Groundwater Quality: The Application of Artificial Intelligence

Area of Science:

• Environmental Science
• Water Resource Management
• Artificial Intelligence

Background:

• Clean water is essential for all life, making future water quality prediction vital.
• Developing accurate predictive models for water quality is crucial for societal and economic well-being.
• Groundwater quality in the Al-Baha region of Saudi Arabia requires careful monitoring and prediction.

Purpose of the Study:

• To develop and evaluate a hybrid artificial neural network (ANN) model for predicting groundwater quality.
• To assess the efficacy of a single exponential smoothing (SES) combined with bidirectional long short-term memory (BiLSTM) and adaptive neurofuzzy inference system (ANFIS) for water quality index (WQI) prediction.
• To determine the suitability of groundwater in Al-Baha for drinking and irrigation purposes through accurate WQ prediction.

Main Methods:

• A hybrid model integrating SES for data preprocessing with BiLSTM and ANFIS for prediction was developed.
• The dataset was randomly split into 70% for training and 30% for testing.
• Model performance was evaluated using efficiency statistics, including accuracy (R) and root-mean-square error (RMSE).

Main Results:

• Both SES-BiLSTM and SES-ANFIS models demonstrated high accuracy in predicting WQI.
• The SES-BiLSTM model exhibited superior performance with R = 99.95% and RMSE = 0.00910 in the testing phase.
• The SES-ANFIS model achieved R = 99.95% and RMSE = 2.2941 × 10⁻⁰⁷, indicating strong predictive capabilities for both models.

Conclusions:

• The SES-BiLSTM and SES-ANFIS models are effective tools for accurately predicting WQI, aiding in water quality enhancement.
• The proposed models provide reliable forecasts for groundwater suitability for drinking and irrigation in Al-Baha.
• These models can be beneficial for future research on groundwater quality prediction for various purposes.