Physical Model Informed Fault Detection and Diagnosis of Air Handling Units Based on Transformer Generative Adversarial Network

Yan, Ke; Chen, Xinke; Zhou, Xiaokang; Yan, Zheng; Ma, Jun

doi:10.1109/tii.2022.3193733

Cited by 42 publications

(3 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Firstly, the reasons for the gradient disappearance of GAN were analyzed theoretically, and then the difference between the generated distribution and the real distribution was calculated by using Wasserstein distance instead of the original JS divergence, which basically solved the problem of the gradient disappearance of GAN in theory. In addition, WGAN also theoretically analyzes the reasons why the samples generated by GAN are concentrated in a few categories and carries out experiments to show the superiority of WGAN in this respect [5]. In order to alleviate the problem of mode collapse of GAN, Pan, T. and others put forward a new design: UnrolledGAN, which believes that the reason of mode collapse of GAN is that the generator only considers the optimal solution of the current state when updating parameters.…”

Section: Literature Reviewmentioning

confidence: 99%

Product Design Process Based on GANs Model

Hu,

Chen

2024

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

In order to solve the difficulties and shortcomings when users use images to describe their requirements, a product design process based on GANs model is proposed. Build a generation countermeasure network and train an intelligent image generation model, so that the computer can quickly generate design cases, and can artificially control the generation results to realize the establishment of case base and the adjustment of image content. In addition, an image artistic style similarity measurement model is proposed to calculate the style differences between images. The results show that the measurement results of this method are consistent with the subjective evaluation of designers. The statistical results of 30 groups of data show that the probability that this method is consistent with the designer’s evaluation results is 90%. So far, in the above performance evaluation experiments, the similarity measurement results of the same/different style works are obviously different, which conforms to the rules of style similarity. In the task of ranking style similarity, this model is 90% consistent with the ranking results of designers. Conclusion: The model is effective in performance evaluation experiments, which lays a foundation for developing style-based design case retrieval and recommendation.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Product Design Process Based on GANs Model

Hu,

Chen

2024

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

show abstract

“…Zhang et al adopted a modified denoising autoencoder (DAE) with a self-attention mechanism into multi-task learning to identify imbalanced sample conditions [26]. Yan et al proposed a physical model based on transformer Wasserstein GAN to recognize fault diagnosis under imbalanced samples [27].…”

Section: Introductionmentioning

confidence: 99%

Cross-domain fault diagnosis network based on attributes and features transfer with dual classifier under limited and unbalanced datasets

Song,

Zhang,

Liu

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

Deep learning-based methods have shown great success in fault diagnosis due to their powerful feature extraction and non-linear fftting capabilities. Meanwhile, their remarkable performance is accompanied by constant operating conditions and sufffcient monitoring data. However, in real engineering environments, variable working conditions and limited and unbalanced data are common, which can widen the gap between fault diagnosis methods and real industrial applications. In this paper, we proposed a cross-domain fault diagnosis network based on a dual classiffer (CFDNet) with input being limited and unbalanced data to learn attributes and features for unsupervised domain adaptation (UDA). We found that the diagnostic performance is commonly bounded by the underlying knowledge, especially feature extraction from original data. Therefore, we designed a new feature encoder with features and relations, i.e., using a convolutional neural network and graph convolutional network, which improves extraction efffciency while retaining valuable information. Then, we discovered that enforced feature transfer can lead to negative transfer. To address this, we present a feature and attribute transfer framework, which not only achieves features transfer but also enables attributes transfer. Moreover, we observed that the limited and unbalanced datasets can cause label bias and biased training in the model. Hence, we designed dual classiffers to improve the probability of high-conffdence ffnal prediction by synthesizing diagnostic results. Comprehensive experiments conducted on three case studies demonstrate the effectiveness and superiority of our method for cross-domain fault diagnosis under limited and unbalanced datasets, which outperforms state-of-the-art (SOTA) methods in this study.

show abstract

“…The traditional ABE is an encryption process depending on multiple attributes of the private and public keys [9]. However, the scheme is inappropriate for end devices due to limited computational resources, complex management and high computational times [10][11][12]. Nevertheless, the encryptor may not be required to recognize the end user's precise details and only needs to embed the attribute into the ciphertext.…”

Section: Introductionmentioning

confidence: 99%

Multiauthority Ciphertext Policy-Attribute-Based Encryption (MA-CP-ABE) with Revocation and Computation Outsourcing for Resource-Constraint Devices

Yan,

Tu,

Alasmary

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

Fog computing accredits by utilizing the network edge while still rendering the possibility to interact with the cloud. Nevertheless, the features of fog computing are encountering several security challenges. The security of end users and/or fog servers brings a significant dilemma in implementing fog computing. The computational power of the resources constrains Internet of Things (IoT) devices in the fog-computing environment. Therefore, an attacker can easily attack. The traditional methods like attribute-based encryption (ABE) techniques are inappropriate for resource-constraint devices with protracted computing and limited computational capabilities. In this regard, we investigate a multiauthority ciphertext policy-attribute-based encryption (MA-CP-ABE) method that enables multiauthority attribute revocation and computation outsourcing. Moreover, the encryption and decryption processes of resource-constraint IoT devices are outsourced to the fog nodes. In this way, it also reduces the computational burden of the resource-constraint IoT devices. Hence, we propose MA-CP-ABE for encryption and decryption, attribute revocation and outsourcing by reducing the computational burden and securing the system. We compare the computational offloading approach with the existing techniques to prove that the proposed approach outperforms the existing approaches. The proposed method reduces the operation time for the encryption and decryption process. We outsource cryptography operations to the fog node, reducing the end user’s computational cost. Eventually, simulated outcomes are used to assess the algorithm’s computational cost.

show abstract

Physical Model Informed Fault Detection and Diagnosis of Air Handling Units Based on Transformer Generative Adversarial Network

Cited by 42 publications

References 21 publications

Product Design Process Based on GANs Model

Product Design Process Based on GANs Model

Cross-domain fault diagnosis network based on attributes and features transfer with dual classifier under limited and unbalanced datasets

Multiauthority Ciphertext Policy-Attribute-Based Encryption (MA-CP-ABE) with Revocation and Computation Outsourcing for Resource-Constraint Devices

Contact Info

Product

Resources

About