The SUBNET neural network architecture has been developed to identify nonlinear state-space models from input-output data. To achieve this, it combines the rolled-out nonlinear state-space equations and a state encode...
详细信息
The SUBNET neural network architecture has been developed to identify nonlinear state-space models from input-output data. To achieve this, it combines the rolled-out nonlinear state-space equations and a state encoder function, both parameterised as neural networks The encoder function is introduced to reconstruct the current state from past input-output data. Hence, it enables the forward simulation of the rolled-out state-space model. While this approach has shown to provide high-accuracy and consistent model estimation, its convergence can be significantly improved by efficient initialization of the training process. This paper focuses on such an initialisation of the subspace encoder approach using the Best Linear Approximation (BLA). Using the BLA provided state-space matrices and its associated reconstructability map, both the state-transition part of the network and the encoder are initialized. The performance of the improved initialisation scheme is evaluated on a Wiener-Hammerstein simulation example and a benchmark dataset. The results show that for a weakly nonlinear system, the proposed initialisation based on the linear reconstructability map results in a faster convergence and a better model quality.
This study proposes a method for recognizing the engravings on the spigot ends of ductile iron pipes based on an improved YOLOX model. Ductile iron pipes, known for their excellent mechanical properties and corrosion ...
详细信息
The paper considers the problem of fault identification in dynamic systems described by linear differential equations. To solve this problem, diagnostic observers constructed on the basis of optimal control methods ar...
详细信息
Three-phase inverters have been widely implemented for stand-alone power conversion applications where the utility grid is not available. In these applications, high-quality output voltage regulation of inverters is c...
详细信息
This paper proposes, for the first time, closed-form stability conditions and differentiation error upper bounds for arbitrary orders of Levant's robust exact Differentiator. Based on these conditions, a tuning ru...
详细信息
This paper proposes, for the first time, closed-form stability conditions and differentiation error upper bounds for arbitrary orders of Levant's robust exact Differentiator. Based on these conditions, a tuning rule is provided to select the Differentiator parameters. Numerical examples demonstrate the application of the proposed tuning rule and compare the conservativeness of the obtained conditions to existing results.
In this study, we propose a novel nanorobot model for the early detection of Alzheimer's disease. As a first step, we explored the polymorphism of the TREM2 gene's exons to assess its association with Alzheime...
详细信息
Co-salient object detection (CoSOD) is to find the salient and recurring objects from a series of relevant images, where modeling inter-image relationships plays a crucial role. Different from the commonly used direct...
详细信息
This paper focuses on the output trajectory tracking problem for linear parabolic distributed parameter systems (DPSs) using sampled-data iterative learning control (ILC) approach. In addition, in order to reduce the ...
详细信息
The tracker based on Siamese neural network is currently a technical method with high accuracy in the tracking field. With the introduction of transformer in the visual tracking field, the attention mechanism has grad...
详细信息
Annular gate MOSFETs have emerged as potential solutions for radiation-hardened electronics, demonstrating exceptional total ionizing dose tolerance. However, accurate modeling of annular MOSFETs is crucial when desig...
详细信息
暂无评论