Data Center Power Distribution via Renewable Energy Prediction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Data centers relying exclusively on renewable energy face instability in power supply due to variable energy generation from sources like solar and wind, leading to high installation costs and emergency power supply concerns, especially in locations without transmission lines.

Innovation Solution

A data center system that includes a controller for predicting electric power generation and storage using renewable energy, switching operating modes based on predicted values, and a method for load distribution between data centers to ensure stable job execution by assigning tasks to servers with sufficient power reserves, utilizing storage batteries and weather forecast data to manage power supply.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a data center operates exclusively on renewable energy, then environmental sustainability is improved, but power supply stability deteriorates

Engineering Contradiction:
Improvepower supply stabilityVSAvoidrenewable energy dependency
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system performs preliminary actions by predicting future power generation amounts from renewable sources and storing this information in advance. The load distribution controller uses these predictions to proactively adjust job assignments between data centers before power shortages occur, ensuring continuous operation without interruption.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces a load distribution controller as an intermediary that manages power allocation between multiple data centers. This controller receives power generation predictions, determines optimal load distributions, and coordinates job migrations, acting as a mediator that balances renewable energy variability with stable power supply across the system.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If solar panels and storage batteries are increased to ensure constant power supply, then power supply stability is improved, but installation costs increase

Engineering Contradiction:
Improvepower supply stabilityVSAvoidinstallation scale
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system segments the power supply function across multiple data centers rather than relying on a single large-scale renewable energy installation. Each data center can operate with smaller, more manageable solar panels and storage batteries, while the load distribution controller coordinates between them to ensure overall system stability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent merges multiple data centers into a coordinated system where excess power from one center can support another. By combining the power resources of multiple centers with moderate solar panel and battery installations, the system achieves greater overall reliability than any single center could provide alone.

Inventive Principle:
Principle #5Merging (Combining)

3Reliability

If load distribution is implemented between data centers, then power supply stability is improved, but system complexity increases

Engineering Contradiction:
Improvepower supply stabilityVSAvoidcontrol system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The load distribution controller operates autonomously by automatically acquiring power generation predictions, calculating optimal load distributions, and executing job migrations without manual intervention. The system serves itself by monitoring its own power status and making real-time adjustments to maintain stability.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system dynamically changes operational parameters such as job assignment locations and power distribution ratios based on real-time power generation predictions. The load distribution controller adjusts these parameters automatically in response to changing renewable energy availability, maintaining stability through adaptive parameter optimization.

Inventive Principle:
Principle #35Parameter changes

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

Enables stable and continuous power supply to data centers, allowing flexible job assignment and reliable operation even during fluctuations in renewable energy generation, ensuring minimal downtime and efficient resource allocation.

Implementation Method 1

a storage battery for storing the electric power

Methodology Applied
Scientific EffectEnergy storage in battery: Battery (electricity)

Implementation Method 2

acquiring a predicted value for an amount of electric power to be stored in the storage battery... determined on the basis of weather forecast data

Methodology Applied
Scientific EffectWeather-dependent energy generation prediction: Solar Energy

Data Source

PatentUS10320195B2Distributing power between data centers
Publication Date: 2019.06.11 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US10320195B2 patent drawing
  • US10320195B2 patent drawing
  • US10320195B2 patent drawing

AI summary

A data center operable using only electric power based on renewable energy. The data center includes at least one device driven by the electric power, a storage battery for storing the electric power, and a controller for switching the operating mode of the device over the course of time on the basis of predicted values for the amount of electric power generated using renewable energy, the amount of electric power stored in the storage battery, and the amount of electric power consumed by the device.