Real-time RSET Prediction Based on Simulation Dataset Using Machine Learning: A Complex Geometry Case Study

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26110%2F24%3APU155388" target="_blank" >RIV/00216305:26110/24:PU155388 - isvavai.cz</a>

Výsledek na webu

<a href="https://files.thunderheadeng.com/femtc/2024_pdf-archive.zip" target="_blank" >https://files.thunderheadeng.com/femtc/2024_pdf-archive.zip</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Real-time RSET Prediction Based on Simulation Dataset Using Machine Learning: A Complex Geometry Case Study

Popis výsledku v původním jazyce

Agent-based evacuation model simulations are not suitable for real-time estimates due to their complexity and computational demands. Machine learning models allow for the approximation of simulations through estimates, creating a metamodel whose outputs can be used in real-time for effective decision-making in object safety management. The article presents a case study demonstrating the process of training the metamodel on a dataset with seven input features and simulations of evacuation model generated by a quasi-random sequence. Among the compared machine learning regression models, the ANN metamodel achieved the best results.

Název v anglickém jazyce

Real-time RSET Prediction Based on Simulation Dataset Using Machine Learning: A Complex Geometry Case Study

Popis výsledku anglicky

Agent-based evacuation model simulations are not suitable for real-time estimates due to their complexity and computational demands. Machine learning models allow for the approximation of simulations through estimates, creating a metamodel whose outputs can be used in real-time for effective decision-making in object safety management. The article presents a case study demonstrating the process of training the metamodel on a dataset with seven input features and simulations of evacuation model generated by a quasi-random sequence. Among the compared machine learning regression models, the ANN metamodel achieved the best results.

Druh

O - Ostatní výsledky

CEP obor

—

OECD FORD obor

20205 - Automation and control systems

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů