检索结果-内蒙古大学图书馆

Synthesis of CaTiO3@Ag@ZnO and its photocatalytic degradation of levofloxacin

Ceramics International 2025年

作者： Huang, Liang Zhang, Leilei Chang, Zhiwen Chen, Cheng Fan, Fengyao Lu, Chunhai No.1 Dongsanlu Erxianqiao Sichuan Chengdu610059 China College of Nuclear Technology and Automation Engineering Chengdu University of Technology No.1 Dongsanlu Erxianqiao Sichuan Chengdu610059 China

This study presents a synthesis strategy for Z-scheme CaTiO3@Ag@ZnO (CTAZ) heterojunction photocatalysts and evaluates their performance in levofloxacin (LVF) degradation. Key operational parameters, including component ratios and solution pH, were systematically optimized. The optimized CTAZ-58 photocatalyst achieved over 90 % LVF removal efficiency within 100 min under visible light irradiation, demonstrating stable performance across four consecutive cycles. Comprehensive characterization via XRD, SEM, TEM, EPR, UV–Vis DRS, Mott-Schottky analysis, and EIS elucidated the material's charge transfer mechanism. This work advances the design of CaTiO3-based photocatalysts and provides a solar-driven solution for sustainable antibiotic remediation. © 2025

关键词： Charge transfer

来源：评论

学校读者我要写书评

暂无评论

Modeling of the cooperative operation process of aircraft ground centralized deicing 4

Modeling of the cooperative operation process of aircraft gr...

引用

2024 4th International Conference on Mechatronics technology and Aerospace engineering, ICMTAE 2024

作者： Xing, Zhiwei Wang, Yuning Yan, Chunyu Sun, Ke Zhang, Tao Zhao, Liang College of Electronic Information and Automation Civil Aviation University of China Tianjin300300 China College of Safety Science and Engineering Civil Aviation University of China Tianjin300300 China Engineering Technology Research Center Second Research Institute of Civil Aviation Administration of China Chengdu610041 China Sichuan Branch of China Eastern Airlines Technology Company Limited Chengdu610041 China

Aiming at the problems of insufficient refinement and low efficiency of aircraft ground centralized deicing operation under snow and ice weather, a description method of aircraft ground centralized deicing operation process based on a spatiotemporal correlation network was proposed. Based on the allocation and spatiotemporal distribution of airport deicing resources, a colored spatiotemporal network model was established to describe the dynamic process of aircraft ground deicing through the evolution of a colored spatiotemporal network. A resource optimization method was designed. The traditional genetic algorithm was improved by Sine chaotic mapping technology and differential evolution, and verified by experiment. The results show that the root-mean-square error of the idle time of each deicing site is 18.29 after optimization, which is 4.88 less than before optimization, suggesting that the utilization rate of the deicing site is increased by 21.06%. The improved NSGA-Ⅱ algorithm can increase the completion rate of deicing potential resource allocation by 7.62% and has a faster convergence rate compared with the traditional algorithm, which proves the feasibility of the proposed method. © 2025 Institute of Physics Publishing. All rights reserved.

关键词： Mean square error

来源：评论

学校读者我要写书评

暂无评论

A lightweight YOLOv5 garbage detection and classification method 2

A lightweight YOLOv5 garbage detection and classification me...

引用

2nd International Conference on Optics and Communication technology, ICOCT 2022

作者： Huang, Mei Chang, Yongxin Zhang, Liangbao Jiao, Shuaifeng School of Electrical Engineering and Information Southwest Petroleum University Chengdu China School of Automation University of Electronic Science and Technology Chengdu China

ISBN: (纸本)9781510660533

Aiming at the problems of unclear, difficult, and inefficient classification of traditional manual waste, and the difficulty of deploying large existing garbage classification network models, a lightweight garbage detection and classification network S-YOLOv5 is designed based on YOLOv5s. First, a garbage dataset containing 18 types of common household garbage is constructed and labeled according to the principles of garbage classification;secondly, a module combining shufflenetv2 and CoordAttention was introduced to replace the YOLOv5s backbone network, and the ReLU activation function in the shufflenet module was substituted by FReLU;finally the PANet structure was replaced by the BiFPN structure, so as to reduce the model complexity and achieve lightweight while maintaining a high mAP. The experimental results show that the size of S-YOLOv5 is only 2.6MB, which is about 1/6 of the original network size, and the mAP is 80.2%. The size of the proposed network is reduced while maintaining high accuracy, making it more suitable for deployment in smart devices. © 2022 SPIE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Random Fourier Features Based Kernel Recursive Least Logarithmic Hyperbolic Cosine Algorithm

Random Fourier Features Based Kernel Recursive Least Logarit...

