检索结果-内蒙古大学图书馆

Model Predictive Control for Electrochemical Impedance Spectroscopy Measurement of Fuel Cells Based on neural network optimization

引用

IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION 2019年第2期5卷 524-534页

作者： Zhang, Liyan Zhou, Ze Chen, Qihong Long, Rong Quan, Shuhai Wuhan Univ Technol Sch Automat Wuhan 430070 Hubei Peoples R China Huazhong Agr Univ Coll Sci Wuhan 430070 Hubei Peoples R China

Electrochemical impedance spectroscopy (EIS) is a key specification of fuel cells, which can reflect the healthy state. EIS could be measured by using the ripple modulation of dc/dc converter connected to fuel cells. This method has the advantage of no external excitation sources and low volume. In ripple modulation, the mixed signals composed of direct current and alternate current are difficult to track accurately, so a model predictive control (MPC) method with favorable stability and dynamic response is proposed. Considering that the traditional optimization algorithm takes a long time to calculate, a recurrent neural network (RNN) optimization is used to find a solution of the quadratic programming (QP) problem in order to reduce online computation time. Moreover, field programmable gate array (FPGA) is employed to implement the proposed MPC method. Experimental results demonstrate that the proposed method could measure the EIS of fuel cell effectively.

关键词： DC/DC converter electrochemical impedance spectroscopy (EIS) field programmable gate array (FPGA) model predictive control (MPC) neural network optimization

来源：评论

学校读者我要写书评

暂无评论

Meta-learning approach to neural network optimization

引用

neural networkS 2010年第4期23卷 568-582页

作者： Kordik, Pavel Koutnik, Jan Drchal, Jan Kovarik, Oleg Cepek, Miroslav Snorek, Miroslav Czech Tech Univ Dept Comp Sci & Engn FEE CR-16635 Prague Czech Republic

optimization of neural network topology, weights and neuron transfer functions for given data set and problem is not an easy task. In this article, we focus primarily on building optimal feed-forward neural network classifier for i.i.d, data sets. We apply meta-learning principles to the neural network structure and function optimization. We show that diversity promotion, ensembling, self-organization and induction are beneficial for the problem. We combine several different neuron types trained by various optimization algorithms to build a supervised feed-forward neural network called Group of Adaptive Models Evolution (GAME). The approach was tested on a large number of benchmark data sets. The experiments show that the combination of different optimization algorithms in the network is the best choice when the performance is averaged over several real-world problems. (C) 2010 Elsevier Ltd. All rights reserved.

关键词： neural network optimization Ant colony Evolutionary algorithms Meta-learning Meta-optimization optimization algorithm benchmark Diversity promotion Inductive modeling Self-organization

来源：评论

学校读者我要写书评

暂无评论

ECG Authentication Hardware Design With Low-Power Signal Processing and neural network optimization With Low Precision and Structured Compression

引用

IEEE TRANSACTIONS ON BIOMEDICAL CIRCUITS AND SYSTEMS 2020年第2期14卷 198-208页

作者： Cherupally, Sai Kiran Yin, Shihui Kadetotad, Deepak Srivastava, Gaurav Bae, Chisung Kim, Sang Joon Seo, Jae-sun Arizona State Univ Sch Elect Comp & Energy Engn Tempe AZ 85281 USA Samsung Adv Inst Technol Suwon 16678 South Korea

Biometrics such as facial features, fingerprint, and iris are being used increasingly in modern authentication systems. These methods are now popular and have found their way into many portable electronics such as smartphones, tablets, and laptops. Furthermore, the use of biometrics enables secure access to private medical data, now collected in wearable devices such as smartwatches. In this work, we present an accurate low-power device authentication system that employs electrocardiogram (ECG) signals as the biometric modality. The proposed ECG processor consists of front-end signal processing of ECG signals and back-end neural networks (NNs) for accurate authentication. The NNs are trained using a cost function that minimizes intra-individual distance over time and maximizes inter-individual distance. Efficient low-power hardware was implemented by using fixed coefficients for ECG signal pre-processing and by using joint optimization of low-precision and structured sparsity for the NNs. We implemented two instances of ECG authentication hardware with 4X and 8X structurally-compressed NNs in 65 nm LP CMOS, which consume low power of 62.37 mu W and 75.41 mu W for real-time ECG authentication with a low equal error rate of 1.36% and 1.21%, respectively, for a large 741-subject in-house ECG database. The hardware was evaluated at 10 kHz clock frequency and 1.2 V voltage supply.

