在設備故障之前，
就知道它何時會停。
Know when a machine will fail — before it does.
機智雲以 PHM（故障預測與健康管理）技術，為製造、航太與能源產業建置 AI 預測維護系統，把非預期停機變成可規劃的保養。 imCloud builds AI-driven predictive maintenance systems using Prognostics and Health Management (PHM), turning unplanned downtime in manufacturing, aerospace and energy into scheduled maintenance.
您的產業，我們已經做過We've already done it in your industry
從飛機起落架到沖床、壓縮機，機智雲的 PHM 模型已在各種設備上實際運作。From aircraft landing gear to stamping presses and compressors, our PHM models run on real equipment in the field.
航太與國防Aerospace & Defense
關鍵零組件故障直接影響飛安，傳統定期維修成本高。Component failures threaten flight safety; time-based maintenance is costly.
- 起落架擺振預測維護系統Landing-gear shimmy predictive maintenance
- 轉子引擎故障診斷Rotary engine fault diagnosis
工具機與金屬加工Machine Tools & Metalworking
刀具斷刀毀損昂貴工件，沖床突發停機打亂排程。Tool breakage ruins costly workpieces; press breakdowns disrupt schedules.
- 沖床預測維護（金豐機器）Stamping press PdM (Chin Fong)
- 刀具崩刃預警（精密機械中心）Tool chipping alerts (PMC)
- 智慧預測性維護（東台 × 研華）Predictive maintenance (Tongtai × Advantech)
石化與能源Petrochemical & Energy
壓縮機、發電機停機代價極高，需要提前預兆。Compressor and generator outages are extremely costly; early warning is essential.
- 壓縮機 AI 預知保養（台灣中油）Compressor AI maintenance (CPC)
- PI 系統資料 AI 預兆診斷AI anomaly diagnosis on PI data
電子與製造業Electronics & Manufacturing
產線設備多、資料分散，需要一套可落地的方法與人才。Many machines and scattered data call for a practical method — and trained people.
- 數位轉型與人才培育（友達光電）Digital transformation & training (AUO)
- 烘箱預測維護（寶成工業）Oven predictive maintenance (Pou Chen)
- 堆高機損壞預警（恆智重機）Forklift failure alerts
成果，用數字說話Results you can measure
每一個專案都經過實際資料驗證，並發表於國際期刊。Every project is validated on real field data, and published in international journals.
起落架擺振預測維護系統Landing Gear Shimmy Predictive Maintenance
以機載感測訊號建立故障診斷與剩餘壽命模型，經實際進廠維修紀錄驗證。Fault diagnosis and RUL models built from on-board sensor data, validated against actual maintenance records.
閱讀案例 →Read case →PHM 智慧沖床預測維護系統PHM-based Smart Stamping Press
即時監控沖床健康度，獲 2023 年第 16 屆工具機「研究發展創新產品」佳作。Real-time press health monitoring; honored at the 16th Machine Tool R&D Innovation Awards, 2023.
閱讀案例 →Read case →工具機刀具崩刃預警系統Cutting Tool Chipping Early Warning
自我感知演算法可辨識新的崩刃型態並自動重新訓練模型。A self-aware algorithm detects new chipping patterns and retrains the model.
閱讀案例 →Read case →PHM 六大步驟，從資料到決策Six steps of PHM — from data to decisions
以數據驅動為基礎，系統化地把機台訊號轉換成「還能用多久」與「該怎麼修」。A data-driven method that systematically turns machine signals into "how long is left" and "what to fix."
數據收集Data Acquisition
Data Acquisition佈建振動、電流、溫度等感測器，或串接既有資料庫。Install vibration, current and temperature sensors, or connect to existing databases.
資料前處理Pre-processing
Pre-processing去除雜訊、切分運轉狀態、對齊時間。Denoise, segment operating states and align timestamps.
特徵提取Feature Extraction
Feature Extraction從時域、頻域萃取能反映劣化的關鍵特徵。Extract time- and frequency-domain features that track degradation.
模型訓練Modeling
Modeling機器學習、遷移學習與超參數最佳化。Machine learning, transfer learning and hyperparameter optimization.
