Pymoo: Multi-Objective Optimization in Python

Blank, Julian; Deb, Kalyanmoy

doi:10.1109/access.2020.2990567

Cited by 1,126 publications

(531 citation statements)

References 49 publications

Supporting

Mentioning

426

Contrasting

Unclassified

Order By: Relevance

“…A genetic algorithm run through pymoo, specifically a BRKGA (Biased Random Key Genetic Algorithm), was used for this exploration, which utilized both elitism and random restarts [8]. A brief design of experiments of the algorithm parameters was conducted to obtain good convergence performance with this algorithm.…”

Section: B Optimizationmentioning

confidence: 99%

Mobile Target Tracking Using a Reconfigurable Low Earth Orbit Constellation

2020

View full text Add to dashboard Cite

Traditional Earth-observing satellite constellations must preselect orbits to match established mission objectives, inherently limiting the capability of the system to respond to dynamic events. A more flexible approach to constellation design which maneuvers satellites to new orbits throughout the mission lifetime can enable the tracking of mobile targets; such a concept will be explored and evaluated in the following paper. Currently, an existing approach for creating a reconfigurable constellation of satellites (ReCon) demonstrates some increased functionality. However, the repeating ground track (RGT) orbits used are only well-suited for static ground targets (i.e. particular latitude and longitude points) as they are defined by their repetitious passes over the particular points on the globe; inherently these orbits are not ideal for tracking mobile targets such as ships, oil spills, hurricanes, and other weather events. By addressing this gap and designing the constellation to follow these mobile targets, data return can be of higher quality and quantity leading toward improved predictions of future movements. Traditional non-maneuverable satellites may of course capture mobile feature data; however, the quality and quantity of data returned is driven by the satellites' orbits which were not designed for the particular mobile target. A proposed concept of operations that models possible sequential satellite maneuvers as a graph has shown promise for tracking mobile targets in the case of a single satellite. In the following paper, this concept is further explored by the addition of an optimizer to this method, applied to both a single satellite and a constellation of maneuverable satellites, like ReCon. Results for a tropical storm case study indicate that with relatively low amounts of delta-V per satellite, many possible maneuver combinations are found which result in improved accesses with greater total access time and closer passes. Trade-offs are shown between total target access time, mean distance to targets upon access, and total delta-V used. The performance of increased constellation size is also explored. By having satellites dynamically respond to mobile targets such as hurricanes, satellite imagery and remote sensing can be of better quality and even contribute to better predictions for developing storms.

show abstract

Section: B Optimizationmentioning

confidence: 99%

Mobile Target Tracking Using a Reconfigurable Low Earth Orbit Constellation

2020

View full text Add to dashboard Cite

show abstract

“…The study is performed using pymoo [36]. NSGA-II is used for all problems with m = 2 and NSGA-III for all problems with more objectives.…”

Section: B Convergence Studymentioning

confidence: 99%

Realistic Constrained Multiobjective Optimization Benchmark Problems From Design

Picard

Schiffmann

2021

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

Multi-objective optimization is increasingly used in engineering to design new systems and to identify design tradeoffs. Yet, design problems often have objective functions and constraints that are expensive and highly non-linear. Combinations of these features lead to poor convergence and diversity loss with common algorithms that have not been specifically designed for constrained optimization. Constrained benchmark problems exist, but they do not necessarily represent the challenges of engineering problems. In this paper, a framework to design electro-mechanical actuators, called MODAct, is presented and 20 constrained multi-objective optimization test problems are derived from the framework with a specific focus on constraints. The full source code is made available to ease its use. The effects of the constraints are analyzed through their impact on the Pareto front as well as on the convergence performance. A constraint landscape analysis approach is followed and extended with three new metrics to characterize the search and objective spaces. The features of MODAct are compared to existing test suites to highlight the differences. In addition, a convergence analysis using NSGA-II, NSGA-III and C-TAEA on MODAct and existing test suites suggests that the design problems are indeed difficult due to the constraints. In particular, the number of simultaneously violated constraints in newly generated solutions seems key in understanding the convergence challenges. Thus, MODAct offers an efficient framework to analyze and handle constraints in future optimization algorithm design.

show abstract

“…(4) to fixed values in the range 0.5 • ≤ θ max ≤ 5 • , which we assume to encompass the region of interest between high-accuracy and low-accuracy tracking. The MOEA/D implementation from Pymoo [34] was extended with the differential evolution strategy and the diversity preservation scheme proposed in the MOEA/D-DE algorithm [33]. The algorithm was then further extended to augment the global differential evolution with local searches based on SciPy's [35] implementation of the SLSQP optimization algorithm [36], as shown in Fig.…”

Section: Numerical Optimizationmentioning

confidence: 99%

High-performance stationary solar tracking through multi-objective optimization of beam-steering lens arrays

Johnsen¹,

Aksnes²,

Torgersen³

2020

Opt. Express

View full text Add to dashboard Cite

Beam-steering lens arrays enable solar tracking using millimeter-scale relative translation between a set of lens arrays. This may represent a promising alternative to the mechanical bulk of conventional solar trackers, but until now a thorough exploration of possible configurations has not been carried out. We present an approach for designing beam-steering lens arrays based on multi-objective optimization, quantifying the trade-off between beam divergence and optical efficiency. Using this approach, we screen and optimize a large number of beam-steering lens array configurations, and identify new and promising configurations. We present a design capable of redirecting sunlight into a <2°divergence half-angle, with 73.4% average yearly efficiency, as well as a simplified design achieving 75.4% efficiency with a <3.5°d ivergence half-angle. These designs indicate the potential of beam-steering lens arrays for enabling low-cost solar tracking for stationary solar concentrators.

show abstract

Pymoo: Multi-Objective Optimization in Python

Cited by 1,126 publications

References 49 publications

Mobile Target Tracking Using a Reconfigurable Low Earth Orbit Constellation

Mobile Target Tracking Using a Reconfigurable Low Earth Orbit Constellation

Realistic Constrained Multiobjective Optimization Benchmark Problems From Design

High-performance stationary solar tracking through multi-objective optimization of beam-steering lens arrays

Contact Info

Product

Resources

About