201120419 六、發明說明： 【發明所屬之技術領域】 本發明係有關於一種污染物來源預測的方法及系統。 【先前技術】 傳統確定污染源可以利用逐步搜尋的方式直接尋找污 染物濃度最高的位置，或者利用建立污染物濃度分布的方 式推$污染濃度最高的位置。這兩種方式都需要在現場建 置大f的監測設備，或是重複地移動測量位置才能達到搜 寻〉可染源的效果。舉例來說，建立污染物濃度分布的方式 一般包含兩種方式，一種是使用低維度的點偵測進行單— 點濃度的採樣分析或即時監測。另一種是中維度的線偵 測，例如開放光徑遙測技術所獲得的濃度資訊。至於整個 平面的濃度偵測則是依據點偵測或線偵測的結果進行内插 補值方式或電腦斷層方式間接計算獲得。因此實務上，需 要在許多定點設置儀器偵測污染濃度，然後根據這些儀= 所測量的污染濃度計算出區域的污染濃度分布情況，再藉 由污染濃度分布情況確定污染源。 為了更有效率地找出污染源，有必要提供一種使用更 少污染監測儀器預測污染物來源的方法與系統。 【發明内容】 本發明提供一種污染物來源預測方法，包括:根據設於 一第一位置之一第一污染監測儀器所測量的污染濃度數據 201120419 及氣象站的—第一風向數據建立一第一既定數量的污染對 風=資料；根據設於一第二位置之一第二污染監測儀器所 測量$污染濃度數據及氣象站的一第二風向數據建立一第 二既定數量的污染對風向資料；根據至少該第一既定數量 的污染對風向資料以及至少該第二既定數量的污染對風向 貧料建構—污染機率區域分布資訊；以及根據該污染機率 區域分布資訊判斷污染物來源。 一本發明更提供一種污染物來源預測系統，包括：一第201120419 VI. Description of the Invention: TECHNICAL FIELD OF THE INVENTION The present invention relates to a method and system for predicting the source of a pollutant. [Prior Art] Traditionally, the source of pollution can be directly searched for the position with the highest concentration of pollutants by means of step-by-step search, or the position with the highest concentration of pollution can be pushed by establishing the concentration distribution of pollutants. Both of these methods require the construction of a large f-monitoring device on site or the repeated movement of the measurement location to achieve the search for a dyeable source. For example, the way to establish a contaminant concentration distribution generally involves two methods, one is to use low-dimensional point detection for single-point concentration sampling analysis or real-time monitoring. The other is mid-dimensional line detection, such as concentration information obtained by open-path telemetry. As for the density detection of the entire plane, it is obtained by interpolation interpolation or computer tomography based on the result of point detection or line detection. Therefore, in practice, it is necessary to set the instrument to detect the pollution concentration at many fixed points, and then calculate the pollution concentration distribution of the area according to the measured pollution concentration of these instruments, and then determine the pollution source by the distribution of the pollution concentration. In order to find sources of pollution more efficiently, it is necessary to provide a method and system for predicting the source of contaminants using fewer pollution monitoring instruments. SUMMARY OF THE INVENTION The present invention provides a method for predicting a source of pollutants, comprising: establishing a first according to pollution concentration data 201120419 measured by a first pollution monitoring instrument disposed at a first location and a first wind direction data of a weather station a predetermined quantity of pollution versus wind = data; a second predetermined amount of pollution versus wind direction data is established according to the pollution concentration data measured by the second pollution monitoring instrument and the second wind direction data of the weather station; And according to at least the first predetermined quantity of pollution to wind direction data and at least the second predetermined quantity of pollution to the wind to lean material construction - pollution probability regional distribution information; and determining the source of the pollutant according to the pollution probability regional distribution information. A invention further provides a pollutant source prediction system, including:
