Resilient Hydrogen Production via Machine Learning Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The challenge in hydrogen production from renewable energy sources lies in the lack of resilient production systems that can accurately match supply and demand, particularly due to the intermittent nature of renewable energy sources like solar and wind, which requires continuous data-driven monitoring and control to ensure reliable and efficient hydrogen production.

Innovation Solution

A method and system utilizing machine learning models, specifically reinforcement learning agents, to predict future power output from renewable energy sources and adjust hydrogen production parameters in real-time, optimizing production rates, storage, and distribution to maintain a resilient hydrogen production system.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Object-affected harmful factors

If renewable energy sources are used for hydrogen production, then environmental sustainability is improved, but production reliability deteriorates due to intermittent supply

Engineering Contradiction:
ImproveCO2 emissionsVSAvoidproduction reliability
Core Design Contradiction:
Object-affected harmful factorsVSReliability

Solution Approach 1:

The system performs preliminary actions by predicting future power output using machine learning models before making production decisions. This allows the system to proactively adjust hydrogen production parameters in advance, ensuring reliable production despite intermittent renewable energy supply by preparing for future supply variations ahead of time

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements continuous feedback loops by monitoring actual power output, comparing it with predictions, and using this information to dynamically adjust production parameters. This feedback mechanism enables the system to adapt to real-time variations in renewable energy supply while maintaining production reliability

Inventive Principle:
Principle #23Feedback

2Reliability

If real-time adjustment of production parameters is implemented, then production reliability is improved, but system complexity increases

Engineering Contradiction:
Improveproduction reliabilityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system employs self-service by using machine learning models that automatically predict power output and determine optimal production parameters without human intervention. The reinforcement learning agent autonomously adjusts production settings based on predicted conditions, reducing the need for complex manual control systems while maintaining high reliability

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system manages complexity by focusing on changing key production parameters dynamically rather than controlling all system aspects. By identifying and adjusting the most critical parameters based on predicted power output, the system achieves reliable production with manageable complexity

Inventive Principle:
Principle #35Parameter changes

3Productivity

If machine learning prediction is used, then production optimization is improved, but data processing time increases

Engineering Contradiction:
Improveproduction optimizationVSAvoiddata processing time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary data processing by training machine learning models on historical data in advance. Once trained, these models can rapidly predict future power output without requiring extensive real-time computation, thus optimizing production decisions while minimizing data processing time during operation

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system applies partial action by using simplified prediction models that focus on the most relevant features for production optimization. Rather than processing all available data comprehensively, the system identifies and processes only the critical data elements needed for timely production decisions

Inventive Principle:
Principle #16Partial or excessive action

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

This approach enables timely and accurate adjustment of hydrogen production parameters, optimizing hydrogen production and storage to ensure a reliable and resilient energy supply, even under adverse environmental conditions, by learning from past experiences and predicting future energy availability.

Implementation Method 1

predicting, using a machine learning model, a future power output of the renewable energy sources based on the measurements

Methodology Applied
Scientific EffectMachine learning prediction:

Implementation Method 2

Green hydrogen can be produced from a renewable source such as solar or wind by the electrolysis of water

Methodology Applied
Scientific EffectElectrolysis: Electrolysis

Data Source

PatentEP4249427A1Method and system for resilient hydrogen production
Publication Date: 2023.09.27 CIBUSCELL TECH GMBH
  • EP4249427A1 patent drawingFigure 1
  • EP4249427A1 patent drawingFigure 2
  • EP4249427A1 patent drawingFigure 3

AI summary

The present invention provides a method for supporting resilient hydrogen production from one or more renewable energy sources, the method comprising: - obtaining, at a plurality of time points, measurements from environmental sensors that indicate a current environmental condition of the renewable energy sources, - predicting, using a machine learning model, a future power output of the renewable energy sources based on the measurements, and - adjusting, based on the machine learning model, one or more hydrogen production parameters based on the predicted future power output.