Research On Residence Indoor Air Quality Evaluation Based On The GA-BP Algorithm

It is a great significance to study on the method of evaluate and improvement for indoor air quality, which is also the foundation of research and practice. There are many methods on indoor air quality assessment at present. Most of them need to construct subject functions to confirm the power values of evaluation index. For not only the designs of those subject functions contain certain artificial factors, much important inforation will be lost in some circumstance. the sureness of every evaluation index right value will bring some human factors too.The research combines the advantage of fuzzy evaluation method and neural network evaluation method , set up the indoor air quality evaluation method based on fuzzy neural network, this model not only clears at physics meaning, but also can arrange in an order according to the result of evaluation.The author combines the advantage of neural network and genetic algorithm and sets up the indoor air quality based on GA-BP models, this model has good approaching the result and very strong learning ability, can reflect the quality state of indoor air accurately, get a quantitative index, avoid intervention that is come from human. And this model has very good portability and good adaptability, the result is accurate and intuitionistic.Using a case study, the research apply fuzzy neural network and genetic neural network two models to evaluate indoor air quality,, have proved the validity and adaptability of the model; have analyse the relation between every influence factor of indoor air quality, has put forward the improvement indoor air quality measure. The pre-warning mechanism of indoor air quality based on GA-BP is studied. The research has probed into the improvement and prewarning tactics and method of indoor air quality. This pre-warning mechanism has greater meanings for the basic research on auto-controle the flux of fresh air in a air conditioner room.