# US 2020/0404754 A1 · Lighting load classification and dimmer configuration based thereon > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2020/0404754 A1 - Document ID: US20200404754A1 - Status: Published application - Kind: A1 - Application number: 16/444,788 - Filed: 2019-06-18 - Published: 2020-12-24 - Earliest claimed: 2019-06-18 - Assignees: Leviton Manufacturing Co., Inc. - Applicants: Leviton Manufacturing Company, Inc. - Inventors: A M Sarwar JAHAN; Wen GU; Tianjian HUANG - Web page: https://patentgrep.com/patent/US20200404754A1 ## CPC - G06N 3/08: Learning methods - G06N 20/00: Machine learning - G06N 3/0499: Feedforward networks - G06N 3/09: Supervised learning - H05B 47/10: Controlling the light source - H05B 47/175: by remote control - H05B 47/1965: {using handheld communication devices} - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - H05B: ELECTRIC HEATING; ELECTRIC LIGHT SOURCES NOT OTHERWISE PROVIDED FOR; CIRCUIT ARRANGEMENTS FOR ELECTRIC LIGHT SOURCES, IN GENERAL ## Abstract A machine learning model for classifying different lighting loads based on properties of electrical current through the loads is built. The model is applied to electrical data gathered based on powering a target lighting load in order to classify the load. Based on load classification, operating parameter(s) to control operation of a dimmer to control power to the load are selected and the dimmer is configured with the selected parameter(s) to achieve desired dimmer operation in dimming the load. ## Claims ### Claim at position 1 1-4. (canceled) ### Claim 5 (depends on 112) The dimmer of claim 112, wherein the machine learning model is an initial machine learning model, and wherein the performing processing further comprises: receiving, in response to the sending, an updated machine learning model; using the updated machine learning model, repeating the applying to classify the lighting load, wherein the repeating the applying classifies the lighting load into a lighting load class indicated by the updated model; identifying, based on the lighting load class indicated by the updated model, the one or more dimmer operating parameters; and performing the configuring the dimmer with the identified one or more dimmer operating parameters. ### Claim 6 (depends on 112) The dimmer of claim 112, wherein the performing processing further comprises: receiving, in response to the sending, an indication of a class of the lighting load; identifying, based on the lighting load class indicated by the updated model, the one or more dimmer operating parameters; and performing the configuring the dimmer with the identified one or more dimmer operating parameters. ### Claim at position 4 7-9. (canceled) ### Claim 10 (depends on 114) The dimmer of claim 114, wherein the dimmer further comprises a network interface through which the dimmer is connected to a network, and wherein the obtaining obtains the machine learning model through the network interface from a remote entity connected to the network. ### Claim 11 (canceled) ### Claim 12 (depends on 114) The dimmer of claim 114, wherein the different lighting load classes correspond to different lamp types having differing internal circuitry and electrical performance thereof, and wherein the different lighting load classes comprise (i) a class for incandescent lamps, (ii) a class for compact fluorescent lamps, and (iii) more than one class for light emitting diode (LED) lamp types, wherein each LED lamp type of the LED lamp types differs from the other LED lamp types based at least on its drive circuitry, and wherein each class of the more than one class corresponds to the drive circuitry of a respective LED lamp type of the LED lamp types. ### Claim 13 (depends on 114) The dimmer of claim 114, wherein the memory stores programs instructions for execution by the processing circuit to perform the method. ### Claim at position 9 14-26. (canceled) ### Claim 27 (depends on 115) The method of claim 115, further comprising: maintaining a database of dimmer operating parameters and associated lighting load classes, each lighting load class of the associated lighting load classes being associated with a respective one or more dimmer operating parameters on a basis that the respective one or more dimmer operating parameters optimizes performance of drive circuitry of lighting loads of that lighting load class to minimize flicker on the lighting load; receiving from the remote entity an indication of a lighting load class into which the lighting load is classified; and identifying from the maintained database the one or more dimmer operating parameters with which to configure the dimmer as the one or more operating parameters associated with the lighting load class into which the lighting load is classified. ### Claim 28 (depends on 115) The method of claim 115, further comprising receiving from the remote entity the one or more dimmer operating parameters, the one or more dimmer operating parameters identified based on a lighting load class into which the lighting load is classified. ### Claim 29 (depends on 115) The method of claim 115, wherein the classification of the lighting load comprises a classification into a lighting load class of a plurality of different lighting load classes that correspond to different lamp types having differing internal circuitry and electrical performance thereof, wherein the different lighting load classes comprise (i) a class for incandescent lamps, (ii) a class for compact fluorescent lamps, and (iii) more than one class for light emitting diode (LED) lamp types, wherein each LED lamp type of the LED lamp types differs from the other LED lamp types based at least on its drive circuitry, and wherein each class of the more than one class corresponds to the drive circuitry of a respective LED lamp type of the LED lamp types. ### Claim 30 (canceled) ### Claim 34 (depends on 117) The dimmer of claim 117, wherein the method further comprises: maintaining a database of dimmer operating parameters and associated lighting load classes, each lighting load class of the associated lighting load classes being associated