There are several risks to using artificial intelligence and machine learning at a water treatment plant. These include:
Coagtech addresses each of these risks through our plant-minded, engineered approach. Reach out to us to learn how.
The coagulation and treatment process is used to support operators in protecting the health and well being of the public. When modelling this process, specialized engineering knowledge is required. We feel strongly that it is inappropriate for anyone other than an engineer to complete these exercises, and that the models should be backed by an authenticated and validated engineering report.
Read more about the importance of engineers roles in the tech evolution here.
Optimizing the coagulant and polymer dosing to maintain high quality treatment reduces financial and operational impacts to the water treatment plant and its customers. Water treatment plants are facing significant inflationary pressures in chemical costs, which ultimately financially impacts the end user.
Another benefit optimizing coagulation is a reduction in treatment residuals. In most treatment plants, particles removed in the process and the associated coagulants are either discharged back to the water body or handled as part of a residuals management program. If they are discharged back into water body the chemicals may have a negative environmental impact. If they are diverted to a residuals management process then additional energy and chemicals are expended for removal, after which the dewatered residual is typically disposed of at a landfill. By reducing the amount of coagulant the environmental impact of treatment is reduced.
With the right instrumentation and adequate historical data, this technology can provide valuable recommendations and be a significant benefit to any water treatment plant. The first phase of the adoption process is a system evaluation, where an on-site process review is completed. This review will be used to identify any process blind spots where additional instrumentation is required. As part of the system evaluation the available historical data is reviewed for data integrity issues, and preliminary model development is completed to assess the feasibility of implementation.
Should the report find that there are process blind spots or data integrity issues it will flag these areas and the required changes to obtain the needed data for a successful implementation. After the recommendations have been implemented and data is recorded for two years the system can be re-evaluated for implementation.
Yes. We use hardware and communication protocols that are standard across the industry. Reach out to us to discuss your specific requirements further.