Inverter PHL Estimation for Curtailed Power Setpoint Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional power plant controls lack reliable estimates of the potential high limit (PHL) of power generation during curtailment periods, hindering effective power setpoint management and visibility into additional generation potential if curtailment restrictions are lifted.
Innovation Solution
Employing a machine learning or artificial intelligence algorithm to estimate the PHL by generating synthetic data based on predetermined models, training the algorithm, and using current and historical measurement values to build a functional relationship between current and voltage, allowing for PHL estimation and control of the power plant.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional power plant controls are used during curtailment periods, then power output is limited according to arbitrary upper limits, but reliable estimates of the potential high limit (PHL) are not available
Solution Approach 1:
The patent creates a virtual copy of the power plant's operational characteristics through a trained machine learning model. This model replicates the relationship between environmental conditions and power generation, allowing estimation of PHL during curtailment without direct measurement. The model is trained on historical data containing environmental parameters (irradiance, temperature, wind speed) and corresponding power output, enabling it to predict what the power plant would generate under current conditions if not curtailed.
Solution Approach 2:
The patent performs preliminary training of the machine learning model using historical operational data before curtailment events occur. This advance preparation ensures the model is ready to immediately estimate PHL when curtailment begins, without delay. The model learns from past patterns of environmental conditions and power generation, building a predictive capability that activates immediately when needed during curtailment periods.
2Ease of operation
If arbitrary upper limits are set on power output during curtailment, then grid balancing is achieved, but effective power setpoint management is hindered
Solution Approach 1:
The patent implements a feedback mechanism where the estimated PHL information is continuously fed back to the power plant control system. This feedback enables operators to see the difference between current curtailed output and potential maximum output, allowing for more informed and precise power setpoint adjustments. The system can dynamically adjust setpoints based on real-time PHL estimates rather than relying on static arbitrary limits, improving both ease of operation and control accuracy.
3Measurement precision
If machine learning algorithms are trained with synthetic data from predetermined models, then accurate PHL estimation is achieved during curtailment, but system complexity increases
Solution Approach 1:
The patent develops a universal machine learning model that can estimate PHL across different curtailment scenarios and environmental conditions using a single trained algorithm. The model is designed to handle multiple types of environmental data (irradiance, temperature, wind speed) and various curtailment levels, making it multi-functional. This universal approach reduces the need for multiple specialized models or complex rule-based systems, thereby limiting the increase in system complexity while maintaining high estimation accuracy across diverse operating conditions.
Data Source
AI summary
A method is for controlling a power plant based on a potential high limit (PHL) of the power plant. The method includes generating synthetic data based on a plurality of predetermined models, each of which is for a specific environment in a power plant, training the machine learning algorithm with the synthetic data, receiving current measurement values of current and voltage at the inverters during a curtailment period, building a model of the function relationship between current and voltage, by the machine learning algorithm, based on previous measurement values and current measurement values, based on the built model, estimating a PHL of each inverter in the power plant, by the machine learning algorithm, and controlling the power plant, based on the estimated PHL.