201118362 六、發明說明： 【發明所屬之技術領域】 本發明係有關於一種廢水處理之監測技術’特別疋有 關於一種監測化學沉澱成效之方法。 【先前技術】 水及廢水處理是由許多不同的物理化學及生物程序所 組成，在眾多處理程序中，化學沉澱是一相當重要的程序。 一般的廢水處理或是學術研究利用以了方式分析膠羽長 成：傳統的粒徑檢測分析儀器、線上濁度計判斷廢水處理 過程中膠羽長成趨勢及光纖膠羽偵測儀（Photometric dispersion analyzer，以下可簡稱PDA)。傳統的粒徑檢測分 析儀器雖然能夠精確的測量出廢水處理過程中膠羽的長 成，但有下列限制： (a) .採樣過程可能破壞膠羽，無法真實反應系統中膠 φ 羽長成。 (b) .樣品必須以經驗或嘗試錯誤法(trial and error)稀 釋至適當濃度才可分析。 (c) .雖然能夠準確的分析出粒徑大小，但設備相當昂 貴（約150〜200萬新台幣），對小型的處理系統而言，是相 當大的負擔。又對混凝或化學沉澱而言’粒徑資訊雖然直 接反應出處理成效，但實際上不是必要資訊，因此在實廠 廢污水處理中亦非要求之水質項目。 (d) .膠羽過大過重時，多在監測槽發生沉澱，因此無 201118362 法量測。 線上濁度計可判斷廢水處理過程Ψ膠羽長成趨勢，然 而，其具有以下缺點： (a) .濁度本身受色度干擾，雖可利用紅外光為光源匠 低色度干擾’減少混凝過程中色度變化對於懸浮固體物散 射光訊號的影響，但以銅的化學沉殿為例，膠羽形成及沉 澱時，色度亦可反應出銅去除效率，因此單純的去除色度 因子無法真實地判斷化學沉澱之成效。 (b) .由於濁度計之刻度較粗略，例如濁度為1〇〇 NTU 時，水樣貫已相當混濁，因此濁度計在膠羽形成之初期， 其訊號跳動範圍較小，不易觀察光訊號之變化。 光纖勝羽摘測儀(PDA)係運用彎管模擬慢混，將水樣送 入積測裝置’研究_示模擬慢混之結果與瓶杯試驗之慢 混結果相似，但仍非真實反絲羽在慢混池之長成。光纖 勝羽偵測儀(PDA)之輪出圖乃經過統計分析後而得，監測時 所選疋參數@0^差等）會影響統計後之數據，影響混凝 結果之判讀。 【發明内容】 根據上述問題’業界需要於快混及/或慢現初期就可以觀 察到廢水處理纽的杨，此方法*僅可以數據即時分析 廢水處理H並可以提供處理廢水的影像，使工薇人員 可由影像直接觀看膠料小與結構，立即得知膠羽之大小 與分佈。 201118362 本發明提供一種監測化學沉澱成效之方法，包括：將一 廢水抽至一容器中；以一光源照射容器中的廢水；以一影 像擷取裝置’擷取廢水的影像；及分析影像中RGB值，以 檢測廢水之化學沉澱成效。 為讓本發明之上述目的、特徵及優點能更明顯易懂，下 文特舉一較佳實施例，並配合所附圖式，作詳細說明如下：201118362 VI. Description of the Invention: TECHNICAL FIELD OF THE INVENTION The present invention relates to a monitoring technique for wastewater treatment, particularly to a method for monitoring the effectiveness of chemical precipitation. [Prior Art] Water and wastewater treatment are composed of many different physical and chemical processes, and chemical precipitation is a very important procedure in many processes. General wastewater treatment or academic research uses the way to analyze the growth of rubber feathers: traditional particle size detection and analysis instruments, online turbidity meter to judge the growth trend of rubber feathers in the process of wastewater treatment and the optical fiber feather detector (Photometric dispersion) Analyzer, the following can be referred to as PDA). Although the traditional particle size analyzer can accurately measure the growth of rubber feathers during wastewater treatment, it has the following limitations: (a) The sampling process may damage the rubber feathers and fail to truly reflect the gel φ plume in the system. (b) Samples must be diluted to the appropriate concentration by experience or trial and error before analysis. (c) Although the particle size can be accurately analyzed, the equipment is quite expensive (about 150 to 2 million NT dollars), which is a considerable burden for small processing systems. In addition to the coagulation or chemical precipitation, the particle size information directly reflects the treatment effect, but it is not necessary information. Therefore, it is not required for water treatment in the actual waste water treatment. (d) When the rubber feather is too large and too heavy, precipitation occurs in the monitoring tank, so there is no 201118362 method. The line turbidity meter can judge the tendency of the rubber feathers to grow in the wastewater treatment process. However, it has the following disadvantages: (a) The turbidity itself is interfered by the chromaticity, although the infrared light can be used as the light source to reduce the chromatic interference. The effect of chromaticity change on the scattered solid light scattering signal during the condensation process, but taking the chemical deposition