Edge Power Monitoring With Dual-State Profiles and Adaptive Polling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing smart home network systems focus primarily on human safety within the home, neglecting potential hazards in the surrounding environment due to insufficient monitoring of power consumption patterns.
Innovation Solution
A power monitor apparatus comprising power meters and an edge controller that utilizes dual memory spaces and adjustable polling rates to identify normal, alert, and alarm states based on power consumption thresholds, triggering alarms when necessary to address potential hazards.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single memory space is used for all power profile storage, then device complexity is reduced, but measurement precision and reliability of power anomaly detection deteriorate due to inability to distinguish between normal and alert state profiles
Solution Approach 1:
The patent divides the memory space into two distinct segments: a first memory space for storing normal state power profiles and a second memory space for storing alert state power profiles. This segmentation allows the system to maintain different sets of baseline profiles separately, improving the accuracy of anomaly detection by preventing contamination between normal and alert state data while managing memory resources efficiently.
2Reliability
If a fixed high polling rate is used for all power loops, then reliability of hazard detection is improved, but energy consumption increases significantly
Solution Approach 1:
The patent implements dynamic polling rate adjustment where the system transitions between different polling rates based on the detected state of power loops. When power consumption remains within normal thresholds, a first (lower) polling rate is used to conserve energy. When alert thresholds are exceeded, the system automatically switches to a second (higher) polling rate to improve hazard detection reliability, thus adaptively balancing energy consumption and detection reliability.
3Measurement precision
If power profiles are continuously monitored at high frequency, then measurement precision of power anomalies is improved, but loss of energy increases due to frequent processing
Solution Approach 1:
The patent employs periodic action by implementing state-based polling intervals. Instead of continuous high-frequency monitoring, the system uses periodic polling at adjusted frequencies based on the current state. Normal state power loops are polled at a first periodic interval while alert state power loops are polled at a second (shorter) periodic interval. This approach maintains measurement precision for anomaly detection while significantly reducing energy loss by avoiding unnecessary high-frequency processing during normal operation.
4Reliability
If alert state power profiles are stored separately from normal state profiles, then reliability of alarm triggering is improved, but device complexity increases due to dual memory spaces
Solution Approach 1:
The patent segments memory into distinct first and second memory spaces for storing normal state and alert state power profiles respectively. This segmentation improves alarm triggering reliability by ensuring that alert state profiles are not contaminated by normal state data, allowing for more accurate anomaly detection. The system manages this increased memory structure complexity through automated state transition logic that handles the organization and retrieval of profiles from appropriate memory segments.
Data Source
AI summary
A power monitor apparatus including power meters and an edge controller is provided. The power meters are configured to generate a plurality of power profiles associated with a plurality of power loops. The edge controller includes a memory and a processor. The processor is configured to move the power profiles into a normal state array and store them as normal process power profiles (NPPPs); compare each NPPP with a first criterion at a first polling rate; identify an alert state subset from the NPPPs that meet the first criterion; move the alert state subset into an alert state array and stored as alert process power profiles (APPPs); compare each APPP with a second criterion at a second polling rate higher than the first polling rate; identify an alarm state subset that meet the second criterion and trigger an alarm event associated with the alarm state subset.