检索结果-内蒙古大学图书馆

An improved transformer model with multi-head attention and attention to attention for low-carbon multi-depot vehicle routing problem

引用

ANNALS OF OPERATIONS RESEARCH 2024年第1-2期339卷 517-536页

作者： Zou, Yang Wu, Hecheng Yin, Yunqiang Dhamotharan, Lalitha Chen, Daqiang Tiwari, Aviral Kumar Nanjing Univ Aeronaut & Astronaut Coll Econ & Management Nanjing 210016 Peoples R China Univ Elect Sci & Technol China Sch Econ & Management Chengdu 611731 Peoples R China Univ Exeter Business Sch Exeter EX4 4PU Devon England Zhejiang Gongshang Univ Sch Management & E Business Hangzhou 310018 Peoples R China Rajagiri Business Sch RBS Kochi 682039 Kerala India

Low-carbon logistics is an emerging and sustainable development industry in the era of a low-carbon economy. The end-to-end deep reinforcement learning (DRL) method with an encoder-decoder framework has been proven effective for solving logistics problems. However, in most cases, the recurrent neural networks (RNN) and attention mechanisms are used in encoders and decoders, which may result in the long-distance dependence problem and the neglect of the correlation between query vectors. To surround this problem, we propose an improved transformer model (TAOA) with both multi-head attention mechanism (MHA) and attention to attention mechanism (AOA), and apply it to solve the low-carbon multi-depot vehicle routing problem (MDVRP). In this model, the MHA and AOA are implemented to solve the probability of route nodes in the encoder and decoder. The MHA is used to process different parts of the input sequence, which can be calculated in parallel, and the AOA is used to deal with the deficiency problem of correlation between query results and query vectors in the MHA. The actor-critic framework based on strategy gradient is constructed to train model parameters. The 2opt operator is further used to optimize the resulting routes. Finally, extensive numerical studies are carried out to verify the effectiveness and operation efficiency of the proposed TAOA, and the results show that the proposed TAOA performs better in solving the MDVRP than the traditional transformer model (Kools), genetic algorithm (ga), and Google OR-Tools (Ortools).

关键词： End-to-end deep reinforcement learning Transformer model Multi-head attention mechanism Low-carbon multi-depot vehicle routing problem ga algorithm

来源：评论

学校读者我要写书评

暂无评论

Optimizing the wells location using the Metaheuristic algorithms, as well as optimizing the drilling time of production and injection wells in one of the reservoirs in south west of Iran

引用

EARTH SCIENCE INFORMATICS 2024年第2期17卷 1393-1410页

作者： Nikravesh, Hamed Ranjbar, Ali Azin, Reza Persian Gulf Univ Fac Petr Gas & Petrochem Engn Petr Engn Dept Bushehr Iran

In this study, the optimization of the location of oil production and injection wells, the number of production and injection wells, and the economic evaluation in the water injection process has been carried out using three algorithms: PSO, ga, and HgaPSO. Optimizing the oil extraction process by optimizing well locations, the optimal number of production and injection wells, and economic evaluation can improve the performance and efficiency of the process. In this study, the optimization of the water injection process into oil reservoirs using the PSO, ga, and HgaPSO algorithms has been addressed. Additionally, economic evaluation, considering costs and profits from oil production using the best-optimized parameters, has been conducted, and the best Net Present Value (NPV) has been calculated. Initially, the optimal well location problem is modeled, taking into account reservoir characteristics and geological conditions. Using the PSO and ga algorithms, the best well location and number of water injection wells have been obtained, leading to improved efficiency and cost reduction. Then, the ga algorithm is used to combine and modify viewpoints and optimize the number and location of wells. This algorithm explores optimal solutions in the well location space. Economic evaluation is performed using the best-optimized parameters by the algorithms. Costs of water injection, oil production costs, and profits from increased oil production are considered. By calculating the best Net Present Value (NPV), optimal decisions regarding the water injection process and enhancing the performance of oil reservoirs are made. Simulation results indicate that the ga algorithm outperforms other algorithms and can be used as a robust and effective method for optimizing well locations and the number of production and injection wells in the oil injection process. Ultimately, this study assists the oil industry in making better decisions about optimizing water injection and improving the

关键词： PSO algorithm ga algorithm Hga PSO algorithm Well location optimization

来源：评论

学校读者我要写书评

暂无评论

Research on Improved Obstacle Avoidance algorithm

引用

RECENT ADVANCES IN ELECTRICAL & ELECTRONIC ENGINEERING 2014年第2期7卷 81-89页

作者： Zhang, Liguang Chen, Chao Xian Technol Univ Sch Elect Informat Engn Xian 710032 Peoples R China

The traditional artificial potential field algorithm has the shortcomings of appearing trap area easily and shaking in front of obstacles. This paper, studying traditional artificial potential field deeply, optimizes the sensor data with the introduction of the BP network and the ga algorithm, unmanned vehicles achieving autonomous obstacle avoidance and prognoses in advance when passing the path with obstacles, and effectively eliminating system instability caused by shock and dead zone. Comparing with traditional artificial potential field algorithm, the improved algorithm makes a further improvement in the respects of system stability and efficiency.

