Accelerated topology optimization by means of deep learning

Kallioras, Nikos Ath.; Kazakis, Georgios; Lagaros, Nikos D.

doi:10.1007/s00158-020-02545-z

Cited by 78 publications

(43 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Microgrid trading systems based on blockchains, such as power grid switching platforms, large power grids, distribution institutions and banks, and distributed energy, energy storage systems, energy converters, related power loads, and monitoring and protection equipment use a set of blockchain technology recording digital information and intelligent transactions of all parties and an innovative energy transmission and distribution system. The microgrid trading system based on blockchain is an autonomous system, which can realize self-control, protection, management, and trading [16]. Microgrid energy distribution means optimizing distributed generation characteristics, power quality requirements, and demand management.…”

Section: Microgrid System Based On Blockchainmentioning

confidence: 99%

Deep Learning Optimization of Microgrid Economic Dispatch and Wireless Power Transmission Using Blockchain

Chen

Guo

Zhao

et al. 2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

The purpose is to realize the decentralized microgrid economic dispatch, improve the information transparency and security of microgrid systems, and make the power grid move towards a clean, safe, efficient, and reliable development path. Deep learning optimization of microgrid economic dispatch and wireless power transmission based on blockchain technology are studied. First, the related theories and methods of microgrid systems, wireless power transmission, and deep learning optimization based on blockchain technology are introduced. Next, the microgrid economic dispatch is simulated and analyzed on a large scale. Finally, the comparison results between microgrid economic dispatch and common radio energy transmission technologies are analyzed. The results show that daily planning can better coordinate the state of distributed generation, energy storage system, and public connection. The operation results of the previous day correspond to the long-term operation economy of the microgrid. The total operation cost of the microgrid is 4668 yuan/day, and the remaining power is maintained between 500 and 600 kW, which helps to prevent excessive battery discharge, prolong battery life, and reduce operation cost. The simulation results show that the total power imbalance of the microgrid can reduce the output fluctuation of controllable load shedding of distributed generation. When the load characteristics are not important, the output fluctuation of controllable distributed generation can be reduced. The proposed economic dispatch model can optimize the data security, information storage, and information release of the microgrid and has a certain guiding role for the development of the national power grid and power industry.

show abstract

Section: Microgrid System Based On Blockchainmentioning

confidence: 99%

Deep Learning Optimization of Microgrid Economic Dispatch and Wireless Power Transmission Using Blockchain

Chen

Guo

Zhao

et al. 2022

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

show abstract

“…where Z is calculated from the summation of all probabilities of existing pairs [1], θ represents the weights and biases of connections between layers of the network. The training of a DBM is accomplished with the use of a greedy layer-wise pre-training, as in the case of DBNs, by decomposing the DBM into RBMs and using back-propagation.…”

Section: Deep Learningmentioning

confidence: 99%

“…In cases of pattern recognition problems, DBNs are able to discover the different levels of representation of the data nonlinearity. This capability of DBNs motivated the recent development of the so-called Deep Learning-assisted Topology Optimization (DLTOP) methodology by the authors of [1], which enables reducing the iterations required by SIMP. The novelty of DLTOP lies in the use of deep learning techniques (DBNs) to propose a close-to-final optimized configuration of a structural system, based upon the system configuration after only a few SIMP iterations.…”

Section: The Deep Learning-assisted Topology Optimization (Dltop) Met...mentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning-Based Accuracy Upgrade of Reduced Order Models in Topology Optimization

2021

Self Cite

View full text Add to dashboard Cite

Topology optimization problems pose substantial requirements in computing resources, which become prohibitive in cases of large-scale design domains discretized with fine finite element meshes. A Deep Learning-assisted Topology OPtimization (DLTOP) methodology was previously developed by the authors, which employs deep learning techniques to predict the optimized system configuration, thus substantially reducing the required computational effort of the optimization algorithm and overcoming potential bottlenecks. Building upon DLTOP, this study presents a novel Deep Learning-based Model Upgrading (DLMU) scheme. The scheme utilizes reduced order (surrogate) modeling techniques, which downscale complex models while preserving their original behavioral characteristics, thereby reducing the computational demand with limited impact on accuracy. The novelty of DLMU lies in the employment of deep learning for extrapolating the results of optimized reduced order models to an optimized fully refined model of the design domain, thus achieving a remarkable reduction of the computational demand in comparison with DLTOP and other existing techniques. The effectiveness, accuracy and versatility of the novel DLMU scheme are demonstrated via its application to a series of benchmark topology optimization problems from the literature.

show abstract

“…The DL-TO methods found in the literature can be divided into two parts. The first part used DL to assist traditional finite-element-TO [25,11,17] while the second part tried to replace completely TO's formulation by DL [28,30,22,19,1,14,6,10]. None of these DL-based methods included any geometrical constraints; on the contrary, they were left to the re-interpretation phase.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning Architecture for Topological Optimized Mechanical Design Generation with Complex Shape Criterion

Almasri

Bettebghor

Danglade

et al. 2021

Advances and Trends in Artificial Intelligence. Artificial Intelligence Practices

View full text Add to dashboard Cite

Topology optimization is a powerful tool for producing an optimal free-form design from input mechanical constraints. However, in its traditional-density-based approach, it does not feature a proper definition for the external boundary. Therefore, the integration of shaperelated constraints remains hard. It requires the experts' intervention to interpret the generated designs into parametric shapes; thus, making the design process time-consuming. With the growing role of additive manufacturing in the industry, developing a design approach considering mechanical and geometrical constraints simultaneously becomes an interesting way to integrate manufacturing and aesthetics constraints into mechanical design. In this paper, we propose to generate mechanically and geometrically valid designs using a deep-learning solution trained via a dual-discriminator Generative Adversarial Network (GAN) framework. This Deep-learning-geometrical-driven solution generates designs very similar to traditional topology optimization's outputs in a fraction of time. Moreover, it allows an easy shape fine-tuning by a simple increase/decrease of the input geometrical condition (here the total-bar-count), a task that a traditional topology optimization cannot achieve.

show abstract

Accelerated topology optimization by means of deep learning

Cited by 78 publications

References 47 publications

Deep Learning Optimization of Microgrid Economic Dispatch and Wireless Power Transmission Using Blockchain

Deep Learning Optimization of Microgrid Economic Dispatch and Wireless Power Transmission Using Blockchain

Deep Learning-Based Accuracy Upgrade of Reduced Order Models in Topology Optimization

Deep Learning Architecture for Topological Optimized Mechanical Design Generation with Complex Shape Criterion

Contact Info

Product

Resources

About