文献类型：专利

英文题名：Soft measurement of ammonia net value in ammonia synthesis using uniaxial three-radial chill style ammonia synthesis tower, by obtaining soft measuring values of ammonia content using back-propagation neural network

作者：XU W;GU X;CAO C;ZHANG L;LV Y;YE S;WANG D;GAO X;SUN S;RUI S;SUN Y;LIU Z;CAO Z;DENG G

机构：[1]UNIV EAST CHINA SCI&TECHNOLOGY;[2]UNIV EAST CHINA SCI & TECHNOLOGY;[3]YANKUANG LUNAN CHEM FERTILISER PLANT

申请号：CN101930008-A

申请日：2010-08-11

公开日：2010-12-29

语种：英文

收录：DERWENT

摘要：NOVELTY - Soft measurement of ammonia net value in ammonia synthesis using uniaxial three-radial chill style ammonia synthesis tower includes obtaining soft measuring value of ammonia content in the mixed gases entering the tower of nitrogen and hydrogen after normalizing the output variable of back-propagation (BP) neural network; obtaining soft measuring value of ammonia content in the ammonia synthesis tower two-out gases after normalizing the output variable of BP neural network; and calculating the differential value of the obtained soft measuring values. USE - Soft measurement of ammonia net value in ammonia synthesis using uniaxial three-radial chill style ammonia synthesis tower (claimed). ADVANTAGE - The method provides favorable evidences for product quality of ammonia production industry. Because of using the neural network models in measurement, the method greatly improves the measuring precision of ammonia content at the entrance and exit of the ammonia synthesis tower and fault tolerance performance, and overcomes the influences to the production procedure and the product quality resulting from the measuring lag of the ammonia content at the entrance and exit of the ammonia synthesis tower. DETAILED DESCRIPTION - Soft measurement of ammonia net value in ammonia synthesis using uniaxial three-radial chill style ammonia synthesis tower comprises: (A) obtaining hydrogen content, pressure, and temperature of mixed gases entering the tower of nitrogen and hydrogen; obtaining on-line measuring value of methane (CH4) content in air in ammonia synthesis; normalizing them as input variable of BP neural network of ammonia content in the mixed gases entering the tower of nitrogen and hydrogen; and obtaining soft measuring value of ammonia content (NH3in) in the mixed gases entering the tower of nitrogen and hydrogen after normalizing the output variable of BP neural network; (B) using the flow rate (F) of hydrogen (H2), phosphorus (P), CH4, and mixed gases entering the tower of nitrogen and hydrogen, the on-line measuring values of flow rate (F1) of first chill gas and flow rate (F2) of second chill gas, and the soft measuring value of NH3in obtained in step (A); normalizing them as input variable of BP neural network of ammonia content (NH3out) in the ammonia synthesis tower two-out gases; and obtaining soft measuring value of NH3out of the mixed gases entering the tower of nitrogen and hydrogen after normalizing the output variable of BP neural network; and (C) calculating the differential value of the soft measuring value of NH3out and the soft measuring value of NH3in to obtain ammonia net value.