关键词： DNN compression ECG authentication fixed-coefficient filtering neural network optimization structured sparsity

来源：评论

学校读者我要写书评

暂无评论

LATTICE neural-network MINIMIZATION - APPLICATION OF neural-network optimization FOR LOCATING THE GLOBAL-MINIMUM CONFORMATIONS OF PROTEINS

引用

JOURNAL OF MOLECULAR BIOLOGY 1993年第4期232卷 1157-1168页

作者： RABOW, AA SCHERAGA, HA Baker Laboratory of Chemistry Cornell University Ithaca NY 14853-1301 U.S.A.

A way of formulating the protein-folding problem in neural network optimization terms is presented in this paper. This is accomplished by representing the conformation of a protein as an array of the amino acid sequence versus position on a three-dimensional face-centered cubic lattice with an energy function defined in terms of the array variables. The method is called lattice neural network minimization (LNNM). Using the neural network minimization method, the energy function is minimized to locate the global minimum energy for the conformation of the protein. The energy function consisted of site exclusion and bond connectivity penalty terms and a pairwise contact energy potential. The contact energy potential used in the procedure is the united-residue potential of Miyazawa, Jernigan and Covell. The LNNM method found the global minimum for a seven residue peptide in all of the 15 runs carried out. The time for each run was ∼30 seconds on one processor of an IBM 3090 computer. For a nine-residue peptide, the global minimum was found in 7 out of 15 runs (47%) in ∼50 seconds per run. For this peptide, LNNM found the global minimum or the second lowest minimum in 10 of the runs. In the same total CPU times (∼750 seconds), a Monte Carlo simulated annealing method found the global minimum or the second lowest minimum in only two runs, demonstrating the superiority of LNNM over the standard Monte Carlo simulated annealing method for this nine-residue peptide. Starting from a uniform array for the protein crambin (46 residues) on the lattice, the energy of the crambin array was minimized and a compact low-energy structure was found in ∼25 minutes of CPU time. Its energy was much lower than that of the native protein, suggesting that there are inadequacies in the Miyazawa-Jernigan-Covell potential. The LNNM method was applied to the prediction of what was previously called nucleation but more properly called chain-folding initiation sites (CFIS) of a prot

关键词： PROTEIN FOLDING GLOBAL optimization neural network optimization LATTICE MODELS CHAIN-FOLDING INITIATION SITES

来源：评论

学校读者我要写书评

暂无评论

Experimental Verification of the neural network optimization Algorithm for Identifying Frequency-Dependent Constitutive Parameters of Viscoelastic Materials

引用

JOURNAL OF VIBRATION ENGINEERING & TECHNOLOGIES 2024年第2期12卷 2147-2173页

作者： Prusty, Jagesh Kumar Papazafeiropoulos, George Mohanty, Sukesh Chandra Natl Inst Technol Rourkela Dept Mech Engn Rourkela 769008 Odisha India Natl Tech Univ Athens Dept Struct Engn Athens 15780 Greece

