# US 11,029,218 B2 · Determination device, determination method, and non-transitory computer-readable recording medium > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,029,218 B2 - Document ID: US11029218B2 - Status: Granted - Kind: B2 - Application number: 15/676,496 - Filed: 2017-08-14 - Published: 2018-02-01 - Granted: 2021-06-08 - Earliest claimed: 2015-02-17 - Assignees: Fujitsu Limited - Applicants: Fujitsu Limited - Inventors: Kazushi Uno; Takahiro Arioka; Takeo Kasajima; Hiroyuki Fukuda - Web page: https://patentgrep.com/patent/US11029218B2 ## CPC - G01K 11/32: using changes in transmittance, scattering or luminescence in optical fibres - G01K 3/005: {Circuits arrangements for indicating a predetermined temperature (fire detection G08B17/00)} - G01M 3/002: {by using thermal means} - G05B 23/0243: {model based detection method, e.g. first-principles knowledge model} - G01K 11/324: using Raman scattering - G05B 19/0425: {Safety, monitoring} - G05B 23/0235: {based on a comparison with predetermined threshold or range, e.g. "classical methods", carried out during normal operation; threshold adaptation or choice; when or how to compare with the threshold} - G01K: MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR (radiation pyrometry G01J5/00) - G01M: TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR - 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 determination device includes: a memory; and a processor coupled to the memory and the processor configured to executes a process, the process comprising: generating a reference model of a sensor detection value; determining whether a time from a predetermined point in time until a deviation between the reference model and the sensor detection value exceeds a threshold is shorter than a predetermined time; and outputting a signal associated with an abnormality when the time is determined to be shorter. ## Claims ### Claim 1 A determination device comprising: a memory; and a processor coupled to the memory and the processor configured to execute a process, the process comprising: generating a reference model of a sensor detection value, after generating the reference model, starting a measuring of a time duration at a predetermined point in time before detecting a deviation between the reference model and the sensor detection value exceeds a threshold and continuing the measuring of the time duration until the deviation between the reference model and the sensor detection value exceeds the threshold, determining whether the time duration measured in the measuring is shorter than a predetermined time duration; and outputting a signal associated with an abnormality when the time duration is determined to be shorter than the predetermined time duration. ### Claim 2 (depends on 1) The determination device as claimed in claim 1, wherein in the generating, the reference model is generated with use of the sensor detection value and detection values of a plurality of other sensors having a correlation with the sensor detection value. ### Claim 3 (depends on 2) The determination device as claimed in claim 2, wherein in the generating, the reference model is generated by a regression analysis with use of the sensor detection value and the detection values of the plurality of other sensors. ### Claim 4 (depends on 2) The determination device as claimed in claim 2, wherein the process further comprises re-generating the reference model with use of the sensor detection value and the detection values of the plurality of other sensors of a predetermined past time from a time when a deviation between the reference model and the sensor detection value exceeds a threshold. ### Claim 5 (depends on 4) The determination device as claimed in claim 4, wherein in the determining, further determining whether a time from a predetermined time until a deviation between the sensor detection value and the reference model that is re-generated exceeds the threshold is shorter than a predetermined time, with use of the sensor detection value before re-generating of the reference model. ### Claim 6 (depends on 2) The determination device as claimed in claim in claim 2, wherein the deviation between the reference model and the sensor detection value is a difference between the reference model and the sensor detection value. ### Claim 7 (depends on 1) The determination device as claimed in claim 1, wherein: in the generating, the reference model of the detection value is generated with use of a correlation between the detection values of a plurality of sensors; and in the determining, further determining whether a time until the deviation between the reference model and any one of the detection values of the plurality of sensors exceeds a threshold is shorter than a predetermined time. ### Claim 8 (depends on 7) The determination device as claimed in claim 7, wherein in the generating, the reference model is generated with use of a center of variability on which a correlation degree of the detection values of the plurality of sensors influences. ### Claim 9 (depends on 7) The determination device as claimed in claim 7, wherein the process further comprises re-generating the reference model with use of the detection values of the