Constructive Neural Networks and Their Application to Ship Multidisciplinary Design Optimization

Besnard, Eric; Schmitz, Adeline; Hefazi, Hamid; Shinde, Rahul

doi:10.5957/jsr.2007.51.4.297

Cited by 17 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Conversely, approaches based on cascade correlation have offered a constructive approach that focused on growing network depth instead of width. These approaches use a cascaded architecture where the network grows many hidden layers the width of a single unit [31,[48][49][50][51][52].…”

Section: Introductionmentioning

confidence: 99%

Dynamic multilayer growth: Parallel vs. sequential approaches

Ross,

Berberian,

Nikolla

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

The decision of when to add a new hidden unit or layer is a fundamental challenge for constructive algorithms. It becomes even more complex in the context of multiple hidden layers. Growing both network width and depth offers a robust framework for leveraging the ability to capture more information from the data and model more complex representations. In the context of multiple hidden layers, should growing units occur sequentially with hidden units only being grown in one layer at a time or in parallel with hidden units growing across multiple layers simultaneously? The effects of growing sequentially or in parallel are investigated using a population dynamics-inspired growing algorithm in a multilayer context. A modified version of the constructive growing algorithm capable of growing in parallel is presented. Sequential and parallel growth methodologies are compared in a three-hidden layer multilayer perceptron on several benchmark classification tasks. Several variants of these approaches are developed for a more in-depth comparison based on the type of hidden layer initialization and the weight update methods employed. Comparisons are then made to another sequential growing approach, Dynamic Node Creation. Growing hidden layers in parallel resulted in comparable or higher performances than sequential approaches. Growing hidden layers in parallel promotes growing narrower deep architectures tailored to the task. Dynamic growth inspired by population dynamics offers the potential to grow the width and depth of deeper neural networks in either a sequential or parallel fashion.

show abstract

Section: Introductionmentioning

confidence: 99%

Dynamic multilayer growth: Parallel vs. sequential approaches

Ross,

Berberian,

Nikolla

et al. 2024

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…In the field of hull optimization, neural network approximation technology is widely used in approximating resistance performance. 3,7,20,34 Among various types of neural networks, BP is more commonly used. 6,16,22,28,35 In addition, neural networks are also applied to approximate seakeeping, 2,8 stability, 9 engine's propulsion power.…”

Section: Introductionmentioning

confidence: 99%

“…In recent years, the application of neural networks in approximation technology has become a hot topic. In the field of hull optimization, neural network approximation technology is widely used in approximating resistance performance 3,7,20,34 . Among various types of neural networks, BP is more commonly used 6,16,22,28,35 .…”

Section: Introductionmentioning

confidence: 99%

Hull form optimization research based on multi‐precision back‐propagation neural network approximation model

Liu,

Zhang,

Lai

et al. 2024

Numerical Methods in Fluids

View full text Add to dashboard Cite

In order to shorten the optimization cycle of ship design optimization and solve the time‐consuming problem of computational fluid dynamics (CFD) numerical calculation, this paper proposes a multi‐precision back‐propagation neural network (MP‐BP) approximation technology. Fewer high‐precision ship samples and more low‐precision ship samples were used to construct an approximate model, back‐propagation (BP) neural network was used to train multi‐precision samples. So that the approximate model is as close as possible to the real model, and achieving the effect of high‐precision approximation model. Subsequently, numerical verification and typical hull form verification are given. Based on CFD and Rankine theory, the multi‐objective design optimization framework for ship comprehensive navigation performance is constructed. The multi‐objective approximation model of KCS ship is constructed by MP‐BP approximation technology, and optimized by particle swarm optimization (PSO) algorithm. The results show that the multi‐objective optimization design framework using the MP‐BP approximation model can capture the global optimal solution and improve the efficiency of the entire hull form design optimization. It can provide a certain degree of technical support for green ship and low‐carbon shipping.

show abstract

“…The reduced speed during the berthing of unmanned ships results in a decrease in maneuvering performance. At present, the main research methods for berthing motion are the empirical formula estimation, the maneuvering test method, the hydrodynamic CFD simulation and the combination of these methods [12][13][14]. A limited number of experimental cases are available in the literature.…”

Section: Introductionmentioning

confidence: 99%

CFD Simulation of the Safety of Unmanned Ship Berthing under the Influence of Various Factors

et al. 2021

View full text Add to dashboard Cite

The safety of unmanned ship berthing is of paramount importance. In order to explore the influence of wind and wave coupling, a berthing computational fluid dynamics (CFD) model was established, and the characteristics of speed field, pressure field, and vortex have been obtained under different speed, wind direction, and the quay wall distances. The results show that the total resistance of the hull against the current can be about 1.60 times higher compared to the downstream resistance, water flow resistance is the dominant factor, accounting for more than 80% of the total resistance. When changing the distance between ship and shore at fixed speed, the results found that the torque is small, but the growth rate is very large when driving below 2 m/s, and the torque growth rate is stable above 2 m/s. Based on the established coupling model, a multi-factor berthing safety study is carried out on an actual unmanned ship. The results show that when the speed increases from 4 m/s to 12 m/s, the curve slope is small, the resistance increases from 3666 N to 18,056 N, and the rear slope increases. The pressure increases with the speed, and when the speed is 24 m/s, the maximum pressure is up to 238,869 Pa. When the wind speed is fixed, the vertical force of the unmanned ship increases first and then decreases to zero and then reverses the same law change, and the maximum resistance is about 425 N at the wind angle of about 45 degrees; At 90 degrees, the maximum lateral force on an unmanned boat is about 638 N. The above results can provide control strategy for unmanned ship berthing safety, and provide theoretical basis for unmanned ship route planning and obstacle avoidance, safety design, etc.

show abstract

Constructive Neural Networks and Their Application to Ship Multidisciplinary Design Optimization

Cited by 17 publications

References 0 publications

Dynamic multilayer growth: Parallel vs. sequential approaches

Dynamic multilayer growth: Parallel vs. sequential approaches

Hull form optimization research based on multi‐precision back‐propagation neural network approximation model

CFD Simulation of the Safety of Unmanned Ship Berthing under the Influence of Various Factors

Contact Info

Product

Resources

About