TW200928643A - Multi-stages event monitoring method for detecting abnormal manufacturing process - Google Patents
Multi-stages event monitoring method for detecting abnormal manufacturing process Download PDFInfo
- Publication number
- TW200928643A TW200928643A TW96150033A TW96150033A TW200928643A TW 200928643 A TW200928643 A TW 200928643A TW 96150033 A TW96150033 A TW 96150033A TW 96150033 A TW96150033 A TW 96150033A TW 200928643 A TW200928643 A TW 200928643A
- Authority
- TW
- Taiwan
- Prior art keywords
- stage
- event
- abnormal process
- abnormal
- monitoring
- Prior art date
Links
Landscapes
- Testing And Monitoring For Control Systems (AREA)
Abstract
Description
L3PA 200928643 九、發明說明： 【發明所屬之技術領域】 於一種異常製程多階段事件監控方、去 ，且特別是有 關 本發明是有關於-種異常製料控方法， 【先前技術】 因警報管理不當所造成的災害事件 導致的損失更是可觀。於西元1979年3 θ 35不窮，而且所 (Three Mile Island )所發生的核能 ^ 28 *在三沒島 損失高達約10億美金，並使全薇永久停機的經濟 釋放到外界的輻射物質對環境所產生的_ ^厫重的是 :的=□，在的因素卻是由於:作介以 良’而主要原因在於操作貝需面對太多不同的❹十不 忽略關鍵警報，甚或被警報誤導。 胃報以致於 參 一般工廠在正常狀況時，現場操作員應遵循正常择作 策略，^過控㈣統維私歧常運作； = =U也=制r標準程序，甚至於多次= 狀::，m緊急應變之間的 ,. ® ^,、 來直疋工廠的操作的灰色地L3PA 200928643 IX. Description of the invention: [Technical field of the invention] In an abnormal process multi-stage event monitoring, go, and especially related to the present invention is related to an abnormal material control method, [prior art] due to alarm management The damage caused by improper disasters is even more impressive. In 1979, 3 θ 35 is endless, and the nuclear energy that occurred in Three Mile Island was lost to about 1 billion US dollars in Sandao Island, and the economy of Quanwei’s permanent shutdown was released to the outside world. The _ ^ 厫 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境 环境Misleading. Stomach report so that when the general factory is in normal condition, the on-site operator should follow the normal selection strategy, ^ control (4) the general operation of the private sector; = = U also = standard procedure of the system, even multiple times = shape: :,m emergency strain between,. ^ ^,, the gray ground of the operation of the factory
At 64異常損失’首先就必須確實掌握製程的「異 常狀，」’這就有賴製程安全監控技術的協助。另外，為 ，現場工作安全、平穩的進行，常須準備各種必要的標準 操作程序作為現場操作、人負自丨丨練乃至於國際標準組織 (International Standards Organization，ISO)稽核之用。 6 200928643At 64 abnormal loss, you must first grasp the "abnormality of the process," which depends on the assistance of process safety monitoring technology. In addition, in order to carry out on-site work safely and smoothly, it is often necessary to prepare various necessary standard operating procedures for on-site operation, self-sufficiency, and even the International Standards Organization (ISO) audit. 6 200928643
—….— 13PA 許多調查發現，事故發生的原因通常不只一個，起使 事件一般是由於設計或操作程序上的問題。當同時發生了 不可預期的另一事件後，就會產生無法控制的情況。現行 的操作安全檢查通常有瑕疵，無法考慮到事件組合衝擊， 故具備良好的事件組合支援工具與事件監控將是不可避 免的趨勢。另一方面’亦能早期察覺不正常的突發事件、 縮知·應變決策時間’並將製程導引至正確方向，同時能管 制各種不良之副作用以減少錯誤發生。 蠼 【發明内容】 本發明係有關於一種異常製程多階段事件監控方 法’係利用製程中的即時資訊判斷製程中是否有異常問題 發生’且於異常問題發生時即時提供對應的處理程序予操 作人員以輔助操作人員處理問題，以避免重大危害發生。 本發明k出一種異常製程多階段事件監控方法，此方 鬌 法包括步驟如下。先偵測一製程站之一異常製程事件群組 第一階段異常製程事件，以擷取並分析一第一階段異 *製程事件監控點群訊號。接著，於第一階段異常製程事 件監控點群訊號滿足一第一階段異常製程事件觸發條件 時’提供第一階段異常製程事件之至少一第一處理動作。 然後’於第-階段異常製程事件之至少一第一處理動作完 成後，轉移偵測一第二階段異常製程事件以擷取並分析一 第段異常製程事件監控料崎。接著，於第二階段 異常製程事件監控點群喊滿足—第二階段異常製程事 7—....— 13PA Many surveys have found that there are usually more than one cause of an accident, and the event is generally caused by a design or operational procedure. When another unexpected event occurs at the same time, there is an uncontrollable situation. Current operational safety checks are often flawed and cannot take into account the combined impact of events, so it is inevitable that good event portfolio support tools and event monitoring will be available. On the other hand, it can also detect abnormal emergencies, shrinking and responding to decision-making times, and guide the process to the correct direction. At the same time, it can control various adverse side effects to reduce errors.蠼[Summary of the Invention] The present invention relates to an abnormal process multi-stage event monitoring method 'Using the instant information in the process to judge whether an abnormal problem occurs in the process' and providing a corresponding processing program to the operator immediately when the abnormal problem occurs Help the operator handle the problem to avoid major hazards. The invention provides an abnormal process multi-stage event monitoring method, and the method includes the following steps. First, an abnormal process event group of one process station is detected, and the first stage abnormal process event is captured to analyze and analyze a first stage hetero-process event monitoring point group signal. Then, at least one first processing action of the first stage abnormal process event is provided when the first stage abnormal process event monitoring point group signal satisfies a first stage abnormal process event trigger condition. Then, after at least one first processing action of the first-stage abnormal process event is completed, a second-stage abnormal process event is detected to capture and analyze a first-stage abnormal process event monitoring. Then, in the second stage, the abnormal process event monitoring point group shouts satisfaction - the second stage abnormal process matter 7
