Ứng dụng mô hình học sâu trong dự báo chất lượng nước mặt hệ thống thủy lợi Bắc Hưng Hải

0 Người đánh giá. Xếp hạng trung bình 0

Tác giả: Hải Dương Hà, Đức Phong Nguyễn

Ngôn ngữ: Vie

Ký hiệu phân loại: 004 Data processing || Computer science

Thông tin xuất bản: Tài nguyên nước 2023

Mô tả vật lý: 61-72

Bộ sưu tập: Metadata

ID: 408690

Thêm vào giỏ Liên kết toàn văn

Tóm tắt
Chủ đề

Hệ thống thuỷ lợi Bắc Hưng Hải nằm ở Trung tâm đồng bằng Bắc Bộ thuộc vùng châu thổ sông Hồng và là hệ thống lớn nhất miền Bắc. Nguồn nước của hệ thống rất quan trọng đối với phát triển kinh tế - xã hội của các địa phương trong vùng nghiên cứu. Tuy nhiên, ô nhiễm nguồn nước có xu hướng gia tăng cả về phạm vi và mức độ, các chỉ tiêu COD, NH4+, PO43- và Coliform có mức tăng cao hơn so với các chỉ tiêu khác. Hàm lượng COD tăng 8,6 lần, NH4+ tăng 2,48 lần
PO43-, tăng 4,15 lần và Coliform tăng 91,6 lần. Hằng năm, chất lượng nước mặt được giám sát (chủ yếu vào vụ Đông - Xuân) để phục vụ đổ ải. Phương pháp giám sát chất lượng nước chủ yếu là phương pháp truyền thông tốn nhiều thời gian, công sức và thường cung cấp thông tin hạn chế về chất lượng nước. Trong nghiên cứu này sẽ đảnh giả tiềm năng của các thuật toán học sâu trong dự báo các thông sổ chất lượng nước của hệ thống thủy lợi Bắc Hưng Hải dựa vào các thông số đo hiện trường (nhiệt độ, pH, DO và độ đục). Kết quả nghiên cứu cho thấy các thuật toán học sâu có thể đưa ra các dự báo chính xác và nhanh chóng đối với các thông số BOD5, NH4+, PO43- và COD. Các mô hình có độ chính xác cao (từ 75% đến 85%), đồng thời cung cấp các kết quả có ý nghĩa. Nghiên cứu cũng đã chứng minh được tiềm năng của các thuật toán học sâu để cải thiện việc dự báo chất lượng nước và nhấn mạnh tầm quan trọng của việc lựa chọn các thuật toán học sâu phù hợp cho các thông số chất lượng nước khác nhau., Tóm tắt tiếng anh, Bac Hung Hai Irrigation System is located in the center of the Northern Delta in the Red River Delta and is the largest system in the North. The water source of the system is very important for the socio-economic development of the localities in the study area. However, water pollution tends to increase in both scope and extent, indicators of COD, NH4+, PO43- and Coliform have a higher increase than other indicators. COD content increased 8.6 times NH4+ increased 2.48 times
PO43- increased 4.15 times and Coliform increased 91.6 times. Every year, surface water quality is monitored (mainly in the Winter-Spring crop) to serve the drainage. Water quality monitoring methods are mainly traditional methods that are time consuming and labor intensive and often provide limited information on water quality. In this study, we will evaluate the potential of deep learning algorithms in predicting water quality parameters of Bac Hung Hai irrigation system based on field measurements (temperature, pH, DO and turbidity). Research results show that deep learning algorithms can make accurate and fast predictions for parameters BOD5, NH4+, PO43- and COD. The models are highly accurate (from 75% to 85%), and provide meaningful results. The study also demonstrated the potential of deep learning algorithms to improve water quality forecasting and highlighted the importance of selecting suitable deep learning algorithms for different water quality parameters.

Tạo bộ sưu tập với mã QR