The Wervershoof digital twin: virtual ozonation plant for real-time prediction and training
The Dutch drinking water company PWN and Dutch wastewater authority HHNK joined forces to protect the IJsel lake from micropollutants, including farmaceutical residues. An advanced treatment plant consisting of ozonation was recently built at the Wervershoof wastewater treatment plant (NL).
"At PWN, source protection is key. What does not enter the source, does not have to be removed" - Ruud van der Neut, drinking water technologist PWN
A digital twin (DT) of that demonstration plant was built based on AM-Team's AMOZONE solution, with the aim of accelerating training and optimising the process. The DT is a virtual copy of the real process, running on a computer. In a second stage, the DT will run in real-time with the real plant to predict key variables that cannot be measured.
Objective of this project
Ozonation is a chemical oxidation process that destroys the micropollutants. This process is very different compared to a biological treatment process which requires fast learning of onsite staff. In addition, some key variables such as micropollutants and bromate (which is an undesired byproduct of ozonation) cannot be measured in real-time. Furthermore, process design and operation need to be optimised at all times, which is not an easy task within the dynamics of a full-scale wastewater treatment plant (WWTP).
"We are applying computer simulations to accelerate our learning when it comes to the removal of various micropollutants. Even before the real plant went into operation, we could already obtain significant knowledge with the model." - Ronald Koolen, project manager HHNK
The final aim is to obtain a real-time digital twin, which allows advanced monitoring and smart plant control.
"The real-time digital twin will enable us to optimally operate the plant and to accurately obtain micropollutant removal targets while minimising undesired byproduct formation. When the real-time coupling is made, we can really talk about a smart treatment plant" - Maaike Hoekstra, treatment technologist HHNK
Using the engineering simulation solution AMOZONE, AM-Team built digital twins of various treatment units used in this project:
- The lab- and pilot-scale units used in preliminary research (data used for model calibration and setup)
- The two 350 m³/h demonstration plants with different ozone injection systems
- The PWNT pilot for advanced water reuse
In a first stage, the 'offline digital twin' is used for learning and training purposes, next to supporting and reducing onsite testing campaigns. In a second stage, the digital twin will run in real-time with the plant for smart monitoring and control of variables that are hard to measure (eg micropollutants and byproducts).
Value for PWN and HHNK
A virtual plant allows 'what-if' testing. For example 'what if we would apply different ozone dosing strategies?' 'What if the water quality would change?'. Also, once running in real-time, smart monitoring and control are enabled.
"The digital twin will help us running a multitude of 'what if scenarios', monitoring in real-time and to enhance process understanding for better operation and design" - Maaike Hoekstra (HHNK)
The real-time coupling is foreseen in 2023.