3PA 200928643 件觸發條件時，提供第二階段異常製程事件之至少一第二 處理動作。然後，於第二階段異常製程事件之至少一第二 處理動作完成後，轉移偵測下一階段異常製程事件，並依 此方式循序執行直至此異常製程事件群組之一第N階段 異常製程事件之至少一第N個處理動作完成。而當第N 階段異常製程事件之至少一第N個處理動作完成，又回到 此異常製程事件群組之一第一階段異常製程事件，持續保 持異常製程監控。 ❹ 為讓本發明之上述内容能更明顯易懂，下文特舉較 佳實施例，並配合所附圖式，作詳細說明如下： 【實施方式】 請參照第1圖，其繪示依照本發明較佳實施例的異常 製程多階段事件監控方法之流程圖。如第1圖所示，異常 製程多階段事件監控方法包括步驟S110〜S150。從步驟 S110開始，先偵測一製程站之一異常製程事件群組之一第 ® 一階段異常製程事件，以擷取並分析一第一階段異常製程 事件監控點群訊號。接著，進入步驟S120，於第一階段異 常製程事件監控點群訊號滿足一第一階段異常製程事件 觸發條件時，提供第一階段異常製程事件之至少一第一處 理動作。然後，進入步驟S130，於第一階段異常製程事件 之至少一第一處理動作完成後，轉移偵測一第二階段異常 製程事件以擷取並分析一第二階段異常製程事件監控點 群訊號。接著，進入步驟S140，於第二階段異常製程事件 8When 3PA 200928643 trigger condition, at least one second processing action of the second stage abnormal process event is provided. Then, after at least one second processing action of the second stage abnormal process event is completed, the next stage abnormal process event is detected, and the Nth stage abnormal process event of one of the abnormal process event groups is sequentially executed in this manner. At least one Nth processing action is completed. When at least one Nth processing action of the Nth phase abnormal process event is completed, the process returns to the first stage abnormal process event of the abnormal process event group, and the abnormal process monitoring is continuously maintained. BRIEF DESCRIPTION OF THE DRAWINGS In order to make the above description of the present invention more comprehensible, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings. FIG. A flow chart of the abnormal process multi-stage event monitoring method of the preferred embodiment. As shown in Fig. 1, the abnormal process multi-stage event monitoring method includes steps S110 to S150. Starting from step S110, one of the abnormal process event groups of one of the process stations is detected to detect and analyze a first-stage abnormal process event monitoring point group signal. Then, proceeding to step S120, when the first stage abnormal process event monitoring point group signal satisfies a first stage abnormal process event trigger condition, at least one first processing action of the first stage abnormal process event is provided. Then, proceeding to step S130, after at least one first processing action of the first phase abnormal process event is completed, a second phase abnormal process event is detected to capture and analyze a second phase abnormal process event monitoring point group signal. Then, proceeding to step S140, in the second stage abnormal process event 8
13PA 200928643 監控點群訊號滿足一第二階段異常製程事件觸發條件 時，提供第二階段異常製程事件之至少一第二處理動作。 然後，進入步驟S150,於第二階段異常製程事件之至少一 第二處理動作完成後，轉移偵測下一階段異常製程事件， 並依此方式循序執行直至此異常製程事件群組之一第N 階段異常製程事件之至少一第N個處理動作完成。而當第 N階段異常製程事件之至少一第N個處理動作完成，又回 到此異常製程事件群組之步驟S110之一第一階段異常製 ❿ 程事件，持續保持異常製程監控。 於任一製程站中，係可設定多個異常製程事件群組以 因應不同的異常製程問題。且這些異常製程事件群組係各 自具有多階段異常製程事件，藉此以依序判斷各階段異常 製程事件的排除與否，而最終可將異常製程問題排除。 請參照第2圖，其繪示一異常製程事件群組之示意 圖。如第2圖所示，一個異常製程事件群組中，係具有多 階段異常製程事件，其例如包括第一階段異常製程事件、 ❿ 第二階段異常製程事件、以及其他第N(N等於或大於二） 階段異常製程事件。於此，係以不同的觸發條件去判斷一 階段異常製程事件是否發生，且於該階段的觸發條件被滿 足時，進一步提供對應該階段異常製程事件之至少一個處 理動作。此外，當一階段異常製程事件之處理動作完成 後，將自動接續至下一階段異常製程事件之判斷。 舉例來說，如第2圖以及步驟S110所示，例如是先 偵測多個第一階段監控點以擷取到第一階段異常製程事 9 20092864313PA 200928643 When the monitoring point group signal satisfies a second stage abnormal process event trigger condition, at least one second processing action of the second stage abnormal process event is provided. Then, proceeding to step S150, after at least one second processing action of the second stage abnormal process event is completed, the next stage abnormal process event is detected and executed in this manner until one of the abnormal process event groups is N At least one Nth processing action of the phase abnormal process event is completed. When at least one Nth processing action of the Nth phase abnormal process event is completed, the first stage abnormal process event of step S110 of the abnormal process event group is returned, and the abnormal process monitoring is continuously maintained. In any process station, multiple abnormal process event groups can be set to respond to different abnormal process problems. Moreover, these abnormal process event groups each have a multi-stage abnormal process event, thereby sequentially judging the exclusion of abnormal process events in each stage, and finally the abnormal process problem can be eliminated. Please refer to FIG. 2, which shows a schematic diagram of an abnormal process event group. As shown in FIG. 2, an abnormal process event group has a multi-stage abnormal process event, which includes, for example, a first-stage abnormal process event, a second-stage abnormal process event, and other Nth (N is equal to or greater than b) Stage abnormal process events. Here, different trigger conditions are used to determine whether a phase abnormal process event occurs, and when the trigger condition of the phase is satisfied, at least one processing action corresponding to the phase abnormal process event is further provided. In addition, when the processing of the one-stage abnormal process event is completed, it will automatically continue to the judgment of the next stage abnormal process event. For example, as shown in FIG. 2 and step S110, for example, multiple first-stage monitoring points are detected to capture the first-stage abnormal process. 9 200928643
