Skip to content
DraconiticResearch
Search Sources API
178,033 studies31 sourcesUpdated 41 min ago
Peer-reviewedCC BY 4.0SoftwareX

Enhancement of the numerical solvers of the pyrolysis code Gpyro

  • Youssouf Abdelhafiz
  • Abdenour Amokrane
  • Gerald Debenest
  • Serge Bourbigot
Open at sourcePDFDOICite

Abstract

Gpyro is a reference code for pyrolysis modeling in fire science. This paper presents a comprehensive revision ofthe Gpyro numerical framework. The revision replaces legacy solvers with more stable and computationallyefficient algorithms, corrects critical numerical and physical-modeling issues, and restructures the codebase toimprove modularity and maintainability. These developments result in substantial performance gains, with thesimulations reported here running between 4 and 200 times faster than in the previous version, depending on theconfiguration and spatial resolution. Additionally, a comprehensive suite of verification and validation cases hasbeen developed, providing a robust foundation for assessing the accuracy and reliability of the simulations.

Related

  1. The Symbol of the Surrogate: Measuring Numerical Provenance in Neural PDE Solvers
    Ridham Patel
    arXivPreprintCC BY 4.0Full textAnalysis & PDEsStatisticsarXiv
  2. Landau–Lifshitz–Gilbert equation written in a suitable form for numerical solution via explicit solvers
    André E. Botha
    Scientific ReportsPeer-reviewedCC BY 4.0Full textNumerical analysis & scientific computingCrossref
  3. Comparative Analysis of Large-Scale SVM Solvers: Convergence, Resource Consumption, and Energy Footprint
    Mamadou Sow, Christophe Cérin
    HALPreprintPublisher termsFull textOptimisation & controlHAL
  4. Impact of Feature Scaling on the Convergence Speed of Linear SVM Solvers : An Empirical Benchmark
    Mamadou Sow, Christophe Cérin
    HALPreprintPublisher termsFull textOptimisation & controlHAL
  5. Comparative Performance Analysis of Optimization Solvers for Linear Support Vector Machines: Impact of Preconditioning and Dimensionality An Empirical Study Using the Benchopt Framework
    Mamadou Sow, Christophe Cérin
    HALPreprintPublisher termsFull textOptimisation & controlHAL

Cite

BibTeXRIS
BibTeX
@article{abdelhafiz2026enhancement,
  title = {{Enhancement of the numerical solvers of the pyrolysis code Gpyro}},
  author = {Youssouf Abdelhafiz and Abdenour Amokrane and Gerald Debenest and Serge Bourbigot},
  journal = {SoftwareX},
  year = {2026},
  month = sep,
  doi = {10.1016/j.softx.2026.103058},
  url = {https://hal.science/hal-05763242v1},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - Enhancement of the numerical solvers of the pyrolysis code Gpyro
AU  - Youssouf Abdelhafiz
AU  - Abdenour Amokrane
AU  - Gerald Debenest
AU  - Serge Bourbigot
PY  - 2026
DA  - 2026/09/22
JO  - SoftwareX
DO  - 10.1016/j.softx.2026.103058
UR  - https://hal.science/hal-05763242v1
AB  - Gpyro is a reference code for pyrolysis modeling in fire science. This paper presents a comprehensive revision ofthe Gpyro numerical framework. The revision replaces legacy solvers with more stable and computationallyefficient algorithms, corrects critical numerical and physical-modeling issues, and restructures the codebase toimprove modularity and maintainability. These developments result in substantial performance gains, with thesimulations reported here running between 4 and 200 times faster than in the previous version, depending on theconfiguration and spatial resolution. Additionally, a comprehensive suite of verification and validation cases hasbeen developed, providing a robust foundation for assessing the accuracy and reliability of the simulations.
SN  - 2352-7110
ER  -