基于热扩散核密度确定密度峰值法的历史工况识别
毕荣山,韩智慧,陶少辉,孙晓岩,项曙光

Recognizing historical operating conditions by determining the density peaks at kernel density estimation of heat diffusion
Rongshan BI,Zhihui HAN,Shaohui TAO,Xiaoyan SUN,Shuguang XIANG
表4 TE多模态过程变量的识别结果
Table 4 Recognition results of TE multi-modal process variables
变量本文方法K-均值法(K = 4)
平均相对偏差最大相对偏差最小相对偏差平均相对偏差最大相对偏差最小相对偏差
D物料流量-0.081-0.4104-0.0225-0.0814-0.412-0.0225
回收流量0.09120.27860.00620.09090.27780.0062
放空率-1.3681-4.748-0.9476-1.4371-4.9847-0.9567
反应器进料量0.06450.15050.03570.06440.15030.0357
产品分离器压力-0.0043-0.0148-0.0014-0.0043-0.0148-0.0014
汽提塔温度0.01890.23430.06130.01860.23370.0613
压缩机工作功率0.10750.32170.04410.10710.32070.0441
反应器组分B流量-0.1046-0.4776-0.1074-0.1054-0.4799-0.1075
放空气体中G组分流量-0.0142-0.93560.1306-0.0173-0.94440.1305
产品中组分H流量-0.04240.46180.1825-0.04380.45970.1821