<style>
.red {
color: red;
}
.blue{
color: blue;
}
.green{
color: green;
}
</style>
# [Modeling of Cyber-Physical Systems and Digital Twin Based on Edge Computing, Fog Computing and Cloud Computing Towards Smart Manufacturing](https://asmedigitalcollection.asme.org/MSEC/proceedings-abstract/MSEC2018/51357/V001T05A018/277806)
:::danger
**Keywords:** Cyber-physical system (CPS), digital twin (DT), edge computing, fog computing, cloud computing, smart manufacturing.
:::
## 1. Introduction
- The advances in new generation information technologies (New IT), such as Internet of Things (IoT), big data, cloud computing, artificial intelligence (AI), etc., have had a profound impact on manufacturing.
- In this context, many countries have issued their advanced manufacturing strategies, e.g., Industry 4.0 in Germany, Industrial Internet in USA, Made in China 2025, etc.
- Although each of these strategies was proposed under different circumstances, <span class='red'>one of the common purposes of these strategies is to achieve smart manufacturing</span>.
- However, <span class='red'>to implement smart manufacturing, one specific challenge is how to converge the physical and cyber worlds of manufacturing</span>.
- **Cyber-physical integration** has gained extensive attentions from academia, industry, and the government.
- From the perspective of the manufacturing development, a fusion process between the physical and cyber worlds of manufacturing is evolving over time:
- **First stage:**
1. Firstly, before the advent of information technologies, manufacturing only involved the physical world.
2. Specifically, all manufacturing activities are done by human and/or machines.
3. The lack of information limited the manufacturing efficiency and production capacities.
- **Second stage:**
1. The **cyber world has appeared**.
2. A variety of manufacturing-related software is developed and applied in production.
3. However, due to the technological limitation, the advantage and potential of cyber world had not been fully exploited.
- **Third stage:**
1. <span class='red'>The physical world and the cyber world began to interact with each other</span>.
2. The cyber world strengthened the management of the manufacturing process, optimizing the logistics, capital flow, and information flow.
3. However, the physical and cyber worlds of manufacturing are **not synchronized**.
- **Fourth stage(Present):**
1. With the continuous developments of New IT, the real-time interaction and further integration between the physical and cyber worlds of manufacturing is becoming the inevitable trend towards smart manufacturing.
- According to either the <span class='red'>horizontal analysis from various manufacturing strategies</span>, or <span class='red'>vertical analysis from manufacturing development</span>, the cyber-physical integration of manufacturing is becoming more and more important.
- To this end, **cyber-physical systems (CPS)** and **digital twin (DT)** are the preferred technologies.
- Integrating computing, communication and control capabilities with the dynamics of physical systems:
1. **CPS** offers close interaction between cyber and physical components.
2. **Digital twin** is to <sapn class='red'>create the virtual models for physical objects in the digital way to simulate their behaviors</sapn>, serving as a bridge between the physical and the cyber worlds.
3. Through the **cyber-physical closed loop**, CPS and DT could promote the smartness of all manufacturing processes.
- Smart manufacturing is the value creation process, in which multiple subjects participate.
:::info
根據文章,「Smart manufacturing is the value creation process, in which multiple subjects participate」是指:
智慧製造是一個多方參與的價值創造過程。其中的關鍵點包括:
1. 智慧製造是一個價值創造的過程,而非單純的生產過程。
2. 在這個過程中,會有多個不同的主體(subject)參與,例如設計者、製造商、售後服務人員等。
3. 這些不同主體之間需要緊密配合協作,才能共同創造出價值。
換句話說,智慧製造已不再是單一企業內部的封閉生產活動,而是涉及多方利益相關方的開放式價值鏈。各方主體通過訊息技術的應用,實現跨系統、跨平台的互聯互通和協同最佳化,從而創造出更高的產品和服務價值。這就是文章強調的「Smart manufacturing is the value creation process, in which multiple subjects participate」的核心含義。
:::
- Depending upon <span class='red'>different magnitude of the data in smart manufacturing</span>, and <span class='red'>different time-sensitivity of smart applications</span>, CPS and DT can be divided into three different levels, i.e., the **unit level**, **system level**, and SoS (**system of system**) level.
- Each level has the data closed-loop, including **status perception**, **data analysis**, **decision making**, and **execution**.
- However, as <span class='red'>various levels have different roles and infrastructures</span>, how to build various levels of CPS and DT towards smart manufacturing becomes a central issue.
:::success
- **Edge computing**, **fog computing** and **cloud computing** may offer solutions to the above-mentioned issue.
:::
- **Cloud computing:**
1. Enables ubiquitous, convenient, on-demand network access to the shared resources pool.
2. Because of the on-demand resources sharing, high computational and storage capabilities, as well as low costs, cloud computing is a perfect technological means for **SoS-level CPS and DT**.
