IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0771570
(2004-02-04)
|
등록번호 |
US-8214271
(2012-07-03)
|
발명자
/ 주소 |
- Lefebvre, W. Curt
- Kohn, Daniel W.
|
출원인 / 주소 |
|
대리인 / 주소 |
|
인용정보 |
피인용 횟수 :
2 인용 특허 :
37 |
초록
▼
Various embodiments of the invention are directed to methods and systems for assigning credit to the external and intermediate inputs of an enterprise-level system or other aggregate process with one or more global outputs that is composed of a number of interconnected local processes. Assigning cre
Various embodiments of the invention are directed to methods and systems for assigning credit to the external and intermediate inputs of an enterprise-level system or other aggregate process with one or more global outputs that is composed of a number of interconnected local processes. Assigning credit is a mechanism for evaluating the impact of a particular variable, e.g., an input to one of the local processes, on the final output of the aggregate process. In certain embodiments of the invention, credit is assigned to the local inputs of each local process in two steps. First, for each local input, a local credit assignment is obtained. For chained outputs, credit assignment data is provided as calculated for the later stage processes to which the outputs are chained. Second, a global credit assignment is derived for the local inputs from the local credit assignment and the credit assignment information from later stage processes.
대표청구항
▼
1. A method for assigning a profit contribution value to a first input of a first process for operation of the fossil fuel power plant in a network comprised of a plurality of interconnected process management modules respectively associated with processes for operation of a fossil fuel power plant,
1. A method for assigning a profit contribution value to a first input of a first process for operation of the fossil fuel power plant in a network comprised of a plurality of interconnected process management modules respectively associated with processes for operation of a fossil fuel power plant, the profit contribution value indicative of a contribution of the first input to a global output of the network, wherein the global output is a profit generated by operation of the fossil fuel power plant, the first process having a plurality of inputs and outputs, at least one of said outputs of the first process being a chained output that is an input to a second process for operation of the fossil fuel power plant and contributes to the global output of the network, the method comprising: obtaining profit contribution values for assignment to each of the chained outputs of the first process for operation of the fossil fuel power plant with respect to the global output, wherein the profit contribution values assigned to each of the chained outputs of the first process are a measure of the contribution of the chained outputs on the global output;using a model-based controller having a first-order differentiabte model of the first process to derive a local contribution value for assignment to the first input of the first process, wherein the local contribution value assigned to the first input is a measure of the contribution of the first input on outputs of the first process; and using a local processor to apply a chain rule for ordered partial derivatives using (a) the first-order differentiable model of the first process, (b) the local contribution value assigned to the first input, and (c) the profit contribution values assigned to the chained outputs of the first process in order to assign the profit contribution value to the first input of the first process. 2. The method of claim 1, wherein the first-order differentiable model is a neural network. 3. The method of claim 1, wherein the first-order differentiable model is a first-principles model. 4. The method of claim 1, wherein the method includes managing the first process using a first process management module and determining the profit contribution value assigned to the first input using the first process management module. 5. The method of claim 1, further comprising: managing the first process using a first process management module;transmitting the local contribution value over the network, from the first process management module, to a second process management module, wherein the second process management module computes the profit contribution value assigned to the first input. 6. A computer program product stored on a computer readable medium for use in analyzing a first process for operation of a fossil fuel power plant, the first process having a plurality of inputs and at least one output, at least one of said outputs being a chained output that is an input to a second process in a network of process management modules respectively associated with processes for operation of the fossil fuel power plant, and contributes to a global output of the network, wherein the global output is a profit generated by operation of the fossil fuel power plant, the computer program product containing instructions for causing a computer to: obtain profit contribution values for assignment to each of the chained outputs of the first process for operation of the fossil fuel power plant with respect to the global output using an application program interface, wherein the profit contribution values assigned to each of the chained outputs of the first process are a measure of the contribution of the chained outputs on the global output;obtain a first-order-differentiable model of the first process; andapply a chain rule for ordered partial derivatives to the first-order-differentiable model using the profit contribution values assigned to each of the chained outputs of the first process to determine a profit contribution value assigned to a first input of the first process with respect to the global output of the network, wherein the profit contribution value assigned to the first input is a measure of the contribution of the first input on the global output. 7. The computer program product of claim 6, wherein the first-order-differentiable model is a neural network. 