-污染監測儀器，於—第—位置收集—第—污染濃度數 !产：：二污染監測儀器，於一第二位置收集-第二污染 ；㈣’ «該第—污染濃度數據及來自 風向數據建立一第一既定數量的污染對風 根據該第二污染濃度數據及來自氣象站的一第二 第二既定數量的污染對風向資料、根據至 ^亥第-既疋數量的污染對風向資料以及至少 數里的W對風向#料建構—污染機率區域分布資訊，以 及根據該污染機率區域分布資訊判 ”本發明的污染物來源預測的方法與系統利用染 監測儀器結合氣象監測數據透過數值、 的分布情況。因此衫增加額外 析“木機羊 【實施方式】 特徵和優點能更明顯易懂， 為使本發明之上述目的 201120419 下文特舉較佳實施例，並配合所附圖式，作詳細說明如下. 第1圖係根據發明實施例說明污染物來源預測的方法 的流程圖。在步驟102與104 ’分別根據污染監測儀器的 污染濃度數據與氣象站的風向數據建立第一既定數量與第 二既定數量的特定方向的污染濃度機率。於本發明的實施 例中，在一個位置設置一台污染監測儀器測量污染濃度， 配合氣象站的風向數據，建立第一個污染濃度_風向=瑰 圖，如第2A圖所示。再另一位置同樣設置一台污染監測 儀器測量污染濃度，配合氣象站的風向數據，建立第二個 污染濃度-風向玫瑰圖。於本實施例中，正規風向的既定數 量是16,污染濃度-風向玫瑰圖表示所在位置相對16個正 規方向所產生的污染濃度的機率的示意圖。舉例來說，在 第2A®中的E方向，長條狀的長度表示機率大小，對於£ 方向來說是3鮮位個大小，而祕㈣顏色深度與寬度 :示污染濃度的高低，αΕ方向來說，此長條狀表示污； =度在範圍’因此^向在污染濃度8_16响的 ，率是3個單位’其他污㈣度機率職乎為零。同樣地， ⑽_方向’就呈現4種污染濃度機率，加總大約為4個 =，因此，NW方向的污染濃度機率為4個單位。於本 建:例4濃度-風向輕圖所呈現的數據是用於往後 冓/可染機率區域分布資訊時的重要計算依據。 的特5 :二〇6二別根據第-既定數量與第二既定數量 更多特定方向的;施:二雲規内插值法以求得 201120419 次雲規内插值法，如第2B®所示，根據原本 6個正規方向的污染敍機率求出360度方㈣污染漠度 機率。當然使用者可依據需喪呻曾 又 嫘而求计异某些特定方向的污染濃 度，率:因此可以分別取得至少第一既定數量或至少第二 既疋數量的特定方向的污染濃度機率。 在v驟108，根據第一位置的一個特定 度機率與第二位置的—個特定方向的污染濃度機率^出 :個既定位置的污染機率，然後建構一個污染機率區域分 布資訊。參考第2C圖，舉例來說，有兩個污染濃度—風向 玫瑰圖’分別具有至少第一既定數目的特定方 度機率與至少第二既定數目的特定方向的污染濃度機率。 例如左邊的 >可染濃度-風向玫瑰圖q〇_q15(q7_qi5因簡化 而未標示）代表這個方向的污染濃度機率（包含各個濃度範 圍的污染濃度機率），同理抓P15是另—組污㈣度_風向 玫瑰圖中對應方向的污染濃度機率。於另—個實施例中， 兩組污染濃度·風向玫魂圖可能有超過16個正規方向的污 染濃度機率。於本實施例中，假如需要A〇位置的污染機 率’則以Q3與m的污染濃度機率的乘積數值作為a〇位 置的汚染機率。同理’ A1位置的濃度機率則是Q4與pi3 的乘積數值。如此就可以建構—個污染機率區域分布資 況。於另-實施例中’可依需要建構更完整與密集的污染 機率，域分布的資訊。承上一段所描述’使用者可依據需 求計算出更多特定方向的污染濃度機率。參考第2C圖，當 存在Qx與Px方向的污染濃度機率時。則可以計算出位置 A999的污染機率（Qx乘以ρχ)，同樣地，位置B999的污染 201120419 機ί亦可求得。另外，於一實施例中，為有效率計算所有 既定位置的濃度機率，只要特定方向的濃度機率為二士， 則可以略過計算其所對應位置的污染機率。 〜才 最後，在步驟110,根據污染機率區域分 =立當=細率分布區域資訊時，可依： 21° 個4馬圖並且標示各種顏色以表示各種污染 機率’則可以清楚確定污染機率高的位置，如第扣圖所 /j\ ° 第3圖係顯示本發明污染物來源預測 預測的系統3。。包括第一污染監= 々第- 測儀器32〇、氣象站33〇以及主控電腦⑽。 ：污；監測儀器310設置於第一位置收集第-污染 ，度數據’第二污染監測儀器320則設置於第二位置 農ΐ數據。主控電腦34〇會根據第一污染濃度數 據及來自乳象站330的第一風向數據建立第一既定數量的 2=資料，亦即是第一位置上的16個正規方向的污 ^ 讀，34()亦會根據第二污染濃度數據及 斜— ί 330的第二風向數據建立第二既定數量的污染 對風向貧料，亦即是第二位置上的16個正規方向的污染濃 度機率。污染監測儀器可能是傅利葉轉換紅外光譜儀、光 游離侧n、氣相層析質譜分析儀或其他類m置之一。 龍ίΐ祖主控電腦,340會根據至少第一既定數量的污染 "f以及至少第二既定數量的污染對風向資料建構 -污染機率區域分布資訊，以及根據污染機率區域分布資 關斷〉5染物來源。於本實施例中，主控電腦340另包含 201120419 :個運异早兀342，用以計算污染機率區域分布資訊中既 疋^置之π染機率。舉例來說，運算單元342根據第一個 :上16個污染對風向資料中對應於既定位置的方向的 固機率數值乘以另—位置上16個污染對風向資料中對 應=既定位置的方向的—個機率數值得到污染機率區域 =布:貝訊中此既定位置之污染機率。綜合所有既定位置的 污:機率則可以求得污染機率區域分布資訊。實務上，會 ，污染機率區域分布資訊建立成以顏色表示機率高低之^ 南圖’使用者可㈣地根據等高圖確認污染來源。 士 外，主控電腦340更包括一第一計算單元與一第二 單70刀別根據第—位置上16個污染對風向資料以及- pollution monitoring equipment, in the - position - the first - pollution concentration number! Production:: two pollution monitoring equipment, in a second location - second pollution; (four) ' « the first - pollution concentration data and from wind direction data Establishing a first predetermined quantity of pollution versus wind according to the second pollution concentration data and a second and a second predetermined amount of pollution-to-wind direction data from the weather station, according to the pollution-to-wind direction data to the number of the first and second At least a few miles of wind-to-wind material construction--contamination probability distribution information, and according to the pollution probability regional distribution information, the method and system for predicting the source of pollutants of the present invention use the dye monitoring instrument to combine the meteorological monitoring data with the numerical value. Distribution. Therefore, the addition of the additional analysis of the "wooden sheep" [embodiment] features and advantages can be more clearly understood, in order to make the above-mentioned purpose of the present invention 201120419, the preferred embodiment is described below, and with the drawings, detailed The description