MIMO soft sensors for estimating product quality with on-line correction

The main difficulties of on-line quality control are the availability of on -line product quality measurements. Soft-sensing techniques supply attracti ve and efficient methods to deal with these difficulties. Soft sensors refe r to the modelling approaches to estimating hard-to-measure process variabl es (e.g. quality variables) from other easy-to-measure variables (e.g. temp erature, pressure and flowrate measurements). At present, much more researc h is concerned with multi-input single-output (MISO) systems than with MIMO systems in the field of soft-sensing modelling. In this paper, some MIMO s oft-sensing techniques are studied for estimating multiple product quality variables simultaneously in a hydrocracking fractionator. RBF and fuzzy ART MAP networks are used to build the models and the latter is shown to be mor e suitable for MIMO soft-sensing modelling. The issues of data pretreatment and on-line correction, which are very important for the industrial implem entation of MIMO soft sensors, are discussed in detail. A useful method usi ng a multivariable fuzzy PID (MFPID) on-line correction algorithm is propos ed for the MIMO soft sensors enabling them to adapt with the fluctuation of process operating conditions and uncertain system disturbances. The real a pplication results show that the proposed methods are effective for MIMO so ft-sensing modelling and have great promise in industrial process applicati ons.