| 1. |
|
|
| 2. |
- Lai, Zou, et al.
(författare)
-
BearingFM: Towards a foundation model for bearing fault diagnosis by domain knowledge and contrastive learning
- 2024
-
Ingår i: International Journal of Production Economics. - : Elsevier BV. - 0925-5273 .- 1873-7579. ; 275
-
Tidskriftsartikel (refereegranskat)abstract
- Monitoring bearing failures in production equipment can effectively prevent finished product quality issues and unplanned factory downtime, thereby reducing supply chain uncertainties and risk. Therefore, monitoring bearing failures in production equipment is important for improving supply chain sustainability. Due to the generalization limitations of neural network models, specific models must be trained for specific tasks. However, in real industrial scenarios, there is a severe lack of labeled samples, making it difficult to deploy fault diagnosis models across massive amounts of equipment in workshops. In order to solve the above issue, this paper proposes a cloud-edge-end collaborative semi-supervised learning framework, which provides multi-level computing power and data support for building a foundation model. A data augmentation method based on the bearing fault mechanism is proposed, which effectively preserves the inherent essential characteristics in vibration signals by normalizing frequency and adding noise in specific frequency bands. A novel contrastive learning model is designed, which narrows the distances between positive samples and widens the distances between negative samples in the high-dimensional space through cross comparisons in the time dimension and knowledge dimension, thereby extracting the most essential characteristics from the unlabeled signals. Multiple sets of experiments conducted on four datasets demonstrate that the proposed approach achieves an approximately 98% fault classification accuracy with only 1.2% labeled samples.
|
|
| 3. |
- Yang, Chen, et al.
(författare)
-
A Novel Bearing Fault Diagnosis Method based on Stacked Autoencoder and End-edge Collaboration
- 2023
-
Ingår i: Proceedings of the 2023 26th International Conference on Computer Supported Cooperative Work in Design, CSCWD 2023. - : Institute of Electrical and Electronics Engineers (IEEE). ; , s. 393-398
-
Konferensbidrag (refereegranskat)abstract
- The deep learning based fault diagnosis methods show excellent performance. However, cost and delay factors make it difficult for their widespread industrial application. Microcontroller units (MCUs) in industrial equipment have the advantages of real-time response and high reliability and usually have some redundant computational resource. However, even lightweight deep learning models cannot be deployed in MCUs due to severely limited computational resources. This paper proposes an end-edge collaborative fault diagnosis framework, by combining real-time decision-making at the end with dynamic adaptive diagnosis at the edge to improve inference performance. The model's minimum input size is deduced through theoretical analysis of the bearing working mechanism, and to make the model suitable for MCUs, we leverage the differential characteristics of the bearing vibration data and proposed a TinyML model based on stacked autoencoders. The pre-autoencoder extracts differential features, while the post-autoencoder performs fault diagnosis based on pooled differential features. Finally, the stacked-autoencoder model and collaborative framework were evaluated using the CWRU bearing dataset, achieving 384x compression in parameter size and 100% accuracy for binary fault classification, requiring only 6.44kB RAM. With the dynamic adaptive collaboration mechanism, the proposed fault diagnosis framework can reduce the edge load by approximately 94%.
|
|
| 4. |
- Yang, Chen, et al.
