论文部分内容阅读
为实现对电厂选择性催化还原(SCR)脱硝装置喷氨的优化控制,以广东某电厂350 MW锅炉为研究对象,采用径向基函数(RBF)神经网络法,以锅炉负荷、烟气体积流量、SCR烟气温度、脱硝进口NO_x 质量浓度以及喷氨质量流量等为输入变量,以SCR脱硝效率为输出变量,建立输入变量与输出变量之间的关系模型,实现对SCR脱硝效率及脱硝出口NO_x 质量浓度的预测.在满足NO_x 排放标准的前提下,以SCR系统运行成本最小为目标,利用Matlab对该模型进行仿真实验,寻求氨耗成本和电耗成本与NO_x 排放费用的临界点,得到最佳喷氨质量流量.结果表明:最佳喷氨质量流量计算值比实测值或高或低,但在满足NO_x 排放标准的前提下,其SCR系统运行成本呈降低趋势.
In order to realize the optimal control of ammonia injection in selective catalytic reduction (SCR) denitrification plant, a 350 MW boiler in a power plant in Guangdong was used as research object. Radial Basis Function (RBF) neural network method was used in this paper. The boiler load, flue gas volume flow , SCR flue gas temperature, NO_x concentration of denitrification inlet and mass flow rate of ammonia injection are taken as input variables, and SCR denitrification efficiency is used as output variable to establish the relationship model between input variables and output variables, so as to realize SCR denitration efficiency and NO_x Mass concentration.Under the premise of meeting the NO_x emission standard, taking the minimum operating cost of the SCR system as a target, this model is simulated by Matlab to find the critical point of the ammonia consumption cost, the electricity consumption cost and the NO_x emission cost, The results show that the optimum ammonia injection mass flow rate is higher or lower than the measured value, but the operating cost of SCR system decreases when the NO_x emission standard is satisfied.