i «JPA 件監控點群訊號。如步驟sl2〇所示，當此監控點群訊號 於分析後係滿足第一階段異常製程事件觸發條件時，則判 斷該第階段異常製程事件已發生，並即時提供對靡該第 -階段異常製程事件之至少-第—處理動作，例戈= 動作H11〜H115。此外，亦可提供一警報訊號以告知人員 製程異常的發生。而於處理動作H11〜H15完成後，如步 驟S130所示，將自動或手動轉移接續至第二階段異常製 程事件之判斷。 〃 ❹ 此時，係接著偵測多個第二階段監控點以擷取第二階 段異常製程事件監控點群訊號。如步驟sl4〇所示，當此 監控點群訊號於分析後係滿足第二階段異常製程事件觸 發條件時，則判斷第二階段異常製程事件已發生，並即時 提供對應第二階段異常製程事件之至少一第二處理動 作，例如是處理動作H21〜H25。另外，亦可提供—警報 訊號以告知人員製程異常的發生。而於這些處理動作完成 ⑩ 後，如步驟sl5〇所示，轉移至下一階段異常製程事件， 例如疋第N階段異常製程事件之彳貞測，直至該階段之處理 動作Hnl〜Hn5完成。若是該第n階段之處理動作完成 後，即可判疋一異常問題已完全排除。而當第N階段異常 製程事件之至少一第>^個處理動作完成，又回到此異常製 程事件群組之步驟sll〇之一第—階段異常製程事件，持 續保持異常製程監控。 較佳地，係於開始偵測一製程站之一異常製程事件群 組之前，先設定好上述異常製程事件群組之基本資料。另i «JPA piece monitoring point group signal. As shown in step sl2, when the monitoring point group signal satisfies the first stage abnormal process event trigger condition after the analysis, it is determined that the first stage abnormal process event has occurred, and the first stage abnormal process is provided immediately. At least the - event processing action, example Ge = action H11 ~ H115. In addition, an alarm signal can be provided to inform personnel of the occurrence of process anomalies. After the processing operations H11 to H15 are completed, as shown in step S130, the automatic or manual transfer is continued to the judgment of the second-stage abnormal process event. 〃 ❹ At this point, the system then detects multiple second-stage monitoring points to capture the second-stage abnormal process event monitoring point group signal. As shown in step sl4, when the monitoring point group signal satisfies the trigger condition of the second stage abnormal process event after the analysis, it is determined that the second stage abnormal process event has occurred, and the second stage abnormal process event is immediately provided. At least one second processing operation is, for example, processing operations H21 to H25. In addition, an alarm signal can be provided to inform personnel of the occurrence of process anomalies. After these processing actions are completed 10, as shown in step sl5, the process proceeds to the next stage abnormal process event, for example, the detection of the Nth stage abnormal process event until the processing actions Hnl~Hn5 of the stage are completed. If the processing of the nth stage is completed, it can be judged that the abnormal problem has been completely eliminated. When at least one of the processing operations of the Nth phase abnormal process event is completed, and returning to the abnormal process event of the step sll of the abnormal process event group, the abnormal process monitoring is continuously maintained. Preferably, before starting to detect an abnormal process event group of a process station, the basic data of the abnormal process event group is set. another
13PA 200928643 號、第二阡俨好第一階段異常製程事件監控點群訊 ;製程事;半吊製程事件監控點群訊號以及第N階段異 本太料。N階段異常製程事件監控點群訊號之基 當：程事件S'"階段異常製程事件觸發條件、第二階段異 昱常製程事彳φΓ條件、第N階段異常製程事件之第N階段 條件、第一階段異常製程事件之至少第 作以及第心ΐ—階段異常製程事件之至少—第二處理動 必須先設定好段㈣製程事件之至少—fN個處理動作亦 之示音H、3A圖’其緣示—製造薇區及其多個製程站 製程站控模組1G去管控多個 減不嶋 ^ ，且於各個衣裎站中裝設有多個用以 擷取δ 件，稱之為標籤（ Φ 是設置於製程設備上，或是設置於製程糊i:工= 境中之監控點’以_不⑽號源，並將所擷取到的訊號 傳送到監控模組1G中，藉此使監_組1G得以即時且全 面性地監控製程站10〇〜500是否產生異常狀況。 根據標藏τ所處製程站的不同、標籤τ於製程站中 的工作環境不同，亦或是標籤T於一製程設備上之設置位 置不同，標藏τ所擷取到的訊號種類可以是生產製造、品 質檢驗、火警訊號、極早期偵煙訊號、漏液訊號、地震訊 號、特殊氣體訊號等。另外，標籤T所擷取到的訊號格式 係可包含整數型（Integer)、浮點數型（Float)、布林型 11 200928643 .--Ι*·|.Γ *l*f*TT-W U — - . · (Boolean)、及文字型（String)之訊號格式。 於設定一異常製程事件群組之不同階段異& 4 θ 件觸發條件時，係可根據訊號源或是訊號種類，將各1段 觸發條件透過内建數學函式（Formula )及函气、=子 (Formula operator )，請參照第3B圖所綠示々運算 數學 與運异子之不意圖’而可制§丁為複合訊號之管制條 ^ 例來說’標籤T1、T2所擷取的訊號種類例如為生產/步， 且標籤Τ1、Τ2係擷取一製程設備的運轉速度量值，產， 可將異常製程之觸發條件設定為Τ1「高於—運 疋以 值」’ Τ2「低於一運轉速度最大值」兩者再透過内學 函式AND制訂為複合訊號之管制條件。 請參照第4圖，其繪示第3A圖的監控 裎 站之方塊圖。監控模組1Q包括—判斷單 程 事件眘％L/i , A辦早兀U、一多階段 =牛資枓庫U及一警報通訊單元15。製程站2〇〇之標藏 事件資^座等揭取之訊號係傳送至判斷單元U。於多階段 ❿本資料 中係可建立前述之異常製程事件群組之基 異常製Ζ外’亦可建立有各異常製程事件群組之多階段 段之資料’用以供判斷單元u邦定各階 可建立^事件發生之依據。多階段事件資料庫13中更 觸發條^應判斷單元11判定不同階段異常製程事件的 訊單元彳滿足時，所需提供的處理動作之資料。警報通 理動=尚^ ^是於任一階段異常製程事件發生或是任一處 工I未完成時，用以產生警報訊號去即時提醒製程的 作或管理人〇 貝。另外，警報通訊單元15亦可藉由發送 1213PA 200928643, the second good first stage abnormal process event monitoring point group news; process matters; semi-lift process event monitoring point group signal and the Nth phase of the difference. The basis of the N-stage abnormal process event monitoring point group signal: the event event S'" phase abnormal process event trigger condition, the second stage different process rule 彳 Γ condition, the Nth stage abnormal process event N phase condition, At least the first stage of the abnormal process event and at least the first stage of the heart-stage abnormal process event - the second process must first set the segment (four) process events at least - fN processing actions also show the sound H, 3A map ' The reason is that the manufacturing area of the Wei area and its multiple processing stations is controlled by a 1G control module, and a plurality of δ pieces are installed in each clothing station. The label ( Φ is set on the process device, or is set in the process paste i: the monitoring point in the work area = _ not (10) source, and the captured signal is transmitted to the monitoring module 1G, borrowed This enables the supervisory group 1G to monitor the process station 10〇~500 for an abnormal situation in an instant and comprehensive manner. According to the different processing stations of the standard τ, the label τ is different in the working environment in the processing station, or is the label T is not set on a process device The type of signal captured by the standard τ can be manufacturing, quality inspection, fire alarm signal, very early detection smoke signal, liquid leakage signal, seismic signal, special gas signal, etc. In addition, the signal captured by the label T The format system can include integer type (Integer), floating point number type (Float), Bolling type 11 200928643 .