# US 2010/0168951 A1 · Apparatus and method for detecting operational issues based on single input single output system dynamics > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2010/0168951 A1 - Document ID: US20100168951A1 - Status: Published application - Kind: A1 - Application number: 12/346,469 - Filed: 2008-12-30 - Published: 2010-07-01 - Earliest claimed: 2008-12-30 - Assignees: Honeywell International Inc. - Inventors: Dinkar Mylaraswamy; Girija Parthasarathy - Web page: https://patentgrep.com/patent/US20100168951A1 ## CPC - G06Q 10/06: Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling - Y10T 137/86558: Plural noncommunicating flow paths - Y10T 137/87161: With common valve operator - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR - Y10T: TECHNICAL SUBJECTS COVERED BY FORMER US CLASSIFICATION ## Abstract Methods and apparatus are provided for detecting an anomaly in fuel demand data and fuel supply data generated by a fuel metering system for an engine. The method comprises collecting the fuel demand data and the fuel supply data during operation of the fuel metering system, generating expected fuel supply data based on the collected fuel demand data and a nominal response model describing the expected behavior of the fuel metering system, and detecting the anomaly if a difference between the expected fuel supply data and the collected fuel supply data exceeds a predetermined threshold. ## Claims ### Claim 1 A method for detecting an anomaly in fuel demand data and fuel supply data generated by a fuel metering system for an engine, the method comprising: collecting the fuel demand data and the fuel supply data during operation of the fuel metering system; generating expected fuel supply data based on the collected fuel demand data and a nominal response model describing the expected behavior of the fuel metering system; and detecting the anomaly if a difference between the expected fuel supply data and the collected fuel supply data exceeds a predetermined threshold. ### Claim 2 (depends on 1) The method of claim 1, further comprising the step of generating the nominal response model based on historical data comprising fuel demand data and fuel supply data generated during previous operation of the fuel metering system. ### Claim 3 (depends on 1) The method of claim 1, further comprising the step of isolating the cause of the detected anomaly based on additional analysis of the collected historical data. ### Claim 4 (depends on 3) The method of claim 3, wherein step of collecting further comprises: collecting fuel demand data and fuel supply data at predetermined time intervals during a sample time period having a predetermined length. ### Claim 5 (depends on 4) The method of claim 4, wherein the step of detecting further comprises detecting the anomaly if the root means square of the difference between the expected fuel supply data and the collected fuel supply data exceeds a predetermined threshold. ### Claim 6 (depends on 5) The method of claim 5, wherein the step of isolating further comprises: applying a low pass filter having a predetermined frequency threshold to the collected fuel demand data; and classifying the anomaly as an oscillation anomaly if the difference between the filtered fuel demand data and the unfiltered fuel demand data exceeds a predetermined threshold. ### Claim 7 (depends on 6) The method of claim 6, further comprising a fuel temperature sensor for detecting the temperature of the fuel that flows to the engine and wherein the step of isolating further comprises determining the cause of the oscillation anomaly based on the fuel temperature. ### Claim 8 (depends on 5) The method of claim 5, wherein the step of isolating further comprises: determining the total amount of fuel demanded during a predetermined time period based on the collected fuel demand data; determining the total amount of fuel supplied during the predetermined time period based on the collected fuel supply data; and detecting an offset anomaly if the difference between the total amount of fuel demanded and the total amount of fuel supplied is greater than a predetermined threshold. ### Claim 9 (depends on 8) The method of claim 8, further comprising an engine exhaust heat sensor and wherein the step of isolating further comprises: determining the total amount of heat expelled from the engine during the predetermined time period; and detecting the cause of the offset anomaly based on the total amount of heat expelled from the engine during the predetermined time period. ### Claim 10 A computer readable medium having instructions stored thereon that when executed by a processor cause the processor to perform a method for determining a nominal response model describing a single input single output (SISO) system, the method comprising: selecting non-anomalous data sets from a plurality of data sets stored on a historical data source, wherein each selected data set includes system input data and system output data for the SISO system; and determining a nominal response model for generating expected system output data based on the system input data from the selected data sets, wherein the difference between the expected system output data and the system output data from the selected data sets is minimized. ### Claim 11 (depends on 10) The computer readable medium of claim 10, wherein the step of selecting further comprises: a) generating a preliminary nominal response model that describes a predetermined number of data sets that were randomly selected from the plurality of data sets and include non-anomalous data; b) for each of the plurality of data sets, determining a data fitness level describing the correlation between the data set and the preliminary nominal response model; c) identifying each data set that has a data fitness level exceeding a predetermined threshold; d) repeating steps a-c a predetermined number of times; and e) selecting data sets that are identified during step c with a threshold frequency level. ### Claim 12 (depends on 11) The computer readable medium of claim 11, wherein the step of generating further comprises: 1) randomly selecting a predetermined number of data sets from the plurality of stored data sets; 2) determining a preliminary nominal response model for generating expected system output data based on the system input data for the randomly selected data sets, wherein the difference between the expected system output data and the system output data from the randomly selected data sets is minimized; 3.) for each randomly selected data set, determining a data fitness level describing the correlation between the data set and the preliminary nominal response model; and 4) repeating steps 1-3 if the data fitness level for each randomly selected data set does not exceed a predetermined threshold. ### Claim 13 (depends on 12) The method of claim 12, wherein: the step of randomly selecting further comprises merging the randomly selected data sets; and the step of determining further comprises determining the preliminary nominal response model based on the merged data set, wherein the preliminary nominal response model is an ARX (auto-regressive) model. ### Claim 14 (depends on 13) The method of claim 13, wherein further comprising: merging the selected data sets; and wherein the step of determining further comprises determining the nominal response model based on the merged data set, wherein the nominal response model is an ARX (auto-regressive) model. ### Claim 15 (depends on 14) The method of claim 14, wherein the step of determining further comprising using MATLABS system ID tool box to determine the ARX model. ### Claim 16 A fuel metering system for regulating the flow of fuel to an engine, the fuel metering system comprising: a fuel delivery system configured to receive a signal representative of a fuel demand and generate a signal representative of a fuel supply; and a controller coupled to the fuel delivery system and configured to: collect fuel demand data provided to the engine and fuel supply data generated by the fuel delivery system during a predetermined sample period; generate expected fuel supply data based on the collected fuel demand data and a nominal response model describing the expected behavior of the fuel metering system; and detect an anomaly if the difference between the expected fuel supply data and the collected fuel supply data exceeds a predetermined threshold. ### Claim 17 (depends on 16) The fuel metering system of claim 16, wherein the nominal response model is generated based on historical fuel demand data and fuel supply data. ### Claim 18 (depends on 16) The fuel metering system of claim 16, wherein the processor is further configured to isolate the cause of a detected anomaly based on additional analysis of the collected fuel demand data. ### Claim 19 (depends on 16) The fuel metering system of claim 16, wherein the processor is further configured to isolate the cause of a detected anomaly based on additional analysis of the collected fuel demand data, the collected fuel supply data, and data collected from one or more external data sources. ### Claim 20 (depends on 16) The fuel metering system of claim 16, wherein in the processor is further configured to detect the anomaly if the root means square of the difference between the expected fuel output data and the collected fuel supply data exceeds a predetermined threshold.