检索结果-南通市图书馆

Deep learning-based battery state of charge estimation:Enhancing estimation performance with unlabelled training samples

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Energy Chemistry 2023年第5期80卷 48-57,I0002页

作者： Liang Ma Tieling Zhang School of Mechanical MaterialsMechatronic and Biomedical EngineeringUniversity of WollongongNSW 2522Australia

The estimation of state of charge(SOC)using deep neural networks(DNN)generally requires a considerable number of labelled samples for training,which refer to the current and voltage pieces with knowing their corresponding SOCs.However,the collection of labelled samples is costly and time-consuming.In contrast,the unlabelled training samples,which consist of the current and voltage data with unknown SOCs,are easy to obtain.In view of this,this paper proposes an improved DNN for SOC estimation by effectively using both a pool of unlabelled samples and a limited number of labelled samples.Besides the traditional supervised network,the proposed method uses an input reconstruction network to reformulate the time dependency features of the voltage and current.In this way,the developed network can extract useful information from the unlabelled samples.The proposed method is validated under different drive cycles and temperature conditions.The results reveal that the SOC estimation accuracy of the DNN trained with both labelled and unlabelled samples outperforms that of only using a limited number of labelled samples.In addition,when the dataset with reduced number of labelled samples to some extent is used to test the developed network,it is found that the proposed method performs well and is robust in producing the model outputs with the required accuracy when the unlabelled samples are involved in the model training.Furthermore,the proposed method is evaluated with different recurrent neural networks(RNNs)applied to the input reconstruction module.The results indicate that the proposed method is feasible for various RNN algorithms,and it could be flexibly applied to other conditions as required.

关键词： Deep learning State of charge estimation data-driven methods Battery management system Recurrent neural networks

A data-driven Distributionally Robust Operational Model for Urban Integrated Energy Systems

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

CSEE Journal of Power and Energy Systems 2022年第3期8卷 789-800页

作者： Hongjun Gao Zhenyu Liu Youbo Liu Lingfeng Wang Junyong Liu College of Electrical Engineering Sichuan UniversityChengdu 610065China Department of Electrical Engineering and Computer Science University of Wisconsin-MilwaukeeMilwaukeeWI 53211USA

A multi-energy conversion can effectively increase the utilization of renewable energy in the urban integrated energy system(UIES).Meanwhile,the uncertainties of renewable energy resources(e.g.,wind energy)also bring increased challenges to the operation of UIES.In this study,a typical two-stage datadriven distributionally robust operation(DDRO)model based on finite scenarios is proposed for UIES including power,gas and heat networks to obtain a salient strategy from both an economic and robustness perspective.In the first stage,the forecasted information for wind power is especially included to improve the economic aspect of robust decisions.The worst probability distribution for the selected known real-time wind power scenarios can be identified in the second stage where the power differences caused by the real-time uncertainties of wind power can be mitigated by flexible regulation of energy purchasing and coupling units(such as gas turbine,power to gas equipment,electric boiler and gas boiler).Moreover,norm-1 and norm-inf co-constraints are utilized to construct a confidence set for the probability distributions of uncertain wind power.The whole two-stage model is solved by the column-and-constraint generation(CCG)algorithm.Finally,case studies are conducted to show the performance of the proposed model and various approaches.Index Terms-data-driven methods,distributionally robust optimization,urban integrated energy system,wind power.

关键词： data-driven methods distributionally robust optimization urban integrated energy system wind power

Kernel-Based State-Space Kriging for Predictive Control

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2023年第5期10卷 1263-1275页

作者： A.Daniel Carnerero Daniel R.Ramirez Daniel Limon Teodoro Alamo the Department of System Engineering and Automation University of SevilleSevilla 41020Spain the School of Engineering Tokyo Institute of TechnologyTokyo 152-8552Japan Department of System Engineering and Automation University of SevilleSevilla 41020Spain

In this paper, we extend the state-space kriging(SSK) modeling technique presented in a previous work by the authors in order to consider non-autonomous systems. SSK is a data-driven method that computes predictions as linear combinations of past outputs. To model the nonlinear dynamics of the system, we propose the kernel-based state-space kriging(K-SSK), a new version of the SSK where kernel functions are used instead of resorting to considerations about the locality of the data. Also, a Kalman filter can be used to improve the predictions at each time step in the case of noisy measurements. A constrained tracking nonlinear model predictive control(NMPC) scheme using the black-box input-output model obtained by means of the K-SSK prediction method is proposed. Finally, a simulation example and a real experiment are provided in order to assess the performance of the proposed controller.

关键词： data-driven methods model identification Kernel methods predictive control

Lean and interpretable digital twins for building energy monitoring - A case study with smart thermostatic radiator valves and gas absorption heat pumps

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Energy and AI

Energy and AI 2023年第4期14卷 645-659页

作者： Massimiliano Manfren Patrick AB James Victoria Aragon Lamberto Tronchin Faculty of Engineering and Physical Sciences University of SouthamptonBoldrewood Campus–SO167QF SouthamptonUnited Kingdom Department of Architecture(DA) University of BolognaVia Cavalcavia 6147521 CesenaItaly

