Automatic optimization of continuous processes

A system for performing an optimization operation on continuous processes generated from a plurality of system sensors, the system comprising:

• a processor;
• a data bus coupled to the processor;
• and
• a non-transitory, computer-readable storage medium embodying computer program code, the non-transitory, computer-readable storage medium being coupled to the data bus, the computer program code interacting with a plurality of computer operations and comprising instructions executable by the processor and configured for:
• transforming a continuous stream of data comprising process states associated with specific ranges of a subset of control parameter inputs, wherein transforming comprises:
• dividing the control parameter inputs into outputs, controllable inputs, and uncontrollable inputs;
• and
• identifying constraints for the outputs, controllable inputs, and uncontrollable inputs;
• performing a clustering operation on select uncontrollable inputs, wherein the clustering operation includes dividing the ranges into a plurality of reliable repeatable process states;
• identifying reliably repeatable optimal states from the reliable repeatable process states, the identifying the -reliably repeatable optimal states comprising defining a key performance indicator as a continuous process output by applying a recursive-partitioning operation to identify input parameter settings that are associated with a desirable process state, the recursive-partitioning operation comprising identifying surrogate parameters and alternative recursive-partitioning models to obtain a pool of combinations of possible parameter settings associated with an optimized process performance;
• characterizing relationships among the control parameter inputs for the reliably repeatable optimal states and for non-optimal states;
• and
• determining operational states for the control parameter inputs based on the characterized relationships;
• wherein the operational states describe controllable parameters and settings to control the system.

A non-transitory, computer-readable storage medium embodying computer program code, the computer program code comprising computer executable instructions configured for:

• transforming a continuous stream of data comprising process states associated with specific ranges of a subset of control parameter inputs, wherein transforming comprises:
• dividing the control parameter inputs into outputs, controllable inputs, and uncontrollable inputs;
• and
• identifying constraints for the outputs, controllable inputs, and uncontrollable inputs;
• performing a clustering operation on select uncontrollable inputs, wherein the clustering operation includes dividing the ranges into a plurality of reliable repeatable process states;
• identifying reliably repeatable optimal states from the reliable repeatable process states, the identifying the -reliably repeatable optimal states comprising defining a key performance indicator as a continuous process output by applying a recursive-partitioning operation to identify input parameter settings that are associated with a desirable process state, the recursive-partitioning operation comprising identifying surrogate parameters and alternative recursive-partitioning models to obtain a pool of combinations of possible parameter settings associated with an optimized process performance;
• characterizing relationships among the control parameter inputs for the reliably repeatable optimal states and for non-optimal states;
• and
• determining operational states for the control parameter inputs based on the characterized relationships;
• wherein the operational states describe controllable parameters and settings to control the system.