Research on Fire Alarm System Based on Entropy Power Method and Topsis Superiority and Inferiority Solution Distance Method

Abstract

According to the statistics of China Fire and Rescue Bureau, in 2021, 748,000 fires were reported to the national fire and rescue teams, with 1,987 deaths, 2,225 injuries and direct property losses of 6.75 billion yuan. The loss of these innocent lives and the loss of a large amount of property make us think while deploring: is there a more effective, scientific and accurate way to prevent the great harm caused by the occurrence of fire. The fire alarm system studied in this paper can be used to reduce the hazards. Firstly, the number of real fires is given 430 times by using three indicators of fire occurrence: address, machine number and loop. After selecting the high-frequency parts in the loop, machine number and component type respectively and finding their corresponding false alarm rate and failure rate, the evaluation weights are obtained using the entropy weight method. A comprehensive evaluation model based on the entropy weight method is established to score each major component. Two most reliable components of the fire alarm system are obtained: the intelligent photoelectric probe and the manual alarm button. Next, the normalized management level of each fire brigade needs to be analyzed. First, the five important parameters of normalized average coverage rate, normalized component reliability, normalized component failure rate, normalized jurisdictional fire frequency (times/hour), and normalized false alarm rate are derived for each fire brigade. The comprehensive score index of each brigade was derived and ranked by the Topsis superiority and inferiority solution distance method, and it was concluded that M brigade, L brigade, and F brigade were the three jurisdictions with the lowest comprehensive management level.