預測與診斷Prognostics & Diagnostics
Prognostics & Diagnostics判斷故障類型，預測剩餘可用壽命（RUL）。Classify fault types and predict remaining useful life (RUL).
數據可視化Visualization
Visualization健康度儀表板與維修建議，串接 ERP 排程。Health dashboards and maintenance advice, linked to ERP scheduling.
產品與系統Products & Systems
AI 預測維護管理資訊系統AI Predictive Maintenance Management System
- 設備健康度即時監控與異常預警Real-time equipment health monitoring and alerts
- 故障診斷（含 CNN 深度學習）與剩餘壽命（RUL）預測Fault diagnosis (including CNN deep learning) and remaining useful life (RUL) prediction
- 與 ERP、MES、PI 等企業系統串接，輔助維修與排程決策Integrates with ERP, MES and PI to support maintenance and scheduling decisions
- 整合碳排資訊，支援 ESG 目標（專利 I776596）Carbon-emission integration for ESG goals (Patent I776596)
合作流程How we work with you
從一台設備開始，先驗證效益，再擴大導入。Start with one machine, prove the value, then scale.
可行性評估Feasibility Assessment
現場訪談與資料盤點，評估 PHM 適用性與預期效益。On-site interviews and data review to assess PHM fit and expected value.
概念驗證 PoCProof of Concept
佈建感測器、擷取資料，建立並驗證診斷與 RUL 模型。Deploy sensors, collect data, then build and validate diagnosis and RUL models.
系統建置與串接Deployment & Integration
PDMS 上線，與企業內部 ERP / MES 等系統串接。Roll out PDMS and connect it to in-house ERP / MES systems.
人才培育與優化Training & Optimization
企業包班培訓內部團隊，模型持續更新。Train your in-house team and keep models up to date.
PHM 人才培育PHM Talent Development
系統要有人會用，技術才能真正落地。我們為企業量身規劃工業 AI 與大數據課程。Technology only sticks when people can use it. We design industrial AI and big-data courses for your team.
- 2023工業人工智慧與工業大數據分析菁英班Industrial AI & Big Data Elite Program友達光電企業包班Corporate program for AUO
- 2021AI 產業實戰應用人才淬鍊計畫AI Industry Application Talent Program經濟部工業局 · 友達光電企業包班Industrial Development Bureau, MOEA · AUO
- 2019預測維護系統課程暨 PHM 人才培育Predictive maintenance & PHM training寶成工業Pou Chen Corporation
- 2018全國第一門工業大數據分析課程Taiwan's first university course in industrial big data analytics逢甲大學Feng Chia University
友達大學工學院經典課程講師AUO University Annual Course Award
2022
友達大學績優菁英講師AUO University Distinguished Lecturer
2019 · 友達光電彭双浪董事長頒發2019 · Presented by AUO Chairman Paul Peng
學術深度 × 產業落地Academic depth, industrial delivery
機智雲股份有限公司成立於 2020 年 4 月，是逢甲大學第 4 家衍生新創公司。核心技術 PHM 源自美國辛辛那提大學李傑教授指導的智慧維護系統中心（IMS Center）。 Founded in April 2020, imCloud is the fourth spin-off company of Feng Chia University. Our core PHM technology originates from the Center for Intelligent Maintenance Systems (IMS Center) at the University of Cincinnati, led by Prof. Jay Lee.
創辦人Founder
- 張淵仁 博士Dr. Yuan-Jen Chang
逢甲大學航太與系統工程學系教授Professor, Dept. of Aerospace & Systems Engineering, Feng Chia University - 台灣唯一親赴 IMS Center 訪問研究並學習 PHM 技術的大學教授（2018/1–2018/7）The only Taiwanese professor to conduct visiting research on PHM at the IMS Center (Jan–Jul 2018)
- 受工業局委託開辦 PHM 人才培育課程Commissioned by the Industrial Development Bureau to run PHM training
發明專利Invention Patents
- TW I724871
智慧製造與先進排程決策輔助資訊管理系統Smart manufacturing & advanced scheduling decision-support system - TW I776596
整合智慧製造與碳排放最佳化決策輔助資訊管理系統Decision-support system integrating smart manufacturing and carbon-emission optimization
產學合作研究發表Publications from Industry Projects
每一項產學合作成果，都經過同儕審查並公開發表。Results from our industry projects are peer-reviewed and published.
