AI for the monitoring and classification of energy flexibilities
Challenge
Companies in the energy sector must manage and interpret large volumes of data to optimize resource use and ensure operational reliability.
Artificialy has developed an AI solution to analyze so-called energy signatures, i.e. the specific energy consumption pattern of each device, whether a heat pump, a production machine, or other equipment. This makes it possible to automatically identify connected devices and improve their management.
Specifically, the client needed to improve the monitoring of energy flexibilities. In particular:
a machine learning algorithm capable of analyzing power distribution to verify synchronization with relays;
a model based on historical and weather data to estimate the nominal power of devices;
a Machine Learning classification model to automatically identify the type of device (e.g. boiler, heat pump)
The solution makes it possible to transform raw data into operational insights, improving control and information quality.
Our AI Solution
Artificialy has developed a solution that combines supervised and unsupervised Machine Learning algorithms for the automatic analysis of consumption time series.
The system:
assesses the synchronization between consumption and control signals;
estimates nominal power using historical data and external variables (e.g. weather conditions);
automatically classifies devices through dedicated ML models.
This approach makes it possible to obtain reliable and scalable insights from operational data.
Results
All the metrics considered are showing growth, highlighting an overall improvement in data quality and monitoring capability.
The greatest benefits are seen in terms of time savings and detection accuracy for each element, in particular:
84% accuracy in synchronization verification;
90% recall in anomaly detection;
69% accuracy in power estimation;
84% recall for identifying inconsistent data;
76% accuracy in device classification.
You may also be interested in
AI for automating ticket routing and responses in customer service
