Electrolyzer Plant Simulation Refinement Using Measured Operating Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Simulating the behavior of a large water electrolyzer plant to run it optimally is challenging due to the vast amount of required information, incomplete understanding of effects, and unavailability of data from component suppliers, making it difficult to establish accurate models for optimal operation and anomaly detection.
Innovation Solution
A method to improve the accuracy of simulation models by mapping operating variables to performance indicators, using measurement values to adjust and refine existing models, and introducing controlled variations to measure responses, allowing for better prediction and remediation of abnormal states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a detailed simulation model is created using first level principles, then the model accuracy would be high, but the amount of information required becomes too large and data availability becomes insufficient
Solution Approach 1:
The patent transforms the simulation approach by changing from first-principles physical parameters to empirically derived behavioral parameters. Instead of modeling detailed physical processes requiring extensive data, the invention uses measured polarization curves, efficiency values, and degradation data to create simplified parameter sets that capture essential plant behavior with far fewer data requirements.
Solution Approach 2:
The patent creates simplified copies of the complex electrolyzer system behavior through empirical models. Rather than simulating the full physical complexity, behavioral information from actual plant measurements is copied and used to represent system characteristics, maintaining accuracy while reducing information needs.
2Loss of information
If behavioral information is obtained from suppliers based on measurements, then the data availability improves, but the model accuracy for optimal operation becomes insufficient
Solution Approach 1:
The patent implements a feedback mechanism where initial simulation models using supplier data are continuously refined through comparison with actual plant measurements. Deviations between predicted and measured values are used to adjust and improve model parameters, progressively enhancing accuracy while maintaining data availability from practical sources.
Solution Approach 2:
The patent performs preliminary model establishment using available supplier behavioral information before actual plant operation. This initial model serves as a foundation that is then refined through subsequent measurement campaigns, allowing the system to benefit from both supplier data and empirical validation.
3Ease of operation
If simplified models are used for computation tractability, then the ease of operation improves, but the ability to detect and remediate abnormal states becomes less reliable
Solution Approach 1:
The patent applies partial action by focusing the simplified model on capturing only the most critical behavioral aspects necessary for anomaly detection. Rather than attempting to model all system complexities, the approach selectively represents key performance indicators and degradation patterns that are sufficient for reliable operation monitoring and fault detection.
Data Source
AI summary
A method for establishing and/or improving a simulation model of an electrolyzer plant includes predicting one or more prediction values of at least one quantity; obtaining one or more measurement values of the same at least one quantity that relate to the same given operating state of the electrolyzer plant; determining at least one deviation of the one or more prediction values from the respective measurement values; determining from the deviation and the simulation model a contribution and/or an adjustment to the simulation model; and applying the contribution and/or the adjustment to the simulation model.