Italian plant finds a better way to use its data
An Italian WwTP has improved plant processing and energy efficiency by better use of sensor data. Marcel van Helten, market director of infrastructure for GE Fanuc Intelligent Platforms (Europe), explains how
The plant diverts wastewater flowing into the Lamone River, splitting it into two parallel lines, which are directed to the treatment tanks. The water is pumped back downstream into the river after treatment.
The activated sludge purification system is a biological type where organic substances and ammonia are oxidised. The nitrate products, typically eutrophying nutrients, are later removed in absence of oxygen.
Consequently, the oxygen content, the active sludge concentration, the nitrates and the ammonia are key data inputs of the plant process control system. The first steps upon entrance to the plant are grit removal and de-oiling (not managed by the control system).
The equalisation and primary decantation tanks form a vessel that controls the sewage flow rate into the various tanks by means of sluices (a simple level gauge is used for this). Then sewage reaches the oxidation and pre-denitrification tanks, where the level of oxygen in the slurry is measured at the inlet and at the outlet. The nitrates and suspended solids are also measured in these tanks (by means of turbidimeters, designed specifically for measuring using optical techniques), along with the phosphorous and ammonia contents, the level of decanted sludge and the inlet and outlet water flow rate.
Some of the output sludge is recirculated back to the inlet and re-introduced to improve the biological processes. After oxidation, the water flows to the secondary decantation tanks where the sludge deposited on the bottom is collected and conveyed to the thickener. Here, the sludge is prepared for drying and disposal. The clarified water is released into the river.
The water treatment plant is, due to its intrinsic nature, subject to seasonal rainfall. Consequently, one of the process criticalities is that the quality of the water to be treated cannot be determined beforehand. Furthermore, the plant collection basin includes a number of industries, which introduce large amounts of waste, thus the water chemistry and flow varies greatly.
Another criticality of a plant like this, with such an extensive coverage, is that it is always on. This is essential to prevent the release of polluted water and to prevent being fined by the water quality monitoring authority.
Previously the plant was run according to a fixed-time logic. This consisted of leaving the sewage to stand in the various vessels for a predetermined length of time and controlling the operation of the process-related machines such as aerators, blowers and pumps. This was judged according to dissolved oxygen measurements and laboratory test data only.
The goal set by Giovanni Tedioli, water treatment manager of Hera Imola - Faenza, was to use the data collected by various sensors to control the transit times of the sewage in the tanks. This involved adjusting the machine operation according to the values of oxygen, ammonia, suspended solids, and nitrates, to improve plant processing and energy efficiency.
Furthermore, the new control system had to allow an operator to work at the plant as well as relaying data to a universal control room, which monitors all Hera plants. The plants are manned during the day, but the control room alone monitors the operation of all plants during the night.
Massimo Zanoni, electrical maintenance, automation & remote control manager of Hera Imola - Faenza, recollects the project start-up: "When we decided to refurbish the plant, we asked ourselves how to make sure that the new automation system would guarantee our peace of mind. The water treatment plant releases water into our own rivers and this implies additional responsibilities towards society; we need to guarantee faultless operation, for ourselves and for our environment."
The "peace of mind" Zanoni mentions had to be translated into high plant availability and reliability, data access by operators and improved process management, in terms of better results and more efficient use of energy resources. To reach these goals, Hera called Pastorelli's Environmental Engineering to establish the project guidelines.
The process was made by Novanet, a local system integration and engineering firm with major expertise in the construction of large control and automation procedures.
Hera asked Novanet to use GE Fanuc products for implementing the control system.
The "brain" of the system is a GE Fanuc PACSystems RX3i in redundant hot backup configuration, which interfaces with all the field instruments on a Profibus network (part optical fibres and part copper wires); there are several new and old sensors in total, amounting to approximately 600 controlled tags. The two redundant central processing units ensure the high plant availability required by the application criticality.
The programmable automation controller establishes the standing times of the slurry in the various stages of the plant. By means of a direct Modbus/transmission control protocol link, the PAC communicates data to the Hera control room, where they are stored in a structured query language (SQL) database and concisely displayed so that the operator (present 24 hours a day) can be warned of faults and act accordingly.
At the Formellino plant, a local computer, running GE Fanuc's Proficy HMI/Scada Cimplicity software, collects, monitors and displays information and data in the form of trend or log. In addition, it monitors alarms, which may be silenced or not by the users according to their access levels.
Ten profiles corresponding to ten different operative and data access levels have been created according to the privileges established for each user class. The application allows set up and programme control parameters (the plant has been running only for a few months and the control logic is still being fine-tuned). Many fault-detecting functions have been implemented in programmes running at the WwTP to signal measurements deviating from expected values and to collect and use self-diagnostic data from the field sensors.
Novanet's technical manager, Alberto Tabanelli, says: "We used GE Fanuc products in this plant for the first time and, despite the complexity of the logics and the installations, we encountered no problems at all. The PLC hot backup function provided default hardware redundancy, which avoided further complications, and the system performance allowed us to introduce a predictive control that has greatly improved plant performance."
The new system collects plant data for constantly monitoring everything in detail. Predictive control, sensor data collection and use and control system response rapidity have been exploited to optimise machine running times and consequently decrease energy consumption while keeping the water quality high.
Before installing the new system, for example, the water was over-oxygenated and this was pointless from a microbiological point of view.
The Formellino automation system has been running for only a few months and the implemented logics are still being optimised. After only 50 days, an energy consumption of 30% has already been observed.
The plant was shut down for approximately half an hour for installation. Personnel training was swift, thanks to intuitive, self-explanatory graphic displays and was carried out over several shifts to account for staff turnover.
New actuators, which will be controlled continuously instead of in steps, will be added in the future. They will be installed on the Profibus field network and controlled directly by the PACSystems Rx3i. These improvements will provide the best results where the processed matter is kept moving such as the oxidation and sludge recirculation tanks.
In summary, the benefits of the project at Formellino WwTP are that all plant data is collected and used for predictive calculations and for optimising process efficiency. The plant's water purification process means improved water quality and better control of crucial river habitat parameters. The plant has also achieved a 30% energy saving.