PurposeThe constitutive parameters of viscoelastic materials, such as storage modulus and loss factor, usually have frequency-dependent characteristics. The combination of polymers with different reinforcement and fillers usually exhibits various mechanical characteristics, which makes the identification of the material properties of viscoelastic materials a challenging task. The present study proposes an inverse identification technique based on a neural network optimization (NNO) algorithm to characterize the frequency-dependent material properties of a viscoelastic *** this end, a symmetric three-layered sandwich plate is considered having face layers of isotropic elastic material and a core layer of viscoelastic material. The experimental free vibration tests are performed using the impact hammer method to determine resonant frequencies and modal loss factors for various eigenmodes. In addition, a numerical model of the sandwich plate is developed to determine vibrational responses utilizing the finite element method. The vibration-based material parameter identification technique is implemented based on the NNO algorithm. The identified material parameters are then compared with the experimental dynamic mechanical analysis (DMA) test results. Furthermore, a numerical parametric study is performed considering the optimized viscoelastic material properties to investigate the influence of various geometrical and structural factors on the free vibration response of the sandwich *** identified results are in excellent agreement with the experimental DMA test results affirming the robustness of the proposed inverse technique. The parametric study not only investigates the effect of various structural and geometric parameters on the dynamic response of the sandwich plate but also verifies that the calibrated properties are both realistic and physically *** proposed algorithm is a useful and efficient tool for inverse i

关键词： Viscoelastic material Impact hammer method Finite element method Dynamic mechanical analysis neural network optimization Inverse identification technique

来源：评论

学校读者我要写书评

暂无评论

Fast-Accurate Dual-Axis Solar Tracker Controlled by P&O Technique with neural network optimization

Fast-Accurate Dual-Axis Solar Tracker Controlled by P&O Tech...

引用

20th IEEE International Conference on Environment and Electrical Engineering (EEEIC) / 4th IEEE Industrial and Commercial Power Systems Europe Conference (I and CPS Europe)

作者： Faraji, Jalal Khanjanianpak, Mahdi Rezaei, M. Kia, Mohsen Aliyan, Ehsan Dehghanian, Payman Shahid Beheshti Univ Dept Elect Engn Tehran Iran Bu Ali Sina Univ Dept Elect Engn Hamadan Hamadan Iran Iran Univ Sci & Technol Dept Elect Engn Tehran Iran Islamic Azad Univ Dept Elect & Comp Engn Pardis Iran George Washington Univ Dept Elect & Comp Engn Washington DC USA

ISBN: (纸本)9781728174556

This paper proposes a fast-accurate method to locate the sun through dual-axis sun tracker rapidly and accurately. The tracking time is reduced to decrease the power consumption and improve the dynamic performance. The controlling procedure is closed-loop combination of Multi-Layer Perceptron (MLP) neural network and Perturbation & Observation (P&O). If the sun deviates from solar panels significantly, the sun tracker will operate quickly, but inaccurately by using the neural network. But if the sun gets away slightly, the sun tracker will operate accurately, but slowly by using P& O. So, by combining these two given cases will cause to track the sun fast and accurately. The error back propagation training algorithm is firstly used to train the neural network and in the next step the Modified Particular Swarm optimization (MPSO) method used to optimize the neural network. The results indicate that the tracking time of proposed method is 42%-49% less than other which is used for comparison. So, the sun tracker has better dynamic performance, lower power consumption and increased efficiency.

关键词： Perturbation & Observation (P&O) dual-axis solar tracker neural network optimization Modified Particular Swarm optimization (MPSO) method MPPT Solar PV

来源：评论

学校读者我要写书评

暂无评论

Control Strategy of Hybrid Power System for Fuel Cell Electric Vehicle based on neural network optimization

Control Strategy of Hybrid Power System for Fuel Cell Electr...

引用

IEEE International Conference on Automation and Logistics

作者： Xie Chang-jun Quan Shu-hai Chen Qi-hong Wuhan Univ Technol Coll Automat Wuhan Hubei Peoples R China

ISBN: (纸本)9781424425020

Fuel Cell Electric Vehicle has very good application foreground for it has some advantages, such as high efficiency and little emission, etc. In order to characterize the structure of hybrid power system for Fuel Cell Electric Vehicle, the project of parallel hybrid power system was brought forward, which consisted of Fuel Cell, NIH battery and DC/DC converter. The power flow of hybrid power system was analyzed, and besides, the issue for energy management of power system was converted to combinatorial optimization problem. The energy management strategy based on neural network optimization was designed by applying three-layer neural network optimization control structure and optimizing neural network connect weight via genetic algorithms. In the end, the simulation results were presented, which show the effectiveness of the control strategy, for it can improve economy performance of vehicles in comparison with fuzzy control strategy.