--Ι*·|.Γ *l*f*TT-W U — - . · (Boolean), and The signal format of the character string (String). When setting the trigger condition of the different & 4 θ parts in different stages of an abnormal process event group, each segment trigger condition can be transmitted through built-in mathematics according to the signal source or signal type. For the formula (Formula) and the qi, the formula operator, please refer to the green 々 々 mathematics and the operation of the singularity of the 3D diagram, and the § D is the control strip of the composite signal. The types of signals captured by the tags T1 and T2 are, for example, production/step, and the tags Τ1 and Τ2 are used to capture the operating speed of a process device. The trigger condition for the abnormal process can be set to Τ1 “above— The value of "" Τ 2 "below a maximum speed of operation" The internal learning function AND is formulated as the control condition of the composite signal. Please refer to Figure 4, which shows the block diagram of the monitoring station of Figure 3A. The monitoring module 1Q includes - judging the one-way event caution %L/i, A Long-term U, a multi-stage = cattle treasury U and an alarm communication unit 15. Signals of the standardization of the process station 2 are transmitted to the judgment unit U. In the data, the basic abnormality of the above-mentioned abnormal process event group can be established. 'The data of the multi-stage segment of each abnormal process event group can also be established' for the judgment unit to be able to establish various events. The basis. The more trigger bar in the multi-stage event database 13 determines the information of the processing action to be provided when the judging unit 11 judges that the signal unit of the abnormal process event in different stages is satisfied. Alarm Transaction = Yes ^ ^ is used to generate an alarm signal to promptly remind the process or the manager of the process when an abnormal process event occurs at any stage or when the work I is not completed. In addition, the alarm communication unit 15 can also transmit 12
3 PA 200928643 短訊、電子郵件或通話等方式告知，以進行後續的必要處 理機制。 監控模組10較佳地更包括一操作紀錄單元17，係用 於將所有操作過程記錄起來，以建立一警報監控日誌檔 案，供作後續作事件追蹤處理。使得監控模組10能夠動 態並即時且持續性的監控各標籤與異常製程問題的資 料，且可依全廠、製程站別、優先等級等，去紀錄、統計 與發佈這些警報資料予製程工作或管理人員。之後將附圖 ® 說明記錄操作流程之方法。 第1圖所示之異常製程多階段事件監控設定方法，必 須在執行製程步驟前，先將製造廠區及其各製程站之基本 資料、標籤的基本資料建立起來。舉例來說，先依照製造 廠區實際製程區域劃分出不同的製程站，如第3圖之製程 站100〜500，甚或可將各製程站排序出來。 接著，於多階段事件資料庫13中設定各製程站及其 異常製程事件群組的資料，並設定異常製程事件群組用以 ® 循序判斷之第一階段異常製程事件觸發條件、第二階段異 常製程事件觸發條件甚或第N階段異常製程事件觸發條 件。如第3B圖所示，多階段事件資料庫13中係建立有多 個數學函式（Formula)及函式運算子（Formula operator) ’ 可用以組合成一任一階段異常製程事件之觸發條件的定 義。另外，多階段事件資料庫13中較佳地亦具有函式檢 查功能，以確認各階段異常製程事件觸發條件之定義正確 無誤。 13 2009286433 PA 200928643 We will inform you by SMS, email or call to carry out the necessary processing mechanism. The monitoring module 10 preferably further includes an operational logging unit 17 for recording all operational procedures to establish an alert monitoring log file for subsequent event tracking processing. The monitoring module 10 can dynamically and continuously and continuously monitor the data of each label and abnormal process problem, and can record, count and publish the alarm data to the process work according to the whole plant, the process site, the priority level, and the like. manager. Figure 1 below illustrates the method of recording the operational flow. The abnormal process multi-stage event monitoring setting method shown in Fig. 1 must first establish the basic data of the manufacturing plant area and its various process stations and the basic data of the label before executing the process steps. For example, different process stations can be divided according to the actual process area of the manufacturing plant, such as the process stations 100 to 500 in Fig. 3, or even the process stations can be sorted out. Then, in the multi-stage event database 13, the data of each process station and its abnormal process event group are set, and the abnormal process event group is set to use the first stage abnormal process event trigger condition and the second stage abnormality. The process event trigger condition or even the Nth stage abnormal process event trigger condition. As shown in FIG. 3B, the multi-stage event database 13 is constructed with a plurality of mathematical formulas and a formula operator 'can be used to combine the definition of a trigger condition for an abnormal process event of any stage. . In addition, the multi-stage event database 13 preferably also has a function check function to confirm that the definition of the trigger condition of the abnormal process event in each stage is correct. 13 200928643
