UPS DC Link Voltage Control for AI Load Cycle Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The cycling of high-load and low-load elements in AI-based server systems leads to repeated battery discharges, reducing battery life in power components of the powertrain.
Innovation Solution
A power component utilizing a trained AI/ML model to predict load cycles and adjust DC link voltage accordingly, anticipating high-load elements to reduce battery discharges.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If the power component uses batteries to provide continuous power through load fluctuations, then the power delivery capability is improved, but the battery life is reduced due to repeated discharge cycles
Solution Approach 1:
The AI/ML model predicts future load cycles in advance, allowing the power component to proactively adjust DC link voltage before high-load elements occur. This preliminary action prevents unnecessary battery discharges by maintaining optimal voltage levels ahead of time, thus extending battery life while ensuring power delivery capability is maintained when needed.
Solution Approach 2:
The system dynamically adjusts the DC link voltage based on predicted load cycles rather than maintaining a static voltage level. The power component modifies operating parameters in real-time according to anticipated load variations, optimizing both power delivery and battery preservation through adaptive control.
2Duration of action of stationary object
If the DC link voltage is adjusted dynamically based on predicted load cycles, then the battery discharge cycles are reduced, but the device complexity increases due to AI/ML model integration
Solution Approach 1:
The power component integrates multiple functions into a single system: load monitoring, AI/ML prediction, DC link voltage control, and battery management. This multi-functionality reduces the need for separate dedicated systems, thereby limiting the increase in device complexity while achieving extended battery life through intelligent prediction and control.
3Duration of action of stationary object
If the power component anticipates high-load elements by adjusting DC link voltage, then the number of battery discharges is minimized, but the measurement precision requirements increase for load data
Solution Approach 1:
The system continuously monitors load data and uses this feedback to train and refine the AI/ML model. By incorporating real-time load information and historical patterns, the model improves its prediction accuracy over time, reducing the stringent precision requirements for individual measurements while still achieving effective battery discharge minimization.
Data Source
AI summary
A system may include a power component, the power component including at least one device processor configured to: obtain load data, wherein the load data comprises a power characteristic associated with at least one of a high-load element or a low-load element, obtain a trained power management artificial intelligence (AI) and/or machine learning (ML) model, based at least on the load data and the trained power management AI and/or ML model, infer a direct current (DC) link voltage adjustment, wherein the DC link voltage adjustment is correlated with a predicted load cycle parameter; and cause the power component to alter a DC link voltage from an initial level to an adjusted level based on the DC link voltage adjustment.