Fuel Cell Compressor Control via Predictive Pressure Drop
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fuel cell power systems face challenges in quickly adjusting air flow to meet fluctuating electricity demands, leading to unwanted noise and reduced compressor lifetime due to dynamic response issues in air compressor control.
Innovation Solution
A controller is implemented in the fuel cell power system that uses a combination of feedback and feedforward gain parameters to adjust the command signal for the air compressor, stabilizing air flow and reducing noise by determining a desired pressure drop and adjusting the control command accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If the controller quickly adjusts the air compressor speed in response to electricity demand fluctuations, then the dynamic response speed is improved, but unwanted noise is generated and compressor lifetime is reduced
Solution Approach 1:
The controller predicts future air flow requirements based on the relationship between current electrical load and historical air flow data, then proactively adjusts compressor speed before the actual demand occurs. This preliminary action allows smooth, planned transitions rather than reactive adjustments, reducing noise and mechanical stress while maintaining fast response to load changes.
Solution Approach 2:
The system dynamically adapts the compressor control strategy by continuously learning from operational data and adjusting the predicted air flow requirements based on changing operating conditions. This dynamic adaptation enables optimal balance between response speed and noise reduction across different operating scenarios.
2Speed
If the controller quickly adjusts the air compressor speed in response to electricity demand fluctuations, then the dynamic response speed is improved, but compressor lifetime is reduced
Solution Approach 1:
By predicting air flow requirements in advance and pre-adjusting compressor speed, the system avoids sudden, aggressive speed changes that cause mechanical stress. This proactive approach extends compressor lifetime while maintaining the ability to respond quickly to electricity demand fluctuations.
Solution Approach 2:
The controller uses learned operational patterns to cushion against abrupt load changes by smoothing compressor speed transitions. This cushioning effect protects the compressor from mechanical shocks and extends its operational life while preserving dynamic response capability.
3Measurement precision
If traditional feedback control is used to adjust air compressor speed, then measurement precision is maintained, but excessive noise and compressor revving occur
Solution Approach 1:
Instead of reacting to pressure drop measurements with immediate compressor adjustments, the system uses predicted air flow requirements to proactively determine compressor speed changes. This eliminates the excessive revving caused by tight feedback control while maintaining accurate air flow management through predictive modeling.
Solution Approach 2:
The predictive model acts as an intermediary between pressure drop measurements and compressor control commands. It translates precise measurements into smooth, noise-free control actions by considering the learned relationship between electrical load and air flow requirements, rather than directly translating pressure variations into compressor adjustments.
Data Source
AI summary
One aspect of the present disclosure is directed to a fuel cell power system. The system may include one or more fuel cells configured to generate electric power and a compressor configured to supply compressed air to the one or more fuel cells. The system may further include one or more sensors. The sensors may be configured to generate a signal indicative of at least one measured parameter of air flow across the one or more fuel cells. The system may also include a controller in communication with the one or more fuel cells, the compressor, and the sensors. The controller may be configured to determine a desired pressure drop based on at least one calculated parameter, determine a control command for the compressor based on the desired pressure drop, and adjust the control command based on a feedback gain parameter and a feed forward gain parameter.