Evaluating the quality of online optimization algorithms by discrete event simulation

Dunke, Fabian; Nickel, Stefan

doi:10.1007/s10100-016-0455-6

Cited by 12 publications

(4 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Simulations allow evaluating the performances of such online algorithms based on multiple criteria. Furthermore, simulations can help with developing better online optimization algorithms for complex dynamic problems [9].…”

Section: Introductionmentioning

confidence: 99%

Benchmarking Flexible Electric Loads Scheduling Algorithms

Linden¹,

Romero²,

Weerdt³

2021

Energies

View full text Add to dashboard Cite

Due to increasing numbers of intermittent and distributed generators in power systems, there is an increasing need for demand responses to maintain the balance between electricity generation and use at all times. For example, the electrification of transportation significantly adds to the amount of flexible electricity demand. Several methods have been developed to schedule such flexible energy consumption. However, an objective way of comparing these methods is lacking, especially when decisions are made based on incomplete information which is repeatedly updated. This paper presents a new benchmarking framework designed to bridge this gap. Surveys that classify flexibility planning algorithms were an input to define this benchmarking standard. The benchmarking framework can be used for different objectives and under diverse conditions faced by electricity production stakeholders interested in flexibility scheduling algorithms. Our contribution was implemented in a software toolbox providing a simulation environment that captures the evolution of look-ahead information, which enables comparing online planning and scheduling algorithms. This toolbox includes seven planning algorithms. This paper includes two case studies measuring the performances of these algorithms under uncertain market conditions. These case studies illustrate the importance of online decision making, the influence of data quality on the performance of the algorithms, the benefit of using robust and stochastic programming approaches, and the necessity of trustworthy benchmarking.

show abstract

Section: Introductionmentioning

confidence: 99%

Benchmarking Flexible Electric Loads Scheduling Algorithms

Linden¹,

Romero²,

Weerdt³

2021

Energies

View full text Add to dashboard Cite

show abstract

“…Simulation allows evaluating the performance of such online algorithms on multiple criteria. Furthermore, simulation can help developing better online optimization algorithms for complex dynamic problems [9].…”

Section: Introductionmentioning

confidence: 99%

Benchmarking Flexible Electric Loads Scheduling Algorithms under Market Price Uncertainty

van der Linden,

Romero,

de Weerdt

2020

Preprint

View full text Add to dashboard Cite

Because of increasing amounts of intermittent and distributed generators in power systems, many demand response programs have been developed to schedule flexible energy consumption. However, proper benchmarks for comparing these methods are lacking, especially when decisions are made based on information which is repeatedly updated. This paper presents a new benchmarking tool designed to bridge this gap. Surveys that classify flexibility planning algorithms were an input to define the benchmarking standards. The framework can be used for different objectives and under diverse conditions faced by electricity energy stakeholders interested in flexibility scheduling algorithms. It includes a simulation environment that captures the evolution of look-ahead information, which enables comparing online planning and scheduling algorithms. The benchmarking tool is used to test seven planning algorithms measuring their performance under uncertain market conditions. The results show the importance of online decision making, the influence of data quality on the algorithm performance, the benefit of using robust and stochastic programming approaches, and the necessity of trustworthy benchmarking.

show abstract

“…Metaheuristic algorithms can be divided into population-based-e.g., genetic algorithms (GA), particle swarm optimization (PSO), or ant colony optimization (ACO)-and single-solution-e.g., tabu search (TS), simulated annealing (SA), or iterated local search (ILS), just to name a few. According to Dunke and Nickel (2017), there are four different combinations of simulation with optimization techniques:…”

mentioning

confidence: 99%

Modern Optimization and Simulation Methods in Managerial and Business Economics: A Review

Calvet

Torre

Goyal

et al. 2020

Administrative Sciences

View full text Add to dashboard Cite

Managerial and Business Economics (ME/BE) aims at using quantitative and computational methods to make an efficient (ideally optimal) assignment of the scarce resources owned by firms and organizations. In the current global market, characterized by a fierce competition, an optimal use of the available resources is more important than ever for guaranteeing the economical sustainability of organizations and enterprises of any size. Heuristic optimization algorithms and simulation methods have been successfully employed to analyze and enhance complex systems and processes in a myriad of ME/BE-related fields. This paper reviews recent works on the use of these methodologies in competitive markets, as well as in imperfect markets considering externalities. The paper also discusses open challenges and how state-of-the art methods combining optimization, simulation, and machine learning can contribute to properly address them.

show abstract

Evaluating the quality of online optimization algorithms by discrete event simulation

Cited by 12 publications

References 37 publications

Benchmarking Flexible Electric Loads Scheduling Algorithms

Benchmarking Flexible Electric Loads Scheduling Algorithms

Benchmarking Flexible Electric Loads Scheduling Algorithms under Market Price Uncertainty

Modern Optimization and Simulation Methods in Managerial and Business Economics: A Review

Contact Info

Product

Resources

About