AI to monitor and automate gas network pressures: the innovative project for business and environmental protection

In the gas sector, one of the most sensitive issues for network operators is the monitoring and regulation of the pressure with which gas circulates in the pipes. On this, in fact, depends the balance of the gas flow as well as the quality and safety of the service offered to the end user. With a pressure above or below a certain threshold, in fact, physiological losses increase (if it is too high) or the gas does not reach the user in the right quantity (if it is too low). It follows that the monitoring and regulation of pressures is of paramount importance and that the distributor's attention must be maximised.

The network pressure

As mentioned earlier, in order to arrive correctly at its destination, the gas must be transmitted at the correct pressure, identified by technical standards in the range between 19 and 32 mBar (for non-stabilised networks). With pressures that are too high, we have a situation that is harmful to the environment and a waste of a precious resource, up to and including, in extreme cases, dangerous operating conditions. If, on the other hand, the pressure is too low, there is a risk that the gas will not reach the end user in adequate quantity or that air will enter the system. The latter case involves shutting down the system, which has to be emptied and refilled, an inconvenience that the network distributor must avoid at all costs. 

The pressure and flow rate of the gas networks also vary according to the seasons, the time of day or the vocation of the area, which influences the amount of gas used by the user. For example, during the morning, lunch and dinner hours, the amount of gas in households must be higher, because people use it for cooking or heating water. Or, in the autumn and winter seasons, the flow is higher due to heating, just as in industrial areas or densely populated centres, more gas is needed than in other sparsely populated areas. Also, during the night, consumption is lower than during the day, so the pressure may drop and the gas flow may be reduced.

Monitoring and regulating pressures using AI

In collaboration with the Politecnico di Milano, one of Europe's most prestigious universities, for some years now we have been conducting a fruitful project for the automatic monitoring and regulation of pressures through Artificial Intelligence which, fed by a number of data sources (atmospheric temperature, area vocation, consumption, pressures detected by data loggers, etc.), monitors network pressure without necessarily knowing the topology, which is often absent even from the documentation held by the operator. 

The significant advantage for distributors is the drastic reduction in the risk of supply interruption: failure to deliver gas to the user due to low pressure leads to a failure of service as well as a potentially dangerous operating condition and, for this reason, operators raise the pressure to prevent this from happening. Thanks to this intelligent threshold management, the pressure can be kept at an optimal level at all times without taking any risks.

Once again, digitisation and innovation of network infrastructures are the ideal solution to increase the quality of services, prevent economic and resource losses and protect the environment. A more structured investment in digitisation today can facilitate the management of tomorrow, in order to contribute to the sustainable management of all gas network infrastructure processes.