:::info
由於雲端計算具有以下特點,因此它是實現system of system level的CPS(cyber-physical systems)和DT(digital twin)的理想技術手段:
1. 隨需資源共享(on-demand resources sharing)
- 雲端計算可以根據需求彈性提供計算、儲存等各種資源。
2. 強大的計算和儲存能力(high computational and storage capabilities)
- 雲端擁有海量的計算和儲存資源,能滿足SoS層級CPS和DT的大數據處理需求。
3. 低成本(low costs)
- 雲計算採用按需付費、資源共享等模式,可以降低SoS層級CPS和DT的建設和運營成本。
由於上述優勢,雲端計算技術非常適合支撐複雜的SoS層級CPS和DT系統。SoS層級涉及多個不同系統的協作優化,需要海量數據儲存、分析處理能力,因此雲端計算是實現這一目標的理想技術手段。
:::
- **Fog computing:**
1. Shifts the compute, storage, and networking capabilities of cloud to the edge network.
2. By directly processing data on the network (e.g., network routers, various information systems, etc.), fog computing could help **implement system-level CPS and DT**.
- **Edge computing:**
1. Allows data processing to be performed at closer proximity to the data sources.
2. The edge can be defined as the end points with both data producers and data consumers included.
3. For **unit-level CPS and DT**, edge computing would be very beneficial.
:::info
邊緣運算(Edge Computing)中的「邊緣」(edge)可以定義為包含了數據生產者和數據消費者的終端點。
這裡的「終端點」指的是靠近數據源頭的設備,比如機器、傳感器等。<span class='red'>這些設備既是數據的產生者,也是數據的使用者</span>。
換句話說,邊緣運算是發生在這些靠近數據源頭的終端節點上的計算和處理活動,而不是發生在遠端的雲端中心。
這種靠近數據源頭的運算方式,可以降低數據傳輸延遲,提高響應速度,同時也可以減輕雲端中心的計算負載。因此,邊緣運算對於單元層級的CPS(cyber-physical system)和DT(digital twin)非常適合。
:::
:::success
- Therefore, edge computing, fog computing and cloud computing are introduced as technologies for implementing the unit-level, system-level and SoS-level of CPS and DT, respectively, towards smart manufacturing.
:::
## 2. Cyber-Physical Systems (CPS) and Digital Twin (DT) in Manufacturing
- Manufacturing is no longer the simple standalone physical machining process, but the interactive process of the physical and cyber worlds by mutual feedback.
- The **CPS** and **DT** provide technical basis for the cyber-physical fusion.
- By integrating advanced information, communication and automatic control technologies, CPS and DT enable mutual mapping, timely interaction and efficient coordination between the physical and cyber worlds.
### A. CPS in manufacturing
- CPS integrate computing, communication, control, and other technologies.
- The powerful computing and communicating capabilities in the cyber world are used to monitor, control, and coordinate the operations of engineered systems in the physical world.
- The <span class='red'>seamless integration</span> and <span class='red'>interaction of computational elements</span> and <span class='red'>physical components</span> is the key to CPS.
- With higher degree of interconnectivity and feedback, computational entities can affect physical processes and vice versa, which are in intensive communication with each other.
:::info
隨著互連性和反饋機制的不斷提高,計算實體(computational entities)和physical processes能夠相互影響,並且它們之間存在緊密的雙向通訊。
具體來說:
1. 計算實體(如嵌入式系統、控制器等)可以感知Physical processes的狀態,並根據Physical processes的變化調整自身的計算和控制行為。
2. 同時,Physical processes也會受到計算實體的影響和控制。計算實體可以對Physical processes施加影響,改變Physical processes的運行狀態。
3. 計算實體和Physical processes之間存在著密切的雙向交互和溝通,密切配合、相互作用,形成緊密的cyber-physical system。
這種高度互聯和反饋的特點,是CPS得以實現現實世界和虛擬世界深度融合的關鍵所在。Physical processes和計算過程可以密切合作,共同完成智慧製造等任務。
:::
- CPS enable the cyber world to control the physical entities in the physical world in a remote, reliable, real-time, safe, collaborative, robust, and efficient way.
- In the manufacturing environment, CPS are increasingly used as networked processing elements to control physical processes or devices, and can be characterized by smartness, connectedness and responsiveness towards internal and external changes.
:::info
CPS越來越被用作網路化的處理單元,來控制Physical Processes或設備。CPS的特點可以包括以下三點:
1. 智能性(smartness):
CPS具備感知、分析、決策和執行的能力,能夠根據內部和外部環境的變化做出適當的反應和調整。
2. 連通性(connectedness):
CPS通過網路連接,實現實體設備與計算系統的互聯互通,形成現實世界和虛擬世界的無縫融合。
3. 響應性(responsiveness):
CPS能夠對內部系統狀態變化和外部環境變化做出快速、靈活的響應。
總體而言,CPS作為智慧製造的核心技術,具備感知、分析、決策、執行的全流程智能化能力,通過網路連接實現真實世界和虛擬世界的緊密融合,並能對內外部變化做出敏捷應對。這些特點使CPS在智慧製造中扮演著日益重要的角色。
:::
1. On the basis of perception based on sensor network, the physical objects (e.g., material,machines, factory, products, etc.) are connected to the cyber world.
