SHMPC

Model Predictive Control of Batch Processes (SHMPC). 279... 

Figure 9.2 illustrates the inner workings of SHMPC during a given phase In this figure, let Xq be the initial state, d, be a vector of measured disturbances available after phase i, y, be a... 

Figure 9.2 Shrinking horizon model predictive control (SHMPC)...

In SHMPC, newly available values of primary and secondary measurements are combined with previously obtained values, as well as with manipulated input settings... 

For use in a strategy such as SHMPC, a model must also be relatively small because it must provide real-time output in an on-line application. In a batch process, such as autoclave curing, model output must be available before the process moves into its next phase (i.e., before the next measurement is recorded). This requirement for real-time models in an on-line application limits the types of models that can serve in SHMPC. 

Artificial neural networks arose from efforts to model the functioning of the mammalian brain. The most popular ANN — the feedforward ANN — has deeper roots in statistics than in neurobiology, though. A form of ANN (a Probability Neural Network) has been used within a QPA context to improve sensor data reliability, but not as an on-line quality model [57]. The best way to represent a feedforward ANN as an on-line quality model for SHMPC is... 

---