(författare)
-
Big Data Driven Edge-Cloud Collaboration Architecture for Cloud Manufacturing : A Software Defined Perspective
- 2020
-
Ingår i: IEEE Access. - : IEEE. - 2169-3536. ; 8, s. 45938-45950
-
Tidskriftsartikel (refereegranskat)abstract
- In the practice of cloud manufacturing, there still exist some major challenges, including: 1) cloud based big data analytics and decision-making cannot meet the requirements of many latency-sensitive applications on shop floors; 2) existing manufacturing systems lack enough reconfigurability, openness and evolvability to deal with shop-floor disturbances and market changes; and 3) big data from shop-floors and the Internet has not been effectively utilized to guide the optimization and upgrade of manufacturing systems. This paper proposes an open evolutionary architecture of the intelligent cloud manufacturing system with collaborative edge and cloud processing. Hierarchical gateways connecting and managing shop-floor things at the "edge" side are introduced to support latency-sensitive applications for real-time responses. Big data processed both at the gateways and in the cloud will be used to guide continuous improvement and evolution of edge-cloud systems for better performance. As software tools are becoming dominant as the "brain" of manufacturing control and decision-making, this paper also proposes a new mode - "AI-Mfg-Ops" (AI enabled Manufacturing Operations) with a supporting software defined framework, which can promote fast operation and upgrading of cloud manufacturing systems with smart monitoring-analysis-planning-execution in a closed loop. This research can contribute to the rapid response and efficient operation of cloud manufacturing systems.
|
|
| 5. |
- Yang, Chen, et al.
(författare)
-
Cloud-edge-device collaboration mechanisms of deep learning models for smart robots in mass personalization
- 2022
-
Ingår i: Robotics and Computer-Integrated Manufacturing. - : Elsevier BV. - 0736-5845 .- 1879-2537. ; 77, s. 102351-
-
Tidskriftsartikel (refereegranskat)abstract
- Personalized products have gradually become the main business model and core competencies of many enterprises. Large differences in components and short delivery cycles of such products, however, require industrial robots in cloud manufacturing (CMfg) to be smarter, more responsive and more flexible. This means that the deep learning models (DLMs) for smart robots should have the performance of real-time response, optimization, adaptability, dynamism, and multimodal data fusion. To satisfy these typical demands, a cloud-edge-device collaboration framework of CMfg is first proposed to support smart collaborative decision-making for smart robots. Meanwhile, in this context, different deployment and update mechanisms of DLMs for smart robots are analyzed in detail, aiming to support rapid response and high-performance decision-making by considering the factors of data sources, data processing location, offline/online learning, data sharing and the life cycle of DLMs. In addition, related key technologies are presented to provide references for technical research directions in this field.
|
|
| 6. |
- Yang, Chen, et al.
(författare)
-
Research on coordinated development between metropolitan economy and gistics using big data and Haken model
- 2019
-
Ingår i: International Journal of Production Research. - : Taylor & Francis. - 0020-7543 .- 1366-588X. ; 57:4, s. 1176-1189
-
Tidskriftsartikel (refereegranskat)abstract
- To quantitatively study the relationship and mutual effects between tropolitan economy and logistics is an important, yet pending issue, ich can scientifically guide the urban planning and investment. rough the identified evaluation indexes of metropolitan logistics and onomic development, this paper first builds up an evaluation process del of metropolitan economic and logistics development, based on big ta analytics (BDA), the entropy evaluation method, and the maximum viation method. BDA can help extract the exact data about the dicators of metropolitan economy and logistics. Then a Haken model is opted to ravel out the dynamic co-evolutionary law of economy and gistics in five Chinese cities, which complements the above static aluation. The results show that the economic development is an order rameter and plays a key role in the coordinated development of tropolitan logistics and economy. However, from 2013 to 2014, these ve cities had not established an orderly evolved positive-feedback chanism through which economic development promotes the coordinated velopment of metropolitan logistics and economic development.
|
|
| 7. |
- Yang, Chen, et al.
