{"id":94467,"date":"2026-05-19T07:43:44","date_gmt":"2026-05-19T07:43:44","guid":{"rendered":"https:\/\/teeptrak.com\/predictive-maintenance-vibration-monitoring-2027\/"},"modified":"2026-06-08T13:24:16","modified_gmt":"2026-06-08T13:24:16","slug":"predictive-maintenance-vibration-monitoring-2027","status":"publish","type":"post","link":"https:\/\/teeptrak.com\/en\/predictive-maintenance-vibration-monitoring-2027\/","title":{"rendered":"Predictive maintenance vibration monitoring 2027: ISO 10816, FFT, ML algorithms, RUL prediction"},"content":{"rendered":"
\nTL;DR \u2014 Predictive maintenance vibration monitoring in 60 words<\/strong>
\nVibration monitoring is the foundation of predictive maintenance for rotating equipment. ISO 10816 \/ ISO 20816 defines vibration severity thresholds. Analysis: FFT spectrum, envelope analysis (bearings), order tracking (variable speed). ML algorithms: RUL (Remaining Useful Life) prediction, anomaly detection, RCA. Major vendors: Augury, Senseye (Siemens), GE Vernova SmartSignal, AspenTech, ABB Ability. ROI -20-40% maintenance cost, -30-50% unplanned downtime.\n<\/div>\n

Predictive maintenance (PdM) has emerged as the most impactful Industry 4.0 application for manufacturing, with proven ROI of -20-40% maintenance cost<\/strong>, -30-50% unplanned downtime<\/strong>, and +5-10 OEE points<\/strong> across diverse industries (automotive, chemicals, food & beverage, pharma, aerospace, mining, oil & gas, power generation). Vibration monitoring<\/strong> is the foundational PdM technique for rotating equipment (pumps, fans, motors, gearboxes, compressors, turbines), addressing 60-80% of rotating equipment failure modes. This guide details the ISO 10816 \/ ISO 20816 framework, vibration analysis techniques (FFT spectrum, envelope analysis, order tracking), ML algorithms (RUL prediction, anomaly detection, RCA), major vendor landscape 2027 (Augury, Senseye, GE Vernova SmartSignal, AspenTech, ABB Ability, IBM Maximo APM), implementation roadmap, and integration patterns with MES + OEE specialists (TeepTrak Pulse). Stellantis \u20ac4.8M case demonstrates compound value: real-time OEE measurement + targeted predictive maintenance on critical equipment.<\/p>\n

ISO 10816 \/ ISO 20816: vibration severity standards<\/h2>\n

ISO 10816<\/strong> (now being progressively replaced by ISO 20816<\/strong>, published 2017+) is the international standard for mechanical vibration evaluation of machines by measurements on non-rotating parts. Multiple parts cover different machine types:<\/p>\n

    \n
  • ISO 20816-1:2016<\/strong>: General guidelines (replaces ISO 10816-1)<\/li>\n
  • ISO 20816-2:2024<\/strong>: Land-based gas turbines, steam turbines, generators with power > 40 MW<\/li>\n
  • ISO 20816-3:2022<\/strong>: Industrial machines with rated power 15 kW to 50 MW (most common industrial use case)<\/li>\n
  • ISO 20816-4:2018<\/strong>: Gas turbines with fluid-film bearings<\/li>\n
  • ISO 20816-5:2018<\/strong>: Hydraulic power generating and pump-storage plants<\/li>\n
  • ISO 20816-6:2024<\/strong>: Reciprocating machines<\/li>\n
  • ISO 20816-7:2016<\/strong>: Rotodynamic pumps for industrial applications<\/li>\n
  • ISO 20816-8:2018<\/strong>: Reciprocating compressor systems<\/li>\n
  • ISO 20816-9:2020<\/strong>: Gear units<\/li>\n
  • ISO 20816-21:2015<\/strong>: Onshore wind turbine gearbox-driven generators<\/li>\n<\/ul>\n

    Vibration severity zones (ISO 20816-3 for industrial machines 15 kW – 50 MW):<\/p>\n\n\n\n\n\n\n\n\n
    Zone<\/th>\nDescription<\/th>\nAction<\/th>\n<\/tr>\n<\/thead>\n
    Zone A<\/strong><\/td>\nNewly commissioned machines, normal operation<\/td>\nNo action \u2014 monitor<\/td>\n<\/tr>\n
    Zone B<\/strong><\/td>\nAcceptable for long-term operation<\/td>\nNo immediate action \u2014 continue monitoring<\/td>\n<\/tr>\n
    Zone C<\/strong><\/td>\nUnsatisfactory for long-term continuous operation<\/td>\nPlan maintenance intervention within reasonable timeframe<\/td>\n<\/tr>\n
    Zone D<\/strong><\/td>\nSufficient severity to cause damage to machine<\/td>\nImmediate corrective action required \u2014 risk of equipment failure<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

    Specific thresholds in mm\/s RMS velocity vary by machine class (rigid foundation vs flexible foundation) and machine power. Example for medium-size machines (300 kW – 50 MW, rigid foundation): A\/B boundary 2.3 mm\/s, B\/C boundary 4.5 mm\/s, C\/D boundary 7.1 mm\/s.<\/p>\n

    Vibration analysis techniques<\/h2>\n

    FFT (Fast Fourier Transform) spectrum analysis<\/h3>\n

    FFT converts time-domain vibration signal into frequency domain, revealing dominant frequencies that correspond to specific failure modes:<\/p>\n