3PA ·Κ· ·· ΑΧ 一異常製程事件群組之一階段異常製程事件係依照 其觸發條件的成立與否、以及處理動作的完成與否，以決 定是否轉移至下一階段異常製程事件的判斷，如此以完成 整個警報監控。於此係以異常製輕事件群組具有三階$異 常製程事件為例作說明。 ^ 對應三階段之異常製程事件群組’並依照製程實際運 作情形，於製程站中設定多個第一階段監控點、多個^二 階段監控點與多個第三階段監控點。第一階段監控點、& Φ 二階段監控點及第三階段監控點所產生的訊號^各別用 於判斷是否滿足該異常製程事件群組之三階段觸發條 件。以第4圖中之製程站200為例，標籤τΐ至Τ12係各 別偵測監控點201至212之製程狀態，並對應產生監控訊 號。在此僅舉例說明’於此例如是將監控點2〇1至212中 之監控點201、204、207、210設定為第一階段監控點， 監控點202、205、208、211設定為第二階段監控點，其 ⑩ 餘之監控點2〇3、206、209、212則設定為第三階段監控 點。第一階段監控點2〇1、204、207、210之標籤ΤΙ、Τ4、 τ_7、τιο所擷取到的訊號係歸類為第一階段異常製程事件 I控點群訊號，其係用以判斷是否滿足第一階段異常製程 觸發條件。第二階段監控點202、2。5、2。8、211之 I，Τ2、T5、T8、Til所拇取到的訊號係歸類為第二階段 ”系裝程事件監控點_訊號，其係用以判斷是否滿足第二 階段異常製程事件觸發條件。而第三階段監控點2〇3、 206 209、212之標鞔T3、T6、T9、T12所擷取到的訊號 143PA ·Κ··· 之一 One of the abnormal process event groups is based on the establishment of the trigger condition and the completion of the processing action to determine whether to transfer to the next stage of the abnormal process event. So to complete the entire alarm monitoring. This is an example of a case where the abnormal light-making event group has a third-order $ abnormal process event. ^ Corresponding to the three-stage abnormal process event group' and according to the actual operation of the process, multiple first-stage monitoring points, multiple second-stage monitoring points and multiple third-stage monitoring points are set in the processing station. The signals generated by the first stage monitoring point, & Φ two-stage monitoring point and the third stage monitoring point are used to determine whether the three-stage trigger condition of the abnormal process event group is satisfied. Taking the process station 200 in Fig. 4 as an example, the labels τ ΐ Τ 系 12 respectively detect the process states of the monitoring points 201 to 212, and correspondingly generate monitoring signals. Here, for example, only the monitoring points 201, 204, 207, 210 in the monitoring points 2〇1 to 212 are set as the first-stage monitoring points, and the monitoring points 202, 205, 208, 211 are set as the second. At the stage monitoring point, more than 10 monitoring points 2〇3, 206, 209, and 212 are set as the third stage monitoring points. The signals extracted by the labels ΤΙ, Τ4, τ_7, and τιο of the first stage monitoring points 2〇1, 204, 207, 210 are classified into the first stage abnormal process event I control point group signal, which is used to judge Whether the first stage abnormal process trigger condition is satisfied. The signals of the second stage monitoring points 202, 2. 5, 2. 8, 211, I, Τ2, T5, T8, Til are classified into the second stage "system processing event monitoring point _ signal, which It is used to judge whether the second stage abnormal process event trigger condition is satisfied, and the signal 14 of the third stage monitoring point 2〇3, 206 209, 212 is marked by T3, T6, T9, T12.
3PA 以3PA to
200928643 則歸類為第三階段異常製程事件監控，⑽，豆係用 判斷是否滿足第三階段異常製程事件鹎發條件 u ^ 點，二Γ:二，、第二階段監‘第毫監控 點選疋後’係可於多階段事件資料庫13中將這些監控點 之標戴的基本資料建立起來，例如是榡職名稱、標籤擷取 之訊號種類等。同時，也必鮮定料軸所擷取訊號之 訊號格式。如前所述，訊號格式可包括整數型、浮點數型、 布林型及文字型之訊號格式。另外，依據製程實際運作情 形，不同階段所設定之監控點是可以相同或是不相同。 接著，可在多階段事件資料庫13中設定因應第一階 段異常製程事件觸發條件被滿足時的至少一第一處理動 作、第二階段異常製程事件觸發條件之至少一第二處理動 作以及第三階段異常製程事件觸發條件之至少一第三處 理動作。這些處理動作係根據一異常製程事件群組之不同 階段所設計的操作指引，當人員依照排除異常之操作指示 將一階段之處理動作完成後，則會轉移至下一階段之觸發 條件判斷。 如第1圖所示，異常製程多階段事件監控方法的步驟 S110及S120中，判斷單元11係先藉由偵測第一階段監控 點201、204、207、210以擷取第一階段異常製程事件監 控點群訊號。當判斷單元11判斷第一階段監控點201、 204、207、210之第一階段異常製程事件監控點群訊號滿 足該異常製程事件群組之第一階段異常製程事件觸發條 件時，判斷單元11係提供至少一第一處理動作，以作為 15200928643 is classified as the third stage abnormal process event monitoring, (10), the bean system judges whether the third stage abnormal process event issuance condition u ^ point, the second: second, the second stage supervision 'the first monitoring point selection Afterwards, the basic data of these monitoring points can be established in the multi-stage event database 13, such as the name of the job, the type of the signal captured by the tag, and the like. At the same time, the signal format of the signal taken by the axis is also determined. As mentioned above, the signal format can include integer, floating point, Boolean, and text signal formats. In addition, depending on the actual operation of the process, the monitoring points set in different stages may be the same or different. Then, at least one first processing action, at least one second processing action, and a third phase of the second phase abnormal process event triggering condition, when the first phase abnormal process event triggering condition is satisfied, may be set in the multi-stage event database 13 At least one third processing action of the phase abnormal process event trigger condition. These processing actions are based on the operational guidelines designed for different stages of an abnormal process event group. When the person completes the processing of the first stage according to the operation instruction for excluding the abnormality, it will shift to the trigger condition judgment of the next stage. As shown in FIG. 1 , in steps S110 and S120 of the abnormal process multi-stage event monitoring method, the determining unit 11 first detects the first-stage abnormality process by detecting the first-stage monitoring points 201, 204, 207, and 210. Event monitoring point group signal. When the determining unit 11 determines that the first stage abnormal process event monitoring point group signal of the first stage monitoring point 201, 204, 207, 210 satisfies the first stage abnormal process event trigger condition of the abnormal process event group, the determining unit 11 is Providing at least one first processing action as 15
