Digital Twin-Based Real-Time CNC Monitoring – Comprehensive Study Notes

Page 1

Citation & Publication Details

  • Full reference: Daraba, D.; Pop, F.; Daraba, C. (2024) “Digital Twin Used in Real-Time Monitoring of Operations Performed on CNC Technological Equipment”, Applied Sciences, 14, 10088.

  • DOI: https://doi.org/10.3390/app142210088

  • Editorial timeline: Received: 9 Oct 2024 → Revised: 26 Oct 2024 → Accepted: 28 Oct 2024 → Published: 5 Nov 2024

  • Open-access under CC-BY 4.0 (MDPI, Basel).

  • Authors & affiliations

    • Dinu Daraba – Engineering & Technology Mgmt. Dept., Technical University of Cluj-Napoca (UTCluj)

    • Florina Pop* (corresponding) – Operational Excellence, XD Connects Printmasters

    • Catalin Daraba – Engineering & Technology Mgmt., UTCluj

Featured Application

Digital Twin (DT) system for real-time retrieval of CNC machine parameters during industrial printing.

Abstract – Core Findings

  • Prototype DT solution deployed on 150 industrial printing/CNC machines across 10 000 m².

  • Architecture: SQL DB + Android clients + .NET (WPF) desktop → 1 s average data-sync rate per machine.

  • Observed benefits: ≈ 10 % reduction in production time, minimized operator/machine/product-defect delays.

  • Embedded analytics enable predictive maintenance & align with Industry 4.0/5.0.

Keywords

CNC • Digital Twin • Real-time monitoring • Process optimization • Human–machine collaboration

1 Introduction – Context

  • Rapid tech & market pressure → need productivity & process optimization.

  • Pain-point: leadership lacks real-time visibility into machine health when no DT is present.

1.1 Background

  • DT-enabled monitoring tackles inefficiencies by providing continuous insights (status, performance, predictive maintenance) \Rightarrow lower downtime & improved OEE.

1.2 Research Topic, Goals, Objectives

  • Deliver Monitoring SW v1.0:

    • SQL for transactional data flow.

    • .NET UI for operators/managers.

    • Continuous sync \approx1\;\text{s} across 150 nodes (Windows Server 2022).

  • Supports KPI tracking, preventive maintenance & Industry-4.0 adoption.

1.3 Novel Contribution

  • First study emphasizing 1-second real-time sync for 150 distributed resources.

  • Bridges theory & practice, yields: shorter cycle-times, cost minimization, competitiveness.

  • Offers integration strategies for legacy manufacturing lines.

Page 2

Real-Time DT Monitoring – Extended Narrative

  • DT builds a virtual representation for tracking key parameters, analytics & issue detection before disruption.

  • Primary objective: raise efficiency/responsiveness using SQL + .NET + 1 s sync (150 machines, Win Server 2022).

  • Direct support for preventive maintenance \Downarrow cost reduction & smoother ops.

Page 3

2 Related Work

2.1 Digital Twin & CNC Markets

  • CNC global CAGR >6\%; DT in mfg CAGR \approx25\%.

  • Traditional systems lack live KPIs (utilization, tool-wear, quality).

  • Proposed SW v1.0 fills gap by virtual replicas → bottleneck discovery, ML-predicted maintenance.

  • Market dominance: North America (≈ 36 % share) – majors: Rockwell, Microsoft, IBM, Siemens.

2.2 Early DT

  • NASA Apollo era: “DT-like” spacecraft monitoring despite limited compute.

2.3 Growth in Manufacturing

  • GE & Tesla use DT across lifecycle.

  • Gartner: >50 % of large industrials to adopt DT by 2025 → efficiency gains up to 10 %.

Page 4

  • Market forecast: \text{USD }39\,\text{bn (2026)} \rightarrow 110\,\text{bn (2028)}.

2.4 DT in CNC

  • Vendors: FANUC, Siemens (Sinumerik ONE) with real-time feedback.

