Evolution of digitalization in the field of wastewater treat

From the first until today, wastewater treatment plants (WWTP) have had to adapt to increasingly demanding functional requirements . Until practically the end of the last century, most WWTPs were built with a single objective, that the treated water met the quality requirements imposed by current regulations. At the end of the last century, a period marked by increasingly strict regulations began, not only in the quality of treated water, but in penalties for non-compliance. This means that WWTPs see automation and control as a solution to improve their operational robustness and at the same time reduce energy consumption. It is precisely during this period that purification facilities, especially large ones, experience a notable deployment of both online instrumentation and more sophisticated automatic control systems. In this way, WWTPs become more autonomous and begin to see digitalization as an indispensable technology. The latter becomes even more evident today, once WWTPs have accepted the challenge of becoming energy-neutral facilities aimed at extracting valuable resources from wastewater.

The WWTPs that we will have within a decade will largely be a consequence of how we approach the digitalization of these facilities today.
That said, it must be noted that WWTPs are complex systems made up of multiple treatment units that interact with each other, and where phenomena of a physical, chemical and biological nature intervene. Furthermore, WWTPs have three characteristics that make their operation a truly demanding Special Data task. On the one hand, they have a very marked non-linear behavior and, therefore, the effect of the operating variables cannot be interpreted individually. On the other hand, they present a lot of uncertainty , especially in biological treatment units, where biological reactions are altered by temperature, the entry of inhibitory substances, etc. Thirdly, WWTPs are systems subject to permanent disturbances caused by the variability of the wastewater to be treated. The way these disturbances are managed has a direct implication on operational costs. Therefore, given current and future perspectives of increasing complexity of WWTPs, it does not seem feasible to optimize all their functional objectives (quality of discharge, energy consumption, consumption of reagents, methane production, nutrient recovery, reuse, etc.) without exploit the immense potential of current digital technologies to store, process and infer knowledge from data.

The WWTPs that we will have within a decade will largely be a consequence of how we approach the digitalization of these facilities today. Therefore, it is necessary that digitalization be included explicitly and with a global vision in all the remodeling and in all the new WWTPs that are to come. Furthermore, for the implementation of tools such as digital twins, recommendation systems, reinforcement learning agents, anomaly detection systems, etc., it will be essential to ensure that experts with in-depth knowledge of debugging understand each other perfectly and Collaborate with artificial intelligence (AI) specialists. On the other hand, it will first be necessary to ensure that the available sensors and data sources are sufficient in quality and quantity. In addition, both advanced and low-level control systems will have to be correctly tuned to maximize their performance. Likewise, the action teams must be adequate so that the operational decisions established from the higher intelligence layers are effectively executed. In short, rethinking and responding to all these questions will be decisive in achieving an AI that, beyond the current commotion, lasts over time and is essential to operate the WWTPs of the future.

_________________
Special Data