The invention provides a water quality detection method based on a gated recursive array. The detection method comprises the following steps: S1, acquiring water quality parameter values including a PH value, conductivity, turbidity, dissolved oxygen, water temperature, content of chlorophyll A, content of blue-green algae, content of ammonia nitrogen, content of total phosphorus, content of copper, content of lead and content of tin; acquiring environmental indexes including atmospheric pressure, atmospheric humidity, atmospheric temperature and wind speed; S2, performing normalization and feature selection on the obtained water quality parameter values and environmental indexes to obtain preprocessed data; S3, selecting a neural network based on a gated recursive array as a prediction model, and setting a loss function and an optimization iteration method; S4, inputting the preprocessed data into a prediction model for model training to obtain a prediction result; and S5, comparing the prediction result with a set five-stage water quality classification to obtain a classification result. According to the method, the water quality attributes can be more effectively utilized for prediction and classification.