WBRPLSR方法及其在化工软测量中的应用

及时测定化工过程变量，对确保生产过程稳定、有效控制产品质量具有重要意义.基于实时样本数据，采用偏最小二乘方法，以分块递归的方式，为过程变量建立软测量模型.在分析时序数据特性的基础上，引入加权策略，并提出选定相关参数的方法步骤，推导构建了加权分块递归偏最小二乘回归方法（WBRPLSR）.将该法实际应用于某公司PTA装置溶剂脱水塔，为塔釡排出液H₂O含量建立软测量模型，效果良好.与已有方法相比，它提高了建模效率，改进了预测性能.

WBRPLSR method and its application to dynamic chemical process modeling

It is well known that to measure and estimate the ch em ical process variables in time has vital significance in ensuring process stabil ization and effectively controlling its product quality.In this study, a soft se nsor model of a chemical process was established by partial least squares method based on its time series data, and the model could be adjusted in the block-wi se recursive way in the presence of new sample data. With a view to the time ser ies data characteristics,a strategy of allotting different weight coefficients t o the time series data was introduced, and an approach of how to ascertain the w eight coefficients was provided in the meantime.Subsequently, the weighted block -wise recursive partial least squares regression (WBRPLSR) algorithm was develo ped and used to model the water content of solvent dehydration tower bottom drai nage in a commercial purified terephthalic acid (PTA)unit.The experimental resul t showed that the algorithm was rapid and effective.Compared with some other met hods, the WBRPLSR method increased modeling efficiency and improved prediction p erformance.