关键词： BP network ga algorithm multi sensor potential field method unmanned vehicle

来源：评论

学校读者我要写书评

暂无评论

TRI-SU-L ADWT-FCM: TRI-SU-L-Based Change Detection in SAR Images with ADWT and Fuzzy C-Means Clustering

引用

JOURNAL OF THE INDIAN SOCIETY OF REMOTE SENSING 2022年第9期50卷 1667-1687页

作者： Kumar, Jakka Thrisul Yennapusa, Mallikarjuna Reddy Rao, Bhima Prabhakara VNITSW Guntur Andhra Pradesh India VVIT Guntur Andhra Pradesh India JNTUK Dept ECE Kakinada Andhra Pradesh India

The Detection of Changes in Synthetic Aperture Radar (SAR) images is indispensable in the field of remote sensing. This paper presents an eccentric method to improve the overall performance of change detection in SAR images by increasing the smoothness and to achieve symmetry between changed pixels and unchanged pixels of the difference image in terms of accuracy. In this regard, three data sets were captured by SAR sensor using RADAASAT-1, RADAASAT-2 and ERS-2 satellites where each data set consisting two multitemporal SAR images, Median-subtraction operator is proposing along with the existing Log-Ratio and Mean-Ratio operators. Meanwhile, the proposing method is named as the Tri Subtraction Level (TRI-SU-L). Difference images from two ratio operators and median subtraction image are merged based on a novel image fusion technique (DWT with optimized filter coefficients), which is named as Adaptive Discrete Wavelet Transform (ADWT). A novel optimization technique which is named as Transformed Dragonfly algorithm algorithm has been implemented to choose Optimized filter coefficients along with existing optimization techniques. FCM-Clustering has been employed to segment changed area as foreground and unchanged area as background. The segmented image is compared with the ground truth image and performance is computed in terms of accuracy. To crosscheck the validity of the proposed method, optimized filter coefficients have been chosen through genetic algorithm , particle swarm optimization algorithm and conventional dragonfly algorithm. Finally, the proposed method improved the accuracy over existing methods of (0.84%, 0.81%, 1.17%, 1.13%, 1.17%), (6.26%, 3.37%, 11.31%, 9.67%, 5.34%), ( 2.45%, 2.86%, 2.32%, 1.56%, 1.88%) for first data set, second data set and third data set, respectively.

关键词： ADWT ga algorithm FCM PSO algorithm SAR TDA

来源：评论

学校读者我要写书评

暂无评论

Reinforced Genetic algorithm using Clustering based on Statistical Estimation

引用

IFAC-PapersOnLine 2018年第18期51卷 287-291页

作者： Park, Taekyoon Kim, Yeonsoo Lee, Jong Min School of Chemical and Biological Engineering Engineering Development Research Center Seoul National University 1 Gwanak-ro Gwanak-gu Seoul08826 Korea Republic of

Genetic algorithm(ga) has beend widely used to obtain solution in various optimization problems because of its robustness and convergence property. ga algorithm, however, has limitations; the computational time increases sharply as the complexity of the problem increases and the user's arbitrary judgement is involved especially in its termination step. In order to solve these limitations, we suggest a modified reinforced ga using clustering based on the statistical estimation. In this method, the mathematical reliability is gradually increased for each generation to find the solution vector to be finally obtained. The similarity between each solution vector generated by ga is determined and the inefficient repetitive calculation is remarkably reduced. In addition, the statistical reliability of the obtained solution vectors can be calculated to reduce the randomness of the user in the conventional termination step. © 2018

关键词： Clustering algorithms Estimation Genetic algorithms Optimization Parameter estimation Reinforcement Computational time Convergence properties ga algorithm Optimization problems similarity determination Solution vectors Statistical estimation Statistical reliability

来源：评论

学校读者我要写书评

暂无评论

Salinity Prediction Model in Estuary Area Based on Neural Network 7th

Salinity Prediction Model in Estuary Area Based on Neural Ne...

引用

7th International Conference on Sustainable Development of Water and Environment

作者： Gu, Chen He, Kun Ke, Keteng Yu, Wenbo Shanghai Invest Design & Res Inst Shanghai 200335 Peoples R China Zhejiang Inst Hydraul & Estuary Zhejiang Inst Marine Planning & Design Hangzhou 310020 Zhejiang Peoples R China

ISBN: (纸本)9783031761508;9783031761485;9783031761478

The salinity problem in the estuary area is related to industrial, agricultural and domestic water use, and is an important indicator of the water environment in the estuary area. In order to improve the prediction accuracy of salinity in the estuary area, based on the collection of a large amount of actual measured data and the neural network method, taking the daily maximum chlorine concentration at Qibao Station in the Qiantang River Estuary as an example, the estuary area salinity prediction model was established and the algorithm was optimized. The results show that it is feasible to predict the salinity in the estuary area based on the neural network model. As far as the prediction effect is concerned, the segmented multi-model BP neural network is the best, followed by the Narx neural network. The ga-BP neural network is slightly improved compared to the BP neural network, but the improvement effect is not as good as the segmented multi-model BP neural network.

