Dynamic demand response event assessment

1. A method of demand response (DR) event issue assessment, the method comprising:

receiving parameters that include one or more of contractual parameters, ambient condition data, historical data, and energy price data related to two or more sites;

predicting a customer demand and a customer energy curtailment based on one or more of the parameters;

calculating a customer participation likelihood based on one or more of the parameters;

determining a price threshold based on one or more of the parameters, the customer energy curtailment, the customer demand, and the customer participation likelihood;

comparing the price threshold to a predicted energy price;

responsive to the price threshold being greater than the predicted energy price, implementing a DR event, the implementing the DR event including broadcasting DR event criteria to appliances at the sites that include commands that turn the appliances off or on according to a control strategy of the DR event,

wherein:

the price threshold represents an energy price at which issuing a DR event is more profitable than not issuing the DR event; and

the determining the price threshold is performed according to an optimization equation:

min( P iTH )|{ E[R DR ({right arrow over ( A i )})−Cost DR ({right arrow over ( A i )}, P iTH )−Inc DR ({right arrow over ( A i )})]≥ E[R −Cost( P iTH )]}, in which:

min( )|{ } represents an operation to determine a minimum value of a variable in the parentheses ( ) such that a condition in the brackets { } is satisfied;

E[ ] represents an expected value;

P iTH represents a price threshold;

R DR ( ) represents a DR event revenue;

Cost DR ( ) represents a DR event cost;

Inc DR ( ) represents an incentive paid to participate in a DR event;

R represents a non-DR event revenue;

Cost( ) represents a non-DR event cost;

i represents an interval in which a price threshold is determined; and

{right arrow over (A i )} represents a random variable that denotes the customer participation likelihood at an interval i.

2. The method of claim 1 , further comprising predicting an energy price for a particular time period.

3. The method of claim 1 , wherein:

the ambient condition data includes a temperature-humidity index (THI) and the historical data includes historical energy usage data of one or more customers;

the customer demand is predicted as a function of the THI; and

the customer energy curtailment is predicted as a function of the THI and a control strategy.

4. The method of claim 1 , wherein the customer participation likelihood includes a random variable with a particular probability distribution based on the historical data.

5. The method of claim 1 , further comprising:

estimating a risk of missing a future DR event opportunity;

determining whether the risk is acceptable; and

broadcasting the DR event in response to the risk being acceptable.

6. The method of claim 5 , wherein the risk is estimated according to an equation:

risk

=

T

-

E

i

T

×

α

;

and wherein:

risk represents an estimated risk;

T represents a number of future days that an energy price is expected to be more than a price threshold, which is predicted at least partially from the historical data;

α represents a reliability of the predicted number of future days that the energy price is expected to be more than the price threshold; and

E i represents a number of future events indicator.

7. The method of claim 1 , further comprising generating a table of price thresholds relative to temperature-humidity index (THI) and control strategy.

8. The method of claim 1 , further comprising estimating a profit if the DR event is issued according to one or more of equations:

⁢

savings

⁢

⁢

%

=

E

⁡

[

R

DR

⁡

(

A

i

→

)

-

Cost

DR

⁡

(

A

i

→

,

P

R

)

-

Inc

DR

⁡

(

A

i

→

)

]

-

E

⁡

[

R

-

Cost

⁡

(

P

R

)

]

E

⁡

[

R

-

Cost

⁡

(

P

R

)

]

×

100

;

⁢

and

savings

⁢

⁢

$

=

E

⁡

[

R

DR

⁡

(

A

i

→

)

-

Cost

DR

⁡

(

A

i

→

,

P

R

)

-

Inc

DR

⁡

(

A

i

→

)

]

-

E

⁡

[

R

-

Cost

⁡

(

P

R

)

]

,

and wherein

P R represents a predicted energy price;

savings % represents an estimated percentage savings; and

savings$ represents an estimated net savings.

