Sensor based system and method for drift analysis to predict equipment failure

• US 10,354,332 B2
• Filed: 09/30/2015
• Issued: 07/16/2019
• Est. Priority Date: 09/30/2015
• Status: Active Grant

• Current Assignee: Johnson Controls Tyco IP Holdings LLP (Johnson Controls International plc)
• Original Assignee: Sensormatic Electronics LLC (Johnson Controls International plc)
• Inventors: Trainor, Christine, Vavrasek, David
• Primary Examiner(s): Kazimi, Hani M
• Assistant Examiner(s): Ali, Hatem M
• Application Number:
  US14/870,551
• Publication Number:
  US 20170091870A1
• Time in Patent Office: 1,385 Days
• Field of Search: 705 4
• CPC Class Codes: G06Q 40/08 Insurance
• CPC Class Codes: G08B 13/08 by opening, e.g. of door, of window, of drawer, of shutter, of curtain, of blind
• CPC Class Codes: G08B 25/08 using communication transmission lines G08B13/19658, G08B21/0286, G08B25/016 take precedence
• CPC Class Codes: G08B 29/185 Signal analysis techniques for reducing or preventing false alarms or for enhancing the reliability of the system

Abstract

Techniques for detecting physical conditions at a physical premises from collection of sensor information from plural sensors execute one or more unsupervised learning models to continually analyze the collected sensor information to produce operational states of sensor information, produce sequences of state transitions, detect during the continual analysis of sensor data that one or more of the sequences of state transitions is a drift sequence, correlate determined drift state sequence to a stored determined condition at the premises, and generate an alert based on the determined condition. Various uses are described for these techniques.

Claims

1. A computer program product tangibly stored on a non-transitory computer readable hardware storage device, the computer program product for forming predictions of when equipment service is coming due or when equipment is about to fail, the computer program product comprising instructions to cause a processor to:

collect sensor information from plural sensors deployed in a premises with sensor information including a sensor data value, an identity of the premises, and identity of a physical item of equipment being monitored by the plural sensors in the identified premises;

convert the sensor signals into semantic representation of sensor states;

continually analyze the collected sensor information by applying one or more unsupervised learning models to the sensor states to determine normal sensor states and drift sensor states, with respect to equipment service of the physical item of equipment;

retrieve one or more predefined sensor state sequences that correspond to drift sensor states;

detect during the continual analysis of the sensor states by comparing the determined drift sensor states to the retrieved sensor state sequences, presence of a detected drift sensor state that corresponds to a state that equipment service is due or the physical item of equipment is about to fail in its operation;

determine a prediction based on the detected drift sensor state; and

send the prediction to an external device.

2. The computer program product of claim 1 further comprising determining an alert to send to the external device based on the prediction.

3. The computer program product of claim 2 wherein when the prediction is a failure prediction, the alert sent to the external device is to schedule a service call.

4. The computer program product of claim 1 wherein when the prediction is a failure prediction, the prediction is sent to a sensor-based underwriting module for modification of insurance rates.

5. A system comprises:

plural sensor devices installed at a premises;

a gateway to couple the plural sensors to a network;

a server computer comprising processor and memory, the sever computer coupled to the network;

a storage device storing a computer program product for forming predictions of when equipment service is coming due or when equipment is about to fail, the computer program product comprising instructions to cause a processor to;

collect sensor information from the plural sensors deployed in the premises with sensor information including a sensor data value, an identity of the premises, and identity of a physical item of equipment being monitored by the plural sensors in the identified premises;

convert the sensor signals into semantic representations of sensor states;

continually analyze the collected sensor information by applying one or more unsupervised learning models to the sensor states to determine normal sensor states and drift sensor states, with respect to equipment service of the physical item of equipment;

retrieve one or more predefined sensor state sequences that correspond to drift sensor states;

detect during the continual analysis of the sensor states by comparing the determined drift sensor states to the retrieved sensor state sequences, presence of a detected drift sensor state that corresponds to a state that equipment service is due or the physical item of equipment is about to fail in its operation;

determine a prediction based on the detected drift sensor state; and

send the prediction to an external device.

6. The system of claim 5 wherein the computer program product further comprises instructions to:

determine an alert to send to the external device based on the prediction.

7. The system of claim 6 wherein when the prediction is a failure prediction, the alert sent to the external device is to schedule a service call.

8. The system of claim 6 wherein when the prediction is a failure prediction, the prediction is sent to a sensor-based underwriting module for modification of insurance rates.

9. A computer implemented method comprises:

collecting from plural sensors that sense physical conditions, sensor information from the plural sensors deployed in a premises, with the sensor information including a sensor data value, an identity of the premises, and identity of a physical item of equipment being monitored by the plural sensors in the identified premises;

sending by a gateway the collected sensor data configured with an identity of the premises and identity of physical objects monitored by the sensors in the identified premises to one or more server computers that comprise processor devices and memory;

converting by the one or more server computers the sensor signals into semantic representation of sensor states;

continually analyzing by the one or more server computers the collected sensor information by applying one or more unsupervised learning models to the sensor states to determine normal sensor states and drift sensor states, with respect to equipment service of the physical item of equipment;

retrieving by the one or more server computers one or more predefined sensor state sequences that correspond to drift sensor states;

detecting by the one or more server computers during the continual analysis of the sensor states by comparing the determined drift sensor states to the retrieved sensor state sequences, presence of a detected drift sensor state that corresponds to a state that equipment service is due or the physical item of equipment is about to fail in its operation;

determining by the one or more server computers a prediction based on the detected drift sensor state; and

sending by the one or more server computers the prediction to an external device.

10. The method of claim 9 further comprising:

determining by the one or more server computers an alert to send to the external device based on the prediction.

11. The method of claim 10 wherein when the prediction is a failure prediction, the alert sent to the external device is to schedule a service call.

12. The method of claim 9 wherein when the prediction is a failure prediction, the prediction is sent to a sensor-based underwriting module for modification of insurance rates.