EIPGRID has been developing the various of protocols and functions for develop optimal communication between electricity participants such as utilities, load aggregators, distributed energy devices

AI - Intelligent Energy Management

Energy management is nowadays a subject of great importance. The intelligent energy management service takes heterogeneous data as input from different sources such as smart electricity meter, gas meter, water meter, power, and energy meter. These data are processed accordingly and generate an adaptive recommendation based on user preferences. Operators(experts) can support in the knowledge creation, knowledge evolution, knowledge verification, and knowledge validation.

<Intelligent energy management service concept>

The intelligent energy management service consists of two parts such as knowledge acquisition and knowledge execution. The knowledge acquisition is considered an offline process, which is used to acquire knowledge from domain experts in the form of mind map. The mind map is then converted into decision tree for a better understanding of contributing factors. The knowledge engineer refines the decision tree under the supervision of domain expert and then proceeds to the rule generation. In the knowledge execution, the extracted rules are inferred in real-time to generate a recommendation.

<Intelligent energy management service – Knowledge Creation Process>

The intelligent energy management service uses a hybrid model, which is the combination of expert-driven and data-driven. The expert-driven uses the mechanism of mind map and decision tree for rule generation. While the data-driven approach uses machine learning model to predict a specific situation. The purpose of using a hybrid approach is to improve the accuracy and evolve the knowledge base using the ripple down rule mechanism.

<Hybrid Model of AI based Intelligent Energy Management Service>