9. The method of claim 1 , wherein:

the DR event revenue is calculated according to a first equation:

R

DR

⁡

(

A

i

→

)

=

∑

j

⁢

∑

c

⁢

R

⁡

(

c

)

⁢

A

i

⁡

(

c

,

j

)

⁡

[

D

i

⁡

(

c

,

THI

)

-

K

i

⁡

(

c

,

THI

,

s

)

]

+

∑

j

⁢

∑

c

⁢

R

⁡

(

c

)

⁡

[

1

-

A

i

⁡

(

c

,

j

)

]

⁢

D

i

⁡

(

c

,

THI

)

;

the DR event cost is calculated according to a second equation:

Cost

DR

⁡

(

A

i

→

,

P

iTH

)

=

∑

j

⁢

∑

c

⁢

P

iTH

⁢

A

i

⁡

(

c

,

j

)

⁡

[

D

i

⁡

(

c

,

THI

)

-

K

i

⁡

(

c

,

THI

,

s

)

]

+

∑

j

⁢

∑

c

⁢

P

iTH

⁡

[

1

-

A

i

⁡

(

c

,

j

)

]

⁢

D

i

⁡

(

c

,

THI

)

+

∑

j

⁢

∑

c

⁢

Cost

DR

-

Op

⁡

(

c

)

⁢

A

i

⁡

(

c

,

j

)

;

the non-DR event revenue is calculated according to a third equation:

R

=

E

i

⁢

∑

c

⁢

R

⁡

(

c

)

⁢

D

i

⁡

(

c

,

THI

)

;

the non-DR event cost is calculated according to a fourth equation:

Cost

⁡

(

P

iTH

)

=

E

i

⁢

∑

c

⁢

P

iTH

⁢

D

i

⁡

(

c

,

THI

)

;

and wherein

c represents an indexing variable related to customers;

j represents a second indexing variable designating future DR events;

R represents non-DR event revenue;

R(c) represents an energy price for a specific customer identified by the indexing variable c;

D i ( ) represents a predicted customer demand at an interval designated by a variable i;

K i ( ) represents the predicted customer energy curtailment at an interval designated by a variable i;

THI represents a temperature-humidity index (THI);

s represents a control strategy;

E i represents a number of future events indicator;

Cost DR-OP ( ) represents operational costs for issuing a DR event;

A i (c, j) represents a probability that a specific customer assigned an indexing variable c will participate in a next j events; and

1-A i (c, j) represents a probability that a specific customer assigned an indexing variable c will not participate in the next j events.

10. A non-transitory computer-readable medium having encoded therein programming code executable by a processor to perform operations comprising:

receiving parameters that include one or more of contractual parameters, ambient condition data, historical data, and energy price data;

predicting a customer demand and a customer energy curtailment based on one or more of the parameters related to two or more sites;

calculating a customer participation likelihood based on one or more of the parameters;

determining a price threshold based on one or more of the parameters, the customer energy curtailment, the customer demand, and the customer participation likelihood;

comparing the price threshold to a predicted energy price;

responsive to the price threshold being greater than the predicted energy price, implementing a DR event, the implementing the DR event including broadcasting DR event criteria to appliances at the sites that include commands that turn the appliances off or on according to a control strategy of the DR event,

wherein:

the price threshold represents an energy price at which issuing a DR event is more profitable than not issuing the DR event; and

the determining the price threshold is performed according to an optimization equation:

min( P iTH )|{ E[R DR ({right arrow over ( A i )})−Cost DR ({right arrow over ( A i )}, P iTH )−Inc DR ({right arrow over ( A i )})]≥ E[R −Cost( P iTH )]}, in which:

min( )|{ } represents an operation to determine a minimum value of a variable in the parentheses ( ) such that a condition in the brackets { } is satisfied;

E[ ] represents an expected value;

P iTH represents a price threshold;

R DR ( ) represents a DR event revenue;

Cost DR ( ) represents a DR event cost;

Inc DR ( ) represents an incentive paid to participate in a DR event;

R represents a non-DR event revenue;

Cost( ) represents a non-DR event cost;

i represents an interval in which a price threshold is determined; and

{right arrow over (A i )} represents a random variable that denotes the customer participation likelihood at an interval i.

11. The non-transitory computer-readable medium of claim 10 , wherein the operations further comprise predicting an energy price for a particular time period.

12. The non-transitory computer-readable medium of claim 10 , wherein:

the ambient condition data includes a temperature-humidity index (THI) and the historical data includes historical energy usage data of one or more customers;

the customer demand is predicted as a function of the THI; and

the customer energy curtailment is predicted as a function of the THI and a control strategy.

13. The non-transitory computer-readable medium of claim 10 , wherein the customer participation likelihood includes a random variable with a particular probability distribution based on the historical data.

