A study on global optimization and deep neural network modeling method in performance-seeking control

Zheng, Qiangang; Fu, Dawei; Wang, Yong; Chen, Haoying; Zhang, Haibo

doi:10.1177/0959651819852477

Cited by 10 publications

(7 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…e authors described the layerwise literacy-based stochastic grade descent system (LLb-SGD) for grade-based optimization of objective functions in deep literacy, which is simple and computationally effective [17].…”

Section: Related Workmentioning

confidence: 99%

Bio-Optimization of Deep Learning Network Architectures

Shanmugavadivu

Rani

Chitra

et al. 2022

Security and Communication Networks

View full text Add to dashboard Cite

Deep learning is reaching new heights as a result of its cutting-edge performance in a variety of fields, including computer vision, natural language processing, time series analysis, and healthcare. Deep learning is implemented using batch and stochastic gradient descent methods, as well as a few optimizers; however, this led to subpar model performance. However, there is now a lot of effort being done to improve deep learning’s performance using gradient optimization methods. The suggested work analyses convolutional neural networks (CNN) and deep neural networks (DNN) using several cutting-edge optimizers to enhance the performance of architectures. This work uses specific optimizers (SGD, RMSprop, Adam, Adadelta, etc.) to enhance the performance of designs using different types of datasets for result matching. A thorough report on the optimizers’ performance across a variety of architectures and datasets finishes the study effort. This research will be helpful to researchers in developing their framework and appropriate architecture optimizers. The proposed work involves eight new optimizers using four CNN and DNN architectures. The experimental results exploit breakthrough results for improving the efficiency of CNN and DNN architectures using various datasets.

show abstract

Section: Related Workmentioning

confidence: 99%

Bio-Optimization of Deep Learning Network Architectures

Shanmugavadivu

Rani

Chitra

et al. 2022

Security and Communication Networks

View full text Add to dashboard Cite

show abstract

“…During aero-engine service period, the component performances of engine will gradually degrade. 9–11 Therefore, how to establish a high-accuracy real-time aero-engine dynamic model even though in a small flight envelope it is intractability. Nevertheless, it is an urgent need to establish an onboard adaptive real-time dynamic model to track the engine performance as preliminarily done of famous F119 turbofan engine.…”

Section: Introductionmentioning

confidence: 99%

Aero-engine dynamic model based on an improved compact propulsion system dynamic model

Zheng

Wang

Jin

et al. 2021

Proceedings of the Institution of Mechanical Engineers, Part I:

Self Cite

View full text Add to dashboard Cite

The modern advanced aero-engine control methods are onboard dynamic model–based algorithms. In this article, a novel aero-engine dynamic modeling method based on improved compact propulsion system dynamic model is proposed. The aero-engine model is divided into inlet, core engine, surge margin and nozzle models for establishing sub-model in the compact propulsion system dynamic model. The model of core engine is state variable model. The models of inlet, surge margin and nozzle are nonlinear models which are similar to the component level model. A new scheduling scheme for basepoint control vector, basepoint state vector and basepoint output vector which considers the change of engine total inlet temperature is proposed to improve engine model accuracy especially the steady. The online feedback correction of measurable parameters is adopted to improve the steady and dynamic accuracy of model. The modeling errors of improved compact propulsion system dynamic model remain unchanged when engine total inlet temperature of different conditions are the same or changes small. The model accuracy of compact propulsion system dynamic model, especially the measurable parameters, is improved by online feedback correction. Moreover, the real-time performance of compact propulsion system dynamic model and improved compact propulsion system dynamic model are much better than component level model.

show abstract

“…5,6 However, the control error is inevitably existing if the engine model is linearized due to the strong nonlinear characteristic of engine. Therefore, some scholars proposed a series of PSC optimization algorithms, [7][8][9][10][11][12][13][14][15][16][17] such as MAPS (Model-Assisted Pattern Search), 18 SQP (Sequential Quadratic Programming), 19,20 FSQP, 7 PSMA (Particle Self-Migrating Algorithm), 8,9 GA (Genetic Algorithm), 10 PSO (particle swarm optimization), 11 IA (Interval Analysis). 12 During these algorithms, the probability-based algorithms, such as GA, PSO, make engine get better engine performance.…”

Section: Introductionmentioning

confidence: 99%

Research on performance seeking control based on Beetle Antennae Search algorithm

Zheng

Xiang

Fang

et al. 2020

Measurement and Control

Self Cite

View full text Add to dashboard Cite

A novel performance seeking control) method based on Beetle Antennae Search algorithm is proposed to improve the real-time performance of performance seeking control. The Beetle Antennae Search imitates the function of antennae of beetle. The Beetle Antennae Search has better real-time performance because of the objective function only calculated twice in Beetle Antennae Search at each iteration. Moreover, the Beetle Antennae Search has global search ability. The performance seeking control simulations based on Beetle Antennae Search, Genetic Algorithm and particle swarm optimization are carried out. The simulations show that the Beetle Antennae Search has much better real-time performance than the conventional probability-based algorithms Genetic Algorithm and particle swarm optimization. The simulations also show that these three probability-based algorithms can get better engine performance, such as more thrust, less specific fuel consumption and less turbine inlet temperature.

show abstract

A study on global optimization and deep neural network modeling method in performance-seeking control

Cited by 10 publications

References 17 publications

Bio-Optimization of Deep Learning Network Architectures

Bio-Optimization of Deep Learning Network Architectures

Aero-engine dynamic model based on an improved compact propulsion system dynamic model

Research on performance seeking control based on Beetle Antennae Search algorithm

Contact Info

Product

Resources

About