How i2o works - The System
As described in the section above, the aim of Advanced Pressure Management is to keep P3 as stable as possible and just above P3ref. As the head loss in the DMA between the PRV and the critical point changes with changing demand patterns, P2 must be continually adjusted in order to achieve this.
Controllers used to date for Advanced Pressure Management have tended to use a fixed table that gives a value for P2 depending on the flow. There are a number of drawbacks to this system; the table has to be entered manually following a logging exercise, the controller does not take into account demand patterns that vary depending on the day of the week or time of year, neither does it take into account changes to the demand pattern in the DMA caused by a factory closing or new house building. Because of these uncertainties, it is usual to enter a very conservative table which limits the water that can be saved.
The i2O system overcomes these problems by using self learning algorithms. These algorithms are created automatically by the i2O server and downloaded daily to the controller. They are updated on a daily basis, taking into account the latest data available. In this way they are always optimised in order to save the maximum amount of water.
The controller communicates with the i2O server once a day using the GSM network. During each communication, the previous 24 hours pressure and flow data are uploaded to the server and an updated control algorithm is downloaded to the controller.
A sensor at the critical point (P3 sensor) also sends pressure data for the previous 24 hours to the i2O server over the GSM network. The P3 data is required to update the control algorithm and to monitor the performance of the system. The P3 sensor also sends an alarm if P3 falls below P3ref.
A further weakness in existing Advanced Pressure Management systems is the method of adjusting P2. This is done by using solenoids to pulse either water or compressed air into a bias chamber that exerts a variable force onto the conventional pilot valve spring. The compressed air version of this system has a short battery life, limited range and is vulnerable to valve chamber flooding whereas both systems rely on solenoids which can be unreliable.
i2O has overcome these issues by developing the APV (advanced pilot valve). This has been designed to provide an adjustable P2 and is simply connected to the PRV in place of its conventional pilot valve. The APV smoothly adjusts P2 in response to instructions from the i2O controller.
The i2O server plays a number of vital roles; it receives and stores data, it updates the control algorithms on a daily basis and downloads these to the controller, it handles alarms and provides reports both to the water company and to i2O.
Where required, a sensor can also be installed at the AZP point (the AZP sensor). This can measure the AZP pressure and send the data back on a daily basis to the i2O server. Where an AZP sensor is not installed, the average of P2 and P3 is calculated (P4) and is used as a proxy for the AZP. The change in AZP or P4 is measured or calculated in order to estimate the level of water saving from the system.
In some DMAs, the critical point may be at different locations depending on the time of day. In this case, a P3 sensor is installed at each critical point. The control algorithm works in the same way, maintaining P3 just above P3ref at each different critical point in turn.