# US 2021/0240177 A1 · Learning apparatus, learning method, computer readable medium having recorded thereon learning program, determination apparatus, determination method, and computer readable medium having recorded thereon determination program > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0240177 A1 - Document ID: US20210240177A1 - Status: Published application - Kind: A1 - Application number: 17/144,063 - Filed: 2021-01-07 - Published: 2021-08-05 - Earliest claimed: 2020-01-31 - Assignees: Yokogawa Electric Corporation - Applicants: Yokogawa Electric Corporation - Inventors: Ryohei FUJII; Yanyan LIN - Web page: https://patentgrep.com/patent/US20210240177A1 ## CPC - G05B 17/02: electric - G05B 13/02: electric - G05B 23/0221: {Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods} - G05B 23/0232: {based on qualitative trend analysis, e.g. system evolution} - G06F 18/2148: {characterised by the process organisation or structure, e.g. boosting cascade} - G06N 20/10: using kernel methods, e.g. support vector machines [SVM] - G05B 23/0272: {Presentation of monitored results, e.g. selection of status reports to be displayed; Filtering information to the user} - 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) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract A learning apparatus is provided, which comprises: a learning data acquiring unit for acquiring learning data including measurement data obtained by measuring a facility and a state of the facility; a learning pre-processing unit for performing a pre-processing for reducing a drift of the measurement data in the learning data and outputting pre-processed learning data; and a learning processing unit for performing a processing for learning a model for determining the state of the facility from the pre-processed measurement data, by using the pre-processed learning data. ## Claims ### Claim 1 A learning apparatus comprising: a learning data acquiring unit for acquiring learning data including measurement data obtained by measuring a facility and a state of the facility; a learning pre-processing unit for performing a pre-processing for reducing a drift of the measurement data in the learning data and outputting pre-processed learning data; and a learning processing unit for performing a processing for learning a model for determining the state of the facility from the pre-processed measurement data, by using the pre-processed learning data. ### Claim 2 (depends on 1) The learning apparatus according to claim 1, wherein the learning pre-processing unit comprises: a feature quantity calculating unit for calculating a feature quantity representing the drift of the measurement data in the learning data; and a data converting unit for removing the drift according to the feature quantity from the measurement data in the learning data. ### Claim 3 (depends on 1) The learning apparatus according to claim 1, further comprising a designation input unit for inputting a designation of a type of pre-processing that the learning pre-processing unit should perform, wherein the learning pre-processing unit is configured to perform a pre-processing corresponding to the inputted designation. ### Claim 4 (depends on 3) The learning apparatus according to claim 3, wherein the learning pre-processing unit is configured to perform, according to the designation, at least one of a pre-processing for reducing a gravity center movement of the measurement data in the learning data or a pre-processing for calculating the difference of the measurement data in the learning data within a sliding window. ### Claim 5 (depends on 1) The learning apparatus according to claim 1, further comprising: a determination data acquiring unit for acquiring determination data including measurement data obtained by measuring the facility; a determination pre-processing unit for performing a pre-processing for reducing a drift of the determination data and outputting pre-processed determination data; and a determining unit for determining the state of the facility from the pre-processed determination data by using the model learned. ### Claim 6 (depends on 5) The learning apparatus according to claim 5, further comprising a display processing unit for performing a processing for displaying a distribution of the pre-processed learning data and the pre-processed determination data. ### Claim 7 (depends on 6) The learning apparatus according to claim 6, wherein the learning data and the determination data include measurement values of a plurality of sensors provided in the facility, wherein the learning apparatus further comprises a selecting unit for selecting measurement values of at least one sensor having a relatively high degree of contribution to a determination result of the model among the measurement values of the plurality of sensors, wherein the display processing unit is configured to perform a processing for displaying a distribution of the measurement values of the at least one sensor selected by the selecting unit among the pre-processed learning data and the pre-processed determination data. ### Claim 8 A learning method comprising: by a learning apparatus, acquiring learning data including measurement data obtained by measuring a facility and a state of the facility; by the learning apparatus, performing a pre-processing for reducing a drift of the measurement data in the learning data and outputting pre-processed learning data; and by the learning apparatus, performing a processing for learning a model for determining the state of the facility from the pre-processed measurement data, by using the pre-processed