hall of copper as an example, when the rubber feather is formed and precipitated, the chromaticity can also reflect the copper removal efficiency, so the color removal factor is simply removed. The effect of chemical precipitation cannot be truly judged. (b) Since the scale of the turbidimeter is relatively rough, for example, when the turbidity is 1〇〇NTU, the water sample is quite turbid, so the turbidity meter has a small range of signal jitter at the initial stage of the formation of the rubber feather, which is difficult to observe. Changes in optical signals. The fiber-optic Shengyu measuring instrument (PDA) uses a curved tube to simulate slow mixing, and sends the water sample into the integrated measuring device. 'Research _ shows that the result of the simulated slow mixing is similar to the slow mixing result of the bottle cup test, but it is still not true. The feather grows in the slow mixing pool. The wheel-out diagram of the fiber-optic Shenge detector (PDA) is obtained after statistical analysis. The selected parameter @ parameter @0^差, etc. will affect the statistical data and affect the interpretation of the coagulation result. SUMMARY OF THE INVENTION According to the above problem, the industry needs to observe the wastewater treatment of the yang in the early stage of rapid mixing and/or slow-moving. This method* can only analyze the wastewater treatment H in real time and can provide an image of the treated wastewater. Wei staff can directly view the small size and structure of the rubber from the image, and immediately know the size and distribution of the rubber feather. 201118362 The present invention provides a method for monitoring the effect of chemical precipitation, comprising: pumping a waste water into a container; irradiating the waste water in the container with a light source; capturing an image of the waste water by an image capture device; and analyzing the RGB in the image Value to detect the chemical precipitation of wastewater. The above described objects, features and advantages of the present invention will become more apparent and understood.
【實施方式】 依據Beer-LambertLaw，當水中懸浮微粒濃度越高，則 穿透光強度越弱，散射越多，即水越混濁。以一光源照射 有懸浮固體物之溶液時，若濃度相當低，水樣之影像僅有 少部份為懸浮固體物之影像’其餘則為背景色，例如零色。 當高濃度時，則影像中許多位置被懸浮固體物佔據，因此[Embodiment] According to Beer-LambertLaw, the higher the concentration of suspended particles in water, the weaker the transmitted light intensity, and the more scattering, that is, the more turbid water. When a solution of suspended solids is irradiated with a light source, if the concentration is relatively low, only a small portion of the image of the water sample is an image of suspended solids, and the rest is a background color, such as zero color. At high concentrations, many locations in the image are occupied by suspended solids, so
水樣背景色出現的比例會降低’在視覺上有水中顆粒的顏 色越來越明顯的情形。若以南領土為例，高濃度時，水才篆 便呈現白色，且濃度越高，散射出越多光線，水樣越白， 影像呈色便與濃度之高低有關，也就是在相當漢度的差 下’憑肉眼便可知濃度之高低。若將膠羽影像之色來數值 化’濃度高低之辨識便可更精確。另一方面，水體^若^ 粒之重量濃度相同時’平均粒徑較大的水樣，由於顆 較少’散射出來的光較少’並且在視覺上有較少的白零 視野中，.因此水樣的顏色較灰。 ' 一沉澱池 析（floe image color analysis)系統之裝置圖 201118362 其中配置有一攪拌裝置102，一蠕動幫浦106可將沉澱池 104中的廢水抽至一石英管110中。一穩定均勻之光源108 對裝有廢水之石英管110照射，並於另一端以一網路攝影 機112擷取廢水的影像，其中光源108與網路攝影機112 呈90°角，且光源108與網路攝影機112間隔絕外界光源 干擾，由網路攝影機112擷取影像後傳至例如電腦之資料 儲存及處理設備114，以軟體分析影像中由於懸浮溶液中 的懸浮顆粒散射所致之RGB值。 第2圖顯示本發明一實施例影像色彩分析監測化學沉澱 成效之方法的流程圖。進行步驟S201，連續擷取廢水之影 像，接著，進行步驟S202，對影像進行分析，並輸出影像 之RGB值S202。