1306549 九、發明說明： 【發明所屬之技術領域】 本發明係關於一種錯誤價測系統及其管理方法，特別係 關於一種可進行自我監控㈤f_m〇nit〇Hng)及錯誤診斷 (em^diagnosing)之錯誤偵測系統及其管理方法。 【先前技術】 積體電路製造技術之持續進步已經發展出許多新的半導 體製程，其通常需要許多的輪入，必需予以精密地調控以 維持較佳的製程品質。從半導體原料至完成封裝之積體電 路元件需要許多不同的製程，例如半導體材料的成長製程 、從半導體晶棒變成晶圓的切割製程、薄膜沈積製程、# 刻製程、佈植製程、封裝製程以及測試程序，而這些製程 彼此具有相當的差異性，大都在不同的區域進行，且不同 的製程大都採用不同的控制技術。 '圖“丁八有即時:貝料庫之錯誤偵測系統，其係揭示於 美國專利號仍6,868,512。該錯誤_系統包含至少一資料 荒集來源125、一即時資料庫⑴以及-資料庫管理單元120 。該資料笼集來源125係用以產生錯誤谓測及更正㈣ detection and correcU〇n ’ FDC)資料，而該資料庫管理單元 :20則係用以將該咖資料儲存於該即時資料庫⑴，其中 /貝料庫S理單χ 12〇亦具有—進人資料目錄W。簡言之 ’該錯誤偵測系統之運作大略為接收進人之㈣資料、儲 存進入之FDC資料於該即時資料庫丨丨〇並產生—進入資料 、亲13〇( .、#曰不進入之FDC資料由一預設之進入資料處理 P26164 PD0105 005101214 -6- 1306549 器140負責處理）、以 13〇所指之資_ _ mDC資料至該進入資料目錄 未設計任二^：°。然而，習知之錯誤偵測系統並 部模組發生錯誤==誤診斷機制’因而在系統之内 【發明内容】 子補求而造成更大的損害。 本發明之主要目Μ & ，”如供一種可進行自我監控及錯誤診 斷之錯誕偵測系統及其管理方法。 人=成亡述目的’本發明提出一種錯誤偵測系統，其包 3 一以鬼集至少—機台之參數的資料舰器、至少一可 該參數超出—預設規範時發出—異常訊號的錯誤感測模 ^ 一可在該錯誤感測模組不正常運作時予以重啟動的監 控U、一用以控制該資料伺服器、該錯誤感測模組及該 監控模組的遠端控制器。較佳地，該機台、該資料伺服器 、該錯誤感測模組、該監控模組及該遠端控制器係藉由網 路相互麵合。此外’該遠端控制器亦可用以診斷該資料伺 服器、該錯誤感測模組及該監控模組之錯誤。 根據上述目的’本發明提出一種錯誤價測系統之管理方 7 ’其包含館存至少-機台之參數於―資制服^、一錯 誤感測模組採用後進先出方式檢查該參數是否超過一預設 規範並在該參數超過該預設規範時產生—異常訊號、一監 控模組檢查該錯誤感測模組是否正常運作並在該錯誤感測 模組不正常運作時予以重新啟動。 較佳地，該錯誤感測模組在讀取該資料伺服器儲存之參 數時-併更新-第-存取時間，而該監控模組則藉由檢查 Ρ2όιό4 PDoi〇5 005101214 1306549 “存取時間是否在—苐—間隔内被更新，判斷該錯誤 ί:拉組是否異常。此外，該遠端控制器亦可藉由檢查該 弟：存，時間是否在一第一間隔内被更新，診斷該錯誤感 測模Τ是否發生錯誤。同樣地，該監控模組在讀取該錯誤 感測模Μ之第—存取時間時亦—併更新—第二存取時間， 而该遠端控制器則藉由檢查該第二存取時間是否在一第二 間内被更新，診斷該監控模組之錯誤。1306549 IX. Description of the invention: [Technical field to which the invention pertains] The present invention relates to a false price measurement system and a management method thereof, in particular to a self-monitoring (5) f_m〇nit〇Hng) and error diagnosis (em^diagnosing) Error detection system and its management method. [Prior Art] The continuous advancement of integrated circuit manufacturing technology has developed many new semi-conducting processes, which usually require a lot of rounding and must be precisely controlled to maintain better process quality. From semiconductor materials to packaged integrated circuit components, many different processes are required, such as the growth process of semiconductor materials, the cutting process from semiconductor ingots to wafers, thin film deposition processes, #刻加工, implantation processes, packaging processes, and Test procedures, and these processes are quite different from each other, mostly in different areas, and different processes mostly use different control techniques. 'Figure' Ding Ba has instant: the error detection system of the shell library, which is revealed in the US patent number is still 6,868,512. The error _ system contains at least one data source 125, an instant database (1) and - database management Unit 120. The data collection source 125 is used to generate error prediction and correction (4) detection and correcU〇n ' FDC) data, and the database management unit: 20 is used to store the coffee data in the real-time data. Library (1), in which / 料 料 S χ χ 〇 〇 〇 〇 〇 〇 〇 进 进 进 进 。 。 。 。 。 。 