# US 2019/0317488 A1 · Predicting failures in electrical submersible pumps using pattern recognition > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2019/0317488 A1 - Document ID: US20190317488A1 - Status: Published application - Kind: A1 - Application number: 15/951,798 - Filed: 2018-04-12 - Published: 2019-10-17 - Earliest claimed: 2018-04-12 - Assignees: Saudi Arabian Oil Company - Applicants: Saudi Arabian Oil Company - Inventors: Abdulmonam Al-Maghlouth; Obiomalotaoso Leonard Isichei; Mohammad Fahad Al-Qasimi - Web page: https://patentgrep.com/patent/US20190317488A1 ## CPC - G05B 23/024: {Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks} - G05B 23/0283: {Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL] (preventive maintenance, i.e. planning maintenance according to the available resources without monitoring the system G06Q10/06)} - E21B 43/128: {Adaptation of pump systems with down-hole electric drives} - F04D 15/0088: {Testing machines} - G05B 23/0248: {Causal models, e.g. fault tree; digraphs; qualitative physics} - G05B 23/0275: {Fault isolation and identification, e.g. classify fault; estimate cause or root of failure} - 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) - E21B: EARTH OR ROCK DRILLING; OBTAINING OIL, GAS, WATER, SOLUBLE OR MELTABLE MATERIALS OR A SLURRY OF MINERALS FROM WELLS - F04D: NON-POSITIVE-DISPLACEMENT PUMPS (engine fuel-injection pumps F02M; ion pumps H01J41/12; electrodynamic pumps H02K44/02) ## Abstract Sensor data is received from a plurality of sensors contained in an electrical submersible pump (ESP) deployed in a well hole. An early indication of an ESP failure that is imminent is determined in real time using a pattern recognition methodology based on the sensor data and a pattern detection model that correlates historical ESP failures with historical sensor readings. A notification of the imminence of the ESP failure is provided. ## Claims ### Claim 1 A computer-implemented method, comprising: receiving sensor data from a plurality of sensors, each sensor contained in an electrical submersible pump (ESP) deployed in a well hole; determining, in real time using a pattern recognition methodology based on the sensor data and a pattern detection model that correlates historical ESP failures with historical sensor readings, an early indication of an ESP failure that is imminent; and providing a notification of the imminence of the ESP failure. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, wherein the sensor data includes motor temperature data, current data, vibration data, intake pressure data, and discharge pressure data. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, further comprising: training the pattern detection model to understand normal behaviors, abnormal behaviors, and critical events of each sensor, wherein the training is based on an analysis of a historical behavior of failed pumps, a historical sensor readings, a failure database, and a database of factors for different ESP types; classifying potential ESP failures under low warnings or high warnings based on a severity of an alarm received from a particular sensor of a particular ESP type, wherein the classifying a particular potential ESP failure is based on a single sensor or multiple sensors; and storing information for each potential ESP failure in the pattern detection model. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, wherein classifying a given potential ESP failure includes considering patterns of sensor readings for one or more sensors generating a wide range of consistent highs or consistent lows. ### Claim 5 (depends on 4) The computer-implemented method of claim 4, further comprising generating tree for a given sensor, the tree including a tree pattern of sensor values for the given sensor and acceptable limits for the sensor values. ### Claim 6 (depends on 5) The computer-implemented method of claim 5, wherein determining the early indication of an ESP failure includes determining an abnormal forest of trees based on a time-related existence of abnormal trees, wherein determining an abnormal forest of trees includes considering the acceptable limits of each sensor. ### Claim 7 (depends on 1) The computer-implemented method of claim 1, wherein the ESP failure is one of a burnt motor, a broken shaft, or one or more hydrogen sulfide (H2S) problems. ### Claim 8 A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising: receiving sensor data from a plurality of sensors, each sensor contained in an electrical submersible pump (ESP) deployed in a well hole; determining, in real time using a pattern recognition methodology based on the sensor data and a pattern detection model that correlates historical ESP failures with historical sensor readings, an early indication of an ESP failure that is