检索结果-内蒙古大学图书馆

Online learning for distributed optimal control of an electric vehicle fleet

ELECTRIC POWER SYSTEMS RESEARCH 2022年 212卷

作者： Latimier, R. Le Goff Cherot, G. Ben Ahmed, H. ENS Rennes CNRS SATIE Bruz France

The management of electrical power systems requires the resolution of large-scale problems whose agents are linked by coupling constraints. Nevertheless, decomposition methods cannot provide an exact solution while dealing with temporal dynamics in a stochastic environment. Indeed, each agent would have to solve a local minimisation in which future quantities intervene. However, these quantities depend on other agents' future decisions which are still unknown. In order to enhance the existing approximate approaches to this challenge, the proposed method involves Alternating Direction Method of Multipliers to overcome the large dimension by an iterative resolution of local coordinated minimisations. Uncertain temporal dynamics are handled by a stochastic dynamic programming approach. In order to make local problems solvable, the online learning of a Markov process is added. The agents can then anticipate future global variations in a local probabilistic way. The optimal charging of an electric vehicle fleet paired with a wind power plant is considered as a case study. The expected benefits are highlighted, both at the outset and after training the anticipative models. The discussion addresses the learning parameters allowing the fastest convergence.

关键词： Alternating direction method of multipliers Decomposition method stochastic dynamic programming Virtual power plant Online learning Markov process

来源：评论

学校读者我要写书评

暂无评论

Comparison of Two Convergence Criterion in the Optimization Process Using a Recursive Method in a Multi-Reservoir System

引用

WATER 2022年第19期14卷 2952-2952页

作者： Mendoza-Ramirez, Rosalva Silva, Rodolfo Dominguez-Mora, Ramon Juan-Diego, Eduardo Carrizosa-Elizondo, Eliseo Univ Nacl Autonoma Mexico Inst Ingn Mexico City 04510 DF Mexico

stochastic dynamic programming (SDP) is an optimization technique used in the operation of reservoirs for many years. However, being an iterative method requiring considerable computational time, it is important to establish adequate convergence criterion for its most effective use. Based on two previous studies for the optimization of operations in one of the most important multi-reservoir systems in Mexico, this work uses SDP, centred on the interest in the convergence criterion used in the optimization process. In the first trial, following the recommendations in the literature consulted, the difference in the absolute value of two consecutive iterations was taken and compared against a set tolerance value and a discount factor. In the second trial, it was decided to take the squared difference of the two consecutive iterations. In each of the trials, the computational time taken to obtain the optimal operating policy was quantified, along with whether the optimal operating policy was obtained by meeting the convergence criterion or by reaching the maximum number of iterations. With each optimization policy, the operation of the system under study was simulated and four variables were taken as evaluators of the system behaviour. The results showed few differences in the two operation policies but notable differences in the computation time used in the optimization process, as well as in the fulfilment of the convergence criterion.

关键词： convergence criterion stochastic dynamic programming multi-reservoir system cascade operation discount factor

来源：评论

学校读者我要写书评

暂无评论

dynamic ordering decisions with approximate learning of supply yield uncertainty

引用

INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS 2022年 243卷

作者： Gel, Esma S. Salman, F. Sibel Arizona State Univ Sch Comp Informat & Decis Syst Engn Tempe AZ 85281 USA Koc Univ Coll Engn Istanbul Turkey

We consider the real-life problem of a coach bus manufacturer located in Turkey, facing the problem of setting ordering quantities for a part procured from an unreliable supplier, where the number of items delivered is binomially distributed with an unknown yield parameter, p. We use the well-defined finite-horizon planning context with deterministic demand per period, purchasing, holding, and shortage costs to investigate the effectiveness of a fill-rate based approximate learning scheme in comparison to an exact Bayesian learning scheme, where observations on the supplier's delivery performance are used to update the assumed distribution ofp. We formulate the exact optimal learning problem as a Bayes-adaptive Markov decision process and solve the corresponding finite horizon stochastic dynamic program to provide insights on the value of online learning in comparison to the unrealistic perfect information (PI) and no information (NT) benchmarks. We contrast the performance of the so-called Bayesian Updating (BU) policy to other practical approaches such as using an assumed/guessed value ofp and implementing a constant safety stock. Noting the significant value of learning, we finally study the effectiveness of an approximate learning formulation that does not enjoy the asymptotic consistency and convergence properties but involves much lower computational burden, and demonstrate its confounding performance, at times beating the BU policy with exact Bayesian updates.

