Privacy-Preserving Thermal Load Scheduling Without Home Thermal Models

The patent extracts only the necessary aggregate information (total power consumption of thermal loads) while leaving sensitive individual data (temperature readings, individual appliance consumption patterns) at the consumer premises. This allows the aggregator to perform load management based on aggregate demand signals without accessing or storing private consumer data, thus resolving the contradiction between measurement precision and privacy protection.

The patent introduces a local controller at each consumer premises as an intermediary that processes temperature and power consumption data locally. This intermediary computes control decisions based on local measurements and communicates only aggregate power consumption signals to the aggregator, preventing direct access to sensitive consumer data while enabling effective thermal load management.

The patent replaces complex, computationally intensive thermal models with simple, lightweight control algorithms that can be executed by inexpensive local controllers. Instead of using detailed building thermal dynamics models requiring significant computational resources, the system employs simple on/off control logic based on temperature thresholds and aggregate power signals, making large-scale deployment feasible.

The patent enables each consumer premises to autonomously manage its thermal loads using local measurements and simple control logic. The local controllers independently make control decisions based on local temperature conditions and received aggregate power signals from the aggregator, eliminating the need for centralized complex model-based optimization that would be computationally intractable at scale.