Electrolyzer Plant Simulation Model Updating From Prediction Deviations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Simulating the behavior of an electrolyzer plant for optimal operation and anomaly detection is challenging due to the large amount of required information, incomplete understanding of effects, and unavailable data from component suppliers, leading to inaccuracies in existing simulation models.
Innovation Solution
A method to establish and improve simulation models by mapping operating variables to performance indicators, using measured data to predict and adjust model parameters, and applying these adjustments based on deviations between prediction and measurement values to enhance model accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a detailed simulation model of the electrolyzer plant is created using first level principles, then the model accuracy would be improved, but the amount of required information and data would be too large and some data would not be available from suppliers
Solution Approach 1:
The patent changes the parameters of the simulation model from detailed first-level physical principles to simplified behavioral parameters obtained from suppliers (efficiency curves, polarization curves, degradation parameters). This parameter transformation allows the model to function with limited data while maintaining sufficient accuracy for operational decisions.
Solution Approach 2:
Instead of creating a detailed physical model from first principles, the patent uses simplified copies or representations of the actual system behavior provided by suppliers. These behavioral models are copies that capture essential characteristics without requiring complete physical understanding or all underlying data.
2Device complexity
If behavioral information from suppliers is used to create a simplified model, then the model complexity is reduced and data availability is improved, but the model accuracy deteriorates
Solution Approach 1:
The patent implements feedback mechanisms where the simplified model's predictions are continuously compared with actual plant measurements. Deviations are used to update and refine the behavioral parameters, ensuring the model maintains accuracy despite its simplified structure. This closed-loop approach compensates for the inherent limitations of simplified models.
Solution Approach 2:
The patent makes the simulation model dynamic by allowing behavioral parameters to change over time based on operating conditions, aging, and learned deviations. Rather than static simplified parameters, the model adapts its parameters dynamically to maintain accuracy across different operating states and time periods.
3Ease of manufacture
If the simulation model uses simplified behavioral parameters from suppliers, then the ease of model establishment is improved, but the ability to detect anomalies and determine optimal operation deteriorates
Solution Approach 1:
The patent performs preliminary actions by establishing baseline behavioral parameters from suppliers before actual operation, and then continuously comparing real-time measurements against these baselines. This preliminary setup enables quick anomaly detection without requiring complex real-time analysis, as deviations from expected behavior are immediately apparent.
Solution Approach 2:
The patent introduces an intermediary layer between the simplified supplier models and the actual plant operation. This intermediary continuously monitors deviations and translates them into actionable insights for anomaly detection and optimization, bridging the gap between simplified models and reliable decision-making.
Data Source
Figure 1
Figure 2~3
AI summary
A method (100) for establishing and/or improving a simulation model (2) of an electrolyzer plant (1) comprising at least one electrolyzer, wherein the simulation model (2) is configured to map one or more variables (3) that characterize an operating state (1a) of the electrolyzer plant (1) to one or more performance indicators (4) of the electrolyzer plant (1), the method comprising the steps of: • predicting (110), using the simulation model (2), for a given operating state (1a) of the electrolyzer plant (1), one or more prediction values (5#) of at least one quantity (5); • obtaining (120) one or more measurement values (5*) of the same at least one quantity (5) that relate to the same given operating state (1a) of the electrolyzer plant (1); • determining (130) at least one deviation (Δ) of the one or more prediction values (5#) from the respective measurement values (5*); • determining (140), from the deviation (Δ) and the simulation model (2), a contribution and/or an adjustment (2a) to the simulation model (2); and • applying (150) the contribution, and/or the adjustment (2a), to the simulation model (2).