关键词： Supply uncertainty Optimal learning Bayesian updates stochastic dynamic programming

来源：评论

学校读者我要写书评

暂无评论

Integrated production-transshipment control policy for a two-location unreliable manufacturing system

引用

INTERNATIONAL JOURNAL OF PRODUCTION ECONOMICS 2022年 247卷 108440-108440页

作者： Dhahri, Akrem Gharbi, Ali Ouhimmou, Mustapha Univ Quebec Ecole Technol Super Syst Engn Dept Prod Syst Design & Control Lab 1100 Notre Dame St West Montreal PQ H3C 1K3 Canada Univ Quebec CIRRELT 1100 Notre Dame St West Montreal PQ H3C 1K3 Canada

An important strategy used by two-location firms is lateral transshipment which allows sharing the inventory between the locations to reduce the impact of shortages. Transshipment can particularly be appealing as a policy used between unreliable production facilities as it can reduce the effects of these uncertainties. We study a manufacturing system that allows transshipment between two failure-prone production facilities with different capacities. An integrated production-transshipment control policy is proposed with the aim of minimizing the total cost that comprises holding, backlog and transshipment costs. The structure of the proposed policy is obtained using the stochastic dynamic programming. It consists of a combination of a hedging point policy for production control and a state dependant economic transshipment order quantity. The parameters of the obtained control policy are optimized by adopting a simulation-based optimization approach. The robustness of the results is verified by performing sensitivity analysis. A comparative study that considers our proposed policy with the most relevant policies from the literature that we adapt to our system is conducted and shows that our proposed policy outperforms the others in terms of cost.

关键词： Unreliable manufacturing system Two-location facility Feedback production planning Lateral transshipment Simulation stochastic dynamic programming

来源：评论

学校读者我要写书评

暂无评论

Enhancements to explicit stochastic reservoir operation optimization method

引用

ADVANCES IN WATER RESOURCES 2022年 169卷

作者： Mousavi, S. Jamshid Ponnambalam, Kumaraswamy Celeste, Alcigeimes B. Cai, Ximing Amirkabir Univ Technol Dept Civil & Environm Engn Tehran Iran Univ Waterloo Dept Syst Design Engn Waterloo ON Canada Univ Fed Sergipe Dept Civil Engn Sao Cristovao Brazil Univ Illinois Dept Civil & Environm Engn Urbana IL USA

The Fletcher-Ponnambalam (FP) method is an explicit stochastic optimization method for design and operations management of real-world storage systems including surface water reservoir and groundwater management problems. The FP method faces no curse of dimensionality and no need for scenario generation. The paper introduces a novel implementation for the FP method, named FP-2022 here for clarity, by removing the need for nonlinear constraints and by decreasing the number of decision variables to just one third of its original value, significantly reducing solving time (similar to 27 times faster than the original formulation). Additionally, new expressions derived for the first and second moments of both reservoir release deficit and surplus variables and the already-derived expression for the second moments of reservoir storages are incorporated into the FP-2022 formulation enabling the method to reach an improved optimality for a nonlinear objective function. The enhanced procedure is applied to solving a reservoir operation optimization problem for a major dam in Brazil. The result comparisons made with other methods along with a thorough analysis of release operation policies prove the optimality of this highly numerically efficient and convenient-to-use FP-2022 method. Finally, a multi-reservoir application of the model is also tested with corrective simulations for improved estimates of some additional variables of interest. A specific constraint-handling approach regarding reservoir release lower and upper bounds is also presented. Satisfactory results are obtained for solving the Parambikulam-Aliyar reservoir system, a real world five-reservoir operation optimization problem from India.