imminent; and providing a notification of the imminence of the ESP failure. ### Claim 9 (depends on 8) The non-transitory, computer-readable medium of claim 8, wherein the sensor data includes motor temperature data, current data, vibration data, intake pressure data, and discharge pressure data. ### Claim 10 (depends on 8) The non-transitory, computer-readable medium of claim 8, the operations further comprising: training the pattern detection model to understand normal behaviors, abnormal behaviors, and critical events of each sensor, wherein the training is based on an analysis of a historical behavior of failed pumps, a historical sensor readings, a failure database, and a database of factors for different ESP types; classifying potential ESP failures under low warnings or high warnings based on a severity of an alarm received from a particular sensor of a particular ESP type, wherein the classifying a particular potential ESP failure is based on a single sensor or multiple sensors; and storing information for each potential ESP failure in the pattern detection model. ### Claim 11 (depends on 8) The non-transitory, computer-readable medium of claim 8, wherein classifying a given potential ESP failure includes considering patterns of sensor readings for one or more sensors generating a wide range of consistent highs or consistent lows. ### Claim 12 (depends on 11) The non-transitory, computer-readable medium of claim 11, the operations further comprising generating tree for a given sensor, the tree including a tree pattern of sensor values for the given sensor and acceptable limits for the sensor values. ### Claim 13 (depends on 12) The non-transitory, computer-readable medium of claim 12, wherein determining the early indication of an ESP failure includes determining an abnormal forest of trees based on a time-related existence of abnormal trees, wherein determining an abnormal forest of trees includes considering the acceptable limits of each sensor. ### Claim 14 (depends on 8) The non-transitory, computer-readable medium of claim 8, wherein the ESP failure is one of a burnt motor, a broken shaft, or one or more hydrogen sulfide (H2S) problems. ### Claim 15 A computer-implemented system, comprising: one or more processors; and a non-transitory computer-readable storage medium coupled to the one or more processors and storing programming instructions for execution by the one or more processors, the programming instructions instruct the one or more processors to perform operations comprising: receiving sensor data from a plurality of sensors, each sensor contained in an electrical submersible pump (ESP) deployed in a well hole; determining, in real time using a pattern recognition methodology based on the sensor data and a pattern detection model that correlates historical ESP failures with historical sensor readings, an early indication of an ESP failure that is imminent; and providing a notification of the imminence of the ESP failure. ### Claim 16 (depends on 15) The computer-implemented system of claim 15, wherein the sensor data includes motor temperature data, current data, vibration data, intake pressure data, and discharge pressure data. ### Claim 17 (depends on 15) The computer-implemented system of claim 15, the operations further comprising: training the pattern detection model to understand normal behaviors, abnormal behaviors, and critical events of each sensor, wherein the training is based on an analysis of a historical behavior of failed pumps, a historical sensor readings, a failure database, and a database of factors for different ESP types; classifying potential ESP failures under low warnings or high warnings based on a severity of an alarm received from a particular sensor of a particular ESP type, wherein the classifying a particular potential ESP failure is based on a single sensor or multiple sensors; and storing information for each potential ESP failure in the pattern detection model. ### Claim 18 (depends on 15) The computer-implemented system of claim 15, wherein classifying a given potential ESP failure includes considering patterns of sensor readings for one or more sensors generating a wide range of consistent highs or consistent lows. ### Claim 19 (depends on 18) The computer-implemented system of claim 18, the operations further comprising generating tree for a given sensor, the tree including a tree pattern of sensor values for the given sensor and acceptable limits for the sensor values. ### Claim 20 (depends on 19) The computer-implemented system of claim 19, wherein determining the early indication of an ESP failure includes determining an abnormal forest of trees based on a time-related existence of abnormal trees, wherein determining an abnormal forest of trees includes considering the acceptable limits of each sensor.