Automatic identification of operating parameters for power plants
First Claim
1. A method for operating of a power plant, comprising:
- providing a processor implementing an operating parameter determination algorithm stored in a non-transitory memory accessible by said processor, said algorithm including;
data mining a historical operating condition database for the power plant with a first partitional clustering algorithm to generate a statistical model;
calculating dynamic operating condition target values from the statistical model taking into account current operating condition data of the power plant by performing a second partitional clustering algorithm which comprises a k-means clustering algorithm upon the statistical model and the current operating condition data and sorting said operating condition target values resulting from the k-means clustering algorithm using a bubble sort algorithm;
performing a real-time energy-loss analysis based on the dynamic operating condition target values;
automatically identifying a plurality of operating parameters of the power plant from the energy-loss analysis, andmanipulating a first operating parameter selected from the plurality of operating parameters using an adjustment device in the power plant during operation of the power plant.
1 Assignment
0 Petitions
Accused Products
Abstract
A method for obtaining operating parameters of a power plant includes data mining a historical operating condition database for the power plant with a partitional clustering algorithm to generate a statistical model, and calculating dynamic operating condition target values from the statistical model taking into account current operating condition data of the power plant. The method further includes performing a real-time energy-loss analysis based on the dynamic operating condition target values and automatically identifying at least one operating parameter of the power plant from the energy-loss analysis. The partitional clustering algorithm can be a k-means clustering algorithm.
16 Citations
13 Claims
-
1. A method for operating of a power plant, comprising:
-
providing a processor implementing an operating parameter determination algorithm stored in a non-transitory memory accessible by said processor, said algorithm including; data mining a historical operating condition database for the power plant with a first partitional clustering algorithm to generate a statistical model; calculating dynamic operating condition target values from the statistical model taking into account current operating condition data of the power plant by performing a second partitional clustering algorithm which comprises a k-means clustering algorithm upon the statistical model and the current operating condition data and sorting said operating condition target values resulting from the k-means clustering algorithm using a bubble sort algorithm; performing a real-time energy-loss analysis based on the dynamic operating condition target values; automatically identifying a plurality of operating parameters of the power plant from the energy-loss analysis, and manipulating a first operating parameter selected from the plurality of operating parameters using an adjustment device in the power plant during operation of the power plant. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. A system for operating a power plant, comprising:
-
a non-transitory memory for storing historical operating condition data for the power plant and an operating parameter determination algorithm; a processor coupled to access said operating parameter determination algorithm and for implementing the operating parameter determination algorithm, said algorithm including; data mining a historical operating condition database for the power plant with a partitional clustering algorithm to generate a statistical model; calculating dynamic operating condition target values from the statistical model taking into account current operating condition data of the power plant by performing a second partitional clustering algorithm which comprises a k-means clustering algorithm upon the statistical model and the current operating condition data, and sorting said operating condition target values resulting from the k-means clustering algorithm using a bubble sort algorithm; performing a real-time energy-loss analysis based on the dynamic operating condition target values, and automatically identifying a plurality of operating parameters of the power plant from the energy-loss analysis, and adjustment devices coupled within the power plant for adjusting respective ones of the plurality of operating parameters. - View Dependent Claims (10, 11, 12, 13)
-
Specification