learning data. ### Claim 9 (depends on 8) The learning method according to claim 8, wherein performing the processing for learning the model comprises: calculating a feature quantity representing the drift of the measurement data in the learning data; and removing the drift according to the feature quantity from the measurement data in the learning data. ### Claim 10 (depends on 8) The learning method according to claim 8, further comprising, by the learning apparatus, inputting a designation of a type of pre-processing that should be performed in outputting the pre-processed learning data, wherein outputting the pre-processed learning data includes performing a pre-processing corresponding to the inputted designation. ### Claim 11 (depends on 10) The learning method according to claim 10, wherein outputting the pre-processed learning data includes performing, according to the designation, at least one of a pre-processing for reducing a gravity center movement of the measurement data in the learning data or a pre-processing for calculating the difference of the measurement data in the learning data within a sliding window. ### Claim 12 A computer readable medium having recorded thereon a learning program to be executed by a computer for causing the computer to function as a learning apparatus comprising: a learning data acquiring unit for acquiring learning data including measurement data obtained by measuring a facility and a state of the facility; a learning pre-processing unit for performing a pre-processing for reducing a drift of the measurement data in the learning data and outputting pre-processed learning data; and a learning processing unit for performing a processing for learning a model for determining the state of the facility from the pre-processed measurement data, by using the pre-processed learning data. ### Claim 13 A determination apparatus comprising: a determination data acquiring unit for acquiring determination data including measurement data obtained by measuring a facility; a determination pre-processing unit for performing a pre-processing for reducing a drift of the determination data and outputting pre-processed determination data; and a determining unit for determining a state of the facility from the pre-processed determination data, by using a model learned by using pre-processed learning data obtained by performing a pre-processing on learning data. ### Claim 14 (depends on 13) The determination apparatus according to claim 13, further comprising a display processing unit for performing a processing for displaying a distribution of the pre-processed learning data and the pre-processed determination data. ### Claim 15 (depends on 14) The determination apparatus according to claim 14, wherein the learning data and the determination data include measurement values of a plurality of sensors provided in the facility, wherein the determination apparatus further comprising a selecting unit for selecting measurement values of at least one sensor having a relatively high degree of contribution to a determination result of the model among the measurement values of the plurality of sensors, wherein the display processing unit is configured to perform a processing for displaying a distribution of the measurement values of the at least one sensor selected by the selecting unit among the pre-processed learning data and the pre-processed determination data. ### Claim 16 A determination method comprising: by a determination apparatus, acquiring determination data including measurement data obtained by measuring a facility; by the determination apparatus, performing a pre-processing for reducing a drift of the determination data and outputting pre-processed determination data; and by the determination apparatus, determining a state of the facility from the pre-processed determination data, by using a model learned by using pre-processed learning data obtained by performing a pre-processing on learning data. ### Claim 17 (depends on 16) The determination method according to claim 16, further comprising, by the determination apparatus, performing a processing for displaying a distribution of the pre-processed learning data and the pre-processed determination data. ### Claim 18 (depends on 17) The determination method according to claim 17, wherein the learning data and the determination data include measurement values of a plurality of sensors provided in the facility, wherein the determination method further comprises selecting measurement values of at least one sensor having a relatively high degree of contribution to a determination result of the model among the measurement values of the plurality of sensors, wherein performing the processing for displaying includes performing a processing for displaying a distribution of the measurement values of the at least one sensor selected by the selecting unit among the pre-processed learning data and the pre-processed determination data. ### Claim 19 A computer readable medium having recorded thereon a determination program to be executed by a computer for causing the computer to function as: a determination data acquiring unit for acquiring determination data including measurement data obtained by measuring a facility; a determination pre-processing unit for performing a pre-processing for reducing a drift of the determination data and outputting pre-processed determination data; and a determining unit for determining a state of the facility from the pre-processed determination data, by using a model learned by using pre-processed learning data obtained by performing a pre-processing on learning data.