Abstract: A method of generating a hybrid model predictive control (MPC) simulation model for a plant configured to run a process that processes at least one raw material to generate at least one tangible product. A predictive dynamic MPC sub-model is provided for each of plurality of process units in the plant, the plant including at least one process controller coupled to field devices coupled to the plurality of process units, where the process units comprise equipment for converting the raw material or an intermediate material formed from the raw material into to another material. A piping network diagram is obtained that provides a representation of a piping network for routing of the raw material and the intermediate material during the process. The dynamic MPC sub-models are coupled together using the piping network to generate the hybrid MPC simulation model which models the plant as a whole.

Abstract: A trained machine learning algorithm processes time series production data. The time series production data are representative of a control process within a facility control loop. The machine learning training algorithm is trained using positive training data that are representative of a normal operation of components within the facility control loop and negative training data that are representative of an abnormal operation of components within the facility control loop. Output of the trained machine learning algorithm identifies abnormalities in the facility control loop.

Abstract: A method of generating a hybrid model predictive control (MPC) simulation model for a plant configured to run a process that processes at least one raw material to generate at least one tangible product. A predictive dynamic MPC sub-model is provided for each of plurality of process units in the plant, the plant including at least one process controller coupled to field devices coupled to the plurality of process units, where the process units comprise equipment for converting the raw material or an intermediate material formed from the raw material into to another material. A piping network diagram is obtained that provides a representation of a piping network for routing of the raw material and the intermediate material during the process. The dynamic MPC sub-models are coupled together using the piping network to generate the hybrid MPC simulation model which models the plant as a whole.

Abstract: A system and method for performing management and diagnostic functions in a cloud computing system for advanced process control (APC). A cloud based APC management computer retrieves operating process data from an APC control computer and performs an iterative step test on the APC system. The iterative step test modifies at least one test parameter of the operating process data and identifies changes to a set of remaining parameters of the operating process data resulting from modification of the test parameter. The APC management computer determines at least one process variable from the iterative step test and generates at least one process model based on the process variable. The APC management computer transmits the process model to the APC control computer.

Abstract: A system and method for performing management and diagnostic functions in an advanced process control (APC) system. An APC management computer retrieves operating process data from an APC control computer and performs an iterative step test on the APC system. The iterative step test modifies at least one test parameter of the operating process data and identifies changes to a set of remaining parameters of the operating process data resulting from modification of the test parameter. The APC management computer determines at least one process variable from the iterative step test and generates at least one process model based on the process variable. The APC management computer transmits the process model to the APC control computer.

Abstract: A system and method for performing management and diagnostic functions in a cloud computing system for advanced process control (APC). A cloud based APC management computer retrieves operating process data from an APC control computer and performs an iterative step test on the APC system. The iterative step test modifies at least one test parameter of the operating process data and identifies changes to a set of remaining parameters of the operating process data resulting from modification of the test parameter. The APC management computer determines at least one process variable from the iterative step test and generates at least one process model based on the process variable. The APC management computer transmits the process model to the APC control computer.

Abstract: A system and method for performing management and diagnostic functions in an advanced process control (APC) system. An APC management computer retrieves operating process data from an APC control computer and performs an iterative step test on the APC system. The iterative step test modifies at least one test parameter of the operating process data and identifies changes to a set of remaining parameters of the operating process data resulting from modification of the test parameter. The APC management computer determines at least one process variable from the iterative step test and generates at least one process model based on the process variable. The APC management computer transmits the process model to the APC control computer.

Abstract: A method of dynamic model selection for hybrid linear/non-linear process control includes developing a plurality of process models including at least one linear process model and at least one non-linear process model from inputs including dynamic process data from a processing system that runs a physical process. At least two of the plurality of process models are selected based on a performance comparison based on at least one metric, wherein the selected process models number less than a number of the plurality of process models received. A multi-model controller is generated that includes the selected process models. The physical process is simulated using the multi-model controller by applying the selected process models to obtain closed loop performance test data for each of the selected models. The performance test data is compared. A selected process model is then selected.

Abstract: A model predictive controller (MPC) for controlling physical processes includes a non-linear control section that includes a memory that stores a non-linear (NL) model that is coupled to a linearizer that provides at least one linearized model, and a linear control section that includes a memory that stores a linear model. A controller engine is coupled to receive both the linearized model and linear model. The MPC includes a switch that in one position causes the controller engine to operate in a linear mode utilizing the linear model to implement linear process control and in another position causes the controller engine to operate in a NL mode utilizing the linearized model to implement NL process control. The switch can be an automatic switch configured for automatically switching between linear process control and NL process control.

Abstract: A model predictive controller (MPC) for controlling physical processes includes a non-linear control section that includes a memory that stores a non-linear (NL) model that is coupled to a linearizer that provides at least one linearized model, and a linear control section that includes a memory that stores a linear model. A controller engine is coupled to receive both the linearized model and linear model. The MPC includes a switch that in one position causes the controller engine to operate in a linear mode utilizing the linear model to implement linear process control and in another position causes the controller engine to operate in a NL mode utilizing the linearized model to implement NL process control. The switch can be an automatic switch configured for automatically switching between linear process control and NL process control.

Abstract: A method of dynamic model selection for hybrid linear/non-linear process control includes developing a plurality of process models including at least one linear process model and at least one non-linear process model from inputs including dynamic process data from a processing system that runs a physical process. At least two of the plurality of process models are selected based on a performance comparison based on at least one metric, wherein the selected process models number less than a number of the plurality of process models received. A multi-model controller is generated that includes the selected process models. The physical process is simulated using the multi-model controller by applying the selected process models to obtain closed loop performance test data for each of the selected models. The performance test data is compared. A selected process model is then selected.