-
Hwang, P.-W.; Chang, Y.-J.*; Tsai, H.-C.; Tu, Y.-T.; Yang, H.-P. Comparison and Optimization of Generalized Stamping Machine Fault Diagnosis Models Using Various Transfer Learning Methodologies. Sensors 2025, 25(6), 1779 (2025.3.13). https://doi.org/10.3390/s25061779 (SCI, IF = 3.4, Rank: 122/353 = 34.56%, Q2 in the WoS category of Engineering, Electrical & Electronic.)
產學合作計畫：PHM 智慧沖床預測維護系統（金豐機器）Project: PHM-based smart stamping press (Chin Fong)
-
Chang, Y.-J.*; Hsu, H.-K.; Hsu, T.-H.; Chen, T.-T.; Hwang, P.-W. The Optimization of a Model for Predicting the Remaining Useful Life and Fault Diagnosis of Landing Gear. Aerospace 2023, 10(11), 963 (2024.3.1). (SCI, IF = 2.60, Rank: 8/34 = 23.5%, Q1 in the WoS category of Engineering, Aerospace.)
產學合作計畫：起落架擺振預測維護系統Project: Landing-gear shimmy predictive maintenance system
-
Hsu, T.-H.; Chang, Y.-J.*; Hsu, H.-K.; Chen, T.-T.; Hwang, P.-W. Predicting the Remaining Useful Life of Landing Gear with Prognostics and Health Management (PHM). Aerospace 2022, 9, 462. (SCI, IF = 2.660, Rank: 8/34 = 23.5%, Q1, Engineering, Aerospace.)
產學合作計畫：起落架擺振預測維護系統Project: Landing-gear shimmy predictive maintenance system
-
徐子軒、張淵仁*，〈中小型企業導入 PHM 的解決方案：基於雲端平台搭配低成本邊緣運算實現 PHM 應用〉，智慧自動化產業期刊，No. 50，2024 年 8 月。
產學合作計畫：PI 系統資料 AI 預兆診斷Project: AI anomaly diagnosis on PI system data
-
王淑麗、王朝民、許𩥉、周宜德、張淵仁，〈AI 狀態分類模型應用於轉動機械故障預測與壽命預估〉，石油季刊，第 58 卷第 1 期，2022 年 3 月。（優秀論文營運管理類銀獎）
產學合作計畫：壓縮機 AI 預知保養（台灣中油）Project: AI predictive maintenance for compressors (CPC Corporation)
-
許驥、張淵仁*、謝聲偉，〈機器學習應用於切削刀具異常檢測模型研究〉，技術通報，Vol. 285，pp. 30–36，2022 年 1 月。
產學合作計畫：刀具崩刃預警（精密機械研究發展中心）Project: Cutting tool chipping early warning (PMC)
-
張淵仁*、陳昱翔、許驥、賴政佑、張平昇、謝聲偉，〈線性與非線性迴歸演算法於 PHM 方法預測刀具剩餘可用壽命之比較〉，機械新刊，第 47 期，pp. 34–44，2020 年 10 月。
產學合作計畫：刀具崩刃預警（精密機械研究發展中心）Project: Cutting tool chipping early warning (PMC)
-
張淵仁，〈預測與健康管理技術（PHM）於預測維護系統（PdM）與相關應用〉，尖端材料科技協會季刊，No. 54，pp. 21–26，2019 年 1 月。SAMPE Taiwan Chapter 出版（ISSN: 2307-0706）。
合作客戶與夥伴Clients & Partners
從一台設備開始Start with one machine
告訴我們您的設備與困擾，我們會在 2 個工作天內與您聯繫，安排免費可行性評估。Tell us about your equipment and challenges. We'll reply within 2 business days to arrange a free feasibility assessment.
- COMPANY
- 機智雲股份有限公司imCloud Co., Ltd.
- yjenchang@o365.fcu.edu.tw
- PHONE
- +886-4-24517250 ext. 3950
- ADDRESS
- 407102 台中市西屯區文華路 100 號No. 100, Wenhua Rd., Xitun Dist., Taichung City 407102, Taiwan (R.O.C.)