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该 该The real-time database is generated and entered into - the data, the pro- 13〇 (., #曰 does not enter the FDC data from a preset entry data processing P26164 PD0105 005101214 -6- 1306549 device 140 is responsible for processing), 13 〇 Refers to the capital _ _ mDC data to the entry data directory is not designed for any two ^: °. However, the conventional error detection system parallel module error == misdiagnosis mechanism 'and thus within the system [invention content] Make up for more The main object of the present invention damage Μ &., "Such as for Christmas A detection system for fault management and self-monitoring and diagnosis of errors. The present invention proposes an error detection system, which includes a data ship with a ghost set at least a parameter of the machine, and at least one parameter that can be issued when the parameter exceeds the preset specification - an abnormal signal The error sensing module is a monitoring U that can be restarted when the error sensing module is not operating normally, and a remote controller for controlling the data server, the error sensing module, and the monitoring module . Preferably, the machine, the data server, the error sensing module, the monitoring module and the remote controller are mutually connected by a network. In addition, the remote controller can also be used to diagnose errors of the data server, the error sensing module, and the monitoring module. According to the above object, the present invention proposes a management unit 7' of the error price measurement system, which includes at least the parameters of the library at the time of the machine, and the error sensing module adopts a last in first out mode to check whether the parameter exceeds one. The preset specification is generated when the parameter exceeds the preset specification - an abnormal signal, a monitoring module checks whether the error sensing module is operating normally, and restarts when the error sensing module is not operating normally. Preferably, the error sensing module reads the parameter stored by the data server - and updates the - access time, and the monitoring module checks the access time by checking ό 2 ό ό 4 PDoi 〇 5 005101214 1306549 Whether it is updated in the interval - 间隔 - interval, determine the error ί: whether the pull group is abnormal. In addition, the remote controller can also check whether the time is updated in a first interval by checking the time: The error sensing module detects whether an error has occurred. Similarly, the monitoring module also reads and updates the second access time when reading the first access time of the error sensing module, and the remote controller The error of the monitoring module is diagnosed by checking whether the second access time is updated in a second interval.
該遠端控制器亦可藉由與該資料伺服器之交握 (handShaking)而診斷該資料伺服器之錯誤，亦即該遠端控 制器可傳送一僅測訊號至該資料飼服器，並根據該資料飼 服器是否在一第三間隔内回傳一確收訊號，判斷該資料飼 服是否正常運作並診斷該資料伺服器是否發生錯誤。此外 :該監控模組可檢查進入該資料祠服器之參數的資料流量 是否超過該錯誤感測模組之處理能力，並在檢查結果為真 時啟動另一錯誤感測模組以協助處理該參數之錯誤分析。、 相較於習知技藝，本發明之錯誤偵測系統利用該監控模 組監控該錯誤感測模組之運作，並在該錯誤感測模組不正 常運作時予以重新啟動。若該監控模組嘗試重新啟動該錯 誤感測模組-職:欠數仍職效日U通知㈣端控制 代替該錯誤感測模組進行該參數之錯誤分析工作。另，^ 遠端控制器亦可藉由與該資料伺服器之交握而診斷該資料 祠服益之錯誤。 、 此外’該監控模組可檢查進入該資料伺服器之參數的資 料流量是否超過該錯誤感測模組之處理能力，並在檢杳結 P26164 PD0105 005101214 1306549 果為真時啟動另一錯誤感測模組以協助處理該參數之錯誤 分析。再者，該監控模組亦可設定一存取數於該資料伺服 益’而該錯誤感測模組根據該存取數以批次方式從該資料 祠服器讀取該參數。 【實施方式】 圖1例不本發明之錯誤偵測系統10。