is as follows. Fig. 1 is a flow chart showing a method of predicting the source of pollutants according to an embodiment of the invention. At steps 102 and 104', respectively, based on the pollution concentration data of the pollution monitoring instrument and the wind direction data of the weather station, the pollution concentration probability of the first predetermined quantity and the second predetermined quantity in a specific direction is established. In the embodiment of the present invention, a pollution monitoring instrument is set at one location to measure the pollution concentration, and the wind direction data of the weather station is used to establish the first pollution concentration _ wind direction = map, as shown in Fig. 2A. In another location, a pollution monitoring instrument is also set to measure the pollution concentration, and the wind direction data of the weather station is used to establish a second pollution concentration-wind direction rose diagram. In the present embodiment, the predetermined amount of the normal wind direction is 16, and the pollution concentration-wind direction rose diagram shows the probability of the concentration of the pollution generated by the position relative to the 16 normal directions. For example, in the E direction in 2A®, the length of the strip indicates the probability, and for the £ direction, it is 3 fresh, and the secret (4) color depth and width: the concentration of the pollution, αΕ direction In this case, the length of the strip indicates the stain; the degree is in the range 'thus ^ to the pollution concentration 8_16, the rate is 3 units' other pollution (four) degree probability is zero. Similarly, (10) _ direction 'has four kinds of pollution concentration probability, and the total is about 4 =, therefore, the probability of contamination concentration in the NW direction is 4 units. Yu Benjian: The data presented in the Example 4 Concentration-Wind Map is an important calculation basis for the distribution of information in the future 冓/dyeable area. Special 5: 二〇6二别 according to the first-established number and the second predetermined number more specific directions; Shi: two cloud gauge interpolation method to obtain the 201120419 sub-cloud gauge interpolation method, as shown in 2B® According to the original six normal directions of pollution, the probability of pollution is determined by 360 degrees (4). Of course, the user can calculate the concentration of the pollution in a certain direction according to the need to be sorrowful and ambiguous. Therefore, the probability of contamination concentration in a specific direction of at least the first predetermined quantity or at least the second 疋 quantity can be obtained separately. At v, 108, according to a certain degree of probability of the first position and the pollution concentration probability of the specific direction of the second position: a pollution probability of a predetermined position, and then construct a pollution probability area distribution information. Referring to Figure 2C, for example, there are two pollution concentrations - wind direction rose diagrams - each having at least a first predetermined number of specific tempences and at least a second predetermined number of specific directional contamination concentrations. For example, the left > stainable concentration - wind rose diagram q〇_q15 (q7_qi5 is simplified because it is not marked) represents the probability of contamination concentration in this direction (including the concentration concentration of each concentration