength, determining seismic anomalies by different monomial forecast methods and its efficiency could be different due to the uncertainty of a precursor itself or complexity of the relationship between a precursor and earthquake gestation. It is very difficult to discover and dispose of this difference in actual application in a ＂two-state＂ model. But in a ＂multi-state＂ model, the difference can be easily reflected and the optimal combination of forecasting parameters for a forecast method can also be determined easily. Based on the ＂multi-state＂ precursory model and the optimization method for parameters of earthquake forecast model under the condition of optimal forecast efficiency, the relationship of the spatial location of earthquake with M ≥ 6.0 and active faults in three seismic belts are analyzed. The results demonstrate that in the Hetao Seismic Belt, seismicity is mostly concentrated in the range of 20 km along the fault, the optimization model can forecast the location of potential earthquakes of M ≥ 6.0 near the faults with a relatively high accuracy and the reliability is 0.5 ; while in the Qilian Mt. Seismic Belt, the reliability only reaches 0.14 when we use the model to estimate earthquakes within 30 km range along the faults. The ＂multi-state＂ precursory model, the efficiency-evaluating model and the parameter selection of individual earthquake forecast model based on optimal efficiency are of certain revelatory and practicable meanings for developing knowledge about precursors, investigating the laws of earthquake preparation and searching for optimal forecasting methods.
基金the National Key R&D Program of China(2017YFB0601903).
文摘Solid oxide fuel cell combined with heat and power(SOFC-CHP)system is a distributed power generation system with low pollution and high efficiency.In this paper,a 10 kW SOFC-CHP system model using syngas was built in Aspen plus.Key operating parameters,such as steam to fuel ratio,stack temperature,reformer temperature,air flow rate,and air preheating temperature,were analyzed.Optimization was conducted based on the simulation results.Results suggest that higher steam to fuel ratio is beneficial to the electrical efficiency,but it might decrease the gross system efficiency.Higher stack and reformer temperatures contribute to the electrical efficiency,and the optimal operating temperatures of stack and reformer when considering the stack degradation are 750℃and 700℃,respectively.The air preheating temperature barely affects the electrical efficiency but affects the thermal efficiency and the gross system efficiency,the recommended value is around 600℃under the reference condition.
文摘In the modern era of manufacturing, it is important to optimize every design parameter in product development stage to reduce cost, material usage and to achieve the desired efficacy level. There are various models which serve those purposes, for instance, Design of Experiment (DoE) is used to check the parameters after adopting optimization tactics which results in reduced cost or saving operating time. In this regard, this research aims to construct a DoE model on a portable workstation to optimize its design parameters. The methodology of DOE would be a 2 level 3 factors full factorial DOE which is conducted to determine the optimal value for three design parameters (factors) which are material density, the length of the table and the length of the table stand in terms of the response which is the required time of fold ability function of the portable workstation. Based upon the evaluated interactions between the parameters, the optimized parameters are chosen for responses. Here, the resultant design parameters are at their lowest level, so the goal of time efficiency in fold ability function is achieved. This similar sort of DoE can be implemented in the furniture and other manufacturing industries who wish to optimize their material usage as well as increase efficiency and reduce cycle time.
文摘In designing a canal system, a major problem is to decide what conveyance parameter to apply in the calculations. Since basic knowledge on this subject is lacking, it is usually taken from literatures. Most of the irrigation projects in Ethiopia are found to work below their expectation. One of the main reasons is the conveyance parameters variation from the expected (design) value which ultimately affects the envisioned conveyance efficiency. To evaluate this variation, Fentale irrigation scheme was used as case study. The conveyance efficiency used at the design stage was 80%, which was within Food and Agricultural Organization of the united nations recommendations;while the field survey value was 17%. Such huge variation was due to the fact that the assigned conveyance parameter values (roughness coefficient, hydraulic radius and bed slope) no longer represent the current situation of the scheme. Such variation has resulted in increase in the depth and top width of the water surface which further resulted in 13% and 3% increase in wetted perimeter and top width of the canal, respectively. Thus this study suggests that conveyance parameters shall be derived from history of existing irrigation schemes in a country, rather than adopting it from standard literatures. As such construction quality, maintenance activities and technological transfer activities in a country shall be seen in deciding the conveyance parameters. The study also suggests that the ever increasing water shortage in an irrigation project could be managed by proper maintenance of the entire irrigation system.
基金Project supported by the 985-Engineering Innovation of Graduate Students of Jilin Universitythe Science and Technology Development Foundation of Jilin Province(20070541)
文摘This paper presents methods for computing a second-order sensitivity matrix and the Hessian matrix of eigenvalues and eigenvectors of multiple parameter structures. Second-order perturbations of eigenvalues and eigenvectors are transformed into multiple parameter forms,and the second-order perturbation sensitivity matrices of eigenvalues and eigenvectors are developed.With these formulations,the efficient methods based on the second-order Taylor expansion and second-order perturbation are obtained to estimate changes of eigenvalues and eigenvectors when the design parameters are changed. The presented method avoids direct differential operation,and thus reduces difficulty for computing the second-order sensitivity matrices of eigenpairs.A numerical example is given to demonstrate application and accuracy of the proposed method.
基金Funded by Guangdong Natural Science Foundation (No.00355991220615019)
文摘A numerical model of thermoelectric module （TEM） is created by academic analysis,and the impacts of the resistance ratio and thermoelement size on the output power and thermoelectric efficiency of the TEM are analyzed by the MATLAB numerical calculation.The numerical model is validated by the ANSYS thermal,electrical,and structural coupling simulation.The effects of the variable physical property parameters and contact effect on the output power and thermoelectric efficiency are evaluated,and the concept of aspect ratio optimal domain is proposed,which provides a new design approach for the TEM.
基金the first publication on the Eurobrake Conference in Dresden Germany in Mai 2019It was only published for the participants of the conference.
文摘The simulation of the brake disc temperature is an important tool in the development of passenger cars.Nowadays thermal models of brake discs are real-time applications,running on electronic control units(ECUs)of cars to improve the vehicle safety in several ways.These models are often working with full empirical methods,leading to large deviations between calculation and measurement.To meet the requirements of automotive safety integrity levels(ASILs),these thermal models cannot rely on the state of the art ambient air temperature sensors,which causes unacceptable deviations.Focusing on numerical efficient thermal simulations,a new approach of a semi-analytical thermal network for simulating the brake disc temperature with minimal effort is proposed.The thermal network is based on lumped parameters,using two thermal capacity nodes and a constant ambient temperature.Using semi-analytical correlations,the model can be adapted to different geometries and car lines effortlessly.The empirical parameters of the model result only from two standardized tests.These parameters are used to evaluate the estimation accuracy in real driving situations.Additionally,the adaptability is tested for two different car lines and four brake disc dimensions.These tests are initially performed with unchanged parameters and afterwards with refitted parameters.The model shows a good estimation for the tested load cases.Compared to the state of the art,the proposed model is less accurate than complex finite element method(FEM)models and computational fluid dynamic(CFD)approaches,but shows a higher accuracy and better adaptability than other lumped parameter models with comparable numerical effort.Hence,possible applications can be dimensioning the brake system in the development process of new car lines or a real-time simulation on the latest ECU in the vehicle.
文摘In this article, research was conducted to improve Linter machines that remove short fibers remaining in ginned cotton seeds at cotton ginneries. The study examined the effect of changing the dimensions of the brush drum, guide and mesh surface in the cleaning device proposed for the linting machine on the movement of the peg and the cleaning efficiency, and the highest level of efficiency in separating impurities from the peg was determined. During the study, the main factors influencing the effective operation of the improved linting machine were identified, the limits of their values were determined, and studies were carried out using the mathematical modeling method. As a result, at the values of the given coefficients, efficient operation of the improved linting machine was observed, that is, the lint cleaning efficiency reached 55.1%.