該錯誤偵測系統1 〇 包含一用以蒐集至少一機台12之參數的資料伺服器14、至 少一可在該參數超出一預設規範時發出一異常訊號的錯誤 • 感測模組16、一可在該錯誤感測模組16不正常運作時予以 重啟動的監控模組U、一用以控制該資料伺服器14、該錯 誤感測模組1 6及該監控模組丨8的遠端控制器2〇。較佳地， 遠機台12、該資料伺服器14、該錯誤感測模組16、該監控 模組18及該遠端控制器2〇係藉由網路相互耦合。此外，該 遠端控制器20可藉由另一網路32連接至—工廠3〇(例如新 建廠房）’其包含另一錯誤偵測系統10及機台12，如此該工 廠30即可經由該遠端控制器2〇分享資料，有助於該工廠30 在短時間内建立製程之控制能力。 車X佳地，该遠端控制器2〇亦可用以診斷該資料伺服器i 4 、忒錯誤感測模組16及該監控模組18是否發生錯誤。該機 口可為里測機台、幫浦或製程機台，該製程機台包含微影 步進機台、蚀刻機台、沈積機台、平坦化機台、快速熱處 =機台、離子佈植機台等等。該參數可為機台之感測器測 知之原始數據，例如製程機台之物理量測資料或量測機台 測得待測元件之電性資料。 P26164 PD〇i〇s 005101214 ^9- 1306549 圖3至圖5例示本發明之錯誤偵測系統1〇之管理方法。該 資料伺服器14接收並儲存至少一機台12之參數、該錯誤感 測模組1 6再採用後進先出的方式檢查該參數是否超過一預 設規範’並在該參數超過該預設規範時產生一異常訊號。 此外，該錯誤感測模組丨6在讀取該資料伺服器14儲存之參 數時亦一併更新一第一存取時間。 參考圖4，该監控模組1 8檢查該錯誤感測模組16是否正常 運作，並在該錯誤感測模組16不正常運作時予以重新啟動 。该錯誤感測模組丨6在讀取該資料伺服器丨4儲存之參數時 一併更新一第一存取時間，而該監控模組丨8則可藉由檢查 该第一存取時間是否在一第一間隔内被更新，判斷該錯誤 感測模組16是否發生異常。若該第一存取時間並未在該第 間搞内更新，5亥監控模組18則接著檢查一第一計數器之 儲存值’其記錄該錯誤感測模組丨6之重新啟動次數。 若。亥第一4數器之儲存值並未超過一預設值（代表可容 許該錯誤感測模組16之最大重新啟動次數），該監控模組i 8 即重新啟動該錯誤感測模組16並將該第一計數器之儲存值 增加1。相對地，若該第一計數器之儲存值超過該預設值（ 亦即該監控模組18先前曾嘗試重新啟動該錯誤感測模組16 ，但無效），該錯誤感測模組丨6已經無法正常地運作，該監 控模組1 8即通知該遠端控制器2〇代替該錯誤感測模組丨6進 行該參數之錯誤分析工作，亦即該遠端控制器係作為該錯 誤感測模組之備用裝置。 參考圖5 ’該監控模組1 8在讀取該錯誤感測模組丨6之第一 P26164 PD0105 005101214 •10- 1306549 存取時間時亦一併更新一第二存取時間，而該遠端控制器 20即可藉由檢查該第二存取時間是否在一第二間隔内被更 新，#斷該監控模組1 8之錯誤。若該第二存取時間未在該 苐一間隔内被更新，該运端控制器2 〇即檢查一第二計數器 之儲存值是否超過一預設值（其記錄該監控模組18之重新 啟動·•入數）。若該第二計數器之储存值並未超過一預設值， δ玄遠端控制器2〇即重新啟動該監控模組丨8並將該第二計數 器之儲存值增加1。相對地，若該第二計數器之儲存值超過 • 該預設值，代表該監控模組20己經無法運作，該遠端控制 器20即代替該監控模組18進行監控工作，亦即該遠端控制 器係作為該監控模組之備用裝置。 此外，該資料伺服器14亦可計算進入之資料流量（即在一 •k時間内進入之參數的數量），該監控模組18可檢查進入該 資料伺服器14之資料流量是否超過該錯誤感測模組16之處 理能力，並在檢查結果為真時啟動另一錯誤感測模組丨6以 協助處理該參數之分析。較佳地，判斷進入該資料伺服器 _ 14之資料流量是否超過該錯誤感測模組16之處理能力，可 藉由檢查儲存於該資料伺服器14之參數是否在一第四間隔 内被讀取。 該遠端控制器20亦可藉由與該資料伺服器14之交握而診 斷該資料伺服器14之錯誤，亦即該遠端控制器2〇亦可傳送 一偵測甙號至該資料伺服器丨4，再檢查該資料伺服器丨4是 否在一第二間隔内回傳—確收訊號，藉以判斷該資料伺服 14是否正常運作並診斷該資料伺服器14之錯誤。再者，該 P26164 PDoi〇5 005101214 1306549 監控模組1 8亦可設定一存取數於該資料伺服器i 4，而該錯 誤感測模組16則根據該存取數以批次方式從該資料伺服器 14讀取該參數。若進入該資料伺服器〗4之參數的資料流量 超過該錯誤感測模組之處理能力，該監控模組丨8可藉由提 歼該存取數之數值而提昇該錯誤感測模組丨6之分析效率。 相較於習知技藝，本發明之錯誤偵測系統利用該監控模 組監控該錯誤感測模組之運作，並在該錯誤感測模組不正 常運作時予以重新啟動。若該監控模組嘗試重新啟動該錯 誤感測模組一預定次數仍然無效時，則通知該遠端控制器 代替該錯誤感測模組進行該參數之錯誤分析工作。另，該 遠端控制器亦可藉由與該資料伺服器之交握而診斷該資料 伺服器之錯誤。 此外，該監控模組可檢查進入該資料伺服器之參數的資 料流量是否超過該錯誤感測模組之處理能力，並在檢查結 果為真時啟動另一錯誤感測模組以協助處理該參數之錯誤 分析。再者，該監控模組亦可設定一存取數於該資料伺服 器，而該錯誤感測模組根據該存取數以批次方式從該資料 伺服器讀取該參數。 本發明之技術内容及技術特點已揭示如上，然而熟悉本 項技術之人士仍可能基於本發明之教示及揭示而作種種不 背離本發明精神之替換及修飾。因此，本發明之保護範圍 應不限於實施例所揭示者’而應包括各種不背離本發明之 替換及修飾’並為以下之申請專利範圍所涵蓋。 【圖式簡要說明】 P26164 PD〇l〇5 005101214 •12- 1306549 圖1例示一習知之錯誤偵测系統； 圖2例示本發明之錯誤偵測系統；以及 圖3至圖5例示本發明之錯誤伯測系統之管理方法。 【主要元件符號說明】 P26X64 PD〇l〇5 005101214 10 錯誤偵測系統 12 機台 14 資料伺服器 16 錯誤感測模組 18 ii控模組 20 遠端控制器 30 工礙 32 網路 110 即時資料庫 120 資料庫管理單 125 資料蒐集來源 130 進入資料目錄 140 資料處理器 13·The remote controller can also diagnose the error of the data server by handshaking with the data server, that is, the remote controller can transmit a test signal only to the data feeder, and According to whether the data feeding device returns a confirmation signal within a third interval, it is determined whether the data feeding service is operating normally and whether the data server is diagnosed to have an error. In addition, the monitoring module can check whether the data flow of the parameter entering the data server exceeds the processing capability of the error sensing module, and when the check