AIknow has developed an advanced energy monitoring solution for its client, offering a complete system for the collection, analysis, and visualization of production and energy consumption data. The platform integrates seamlessly with existing infrastructure through IoT gateways installed in the field, capable of connecting directly to devices such as PLCs, smart meters, and sensors.
Client Requirements
The main objective of the solution is to support the end client by providing:
- An advanced support service
- The centralization of information on product performance
- The monitoring of correct product usage by customers
Operational Context
The platform is designed to be deployed in various sectors, including:
- Devices for stabilizing the electrical grid used in industrial production machinery
- Renewable energy power generation plants
- Monitoring of transformation, distribution, and control substations
- Monitoring energy consumption of buildings and industrial production facilities
AIknow Solution
AIknow designed and developed a fully customized platform to meet the specific client needs. The main features include:
Edge Computing and IoT
-
-
- Selection and provisioning of IoT gateways capable of hosting the edge-side application stack
- Use of the Sparkplug standard for data exchange between edge and cloud via MQTT protocol and Protocol Buffer binary payload
- Centralized fleet management of gateways via BalenaOS
- Edge stack on gateways composed of:
- Dedicated C++ services for each industrial protocol
- Orchestration applications in Java/Python for buffering, cloud data sending, and command reception
-
Cloud Application
-
-
- Telemetry reception and time-series data storage in a database (InfluxDB v2)
- Diagnostic alarm generation for gateways and devices
- Monitoring of metrics with configurable alarm thresholds
- Calculation of derived metrics for advanced analysis
- Automated report generation
- Automatic command generation logic for devices through gateways
- Ability for end users to create customized dashboards
-
Cloud Infrastructure
-
-
- Creation, management, and maintenance of cloud infrastructure on Amazon AWS
-
Time Series Study
AIknow also conducted a preliminary study on the time series of energy data, aiming to evaluate the possibility of forecasting future consumption and identifying anomalous behaviors. This analysis allowed exploration of potential future developments related to predictive maintenance and optimization of energy resource usage, thanks to the use of Artificial Intelligence applied to advanced forecasting models such as LSTM (Long Short-Term Memory) for temporal sequence analysis and XGBoost for detecting nonlinear patterns in the data. Furthermore, sentiment analysis applied to feedback and textual data was considered to integrate qualitative context into predictive models.
Thanks to this solution, AIknow guarantees efficient and scalable energy monitoring, improving data management and optimizing energy resource usage for its clients.