2. The cyber world acquires and processes large amounts of data from the physical world.
3. Through data analysis and decision making, these data are devoted to a variety of network-based applications.
:::info
通過對從現實世界獲取的數據進行分析和決策,這些數據最終被用於輔助各種基於網路的應用程式。
具體來說:
1. CPS會利用各種感測器感知現實世界的狀態,獲取大量的原始數據。
2. 這些數據會被送入CPS的計算和分析系統,經過深入的數據分析和智慧決策。
3. 分析和決策的結果,會被用於輔助和驅動各種基於網路技術的應用程式。
4. 這些應用程式可以涵蓋生產排程優化、故障預測、遠程監控等各種智慧製造領域的功能。
5. 透過CPS將現實世界的數據轉化為輔助網路應用程式的有價值訊息,實現了真實世界和網路世界的深度融合。
總之,文章描述的是CPS如何利用數據分析和決策,驅動各種基於網路技術的智慧製造應用程式,達成真實世界和網路世界的緊密合作。
:::
- Ultimately, these applications would adjust the physical world through the actuators.
- The manufacturing system has the functions of **communication**, **precise control**, **remote coordination** and **autonomy**.
- The current state-of-the-art and the latest advancement of CPS in manufacturing is presented in [18] together with directions of future research and applications in the field.
- However, little of the existing research has considered the different granularity of CPS as the basic module of resource organization.
:::info
1. 「The current state-of-the-art and the latest advancement of CPS in manufacturing is presented in [18] together with directions of future research and applications in the field.」
這段話的意思是:
文章[18]總結了CPS在製造業中的現狀和最新進展,並提出了該領域未來的研究方向和應用前景。
也就是說,文章[18]對CPS在製造業的現有技術水平和最新進展做了全面梳理,同時也展望了CPS在未來製造業中的研究重點和應用發展方向。
2. 「However, little of the existing research has considered the different granularity of CPS as the basic module of resource organization.」
這段話的意思是:
然而,現有的研究中,很少有考慮到將CPS的不同粒度(granularity)作為資源組織的基本單元。
也就是說,大部分現有研究都是將CPS整體作為一個單元來進行研究和應用,但並沒有深入思考CPS可以分為不同層級,如Unit-level、System-level和System-to-System level等,並以此作為資源組織的基礎module。
這是當前CPS研究中仍有待解決的一個重要問題,文章認為有必要針對CPS的分層特性,探索如何更好地組織和管理製造資源。
:::
### B. DT in manufacturing
- In the general manufacturing process, designers, manufacturers, and after-sales serviceman are affiliated with different departments or systems, and work independently.
- Therefore,
1. How to feed back the effective information collected from manufacturing process to the product design link.
2. How to realize the correlation and dynamic adjustment between manufacturing planning and implementation.
3. How to promote proactive MRO(Maintence, Repair, Operation) in digital way, are big challenges.
- <span class='red'>Digital twin paves a way to solve the above-mentioned challenges</span>.
- Digital twin can allow companies or users to have a **complete digital footprint** of **product from design to the end of product life cycle**.
- Digital twin is not just a mapping of static physical systems, but shall also be a dynamic simulation of physical systems.
- It contains five parts, i.e. **services**, **physical entities**, **virtual models**, the **connected data**, and the **interaction and connection between them**.
- Digital twin reflects two-way dynamic mapping of physical objects and virtual models.
1. Virtual models of physical objects are created in a digital way to simulate their behaviors in real-world environments.
2. The virtual models could guide the physical entities to respond to manufacturing changes, and to improve the operation.
:::info
數位孿生(Digital Twin)實現了Physical Entities和虛擬模型之間的雙向動態映射關係。
具體來說:
1. 數位孿生包含了physical entities的虛擬模型(virtual models)。這些虛擬模型能夠以數位的方式模擬和重現Physical Entities在現實環境中的行為。
2. 這種Physical Entities與虛擬模型之間是雙向互動的關係。一方面,虛擬模型能夠反映Physical Entities的真實狀態;另一方面,虛擬模型也可以引導和改變Physical Entities的運作。
3. 這種現實-虛擬之間的動態映射是持續的,隨著Physical Entities的變化而即時更新虛擬模型,虛擬模型也能反過來影響Physical Entities的行為。
4. 這種雙向的動態映射,可以實現真實世界和虛擬世界的密切融合,為智慧製造帶來新的可能。
總之,數位孿生實現了Physical Entities和虛擬模型之間的緊密、及時、可雙向調整的映射關係,這是實現現實-虛擬融合的關鍵。
:::
- Besides, services with their characteristics of **interoperability** and **platform independence** pave the way for large-scale smart applications and collaborations.
:::info
除此之外,具有互操作性和跨平台特性的服務,為大規模的智慧應用和合作奠定了基礎。
具體來說:
1. 互操作性(interoperability)指的是服務之間能夠無縫相連,實現不同系統、平台之間的數據和功能交換。
2. 跨平台獨立性(platform independence)意味著服務不受限於特定的硬體或軟體平台,可以在不同環境中靈活部署。
3. 這些服務特性,使得各種智慧製造應用(如個性化設計、遠端維護等)能夠大規模部署和協作運行。
4. 不同的參與主體(如設計、製造、維修等部門)可以通過這些具有互操作性和跨平台性的服務,實現更緊密的協同配合。
總之,文章強調了服務化的特性為實現智慧製造的大規模應用和跨組織協作奠定了基礎。這些服務特性有助於打造開放、靈活的智慧製造生態系統。
:::
- Through services, DT realizes the closed-loop optimization in the entire process from product design, manufacturing execution, to predictive maintenance.
