HVAC analytics

1. A monitoring system that is configured to monitor a property, the monitoring system comprising:

one or more sensors that are located throughout the property and that are configured to generate sensor data;

a thermostat that is configured to control an HVAC system located at the property and that is configured to generate thermostat data;

a monitor control unit that is configured to:

receive, from the thermostat, thermostat data that indicates activity of the HVAC system and temperature history of the property;

receive, from the one or more sensors, the sensor data;

apply, to an HVAC model that is trained using past sensor data, past thermostat data, past errors of the HVAC system, and past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data;

based on applying the thermostat data and the sensor data to the HVAC model, determine an error of the HVAC system;

based on the error of the HVAC system, determine an action for correcting the error of the HVAC system;

provide, for output, data identifying the error of the HVAC system and the action for correcting the error of the HVAC system;

receive data indicating completion of the action; and

in response to receiving the data indicating completion of the action, obtain additional thermostat information by performing a test of the HVAC system,

wherein the monitor control unit is configured to:

based on determining the error of the HVAC system, determine an HVAC error classification that specifies a type of the error;

determine the action for correcting the error of the HVAC system by determining the action for correcting the error of the HVAC system based on the HVAC error classification that specifies the type of the error,

wherein the type of the error of the HVAC error classification comprises a heating vs. cooling category indicating an issue with the HVAC system not properly heating or cooling the property; an acute vs. inefficient category indicating an issue with one or more components of the HVAC system not working at a predetermined efficiency; a filter change category indicating an issue with an air quality in the property; or a long cycling category indicating an issue with a run time of the HVAC system.

2. The system of claim 1 , wherein the monitor control unit is configured to:

receive the sensor data by receiving, from a motion sensor that monitors motion at the property, motion data;

apply, to the HVAC model that is trained using the past sensor data, the past thermostat data, the past errors of the HVAC system, and the past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data by applying, to the HVAC model, the thermostat data and the motion data; and

determine the error of the HVAC system by determining the error of the HVAC system based on applying the thermostat data and the motion data to the HVAC model.

3. The system of claim 1 , wherein the monitor control unit is configured to:

receive the sensor data by receiving, from a door or window sensor that monitors whether a door or a window is open or closed, door or window sensor data;

apply, to the HVAC model that is trained using the past sensor data, the past thermostat data, the past errors of the HVAC system, and the past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data by applying, to the HVAC model, the thermostat data and the door or window sensor data; and

determine the error of the HVAC system by determining the error of the HVAC system based on applying the thermostat data and the door or window sensor data to the HVAC model.

4. The system of claim 1 , wherein the monitor control unit is further configured to:

receive timestamp data for the sensor data and the thermostat data; and

apply, to the HVAC model that is trained using the past sensor data, the past thermostat data, the past errors of the HVAC system, and the past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data by applying, to the HVAC model that is trained using the past sensor data collected during a time of day corresponding to the timestamp data, the past thermostat data collected during a time of day corresponding to the timestamp data, the past errors of the HVAC system, and the past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data.

5. The system of claim 1 , comprising:

a monitoring server that is configured to:

receive, from other monitoring systems that are configured to monitor other properties, the past sensor data of each monitoring system, the past thermostat data of each monitoring system, the past errors of each HVAC system of each property, and past actions that corrected the errors of each HVAC system of each property;

train, using machine learning, the HVAC model using the past sensor data of each monitoring system, the past thermostat data of each monitoring system, the past errors of each HVAC system of each property, and past actions that corrected the errors of each HVAC system of each property; and

provide, for output to the monitor control unit, the HVAC model.

6. The system of claim 5 , wherein the past errors of each HVAC system of each property and the error of the HVAC system are a same error.

7. The system of claim 5 , wherein the HVAC system and the HVAC system of the other properties are a same model type.

8. The system of claim 5 , wherein the HVAC system and the HVAC system of the other properties comprise different model types.

9. The system of claim 1 , wherein the monitor control unit is configured to:

provide, for output, data identifying the error of the HVAC system and the action for correcting the error of the HVAC system, the error comprising at least one of a broken thermostat, unresponsive burner, unresponsive compressor, lack of refrigerant, broken blower, broken evaporator, or a low battery; and

provide, for output, data instructing a user to shut off the HVAC system.

10. The system of claim 1 , wherein the monitor control unit is further configured to perform the test by:

determining a first amount of time for the HVAC system to raise a temperature of the property by a temperature amount and a second amount of time to lower the temperature of the property by the temperature amount;

determining that the first amount of time satisfies a first time threshold;

determining that the second amount of time satisfies a second time threshold; and

in response to determining that the first amount of time satisfies the first time threshold and determining that the second amount of time satisfies the second time threshold, determining the error of the HVAC system has been fixed.

11. The system of claim 1 , wherein the monitor control unit is further configured to perform the test by:

determine a difference between a temperature of the property and a desired temperature of the property within a predetermined period;

determine that the difference between the temperature of the property and the desired temperature of the property satisfies a temperature threshold within the predetermined period; and

in response to determining that the difference between the temperature of the property and the desired temperature of the property satisfies the temperature threshold within the predetermined period, determine the error of the HVAC system has been fixed.

