Artificial intelligence (AI) is an emerging, powerful, novel technology that can model real-time problems involving numerous intricacies. The modeling capabilities of AI techniques are quite advantageous in water purification and wastewater treatment processes because the automation of such facilities resulted in easy and low-cost operations, in addition to a significant reduction in the occurrence of human errors. AI technologies involve multi-linear or nonlinear relationships and process dynamics that are usually impractical to model by conventional methodologies. 

Three-quarters of the earth's surface is covered with water, an essential element to sustain human life. Several methods have been employed to analyze and predict water quality and wastewater quality, including coagulation/flocculation techniques, desalination, membrane filtration, biological oxygen demand (BOD), and chemical oxygen demand (COD). These techniques mainly utilize mathematical models and linear regression algorithms as predictive models to simulate and determine the operating system's parametric relationships among various process variables. However, these conventional methodologies are time-consuming, require lengthy procedures, and are unable to map the massive non-linearity and complexity of systems. Furthermore, they are typically simplified based on assumptions and ideal considerations that are not practical. Although the empirical and statistical regression models can show acceptable predictions, these regression models cannot deal with the overall nonlinear relationships and complicated dynamics which largely exist in water treatment processes.

AI in water purification

AI technologies have largely revolutionized today's industrial sector, termed Industry 4.0, and have triggered research in many science and engineering fields, such as intelligent robotics, natural language processing, material design, disease diagnosis, and medicine. The natural intelligence of humans and certain behaviors are impersonated in technical systems constituting non-parametric algorithms imitating natural human brain functions, such as learning and interpreting. AI technologies, such as artificial neural networks (ANNs), deep learning (DL), support vector machine (SVM), genetic algorithm (GA), and fuzzy logic (FL), can perform independent analysis, evaluation, and prediction according to the input data, optimize the system variables, or send out warning signals to analyze parameters and adjust the output accordingly; this greatly reduces human errors and improves productivity.

Phoenix experiments

In this context, the capital of the southwestern US state of Arizona is using AI for improved wastewater monitoring. The city's water services department launched a six-month wastewater treatment pilot program with AI. According to Nazario Prieto, assistant water services director for Phoenix, the system is adding eyes through their sewer mains 24/7.

The city's Water Services Department believes the Arizona Department of Environmental Quality may soon require sewer monitoring, and these devices will help them meet those conditions. With 16 devices installed in sewer drains across Phoenix, the city is the fourth US state to utilize this AI technology.

Conclusion 

The main purpose of AI technologies in a system is to boost computer functions pertinent to human knowledge, such as determination of issues, perception, and cognition. In the last decade, several studies have applied AI models owing to their comfort of use, high-speed process, and acceptable error without comprehending physical problems. 

AI has performed better than conventional modeling procedures in various water treatment processes. The successful implementation of these technologies stimulates further research and innovation in the model structures to overcome some of the limitations that paved the way for the effective operation in water treatment industries.