result is true, start another error sensing module to assist in processing the Error analysis of parameters. Compared with the prior art, the error detection system of the present invention uses the monitoring module to monitor the operation of the error sensing module and restarts when the error sensing module is not operating normally. If the monitoring module attempts to restart the error sensing module - job: the number of still operating days U notification (four) end control instead of the error sensing module for error analysis of the parameter. In addition, the remote controller can also diagnose the error of the data by the handshake with the data server. In addition, the monitoring module can check whether the data flow of the parameter entering the data server exceeds the processing capability of the error sensing module, and initiate another error sensing when the check node P26164 PD0105 005101214 1306549 is true. The module assists in the analysis of errors in this parameter. Moreover, the monitoring module can also set an access number to the data server, and the error sensing module reads the parameter from the data server in batch mode according to the access number. [Embodiment] FIG. 1 illustrates an error detection system 10 of the present invention. The error detection system 1 includes a data server 14 for collecting parameters of at least one machine 12, at least one error that can be sent when the parameter exceeds a preset specification, and a sensing module 16 a monitoring module U that can be restarted when the error sensing module 16 is not functioning properly, and a remote control unit 14 for controlling the data server 14, the error sensing module 16 and the monitoring module 丨8 End controller 2〇. Preferably, the remote station 12, the data server 14, the error sensing module 16, the monitoring module 18, and the remote controller 2 are coupled to each other by a network. In addition, the remote controller 20 can be connected to the factory 3 (for example, a new factory) by another network 32, which includes another error detection system 10 and the machine 12, so that the factory 30 can pass through the The remote controller 2 shares the data and helps the plant 30 to establish process control capabilities in a short period of time. The remote controller 2 can also be used to diagnose whether the data server i 4 , the error sensing module 16 and the monitoring module 18 have an error. The machine mouth can be a measuring machine, a pump or a processing machine. The processing machine comprises a lithography stepping machine, an etching machine, a deposition machine, a flattening machine, a rapid hot place=machine, and an ion. Planting machine and so on. This parameter can be the raw data measured by the sensor of the machine, such as the physical measurement data of the process machine or the electrical data of the component to be tested measured by the measuring machine. P26164 PD〇i〇s 005101214 ^9- 1306549 FIGS. 3 to 5 illustrate a method of managing the error detection system 1 of the present invention. The data server 14 receives and stores the parameters of at least one machine 12, and the error sensing module 16 checks whether the parameter exceeds a