An internet of things device energy consumption optimization method based on a knowledge graph
By constructing a data organization method driven by operational disturbance events and an improved R-GCN model, the modeling challenge of multi-device correlation and influence relationships in energy consumption optimization of IoT devices was solved. This enabled accurate representation of energy consumption propagation relationships and dynamic adaptation of optimization decisions, thereby improving the accuracy and interpretability of energy consumption optimization.
Patent Information
- Application Number
- CN202610587719.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-29
- Publication Date
- 2026-07-21
AI Technical Summary
Existing energy consumption optimization methods for IoT devices are unable to reflect the interrelationships between multiple devices, and lack unified event-based modeling, multi-source heterogeneous data organization structure, and control action impact path evaluation mechanism, resulting in insufficient accuracy and dynamic adaptability of energy consumption optimization results.
By constructing a data organization method driven by disturbance events, a disturbance response observation sequence is generated. Then, by combining time window alignment, differential calculation and delay matching, a relational response representation matrix is generated. A response-enhanced device energy consumption knowledge graph is constructed. The improved R-GCN model is used for path evaluation and optimization decision-making, forming a closed-loop dynamic optimization process.
It improves the accuracy of energy consumption propagation relationship modeling and the interpretability of control action impact assessment in multi-device interconnected scenarios, and enhances the dynamic adaptability of energy consumption optimization decision-making.
Smart Images

Figure CN122432571A_ABST