Implementation of an AI ready BACS system in Treviso school with DCV (Demand Control Ventilation)

The pandemic has highlighted the extreme importance of indoor air quality (IAQ). Especially in schools, so places that have always been critical for the spread of highly contagious viral diseases, IAQ is an increasingly necessary need. In the province of Treviso, a tailored system has been installed: there are both winter and summer air conditioning, plus AHU’s capable of guaranteeing the necessary air exchange to obtain a healthy level of the air inside the classrooms. The county also intends to replicate this experience in all schools of its territory and monitor the operation of these plants continuously. The important matter is not only to implement systems equipped with HRV (heat recovery ventilation), but also to equip them with automation and control systems capable of implementing an high automation standard according to EN 15232-1 and a high SRI or Smart Readiness Indicator defined by the EPBD III directive 844/2018. A high index ensures an implementation of a distributed intelligence system all over the plant, connected to the cloud and capable of extrapolating the physical variables, then processing them with ML (machine learning) and AI (artificial intelligence) techniques optimized for the HVAC algorithms. At the end, it is not enough to detect quantities such as EI (environmental index) but it is also necessary to process the data for the purposes of predictive maintenance and much more.


Introduction
The pandemic has indelibly marked our lives and the approach to the design of air conditioning systems, focusing on the quality of the air. As is well known, we spend 90% of our time inside the buildings where we live, work, study, etc. Before the pandemic, the issue of air quality was not so acute, so the few people who had the foresight to think about it before, now find themselves ahead of the others.
In this case we are talking about the far-sighted foresight of Eng. Fabio Fabris and Eng. Sandro Michielin of the Province of Treviso, who were grappling with the expansion of an agricultural technical institute in which an additional part of 14 classrooms has been built. The systems were designed through the most current methods (BIM) considering both air conditioning and lighting, but above all the quality of the environments at a time when COVID had not yet arrived in Italy.
What then fully realized the futuristic vision of the Province and its designers in this plant is the adoption of controllers based on a truly innovative visual programming system and even more than the adoption of a BMS based on the Haystack system. Through this system it is possible to combine BMS in the field with high-level software such as cloud Computing, Machine Learning and artificial intelligence.

Expansion of sartor institute
As shown in figure 1, the expansion of the State Institute of Secondary Education called "SARTOR INSTITUTE" located in the Municipality of Castelfranco Veneto, via Postioma di Salvarosa, 28, consists in the construction of a new body of n. 14 classrooms and related toilets, inside the complex.
The school complex already includes the following buildings: -Body A Ex Convitto, in which they find accommodations of the classrooms, a boiler room and the kitchen of the school canteen; -Body B Gym for use by the Institute; -Body C School Building Agricultural Building Sartor, with classrooms and administration; The following floor plan shows both the existing buildings and the inclusion of the new school extension covered by this intervention

Air conditioning system
The thermal system aims to perform mainly the winter conditioning of the new rooms constituting the expansion of the school complex and guarantees: • heating of all environments in winter; • supply energy to the coils of heat recovery devices. It is also planned to operate the plant in cooling regime for the period of the midseasons, approximately 15 April-15 June and 15 September-15 October, in order to guarantee better comfort of the occupants in case situations of thermal accumulations arise inside the classrooms due to the high crowding.
The thermal power required for the new plant was determined on the basis of the project data resulting from the calculation of the values of winter thermal dispersion (Law 10/91) and subsequent D.P.R. and UNI EN standards (Annex 2).
The heating system as a whole is mainly composed of: • N°1 reversible water/water type heat pump with condensation by geothermal probes; • N° 14 mechanical ventilation units with heat recovery/free-cooling, one for each classroom, for primary air treatment; • N° 4 secondary circuits starting from the supply manifold divided as follows: ▪ ventilation unit circuit; ▪ circuit plant radiant ceiling classrooms east wing; ▪ circuit system radiant ceiling classrooms west wing; o radiant floor system circuit corridors and toilets; • N°1 circuit plant geothermal probes.
• N°1 dedicated heat pump to produce DHW The emission system in the room will consist of radiant ceiling panels inside the classrooms and radiant floor panels in the corridors and inside the toilets.
Main features related to the heating conditioning system

The school project
The province has a role to play in an innovative design approach, but it should actually be normal nowadays. In fact, the enlarged property is an NZEB (EPC in Annex 1) even if the obligation starts only from 1.1.2021. The generation of the heat transfer fluid is ensured by a geothermal heat pump powered by a photovoltaic plant. The heating is ceiling panels with DCV for single classroom, latest generation fixtures and LED lighting. Let's analyze deeper the aspects.