plurality of sensors of a predetermined past period from a time when the deviation exceeds the threshold. ### Claim 10 (depends on 7) The determination device as claimed in claim 7, wherein the detection values of the plurality of sensors are results obtained from a back scattering lights at different longitudinal direction positions in a single optical fiber. ### Claim 11 (depends on 7) The determination device as claimed in claim 7, wherein in the generating, the reference model is generated with use of a variance-covariance matrix using an average of detection values of the plurality of sensors. ### Claim 12 (depends on 1) The determination device as claimed in claim 1, wherein the threshold of the deviation is determined on a basis of the deviation of a predetermined period after generating of the reference model. ### Claim 13 (depends on 1) The determination device as claimed in claim 1, wherein the predetermined time is determined on a basis of a variability of a time until the deviation exceeds the threshold. ### Claim 14 (depends on 1) The determination device as claimed in claim 1, wherein the predetermined point in time is a starting point of measuring the deviation between the reference model and the sensor detection value. ### Claim 15 (depends on 1) The determination device as claimed in claim 1, wherein, the determining includes whether the deviation exceeds a first threshold, whether an accumulated value of the deviation exceeds a second threshold, and whether one of a first time duration from the predetermined point in time until the deviation exceeds the first threshold and a second time duration from the predetermined point in time until the accumulated value exceeds the second threshold is shorter than the predetermined time duration, and wherein, in the outputting, the signal is output upon determining that the deviation exceeds the first threshold or the accumulated value exceeds the second threshold, and upon determining that one of the first time duration and the second time duration is shorter than the predetermined time duration. ### Claim 16 A determination method comprising: generating a reference model of a sensor detection value; after generating the reference model, starting a measuring of a time duration at a predetermined point in time before detecting a deviation between the reference model and the sensor detection value exceeds a threshold and continuing the measuring of the time duration until the deviation between the reference model and the sensor detection value exceeds the threshold; determining the time duration measured in the measuring is shorter than a predetermined time duration; and outputting a signal associated with an abnormality when the time duration is determined to be shorter than the predetermined time duration. ### Claim 17 (depends on 16) The determination method as claimed in claim 16, wherein in the generating, the reference model is generated with use of the sensor detection value and detection values of a plurality of other sensors having a correlation with the sensor detection value. ### Claim 18 (depends on 17) The determination method as claimed in claim 17, wherein in the generating, the reference model is generated by a regression analysis with use of the sensor detection value and the detection values of the plurality of other sensors. ### Claim 19 (depends on 17) The determination method as claimed in claim 17, further comprising re-generating the reference model with use of the sensor detection value and the detection values of the plurality of other sensors of a predetermined past time from a time when a deviation between the reference model and the sensor detection value exceeds a threshold. ### Claim 20 (depends on 17) The determination method as claimed in claim in claim 17, wherein the deviation between the reference model and the sensor detection value is a difference between the reference model and the sensor detection value. ### Claim 21 (depends on 16) The determination method as claimed in claim 16, wherein: in the generating, the reference model of the detection value is generated with use of a correlation between the detection values of a plurality of sensors; and in the determining, further determining whether a time until the deviation between the reference model and any one of the detection values of the plurality of sensors exceeds a threshold is shorter than a predetermined time. ### Claim 22 A computer readable, non-transitory medium storing a program that causes a computer to execute a process, the process comprising: generating a reference model of a sensor detection value; after generating the reference model, starting a measuring a time duration at a predetermined point in time before detecting a deviation between the reference model and the sensor detection value exceeds a threshold and continuing the measuring of the time duration until the deviation between the reference model and the sensor detection value exceeds the threshold; determining whether the time duration measured in the measuring is shorter than a predetermined time duration; and outputting a signal associated with an abnormality when the time duration is determined to be shorter than the predetermined time duration.