A fundamental challenge when modeling combinatorial optimization problems is that often multiple sub-objectives need to be weighted against each other, but it is not clear how much weight each sub-objective should be ...
详细信息
ISBN:
(数字)9783031605994
ISBN:
(纸本)9783031606014;9783031605994
A fundamental challenge when modeling combinatorial optimization problems is that often multiple sub-objectives need to be weighted against each other, but it is not clear how much weight each sub-objective should be given: consider routing problems that trade off distance and duration where the relative importance of the two is not known a priori. In recent work, it has been proposed to use machine learning algorithms from the domain of structured output prediction to learn such weights from examples of desirable solutions. However, until now such techniques were only evaluated on fast-to-solve optimization problems. We propose and evaluate three techniques that make it feasible to apply the structured perceptron on NP-hard optimization problems: 1) using heuristic solving methods during the learning process, 2) solving well-chosen satisfaction variants of the problems, 3) caching solutions computed during the learning process and reusing them. Experiments confirm the validity and speed-ups of these techniques, enabling structured output learning on larger combinatorial problems than before.
The goal of this paper is to develop an evaluation system that leverages recent advances in language processing to be automatic, adaptive, and more efficient than traditional modelingsystems. From a user perspective,...
详细信息
Optimal sizing of a photovoltaics power system equipped with energy storage is of critical importance to maximize the economic revenue and to reduce the early aging of the storage devices. In this work, a simulation m...
详细信息
ISBN:
(数字)9781665442800
ISBN:
(纸本)9781665442800
Optimal sizing of a photovoltaics power system equipped with energy storage is of critical importance to maximize the economic revenue and to reduce the early aging of the storage devices. In this work, a simulation model for the evaluation of the electrical behavior of a photovoltaic system, connected to the grid and equipped with a battery storage system, is proposed. The model is implemented in Python, and it is directly connected to the European PVGis portal to use temperature and irradiance profiles from arbitrary locations. The simulation system is coupled with a metaheuristic optimization algorithm to determine the optimal sizing of the power plant, considering as cost function the energy taken from the grid. Optimal parameters such as the number of panels (PV area) and battery's capacity are obtained for several boundary conditions.
statistical optimization, or multi stage Monte Carlo optimization (MSMCO), is used here to solve a 175 variable with 175 equations nonlinear system for the true optimal solution. This problem requires that all of the ...
详细信息
Dynamic task assignment involves assigning arriving tasks to a limited number of resources in order to minimize the overall cost of the assignments. To achieve optimal task assignment, it is necessary to model the ass...
详细信息
ISBN:
(纸本)9783031416194;9783031416200
Dynamic task assignment involves assigning arriving tasks to a limited number of resources in order to minimize the overall cost of the assignments. To achieve optimal task assignment, it is necessary to model the assignment problem first. While there exist separate formalisms, specifically Markov Decision Processes and (Colored) Petri Nets, to model, execute, and solve different aspects of the problem, there is no integrated modeling technique. To address this gap, this paper proposes Action-Evolution Petri Nets (A-E PN) as a framework for modeling and solving dynamic task assignment problems. A-E PN provides a unified modeling technique that can represent all elements of dynamic task assignment problems. Moreover, A-E PN models are executable, which means they can be used to learn close-to-optimal assignment policies through Reinforcement Learning (RL) without additional modeling effort. To evaluate the framework, we define a taxonomy of archetypical assignment problems. We show for three cases that A-E PN can be used to learn close-to-optimal assignment policies. Our results suggest that A-E PN can be used to model and solve a broad range of dynamic task assignment problems.
This paper introduces a new web-based system to automate the use of machine learning techniques to support decision-makers in identifying and assessing consumer's behavior. Customer retention is one of the main pi...
详细信息
This paper introduces a new web-based system to automate the use of machine learning techniques to support decision-makers in identifying and assessing consumer's behavior. Customer retention is one of the main pillars of business development in the competitive market, and the timely identification of factors that influence customer opinions is a topic of real interest for both business and academia. Seven machine learning classification methods were applied and evaluated on an established dataset aimed at understanding the behavior of the consumer of banking services. A new web application was developed to automatize the assessment of implemented techniques and reveal the optimum one, in terms of performance, by fine-tuning the parameters of the selected models. This solution can be successfully applied on other datasets gathered to better target the marketing campaign and prioritize the promotion to customers who have a high potential to hire a new service or purchase a particular product.
This paper addresses the optimal control problem of switched nonlinear systems. Indeed, a new mechanistic algorithm called Selfish Herd optimization (SHO) is suggested. The proposed approach consists of minimizing a d...
详细信息
Over the past twenty years, there has been significant diversification and advancement in the quality of construction materials. A notable contribution is the evolution of self-compacting concrete which has, however, ...
详细信息
Collective pitch control (CPC) is normally combined with an active tower damping control (ATDC) and both control loops are commonly found in the control system of very large wind turbines. Normally, each controller is...
详细信息
In this paper, the dynamic mechanical properties of jumper cables in offshore floating photovoltaic power systems are studied. By establishing a mathematical model based on parabola theory and using finite element sim...
详细信息
ISBN:
(纸本)9798350388084;9798350388077
In this paper, the dynamic mechanical properties of jumper cables in offshore floating photovoltaic power systems are studied. By establishing a mathematical model based on parabola theory and using finite element simulation software for experimental analysis, the stability and reliability of jumper cable in Marine environment are discussed in this study. It is found that the design and laying of jumper cables are crucial to the performance of the entire power system, especially in the electrical properties, mechanical properties, weather resistance and corrosion resistance of the cables. In addition, factors such as the bending radius, tension, redundant length and protective measures of the cable also need to be carefully considered to ensure the normal operation and maintenance of the cable, reduce costs and extend the life of the cable. This study provides theoretical support for the scale, commercialization and standardization of offshore photovoltaic power systems, and has guiding significance for future design and performance optimization.
暂无评论