The building and the structures
The province The enlarged portion of the building is structured as follows: • Very low thermal transmittance coating -0.182 W/(m 2 K) • Windows fixtures -max 1.4 W/(m 2 K) All data are in the technical report Italian Law 10/91 in annex 2.
The outside of the classrooms is shaded by the extension of the ceiling.

Classroom
Each classroom is designed to: 1. Hydronic air conditioning with radiant ceiling panels 2. DCV with heat recovery and hot/cold post coil 3. Temperature probes, relative humidity and CO2 Winter and summer air conditioning is ensured by ceiling panels with interception valve controlled by the area controller.
The DCV is made through an AHU dedicated to each classroom equipped with • fans to inverter 0-10V • high-yield heat recovery • post coil • dampers

Fig. 5. Diagram of a classroom
The presence of the DCV is far from obvious because the design of the technology started before the COVID-19 emergency. This means that the Province of Treviso had already taken into account the problem of air quality at a time when many did not care about the problem. This will be discussed in the next section

The distribution
The hydraulic circuit consists of a two-tube manifold with these deliveries: • for the AHU post batteries of the DCV of the classrooms • mixed for the East classroom panels • mixed for West classroom panels • mixed for corridor panels and hygienic services

Production
Hot/cold production is ensured by a heat pump with geothermal probes, 105.4kW power, COP 3.61

Production The photovolitaic fielb
A photovoltaic field has been installed in the roof using photovoltaic cell modules in polycrystalline silicon with a nominal power of 30.8 kW.

Control
Control is one of the most important parts for the efficiency of the whole system while ensuring the correct parameters of comfort and healthiness of the rooms.
A programmable hardware was used with an innovative system for simplicity and immediacy. Such a system is a CAD that allows a normal thermotechnical operator to make programs for PLC without having any preparation of programming languages. To program, in fact, it is enough: • open the programming software • click on the add icon of a form marked with a green "+" In the classrooms, the controlled parameters are: • temperature • relative humidity • air quality in terms of CO2 concentration The most interesting part is related to air quality, which denotes a top-down aeraulic circuit, having vents directed towards the windows and the air recovery done in the opposite side of the classroom, at the bottom. In this way any viral charges and other pathogens are directed down and the probability of contagion is reduced  Fig. 10. Aeraulic diagram of a classroom made with BIM technology.
The regulation was placed near the AHU and accessible by inspection hatch Fig. 11. The AHU adjustment panel of the class.
The temperature regulation logic is achieved through on-off valves with proportional regulation and setpoints according to a time schedule. Room temperature fluctuations are minimal but are greatly affected by the endogenous heat developed by the pupils present.  Fig. 12. Classroom temperature logger The most interesting regulation is done on IAQ. In fact, depending on the concentration of CO2, typically 1000 ppm, the controller acts on the speed of the AHU fan. The probes are connected in ModBus to the management system called "WEBGARAGE" which communicates the information to the field regulation system always through ModBus. The CO2 quantity analyses carried out on May the 4h this year certified the correct functionality of the system. The air terminal regulation in the environment determines the adjustment in the central plant and the distribution in a perspective EN 15232-1 [1] ensures that the demand signals generate just the power necessary for the heat generator and the exact flow rate to the distribution pumps. The setpoint of the heat pump is determined by the demand of the AHU post coils, while the heat panels are governed through climatic curves.

Scada
The architecture of the BMS system is structured with: . This is a data collection standard so that it is very easy and fast for other applications to access it reliably and securely (REST API). If we add this feature to the remote management mentioned earlier, you can easily understand that an external software, based on artificial intelligence and machine learning algorithms, run by the cloud, can process in real time not only this data, but potentially the data of all the other plants that have been made through the same technology and computer structure. For example, thanks to this technology, ENEA is developing a data collection project of different flats aimed at creating a web portal that compares real consumption with theoretical consumption. They can create a reference baseline and showing similar building energy behavior to tenants, anonymously. Thanks to this information, they are enhancing competition between tenants, in order to generate a kind of "champions of energy saving".  Fig. 18. Possibility of using AI and ML software with the Haystack protocol

Conclusions
The Province of Treviso, and its foresight to install the DCV in schools and a very advanced BMS system, has opened up a path which it is hoped will lead the same province to multiply this technology in other schools, but above all it is an example of how the plants in which we all live should always be built to obtain healthy and energetically efficient environments