其後，進行步驟S203，對RGB值進行處 理分析，利用膠羽影像色彩分析監測系統分析影像得到不 同時間所攝得之影像及RGB值後，可進行以下判讀： 1. 重金屬去除效率-計算BR差值與GR差值後，並根據 BR差值與GR差值推估重金屬去除率。 2. 計算慢混區RGB值之下降斜率。 3. 計算慢混之RGB標準偏差值（SD)。 4. 根據RGB值利用關係式計算濁度值。 【實施範例-重金層去除效率】 以銅模擬廢水為例，利用光學影像監測系統對不同pH 與不同混凝劑量之膠羽光學影像之色彩分析。附件一顯示 不同時間之廢水影像，由照片可清楚觀察到膠羽大小影像 201118362 之變化。第3圖為調整PH與添加混凝劑後，在調整η 块此、及慢混時影像RGB之變動，其ρΗ係為9，初 濃度為145〜165PPm，多元氣化鋁ΡΑα加藥量為〇 第4圖為調整ρΗ與添加混凝劑後，在調整ρΗ、快混、 &處時影像RGB之變動，並dh # A 6 S，.， 及 隻軔/、PH係為6.5，初始銅濃度為 45〜165PPm ’多元氣化鋁（PAC1)加藥量為〇 41^几。如 圖和第4圖所示，調整pH後RGB值急遽上升，這是由於 在調整pH前，溶液中主要為溶解的銅離子，在調整後 形成氩氧化銅沉澱物，此時水中的懸浮顆粒數量大增，因 此散射光強度急遽上升。上升後再由於懸浮的沉澱物聚 集，RGB值下降。由第3圖和第4圖可以看到，高pH時， 且B值與R值之差異越大，經過其它不同pH的驗證得到， 將R訊號為基準’計算]gR之差異（即B -R)與Gr之差異（即 G_R)後發現’pH值越高，即銅去除率越高，b-r與g-r 值也越高’經迴歸分析後得到銅去除率與B-R及G-R值 有極高之相關性，相關式如式⑴。 銅去除率=8.185 (B - R) - 6.01 (G - R) + 27.582 (1) 【fe混區RGB值之下降斜率】 再比較第3圖和第4圖之後段（慢混）訊號值’可以看到 第4圖中RGB訊號下降之趨勢較明顯，這是由於第4圖中 的膠羽有效混凝’使得水樣整體開始呈現較清澈的影像。 比較其他不同加藥量也的確發現，膠羽長成越佳時，快慢 混區RGB訊號下降斜率亦較大（見表1)，因此可以利用rGB 下降斜率判定化學沉澱時快慢混之成效。 201118362 表1加藥輿钭率之關係 混凝劑量（mg/U 0.067 0.2 0.4 慢混終了濁度 (NTU) 43.3 78.0 42.8 斜率 -0.017 -0.012 -0.023 【慢混之RGB標準偏差值】 以附件一之影像為例，將化學沉澱各區，即pH調整初 期與調整後期、快混與慢混區之RGB訊號計算其標準偏差 (standard deviation，以下可簡稱SD)，可以得到各區段之 SD值，見第5圖，第5圖可以看到在調整pH時，SD急遽 上升，由附件一也可見到此時膠羽粒徑最大，隨後圖中的 膠羽漸小，S D值也隨之下降。本系統可直接觀看膠羽大小 變化外，並可運用標準偏差之計算瞭解系統中膠羽大小變 化之情形。 【計算濁度值】 在不同的水樣條件下，濁度值可以RGB相關式求出。 以pH分別為6‘5、7及9，混凝劑加藥量為〇.2 mg/L為例， pH 6.5，蜀度754B+2.925R+6.191 R2=〇.9685 pH=7 濁度二4.785R- 3.677B+71.093 R2=0.9959 pH=9 濁度49O6R-0.878B-82.958 R2=0.9805 R相關性鬲代表上述關係式準確度高。 201118362 因此以膠羽影像色彩分析對於欲監測之化學沉澱系統建 立濁度與操作條件之資料庫後，便可利用關係式，連續計 算處理系統之濁度值。 根據上述，本發明膠羽影像色彩分析監測化學沉澱成效 之方法可在化學沉澱之快混或慢混初期得知該化學沉澱之 成效，且可直接根據擷取之廢水的影像了解廢水中膠羽之 大小，監測該化學沉澱成效。 雖然本發明已揭露較佳實施例如上，然其並非用以限定 本發明，任何熟悉此項技藝者，在不脫離本發明之精神和 範圍内，當可做些許更動與潤飾，因此本發明之保護範圍 當視後附之申請專利範圍所界定為準。 201118362 【圖式簡單說明】 第1圖顯示本發明一實施例監測廢水之光學影像監測系 統之裝置圖。 第2圖顯示本發明一實施例影像色彩分析監測化學沉澱 成效之方法的流程圖。 第3圖顯示本發明一範例廢水處理之RGB變動曲線圖。 第4圖顯示本發明另一範例廢水處理之RGB變動曲線 圖。 第5圖顯示RGB標準偏差值曲線圖。 附件一顯示不同時間之廢水影像。 【主要元件符號說明】 102〜攪拌裝置； 104〜沉澱池； 106〜幫浦； 10 8〜光源； 110〜石英管； 112〜攝影機； 114〜資料儲存及處理設備。The proportion of the background color of the water sample is reduced by the fact that the color of the particles in the water is visually more and more obvious. If the southern territory is taken as an example, when the water is high, the water will appear white, and the higher the concentration, the more light is scattered, the whiter the water sample, and the color of the image is related to the concentration, that is, at a considerable level. Under the difference, the concentration can be known by the naked eye. If the color of the rubber feather image is numerically determined, the identification of the concentration can be more accurate. On the other hand, when the water body has the same weight concentration, the water sample with a larger average particle size has less 'scattered light less' and has less white zero field of view. Therefore, the color of the water sample is gray. 'Device diagram of a floe image color analysis system 201118362 A stirring device 102 is disposed therein, and a peristaltic pump 106 can pump the wastewater in the sedimentation tank 104 into a quartz tube 110. A uniformly uniform light source 108 illuminates the quartz tube 110 containing the wastewater, and at the other end draws an image of the wastewater by a network camera 112, wherein the light source 108 is at an angle of 90 to the network camera 112, and the light source 108 and the net The road camera 112 isolates the external light source