Method for non-intrusive load monitoring using a hybrid systems state estimation approach
A method for non-intrusively monitoring a load including a plurality of appliances includes retrieving a plurality of mathematical models for modeling operation of a respective subset of the appliances. A value of a respective operational parameter is predicted for each of the subsets of appliances. An output of the load is measured. A respective value of each of the operational parameters is calculated based on the mathematical models and the outputs of the load. The predicting, measuring and calculating steps are repeated until a metric pertaining to a difference error between the measured output and the predicted output calculated from the operational parameter and the mathematical models is equal to or below a threshold for one of the subsets of appliances. It is decided that the one subset of appliances is currently operating whose metric pertaining to a difference error is equal to or below the threshold.
1. A method for non-intrusively monitoring a load including a plurality of appliances, the method comprising:
retrieving a plurality of mathematical models, each of the models modeling operation of a respective subset of the appliances;
predicting a value of a respective operational parameter for each said subset of the appliances;
measuring an output of the load;
calculating a respective value of each of the operational parameters, the calculating being based on the mathematical models and the output of the load;
repeating the predicting, measuring and calculating steps until a metric pertaining to a difference error between the predicted value and the corresponding calculated value of the operational parameter is equal to or below a threshold for one said subset of the appliances; and
deciding that each said appliance in the one said subset of the appliances whose metric pertaining to the difference error is equal to or below the threshold is currently operating.
2. The method of claim 1 wherein the operational parameter comprises a continuous state vector.
3. The method of claim 1 wherein the output of the load comprises a level of power consumption of the load.
4. The method of claim 1 wherein each of the models functions to model operation of a respective one said subset of the appliances as having continuous dynamics.
5. The method of claim 1 wherein the models function to model operation of the appliances conjunctively as a discrete state system.
6. The method of claim 1 wherein the deciding step is dependent upon known probabilities associated with transitions between appliances or appliance stages and/or behavior of a user.
7. The method of claim 1 wherein the deciding step is dependent upon the metric pertaining to the difference error for the one said subset of appliances being below a threshold percentage of each of the metrics pertaining to the difference error of all other said subsets of appliances.
8. A method for non-intrusively monitoring a load including a plurality of appliances, the method comprising:
retrieving a plurality of mathematical models, each of the models modeling operation of a respective subset of the appliances as having continuous dynamics, the models modeling the appliances conjunctively as a discrete state system;
predicting a value of a respective operational parameter for each said subset of the appliances;
measuring an output of the load;
calculating a respective value of each of the operational parameters, the calculating being based on the mathematical models and the output of the load; and
deciding which of the subsets of appliances has most recently begun operating, the deciding being based upon the predicted and calculated values of the operational parameters of the subsets of appliances.
9. The method of claim 8 wherein the output of the load comprises a level of power consumption of the load.
10. The method of claim 8 wherein the deciding step is dependent upon known probabilities associated with transitions between appliances or appliance stages and/or behavior of a user.
11. The method of claim 8 comprising the further step of repeating the predicting, measuring and calculating steps until a metric pertaining to a difference error between the predicted value and the corresponding calculated value of the operational parameter is below a threshold for one of the subsets of appliances.
12. The method of claim 11 wherein the deciding step includes deciding that the one subset of appliances whose metric pertaining to a difference error is below the threshold is currently operating.
13. The method of claim 11 wherein the deciding step is dependent upon the metric pertaining to a difference error for the one subset of appliances being below a threshold percentage of each of the metrics pertaining to a difference error of all other said subsets of appliances.
14. The method of claim 8 wherein the operation of one of the appliances includes a plurality of phases, each of the phases being modeled by a respective said model.
15. A method for non-intrusively monitoring a load including a plurality of appliances, the method comprising:
obtaining a plurality of mathematical models, each of the models modeling operation of a respective subset of the appliances, each of the models being of the form:
ⅆ
x
n
ⅆ
t
=
A
n
x
n
(
t
)
+
B
n
u
(
t
)
y
=
C
n
x
n
(
t
)
+
D
n
u
(
t
)
wherein x n is a continuous state vector of an n th said subset of appliances, u(t) is a step voltage input to a corresponding said subset of appliances, y is an output of each discrete state of the load, and each of A n , B n , C n and D n is a coefficient corresponding to said n th subset of appliances;
predicting a value of x n for each of the subsets of appliances;
measuring a value of y;
calculating a respective value of x n for each of the subsets of appliances, the calculating being based on the mathematical models and the measured value of y; and
deciding which of the subsets of appliances has most recently begun operating, the deciding being based upon the predicted and calculated values of x n .
16. The method of claim 15 wherein the output of the discrete state of the load comprises a level of power consumption of the discrete state of the load.
17. The method of claim 15 wherein the deciding step is dependent upon known probabilities associated with transitions between appliances or appliance stages and/or behavior of a user.
18. The method of claim 15 comprising the further step of repeating the predicting, measuring and calculating steps until a difference error between the predicted value and the corresponding calculated value of x n is below a threshold for one of the subsets of appliances.
19. The method of claim 18 wherein the deciding step includes deciding that the one subset of appliances whose difference error is below the threshold is currently operating.
20. The method of claim 18 wherein the deciding step is dependent upon the difference error for the one subset of appliances being below a threshold percentage of each of the difference errors of all other said subsets of appliances.