摘要
针对德士古气化炉炉膛温度难以测量这一情况,提出利用软测量技术来解决这一问题。通过建立BP网络模型和RBF网络模型以及基于PCA和CHAOS的神经网络模型,并对其仿真结果进行分析和比较,验证了该方法的可行性。CHAOS-RBF软测量模型在化肥厂的应用效果良好,误差保持在1.5%以内,不但提高了温度测量精度,而且有利于更好的生产控制。
Aiming at the difficulty in measuring the temperature of Texaco gasification furnace, soft-sensing method is applied to solve this question. The BP network model, RBF network model and ANN model based on PCA and CHAOS are established. The simulation results are analyzed and compared, practicability of this method is proved. CHAOS-RBF soft-sensing model is applied in Lunan fertilizer plant;temperature error is maintained below 1.5 percent.
出处
《化工自动化及仪表》
EI
CAS
2006年第5期48-50,54,共4页
Control and Instruments in Chemical Industry