The transition to low carbon energy systems poses challenges in terms of energy efficiency.In building refur-bishment projects,efficient technologies such as smart controls and heat pumps are increasingly being used as a substitute for conventional technologies with the aim of reducing carbon emissions and determining operational energy and cost savings,together with other benefits.Measured building performance,however,often reveals a significant gap between the predicted energy use(design stage)and actual energy use(operation stage).For this reason,lean and interpretable digital twins are needed for building energy monitoring aimed at persistence of savings and continuous performance improvement.In this research,interpretable regression models are built with data at multiple temporal resolutions(monthly,daily and hourly)and seamlessly integrated with the goal of verifying the performance improvements due to Smart thermostatic radiator valves(TRVs)and gas absorption heat pumps(GAHPs)as well as giving insights on the performance of the building as a whole.Further,as part of modelling research,time of week and temperature(TOWT)approach is reformulated and benchmarked against its original implementation.The case study chosen is Hale Court sheltered housing,located in the city of Portsmouth(UK).This building has been used for the field-testing of innovative technologies such as TRVs and GAHPs within the EU Horizon 2020 project THERMOSS.The results obtained are used to illustrate possible extensions of the use of energy signature modelling,highlighting implications for energy management and innovative building technologies development.

关键词： data-driven methods Digital twins Energy signature Thermostatic radiator valves Gas absorption heat pumps Energy management Energy Analytics

Concrete corrosion in wastewater systems:Prediction and sensitivity analysis using advanced extreme learning machine

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Structural and Civil Engineering 2021年第2期15卷 444-460页

作者： Mohammad ZOUNEMAT-KERMANI Meysam ALIZAMIR Zaher Mundher YASEEN Reinhard HINKELMANN Department of Water Engineering Shahid Bahonar University of KermanKerman 76169-13439Iran Department of Civil Engineering Hamedan BranchIslamic Azad UniversityHamedan 65181-15743Iran Institute of Research and Development Duy Tan UniversityDa Nang 550000Vietnam Department of Civil Engineering Technische Universitat BerlinBerlin 13355Germany

The implementation of novel machine learning models can contribute remarkably to simulating the degradation of concrete due to environmental factors.This study considers the sulfuric acid corrosive factor in wastewater systems to simulate concrete mass loss using five machine learning models.The models include three different types of extreme learning machines,including the standard,online sequential,and kernel extreme learning machines,in addition to the artificial neural network,classification and regression tree model,and statistical multiple linear regression model.The reported values of concrete mass loss for six different types of concrete are the target values of the machine learning models.The input variability was assessed based on two scenarios prior to the application of the predictive models.For the first assessment,the machine learning models were developed using all the available cement and concrete mixture input variables;the second assessment was conducted based on the gamma test approach,which is a sensitivity analysis technique.Subsequently,the sensitivity analysis of the most effective parameters for concrete corrosion was tested using three different approaches.The adopted methodology attained optimistic and reliable modeling results.The online sequential extreme learning machine model demonstrated superior performance over the other investigated models in predicting the concrete mass loss of different types of concrete.

关键词： sewer systems environmental engineering data-driven methods sensitivity analysis

An Adaptive ANOVA-Based data-driven Stochastic Method for Elliptic PDEs with Random Coefficient

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Communications in Computational Physics 2014年第8期16卷 571-598页

作者： Zhiwen Zhang Xin Hu Thomas Y.Hou Guang Lin Mike Yan Computing and Mathematical Sciences California Institute of TechnologyPasadenaCA 91125USA. Pacific Northwest National Laboratory RichlandWA 99352USA.

In this paper,we present an adaptive,analysis of variance(ANOVA)-based data-driven stochastic method(ANOVA-DSM)to study the stochastic partial differential equations(SPDEs)in the multi-query setting.Our new method integrates the advantages of both the adaptive ANOVA decomposition technique and the data-driven stochastic method.To handle high-dimensional stochastic problems,we investigate the use of adaptive ANOVA decomposition in the stochastic space as an effective dimension-reduction technique.To improve the slow convergence of the generalized polynomial chaos(gPC)method or stochastic collocation(SC)method,we adopt the data-driven stochastic method(DSM)for speed up.An essential ingredient of the DSM is to construct a set of stochastic basis under which the stochastic solutions enjoy a compact representation for a broad range of forcing functions and/or boundary conditions.Our ANOVA-DSM consists of offline and online stages.In the offline stage,the original high-dimensional stochastic problem is decomposed into a series of lowdimensional stochastic subproblems,according to the ANOVA decomposition technique.Then,for each subproblem,a data-driven stochastic basis is computed using the Karhunen-Lo`eve expansion(KLE)and a two-level preconditioning optimization approach.Multiple trial functions are used to enrich the stochastic basis and improve the accuracy.In the online stage,we solve each stochastic subproblem for any given forcing function by projecting the stochastic solution into the data-driven stochastic basis constructed offline.In our ANOVA-DSM framework,solving the original highdimensional stochastic problem is reduced to solving a series of ANOVA-decomposed stochastic subproblems using the DSM.An adaptive ANOVA strategy is also provided to further reduce the number of the stochastic subproblems and speed up our method.To demonstrate the accuracy and efficiency of our method,numerical examples are presented for one-and two-dimensional elliptic PDEs with random coefficients.

关键词： Analysis of variance stochastic partial differential equations data-driven methods Karhunen-Loeve expansion uncertainty quantification model reduction.

data-driven Forecasting of Subsurface Inclusion Defects for Continuous Casting Slabs

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Data-driven Forecasting of Subsurface Inclusion Defects for ...