interference, and the image is captured by the network camera 112 and transmitted to a data storage and processing device 114 such as a computer to analyze the RGB values in the image due to scattering of suspended particles in the suspension solution. Fig. 2 is a flow chart showing a method for monitoring the effect of chemical precipitation by image color analysis according to an embodiment of the present invention. In step S201, the image of the wastewater is continuously taken, and then, in step S202, the image is analyzed, and the RGB value S202 of the image is output. Then, step S203 is performed to analyze and analyze the RGB values, and the image and RGB values obtained by different times are analyzed by the gel image color analysis monitoring system, and the following interpretation can be performed: 1. Heavy metal removal efficiency - calculation BR After the difference is different from the GR, the heavy metal removal rate is estimated based on the BR difference and the GR difference. 2. Calculate the falling slope of the RGB value of the slow mixing zone. 3. Calculate the RGB standard deviation value (SD) for slow mixing. 4. Calculate the turbidity value using the relationship based on the RGB values. [Examples - Heavy gold layer removal efficiency] Taking copper simulated wastewater as an example, the optical image monitoring system was used to analyze the color of the rubber feather optical images of different pH and different coagulation doses. Annex I shows the wastewater images at different times, and the changes in the size of the rubber feathers 201118362 can be clearly observed from the photos. The third figure shows the change of the image RGB when adjusting the pH and adding the coagulant, and adjusting the η block and the slow mixing. The ρΗ system is 9, the initial concentration is 145~165PPm, and the multi-component gasification aluminum ΡΑα dosage is 〇 Figure 4 shows the change of image RGB when adjusting ρΗ and fast mixing, and dh # A 6 S,., and 轫/, PH system is 6.5, after adjusting ρΗ and adding coagulant. The copper concentration is 45~165PPm 'The amount of polyaluminized aluminum (PAC1) is 〇41^. As shown in Fig. 4 and Fig. 4, after adjusting the pH, the RGB value rises sharply. This is because the copper ions in the solution are mainly dissolved before the pH is adjusted. After the adjustment, the argon oxide copper precipitate is formed, and the suspended particles in the water at this time. The number is greatly increased, so the intensity of the scattered light rises sharply. After rising, the RGB values decrease due to the accumulation of suspended precipitates. It can be seen from Fig. 3 and Fig. 4 that at high pH, the greater the difference between B value and R value, the other difference in pH is obtained, and the R signal is used as the reference 'calculation' gR difference (ie B - R) and Gr difference (ie G_R) found that 'the higher the pH value, that is, the higher the copper removal rate, the higher the br and gr values'. After the regression analysis, the copper removal rate and the BR and GR values are extremely high. Correlation, the correlation is as shown in formula (1). Copper removal rate = 8.185 (B - R) - 6.01 (G - R) + 27.582 (1) [Descent slope of RGB value of fe mixed zone] Compare the 3rd and 4th figure (slow mix) signal value' It can be seen that the tendency of the RGB signal to drop in Figure 4 is more obvious, because the glue coagulation in Figure 4 is effective, so that the water sample as a whole begins to present a clear image. Compared with other different dosages, it is found that the better the gel length is, the lower the slope of the RGB signal in the fast mixing zone is (see Table 1). Therefore, the slope of the rGB can be used to determine the effect of the chemical precipitation. 