- Nowadays, as the application of New IT in manufacturing, the smart manufacturing era is coming and digital twin has also been introduced into manufacturing stage.
- It is worthwhile to note that the above-mentioned digital twin applications focus primarily on production and/or product, and <span class='red'>do not consider the different granularity of DT</span>.
## 3. The Hierarchical Levels of CPS and Digital Twin in Manufacturing
- From the perspective of hierarchy, CPS and DT can be divided into three different levels, which are unit level, system level, and SoS (system of system) level.
- **Unit level of CPS:**
1. The unit level CPS is the smallest granularity with indivisibility.
2. Some small units, such as a component and equipment (e.g., NC machine tools, smart robot, etc.), material (e.g., raw material with RFID, Automated Guided Vehicle with built-in sensors, etc.) and environmental sensors, can form the unit-level of CPS.
- **Unit level of DT:**
1. The unit-level DT (e.g., component, material, equipment, etc.) can also be formed through the description and modeling in the virtual space of geometric shape, identity and function information, as well as operating status of unit-level physical entities.
2. Accompanying the physical machining, assembly, and integration, both the virtual and physical spaces of unit-level DT evolve in the process of virtual-reality interaction.
:::info
Unit-level的數位孿生(unit-level DT)也可以透過以下方式來實現:
1. 對unit-level的現時生活中的實體(如零件、材料、設備等)進行建模和描述,建立它們在虛擬空間中的幾何形狀、屬性標識、功能資訊等模型。
2. 同時也要建立這些實體的運行狀態模型,包括它們在實際生產過程中的各種參數、性能指標等。
3. 將上述虛擬空間中的幾何模型、屬性模型和運行狀態模型相互關聯,構建出生活中的實體的數位孿生。
4. 這些生活中的實體的數位孿生,可以隨著物理實體的變化而動態地演化,實現虛擬世界和真實世界的緊密映射。
簡而言之,通過對unit-level physical entities的幾何、功能和運行狀態等各方面的建模和描述,可以構建出對應的unit-level數位孿生,為智慧製造提供支撐。
:::
:::info
在現實世界的加工、裝配和綜合集成的過程中,Unit-level數位孿生(unit-level DT)的虛擬空間和實體空間會同步演化,並相互作用。
具體來說:
1. 隨著實際的物理加工、組裝和整合活動的進行,對應的物理實體(如零件、設備等)會發生變化。
2. 與此同時,unit-level數位孿生中的虛擬模型也會隨之更新和演化,以反映物理實體的變化。
3. 虛擬模型的變化又會反過來影響物理實體的後續行為,形成虛擬-現實之間的雙向互動。
4. 通過這種虛擬-現實的持續交互,unit-level數位孿生的虛擬空間和物理空間會不斷地共同演化。
簡而言之,這段話強調了unit-level數位孿生在實際生產活動中,會隨著物理實體的變化而動態地更新,並與實體空間形成緊密的反饋互動。這是實現虛擬-現實融合的重要過程。
:::
:::success
1. Compared with CPS, digital twin pays more attention to the construction of the models, including geometric shape model, rules, behavior and other constraints models, etc.
2. At the unit level, the cyber space of CPS can control the physical hardware, whereas the virtual space of DT can further perform high-fidelity visual simulation.
:::
:::info
在Unit-level,CPS(cyber-physical system)的網路空間可以控制實際的硬體,而DT(digital twin,數位孿生)的虛擬空間則可以進一步進行高保真度的視覺模擬。
具體解釋如下:
1. CPS的unit-level包含了感測器、控制器等網路化的計算實體,它們可以直接監測和控制實際的硬體設備。
2. 通過CPS的網路空間,可以及時感知物理實體的狀態,並向其發送控制指令,從而調節和控制物理設備的行為。
3. 相比之下,單元級DT則側重於建立物理實體的高度逼真的虛擬模型。
4. DT的虛擬空間可以對物理實體進行詳細的幾何建模、物理特性建模,從而進行高保真度的模擬和仿真。
5. 這種虛擬仿真有助於預測物理實體的運行狀態,為優化實際操作提供依據。
總之,CPS的網絡空間專注於對實體的及時監控和控制,而DT的虛擬空間則注重構建高度精準的物理模型,為單元級智慧製造提供了不同側重的支撐。
:::
- **System level of CPS:**
1. System level CPS is the <span class='red'>integration of multiple unit-level CPSs, which cooperate with each other</span>.
2. **Through the industrial network** (such as industrial field bus, Industrial Ethernet, etc.), multiple unit-level CPSs achieve a wider range of system-level CPS.
3. Based on the closed-loop of “perception-analysis-decision-execution” of multiple unit-level CPSs, the system-level CPS can improve the optimal allocation of manufacturing resources, and the collaboration efficiency among various resources.
