# US 12,038,728 B2 · Automated embedded tuning of model-less controllers > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,038,728 B2 - Document ID: US12038728B2 - Status: Granted - Kind: B2 - Application number: 17/483,335 - Filed: 2021-09-23 - Published: 2023-03-23 - Granted: 2024-07-16 - Earliest claimed: 2021-09-23 - Assignees: Rockwell Automation Technologies, Inc. - Applicants: Rockwell Automation Technologies, Inc. - Inventors: Bijan Sayyarodsari; Hanyu Shi - Web page: https://patentgrep.com/patent/US12038728B2 ## CPC - G05B 13/024: {in which a parameter or coefficient is automatically adjusted to optimise the performance} - G05B 13/042: {in which a parameter or coefficient is automatically adjusted to optimise the performance} - G05B 11/42: for obtaining a characteristic which is both proportional and time-dependent, e.g. P. I., P. I. D. - G05B 13/0265: {the criterion being a learning criterion} - G05B 2219/42134: Fuzzy logic tuning of controller as function of error - G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G) ## Abstract A method may include receiving data representative of one or more commands generated by a model-less controller to control operations of devices within a system and output parameters associated with the devices of the system. The method may also include determining whether the data is indicative of a change in operational characteristics of the system and generating a model representative of the operational characteristics of the system as a function of the data based on a Bayesian optimization algorithm in response to the data being indicative of the change. The method may also involve transmitting an excitation input to the devices in response to the data not being indicative of the change, receiving updated output parameters associated with the devices of the system after the excitation input is transmitted, and generating the model based on the updated output parameters and the excitation input. ## Claims ### Claim 1 A non-transitory computer-readable medium comprising computer-executable instructions that, when executed, are configured to cause a processor to perform operations comprising: receiving data representative of: one or more commands generated by a model-less controller to control one or more operations of one or more devices within a system; and one or more output parameters associated with the one or more devices of the system; determining whether the data is indicative of a change in one or more operational characteristics of the system; generating a model representative of the one or more operational characteristics of the system as a function of the data based on a Bayesian optimization algorithm in response to the data being indicative of the change; determining an excitation input based on the Bayesian optimization algorithm in response to the data not being indicative of the change, wherein the excitation input is configured to induce the change in the one or more operational characteristics within a threshold range by adjusting at least a first subset of the data; after determining the excitation input: transmitting the excitation input to the one or more devices; after transmitting the excitation input, identifying a time period corresponding to: when a second subset of the data is outside a second threshold range; and when a variation in the second subset of the data exceeds a variation threshold; receiving one or more updated output parameters associated with the one or more devices of the system during the time period; generating the model based on the one or more updated output parameters received during the time period and the excitation input; generating one or more model-less control parameters associated with the model-less controller based on the model; and transmitting the one or more model-less control parameters to the model-less controller, wherein the model-less control is augmented based on the one or more model-less control parameters, wherein the model-less controller is configured to control the one or more devices based on the one or more model-less control parameters to converge the system to a desired state. ### Claim 2 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein the data is representative of manipulated variable disturbance associated with the one or more operational characteristics of the system. ### Claim 3 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein the computer-executable instructions that, when executed, are configured to cause the processor to perform the operations comprising: augmenting a dataset associated with a digital twin representative of the one or more operational characteristics of the system in response to the data not being indicative of the change; and generating the model representative of the one or more operational characteristics of the system as a function of additional data acquired from the digital twin based on the Bayesian optimization algorithm. ### Claim 4 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein the one or more model-less control parameters are configured to cause the model-less controller to generate one or more additional commands configured to cause the system to more efficiently converge to a set point received by the model-less controller as compared to one or more additional parameters previously employed by the model-less controller. ### Claim 5 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein the computer-executable instructions that, when executed, are configured to cause the processor to perform the operations comprising: identifying a time period to transmit the one or more model-less control parameters to the model-less controller based on an operation schedule associated with the system; and transmitting the one or more model-less control parameters to the model-less controller during the time period. ### Claim 6 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein the model-less controller comprises a proportional-integral-derivative (PID) controller. ### Claim 7 A method, comprising: receiving, via at least one processor, data comprising: one or more commands generated by a model-less controller to control one or more operations of one or more devices within a system; and one or more output parameters associated with the one or more devices of the system; determining, via the at least one processor, whether the data is indicative of a change in one or more operational characteristics of the system; generating a model