预测TiO_2光催化烟气同时脱硫脱硝效率的遗传程序设计方法研究

STUDY ON THE GENETIC PROGRAMMING METHOD FOR PREDICTING EFFICIENCY OF SIMUTANEOUS FLUE GAS DESULPHURIZATION AND DENITRIFICATION BY USING TiO_2 PHOTO-CATALYSIS

影响TiO2光催化烟气同时脱硫脱硝效率的因素包括温度、催化剂、含湿量、氧浓度、光、SO2和NOx浓度等,它们呈复杂的非线性关系。在紫外光下,利用催化剂及混合烟气中氧气浓度为8%的条件下,以TiO2光催化烟气同时脱硫脱硝的试验数据为样本,采用遗传程序设计(GP)方法自动找出脱硫脱硝效率随各主要影响因素变化的规律,并通过检验样本数据对模型进行了预测。结果表明,该方法避免了事先确定变量之间函数关系的主观性,预测数据的准确度较高。

The factors affecting efficiency of simutaneous flue gas desulphurization and denitrification by using TiO2 photo - catalysis, including temperature, catalyst, humidity, oxygen concentration etc. , have complex and non - linear relations among them. Under condition of ultraviolet ray and oxygen concentration in the catalyst and flue gas mixture being 8%, taking the data of simutaneous flue gas desulphurization and denitrification as the sample,and using the genetic programming method,the variation regularity of desulphurization and denitrification efficiency along with the main affecting factors has been automatically found out, and the model being predicted through inspection of the sample data. Results show that the said method can avoid the subjectiveness in functional relationship among the variables determined in advance, having higher precesion of the predicted data.