:::info
基於多個Unit-level CPS(cyber-physical system)之間的"感知-分析-決策-執行"的閉環機制,System-level CPS可以提高製造資源的最佳配置,並增強各種資源之間的協作效率。
具體來說:
1. Unit-level CPS能夠透過感知設備狀態、分析數據、做出決策、執行控制命令,形成一個closed-loop。
2. 將多個Unit-level CPS組織成System-level CPS時,這種"感知-分析-決策-執行"的閉環機制也會呈現在System-level。
3. System-level CPS可以整合各個Unit的感知資訊,進行全局的分析和決策優化。
4. 基於這種系統層面的優化,System-level CPS可以實現製造資源(如設備、材料等)的最佳配置,提高資源利用效率。
5. 同時,系統級CPS還可以促進各個unit資源之間的協同合作,進一步提升整個製造系統的運行效率。
總之,文章強調了System-level CPS靠著Unit-level CPS的closed loop,能夠實現製造資源的優化配置和跨單元的高效協作,這是實現智慧製造的關鍵所在。
:::
- **System level of DT:**
1. As for digital twin, system- level physical manufacturing system (e.g., production line, shop floor, factory, etc.) is same as the system-level CPS.
2. Different from system-level CPS, <span class='red'>the virtual models of unit-level DTs need to be combined and collaborate with each other to form a system-level virtual model</span>.
3. In the interaction of the unit-level virtual models, the optimal organization of physical resources may be found.
4. Besides, a complex product may be considered as a system-level DT.
- **System of System level (CPS):**
1. By building a CPS smart service platform, collaborative optimization between system-level CPSs can be achieved.
2. At this level, multiple system-level CPSs constitute the SoS-level CPS.
3. Through the smart service platform, the cross-systems and cross-platforms interconnection and interoperability can be achieved.
4. Besides, based on SoS-level CPS, some applications involved multiple participants, such as personalized customizations, smart design, remote maintenance, etc., can be achieved.
:::warning
**For instance**, the enterprises collaboration networks derived from complex network-based manufacturing services supply-demand matching would provide different kinds of cooperation applications, such as commerce cooperation, supply chain cooperation, manufacturing cooperation, etc.
:::
:::info
透過基於複雜網路的製造服務供需匹配,可以衍生出企業之間的協作網路。這種企業協作網路將為各種合作應用提供支持,包括:
1. 商務合作(commerce cooperation)
- 例如供需雙方的電子商務交易、商業模式創新等。
2. 供應鏈合作(supply chain cooperation)
- 例如供應鏈上下游企業之間的資訊共享、物流協同等。
3. 製造合作(manufacturing cooperation)
- 例如跨企業的生產排程優化、資源共享、聯合研發等。
換句話說,這個企業協作網路是建立在複雜網路技術和製造服務供需匹配之上的,它能夠支撐各種跨企業的商業、供應鏈和製造方面的合作應用。
這種協作網路有助於突破單一企業的局限,整合多方資源,實現生產要素的優化配置和協同創新,是實現智慧製造的重要基礎。
:::
:::info
「Besides, based on SoS-level CPS, some applications involved multiple participants, such as personalized customizations, smart design, remote maintenance, etc., can be achieved.」
這段話的意思是:
除此之外,在SoS(System-of-Systems)層級的CPS(Cyber-Physical System的基礎上,一些涉及多方參與者的應用也能得以實現,比如:
1. 個性化定制(personalized customizations)
- 利用SoS層CPS整合的資源和數據,實現針對客戶的個性化產品設計和製造。
2. 智慧設計(smart design)
- 基於SoS層CPS收集的全生命週期數據,對產品設計進行智慧化優化。
3. 遠程維護(remote maintenance)
- 利用SoS層CPS提供的跨系統數據和分析能力,實現對設備的遠程監控和預防性維護。
換言之,SoS層CPS打造了一個涵蓋多個參與方的智慧製造生態系統,使得一些涉及多方協作的應用成為可能。
這種基於SoS層CPS的跨越式應用,是實現智慧製造轉型的重要體現,標誌著製造業正朝著更開放、協同、個性化的方向發展。
:::
- **System of System level (DT):**
1. It is not simply composed of system-level DTs.
2. <span class='red'>The SoS-level DT is the integration of the various stages of the product lifecycle</span>.
3. It brings the data from all aspects of product lifecycle together, hence laying the foundation for innovative product design and quality traceability.
4. SoS-level DT can not only shorten the design cycle but also greatly reduce the cost of time and money.
:::info
「The SoS-level DT is the integration of the various stages of the product lifecycle」這句話的意思是:
SoS層級的數位孿生(DT)是整合產品全生命週期各個階段的結果。
具體來說:
1. SoS層級的DT並不是單單由某個系統級DT組成,而是更高層面的集成。
2. 這種SoS層DT涵蓋了產品從設計、製造、使用到報廢等整個生命週期的各個階段。
3. 它整合了這些不同階段所產生的各種數據和訊息,形成一個貫穿全生命週期的數位化模型。
4. 將產品全生命週期的訊息融合在一起,可以為創新產品設計、質量追溯等提供強大的支撐。
5. 同時,SoS層DT也能促進不同生命週期階段之間的協同優化,提高整體效率。
總之,SoS層DT體現了對產品全生命週期的數位化集中管理,這是實現智慧製造的關鍵所在。它打造了一個涵蓋產品全生命週期的數位化協同平台。
:::
## 4. Modeling of Various Levels of CPS and DT Based on Edge Computing, Fog Computing and Cloud Computing
- The essence of CPS and DT is to acquire data from physical entities and environment thorough sensors, as well as compute and analyze them in the cyber world, so as to control the physical entities and environment.
