Study On The Theory Of Clean Aerosol Extinguishing Agent Design

Hot aerosol fire extinguishing agent has been widely studied as a kind of efficient and clean fire extinguishing agent. But as a kind of pyrotechnic composition, it would release amounts of extinguishing media form burning, and bring some deficiencies such as high combustion temperature, much residue and slagging. To overcome the deficiencies of the aerosol fire extinguishing agent, this paper has redesigned the formula and the properties of new agent were also studied.First of all, the paper selected sucrose as main combustible in the basic formula of extinguishing agent. By studying the performances of different composition with different oxygen balance, the results showed that the formula has high extinguishing efficiency, less residues and environmentally friendly as well as stable comprehensive performance when the oxygen value was-12g.After that, the back-propagation neural network and genetic algorithm (BP-GA) model was introduced to predict the performance of aerosol fire extinguishing agent, and compared with pure BP model. The results showed that the BP-GA model had a better prediction. The model provides a rapid optimization method.Finally, according to the combustion particles distribution of the flame structure and the distribution of aerosol particles in the fire ground space, the results showed that the flame of agent was constituted by dense micro-scale combustion particles. And through the particle distribution model, due to the aerosol particles size were very small, aerosol particles were involved into flame region by momentum and entrainment primarily to adsorb the free radicals in the fire.