Use case: Weather-predictive energy optimisation
Dynamic, adaptive plant management through a solution combining IoT, analysis of real and forecast weather data, and artificial intelligence
Partners involved
Loogos Srl (solution provider) and the Municipality of Parma.
Benefits
Operating data from the plants and data on indoor and surrounding environmental conditions are acquired, monitored and sent to the Cloud, where an Artificial Intelligence algorithm analyses them together with weather information specific to the area, processes them and determines the necessary corrections, which are then transferred to the system to optimise its overall performance.
The Challenge
Traditional approaches to energy efficiency focus mainly on replacing or modifying existing plants. However, these methods tend to ignore the many external factors that can affect the plants' energy consumption. Usually, the aim is a static, long-term optimisation designed to be effective in any condition. This kind of optimisation, however, does not take environmental variations into account and only adjusts plant settings once conditions have changed. As a result, it fails to achieve an optimal energy efficiency that considers the dynamic behaviour of the environment.
The lack of preventive and continuous adaptation to changing external conditions severely limits the potential for energy savings and the sustainability of the plants themselves.
To achieve better results, what is needed is a more integrated and flexible approach, able to respond proactively to external changes, thus maximising efficiency and reducing waste. Only through dynamic and adaptive plant management is it possible to achieve truly sustainable energy efficiency.
The challenge was proposed by the Municipality of Parma which launched a call for tenders to identify innovative solutions to improve the energy efficiency of some of the city's schools.
The Solution
The call allowed the Municipality of Parma to select the energy optimisation project proposed by Loogos Srl and to apply it to two local schools: the Scuola per L'Europa and the Istituto Comprensivo Albertelli-Newton.
The weather-predictive paradigm, a rewarding and decisive factor in awarding the contract to Loogos, involved an approach to weather forecasting based on theanalysis of historical data, of current observations and on mathematical models to estimate future weather conditions in that specific geographical area.
The adoption of the innovative and advanced technologies used in the project, such as Artificial Intelligence, predictive models and the IoT which can truly revolutionise the way forecasts are generated, made it possible to analyse a wide range of real-time data from the perimeter and external sensors installed and from the weather data provided by satellite updates, producing increasingly accurate forecasts capable of detecting micro-variations in the two very small areas.
The integration of weather and environmental data with the operating data of the HVAC systems (heating, ventilation and air conditioning) enabled a prediction and preventive correction of the behaviour of the two individual plants. This synergy and correlation of data is made possible by a sophisticated Artificial Intelligence algorithm, which processed and analysed the data in real time, allowing immediate corrections.
Thanks to the environmental sensors, it was possible to constantly monitor indoor conditions, guaranteeing an optimal level of comfort while optimising energy consumption. This approach therefore makes it possible not only to prevent inefficiencies or malfunctions of the HVAC systems, but also to maximise the overall energy efficiency of the buildings or spaces where it is installed.
These data are sent to the Cloud via the acquisition system, where an Artificial Intelligence algorithm analyses them together with weather information specific to the area. The algorithm then works out the necessary corrections, which are transferred to the system to optimise its overall performance.
The system constantly processes plant information together with weather information and determines the optimal corrections to apply.
The Slave devices acquire data from the various sensors and send it to the Master device, which has a connectivity module that allows it to send the data to the Cloud processing system.
The Master device then receives the corrections to apply to the plant and, through the Slave devices, sends commands to the actuators installed on the control components.
All data collected by the sensors and the changes made by the system are accessible through a dedicated online platform, which provides a detailed, real-time picture of how the plant is working, allowing users to monitor the energy consumption of every single component.
This platform also makes it possible to detect anomalies or signs of wear in components, dynamically adapting the maintenance schedule and intervening promptly on parts that need servicing. Thanks to this in-depth analysis and timely intervention, the plant's service life is extended.
The EVO Method
A monitoring period is started to evaluate how the system works. Initially, the system is activated on alternate days to check that it is working correctly. This approach makes it possible to use the degree-day method to compare the plant's consumption when the predictive system is on and when it is off. This method, called Evo, makes it possible to quickly obtain an accurate assessment of the level of optimisation achieved in the plant in the short term.
The Degree-Day Method
The degree-day method measures the energy demand needed to heat or cool a building based on the difference between the desired indoor temperature and the outdoor temperature. It then calculates the sum of the daily differences from a balance temperature. It is used to assess energy efficiency and compare consumption.
COMPOSITION OF THE INSTALLED SYSTEMS
MASTER CONTROL UNIT
Real-time indoor temperature function, managed online via App. + 4 PROBES (Outdoor Temperature, Single-floor Temperature, Water Supply Temperature, Water Return Temperature). HOT / COLD circuit.
- Panel power meter with 3 Rogowski probes
- Slave control unit for the hot circuit
- Wireless temperature sensors with Mesh technology
- Wireless temperature and humidity sensors with Mesh technology
The chart above analyses the month of April at the Albertelli Newtonschool, using the system in Evo mode. To make the consumption data comparable, they were normalised using the degree-day differential (kW/degree days).
The chart shows that, on average, daily consumption with the system running is lower. Towards the end of the month, a sharp drop in temperatures was anticipated and managed by the system, reducing consumption peaks and substantially increasing average savings, which by the end of the period reached 27%.
In conclusion, the recorded performance confirms that, in a constantly changing climate, the technologies tested in this project are an effective response. They can adapt indoor climate control in real time, maintaining the right balance between indoor comfort and energy savings.
ADVANTAGES
The main advantage of adopting a system of this kind is that the level of efficiency keeps growing steadily.
Thanks to continuous learning of the behaviour of the indoor spaces and to the improved prediction of external environmental changes, the system is able to progressively increase its precision and reliability.
Ecosystem partners involved

Loogos
Loogos develops solutions based on IoT, data analytics, and AI to make systems and buildings smarter, especially through…
Parma (PR)