201118362 Table 1 Relationship between dosing rate and coagulation dose (mg/U 0.067 0.2 0.4 slow mixing final turbidity (NTU) 43.3 78.0 42.8 slope -0.017 -0.012 -0.023 [slow mixing RGB standard deviation value] For example, the standard deviation (standard deviation, hereinafter referred to as SD) of the RGB signals in the chemical precipitation zone, that is, the initial stage of pH adjustment and the late adjustment, fast mixing and slow mixing zone, can be obtained, and the SD value of each section can be obtained. See Fig. 5, Fig. 5 shows that when pH is adjusted, SD rises sharply. It can be seen from Annex I that the gelatin size is the largest, and then the rubber feathers in the figure become smaller and the SD value decreases. The system can directly watch the change of the size of the rubber feather, and can use the calculation of the standard deviation to understand the change of the size of the rubber feather in the system. [Calculation of turbidity value] Under different water conditions, the turbidity value can be RGB correlation. The pH is 6'5, 7 and 9, respectively, and the coagulant dosage is 〇.2 mg/L, pH 6.5, twist 754B+2.925R+6.191 R2=〇.9685 pH=7 Turbidity II 4.785R- 3.677B+71.093 R2=0.9959 pH=9 Turbidity 49O6R-0.878B-82.958 R2=0.9805 R correlation Representing the above relationship is highly accurate. 201118362 Therefore, after the gel color image analysis is used to establish a database of turbidity and operating conditions for the chemical precipitation system to be monitored, the relationship can be used to continuously calculate the turbidity value of the processing system. In the above, the method for monitoring the chemical precipitation effect of the color analysis of the gel feather image of the present invention can know the effect of the chemical precipitation in the early stage of rapid mixing or slow mixing of chemical precipitation, and can directly understand the rubber feather in the wastewater according to the image of the extracted wastewater. </ RTI> </ RTI> </ RTI> </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; The scope of protection of the present invention is defined by the scope of the appended claims. 201118362 [Simplified Description of the Drawings] Figure 1 shows a device diagram of an optical image monitoring system for monitoring wastewater according to an embodiment of the present invention. Fig. 2 is a flow chart showing a method for monitoring the effect of chemical precipitation by image color analysis according to an embodiment of the present invention. 3 is a graph showing RGB variation of wastewater treatment in an example of the present invention. Fig. 4 is a graph showing RGB variation of wastewater treatment in another example of the present invention. Fig. 5 is a graph showing RGB standard deviation values. Image. [Main component symbol description] 102~ stirring device; 104~ sedimentation cell; 106~ pump; 10 8~ light source; 110~ quartz tube; 112~ camera; 114~ data storage and processing equipment.