2.5 DT in Healthcare & Smart Cities

  • Philips & GE Health – equipment DT for predictive maintenance.

  • Patient-specific twins for personalized medicine.

  • Smart-city exemplar: Virtual Singapore.

2.6 Future Prospects

  • Fusion with AI/ML, autonomous systems, 5G → remote surgery, autonomous vehicles, smart-city mgmt.

Page 5

2.7 DT Architecture

  • Three layers: Physical (sensors, edge, security) – Connectivity (Internet/Bluetooth) – Virtual (replica, ML/AI, DB).

2.8 DT Maturity Levels

  • Levels 0–5 (majority sit 0–3). Barriers: data collection, anomalies, device failures.

2.9 DT in Smart Manufacturing

  • NIST definition: system must react in real time. DTs support lifecycle decisions via IoT data.

Page 6

3 Methodology

3.1 WPF + MVVM for CNC UI

  • Benefits: separation of Model–View–ViewModel → scalability & testability.

  • Capabilities:

    • Real-time graphs (e.g., spindle-speed).

    • Command binding (Start/Stop).

    • Visualization libraries: OxyPlot/LiveCharts.

    • Fault alerts via INotifyPropertyChanged.

3.2 SignalR Hub (patch 2.4)

  • Ensures bi-directional low-latency comms (prefers WebSocket, falls back to SSE or Long-Polling).

  • Key equations:
    T{connect}=T{init}+T{auth}+T{SignalR}
    T{retry}=T{init}\times2^{n}
    M{received}={msg1,\dots,msgn} N{update}=M{push}+M{realtime}

3.3 Dapper ORM

  • Lightweight SQL-mapper → minimal overhead for high-frequency DT data.

  • Suitable for batch updates & multi-DB compatibility.

3.4 Android Java Operator App

  • Mobile benefits: real-time, mobility, interactive dashboards (MPAndroidChart), push notifications, offline-sync.

Page 7

4 Design & Implementation

4.1 Solution Architecture

  • Stack: .NET WPF desktop + Android Java tablets + Swagger API + SignalR + MS SQL.

4.2 DB Schema

  • MySQL, 10 normalized tables (PKs, descriptive naming, optimized datatypes).

4.3 Windows Monitoring Client

  • Project layers: Controls • Helpers • Resources • Models • ViewModels • Windows (main shell).

Real-Time Event Wiring

  • Establish SignalR connection → subscribe “LiveMapRefresh”.

AES Password Handling

D{pass}=E^{-1}{pass}(key)

4.4 Android Module

  • Global constants file, JDBC/MySQL queries, operator data-model for setup/production actions.

4.5 3-D DT Model

  • AutoCAD model + linked real-time data → simulations & continuous improvement.

Page 8 – Page 24 (Condensed Key Outcomes)

5 Results

  • Deployment: 150 machines, Win Server 2022 (Xeon Gold 6346 ×4, 16 GB RAM).

  • Performance: 714 644 actions captured in 1 week (~500 orders/day).

  • Avg sync latency \approx1\;\text{s}/machine; 10 % production-time reduction (Act.Hours comparison).

  • Visualization figures:

    • Actions histogram, connected-resources chart, latency plot, production-time bar graph, competitive-solution comparison.

6 Conclusions & Future Work

  • DT monitoring ↑ productivity, coordination & predictive maintenance.

  • Next milestones: automatic parameter harvesting, advanced analytics reports, predictive maintenance algorithms, camera/sensor fusion for single-equipment twins.

Core Equations Recap

  1. Connection establishment T{connect}=T{init}+T{auth}+T{SignalR}

  2. Retry logic T{retry}=T{init}\times2^{n}

  3. Message set M{received}={msg1\dots msg_n}

  4. Notification aggregation N{update}=M{push}+M_{realtime}

  5. AES decryption D{pass}=E^{-1}{pass}(key)

These notes capture every substantive point, numerical claim, architectural component, formula, example, and future-direction mentioned across pages 1-24, offering a complete standalone study scaffold for exam preparation.