Model-plant mismatch detection using model parameter data clustering for paper machines or other systems

An apparatus for controlling an operation of a paper machine in a web manufacturing system, the apparatus comprising:

• at least one memory configured to store data associated with a process;
• and
• at least one processing device configured to:
• obtain routine operating data associated with a model-based controller;
• repeatedly identify one or more values for one or more model parameters of at least one model associated with the process using the data associated with the process at multiple periods of time, the one or more values for the one or more model parameters identified using data associated with the process by estimating the one or more model parameters at each of the multiple periods of time;
• cluster the values of the one or more model parameters into one or more clusters using a support vector machine, the support vector machine using a kernel function to generate a clustering of data by mapping the data into a space where a cluster boundary is identified;
• identify one or more additional values for the one or more model parameters using additional data which is different than the data associated with the process during one or more additional periods of time;
• and
• detect using a tool, a mismatch between the at least one model and the process in response to determining that at least some of the one or more additional values fall outside of the one or more clusters, the tool analyzing the routine operating data for the model-based controller using the support vector machine for detecting the mismatch;
• and
• taking corrective action in response to detecting the mismatch.

A non-transitory computer readable medium for controlling an operation of a paper machine in a web manufacturing system, containing instructions that, when executed by at least one processing device, cause the at least one processing device to:

• obtain routine operating data associated with a model-based controller;
• repeatedly identify one or more values for one or more model parameters of at least one model associated with a process at multiple periods of time, the one or more values for the one or more model parameters identified using data associated with the process by estimating the one or more model parameters at each of the multiple periods of time;
• cluster the values of the one or more model parameters into one or more clusters using a support vector machine, the support vector machine using a kernel function to generate a clustering of data by mapping the data into a space where a cluster boundary is identified;
• identify one or more additional values for the one or more model parameters using additional data which is different than the data associated with the process during one or more additional periods of time;
• and
• detect using a tool, a mismatch between the at least one model and the process in response to determining that at least some of the one or more additional values fall outside of the one or more clusters, the tool analyzing the routine operating data for the model-based controller using the support vector machine for detecting the mismatch;
• and
• take corrective action in response to detecting the mismatch.