Combined Cycle Plant Setpoint Optimization With ANN and Kalman Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Combined cycle power plants face complexity in determining optimal operational settings to minimize generation costs and maximize profitability, especially with power augmentation devices like chillers and duct firing systems, due to varying demand and fuel costs, which existing systems struggle to address effectively.
Innovation Solution
A method involving digital modeling of gas turbines, steam turbines, chiller systems, and duct firing systems, correlated using an artificial neural network, with a Kalman filter for real-time calibration and optimization, to determine optimal operating settings that minimize costs and maximize profitability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If power augmentation devices (chillers and duct firing systems) are added to increase power output capacity, then the power generation capability is improved, but the device complexity and operational difficulty increase significantly
Solution Approach 1:
The system segments the complex power plant into distinct digital models for each major component (gas turbine, steam turbine, chiller, duct firing system). Each component is modeled independently and then integrated through an artificial neural network, allowing manageable complexity while maintaining overall system optimization capability.
Solution Approach 2:
An artificial neural network acts as an intermediary that correlates the separate digital models of gas turbine, steam turbine, chiller, and duct firing system. This neural network mediator integrates the segmented component models into a unified plant-wide simulation framework, enabling coordinated optimization of all power augmentation devices.
2Productivity
If detailed simulation and optimization systems are implemented to determine optimal operational settings, then the profitability and operational efficiency are improved, but the computational complexity and data processing requirements increase
Solution Approach 1:
The system performs preliminary digital modeling and simulation of all plant components under various operating conditions before actual operation. By pre-establishing digital twins and running simulations in advance, the system determines optimal operational settings ahead of time, reducing real-time computational burden while maintaining high operational efficiency.
Solution Approach 2:
The optimization system continuously compares actual plant performance with simulated performance from digital models. This feedback mechanism allows the system to refine operational settings based on real-world deviations, improving profitability and efficiency while adapting to changing conditions without requiring complete re-simulation.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution enables accurate simulation and optimization of combined cycle power plants, allowing operators to adjust settings for maximum profitability and effective power generation, accounting for thermodynamic and economic factors, thereby enhancing the plant's operational efficiency and grid power delivery.
Implementation Method 1
adjusting the simulation of the operation of the power plant with a Kalman filter that receives and compares data regarding the actual operation of the power plant to data regarding the simulation
Implementation Method 2
correlating the digital models using an artificial neural network
Data Source
Figure 1
Figure 2
AI summary
A method for determining an operating set point for a combined cycle power plant, the method includes: simulating the operation of the power plant (10); correcting the simulation of the operation of the power plant (10); optimizing the simulation of the operation by simulating the operation at different operating settings and selecting at least one of the operating settings as being optimal, and generating the operating set point based on the optimized simulation of the power plant (10).