FULLY AUTOMATIC PEAK DETECTION AND INTEGRATION FOR DSC CURVES. SMOOTHS THE CURVE, DETECTS SIGNIFICANT PEAKS, IDENTIFIES INTEGRATION LIMITS, AND COMPUTES AREA UNDER EACH PEAK. ALSO EMITS A DEGREE-OF-CURE CURVE PER PEAK (NORMALIZED CUMULATIVE AREA, 0 TO 1).¶
Performs fully automatic peak detection and integration on a DSC (Differential Scanning Calorimetry) curve. Smooths the input curve, detects significant peaks, determines integration limits, and computes the area under each peak. Also emits a degree-of-cure curve per peak as a normalized cumulative area (0–1) from left to right integration limit.
When to use¶
Classification: process.
Tagged: adhesive, area_under_curve, calorimetry, curing, degree_of_cure, dsc, heat_flow, peak_detection.
Inputs¶
| Label | ID | Type | Default | Required | Description |
|---|---|---|---|---|---|
| DSC Curve | dsc_curve | vector | — | Input DSC curve where X is temperature (°C or K) and Y is heat flow (mW or W/g); peaks are detected automatically — provide a single preprocessed DSC scan. | |
| Area Calculation Method | area_method | select | baseline | How each peak’s area is integrated. ‘Baseline using Limits’ draws a straight baseline from the left to the right integration limit and integrates (curve - baseline), matching ASTM E2070 DSC convention; this is the chemically meaningful peak energy. ‘Y=0’ integrates against the X axis; only meaningful for curves that stay on one side of zero, since positive and negative regions cancel against y=0. Default is baseline. | |
| Grouping Percentage | grouping_percentage | string | 5 | Percentage of the total X (temperature) range within which nearby peaks are merged into a single result; default is 5% — set to 0 to disable grouping or 100 to collapse all peaks into one group. | |
| Prominence Threshold (%) | prominence_threshold | string | 20 | Peaks with prominence below this percentage of the maximum prominence are filtered out. Default 20%. Lower values detect more (possibly noisy) peaks; higher values keep only the most prominent peaks. |
Outputs¶
| Label | ID | Type | Description |
|---|---|---|---|
| DSC Peak Analysis Results | curve_dsc_peak_analysis_output_1 | dataset | Tabular dataset with one row per detected peak and columns: peak_temperature, peak_heat_flow, left_limit, right_limit, area (signed integral; if area_method is ‘baseline’ this is the integral against a straight line drawn between the integration limits, i.e. the chemically meaningful peak energy per ASTM E2070; if ‘y_zero’ it is the integral against the X axis, which can cancel positive and negative regions against each other), absolute_area (absolute value of area), and baseline_clean (transparency flag for the integration window: true when every point sits at or above the baseline line within a 1% tolerance of the peak amplitude, so the baseline-integral interpretation can be trusted directly; false when the curve dips below the line by more than that, typically because the automatic integration limits landed slightly past the true valley or because the data is noisy near the endpoints, in which case the peak should be surfaced for human review; the math runs to completion in both cases and the degree_of_cure curve is clamped to stay monotonic, so a false value is informational rather than an error). |
| Degree of Cure Curves | curve_dsc_peak_analysis_output_2 | vector | Collection of degree-of-cure curves (one per detected peak), where X is temperature and Y is the normalized cumulative baseline-subtracted area from 0 at the left integration limit to 1 at the right integration limit. The cumulative integrand is clamped at zero so the curve is guaranteed monotonic non-decreasing even when the underlying curve briefly dips below the baseline line; suitable for adhesive and curing-chemistry workflows. |
Disciplines¶
- data.curve.transform
- data.signal_processing
- data.statistics
- engineering.material.characterization
Runnable example¶
A runnable example is registered for this worker. Open the example workflow on the d3VIEW canvas: /api/workflow/example?id=curve_dsc_peak_analysis
Auto-generated from transformation schema. Worker id: curve_dsc_peak_analysis. Schema hash: e10ebca0f3fd. Hand-curated docs in workerexamples/ override this page when present.