文摘
The problem of finding suitable sensor locations for distributed parameter systems (DPS) is tackled as avariable selection problem. Two existing variable selection methods are used: one is based on principalcomponent analysis (PCA) and the other on the principal variable (PV) method. A new PCA-based variableselection method, called "orthogonal variables in loading space" (OVL) is introduced. The best sensor locationfor DPS is dependent on sensor characteristics and also on the time interval of interest. This is illustrated ina case study where the best point in time to replace a packed bed filter is studied. Sensor positions aredetermined for different time intervals and different types of measurement errors. The resulting sensor positionscharacterize the overall time behavior of the DPS in the selected time interval. As a test, the specific problemof predicting the exit concentration of the packed bed filter is considered. Lagged PLS models are built anda full search is done to determine the best possible sensor locations. These "benchmark" sensor positions arecompared to the sensor locations found by the variable selection methods. The OVL method and the PVmethod both perform well, but the OVL method is additionally computationally less demanding.