Skip to main content
U.S. flag

An official website of the United States government

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

Sensors and Machine Learning Models to Prevent Cooktop Ignition and Ignore Normal Cooking



Amy Mensch, Anthony Hamins, Wai Cheong Tam, John Lu, Kathryn Markell, Christina You, Matthew Kupferschmid


According to a recent NFPA report, 49 % of reported home fires involve cooking equipment, with cooktops accounting for 87 % of cooking-fire deaths and 80 % of the civilian injuries [1, 2]. Between 2014–2018, U.S. fire departments responded to an estimated 172,900 home cooking fires per year, leading to an average of 550 civilian deaths [2]. Electric-coil cooktops manufactured after June 2018 in the U.S. must pass the abnormal cooking test in UL 858 [3]. The test prescribes a maximum temperature of the dry-pan or a performance test for ignition-prevention using 50 mL of canola oil with the coil element on its highest power setting. This standard does not apply to older cooktops or gas cooktops. Therefore, we consider the feasibility of using a variety of sensors as the basis for a retrofit device that would provide early warning or control to automatically shut off the cooktop during unattended cooking, while ignoring normal-cooking activities. Several studies [4–9] have investigated the performance advantages of multiple sensors over a single sensor for detecting generalized fire conditions, preventing cooktop ignition, or resisting nuisance sources. However, a comprehensive study to compare the effectiveness of a wide range of sensors and sensor combinations has not previously been conducted. This article is a compressed version of a recently published study [10] with an objective to apply data-driven, statistical methods and machine learning methods to design a detection algorithm for cooktop ignition prevention. Data was obtained from experiments with a variety of ignition and normal-cooking scenarios to develop and evaluate prediction algorithms using threshold analysis and neural network models.
SFPE Europe


Cooktop ignition, Sensor analysis, Neural networks, ignition prevention


Mensch, A. , Hamins, A. , Tam, W. , Lu, J. , Markell, K. , You, C. and Kupferschmid, M. (2021), Sensors and Machine Learning Models to Prevent Cooktop Ignition and Ignore Normal Cooking, SFPE Europe, [online], (Accessed June 21, 2024)


If you have any questions about this publication or are having problems accessing it, please contact

Created July 28, 2021, Updated November 29, 2022