引用

2023 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2023

作者： Tang, Shenjie Li, Xifeng Liu, Bo Yi, Guo Zheng, Xuepeng Bi, Dongjie Xie, Xuan Xie, Yongle University of Electronic Science and Technology of China School of Automation Engineering Chengdu611731 China Shanghai Spaceflight Precision Machinery Institute Shanghai201600 China

ISBN: (纸本)9798350327182

Via the help of the reproducing kernel Hilbert space (RKHS), kernel adaptive filtering as a remarkable and effective method for the online non-linear learning has drawn considerable attention. It is worth mentioning that the robustness of kernel adaptive filtering algorithm is highly dependent on the pertinent cost criteria. In addition, the logarithmic hyperbolic cosine criterion has been applied in kernel adaptive filtering successfully, since it can offer the exceptional robustness under non-Gaussian noises. However, for the optimization of existing logarithmic hyperbolic cosine based kernel adaptive filters, they almost adopt stochastic gradient descent method, which can not obtain a trade-off between convergence rate and filtering accuracy. Whereas the recursion can provide faster convergence for the optimization of adaptive filtering. Furthermore, the random Fourier feature (RFF) method can achieve effective sparsification for traditional kernel adaptive filtering. Therefore, a novel RFF based kernel recursive least logarithmic hyperbolic cosine (RFFKRLL) algorithm is developed in this paper. Experiments on Mackey-Glass time series prediction and non-linear channel equalization validate the advantages from aspects of accuracy, consumed time and robustness against non-Gaussian noises. © 2023 IEEE.

关键词： Adaptive filtering

来源：评论

学校读者我要写书评

暂无评论

Application of pole-zero cancellation circuit in nuclear signal filtering and shaping algorithm

引用

Nuclear Science and Techniques 2021年第8期32卷 85-94页

作者： Huai-Ping Wang Jian-Bin Zhou Xiao-Ping Ouyang Xian-Guo Tuo Xu Hong Yi Liu Jie Yu You-Xian Jin College of Nuclear Technology and Automation Engineering Chengdu University of TechnologyNo.1 East 3 RoadErxian BridgeChenghua DistrictChengdu 610059China Northwestern Institute of Nuclear Technology Xi’an 710024China Engineering Research Center of Nuclear Technology Application(East China University of Technology) Ministry of EducationNanchang 330013China

In radiation measurement and digital nuclear spectrum systems,traditional nuclear signal processing circuits in nuclear electronics have been gradually replaced by digital algorithm modules with the application of highperformance programmable hardware logic devices(such as FPGA or DSP).Referring to the digital realization method of inverse RC integral circuit systems,the function of the pole-zero cancellation(PZC)circuit was analyzed,a new modified cascade equivalent model of PZC was established,and the time-domain digital PZC(DPZC)recursive algorithm was derived in detail in this *** parameters kIand k_(D)are included in the new algorithm,where kIshould match the exponential decay time constant of the input signal to realize the pole-zero compensation,while the decay time constant of the output signal can be changed with the adjustable parameter k_(D)(which is larger than the decay time constant of the input signal).Based on the new DPZC algorithm module,two trapezoidal(triangular)shaping filters were designed and *** amplitude–frequency characteristics of the output signal of the proposed trapezoidal shaping algorithm and the convolution trapezoidal shaping algorithm were compared,with fixed peaking *** results show that the trapezoidal shaping algorithm based on DPZC can better suppress high-frequency ***,based on the Na I(Tl)scintillator(u75 mm×75 mm)detector and^(137)Cs source,the effect of the k_(D)value on the energy resolution of the DPZC trapezoidal(triangular)shaping algorithm was *** experimental results show that,with an increase in k_(D),the energy resolution of the system improved and reached the maximum when k_(D)was greater than 10,and the optimal energy resolution of the system was 7.72%.

