Skip to content

SignalComparisonEvents¤

Compare two related signals (e.g. setpoint vs actual) and detect divergence.

Module: ts_shape.events.engineering.signal_comparison
Guide: Process Engineering

When to Use¤

Use to compare setpoint vs actual, sensor A vs sensor B, or any paired signals. Detects when signals that should track each other start to diverge. Valuable for sensor validation, control loop monitoring, and redundancy checks.

Quick Example¤

from ts_shape.events.engineering.signal_comparison import SignalComparisonEvents

comparator = SignalComparisonEvents(
    df=process_data,
    uuid="setpoint_pressure_01"
)

# Detect when actual diverges from setpoint beyond tolerance
divergence = comparator.detect_divergence(
    actual_uuid="actual_pressure_01",
    tolerance=2.0,
    min_duration="1m"
)

# Per-hour deviation statistics
dev_stats = comparator.deviation_statistics(
    actual_uuid="actual_pressure_01",
    window="1h"
)

# Rolling correlation to detect decoupling
correlation = comparator.correlation_windows(
    actual_uuid="actual_pressure_01",
    window="1h"
)

Key Methods¤

Method Purpose Returns
detect_divergence(actual_uuid, tolerance, min_duration='1m') Divergence intervals DataFrame of divergence events
deviation_statistics(actual_uuid, window='1h') Per-window deviation stats DataFrame with deviation metrics
tracking_error_trend(actual_uuid, window='1D') Deviation trend over time DataFrame with daily trend
correlation_windows(actual_uuid, window='1h') Per-window correlation DataFrame with correlation values

Tips & Notes¤

Use correlation to catch slow drift

detect_divergence() catches absolute deviations, but two signals can drift apart slowly while staying within tolerance. Monitor correlation_windows() over days — a declining correlation trend signals degrading sensor or actuator performance.

Related modules

See Also¤