Synethetic Time Series Generator For Anamoly Detection

• Author(s):

  Ms.Archana | Parv Jain

• Keywords:

  Synthetic Time Series, Anomaly Detection, Anomaly Injection, Time Series Generator, Machine Learning, Deep Learning, Trend Modeling, Seasonality, Noise Simulation, Python.

• Abstract:

  Industries Increasingly Depend On Continuous Monitoring Systems Where Fast And Accurate Anomaly Detection Is Crucial For Preventing Failures And Ensuring Operational Reliability. This Project Proposes A Lightweight Python-based Synthetic Time Series Generator Capable Of Producing Realistic Data With Trends, Seasonality, Noise, And Multiple Anomaly Types. The System Supports Controlled Anomaly Injection And Automatic Labeling, Making It Suitable For Training And Evaluating Anomaly Detection Models. Experimental Results Show Strong Performance, Achieving ROC-AUC Values Between 0.85–0.95 And High Precision, Recall, And F1-scores.

---

Other Details

• Paper id:

  IJSARTV11I11104304

• Published in:

  Volume: 11 Issue: 11 November 2025

• Publication Date:

  2025-11-18

Download Article