Autonomous machine learning diagonostic system with simplified sensors for home appliances

1. A computer implemented method for remote performance monitoring of a ventilation system comprising:

accessing first sensor data from a first sensor placed on a return air grill for measuring a return temperature and a return humidity of air going into a return of a ventilation system for heating, cooling or a combination thereof;

accessing second sensor data from a second sensor placed on a supply air grill for measuring a supply temperature and a supply humidity of air coming out of the ventilation system;

determining an on-off duty cycle of an

on-time period during which the ventilation system is operating to change a temperature, humidity, or combination thereof, between the air as measured by the first sensor on the return air grill and the air as measured by the second sensor on the supply air grill, and

off-time period during which the ventilation system is not operating to change a temperature, humidity, or combination thereof, between the air as measured by the first sensor on the return air grill and the air as measured by the second sensor on the supply air grill;

creating a first training set of the on-off duty cycle;

training a machine learning algorithm with the first training set;

creating a second training set of return temperature, return humidity, supply temperature, supply humidity during each the on-off duty cycle;

training the machine learning algorithm with the second training set; and

using the machine learning algorithm from the first training set and the second training set, and the on-off duty cycle to predict that a component of the ventilation system is faulty.

2. The computer implemented method of claim 1 , wherein the determining an on-off duty is carried out independent of data from an electric meter measuring current consumed by the ventilation system.

3. The computer implemented method of claim 1 , wherein the determining an on-off duty is carried out independent of data from a ventilation thermostat which controls the ventilation system.

4. The computer implemented method of claim 1 , wherein the determining an on-off duty is carried out independent of data from an electric meter measuring current consumed by the ventilation system and wherein the determining an on-off duty is carried out independent of data from a ventilation thermostat which controls the ventilation system.

5. The computer implemented method of claim 1 , further comprising:

accessing outside air temperature data and outside air relative humidity corresponding to a geographic location in which the ventilation system is operating; and wherein the creating the second training set includes the outside air temperature data and the outside air relative humidity data.

6. The computer implemented method of claim 1 , further comprising:

using supervised machine learning with event models built from previously measured datasets of sensor data with two or more datasets of the first sensor data, the second sensor data and the on-off duty cycle to predict that a component of the ventilation system is faulty.

7. The computer implemented method of claim 6 , wherein the component is at least one of:

a compressor,

a blower,

a refrigerant charge;

a drain line float switch

a filter, and

a heating element.

8. A computer implemented method for remote performance monitoring of a ventilation system comprising:

accessing first sensor data from a first sensor placed on a return air grill for measuring a return temperature and a return humidity of air going into a return of the ventilation system for heating, cooling or a combination thereof;

accessing second sensor data from a second sensor placed on a supply air grill for measuring a supply temperature and a supply humidity of air coming out of the ventilation system;

using only data from the first sensor and data from the second sensor to determine an on-off duty cycle of an

on-time period during which the ventilation system is operating to change a temperature, humidity, or combination thereof, between the air as measured by the first sensor on the return air grill and the air as measured by the second sensor on the supply air grill, and

off-time period during which the ventilation system is not operating to change a temperature, humidity, or combination thereof, between the air as measured by the first sensor on the return air grill and the air as measured by the second sensor on the supply air grill; and

using supervised machine learning with event models built from previously measured datasets of sensor data with two or more datasets of the first sensor data, the second sensor data and the on-off duty cycle to predict that a component of the ventilation system is faulty.

9. The computer implemented method of claim 8 , further comprising:

accessing outside air temperature data and outside air relative humidity data corresponding to a geographic location in which the ventilation system is operating; and wherein the machine learning includes the outside temperature data and outside air relative humidity data.

10. The computer implemented method of claim 8 , wherein the component is at least one of:

a compressor,

a blower,

a refrigerant charge;

a drain line float switch

a filter, and

a heating element.

11. A system for remote performance monitoring of a ventilation system, the system comprising:

a computer memory capable of storing machine instructions; and

a hardware processor in communication with the computer memory, the hardware processor configured to access the computer memory to execute the machine instructions to perform

accessing first sensor data from a first sensor placed on a return air grill for measuring a return temperature and a return humidity of air going into a return of the ventilation system for heating, cooling or a combination thereof;

accessing second sensor data from a second sensor placed on a supply air grill for measuring a supply temperature and a supply humidity of air coming out of the ventilation system;

using data from the first sensor and data from the second sensor to determine an on-off duty cycle of an

on-time period during which the ventilation system is operating to change a temperature, humidity, or combination thereof, between the air as measured by the first sensor on the return air grill and the air as measured by the second sensor on the supply air grill, and

off-time period during which the ventilation system is not operating to change a temperature, humidity, or combination thereof, between the air as measured by the first sensor on the return air grill and the air as measured by the second sensor on the supply air grill; and

using machine learning with two or more datasets of the first sensor data, the second sensor data and the on-off duty cycle to indicating a component of the ventilation system is faulty.

12. The system of claim 11 , wherein the using data from the first sensor and data from the second sensor to determine an on-off duty cycle is carried out independent of data from an electric meter measuring current consumed by the ventilation system.

13. The system of claim 11 , wherein the using data from the first sensor and data from the second sensor to determine an on-off duty cycle is carried out independent of data from a ventilation thermostat which controls the ventilation system.

14. The system of claim 11 , further comprising:

accessing outside air temperature data and outside air relative humidity corresponding to a geographic location in which the ventilation system is operating; and wherein the using machine learning with two or more datasets of the first sensor data, the second sensor data and the on-off duty cycle includes the outside air temperature data and the outside air relative humidity data.

15. The system of claim 11 , further comprising:

accessing outside air temperature data and outside air relative humidity data corresponding to a geographic location in which the ventilation system is operating; and wherein the machine learning includes the outside temperature data and outside air relative humidity data.

16. The system of claim 15 , wherein the component is at least one of:

a compressor,

a blower,

a refrigerant charge;

a drain line float switch

a filter, and

a heating element.

17. The system of claim 10 , further including:

a first wireless temperature and humidity sensor for supplying the first sensor data; and

a second wireless temperature and humidity sensor for supplying the second sensor data.

18. A system for remote performance monitoring of a ventilation system, the system comprising:

a first wireless sensor placed on a return air grill for measuring a return temperature and a return humidity of air going into a return of the ventilation system for heating, cooling or a combination thereof, to provide supply temperature and humidity data to an autonomous fault detection computer system; and

a second wireless sensor placed on a supply air grill for measuring a supply temperature and a supply humidity of air coming out of the ventilation system to provide supply temperature and humidity data to the computer system,

wherein the computer system determines an on-off duty cycle of an

on-time period during which the ventilation system is operating to change a temperature, humidity, or combination thereof, between the air as measured by the first wireless sensor on the return air grill and the air as measured by the second wireless sensor on the supply air grill, and

off-time period during which the ventilation system is not operating to change a temperature, humidity, or combination thereof, between the air as measured by the first wireless sensor on the return air grill and the air as measured by the second wireless sensor on the supply air grill; and

using machine learning with two or more datasets of the data from the first wireless sensor, data from the second wireless sensor, and the on-off duty cycle to predict a component of the ventilation system is faulty.

19. The system of claim 18 , wherein the using data from the first wireless sensor and data from the second wireless sensor to determine an on-off duty cycle is carried out independent of data from an electric meter measuring current consumed by the ventilation system.

20. The system of claim 19 , wherein the using data from the first wireless sensor and data from the second wireless sensor to determine an on-off duty cycle is carried out independent of data from a ventilation thermostat which controls the ventilation system.