Differential Evolution: A review of more than two decades of research

Bilal, B; Pant, Millie; Zaheer, Hira; García-Hernández, Laura; Abraham, Ajith

doi:10.1016/j.engappai.2020.103479

Cited by 492 publications

(122 citation statements)

References 220 publications

Supporting

Mentioning

119

Contrasting

Unclassified

Order By: Relevance

“…DE is a simple yet powerful population-based global searching algorithm for solving various optimization problems over continuous spaces. However, there are two main defects in DE, premature convergence and stagnation [19,29,30]. The improvement of previous studies mainly focuses on the following four aspects: control parameter settings [31][32][33], strategy selection [34][35][36], population topology [37,38] and mixing with other optimization algorithms [39,40].…”

Section: Inverse Design Of Multilayer Nanofilms With De-bp Methodsmentioning

confidence: 99%

Fast Inverse Design of Nanophotonics using Di fferential Evolutionn and Back-Propagation

Qiu

Luo

et al. 2021

Preprint

View full text Add to dashboard Cite

Deep learning technology have been used as a new approach for forward simulation and inverse design of nanophotonic structures. Deep learning technology greatly reduces the time of optical simulation and enables us to use back-propagation (BP) algorithm to optimize design parameters. But BP is very sensitive to the initial values and hard to converge to the optimal value for some initial values. In this research, we propose a hybrid optimization strategy that combined differential evolution (DE) with BP algorithm for the inverse design of multilayer nanofilms structures. The proposed method effectively utilizes the global parallel exploration capability of DE and the local exploitation capability of gradient descent based on BP. It can alleviate the sensitivity of the initial values for the BP algorithm and effectively compensates for the slower convergence properties of the DE. The results suggest that the hybrid DE-BP algorithm can greatly speeds up the inverse design process of multilayer nanofilms and can search in a larger parameter space that even exceeds the parameter range of the training dataset that are used to train the forward prediction neural networks.

show abstract

Section: Inverse Design Of Multilayer Nanofilms With De-bp Methodsmentioning

confidence: 99%

Fast Inverse Design of Nanophotonics using Di fferential Evolutionn and Back-Propagation

Qiu

Luo

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…We integrate model in Equation (9) by classic Runge Kutta scheme and solve the optimization stage with the so-called Differntial Evolution method. Differential Evolution (DE) [22] is an evolutionary algorithm successfully employed for global optimization [23]. The method is designed to optimize functions f : ℝ n → ℝ.…”

Section: Numerical Experimentsmentioning

confidence: 99%

Optimal piecewise constant vaccination and lockdown policies for COVID-19

Salcedo-Varela¹,

Peñuñuri

González‐Sánchez

et al. 2021

Preprint

View full text Add to dashboard Cite

We formulate a controlled system of ordinary differential equations, with vaccination and lockdown interventions as controls, to simulate the mitigation of COVID-19. The performance of the controls is measured through a cost functional involving vaccination and lockdown costs as well as the burden of COVID19 quantified in DALYs. We calibrate parameters with data from Mexico City and Valle de Mexico. By using differential evolution, we minimize the cost functional subject to the controlled system and find optimal policies that are constant in time intervals of a given size. The main advantage of these policies relies on its practical implementation since the health authority has to make only a finite number of different decisions. Our methodology to find optimal policies is relatively general, allowing changes in the dynamics, the cost functional, or the frequency the policymaker changes actions.

show abstract

“…Şekil 4'te görüldüğü gibi DE algoritmasında yeni bir popülasyon oluşturmak için, mutasyon, çaprazlama ve seçim döngüsü her yinelemede birey sayısı (Np) kadar tekrarlanır (Bilal et al, 2020). Bu işlem önceden belirlenen durma koşulu sağlanıncaya kadar devam ettirilir.…”

Section: Farksal Gelişim Algoritmasıunclassified

Oransal İntegral Türevsel Denetleyici Parametrelerinin Sezgisel Optimizasyon Yöntemleri ile Ayarlanması

Cakir¹,

Tekin²

2021

European Journal of Science and Technology

View full text Add to dashboard Cite

Öz Bu makalede, radyal hareket optimizasyonunun (Radial Movement Optimization (RMO)) yeni bir varyantı olarak değiştirilmiş radyal hareket optimizasyonu (Modified Radial Movement Optimization (MRMO)) önerilmiştir. Ayrıca seçilen üç test sistem için PID denetleyicinin kazanç faktörleri, önerilen yöntem (MRMO), parçacık sürüsü optimizasyonu (Particle Swarm Optimization (PSO)), RMO, farksal gelişim (Differential Evolution, (DE)) ve genetik algoritma (Genetic Algorithm (GA)) ile en iyileştirilmiştir. Sezgisel optimizasyon yöntemleri ile parametre ayarında mutlak hata toplamı, hata karesi toplamı, zaman ağırlıklı mutlak hata toplamı ve zaman ağırlıklı hata karesi toplamı olmak üzere dört farklı hata alanı tabanlı başarım kriteri kullanılmıştır. Öncelikle bu beş sezgisel optimizasyon yöntemi (MRMO, PSO, RMO, DE ve GA) destekli PID denetleyicinin başarımı seçilen hata alanı kriterine göre kıyaslanmış olup, ardından optimizasyon yöntemlerinin performansları birbirleri ile karşılaştırılmış ve elde edilen sonuçlar analiz edilmiştir. Literatürde PID denetleyici parametrelerinin PSO, DE, GA ve RMO kullanılarak ayarlanmasına yönelik çok sayıda çalışma mevcuttur. Ancak PID denetleyicinin kazanç değerlerinin önerilen MRMO ile en iyileştirilmesine yönelik bir çalışma bulunmamaktadır. Bu yenilik sunulan çalışmanın mevcut literatüre en önemli katkısıdır.

show abstract

Differential Evolution: A review of more than two decades of research

Cited by 492 publications

References 220 publications

Fast Inverse Design of Nanophotonics using Di fferential Evolutionn and Back-Propagation

Fast Inverse Design of Nanophotonics using Di fferential Evolutionn and Back-Propagation

Optimal piecewise constant vaccination and lockdown policies for COVID-19

Oransal İntegral Türevsel Denetleyici Parametrelerinin Sezgisel Optimizasyon Yöntemleri ile Ayarlanması

Contact Info

Product

Resources

About