8. The computer program product of claim 6, wherein the first-order-differentiable model is a first-principles model. 9. The computer program product of claim 6, wherein the first-order-differentiable model is changed due to (a) a change in operating region of the first process, (b) retraining of the model, or (c) a physical change in the first process. 10. The method of claim 1, wherein said first and second processes for operation of the fossil fuel power plant are selected from the group consisting of the following processes: combustion optimization, sootblowing optimization, boiler performance optimization, selective catalytic reduction (SCR) optimization, flue gas desulfurization (FGD) optimization, and profit optimization. 11. The method of claim 1, wherein the first process is combustion optimization, said first input of the first process is selected from the group consisting of: O2 trim, over fire air (OFA), mill biases, SAD, and cleanliness; and an output of the first process is selected from the group consisting of: boiler losses, boiler NOx and boiler SOx. 12. The method of claim 1, wherein the first process is sootblowing optimization, said first input of the first process is selected from the group consisting of: location, pressure and frequency of sootblowing operations; and an output of the first process is selected from the group consisting of soot losses and cleanliness. 13. The method of claim 1, wherein the first process is SCR optimization, said first input of the first process is selected from the group consisting of: boiler NOx and NH3; and an output of the first process is selected from the group consisting of: SCR losses and NOx. 14. The method of claim 1, wherein the first process is FGD optimization, said first input of the first process is selected from the group consisting of: boiler SOx and limestone; and an output of the first process is selected from the group consisting of: FGD losses and SOx. 15. The method of claim 1, wherein the first process is boiler performance optimization, said first input of the first process is selected from the group consisting of: soot losses, cleanliness, boiler losses, SCR losses and FGD losses; and an output of the first process is selected from the group consisting of: heat rate (HR) and MW. 16. The method of claim 1, wherein the method includes managing the first process using a first process management module, the first management module selected from the group consisting of: a module for optimizing combustion; a module for optimizing sootblowing; a module for optimizing boiler performance; a module for optimizing selective catalytic reduction (SCR); and a module for optimizing flue gas desulfurization (FGD). 17. The method of claim 1, wherein said processes of the process management modules include a third process having a plurality of inputs and an output that is said global output of the network, wherein the third process is profit optimization. 18. The method of claim 17, wherein an input of said third process is selected from the group consisting of: heat rate (HR), MW, NOx, NH3, SO, limestone, emission credits, and fuel costs. 19. The computer program product of claim 6, wherein said first and second processes of the fossil fuel power plant are selected from the group consisting of the following processes: combustion optimization, sootblowing optimization, boiler performance optimization, selective catalytic reduction (SCR) optimization, flue gas desulfurization (FGD) optimization, and profit optimization. 20. The computer program product of claim 6, wherein the first process is combustion optimization, said first input of the first process is selected from the group consisting of: O2 trim, over fire air (OFA), mill biases, SAD, and cleanliness; and an output of the first process is selected from the group consisting of: boiler losses, boiler NOx and boiler SOx. 21. The computer program product of claim 6, wherein the first process is sootblowing optimization, said first input of the first process is selected from the group consisting of: location, pressure and frequency of sootblowing operations; and an output of the first process is selected from the group consisting of: soot losses and cleanliness. 22. The computer program product of claim 6, wherein the first process is SCR optimization, said first input of the first process is selected from the group consisting of: boiler NOx and NH3; and an output of the first process is selected from the group consisting of: SCR losses and NOx. 23. The computer program product of claim 6, wherein the first process is FGD optimization, said first input of the first process is selected from the group consisting of: boiler SOx and limestone; and an output of the first process is selected from the group consisting of: FGD losses and SOx. 24. The computer program product of claim 6, wherein the first process is boiler performance optimization, said first input of the first process is selected from the group consisting of: soot losses, cleanliness, boiler losses, SCR losses and FGD losses; and an output of the first process is selected from the group consisting of: heat rate (HR) and MW. 25. The computer program product of claim 6, wherein the first process is managed by a first process management module, wherein the first management module is selected from the group consisting of: a module for optimizing combustion; a module for optimizing sootblowing; a module for optimizing boiler performance; a module for optimizing selective catalytic reduction (SCR); and a module for optimizing flue gas desulfurization (FGD). 26. The computer program product of claim 6, wherein a third process having a plurality of inputs and an output that is said global output of the network, wherein the third process is profit optimization. 27. The computer program product of claim 26, wherein an input of said third process is selected from the group consisting of: heat rate (FIR), MW, NOx, NH3, SO, limestone, emission credits, and fuel costs.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.