Finite State Machine Load Identification via Transient Waveforms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for identifying and managing plug-in loads in commercial or residential buildings lack effectiveness due to the inability to provide finer granular visibility on energy usage, leading to nuisance trips and potential damage, and are not well-suited for distinguishing between similar load types like office appliances and PCs.
Innovation Solution
A system utilizing sensors to acquire voltage and current waveforms, calculating RMS profiles, quantizing state-values, and generating a state-sequence that describes a finite state machine model to identify different electric load types based on their start-up and transient profiles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing plug-in load control and management products are used, then some level of load management is achieved, but nuisance trips occur causing inconvenience and potential damage
Solution Approach 1:
The system changes the parameter of load identification from generic categorization to type-specific identification by analyzing transient response characteristics. By monitoring how loads respond to disturbances and categorizing them into specific types (resistive, inductive, electronic), the system adapts its control parameters to match the actual load characteristics, preventing nuisance trips caused by mismatched control strategies.
Solution Approach 2:
The patent replaces mechanical/circuit-based protection mechanisms with intelligent software-based identification and control. Instead of using fixed thermal-magnetic breakers or simple overload protection, the system uses processors to analyze voltage and current waveforms, identify load types through transient response patterns, and implement adaptive control algorithms that prevent nuisance trips through intelligent decision-making rather than mechanical response.
2Ease of operation
If generic load management approaches are used, then implementation is simple, but finer granular visibility on energy usage by load types is not achieved
Solution Approach 1:
The system segments the overall load management function into distinct load type categories (resistive, inductive, electronic) based on transient response characteristics. By dividing the monitoring and control process into type-specific segments, the system achieves fine-grained visibility of energy usage by load type while maintaining implementation simplicity through automated classification algorithms that require minimal user configuration.
Solution Approach 2:
The patent introduces transient response analysis as an intermediary mechanism between raw electrical measurements and load type identification. By using the transient response characteristics (how voltage and current behave during switching events) as an intermediary signal, the system automatically categorizes loads without requiring user input or complex setup, thereby achieving detailed load-type visibility while keeping the system easy to implement.
3Device complexity
If load identification methods lack type-specific discrimination, then system complexity is reduced, but distinction between similar load types like office appliances and PCs is not achieved
Solution Approach 1:
The system employs dynamic transient response analysis to distinguish between similar load types. Instead of using static or steady-state measurements that may be identical for different load types, the system captures the dynamic behavior during switching transients. By analyzing how voltage and current evolve over time during load switching events, the system achieves precise discrimination between office appliances, PCs, and other similar loads while keeping the implementation relatively simple through automated waveform analysis.
Data Source
AI summary
A system is for a plurality of different electric load types. The system includes a plurality of sensors structured to sense a voltage signal and a current signal for each of the different electric loads; and a processor. The processor acquires a voltage and current waveform from the sensors for a corresponding one of the different electric load types; calculates a power or current RMS profile of the waveform; quantizes the power or current RMS profile into a set of quantized state-values; evaluates a state-duration for each of the quantized state-values; evaluates a plurality of state-types based on the power or current RMS profile and the quantized state-values; generates a state-sequence that describes a corresponding finite state machine model of a generalized load start-up or transient profile for the corresponding electric load type; and identifies the corresponding electric load type.