关键词： Multireservoir operations optimization Two stage stochastic programming stochastic dynamic programming Fletcher-Ponnambalam Method Monte-Carlo simulation

来源：评论

学校读者我要写书评

暂无评论

Long-term Value of Flexibility from Flexible Assets in Building Operation

引用

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS 2022年 138卷 107811-107811页

作者： Thorvaldsen, Kasper Emil Korpas, Magnus Farahmand, Hossein Norwegian Univ Sci & Technol Dept Elect Power Engn N-7491 Trondheim Norway

In this work, we investigate how flexible assets within a residential building influence the long-term impact of operation. We use a measured-peak grid tariff (MPGT) that puts a cost on the highest single-hour peak import over the month. We apply a mathematical model of a Home Energy Management System (HEMS) together with stochastic dynamic programming (SDP), which calculates the long-term impact of operating as a non-linear expected future cost curve (EFCC) from the end of the scheduling period to the start. The proposed model is applied to a case study for a Norwegian building with smart control of a battery energy storage system (BESS), Electric vehicle (EV) charging and space heating (SH). Each of the flexible assets are investigated individually with MPGT and for an energy-based grid tariff. The results showed that EV charging has the highest peak-power impact in the system, decreasing the total electricity cost by 14.6% with MPGT when controllable compared to a reference case with passive charging. It is further shown how the EFCC helps achieve optimal timing and level of the peak demand, where it co-optimizes both real-time pricing and the MPGT.

关键词： Demand-side management Grid tariff Operational planning stochastic dynamic programming Expected future cost curve

来源：评论

学校读者我要写书评

暂无评论

Gray whale habitat use and reproductive success during seismic surveys near their feeding grounds: comparing state-dependent life history models and field data

引用

ENVIRONMENTAL MONITORING AND ASSESSMENT 2022年第SUPPL 1期194卷 733-733页

作者： Schwarz, Lisa McHuron, Elizabeth Mangel, Marc Gailey, Glenn Sychenko, Olga Univ Calif Santa Cruz Inst Marine Sci Santa Cruz CA 95064 USA Univ Calif Santa Cruz Dept Ecol & Evolutionary Biol Santa Cruz CA 95064 USA Univ Washington Cooperat Inst Climate Ocean & Ecosyst Studies Seattle WA 98195 USA Univ Bergen Dept Biol Theoret Ecol Grp N-9020 Bergen Norway Univ Washington Puget Sound Inst Tacoma WA 98402 USA Cetacean EcoSyst Res Lacey WA 98516 USA

We used a stochastic dynamic programming (SDP) model to quantify the consequences of disturbance on pregnant western gray whales during one foraging season. The SDP model has a firm basis in bioenergetics, but detailed knowledge of minimum reproductive length of females (L-min) and the relationship between length and reproductive success (R-fit) was lacking. We varied model assumptions to determine their effects on predictions of habitat use, proportion of animals disturbed, reproductive success, and the effects of disturbance. Smaller L-min values led to higher predicted nearshore habitat use. Changes in L-min and R-fit had little effect on predictions of the effect of disturbance. Reproductive success increased with increased L-min and with higher probability of reproductive success by length. Multiple seismic surveys were conducted in 2015 off the northeast coast of Sakhalin Island, with concomitant benthic prey surveys, photo-identification studies, and whale distribution sampling, thus providing a unique opportunity to compare output from SDP models with empirical observations. SDP model predictions of reproductive success and habitat use were similar with and without acoustic disturbance, and SDP predictions of reproductive success and large-scale habitat use were generally similar to values and trends in the data. However, empirical estimates of the proportion of pregnant females nearshore were much higher than SDP model predictions (a large effect, measured by Cohen's d) during the first week, and the SDP model overestimated whale density in the south and underestimated density around the mouth of Piltun Bay. Such differences in nearshore habitat use would not affect SDP predictions of reproductive success or survival under the current seismic air gun disturbance scenario.