关键词： Fuel Cell Electric Vehicle hybrid power system neural network optimization Genetic algorithms energy management strategy

来源：评论

学校读者我要写书评

暂无评论

neural network optimization Algorithm Model Combining L1/2 Regularization and Extreme Learning Machine

Neural Network Optimization Algorithm Model Combining L1/2 R...

引用

2018 3rd International Workshop on Materials Engineering and Computer Sciences(IWMECS 2018)

作者： Anzhi Qi Liaoning Jianzhu Vocational College

Extreme Learning Machine(ELM) is a fast learning algorithm that uses random mechanism to reduce parameter setting and selection,thereby greatly improving learning speed and ensuring generalization *** traditional learning methods,ELM The variables are no longer iterative but randomly generated,so that a nonlinear system that expresses a forward neural network can be reduced to a linear system that only needs to compute the output weights,and the least-squares method can be used to solve the linear system *** the ELM algorithm However,the effectiveness of this algorithm in the application of large-scale data needs to be further *** thesis is based on L1/2 regularization theory and full rank Cholesky matrix decomposition respectively Two Improved ELM Algorithms Applied to Large Scale Data.

关键词： ELM L1/2 Regulation neural network optimization

来源：评论

学校读者我要写书评

暂无评论

Ellipse Fitting Based on neural network optimization 41

Ellipse Fitting Based on Neural Network Optimization

引用

第41届中国控制会议

作者： Chenrong Long Peng Liu Qinglei Hu School of Automation Science and Electrical Engineering Beihang University National Key Laboratory of Science and Technology on Test Physics & Numerical Mathematics Beijing Institute of Space Long March Vehicle School of Automation Science and Electrical Engineering Beihang UniversityBeijing 100191China.

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

Ellipse fitting is to estimate the parameters of the ellipse from a set of coplanar points,but the geometric distance from the plane point to the elliptic curve is difficult to calculate *** paper proposes a neural network optimization-based algorithm to estimate the geometric distances from plane points to elliptic curves to solve the ellipse fitting ***,the neural network is firstly applied to train the geometric distance between the plane point and the standard ellipse,and then obtain the geometric distance between the plane point and the general ellipse through the rotation and translation ***,use minimizing the sum of the Euclidean distances from the plane point to the elliptic curve as the objective function,and finally use the gradient descent method to iteratively solve the ellipse ***,the proposed method avoids the direct algebraic solution of the geometric distance from the plane point to the elliptic curve and the performance of the algorithm to gether with its robustness to noise are analysed via numerical simulation.

关键词： Ellipse Fitting Coplanar Points Geometric Distance neural network optimization

来源：评论

学校读者我要写书评

暂无评论

Optimal Power Control of Smart Grid Based on neural network optimization

Optimal Power Control of Smart Grid Based on Neural Network ...

引用

International Conference on Advanced Materials and Computer Science

作者： Zhang, Liyan Quan, Shuhai Wuhan Univ Technol Sch Automat Wuhan 430070 Hubei Province Peoples R China

ISBN: (纸本)9783037850978

This paper presents a neural network based optimal control approach for the power allocation problem of smart grid. Firstly the optimal control framework which consists of center control layer, area control layer and generator control layer is presented. In order to reduce computation time of optimal control, the neural network optimization method and schematic diagram is proposed. Simulation is implemented in a modified IEEE30-bus power system comprising 6 conventional generators and a wind plant. Simulation results demonstrate that the proposed neural network optimization approach is satisfied the demand of computation time and the optimal power controller can allocate the power into loads reasonably and cope with the fluctuation of the load and renewable energy source's output power.

关键词： neural network optimization optimal power control smart grid renewable energy sources

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：