When performing operation of the energy system in a building, it is difficult to accurately represent the uncertain long-term objectives in the short-term window, and the correlation between long-term objectives. This...
详细信息
When performing operation of the energy system in a building, it is difficult to accurately represent the uncertain long-term objectives in the short-term window, and the correlation between long-term objectives. This work investigates a strategic modelling framework for representing long-term incentives during short-term operational planning for an energy system within flexible buildings. Multi -stage backwards stochastic dynamic programming (SDP) is used to decompose the yearly operational problem into smaller stages, and create long-term cost curves that captures the cost of adjusting long-term price signals, which in this work considers seasonal thermal energy storage and recurring monthly capacity -based grid tariff costs. The proposed method is analyzed for a flexible realistic Norwegian building located in Southern Norway, where the influence both price signals have on long-term operational performance is analyzed. Results show that the framework strategically optimizes the longterm seasonal storage together with the recurring monthly grid tariff, complementing each other by using the seasonal storage to reduce peak costs during winter. With both seasonal storage and monthly demand charge, yearly operational costs are reduced by 4.3% compared to only accounting for the grid tariff. Additionally, the operational performance achieved only a 0.9% higher cost compared to yearly simulation with perfect information.
The power system faces more challenges with the development of human society. The battery energy storage system (BESS) can provide more flexibility to power systems and it can participate in multiple services in the p...
详细信息
ISBN:
(纸本)9781665451451
The power system faces more challenges with the development of human society. The battery energy storage system (BESS) can provide more flexibility to power systems and it can participate in multiple services in the power market to gain more profits. We focus on the problem that the BESS participating in both the energy and regulation markets. The BESS needs to determine how to respond to the automatic generation control (AGC) signal in the fine timescale and decide the capacity offering in the energy and regulation markets in the coarse timescale. We propose a two-timescale framework to solve this problem. We use an online method for regulation service while considering the battery's cycle life. And the simulation results of the regulation service are used to estimate the regulation revenue for each hour. Then stochastic dynamic programming (SDP) is used to calculate the capacity in the energy and regulation markets. We validate our method in the case study. It shows that the BESS can gain more profits by participating in both the energy and regulation markets.
In this paper, we focus on the farmer's risk income when using commodity futures, when price and output processes are randomly correlated and represented by jump-diffusion models. We evaluate the expected utility ...
详细信息
In this paper, we focus on the farmer's risk income when using commodity futures, when price and output processes are randomly correlated and represented by jump-diffusion models. We evaluate the expected utility of the farmer's wealth and determine the optimal consumption rate and hedging position at each point in time given the harvest timing and state variables. We find a closed form for the optimal consumption and positioning rate in the case of an investor with CARA utility. This result (see Table 3.3) is a generalization of the result of Ho (J Financ 39:351-376, 1984), which considers the special case in which price and output are diffusion models.
We use the technique of information relaxation to develop a duality-driven iterative approach (DDP) to obtain and improve confidence interval estimates for the true value of finite-horizon stochasticdynamic programmi...
详细信息
We use the technique of information relaxation to develop a duality-driven iterative approach (DDP) to obtain and improve confidence interval estimates for the true value of finite-horizon stochastic dynamic programming problems. Each iteration of the algorithm solves an optimization-expectation procedure. We show that the sequence of dual value estimates yielded from the proposed approach monotonically converges to the true value function in a finite number of dual iterations. Aiming to overcome the curse of dimensionality in various applications, we also introduce a regression-based Monte Carlo algorithm for implementation. The new approach can assess the quality of heuristic policies and, more importantly, improve them if we find that their duality gap is large. We obtain the convergence rate of our Monte Carlo method in terms of the amounts of both basis functions and the sampled states. Finally, we demonstrate the effectiveness of our method using an optimal order execution problem with market friction. The experiments show that our method can significantly improve various heuristics commonly used in the literature to obtain new policies with a satisfactory performance guarantee. When we implement DDP in the numerical example, some local optimization routines are used in the optimization step. Inspired by the work of Brown and Smith [Brown DB, Smith JE (2014) Information relaxations, duality and convex stochasticdynamic programs. Oper. Res. 62:1394-1415.], we propose an ex-post method for smooth convex dynamic programs to assess how the local optimality of the inner optimization impacts the convergence of the DDP algorithm.
In this study, operation policies were obtained for a reservoir in Michoacan, Mexico, used for irrigation and domestic water supplies. The main purpose of these policies is to optimize the uses of the water, an increa...
详细信息
In this study, operation policies were obtained for a reservoir in Michoacan, Mexico, used for irrigation and domestic water supplies. The main purpose of these policies is to optimize the uses of the water, an increasingly scarce resource everywhere. Two optimization methodologies were used;stochastic dynamic programming, that provides release decisions for each stage, and genetic algorithms coupled with a reservoir operation simulation program, to achieve annual release curves. The operation of the reservoir was evaluated using historical inflow records. Monthly requirements for crop cycles, as well as the volumes of spills and deficits were examined. Both methodologies gave inverse relationships between deficits and spilled volumes. While both methodologies proved efficient in achieving the objectives, the results of the stochastic dynamic programming showed a better performance for this system.
One of the most important and effective works of water resource planning and management is determining the specific, applicable, regulated operating policies of the Zayandehroud dam reservoir, as a case study, in whic...