关键词： Bayesian analysis Eschrichtius robustus Russia Population consequences of disturbance PCOD stochastic dynamic programming

来源：评论

学校读者我要写书评

暂无评论

stochastic model predictive control of photovoltaic battery systems using a probabilistic forecast model

引用

EUROPEAN JOURNAL OF CONTROL 2020年 56卷 254-264页

作者： Gross, Arne Wittwer, Christof Diehl, Moritz Fraunhofer Inst Solar Energy Syst Freiburg Freiburg Germany Albert Ludwigs Univ Freiburg Inst Mikrosyst Tech Freiburg Germany

Photovoltaic (PV) battery systems allow citizens to take part in a more sustainable energy system. Using the electric energy produced on-site usually entails a financial benefit for the consumer. Furthermore, feed-in peaks during high photovoltaic generation sometimes cause local voltage violations. Therefore, a feed-in limit applies to PV battery systems. In our study, we present a method to generate an optimal control that takes into account the forecast uncertainties. To that end, a stochastic forecast model is developed and used in a dynamic programming framework. We carry out a simulation study assuming the regulatory constraints in Germany. In this setup, our method is shown to mitigate the effects of the forecast uncertainties better than comparable methods. (c) 2020 European Control Association. Published by Elsevier Ltd. All rights reserved.

关键词： stochastic MPC stochastic dynamic programming Photovoltaic battery systems

来源：评论

学校读者我要写书评

暂无评论

Should environmental constraints be considered in linear programming based water value calculators?

引用

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS 2020年第0期117卷 105662-000页

作者： Guisandez, Ignacio Ignacio Perez-Diaz, Juan Nowak, Wolfgang Haas, Jannik Univ Politecn Madrid Dept Hydraul Energy & Environm Engn E-28040 Madrid Spain Univ Stuttgart Dept Stochast Simulat & Safety Res Hydrosyst IWS D-70569 Stuttgart Germany

We evaluate the practical usefulness of incorporating maximum ramping rates and minimum environmental flows into a linear programming based water value calculator for hydropower plants that participate in the day-ahead electricity market. The methodology consists of three steps: first computing the water value once with and once without environmental constraints, then simulating the plant operations using each water value, and finally comparing the simulation profits. A set of nine representative hydropower plants formed by combinations of three real locations (in Colombia, Norway and Spain) and three turbine configurations (from one to three Francis units) are individually analyzed. Each plant is simulated in two synthetic 10-year long series subject to fifteen combinations of maximum ramping rates and minimum flows with the two above-mentioned water values, totaling 540 simulations. The results indicate that incorporating the analyzed environmental constraints into a linear programming based water value calculator can be significantly profitable only when the hydropower plants have only one or at most two turbines.

关键词： Maximum ramping rate Minimum environmental flow Linear programming Mixed integer linear programming stochastic dynamic programming

来源：评论

学校读者我要写书评

暂无评论

The stochastic container relocation problem with flexible service policies

引用

TRANSPORTATION RESEARCH PART B-METHODOLOGICAL 2020年 141卷 116-163页

作者： Feng, Yuanjun Song, Dong-Ping Li, Dong Zeng, Qingcheng Univ Liverpool Sch Management Chatham St Chatham L69 7ZH Kent England Dalian Maritime Univ Sch Maritime Econ & Management 1 Linghai Rd Dalian 116026 Peoples R China

This paper investigates the stochastic Container Relocation Problem in which a flexible service policy is adopted in the import container retrieval process. The flexible policy al-lows the terminal operators to determine the container retrieval sequence to some extent, which provides more opportunity for reducing the number of relocations and the truck waiting times. A more general probabilistic model that captures customers' arrival preference is presented to describe the randomness for external truck arrivals within their appointed time windows. Being a multi-stage stochastic sequential decision-making problem, it is first formulated into a stochastic dynamic programming (SDP) model to minimize the expected number of relocations. Then, the SDP model is extended considering a secondary objective representing the truck waiting times. Tree search-based algorithms are adapted to solve the two models to their optimality. Heuristic algorithms are designed to seek high-quality solutions efficiently for larger problems. A discrete-event simulation model is developed to evaluate the optimal solutions and the heuristic solutions respectively on two performance metrics. Extensive computational experiments are performed based on instances from literature to verify the effectiveness of the proposed models and algorithms. (C) 2020 Elsevier Ltd. All rights reserved.

关键词： stochastic container relocation problem Appointment time window Flexible service stochastic dynamic programming Tree search-based 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

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

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：