preset specification by using a last-in, first-out mode and exceeds the preset specification in the parameter. An abnormal signal is generated. In addition, the error sensing module 丨6 also updates a first access time when reading the parameters stored by the data server 14. Referring to FIG. 4, the monitoring module 18 checks whether the error sensing module 16 is operating normally and restarts when the error sensing module 16 is not operating normally. The error sensing module 丨6 updates a first access time when reading the parameters stored in the data server 丨4, and the monitoring module 丨8 can check whether the first access time is It is updated in a first interval to determine whether an abnormality has occurred in the error sensing module 16. If the first access time is not updated in the middle, the 5H monitoring module 18 then checks the stored value of a first counter, which records the number of restarts of the error sensing module 丨6. If. The stored value of the first four digitizers does not exceed a preset value (representing the maximum number of restarts of the error sensing module 16), and the monitoring module i 8 restarts the error sensing module 16 And increase the stored value of the first counter by one. In contrast, if the stored value of the first counter exceeds the preset value (that is, the monitoring module 18 has previously attempted to restart the error sensing module 16 but is invalid), the error sensing module 丨6 has The monitoring module 18 notifies the remote controller 2 to perform the error analysis of the parameter instead of the error sensing module ,6, that is, the remote controller acts as the error sensing. A backup device for the module. Referring to FIG. 5, the monitoring module 18 also updates a second access time when reading the first P26164 PD0105 005101214 • 10 - 1306549 access time of the error sensing module 丨6, and the remote access time is updated. The controller 20 can interrupt the error of the monitoring module 18 by checking whether the second access time is updated in a second interval. If the second access time is not updated in the first interval, the terminal controller 2 checks whether the stored value of the second counter exceeds a preset value (which records the restart of the monitoring module 18) ·•Enter the number). If the stored value of the second counter does not exceed a preset value, the δ remote controller 2 restarts the monitoring module 丨8 and increases the stored value of the second counter by one. In contrast, if the stored value of the second counter exceeds the preset value, indicating that the monitoring module 20 has not been operated, the remote controller 20 replaces the monitoring module 18 for monitoring, that is, the far The end controller acts as a backup for the monitoring module. In addition, the data server 14 can also calculate the incoming data traffic (ie, the number of parameters entered in a time of one k), and the monitoring module 18 can check whether the data traffic entering the data server 14 exceeds the error. The processing capability of the module 16 is tested and another error sensing module 丨6 is activated when the check result is true to assist in processing the analysis of the