Fuel Cell Power Control Using IT Workload Shaping

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Fuel cell systems face reliability issues due to their slow response to transient power events in data centers, leading to fluctuations in power consumption and the need for expensive energy storage or complex electrical grid-connected architectures.

Innovation Solution

A power controller is implemented to monitor and adjust the power consumption of computing devices in a data center, optimizing performance and workload distribution to reduce variations in power draw from the fuel cell system, thereby smoothing power consumption and eliminating the need for expensive energy storage or complex architectures.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If fuel cell systems are used to power data centers, then energy efficiency and environmental benefits are improved, but reliability deteriorates due to slow response to transient power events

Engineering Contradiction:
Improveenergy efficiencyVSAvoidresponse speed to transient power events
Core Design Contradiction:
Loss of energyVSReliability

Solution Approach 1:

The power controller performs preliminary actions by predicting upcoming changes in power consumption based on historical statistics before they occur. It proactively adjusts the performance of computing devices ahead of time to prevent power fluctuations, rather than reacting after fluctuations occur. This anticipatory approach resolves the contradiction by preparing the system in advance to maintain both efficiency and reliability.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adjusts the performance of computing devices based on real-time power consumption monitoring and predictions. The power controller continuously modifies operational parameters such as clock speeds and workload distribution to optimize the balance between energy efficiency and reliable power delivery, allowing the fuel cell system to adapt to changing conditions without requiring expensive energy storage infrastructure.

Inventive Principle:
Principle #15Dynamics

2Device complexity

If fuel cell systems operate without power consumption optimization, then system complexity is reduced, but power fluctuations increase requiring expensive energy storage or complex grid connections

Engineering Contradiction:
Improvesystem architecture complexityVSAvoidpower consumption stability
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The power controller implements a feedback mechanism that continuously monitors power consumption of computing devices and uses this information to adjust system performance. Historical statistics are analyzed to predict future power consumption patterns, and control actions are taken based on this feedback loop. This feedback-driven approach maintains stable power consumption without requiring complex energy storage systems or grid connections.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The fuel cell system performs self-service through automated power consumption optimization. The power controller independently monitors, predicts, and adjusts computing device performance without external intervention or complex auxiliary systems. This self-managing capability maintains reliable power delivery while keeping the system architecture simple and avoiding expensive energy storage infrastructure.

Inventive Principle:
Principle #25Self-service

3Reliability

If performance of computing devices is adjusted to reduce power fluctuations, then fuel cell reliability is improved, but productivity of computing devices may deteriorate

Engineering Contradiction:
Improvefuel cell system reliabilityVSAvoidcomputing device performance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

Performance adjustments are made in advance based on predictions of power consumption changes rather than reactive adjustments. The power controller anticipates when power fluctuations will occur and proactively optimizes computing device performance beforehand, minimizing the impact on productivity while ensuring fuel cell reliability during actual power events.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically balances reliability and productivity by continuously adjusting computing device performance based on real-time conditions and historical patterns. Rather than static performance reduction, the system adapts performance levels to match actual power consumption needs, maintaining high productivity during stable periods while ensuring reliability during transient events.

Inventive Principle:
Principle #15Dynamics

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

The power controller effectively reduces power fluctuations, enhancing the reliability of fuel cell systems by managing power consumption based on historical statistics and real-time conditions, ensuring consistent energy supply without the requirement for costly storage or complex grid connections.

Implementation Method 1

a power source, an Information Technology (IT) stack electrically coupled to the power source... In some aspects, the power source is a fuel cell

Methodology Applied
Scientific EffectFuel cell electrochemical conversion: Fuel Cell

Data Source

PatentUS11287863B2Fuel cell power management
Publication Date: 2022.03.29 GOOGLE LLC
  • US11287863B2 patent drawing
  • US11287863B2 patent drawing
  • US11287863B2 patent drawing

AI summary

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for reducing variations in power consumption. In one aspect, a system includes a power controller communicatively coupled to a power source and a plurality of computing devices of an Information Technology (IT) stack electrically coupled to the power source. The controller configured to perform operations including monitoring variations in power consumption of the plurality of computing devices, determining, for the plurality of computing devices and based on the monitored power consumption, a power consumption profile for the plurality of computing devices that describes historical statistics of the power consumption, reducing variations in power consumed by the IT stack by adjusting performance of devices consuming power from the power source or adjusting which devices in the IT stack are consuming power from the power source based on a change in the power consumption and the power consumption profile.