- The various levels of CPS and DT in manufacturing <span class='red'>have different requirements for data processing and data circulation</span>, such as latency, bandwidth, security, and so on.
- The **edge computing**, **fog computing** and **cloud computing** with complementary properties, provide new ideas and way for the implementation of the unit-level, system-level and SoS-level CPS and DT.
### A. Cloud computing, fog computing and edge computing
- **Cloud Computing:**
1. Cloud computing is the computing paradigm that enables ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g., computing and storage facilities, applications, services, etc.).
2. Through **virtualization technology**, cloud computing shields the diversity of underlying devices and provides users with a variety of services in a transparent way, including IaaS (Infrastructure-as- a-Service), PaaS (Platform-as-a-Service) and SaaS (Software- as-a-Service).
3. Due to the increasing number of access devices, cloud computing may face some problems in the bandwidth, latency, network unavailability, security and privacy, etc.
- **Fog Computing:**
1. <span class='red'>Is considered as an extension of cloud computing to the edge network</span>, providing services (e.g., compute, storage, networking, etc.) closer to near-user devices (e.g., network routers, various information systems, etc.), instead of sending data to cloud.
2. In fog computing paradigm, data storage and processing rely more on local devices, rather than on cloud data center.
3. Fog computing makes the applications more convenient, meeting a wider range of node access.
:::info
根據文章,「In fog computing paradigm, data storage and processing rely more on local devices, rather than on cloud data center. Fog computing makes the applications more convenient, meeting a wider range of node access.」這段話的意思是:
在Fog Computing中:
1. 數據的儲存和處理更多依賴於在地設備,而非集中道雲端數據中心。
- 也就是說,Fog Computing將計算資源儲存到靠近數據源頭的設備上,而不是全部集中在遠端的雲端。
2. 這樣可以使相關應用更加便捷和貼近使用者。
- 因為數據處理和應用部署更靠近數據產生的終端節點,可以滿足更多類型節點的訪問需求。
總的來說,與傳統的雲端計算模式相比,Fog Computing通過在網路邊緣部署計算資源,使應用程式能夠就近處理數據,提高了應用的便捷性和靈活性。這對於實現Unit-level CPS和DT而言,是一種非常適合的技術方案。
相比集中在雲端的計算模式,Fog Computing能夠更好地滿足Unit-level CPS和DT對實時性、安全性、可靠性等方面的需求,是一種有效的補充和擴展。
:::
- **Edge-Computing:**
1. Edge computing also allows computation to be performed at the edge of the network, but at closer proximity to the data sources.
2. The difference between fog computing and edge computing is that <span class='red'>fog computing relies on interconnection capabilities among nodes, whereas edge computing runs in isolated edge nodes</span>.
3. Edge computing provides edge services near the source of data to meet the critical requirements in agile connectivity, real-time optimization, smart applications, security and privacy.
:::info
根據文章,「The difference between fog computing and edge computing is that fog computing relies on interconnection capabilities among nodes, whereas edge computing runs in isolated edge nodes.」這段話的意思是:
霧計算(Fog Computing)和邊緣計算(Edge Computing)之間的區別在於:
1. Fog Computing依賴於節點之間的互連能力:
- Fog Computing的環境中,各個節點之間會互相連接,形成一個分散式的計算網路。
- 節點之間可以進行資源共享和協作處理,共同完成任務。
2. 邊緣運算在孤立的邊緣節點上運行:
- 邊緣運算則是在相對獨立的邊緣節點上進行運算和處理。
- 這些邊緣節點雖然靠近數據源頭,但彼此之間並不強調於依賴,是相對孤立的。
簡而言之,Fog Computing強調節點之間的互連和協作,而邊緣計算則側重於單個邊緣節點本身的計算能力。
這種差異使得兩者在應用場景和實現方式上也有所不同。Fog Computing更適合需要節點間協同的應用,而邊緣運算則更適合對即時性、安全性有嚴格要求的獨立應用。
:::
### B. Modeling of various levels of CPS and DT based on edge computing, fog computing and cloud computing
- The basic requirements that need to be met in constructing a unit-level CPS and DT include the following:
1. **State perception**
2. **Compute and process data**
3. **Physical entities control**
:::info
根據文章,在構建Unit-level的CPS(cyber-physical system)和DT(digital twin,數位孿生)時,需要滿足以下基本要求:
1. 狀態感知(State perception)
- 能夠透過各種感測器,感知和採集Unit-level物理實體(如機器、設備、材料等)的狀態資訊。
2. 數據計算和處理(Compute and process data)
- 對從物理世界採集的原始數據進行計算分析,獲取有價值的資訊。
3. 物理實體控制(Physical entities control)
- 能夠根據計算分析的結果,透過執行器等裝置對物理世界的實體進行控制和調整。
這三個基本要求涵蓋了從感知、分析到執行的全流程,呈現出了Unit-level CPS和DT的核心功能:
- 感知物理實體的狀態
- 利用計算分析做出決策
- 反饋控制影響物理實體
滿足這三個基本要求,是實現單元級CPS和DT的關鍵所在,為上層系統級和SoS級提供有力支撐。
:::
- As an architecture that extends computing, networking and storage capabilities from the cloud to the edge, edge computing enables perception, computing, and control for objects through data analysis and processing in edge nodes.