3PA 200928643 排除該第一階段異常製程事件之操作指引。 當第一處理動作完成後，如步驟S130及S140所示， 判斷單元11會接著藉由偵測第二階段監控點202、205、 208、211以擷取第二階段異常製程事件監控點群訊號。當 判斷單元11判斷第二階段監控點202、205、208、211之 第二階段異常製程事件監控點群訊號滿足該異常製程事 件群組之第二階段異常製程事件觸發條件時，判斷單元11 係提供至少一第二處理動作，以作為排除該第二階段異常 ⑩ 製程事件之操作指引。 當該第二處理動作完成後判斷單元11會接著藉由偵 測第三階段監控點203、206、209、212以擷取第三階段 異常製程事件監控點群訊號。當判斷單元11判斷第三監 控點203、206、209、212之第三階段異常製程事件監控 點群訊號滿足該異常製程事件群組之第三階段異常製程 事件觸發條件時，判斷單元11係提供至少一第三處理動 作，以作為排除第三階段異常製程事件之操作指引。 • 以下舉例說明異常製程多階段事件監控方法實際應 用於化工廠區之二個異常事件的排除流程。 製裎異常處理事件一 於製程異常處理事件一中，製程塔槽或設備的編號為 Bl、FI、F2、F3、F4、Cl、C2等；製程監控點的編號貝|J 為 FIC1、FIC1、FIC3、Til、TI2、TI3、TI4 等。以下是 以表列方式說明各階段觸發條件及其處理動作： 163PA 200928643 Excludes the operational guidelines for this first phase of anomalous process events. After the first processing action is completed, as shown in steps S130 and S140, the determining unit 11 will then detect the second-stage abnormal process event monitoring point group signal by detecting the second-stage monitoring points 202, 205, 208, and 211. . When the determining unit 11 determines that the second stage abnormal process event monitoring point group signal of the second stage monitoring point 202, 205, 208, 211 satisfies the second stage abnormal process event trigger condition of the abnormal process event group, the determining unit 11 At least one second processing action is provided as an operational guide to exclude the second stage abnormality 10 process event. After the second processing action is completed, the determining unit 11 will then monitor the point group signal by detecting the third stage monitoring point 203, 206, 209, 212 to capture the third stage abnormal process event. When the determining unit 11 determines that the third stage abnormal process event monitoring point group signal of the third monitoring point 203, 206, 209, 212 satisfies the third stage abnormal process event triggering condition of the abnormal process event group, the determining unit 11 provides At least one third processing action as an operational guide to exclude the third stage abnormal process event. • The following example illustrates the abnormal process multi-stage event monitoring method that is actually applied to the elimination process of two abnormal events in the chemical plant area. In the process exception processing event 1, the process tower slot or equipment number is Bl, FI, F2, F3, F4, Cl, C2, etc.; the number of the process monitoring point is |J is FIC1, FIC1 FIC3, Til, TI2, TI3, TI4, etc. The following is a table showing the trigger conditions of each phase and its processing actions: 16
3PA 2009286433PA 200928643
異常事件之第一階段 觸發條 件 當FIC1、FIC2之訊號皆滿足「設備運轉速 度低於FIC低警界值」之條件 處理動 1 確認B1是否異常，並確認異常原因 作 2 關閉F1至F4以及FIC3控制閥 3 排除上述異常原因，啟動B1 異常事件之第二階段 觸發條 當^、耵之〜丁^〜丁以之訊號係皆滿足厂溫 件 度小於TI低警界值」之條件 處理動 1 啟動F1至F4 作 2 維持C1及C2系統運轉 異常事件之第三階段 觸發條 當Til、TI2、TI3、TI4之訊號係皆滿足「溫 件 度到達TI高警界值」之條件 處理動 1 打開FIC1控制閥 作 製程異常處理事件二 於製程異常處理事件二中，製程塔槽或設備的編號為 UV1、UV2、R1、LV1、B2、FI、VI等；製程監控點的編 號則為PI1、FI1、FI2、PIC1、TIC1等。以下是以表列方 式說明各階段觸發條件及其處理動作： 17 200928643The first stage trigger condition of the abnormal event When the signals of FIC1 and FIC2 satisfy the condition that the device operation speed is lower than the FIC low alarm value, the operation 1 confirms whether B1 is abnormal, and confirms the abnormal cause 2 to close F1 to F4 and FIC3. The control valve 3 excludes the above-mentioned abnormal cause, and the second stage triggering bar of the B1 abnormal event is started. When the signal system of the ^1, 耵~丁~~丁 ” is satisfied, the condition of the factory temperature is less than the TI low police value” Start F1 to F4 for 2. Maintain the third stage trigger of C1 and C2 system abnormal events. When the signal systems of Til, TI2, TI3, and TI4 meet the conditions of "the temperature reaches the TI high police value", the processing is performed. FIC1 control valve for process exception handling event 2 in the process exception handling event 2, the process tower slot or equipment number is UV1, UV2, R1, LV1, B2, FI, VI, etc.; the process monitoring point number is PI1, FI1 , FI2, PIC1, TIC1, etc. The following is a table listing of the trigger conditions for each phase and its processing actions: 17 200928643
3PA 異常事件之第一階段 觸發條 件 當PI1之訊號滿足「小於PI低警界值」之 條件 處理動 作 1 現場確認UV1、UV2是否關斷 2 將現場所有母火、主火手動閥及遮斷閥關 斷 3 開啟R1之管線手動閥 4 確認B2持續運轉 異常事件之第二階段 觸發條 件 當Π1及FI2之訊號皆滿足「大於FI高警 界值」之條件 處理動 作 1 清除R1 2 將UV1、UV2復歸，且重新點燃F1之母 火及主火 異常事件之第三階段 觸發條 件 當PIC1之訊號為「維持在HC管制值」，而 TIC1是「升溫至TI高警界值」時 處理動 作 1 準備R1重新投料 於本實施例及上述製程異常處理事件一與二的說明 中，僅將一異常事件群組設定為三階段之判斷步驟，然亦 可根據各製程站之實際運作情形，將排除過程設定為更多 階段，並設定更多階段之不同觸發條件與處理動作。 18The first phase trigger condition of 3PA abnormal event When the signal of PI1 meets the condition of “less than PI low police value”, the processing action 1 confirms whether UV1 and UV2 are turned off at the scene 2 All the female fire, main fire manual valve and interrupting valve on site Shutdown 3 Open R1 pipeline manual valve 4 Confirm B2 continuous operation abnormal event second stage trigger condition When both Π1 and FI2 signals meet the condition of “greater than FI high police value” Processing action 1 Clear R1 2 UV1, UV2 The third stage trigger condition of re-ignition and re-ignition of F1's mother fire and main fire abnormal event. When PIC1 signal is "maintained at HC control value" and TIC1 is "warmed to TI high police value", action 1 preparation R1 is re-invested in the description of the first embodiment and the second processing exception event, and only one abnormal event group is set as a three-stage determination step, but the elimination process may be performed according to the actual operation situation of each process station. Set to more stages and set different trigger conditions and processing actions for more stages. 18