representative of the one or more operational characteristics of the system as a function of the data based on a Bayesian optimization algorithm in response to the data being indicative of the change; determining an excitation input based on the Bayesian optimization algorithm in response to the data not being indicative of the change, wherein the excitation input is configured to induce the change in the one or more operational characteristics within a threshold range by adjusting at least a first subset of the data; after determining the excitation input: transmitting the excitation input to the one or more devices; after transmitting the excitation input, identifying a time period corresponding to: when a second subset of the data is outside a second threshold range; and when a variation in the second subset of the data exceeds a variation threshold; receiving one or more updated output parameters associated with the one or more devices of the system during the time period; generating the model based on the one or more updated output parameters received during the time period and the excitation input generating one or more model-less parameters associated with the model-less controller based on the model; and transmitting the one or more model-less control parameters to the model-less controller, wherein the model-less control is augmented based on the one or more model-less control parameters, wherein the model-less controller is configured to control the one or more devices based on the one or more model-less control parameters to converge the system to a desired state. ### Claim 8 (depends on 7) The method of claim 7, wherein the data comprises manipulated variable disturbance associated with the one or more operational characteristics of the system. ### Claim 9 (depends on 7) The method of claim 7, wherein the one or more output parameters are affected by one or noise signals, one or more controlled variable disturbances, or both. ### Claim 10 (depends on 7) The method of claim 7, comprising: augmenting a dataset associated with a digital twin representative of the one or more operational characteristics of the system in response to the data not being indicative of the change; and generating the model representative of the one or more operational characteristics of the system as a function of additional data acquired from the digital twin based on the Bayesian optimization algorithm. ### Claim 11 (depends on 7) The method of claim 7, wherein the one or more model-less control parameters are configured to cause the model-less controller to generate one or more additional commands configured to cause the system to more efficiently converge to a set point received by the model-less controller as compared to one or more additional parameters previously employed by the model-less controller. ### Claim 12 (depends on 7) The method of claim 7, comprising: identifying a time period to transmit the one or more model-less control parameters to the model-less controller based on an operation schedule associated with the system; and transmitting the one or more model-less control parameters to the model-less controller during the time period. ### Claim 13 (depends on 7) The method of claim 7, wherein the model-less controller comprises a proportional-integral-derivative (PID) controller. ### Claim 14 A control system comprising one or more processors configured to perform operations comprising: receiving data representative of: one or more commands generated by a model-less controller to control one or more operations of one or more devices within a system; and one or more output parameters associated with the one or more devices of the system; determining whether the data is indicative of a change in one or more operational characteristics of the system; generating a model representative of the one or more operational characteristics of the system as a function of the data based on a Bayesian optimization algorithm in response to the data being indicative of the change; determining an excitation input based on the Bayesian optimization algorithm in response to the data not being indicative of the change, wherein the excitation input is configured to induce the change in the one or more operational characteristics within a threshold range by adjusting at least a first subset of the data; after determining the excitation input: transmitting the excitation input to the one or more devices; after transmitting the excitation input, identifying a time period corresponding to: when a second subset of the data is outside a second threshold range; and when a variation in the second subset of the data exceeds a variation threshold; receiving one or more updated output parameters associated with the one or more devices of the system during the time period; generating the model based on the one or more updated output parameters received during the time period and the excitation input generating one or more model-less control parameters associated with the model-less controller based on the model; and transmitting the one or more model-less control parameters to the model-less controller, wherein the model-less control is augmented based on the one or more model-less control parameters, and wherein the model-less controller is configured to control the one or more devices based on the one or more model-less control parameters to converge the system to a desired state. ### Claim 15 (depends on 14) The control system of claim 14, wherein the model is generated based on identifying a portion of the data associated with an uncertainty level that is less than a threshold, wherein the uncertainty level is determined based on the Bayesian optimization algorithm. ### Claim 16 (depends on 14) The control system of claim 14, wherein the one or more processors are configured to determine whether the data is indicative of the change in one or more operational characteristics of the system based on identifying a correlation between one or more input variable of the system and one or more output variables of the system. ### Claim 17 (depends on 16) The control system of claim 16, wherein the one or more processors are configured to perform the operations comprising: identifying a time period to transmit the one or more model-less control parameters to the model-less controller based on an operation schedule associated with the system; and transmitting the one or more model-less control parameters to the model-less controller during the time period.