with a respective one or more dimmer operating parameters on a basis that the respective one or more dimmer operating parameters optimizes performance of drive circuitry of lighting loads of that lighting load class to minimize flicker on the lighting load; and identifying the one or more dimmer operating parameters with which to configure the dimmer as the respective one or more operating parameters associated with the lighting load class into which the lighting load is classified. ### Claim at position 15 35-36. (canceled) ### Claim 37 (depends on 117) The dimmer of claim 117, wherein the classification of the lighting load comprises a classification into a lighting load class of a plurality of different lighting load classes that correspond to different lamp types having differing internal circuitry and electrical performance thereof. ### Claim 38 (depends on 37) The dimmer of claim 37, wherein the different lighting load classes comprise (i) a class for incandescent lamps, (ii) a class for compact fluorescent lamps, and (iii) more than one class for light emitting diode (LED) lamp types, wherein each LED lamp type of the LED lamp types differs from the other LED lamp types based at least on its drive circuitry, and wherein each class of the more than one class corresponds to the drive circuitry of a respective LED lamp type of the LED lamp types. ### Claim at position 18 39-111. (canceled) ### Claim 112 A dimmer for controlling conduction of a supply of power to a lighting load, the dimmer comprising: a line input terminal and a load output terminal, the line input terminal configured to be electrically coupled to the supply of power, and the load output terminal configured to be electrically coupled to the lighting load; a switching circuit electrically coupled in series between the line input terminal and the load output terminal, the switching circuit configured to be selectively controlled between an ON state and an OFF state; a memory; and a processing circuit in communication with the memory, wherein the dimmer is configured to perform a method comprising: obtaining a machine learning model, the machine learning model configured for classifying lighting loads into a plurality of different lighting load classes based on properties of electrical current through the lighting loads; based on conducting the supply of power to the lighting load, obtaining electrical current data representing properties of electrical current through the lighting load over a duration of time; applying the machine learning model, using the obtained electrical current data representing properties of electrical current through the lighting load, to classify the lighting load; and performing processing based on the applying, wherein the performing processing comprises configuring the dimmer with one or more dimmer operating parameters that control operation of the dimmer, wherein the applying fails to classify the lighting load into any lighting load class of the plurality of different lighting load classes, and wherein the performing processing comprises sending the obtained electrical current data to a remote entity. ### Claim 113 A dimmer for controlling conduction of a supply of power to a lighting load, the dimmer comprising: a line input terminal and a load output terminal, the line input terminal configured to be electrically coupled to the supply of power, and the load output terminal configured to be electrically coupled to the lighting load; a switching circuit electrically coupled in series between the line input terminal and the load output terminal, the switching circuit configured to be selectively controlled between an ON state and an OFF state; a memory; and a processing circuit in communication with the memory, wherein the dimmer is configured to perform a method comprising: obtaining a machine learning model, the machine learning model configured for classifying lighting loads into a plurality of different lighting load classes based on properties of electrical current through the lighting loads; based on conducting the supply of power to the lighting load, obtaining electrical current data representing properties of electrical current through the lighting load over a duration of time; applying the machine learning model, using the obtained electrical current data representing properties of electrical current through the lighting load, to classify the lighting load; and performing processing based on the applying, wherein the performing processing comprises configuring the dimmer with one or more dimmer operating parameters that control operation of the dimmer, wherein the applying fails to classify the lighting load into any lighting load class of the plurality of different lighting load classes, and wherein the performing processing comprises selecting a default one or more dimmer operating parameters as the one or more dimmer operating parameters with which to configure the dimmer. ### Claim 114 A dimmer for controlling conduction of a supply of power to a lighting load, the dimmer comprising: a line input terminal and a load output terminal, the line input terminal configured to be electrically coupled to the supply of power, and the load output terminal configured to be electrically coupled to the lighting load; a switching circuit electrically coupled in series between the line input terminal and the load output terminal, the switching circuit configured to be selectively controlled between an ON state and an OFF state; a memory; and a processing circuit in communication with the memory, wherein the dimmer is configured to perform a method comprising: obtaining a machine learning model, the machine learning model configured for classifying lighting loads into a plurality of different lighting load classes based on properties of electrical current through the lighting loads; based on conducting the supply of power to the lighting load, obtaining electrical current data representing properties of electrical current through the lighting load over a duration of time; applying the machine learning model, using the obtained electrical current data representing properties