13PA13PA
❹ S501 ° 200928643 於製程站中之異常製程問題的完整處理流程係可藉 由第4圖中的操作紀錄單元17記錄起來。請參照第5圖’ 其4會示紀錄異常事件處理程序之流程方塊圖。如步雜S501 與S502所示’當判斷單元u判斷異常事件發生時，操作 紀錄單元17係隨即紀錄「事件發生的時間」。接著，如步 驟S503所示，判斷單元η係判斷是否開始處理此異常事 件？若是，則進入步驟S504，紀錄「事件開始處理的時 間」。然後進入步驟S5〇5，係判斷是否符合事件完成處理 條件？若是，則進入步驟S506，操作紀錄單元17係隨即 紀錄「事件完成處理的時間」。接著，進入步驟s5〇9，邦 斷異常程序處理疋否几成？若是，則如步驟所示， 此異常程序處理完成。若是異常程序處理尚未完成，則進 入步驟S511，其係啟動新的事件監控，並於啟動後回到步 於步騍S503中若是不立即著手處理此異常事件， 則例如是進人贵驟S5G9’進—步判斷異常程序處理是否完 成？由於未處理此^事件，是以會進人步驟S5U，啟動 新的監控，產回到步驟S501。 於步驟S505中’若不符合事件完成處理 入步驟洲，進一步判斷是否停止事件的監控則進 進入” _，紀錄「事件停止處理發及 原因」，旅進入步驟_。若於步驟_中止 監控，則直接進入步驟S509。 〒止事件的 在多階段的製程監控中，可以將異常事件的起始時間 200928643❹ S501 ° 200928643 The complete processing of the anomalous process problems in the process station can be recorded by the operation log unit 17 in Fig. 4. Please refer to Figure 5'. 4 shows the flow block diagram of the record exception event handler. As shown in steps S501 and S502, when the judging unit u judges that an abnormal event has occurred, the operation recording unit 17 immediately records the "time of occurrence of the event". Next, as shown in step S503, the judging unit n judges whether or not to start processing the abnormal event. If so, the process proceeds to step S504, where "the time at which the event starts processing" is recorded. Then, the process proceeds to step S5〇5, and it is determined whether the event completion processing condition is met. If so, the process proceeds to step S506, and the operation recording unit 17 immediately records the "time of event completion processing". Then, proceed to step s5〇9, is the processing of the abnormality program a few? If yes, the exception procedure is processed as shown in the step. If the abnormal program processing has not been completed, proceed to step S511, which starts a new event monitoring, and returns to step S503 after the startup, if the abnormal event is not immediately processed, then for example, enter the person S5G9' Step-by-step to determine if the abnormal program processing is completed? Since the event is not processed, the step S5U is entered, the new monitor is started, and the process returns to step S501. In step S505, if the event completion processing is not completed, the process further determines whether the monitoring of the event is stopped, and then enters "_, records "the event stops processing and causes", and the travel proceeds to step _. If the monitoring is stopped in step _, the process proceeds directly to step S509. In the multi-stage process monitoring, the start time of the abnormal event can be 200928643
SPA 與介入處理動作後的回復正常狀態時間加以記錄，對於曰 後異常事件之處置方法有極具參考的價值。此外，更可以 協助收集記錄曰誌檔中的異常事件起始與結束時間内的 工程數據資訊，將對日後失誤的根源診斷（Fault Diagnosis )有相當的貢獻。 本發明上述實施例所揭露之異常製程多階段事件監 控方法，係利用製程中的即時資訊判斷製程中是否有異常 製程問題發生。製程中如有異常發生，本發明之異常製程 ® 多階段事件監控方法可在極短的時間内提供事件操作指 引的機制，即刻顯示相對應的處理程序，以減少人為作業 的錯誤，始能有充足資訊、有效率且迅速地做出正確的標 準處理程序，以便啟動工廠的緊急應變系統來爭取操作人 員處理製程或校正設備異常的時間，藉以回復製程的正常 操作狀態，輔助操作人員將可能發生的重大事件之損失降 至最低，避免重大的危害。 綜上所述，雖然本發明已以較佳實施例揭露如上，然 ® 其並非用以限定本發明。本發明所屬技術領域中具有通常 知識者，在不脫離本發明之精神和範圍内，當可作各種之 更動與潤飾。因此，本發明之保護範圍當視後附之申請專 利範圍所界定者為準。 20The SPA and the normal state of the response after the interventional action are recorded, which is of great value for the disposal of the anomalous event. In addition, it can assist in collecting and collecting engineering data information during the start and end of abnormal events in the archives, which will contribute significantly to the Fault Diagnosis of future errors. The abnormal process multi-stage event monitoring and control method disclosed in the above embodiments of the present invention uses the instant information in the process to judge whether an abnormal process problem occurs in the process. If an abnormality occurs in the process, the abnormal process® multi-stage event monitoring method of the present invention can provide an event operation guidance mechanism in a very short time, and immediately display the corresponding processing program to reduce the error of the human operation, and can have Adequately, efficiently and quickly make the correct standard processing procedures to start the plant's emergency response system to get the operator to process the process or correct the abnormal time of the equipment, so as to return to the normal operating state of the copying process, the auxiliary operator will likely The loss of major events is minimized to avoid major hazards. In conclusion, the present invention has been disclosed above in terms of preferred embodiments, and is not intended to limit the invention. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the invention. Therefore, the scope of the invention is defined by the scope of the appended claims. 20