详细信息
One of the most important and effective works of water resource planning and management is determining the specific, applicable, regulated operating policies of the Zayandehroud dam reservoir, as a case study, in which it should be user-friendly and straightforward for the operator. For this purpose, different methods have been proposed in which each of them has its limitations. Due to the unique capabilities of the genetic programming (GP) model, here, this method is used to determine the operating rule curves and policies of the dam reservoir. For this purpose, here, two cases are proposed in which, in the first case, each month is individually simulated and modeled. However, in the second case, all months are simulated simultaneously. A second case is proposed here to determine simple and more applicable operation rule curves. In addition, two approaches are suggested for each case in which in the first approach, the influential input variables are selected by presenting the hybrid method. In the proposed hybrid method, the artificial neural network (ANN) model is equipped with non-dominated sorting genetic (NSGA-II) algorithm leading to a hybrid method named the ANN-NSGA-II method. However, in the second approach, the influential input variables are selected automatically using the GP method. Here, the hybrid method is proposed and used to overcome the limitations of existing usual method. In other words, it is proposed to reduce the number of influential input variables of data-driven methods and select the effective ones. The obtained results of all proposed cases and approaches are presented and compared with the standard operation policy method, stochastic dynamic programming, ANN model, and NLP method. Comparison of the results shows the acceptable performance of the proposed cases and approaches. In other words, the best-obtained values of (stability index) SI index and water deficit (objective function value) are 49.3% and 32, respectively.
In this paper, we propose learning-based adaptive control based on reinforcement learning for the booking policy in sea cargo revenue management. The problem setting is that the demand distribution is unknown while th...
详细信息
In this paper, we propose learning-based adaptive control based on reinforcement learning for the booking policy in sea cargo revenue management. The problem setting is that the demand distribution is unknown while the historical data is available, and the problem is formulated as a stochastic dynamic programming model. We demonstrate the existence of an optimal control limit policy and investigate the important properties and optimal policy structures of the model. We then propose a reinforcement learning approach for the data-driven approximation of the optimal booking policy to maximize shipping line revenue. The performance of the proposed approach is very close to that of the optimal policy and superior to that of the EMSR-b algorithm. (c) 2021 Elsevier Inc. All rights reserved.
Coconut plantations throughout the Asia-Pacific region are generally characterised by the presence of low-productivity senile palms over the age of 60, which have negative impacts on farming communities, coconut proce...
详细信息
Coconut plantations throughout the Asia-Pacific region are generally characterised by the presence of low-productivity senile palms over the age of 60, which have negative impacts on farming communities, coconut processors, and the wider economy. In Fiji, despite numerous senile coconut replacement programs, 60% of coconut palms are considered senile. The purpose of this study is to provide preliminary estimates of the financial viability of a market-based approach to senile coconut palm replacement in Fiji by utilising the palms as a feedstock, for the manufacture of rotary peeled veneer, along with plantation pine and mahogany. A mathematical model capable of supporting deterministic and stochasticdynamic optimisation was developed with an objective function to maximise the gross margin of marketable veneer manufacture per hour (GMpz) by procuring the optimal allocation of logs throughout the landscape. The majority of facility location and log processing scale scenarios evaluated found that utilising large volumes of senile coconut palms for the manufacture of veneer was optimal, whilst veneering mills situated near the coconut plantations in Vanua Levu were found to maximise GMpz. Overall, the results indicate that a coconut veneer and engineered wood product (EWP) value chain could present a financially viable opportunity to support large-scale senile coconut palm replacement in Fiji.
When we consider dynamics of pouring viscous liquid, there are multiple modes;e.g., liquid jams when the size of container opening is narrow. Our past work showed that using multiple skills (e.g., tipping and shaking ...
详细信息
ISBN:
(纸本)9781665483773
When we consider dynamics of pouring viscous liquid, there are multiple modes;e.g., liquid jams when the size of container opening is narrow. Our past work showed that using multiple skills (e.g., tipping and shaking a container) is effective to handle such tasks where stochastic neural networks were used to model the multimodal dynamics in a model-based control. It was possible because we could assume the output-state probability distribution is unimodal for an input state-action distribution. However, we have found that the output-state distribution becomes multimodal where the mode of dynamics switches, and the prediction of the stochastic neural networks becomes inaccurate due to the unimodal assumption. As the consequence, the model-based control may choose a risky skill and parameters. This paper explores modeling such multimodal dynamics. Since the output distribution becomes multimodal only where the dynamics mode switches, the number of available samples might be limited. Furthermore, we do not need to explicitly handle the output multimodality since our goal is a stochastic model-based control (dynamicprogramming). Thus, we propose to introduce a Gaussian mixture model to expand the variance of output distribution of the stochastic neural networks. This model can be easily unified into existing stochastic dynamic programming. The simulation experiments of pouring viscous liquid demonstrated that the proposed method improves the risk sensitivity.
In this paper, resource allocation problems are formulated via a set of parallel birth-death processes (BDP). This way, we can model the fact that resources can be allocated to customers at different prices, and that ...
详细信息
In this paper, resource allocation problems are formulated via a set of parallel birth-death processes (BDP). This way, we can model the fact that resources can be allocated to customers at different prices, and that customers can hold them as long as they like. More specifically, a discretisation approach is applied to model resource allocation problems as a set of discrete-time BDPs, which are then integrated into one Markov decision process. The stochasticdynamics of the resulting system are also investigated. As a result, revenue management becomes a stochastic decision-making problem, where price managers can propose suitable prices to the allocation requests such that the maximum expected total revenue is obtained at the end of a predefined finite time horizon. stochastic dynamic programming is employed to solve the related optimisation problem with the support of an ad-hoc Matlab-based application. Several simulations are performed to prove the effectiveness of the proposed model and the optimisation approach.
暂无评论