即時精準多層式感測器改善給水廠污泥操作之研究

Abstract

近年物聯網 (Internet of Things, IoT) 與大數據正蓬勃發展中，然而現今淨水處理過程中，大多數污泥排放都是透過傳統方法來控制，例如：定時排放和經驗豐富的操作人員適時掌控。傳統淨水污泥排放機制並未考慮沉澱池的污泥濃度，因此排放污泥的機制不能依據水質進行調節，傳統控制方法的缺點是低能量效率和低應變能力。 本研究透過使用「即時精準多層式感測器 (Real-Time Precise Multi-Layer Sensor, RPMS)」進行水池污泥監測，以判斷污泥位置。本研究於大湳給水廠二次污泥濃縮池中成功監測並判斷出污泥位置，由RPMS所測得之污泥厚度與現場採樣的污泥厚度位置誤差為50公分以內。並且於乾旱時期、乾旱逐漸解旱期、颱風降雨前後，在平鎮給水廠高速膠凝平板式污泥毯澄清池進行淨水程序加藥、天氣變化與污泥毯澄清池中進行水質監測，發現天氣對原水水質有明顯的影響，但現行淨水操作條件並未能夠即時針對最適加藥量進行調整。 本研究利用RPMS 進行污泥層監測後，得出水中TDS與SS對相對導電率 (Relative Electrical conductivity, R-EC) 均會產生影響，其中TDS對R-EC影響約為SS對R-EC影響的100倍，然而同一日的水池中各深度TDS變化幅度不大，因此在同一水池中，僅SS會隨深度變化影響R-EC，因此可以透過R-EC變化判斷污泥層位置。 RPMS具有良好的污泥層與清水層的判斷能力，同時亦可監測清水層沉澱後的水質狀況，亦具有作為水質預警、物聯網與大數據的發展潛力，以期可為國內自來水供水品質帶來改善。

In recent years, there has been a flourishing development of the Internet of Things (IoT) and big data. However, most sludge discharge in the current water treatment process is controlled using traditional methods, such as timed release and the expertise of experienced operators. The conventional sludge discharge mechanism does not consider the sludge concentration in the sedimentation tank, resulting in an inability to adjust the sludge discharge based on water quality. The drawbacks of traditional control methods include low energy efficiency and limited adaptability. This study introduces the use of a Real-Time Precise Multi-Layer Sensor (RPMS) to monitor sludge levels in water tanks and determine the position of the sludge. We successfully tracked and accurately determined the positions of sludge in the secondary sludge thickening tank of the Da-Nan Water Plant. The deviation between the sludge thickness measured by RPMS and the on-site sampling was within 50 centimeters. Water treatment processes and water quality monitoring occurred in the high-speed coagulation plate settling tank at the Pingzhen Water Plant during dry periods, drought relief, and before and after typhoon rainfall. The study revealed a significant impact of weather on the raw water quality. However, the current water treatment operating conditions could not adjust the optimal dosage in real time. By utilizing RPMS for sludge layer monitoring, this study found that total dissolved solids (TDS) and suspended solids (SS) both had an impact on the relative electrical conductivity (R-EC) of the water. The influence of TDS on R-EC was approximately 100 times greater than that of SS on R-EC. However, the daily TDS variation within the same water tank was insignificant. Therefore, only SS varied with depth and influenced R-EC, enabling the determination of the sludge layer position through changes in R-EC. RPMS demonstrated excellent capability in distinguishing between sludge and clear water layers, as well as monitoring the water quality after sedimentation. It also showed potential in water quality forecasting, IoT, and big data development, aiming to improve the domestic tap water supply quality.