12. A computer-implemented method, comprising:

receiving, by a monitoring system that is configured to monitor a property and from a thermostat that is configured to control an HVAC system located at the property, thermostat data that indicates activity of the HVAC system and temperature history of the property;

receiving, by the monitoring system and from one or more sensors that are located throughout the property, sensor data;

applying, by the monitoring system and to an HVAC model that is trained using past sensor data, past thermostat data, past errors of the HVAC system, and past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data;

based on applying the thermostat data and the sensor data to the HVAC model, determining, by the monitoring system, an error of the HVAC system;

based on determining the error of the HVAC system, determining an HVAC error classification that specifies a type of the error;

based on the error of the HVAC system, determining, by the monitoring system, an action for correcting the error of the HVAC system;

providing, for output by the monitoring system, data identifying the error of the HVAC system and the action for correcting the error of the HVAC system;

receiving data indicating completion of the action; and

in response to receiving the data indicating completion of the action, obtaining additional thermostat information by performing a test of the HVAC system,

wherein determining the action for correcting the error of the HVAC system comprises determining the action for correcting the error of the HVAC system based on the HVAC error classification that specifies the type of the error, and

wherein the type of the error of the HVAC error classification comprises a heating vs. cooling category indicating an issue with the HVAC system not properly heating or cooling the property; an acute vs. inefficient category indicating an issue with one or more components of the HVAC system not working at a predetermined efficiency; a filter change category indicating an issue with an air quality in the property; or a long cycling category indicating an issue with a run time of the HVAC system.

13. The method of claim 12 , wherein:

receiving the sensor data comprises receiving, from a motion sensor that monitors motion at the property, motion data;

applying, to the HVAC model that is trained using the past sensor data, the past thermostat data, the past errors of the HVAC system, and the past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data comprises applying, to the HVAC model, the thermostat data and the motion data; and

determining the error of the HVAC system comprises determining the error of the HVAC system based on applying the thermostat data and the motion data to the HVAC model.

14. The method of claim 12 , wherein:

receiving the sensor data comprises receiving, from a door or window sensor that monitors whether a door or a window is open or closed, door or window sensor data;

applying, to the HVAC model that is trained using the past sensor data, the past thermostat data, the past errors of the HVAC system, and the past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data comprises applying, to the HVAC model, the thermostat data and the door or window sensor data; and

determining the error of the HVAC system comprises determining the error of the HVAC system based on applying the thermostat data and the door or window sensor data to the HVAC model.

15. The method of claim 12 , further comprising:

receiving timestamp data for the sensor data and the thermostat data;

wherein applying, to the HVAC model that is trained using the past sensor data, the past thermostat data, the past errors of the HVAC system, and the past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data comprises applying, to the HVAC model that is trained using the past sensor data collected during a time of day corresponding to the timestamp data, the past thermostat data collected during a time of day corresponding to the timestamp data, the past errors of the HVAC system, and the past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data.

16. A monitoring server, comprising:

one or more computers; and

one or more storage devices storing instructions that, when executed by the one or more computers, cause the one or more computers to perform operations comprising:

receiving, from a thermostat that is configured to control an HVAC system located at the property, thermostat data that indicates activity of the HVAC system and temperature history of the property;

receiving, from one or more sensors that are located throughout the property, sensor data;

applying, to an HVAC model that is trained using past sensor data, past thermostat data, past errors of the HVAC system, and past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data;

based on applying the thermostat data and the sensor data to the HVAC model, determining an error of the HVAC system;

based on determining the error of the HVAC system, determining an HVAC error classification that specifies a type of the error;

based on the error of the HVAC system, determining an action for correcting the error of the HVAC system;

providing, for output, data identifying the error of the HVAC system and the action for correcting the error of the HVAC system;

receiving data indicating completion of the action; and

in response to receiving the data indicating completion of the action, obtaining additional thermostat information by performing a test of the HVAC system,

wherein determining the action for correcting the error of the HVAC system comprises determining the action for correcting the error of the HVAC system based on the HVAC error classification that specifies the type of the error, and

wherein the type of the error of the HVAC error classification comprises a heating vs. cooling category indicating an issue with the HVAC system not properly heating or cooling the property; an acute vs. inefficient category indicating an issue with one or more components of the HVAC system not working at a predetermined efficiency; a filter change category indicating an issue with an air quality in the property; or a long cycling category indicating an issue with a run time of the HVAC system.

17. The monitoring server of claim 16 , wherein:

receiving the sensor data comprises receiving, from a motion sensor that monitors motion at the property, motion data;

applying, to the HVAC model that is trained using the past sensor data, the past thermostat data, the past errors of the HVAC system, and the past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data comprises applying, to the HVAC model, the thermostat data and the motion data; and

determining the error of the HVAC system comprises determining the error of the HVAC system based on applying the thermostat data and the motion data to the HVAC model.

18. The monitoring server of claim 16 , wherein:

receiving the sensor data comprises receiving, from a door or window sensor that monitors whether a door or a window is open or closed, door or window sensor data;

applying, to the HVAC model that is trained using the past sensor data, the past thermostat data, the past errors of the HVAC system, and the past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data comprises applying, to the HVAC model, the thermostat data and the door or window sensor data; and

determining the error of the HVAC system comprises determining the error of the HVAC system based on applying the thermostat data and the door or window sensor data to the HVAC model.

19. The monitoring server of claim 16 :

wherein the operations comprise receiving timestamp data for the sensor data and the thermostat data; and

wherein applying, to the HVAC model that is trained using the past sensor data, the past thermostat data, the past errors of the HVAC system, and the past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data comprises applying, to the HVAC model that is trained using the past sensor data collected during a time of day corresponding to the timestamp data, the past thermostat data collected during a time of day corresponding to the timestamp data, the past errors of the HVAC system, and the past actions that corrected the errors of the HVAC system, the thermostat data and the sensor data.

20. The monitoring server of claim 16 , wherein the operations comprise:

receiving, from other monitoring systems that are configured to monitor other properties, the past sensor data of each monitoring system, the past thermostat data of each monitoring system, the past errors of each HVAC system of each property, and the past actions that corrected the errors of each HVAC system of each property; and

training, using machine learning, the HVAC model using the past sensor data of each monitoring system, the past thermostat data of each monitoring system, the past errors of each HVAC system of each property, and the past actions that corrected the errors of each HVAC system of each property.