- **Unit level / Edge Computing:**
1. The manufacturing resources (e.g., machine, robot, component, AGV car, etc.) with physical device (e.g., machine body, spindle, tools, etc.) and cyber part (e.g., embedded system), form the unit-level CPS and DT.
2. Through the sensors, the cyber part can monitor and perceive information from physical devices, and exert control onto the physical devices through the actuator capable of receiving control instructions.
3. Due to the perception, data analysis and control capacities, <span class='red'>edge computing could be deployed on the unit-level CPS and DT, which can be considered as an edge node</span>.
4. Because of the data circulating on the unit device, the edge computing could implement smaller applications that can help provide more real-time responses.
5. <span class='red'>Edge computing architecture does not rely on the Internet connection</span>, which is beneficial to unit-level CPS and DT.
:::info
根據文章,「Because of the data circulating on the unit device, the edge computing could implement smaller applications that can help provide more real-time responses.」這段話的意思是:
由於在單元設備上會產生大量的數據流通,因此邊緣計算(Edge Computing)技術可以在這些設備上實現一些較小規模的應用程式,從而提供更即時的回應。
具體來說:
1. 單元設備(如機器、機器人等)會產生大量的感測數據、運行狀態資訊等。
2. 這些數據如果全部傳輸到遠端的雲端進行處理,會造成網路延遲和回應時間過長的問題。
3. 而邊緣運算則可以將部分數據處理和分析任務部署到靠近數據源頭的單元設備上。
4. 這樣可以實現對數據的就近處理和即時回應,比將數據上傳雲端然後再下發控制指令更加快捷。
5. 邊緣計算設備上運行的這些小型應用程式,能夠對單元設備的狀態做出快速反應和控制調整。
總之,邊緣計算結合單元設備本地的數據處理能力,可以提供更加即時的回應,這對於Unit level CPS和DT的實現非常重要。
:::
- **System level / Fog Computing:**
1. Multiple unit-level CPSs and DTs are connected to the network through the network interface and the information management systems (e.g., ERP, MES, SCM, CRM, etc.).
2. <span class='red'>The system-level CPS and DT integrate a variety of heterogeneous unit-level CPSs and DTs</span>.
3. Besides, through the **human-computer interface (HMI)**, each unit-level CPS and DT can be accessed and controlled to monitor and diagnose the state and health of the corresponding system.
4. System-level CPS and DT **emphasize the interconnection and interoperability** among its component elements.
5. On this basis, it focuses on the real-time and dynamic collaborative control of different elements, to achieve the coordination and unification of the physical world and the cyber world.
6. In general, system-level CPS and DT are geographically concentrated (mainly within the manufacturing enterprise), which is well suited to the fog computing model.
7. The fog computing environment could be constituted by the network components (such as, routers, proxy servers, base stations and others).
8. <span class='red'>The unit-level CPS and DT use sensors to collect operating data on each device and analyze them. The fog computing environment summarize the data from all unit-level CPSs and DTs, and provide actionable information (such as cooperation with each other to finish task) for smart production.</span>
9.
:::info
根據文章,「In general, system-level CPS and DT are geographically concentrated (mainly within the manufacturing enterprise), which is well suited to the fog computing model.」這段話的意思是:
一般來說,系統級的CPS(cyber-physical system)和DT(digital twin,數位孿生)都集中在地理上相近的區域內(主要位於製造企業內部),這種特性非常適合採用霧計算(Fog Computing)的模式。
具體而言:
1. 系統級CPS和DT涉及多個單元級設備或系統的協同合作,這些設備通常集中在同一個製造企業或工廠內部。
2. 這種地理上的集中性,使得它們更適合採用Fog Computing而非完全依賴遠端的雲端計算。
3. 因為霧計算可以將計算、儲存、網路等資源部署在靠近數據源頭的邊緣節點上,而不是全部集中在雲端數據中心。
4. 這樣可以降低數據在網路上的傳輸延遲,提高系統級CPS和DT的即時性和回應效率。
5. 同時,還可以減輕雲端的計算負荷,提高整體的可靠性和安全性。
總之,系統級CPS和DT的地理局限性與Fog Computing的特點非常契合,是一個理想的組合,有助於實現更加高效的智慧製造。
:::
- **System of System level / Cloud Computing:**
1. On the basis of system-level CPS and DT, it is able to achieve the collaborative optimization between system-level CPSs and DTs, by developing the smart service platform.
2. As a result, multiple system-level CPSs and DTs constitute SoS-level CPS and DT.
3. SoS-level CPS and DT involve various participants and resources, which may be geographically dispersed.
4. The data of SoS-level CPS and DT is richer and more diverse. Therefore, the requirements that need to be met in SoS-level CPS and DT include:
- distributed data storage and processing
- provide data and smart services for enterprise collaboration.