关键词： Pole-zero cancellation C–R inverse system Trapezoidal/triangular shaping Amplitude–frequency characteristics Energy resolution

来源：评论

学校读者我要写书评

暂无评论

An Efficient Parallel Computing Cnn Accelerator Based on Fpga Hardware Resources

SSRN

引用

SSRN 2024年

作者： An, Zhonghao Peng, Kexin Yang, Kailing Lin, Cheng Zeng, LingPeng Yang, Wenbo School of Nuclear Technology and Automation Engineering Chengdu University of Technology Chengdu610059 China School of Computer Science and Cyber Security Chengdu University of Technology Chengdu610059 China School of Business Chengdu University of Technology Chengdu610059 China

Convolutional Neural Networks (CNNs) have achieved remarkable success in a wide range of fields, becoming one of the most widely used and powerful techniques in computer vision. Given that the model structure and convolutional operations of convolutional neural networks are significantly highly parallel, accelerating CNNs using field-programmable gate arrays (FPGAs) has become an important solution to this problem. In this paper, we propose an integrated hardware accelerator architecture based on YOLOv3-tiny (You Only Look Once version 3 tiny) with low latency and high computational efficiency. The accelerator is designed in a parallel way with input partially parallel, full parallelism between convolutional cores, partial parallelism within cores, and output partially parallelism, and utilises a dynamically configured matrix generator to complete the matrix transformation of the Generalised Matrix Multiplication (GEMM) algorithm. In order to optimise the accelerator further: firstly, we improve the original YOLOv3-tiny algorithm by adopting INT8 precision in order to increase the parallelism and reduce the memory space and bandwidth requirements. Secondly, we propose a way to map two INT8 multiplications at a time into the DSP (Digital Signal Processing), which improves the utilisation of the DSP by a factor of 2.25 for low-precision operations. As well as adding a new REQ layer to maintain INT8 while keeping the precision. Based on the experimental results, our proposed accelerator is able to achieve 42ms latency and 120 GOPS data throughput, which outperforms most of the research proposals in terms of comprehensive performance. The architecture is suitable for edge computing applications. © 2024, The Authors. All rights reserved.

关键词： Digital signal processing

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Multi-Step Ahead and Long-Term Probabilistic Predictions for Automotive Fuel Cell Performance Degradation

引用

IEEE Transactions on Transportation Electrification 2025年

作者： Li, Junhao Chen, Junxiong Wang, Renkang Cao, Jishen Gao, Yan Tang, Hao University of Electronic Science and Technology of China School of Automation Engineering Chengdu610023 China University of Electronic Science and Technology of China Hydrogen and Fuel Cell Institute Chengdu610023 China

Accurate prediction of performance degradation is crucial for the timely maintenance of fuel cell. The influence of uncertain factors on automotive fuel cell makes it challenging for traditional point estimation methods to achieve reliable longterm predictive results. To address this, we have designed a datadriven probabilistic prediction model called MambaMixer, which enables multi-step ahead and long-term probabilistic prediction of performance degradation for automotive fuel cells without a mechanism model. The model first combines a dimensional-segment-wise (DSW) strategy to protect historical data. Secondly, the mamba model and multi-layer perceptron (MLP) are used to design a two-stage transform (TST) layer. Then, the decomposition strategy is used to extract multi-scale historical information, and the long-term probability prediction module is designed based on the student's t-distribution. In the prediction experiment of relative power-loss rate (RPLR), our model demonstrates superior performance in multi-step ahead prediction compared with three advanced models. In the long-term probabilistic prediction experiments, the effectiveness of the proposed algorithm is validated through multiple sets of experiments. In general, our model helps to improve the prediction performance of automotive fuel cell performance degradation, providing a new solution for data-driven model design. © 2015 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

A High-Precision Parameter Identification Method Based on Improved DFT for Sub-Super Synchronous Oscillations

A High-Precision Parameter Identification Method Based on Im...

引用

IEEE/IAS Industrial and Commercial Power System Asia (I&CPS Asia)

作者： Feiyu Sun Dongsheng Cai Zhengyuan Zhang Qi Huang The College of Nuclear Technology and Automation Engineering Chengdu University of Technology Chengdu China School of Mechanical and Electrical Engineering University of Electronic Science and Technology of China Chengdu China School of Information Engineering Southwest University of Science and Technology Mianyang China

