Flexible Load Aggregation Using Privacy-Preserving Elasticity Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing demand-side flexibility resources, such as electric vehicles and smart buildings, are difficult to aggregate and optimize due to reliance on precise operational parameters that can be distorted or misreported, leading to inefficiencies and security constraint violations, and there is a need for non-intrusive methods that respect user privacy.
Innovation Solution
A method using machine learning models, specifically neural networks, for non-intrusive aggregation and optimal control of flexible loads, involving feature identification and elasticity estimation models to determine responsive electricity consumption and virtual elasticity matrices, with iterative optimization to ensure system security constraints are met.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If precise operational parameters are collected from users for demand-side flexibility resources, then aggregation and optimization accuracy can be improved, but user privacy protection is compromised and parameter distortion or malicious misreporting may occur
Solution Approach 1:
The patent extracts only the necessary aggregated load response characteristics from user data without collecting individual operational parameters. The feature identification model processes data to extract aggregate-level features that preserve accuracy while removing private information, effectively separating the useful signal from the privacy-containing data.
Solution Approach 2:
The patent introduces an intermediary processing layer (feature identification model and elasticity estimation model) between user data and the aggregation system. This intermediary transforms raw operational parameters into aggregated features and virtual elasticity matrices, preventing direct access to user privacy while maintaining computational accuracy for demand response optimization.
2Reliability
If operational parameters are distorted or maliciously misreported, then system security constraints may be violated and flexibility resources are wasted, but collecting precise parameters requires high-cost pilot projects with limited scalability
Solution Approach 1:
The patent enables the aggregation system to self-verify data quality through the elasticity estimation model, which detects parameter distortion and misreporting automatically. The system uses virtual elasticity matrices to identify inconsistent reports without requiring external verification mechanisms, reducing the need for complex pilot projects and high-cost monitoring infrastructure.
Solution Approach 2:
The patent transforms physical operational parameters into virtual elasticity matrices that are more robust against distortion and misreporting. By changing the parameter representation from direct operational data to derived elasticity characteristics, the system maintains reliability while reducing vulnerability to data quality issues.
3Measurement precision
If refined research methods with high-precision user modeling are implemented, then parameter accuracy can be improved, but user privacy awareness limits applicability and participation willingness
Solution Approach 1:
The patent segments the modeling process into two distinct components: feature identification models that operate on aggregated data without revealing individual user information, and elasticity estimation models that work with virtual parameters. This segmentation allows high-precision modeling while maintaining user privacy, thereby increasing participation willingness without sacrificing accuracy.
Data Source
AI summary
A computer-implemented method is used for non-intrusive aggregation and optimal control of flexible loads. The method includes: constructing first and second models oriented to the flexible loads; generating an incentive price for a current round, and inputting the incentive price respectively into the first and second models to output a real-time response and a real-time matrix; if a constraint is satisfied based on the real-time response and the real-time matrix, determining the incentive price for the current round is optimal, and the real-time consumption is optimal; if the constraint is not satisfied, constructing a third model based on the incentive price for the current round, the real-time response and the real-time matrix, and obtaining an optimal incentive price and an optimal response based on the third model; and performing non-invasive aggregation and optimal control of the flexible loads based on the optimal incentive price and the optimal response.