of electrical current through the lighting load, to classify the lighting load; performing processing based on the applying, wherein the performing processing comprises configuring the dimmer with one or more dimmer operating parameters that control operation of the dimmer; maintaining the machine learning model as a maintained machine learning model on the dimmer for lighting load classification; periodically or aperiodically receiving from a remote entity an updated machine learning model as part of a firmware update to the dimmer; and replacing the maintained machine learning model with the received updated machine learning model such that the updated machine learning model becomes the maintained machine learning model for lighting load classification. ### Claim 115 A method for controlling operation of a dimmer, the method comprising: conducting a supply of power to a lighting load; based on the conducting, obtaining electrical current data representing properties of electrical current through the lighting load over a duration of time; sending the electrical current data to a remote entity; and based on a classification of the lighting load, the classification being based on the electrical current data, configuring the dimmer with one or more dimmer operating parameters that control operation of the dimmer, wherein the classification is performed by the remote entity based on receiving the sent electrical current data. ### Claim 116 A method for controlling operation of a dimmer, the method comprising: obtaining a machine learning model, the machine learning model configured for classifying lighting loads into a plurality of different lighting load classes based on properties of electrical current through the lighting loads; maintaining a database of dimmer operating parameters and associated lighting load classes, each lighting load class of the associated lighting load classes being associated with a respective one or more dimmer operating parameters on a basis that the respective one or more dimmer operating parameters optimizes performance of drive circuitry of lighting loads of that lighting load class to minimize flicker on the lighting load; conducting a supply of power to a lighting load; based on the conducting, obtaining electrical current data representing properties of electrical current through the lighting load over a duration of time; and performing a classification of the lighting load, the performing the classification comprising applying the machine learning model, using the obtained electrical current data representing properties of electrical current through the lighting load, to classify the lighting load; based on the classification of the lighting load, the classification being based on the electrical current data: identifying from the maintained database one or more dimmer operating parameters that control operation of the dimmer, the identified one or more dimmer operating parameters being identified as the one or more operating parameters associated with the lighting load class into which the lighting load is classified; and configuring the dimmer with the identified one or more dimmer operating parameters that control operation of the dimmer. ### Claim 117 A dimmer for controlling conduction of a supply of power to a lighting load, the dimmer comprising: a line input terminal and a load output terminal, the line input terminal configured to be electrically coupled to the supply of power, and the load output terminal configured to be electrically coupled to the lighting load; a switching circuit electrically coupled in series between the line input terminal and the load output terminal, the switching circuit configured to be selectively controlled between an ON state and an OFF state; a memory; and a processing circuit in communication with the memory, wherein the dimmer is configured to perform a method comprising: based on conducting the supply of power to the lighting load, obtaining electrical current data representing properties of electrical current through the lighting load over a duration of time; sending the electrical current data to a remote entity; based on the classification of the lighting load, the classification being based on the sent electrical current data and the classification being by a machine learning model, receiving from the remote entity an indication of a lighting load class into which the lighting load is classified; identifying, based on the indication of the lighting load class, one or more dimmer operating parameters that control operation of the dimmer; and configuring the dimmer with the one or more dimmer operating parameters that control operation of the dimmer. ### Claim 118 A dimmer for controlling conduction of a supply of power to a lighting load, the dimmer comprising: a line input terminal and a load output terminal, the line input terminal configured to be electrically coupled to the supply of power, and the load output terminal configured to be electrically coupled to the lighting load; a switching circuit electrically coupled in series between the line input terminal and the load output terminal, the switching circuit configured to be selectively controlled between an ON state and an OFF state; a memory; and a processing circuit in communication with the memory, wherein the dimmer is configured to perform a method comprising: based on conducting the supply of power to the lighting load, obtaining electrical current data representing properties of electrical current through the lighting load over a duration of time; sending the electrical current data to a remote entity; based on a classification of the lighting load, the classification being based on the sent electrical current data and the classification being by a machine learning model, receiving from the remote entity one or more dimmer operating parameters that control operation of the dimmer, the one or more dimmer operating parameters identified based on a lighting load class into which the lighting load is classified; and configuring the dimmer with the one or more dimmer operating parameters that control operation of the dimmer.