Abstract
To reduce casualties and property damage, timely and accurate fire warning are extreme essential when fire breaks out in the urban underground pipe gallery. This paper aims to propose a multi-GA (Genetic algorithm)-BPNN (Back propagation neural network) fusion prediction algorithm to achieve rapid detection during the initial stage of tunnel fires. In the developed fusion prediction algorithm, genetic-algorithm is introduced to meliorate the traditional back propagation neural network, and a strong multi-neural network is composed by several optimized BP neural networks to accomplish the enhanced fusion forecast. Finally, a full-scale cable fire and a gasoline pool fire test were conducted to validate the forecast capability and effectivity of the algorithm. Thus, two strategies are utilized to improve the detection accuracy of the algorithm, which include the data de-noising method and one criterion procedure coupled in the algorithm. The results indicate that the algorithm proposed in this study can effectively detect the fire during the initial stage by learning the temperature sensor data, and the improved algorithm based on the two strategies can ensure 100 % accuracy for the tunnel fire detection, when the temperature rise is greater than 1 °C. In comparison to the previous methods, the developed algorithm can increase the detection accuracy of tunnel fire, which is of great potential in practical implementation of fire early warning in the underground utility tunnel.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.