关键词： Estuary area Salinity BP neural network ga algorithm Narx neural network

来源：评论

学校读者我要写书评

暂无评论

Design and Simulation for the Single Leg of the Six Degrees of Freedom Vibration Isolator

Design and Simulation for the Single Leg of the Six Degrees ...

引用

Asia Simulation Conference/International Conference on System Simulation and Scientific Computing (AsiaSim and ICSC 2012)

作者： Zhang, Yongliang Guo, Baiwei Meng, Xiuyun Beijing Inst Technol Key Lab Dynam & Control Flight Vehicle Minist Educ Beijing 100081 Peoples R China

ISBN: (纸本)9783642343803

With the development of space technologies, the optical payload requirement on the pointing accuracy becomes higher and higher. The active vibration isolation is an effective way to improve the pointing accuracy of the optical load, and the six freedoms Stewart Platform is widely adopted in the active vibration isolation for spacecrafts. The double-layer structure is adopted for the single leg of the Stewart Platform and the parameters of the passive part are optimized by the ga algorithm. For the active part, the L1 method is used for solving the problems in the vibration isolator because of the impact load. The simulation results indicate the effectiveness of the L1 method for solving the problem.

关键词： ga algorithm parameters optimization L1 method double-layer structure active vibration isolation

来源：评论

学校读者我要写书评

暂无评论

Research of Fractal Artistic Graphics Generation Method

Research of Fractal Artistic Graphics Generation Method

引用

International Conference on Intelligent Transportation, Big Data and Smart City (ICITBS)

作者： Li, Ma Ji, Ma Guilin Univ Aerosp Technol Guangxi Guilin 541004 Peoples R China

ISBN: (纸本)9781509004645

Graphic color which is obtained from traditional escape time algorithm is very single and it cannot effectively reflect structural characteristics of M set and J set so image cannot be effectively applied in generalized scope. This paper introduces genetic algorithm and integrates user preferences into genetic process by adding human-computer interaction in order to reflect designers' design ideas. Meanwhile, it solves difficult expression of fitness function in conceptual design and adopts adaptive method to guarantee optimal property in genetic population. Then, above methods are applied in genetic algorithm-based fractal pattern drawing system and the effectiveness of above methods is verified through actual cases.

关键词： Fractal ga algorithm M-J set Escape time tree

来源：评论

学校读者我要写书评

暂无评论

A Genetic algorithm Optimized MPPT Controller for a PV System with DC-DC Boost Converter

A Genetic Algorithm Optimized MPPT Controller for a PV Syste...

引用

International Conference on Engineering and MIS (ICEMIS)

作者： Badis, Afef Mansouri, Mohamed Nejib Boujmil, Mohamed Habib Univ Monastir Fac Sci Elect & Microelect Lab E E Monastir Tunisia Univ Monastir Ind Syst Study & Renewable Energy Natl Engn Sch Monastir ENIM Monastir Tunisia

ISBN: (纸本)9781509067787

Because of the non-linearity of the P-V characteristic of the Photovoltaic (PV) panel, a power electronic interface is mandatory to reach the desired fixed voltage level. This paper deals with the design of a PID controller in order to control a DC-DC boost converter in the PV system. For the design of the whole system, the used approach is based on the Energetic Macroscopic Representation (EMR). This technique goes from a complex command to a simpler one and is characterized by the use of cascading of one or more controllers. The three parameters of the controller are tuned by using Genetic algorithm (ga). Both of the system responses using PID MPPT controller tuned by genetic algorithm and PID controller tuned by Pole Placement (PP) are tested when fast changing the solar irradiation is experienced. Finally, in order to obtain the best system performance, Perturb and Observe (P&O) MPPT method is applied to generate the output voltage reference Vref. The simulation results are compared in terms of mean squared error (MSE) criterion, percentage overshoot and rising time. The results show that parameters tuned by ga PID controller gives better performance.

关键词： ga PID controller MPPT Photovoltaic system Energetic Macroscopic Representation Boost converter ga algorithm cascaded control

来源：评论

学校读者我要写书评

暂无评论

Parameter Optimization of Neural Network Model Based on Simulated Moving Bed

Parameter Optimization of Neural Network Model Based on Simu...

引用

Chinese Automation Congress (CAC)

作者： Li, Ling Dai, Chang Deng, Jia-Bi Yuan, De-Cheng Shenyang Inst Chem Technol Coll Informat Engn Shenyang Peoples R China

ISBN: (纸本)9781728140940

In recent years Simulated Moving Bed (SMB) chromatography separation technology has been used in petroleum chemicals, food processing and fine chemicals. In this paper, Bayesian regularization BP neural network algorithm is used to fit the input and output data of mechanism model of Simulated Moving Bed. Genetic algorithm and particle swarm algorithm are used to optimize the operating parameters and to find out the best concentration. The results are put into the mechanistic model to verify the point. The simulation results show that the error is small and these methods can solve the problem that the parameter optimization time is too long.

关键词： component formatting style styling insert (key words) Simulated Moving Bed Bayesian regularization BP neural network ga algorithm PSO algorithm

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：