      \n
    • 1\u00d7 rotation frequency<\/strong>: Unbalance (most common defect, 30-40% of vibration cases)<\/li>\n
    • 2\u00d7 rotation frequency<\/strong>: Misalignment, looseness, cracks in shaft<\/li>\n
    • 3\u00d7 and higher harmonics<\/strong>: Misalignment severity, looseness<\/li>\n
    • 0.5\u00d7 subharmonic<\/strong>: Oil whirl in journal bearings<\/li>\n
    • Bearing defect frequencies<\/strong>: BPFO (Ball Pass Frequency Outer race), BPFI (Inner race), BSF (Ball Spin Frequency), FTF (Fundamental Train Frequency) \u2014 calculated from bearing geometry<\/li>\n
    • Gear mesh frequency<\/strong>: Number of teeth \u00d7 shaft rotation frequency \u2014 indicates tooth wear, eccentricity, manufacturing defects<\/li>\n
    • Blade pass frequency<\/strong> (pumps, fans, turbines): Number of blades \u00d7 rotation frequency \u2014 indicates impeller\/blade wear, flow disturbance<\/li>\n
    • Sidebands<\/strong> around fundamental frequencies: modulation indicating non-linear faults (cracks, severe wear)<\/li>\n<\/ul>\n

      Envelope analysis (bearing diagnostics)<\/h3>\n

      Bearing defects produce small high-frequency impacts that get masked by larger low-frequency vibrations in raw FFT. Envelope analysis (also called demodulation):<\/p>\n

        \n
      1. Band-pass filter raw signal to isolate high-frequency resonance band (typically 5-30 kHz)<\/li>\n
      2. Rectify and low-pass filter to extract amplitude envelope<\/li>\n
      3. Apply FFT to envelope \u2014 reveals bearing defect frequencies (BPFO, BPFI, BSF, FTF) that were hidden in raw spectrum<\/li>\n<\/ol>\n

        Envelope analysis is the gold standard for bearing condition assessment, capable of detecting defects 6-12 months before failure.<\/p>\n

        Order tracking (variable speed equipment)<\/h3>\n

        For variable-speed equipment (compressors, EV motors, wind turbines), traditional FFT with fixed frequency bins gets smeared. Order tracking uses tachometer signal to synchronize sampling with rotation, producing speed-independent spectra in “orders” of rotation. Standard for variable-frequency drives (VFD) applications.<\/p>\n

        Cepstrum analysis<\/h3>\n

        Cepstrum = Inverse FFT of log-amplitude FFT. Useful for detecting periodic events in spectrum (gear mesh sidebands, multiple harmonics), separating excitation from path\/transmission characteristics.<\/p>\n

        Time waveform analysis<\/h3>\n

        Direct time-domain analysis reveals: impulsive events (impacts, looseness, cavitation in pumps), beat patterns (frequency modulation), shaft orbit plots (combined X-Y proximity probe data showing journal bearing condition).<\/p>\n

        Machine learning algorithms for predictive maintenance<\/h2>\n

        RUL (Remaining Useful Life) prediction<\/h3>\n

        RUL prediction estimates how many hours\/days\/cycles remain before equipment failure. Common approaches:<\/p>\n

          \n
        • Survival analysis<\/strong>: Cox proportional hazards model, Weibull regression \u2014 established statistical methods<\/li>\n
        • LSTM\/GRU neural networks<\/strong>: Recurrent neural networks for sequence data (sensor time-series)<\/li>\n
        • Transformer models<\/strong>: Attention-based models (TST – Time Series Transformer) \u2014 emerging 2024-2027<\/li>\n
        • Physics-Informed Neural Networks (PINN)<\/strong>: Combine physics-based equations with ML \u2014 gray-box approach<\/li>\n
        • Particle filtering \/ Kalman filtering<\/strong>: Bayesian state estimation tracking equipment degradation<\/li>\n<\/ul>\n

          Performance metric: RMSE (Root Mean Square Error) on RUL prediction, typically expressed in days. Top-performing systems achieve RMSE 5-15 days on industrial pumps\/motors.<\/p>\n

          Anomaly detection<\/h3>\n

          Anomaly detection identifies abnormal equipment behavior without prior knowledge of specific failure modes:<\/p>\n

            \n
          • Isolation Forest<\/strong>: Tree-based unsupervised learning, fast for high-dimensional data<\/li>\n
          • One-Class SVM<\/strong>: Support vector machine for novelty detection<\/li>\n
          • Autoencoders<\/strong>: Neural network learning normal patterns, reconstruction error indicates anomaly<\/li>\n
          • LSTM Autoencoders<\/strong>: For time-series, capturing temporal dependencies<\/li>\n
          • Variational Autoencoders (VAE)<\/strong>: Probabilistic version, more robust to noise<\/li>\n
          • Statistical methods<\/strong>: Mahalanobis distance, Hotelling’s T\u00b2 for multivariate SPC<\/li>\n<\/ul>\n

            RCA (Root Cause Analysis)<\/h3>\n

            When anomaly detected, RCA identifies which subsystem\/component caused it:<\/p>\n

              \n
            • Feature importance<\/strong> (SHAP, LIME): explain ML predictions<\/li>\n
            • Causal discovery algorithms<\/strong>: PC algorithm, NoTears, FCI for inferring causal graphs<\/li>\n
            • Knowledge graph + ML hybrid<\/strong>: combining domain expertise (failure mode trees) with ML predictions<\/li>\n
            • Generative AI \/ LLM-based<\/strong>: emerging 2024-2027 for natural language explanations to maintenance technicians<\/li>\n<\/ul>\n
              \n

              Download the white paper<\/h3>

              Enter your email address to receive our White Paper<\/p> \n

              \n
              <\/div> \n
              \n