Multiparameter water quality trend prediction technique is one of the important tools for water environment management and regulation. This study proposes a new water quality prediction model with better prediction performance, which is combined with improved sparrow search algorithm (ISSA) and support vector regression (SVR) machine. For the problems of low population diversity and easily falling into local optimum of sparrow search algorithm (SSA), ISSA is proposed to increase the initial population diversity by introducing Skew-Tent mapping and to help the algorithm jump out of local optimum by using the adaptive elimination mechanism. The optimal values of the penalty factor C and kernel function parameter g of the SVR model are selected using ISSA to make the model have better prediction accuracy and generalization performance. The performance of the ISSA-SVR water quality prediction model is compared with BP neural network, SVR model, and other hybrid models by conducting water quality prediction experiments with actual breeding-water quality data. The experimental results showed that the prediction accuracy of the ISSA-SVR model was significantly higher than that of other models, reaching 99.2%; the mean square deviation (MSE) was 0.013, which was 79.37% lower than that of the SVR model and 75% lower than that of SSA-SVR model, and the coefficient of determination R 2 was 0.98, which was 5.38% higher than that of the SVR model and 7.57% higher than that of the SSA-SVR model, indicating that the ISSA-SVR water quality prediction model has some engineering application value in the field of water body management.

No abstract

Single-frame circulation aquaculture belongs to the important category of sustainable agriculture development. In light of the visual-detection problem related to survival rate of Portunus in single-frame three-dimensional aquaculture, a fusion recognition algorithm based on YOLOV5, RCN (RefineContourNet) image recognition of residual bait ratio, centroid moving distance, and rotation angle was put forward. Based on three-parameter identification and LWLR (Local Weighted Linear Regression), the survival rate model of each parameter of Portunus was established, respectively. Then, the softmax algorithm was used to obtain the classification and judgment fusion model of Portunus’ survival rate. In recognition of the YOLOV5 residual bait and Portunus centroid, the EIOU (Efficient IOU) loss function was used to improve the recognition accuracy of residual bait in target detection. In RCN, Portunus edge detection and recognition, the optimized binary cross-entropy loss function based on double thresholds successfully improved the edge clarity of the Portunus contour. The results showed that after optimization, the mAP (mean Average Precision) of YOLOV5 was improved, while the precision and mAP (threshold 0.5:0.95:0.05) of recognition between the residual bait and Portunus centroid were improved by 2% and 1.8%, respectively. The loss of the optimized RCN training set was reduced by 4%, and the rotation angle of Portunus was obtained using contour. The experiment shows that the recognition accuracy of the survival rate model was 0.920, 0.840, and 0.955 under the single parameters of centroid moving distance, residual bait ratio, and rotation angle, respectively; and the recognition accuracy of the survival rate model after multi-feature parameter fusion was 0.960. The accuracy of multi-parameter fusion was 5.5% higher than that of single-parameter (average accuracy). The fusion of multi-parameter relative to the single-parameter (average) accuracy was a higher percentage.

No abstract