5. Cloud computing architecture is conducive to the organization and management of a large number of connected devices, as well as the combination and integration of enterprise’s internal and external data.
6. In the cloud computing architecture, a variety of different types of storage devices can work together through the application software, to jointly provide data storage and business access for enterprises.
7. Besides, massive data mining must be supported by the distributed processing, and virtualization technologies, which are the typical characteristics of cloud computing.
8. Services with the characteristics of interoperability and platform independence, pave the way for collaboration among system-level CPSs and DTs (e.g., factories or enterprises, etc.).
:::info
根據文章,「Services with the characteristics of interoperability and platform independence, pave the way for collaboration among system-level CPSs and DTs (e.g., factories or enterprises, etc.)」這段話的意思是:
具有互操作性和跨平台獨立性特徵的服務,為系統級CPS(cyber-physical systems)和DT(digital twin, 數位孿生)之間的協作奠定了基礎。
具體來說:
1. 互操作性(interoperability)指的是服務之間能夠無縫銜接和交互,不受限於特定的系統或平台。
2. 跨平台獨立性(platform independence)則意味著服務可以在不同的硬體和軟體環境中靈活部署。
3. 這些特性使得不同SOS level CPS和DT(如工廠、企業等)能夠通過服務進行有效協作。
4. 例如,設計部門的CPS/DT和生產部門的CPS/DT可以通過互操作的服務交換資訊,實現跨部門的合作。
5. 跨企業的CPS/DT也可以利用平台獨立的服務進行聯動,達成供應鏈協作等。
總之,服務化技術的互操作性和跨平台性,為SOS CPS和DT構建開放、靈活的協作機制提供了基礎,有助於實現智慧製造生態圈的構建。
:::
:::success
In summary, the cloud computing architecture provides more architectural flexibility and utilization of external data for the innovation of product and value creation.
:::
## 5. Summary and Future Works
- Depending on the scope of resource allocation, CPS and DT can be stratified to unit- level, system-level, and SoS-level.
- Because the data generated in a different level has different requirements and uses at time scales, edge computing, fog computing and cloud computing offer ideal technical solutions for different-level CPS and DT.
- Edge computing, fog computing and cloud computing cooperate with each other, better meeting the requirements of each level CPS and DT.
:::info
文章中提到,邊緣運算(Edge Computing)、霧計算(Fog Computing)和雲端計算(Cloud Computing)三種技術相互配合,能更好地滿足不同層級CPS(Cyber-Physical System)和DT(Digital Twin,數位孿生)的需求。具體來說:
1. 邊緣運算: 部署在靠近數據源頭的設備上,能夠實現快速的感知、分析和控制,適合滿足Unit level CPS和DT的即時性要求。
2. 霧計算: 部署在網路邊緣的節點上,可以就近對System-level CPS和DT的數據進行處理和分析,改善延遲和安全性。
3. 雲端計算: 提供強大的計算和儲存能力,適合處理SoS(System-of-Systems)層級CPS和DT的海量數據和跨組織協作需求。
三種計算技術在不同層級上發揮優勢,彼此協作可以形成一個有機的整體:
- 邊緣計算滿足Unit level的實時性需求
- 霧計算處理系統級的局部數據,提高反應速度
- 雲計算則負責SoS層級的大數據分析和智慧服務
這種分工合作的方式,能夠更好地滿足CPS和DT在感知、分析、決策、控制等各個環節的不同需求,促進智慧製造的全面實施。
:::
- However, there still are the following works to be undertaken and improved with the development of New IT:
1. More edge nodes need to be built and applied in manufacturing resources, and the capacities of edge computing need to be further improved.
2. The architecture, platform, and standard of fog computing are more attractive for end users.
3. The data processed by edge computing, fog computing and cloud computing is different. Thus, the algorithms and rules for data filtering between different levels are crucial.
4. More attention should also be paid to the issues of security and energy consumption minimization.
:::info
1. 「More edge nodes need to be built and applied in manufacturing resources, and the capacities of edge computing need to be further improved.」
- 這意味著需要在更多的製造資源(如設備、機器人等)上部署邊緣節點,擴大邊緣計算的覆蓋範圍。
- 同時,還需要提升邊緣計算設備本身的計算能力、儲存能力等,以支持更複雜的應用需求。
2. 「The architecture, platform, and standard of fog computing are more attractive for end users.」
- 這表示需要進一步完善和優化Fog Computing的架構、平台以及標準,使其對終端用戶(如製造企業)更具吸引力。
- 包括提高互操作性、擴展功能性、降低成本等方面的改進,以促進Fog Computing在製造領域的廣泛應用。
3. 「The data processed by edge computing, fog computing and cloud computing is different. Thus, the algorithms and rules for data filtering between different levels are crucial.」
- 由於不同計算層級處理的數據類型和特點不同,需要制定適當的數據過濾算法和規則。
- 這樣可以確保數據在邊緣、霧和雲之間高效流轉,避免重複處理或不必要的數據傳輸。
- 合理的數據分級和分流機制,是實現三層計算合作的關鍵所在。
總之,這三個方面都需要進一步的研究和優化,以更好地支持CPS和DT在智慧製造中的應用。
:::
Sign in with Wallet
Connect another wallet