range), the same reason to grasp P15 is another group Pollution (four) degree _ wind direction rose in the corresponding direction of the concentration of pollution concentration. In another embodiment, the two concentrations of pollution concentration and wind direction may have a probability of contamination concentration in more than 16 normal directions. In the present embodiment, if the pollution probability of the A〇 position is required, the product of the contamination concentration probability of Q3 and m is used as the pollution probability of the a〇 position. Similarly, the concentration probability of the 'A1 position is the product of Q4 and pi3. In this way, it is possible to construct a pollution probability regional distribution. In another embodiment, more complete and intensive pollution probability, domain distribution information can be constructed as needed. According to the description in the previous paragraph, the user can calculate the pollution concentration probability in more specific directions according to the demand. Refer to Figure 2C for the probability of contamination concentration in the Qx and Px directions. Then, the pollution probability of position A999 (Qx multiplied by ρχ) can be calculated. Similarly, the pollution of position B999 201120419 machine ί can also be obtained. Further, in an embodiment, in order to efficiently calculate the concentration probability of all the predetermined positions, as long as the concentration probability in a specific direction is two, the pollution probability of the corresponding position can be skipped. ~ Finally, in step 110, according to the pollution probability area = vertical = fine distribution area information, according to: 21 ° 4 horse map and labeling various colors to indicate various pollution probability ' can clearly determine the high probability of pollution The position, such as the map /j\ ° Figure 3 shows the system 3 for predicting the prediction of the source of the pollutants of the present invention. . Including the first pollution supervisor = 々 first - measuring instrument 32 〇, weather station 33 〇 and the main control computer (10). The monitoring instrument 310 is disposed at the first position to collect the first pollution level data. The second pollution monitoring instrument 320 is disposed in the second position of the farm data. The main control computer 34 will establish a first predetermined number of 2=data according to the first pollution concentration data and the first wind direction data from the breast station 330, that is, the 16 normal directions of the first position. 34() will also establish a second predetermined amount of pollution versus wind lean material based on the second pollution concentration data and the second wind direction data of oblique 330, that is, the probability of contamination concentration in the 16 normal directions at the second location. The pollution monitoring instrument may be one of a Fourier transform infrared spectrometer, a light free side n, a gas chromatography mass spectrometer or other type of m. Long ΐ 主 master computer, 340 will be based on at least the first predetermined quantity of pollution "f and at least a second amount of pollution on the wind direction data construction - pollution probability regional distribution information, and according to the pollution probability regional distribution of capital shutdown> 5 Source of dyes. In this embodiment, the main control computer 340 further includes 201120419: a different operation time 342, which is used to calculate the