parameter. Preferably, it is determined whether the data flow entering the data server _ 14 exceeds the processing capability of the error sensing module 16 by checking whether the parameter stored in the data server 14 is read in a fourth interval. take. The remote controller 20 can also diagnose the error of the data server 14 by the handshake with the data server 14, that is, the remote controller 2 can also transmit a detection nickname to the data servo. Then, it is checked whether the data server 丨4 is returned in a second interval to confirm the signal, thereby determining whether the data server 14 is operating normally and diagnosing the error of the data server 14. Furthermore, the P26164 PDoi〇5 005101214 1306549 monitoring module 18 can also set an access number to the data server i 4, and the error sensing module 16 batches the number according to the access number. The data server 14 reads this parameter. If the data flow entering the parameter of the data server 〖4 exceeds the processing capability of the error sensing module, the monitoring module 丨8 can enhance the error sensing module by raising the value of the access number. 6 analysis efficiency. Compared with the prior art, the error detection system of the present invention uses the monitoring module to monitor the operation of the error sensing module and restarts when the error sensing module is not operating normally. If the monitoring module attempts to restart the error sensing module for a predetermined number of times, it is notified that the remote controller replaces the error sensing module to perform error analysis of the parameter. In addition, the remote controller can also diagnose the error of the data server by the handshake with the data server. In addition, the monitoring module can check whether the data flow of the parameter entering the data server exceeds the processing capability of the error sensing module, and activate another error sensing module to assist in processing the parameter when the check result is true. Error analysis. Moreover, the monitoring module can also set an access number to the data server, and the error sensing module reads the parameter from the data server in batch mode according to the access number. The technical contents and technical features of the present invention have been disclosed as above, and those skilled in the art can still make various substitutions and modifications without departing from the spirit and scope of the invention. Therefore, the scope of the invention is not to be construed as being limited by the scope of the inventions BRIEF DESCRIPTION OF THE DRAWINGS P26164 PD〇l〇5 005101214 • 12-1306549 FIG. 1 illustrates a conventional error detection system; FIG. 2 illustrates the error detection system of the present invention; and FIGS. 3 to 5 illustrate errors of the present invention. The management method of the beta system. [Main component symbol description] P26X64 PD〇l〇5 005101214 10 Error detection system 12 Machine 14 Data server 16 Error sensing module 18 ii Control module 20 Remote controller 30 Obstruction 32 Network 110 Real-time data Library 120 Database Management Sheet 125 Data Collection Sources 130 Access to Data Catalog 140 Data Processor 13·