(författare)
-
Towards IoT-enabled dynamic service optimal selection in multiple manufacturing clouds
- 2020
-
Ingår i: Journal of manufacturing systems. - : Elsevier BV. - 0278-6125 .- 1878-6642. ; 56, s. 213-226
-
Tidskriftsartikel (refereegranskat)abstract
- With the Internet of Things, it is now possible to sense the real-time status of manufacturing objects and processes. For complex Service Selection (SS) in Cloud Manufacturing, real-time information can be utilized to deal with uncertainties emerging during task execution. Moreover, in the face of diversified demands, multiple manufacturing clouds (MCs) can provide a much wider range of choices of services with their real-time status. However, most researchers have neglected the superiority of multiple MCs and failed to make a study of how to utilize the abundant and diverse resources of multiple MCs, let alone the multi-MCs service mode under dynamic environment. Therefore, we first propose a new dynamic SS paradigm that can leverage the abundant services from multiple MCs, real-time sensing ability of the Internet of Things (IoT) and big data analytics technology for knowledge and insights. In this way, providing optimal manufacturing services (with high QoS) for customers can be guaranteed under dynamic environments. In addition, considering that a relatively long time might be spent to complete a complex manufacturing task after SS, a quantified approach, based on the Analytic Hierarchy Process and big data, is proposed to evaluate whether the intended cloud manufacturing services should be reserved to make sure that eligible services are ready to use without compromising cost or time. In this paper, the problem of IoT-enabled dynamic SS across multiple MCs is formulated in detail to enable an event-driven adaptive scheduling when the model is faced with three kinds of uncertainties (of the service market, service execution and the user side respectively). Experiments with different settings are also performed, which show the advantages of our proposed paradigm and optimization model.
|
|
| 8. |
- Yang, Chen, et al.
(författare)
-
Transforming Hong Kong's warehousing industry with a novel business model : A game-theory analysis
- 2021
-
Ingår i: Robotics and Computer-Integrated Manufacturing. - : Elsevier BV. - 0736-5845 .- 1879-2537. ; 68
-
Tidskriftsartikel (refereegranskat)abstract
- The booming e-commerce and intense regional competition are pushing the transformation of Hong Kong's warehousing industry towards more automated and efficient. However, this falls into a dilemma when stakeholders do not have enough technical and operational capability (to use advanced facility and systems) or strong motivation (due to high investment and risk). To resolve this, we first propose a new business model in which a third party - the warehousing equipment supplier (WES) is introduced to the current business between the warehouse owner (WO) and the user. The WO and WES own different advantages, complement each other and have the potential to "make the pie bigger" and promote the transformation. Then we utilize cooperative game-theory approaches (the Cournot game and the Shapley value) to explore the possible equilibrium in the new business model on profit distribution, the essential conditions for this new model to succeed and what factors that determine and affect the efficiency of the game. The experiments using the real data set show that market demand and its sensitivity to service quality, service price, price elasticity, etc. can exert impacts on the co-operative surplus and lead to the feasibility/infeasibility of the business model.
|
|
| 9. |
- Zaher, Usama, et al.
(författare)
-
GISCOD: General Integrated Solid Waste Co-Digestion model
- 2009
-
Ingår i: Water Research. - : Elsevier BV. - 1879-2448 .- 0043-1354. ; 43:10, s. 2717-2727
-
Tidskriftsartikel (refereegranskat)abstract
- This paper views waste as a resource and anaerobic digestion (AD) as an established biological process for waste treatment, methane production and energy generation. A powerful simulation tool was developed for the optimization and the assessment of co-digestion of any combination of solid waste streams. Optimization was aimed to determine the optimal ratio between different waste streams and hydraulic retention time by changing the digester feed rates to maximize the biogas production rate. Different model nodes based on the ADM1 were integrated and implemented on the Matlab-Simulink (R) simulation platform. Transformer model nodes were developed to generate detailed input for ADM1, estimating the particulate waste fractions of carbohydrates, proteins, lipids and inerts. Hydrolysis nodes were modeled separately for each waste stream. The fluxes from the hydrolysis nodes were combined and generated a detailed input vector to the ADM1. The integrated model was applied to a co-digestion case study of diluted dairy manure and kitchen wastes. The integrated model demonstrated reliable results in terms of calibration and optimization of this case study. The hydrolysis kinetics were calibrated for each waste fraction, and led to accurate simulation results of the process and prediction of the biogas production. The optimization simulated 200,000 days of virtual experimental time in 8 h and determined the feedstock ratio and retention time to set the digester operation for maximum biogas production rate. Published by Elsevier Ltd.
|
|