14. The non-transitory computer-readable medium of claim 10 , further comprising:

estimating a risk of missing a future DR event opportunity;

determining whether the risk is acceptable; and

broadcasting the DR event in response to the risk being acceptable.

15. The non-transitory computer-readable medium of claim 14 , wherein the risk is estimated according to an equation:

risk

=

T

-

E

i

T

×

α

;

and wherein:

risk represents an estimated risk;

T represents a number of future days that an energy price is expected to be more than a price threshold, which is predicted at least partially from the historical data;

α represents a reliability of the predicted number of future days that the energy price is expected to be more than the price threshold; and

E i represents a number of future events indicator.

16. The non-transitory computer-readable medium of claim 10 , wherein the operations further comprise generating a table of price thresholds relative to temperature-humidity index (THI) and control strategy.

17. The non-transitory computer-readable medium of claim 10 , wherein the operations further comprise estimating a profit if the DR event is issued according to one or more of equations:

⁢

savings

⁢

⁢

%

=

E

⁡

[

R

DR

⁡

(

A

i

→

)

-

Cost

DR

⁡

(

A

i

→

,

P

R

)

-

Inc

DR

⁡

(

A

i

→

)

]

-

E

⁡

[

R

-

Cost

⁡

(

P

R

)

]

E

⁡

[

R

-

Cost

⁡

(

P

R

)

]

×

100

;

⁢

and

savings

⁢

⁢

$

=

E

⁡

[

R

DR

⁡

(

A

i

→

)

-

Cost

DR

⁡

(

A

i

→

,

P

R

)

-

Inc

DR

⁡

(

A

i

→

)

]

-

E

⁡

[

R

-

Cost

⁡

(

P

R

)

]

,

and wherein

P R represents a predicted energy price;

savings % represents an estimated percentage savings; and

savings$ represents an estimated net savings.

18. The non-transitory computer-readable medium of claim 10 , wherein:

the DR event revenue is calculated according to a first equation:

R

DR

⁡

(

A

i

→

)

=

∑

j

⁢

∑

c

⁢

R

⁡

(

c

)

⁢

A

i

⁡

(

c

,

j

)

⁡

[

D

i

⁡

(

c

,

THI

)

-

K

i

⁡

(

c

,

THI

,

s

)

]

+

∑

j

⁢

∑

c

⁢

R

⁡

(

c

)

⁡

[

1

-

A

i

⁡

(

c

,

j

)

]

⁢

D

i

⁡

(

c

,

THI

)

;

the DR event cost is calculated according to a second equation:

Cost

DR

⁡

(

A

i

→

,

P

iTH

)

=

∑

j

⁢

∑

c

⁢

P

iTH

⁢

A

i

⁡

(

c

,

j

)

⁡

[

D

i

⁡

(

c

,

THI

)

-

K

i

⁡

(

c

,

THI

,

s

)

]

+

∑

j

⁢

∑

c

⁢

P

iTH

⁡

[

1

-

A

i

⁡

(

c

,

j

)

]

⁢

D

i

⁡

(

c

,

THI

)

+

∑

j

⁢

∑

c

⁢

Cost

DR

-

Op

⁡

(

c

)

⁢

A

i

⁡

(

c

,

j

)

;

the non-DR event revenue is calculated according to a third equation:

R

=

E

i

⁢

∑

c

⁢

R

⁡

(

c

)

⁢

D

i

⁡

(

c

,

THI

)

;

the non-DR event cost is calculated according to a fourth equation:

Cost

⁡

(

P

iTH

)

=

E

i

⁢

∑

c

⁢

P

iTH

⁢

D

i

⁡

(

c

,

THI

)

;

and wherein:

c represents an indexing variable related to customers;

j represents a second indexing variable designating future DR events;

R represents non-DR event revenue;

R(c) represents an energy price for a specific customer identified by the indexing variable c;

D i ( ) represents a predicted customer demand at an interval designated by a variable i;

K i ( ) represents the predicted customer energy curtailment at an interval designated by a variable i;

THI represents a temperature-humidity index (THI);

s represents a control strategy;

E i represents a number of future events indicator;

Cost DR-OP ( ) represents operational costs for issuing a DR event;

A i (c, j) represents a probability that a specific customer assigned an indexing variable c will participate in a next j events; and

1-A i (c, j) represents a probability that a specific customer assigned an indexing variable c will not participate in the next j events.