3 PA 200928643 【圖式簡單說明】 第1圖繪示依照本發明較佳實施例的異常製程多階 段事件監控方法之流程圖。 第2圖繪示一異常製程事件群組之示意圖。 第3A圖繪示一製造廠區及其多個製程站之示意圖。 第3B圖繪示數學函式及運算子之示意圖。 第4圖繪示第3A圖的監控模組與製程站之方塊圖。 第5圖繪示紀錄異常事件處理程序之流程方塊圖。 【主要元件符號說明】 10:監控模組 11 :判斷單元 13 :多階段事件資料庫 15 :警報通訊單元 17 :操作紀錄單元 100、200、300、400、500 :製程站 201〜212:監控點 T、T1〜T12 :標籤3 PA 200928643 [Simple Description of the Drawings] FIG. 1 is a flow chart showing a method for monitoring an abnormal process multi-stage event according to a preferred embodiment of the present invention. Figure 2 is a schematic diagram showing an abnormal process event group. FIG. 3A is a schematic diagram showing a manufacturing plant area and a plurality of process stations. FIG. 3B is a schematic diagram showing mathematical functions and operators. Figure 4 is a block diagram of the monitoring module and the process station of Figure 3A. Figure 5 is a block diagram showing the process of recording an abnormal event handler. [Main component symbol description] 10: Monitoring module 11: Judging unit 13: Multi-stage event database 15: Alarm communication unit 17: Operation recording unit 100, 200, 300, 400, 500: Process stations 201 to 212: Monitoring points T, T1~T12: label
Hnl〜Hn5 :處理動作 21Hnl~Hn5: Processing action 21
Claims (1)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW96150033A TW200928643A (en) | 2007-12-25 | 2007-12-25 | Multi-stages event monitoring method for detecting abnormal manufacturing process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW96150033A TW200928643A (en) | 2007-12-25 | 2007-12-25 | Multi-stages event monitoring method for detecting abnormal manufacturing process |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| TW200928643A true TW200928643A (en) | 2009-07-01 |
Family
ID=44864128
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW96150033A TW200928643A (en) | 2007-12-25 | 2007-12-25 | Multi-stages event monitoring method for detecting abnormal manufacturing process |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TW200928643A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9857775B2 (en) | 2010-12-28 | 2018-01-02 | International Business Machines Corporation | Method, computer program, and computer for determining system situation |
| CN111989625A (en) * | 2018-03-29 | 2020-11-24 | 安曼瑞士股份公司 | Emission control for asphalt mixing plant |
| CN113055276A (en) * | 2021-03-09 | 2021-06-29 | 井冈山大学 | Ring chat creating method, display method and system based on smart phone |
-
2007
- 2007-12-25 TW TW96150033A patent/TW200928643A/en unknown
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9857775B2 (en) | 2010-12-28 | 2018-01-02 | International Business Machines Corporation | Method, computer program, and computer for determining system situation |
| CN111989625A (en) * | 2018-03-29 | 2020-11-24 | 安曼瑞士股份公司 | Emission control for asphalt mixing plant |
| CN113055276A (en) * | 2021-03-09 | 2021-06-29 | 井冈山大学 | Ring chat creating method, display method and system based on smart phone |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP2016192000A (en) | Operation support device and operation support method | |
| Shin et al. | STPA-based hazard and importance analysis on NPP safety I&C systems focusing on human–system interactions | |
| CN105678337A (en) | Information fusion method in intelligent transformer station fault diagnosis | |
| TW200928643A (en) | Multi-stages event monitoring method for detecting abnormal manufacturing process | |
| JP6223594B2 (en) | Remote integrated monitoring and operation system | |
| US20180114176A1 (en) | Work planning system, work planning method, decision-making support system, computer program, and storage medium | |
| Zhen et al. | Development and prospects of major accident indicators in the offshore petroleum sector | |
| JP4794228B2 (en) | Emission monitoring system and method for transferring data | |
| CN105574299A (en) | Safety pre-evaluation method for rail transit signal system | |
| Adams et al. | Investigating cyber threats in a nuclear power plant. | |
| CN103164998B (en) | Training device and training method for monitoring operation | |
| Tseng et al. | Planning an emergency response centre in Southern Taiwan Science Park | |
| Kaymakci et al. | Reliability assessment of fire safety systems in railway industry: a case study | |
| JP2009271891A (en) | Maintenance inspection support device | |
| JP2011007294A (en) | Valve monitoring method and system | |
| TWI293162B (en) | ||
| Said et al. | Introduction of an emergency response plan for flood loading of Sultan Abu Bakar Dam in Malaysia | |
| Rees et al. | A discussion on the use of Eliminative Argumentation (EA) to identify Key Performance Indicators (KPIs) for the CERN LHC Machine Protection System. | |
| KR20150096182A (en) | System and Method for Supporting Decision Making and Incident Managing | |
| Kim et al. | Development of new taxonomy of inappropriate communication and its application to operating teams in nuclear power plants | |
| CN109066618B (en) | The switching value method of inspection, device and storage medium based on short-term time scale | |
| Oedewald | History of systemic approach to safety | |
| Skjerve et al. | Human and organizational contributions to safety defences in offshore oil production | |
| JP2010038831A (en) | Anomaly detection equipment in nuclear power plant | |
| Manuele | Getting Off the Plateau: How to Improve Results |