ISBN: (数字)9798350352290

ISBN: (纸本)9798350352306

Wind energy has been employed extensively as a renewable energy source with the aim of carbon neutrality. The power system also makes extensive use of power electronic properties and lead to the sub/super-synchronous oscillation (Sub/Sup-SO) issue, which damages equipment and jeopardizes the power system's ability to operate safely and steadily. To improve the frequency band estimation range to 100 Hz, this research use Phasor Measurement Unit (PMU) instantaneous value data to determine the Sub/Sup-SO associated characteristics. To reduce the influence of spectral leakage and the fence effect, as well as to improve the identification accuracy of the related parameters, this paper adopts the RV self-convolutional window based Discrete Fourier Transform (DFT) and the Hanning double-window based all-phase Fast Fourier Transform (apFFT) hybrid algorithm to identify the Sub/Sup-SO modal parameters. We take into account the identification accuracy in complex cases, such as non-nominal frequencies, with noise effects and multiple Sub/Sup-SO modes. The simulation results show that the proposed method has high discrimination bandwidth, high discrimination accuracy and strong flexibility.

关键词： Damping Accuracy Simulation Discrete Fourier transforms Noise Frequency estimation Phasor measurement units Mathematical models Power systems Oscillators

来源：评论

学校读者我要写书评

暂无评论

Comparative Evaluation of Several Types of Passive Damping Branches for Grid-Connected Inverters under Passivity

Comparative Evaluation of Several Types of Passive Damping B...

引用

International Conference on Power Electronics and Motion Control (IPEMC)

作者： Guangda Ma Chuan Xie Cheng Li Yang Han Weijun Wang Jianxiao Zou School of Automation Engineering University of Electronic Science and Technology of China Chengdu School of Mechanical and Electrical Engineering University of Electronic Science and Technology of China Chengdu

ISBN: (数字)9798350351330

ISBN: (纸本)9798350351347

The passive damping (PD) branches based on passive design can provide sufficient passivity compensation to the output admittance of the grid-connected inverter (GCI) at the cost of low damping loss. However, the type of passivity-based PD branch is only considered on the shunt RC branch in recent reports. This paper investigates the passivity compensation capabilities of several types of PD branches. Furthermore, design guidelines for the different types of PD branches are presented based on their passive properties and damping loss. Finally, the design example is given to verify the correctness of the proposed design guidelines.

关键词： Damping Costs Design methodology Asia Inverters Admittance Motion control

来源：评论

学校读者我要写书评

暂无评论

An Improved Convolutional Networks Model for Carbon Emission Prediction in Power Systems

An Improved Convolutional Networks Model for Carbon Emission...

引用

IEEE/IAS Industrial and Commercial Power System Asia (I&CPS Asia)

作者： Dongsheng Cai Caroline Acen Chiagoziem C. Ukwuoma Nan Zhang Wei Liu Zhenyuan Zhang Qi Huang College of Nuclear Technology and Automation Engineering Chengdu University of Technology Chengdu China School of Computing and Cyber Security Chengdu University of Technology Chengdu China School of Mechanical and Electrical Engineering University of Electronic Science and Technology of China Chengdu China School of Information Engineering Southwest University of Science and Technology Mianyang China

ISBN: (数字)9798350352290

ISBN: (纸本)9798350352306

In response to the imperative need for carbon reduction, tracking the carbon footprint of power systems has emerged as a critical concern. Traditional methods have struggled to accurately capture the nuances of carbon emissions, necessitating a fresh approach. To address this gap, the study introduces a pioneering deep learning-based model. We contend that the rapid advancement of deep learning offers a solution by harnessing neural network models, particularly the temporal convolutional networks (TCN) and gated recurrent units (GRU). These models are adept at handling the intricate patterns and dependencies inherent in carbon emission data, which exhibit both short-term fluctuations and long-term trends. Sequel to that, we propose a novel TCN-GRU-ATT neural network model designed explicitly for carbon footprint data. This innovative model combines TCN, GRU, and attention mechanisms to comprehensively capture both short-term variations and long-term dependencies within carbon emission patterns. Using power generation data from a gas-fired unit within the power grid as a foundation, the model calculates carbon emissions across various time intervals. Results demonstrate the model's exceptional performance, with a low Mean Absolute Percentage Error (MAPE) of 1.7123%, a Root Mean Square Error (RMSE) of 8.8436 kg/CO2, and a high Coefficient of Determination R 2 of 0.9802.

关键词： Training Neural networks Carbon dioxide Predictive models Logic gates Market research Data models Convolutional neural networks Forecasting Carbon footprint

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：