PULsE allows effective treatment of raw data generated in laser flash experiments where conditions may not be ideal for simpler analysis.
PULsE analyses the heating curves, calculates and outputs the thermal properties of the sample, such as the thermal diffusivity, based on the inverse solution of a heat transfer problem. The software is specifically tailored for use
with Netzsch and Linseis LFA systems and allows loading multiple data formats. It is therefore inteded for use with any LFA instruments as an alternative to the standard, often simplistic software.
This software has been written in Java 11 and requires the corresponding virtual machine to run. A javadoc is available with information
on what Java classes and methods it contains, including specification of data formats.
Main features of PULsE:
Calculates the time-temperature profiles for the laser flash experiments by solving heat transfer problems with high-accuracy finite difference schemes and numerical solvers
Solves the reverse heat conduction problems by using advanced numerical optimisation techniques
Ability to fully customise the optimisation procedure by adjusting the number of search variables and fitting domain
May re-construct missing data, e.g. due to detector failure or electronic circuit overheating
Capable of correcting systematic errors e.g. due to incorrect pulse timing
Uses statistical helpers to facilitate and enhance data analysis
Provides an easy to use, light-weight graphical user interface based on Swing and JFreeChart
Automatically estimates the available computing power and tries to use multi-threading for batch processing of tasks
Plots interactive heating curves for the user to adjust the calculation domain and to check the quality of fit
Manages results with dynamically-adjustable format; capable of auto-merging resuts and calculating standard errors on-the-fly
Comprehensive logging that can show both intermediate execution steps and time events
Ability to export data in html or csv format
Problem statements supported in this version:
Classical One-Dimensional
Classical Two-Dimensional
1D - Nonlinear heat losses
Distributed Laser Absorption
Two-Temperature Model
Diathermic Model
Participating Medium (fully coupled radiative-conductive heat transfer)
Advanced statistical toolkit includes:
Normality tests (Kolmogorov, Anderson-Darling)
Correlation tests (Spearman, Pearson)
Model selection via AIC and BIC statistics
Outlier-robust optimisation via Least Absolute Deviations (LAD)
PULsE is distributed under the Apache 2.0 license.