Smart Electrical Protection Using Adaptive EMI Trip Analytics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional protection devices in electric power distribution systems struggle with nuisance trips due to environmental noise interference, such as EMI, and lack adaptive capabilities to characterize noise sources, leading to inefficient fault detection and isolation.
Innovation Solution
Integration of a hybrid machine-learning model (HMLM) with field protection systems (FPS) that utilize real-time data and environmental conditions to learn and adapt to noise parameters, enabling intelligent decision-making and minimizing nuisance trips.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional filters are used to detect noises in electrical circuits, then noise detection capability is provided, but the filters cannot adapt to environmental changes and fail to characterize EMI sources effectively
Solution Approach 1:
The patent implements dynamic adaptability by enabling the protection device to learn from historical trip data and environmental parameters, continuously updating its noise characterization models to adapt to changing environmental conditions while maintaining precise noise detection capabilities
Solution Approach 2:
The system changes parameters by incorporating multiple environmental parameters (temperature, humidity, vibration) and using machine learning to dynamically adjust noise thresholds and characterization criteria based on learned patterns from historical data, enabling effective EMI source characterization across varying conditions
2Ease of operation
If IOT based tripping circuits consider loads for isolation decisions, then tripping decisions are made based on load information, but the system requires Cloud or central server communication and does not characterize the load
Solution Approach 1:
The protection device performs self-characterization of loads and environmental conditions using onboard sensors and machine learning algorithms, making autonomous tripping decisions without requiring external cloud communication or central server infrastructure
Solution Approach 2:
The patent extracts and processes environmental and load parameters directly at the protection device using integrated sensors and machine learning models, removing the dependency on external cloud systems for data processing and decision-making
3Reliability
If conventional protection devices are used for fault detection, then fault isolation is achieved, but nuisance trips occur due to failure to distinguish between true faults and environmental noises
Solution Approach 1:
The system uses feedback from historical trip data and environmental parameters to continuously refine noise characterization models, enabling the protection device to distinguish between true faults and environmental noises while maintaining reliable fault isolation
Solution Approach 2:
The patent converts the harmful effect of environmental noises into a beneficial learning opportunity by using machine learning to analyze patterns in historical trip data caused by various environmental conditions, enabling the system to recognize and ignore similar noise patterns in the future while maintaining sensitive fault detection
Data Source
AI summary
A system and method for protecting an electric power distribution system integrated with a hybrid machine-learning model includes a plurality of field protection systems, a plurality of electrical switching circuits, and a local area network. Data from each of the plurality of field protection systems that is connected to the electrical switching circuits that are configured for protecting the electric power distribution system, is transmitted to the cloud. The received data is processed to identify a change in patterns and compute an error related to the HMLM in comparison with the processed data. The HMLM that is implemented in the plurality of FPS is calibrated to minimize the error and send updates to develop a new decision making firmware thereby to control actuation of the plurality of the electrical switching circuits.