π dyeing rate of the pollution probability regional distribution information. For example, the operation unit 342 multiplies the value of the solid-state rate corresponding to the direction of the predetermined position in the wind direction data by the first one of the first 16 pollutions, and the direction of the corresponding position in the wind direction data corresponding to the corresponding position in the wind direction data. A probability value is obtained for the pollution probability area = cloth: the pollution probability of this predetermined position in Beixun. Combine all the pollution at a given location: the probability can be used to obtain information on the distribution of pollution probability. In practice, the information on the distribution probability of pollution probability is established to indicate the probability of color. The user of the map can confirm the source of pollution according to the contour map. In addition, the main control computer 340 further includes a first computing unit and a second single 70 knife according to the 16 locations of the pollution in the first position and
Ϊ^\16個污染對風Μ料實施三次雲規内插值法 、’、知S ’亏木監測儀&、第二污染監測儀器與氣象站W 固正規方向之外的/亏染對風向資料。於—實施例中，越多 ^亏染，風向資料可以建立更多既定位置的污染機率，產 高二大里的樣率區域分布資訊以助於建構更精密的等 ^最後’熟此技藝者可體認到他們可以輕易地使用揭露 的觀念以㈣定實施例為基礎而變更及設計可以實施同樣 目的之其他結構且不脫離本發明以及巾請專利範圍。 201120419 【圖式簡單說明】 第1圖係根據發明實施例說明污染物來源預測的方法 的流程圖， 第2A圖係發明實施例的濃度-風向玫瑰圖的示意圖； 第2B圖係說明發明實施例的三次雲規内插值法的示 意圖； 第2C圖係說明發明實施例之建構污染機率區域分布 貧訊的不意圖， 第2D圖說明根據污染機率區域分布資訊預測污染來 源的實施例的示意圖；以及 第3圖係顯示發明污染物來源預測的系統的架構圖。Ϊ^\16 pollutions carry out three cloud gauge interpolation methods for wind picking, ', know S' loss wood monitor & second pollution monitoring instrument and weather station W solid outside the normal direction / loss of dye on the wind direction data. In the embodiment, the more the dyes are, the wind direction data can establish more pollution chances at a given location, and the distribution information of the sample rate in the second generation can help to construct more precise ones. It is recognized that they can easily use the concept of disclosure to alter and design other structures that can perform the same purpose, based on the embodiments, without departing from the scope of the invention and the scope of the invention. 201120419 [Simplified description of the drawings] Fig. 1 is a flow chart illustrating a method for predicting the source of pollutants according to an embodiment of the invention, Fig. 2A is a schematic diagram of a concentration-wind direction rose diagram of the inventive example; and Fig. 2B is a diagram illustrating an embodiment of the invention Schematic diagram of the three-dimensional cloud interpolation method; FIG. 2C is a schematic diagram illustrating the construction of the pollution probability region distribution information in the embodiment of the invention, and FIG. 2D illustrates a schematic diagram of an embodiment for predicting the pollution source according to the pollution probability region distribution information; Figure 3 is a block diagram showing the system for predicting the source of the invented contaminants.
I 【主要元件符號說明】 102、104、106、108、110〜流程步驟； 3 0 0〜污染物來源預測的系統； 310〜第一污染監測儀器； 320〜第二污染監測儀器； 330〜氣象站； 340〜主控電腦； 342〜運算單元； 344〜第一計算單元； 346〜第二計算單元。I [Description of main component symbols] 102, 104, 106, 108, 110 ~ process steps; 3 0 0 ~ pollutant source prediction system; 310 ~ first pollution monitoring instrument; 320 ~ second pollution monitoring instrument; 330 ~ weather Station; 340~ master computer; 342~ arithmetic unit; 344~ first computing unit; 346~ second computing unit.