基于MIC和MPA-KELM的脱硫出口SO2浓度预测

建立脱硫出口SO₂浓度预测模型是实现脱硫系统经济运行的基础。针对这一问题，提出了基于最大信息系数(MIC)的变量选择方法和基于海洋捕食算法(MPA)优化核极限学习机(KELM)的脱硫出口SO₂浓度预测模型。首先，采用机理分析法筛选影响出口SO₂浓度的变量，提出循环浆液综合流量表达方法，便于描述浆液循环泵组合的影响特性；在此基础上，通过基于最大信息系数的变量选择算法确定模型输入变量；运用MPA对KELM的正则系数C和核参数S进行寻优，建立MPA-KELM的出口SO₂浓度预测模型；最后，利用电厂运行数据进行仿真实验。实验结果表明，所建立出口SO₂浓度预测模型的均方误差、平均绝对百分比误差分别为1.236 66 mg/m³和4.987 6%,预测精度高，能够为脱硫系统出口SO₂的现场优化控制提供技术支持。

Prediction of SO2 Concentration at Desulfurization Outlet Based on MIC and MPA-KELM

The establishment of SO₂ concentration prediction model at desulfurization outlet is the basis for realizing the economic operation of desulfurization system. Aiming at this problem, a SO2 concentration prediction model at desulfurization outlet based on variable selection of maximum information coefficient (MIC) and marine predation algorithm (MPA) optimized nuclear limit learning machine (KELM) was proposed. Firstly, the mechanism analysis method was used to screen the variables affecting the SO₂ concentration at the outlet, and the expression method of comprehensive flow of circulating slurry was proposed to describe the influence characteristics of slurry circulating pump combination. On this basis the input variables of the model were determined by the variable selection algorithm based on the maximum information coefficient. Then, MPA was used to optimize the regularity coefficient C and nuclear parameter S of KELM, and the outlet SO₂ concentration prediction model of MPA-KELM was established. Finally, the simulation experiment was carried out by using the operation data of the power plant. The results show that after variable selection, the mean square error and average absolute percentage error of MPA-KELM model are 1.23666mg/m³ and 4.9876% respectively. The prediction accuracy is high, which can provide technical support for the on-site optimal control of SO₂ in the desulfurization system.