201226101 六、發明說明： 【發明所屬之技術領域】 本發明係關於-種狀態偵測方法及其裝置尤浐一種 - 用以偵測工具機刀具狀態的偵測方法及其裝置者。θ 【先前技術】 按，既有工具機於運轉時，為了考慮刀具產生斷裂而 無警不的狀況時’會使得生產效率與品質下滑，導致生產 成本提升，而在沒有刀具磨耗的線上監測輔助的狀況下， 操作者必須保守的定時更換刀具，但因每支刀具壽命表現 的差異相當大，相對會造成換刀時刀具並未損冑，且換刀 頻率的增加亦會降低系統運作的效率，其中我國發明專利 第490357號「檢測切削工具機切削刀具異常之裝置與其方 法」一案係揭示一振動檢測裝置、一異常狀態檢測褒置、 —停止指示裝置及一通報裝置，於檢測時主要係透過該振 動檢測裝置檢測工具機刀具進行加工時所發生的振動經 φ 由5亥異常狀態檢測裝置計算所檢測之振動中超過規定值之 峰值發生次數，其中當峰值發生次數超過規定臨界值時， 輸出一信號至該停止指示裝置中並對於該工具機進行停止 加工之操作，再經由該通報裝置通知作業員更換切削刀 具，藉以提供一檢測工具機切削刀具異常之裝置及其方法. 然而’既有檢測工具機刀具異常之裝置及其方法於使 用時’雖可透過該振動檢測裝置對於刀具進行偵測，但僅 以單—振動檢測裝置進行偵測的方式，無法辨識與刀具狀 是化相關说號在能量大小或頻寬相同的雜訊號，亦或因 系統狀態變動產生的訊號變易’造成辨識抗雜訊的能力與 201226101 穩定度變差，容易造成刀具狀態的誤判而使系統無法於實 際的工業上進行應用，誠有加以改良之處。 【發明内容】 因此，本發明人有鑑於目前工具機刀具檢測方法及裝 置，辨識抗雜訊的能力與穩定度差的不足與問題，特經過 不斷的研究與試驗，終於發展出一種能改進現有缺失之本 發明。201226101 VI. Description of the Invention: [Technical Field] The present invention relates to a state detecting method and a device thereof - a method for detecting a tool tool state and a device thereof. θ [Prior Art] Press, when the machine tool is running, in order to consider the tool breakage and no alarm, it will cause the production efficiency and quality to decline, resulting in higher production costs, and on-line monitoring without tool wear. In the case of the operator, the operator must change the tool at a conservative timing, but the difference in the life performance of each tool is quite large, which will result in the tool not being damaged during the tool change, and the increase of the tool change frequency will also reduce the efficiency of the system operation. In the case of the Chinese Patent No. 490357, "A device for detecting an abnormality of a cutting tool of a cutting tool machine and a method thereof", a vibration detecting device, an abnormal state detecting device, a stop indicating device and a notification device are disclosed, which are mainly detected. The vibration generated by the machine tool is detected by the vibration detecting device, and the number of occurrences of the peak value exceeding the predetermined value is calculated by the 5H abnormal state detecting device, wherein the peak occurrence times exceed the predetermined threshold value. , outputting a signal to the stop indicating device and stopping the machine tool The processing operation, and then notifying the operator to replace the cutting tool via the notification device, thereby providing a device for detecting the abnormality of the cutting tool of the machine tool and the method thereof. However, the device and the method for detecting the abnormality of the tool of the tool machine are used. The vibration detecting device can detect the tool, but only the single-vibration detecting device can detect the noise signal with the same energy size or bandwidth as the tool shape. The signal generated by the change of the system state becomes easy. The ability to identify anti-noise is worse than the stability of 201226101. It is easy to cause misjudgment of the tool state and the system cannot be applied in the actual industry. It is improved. SUMMARY OF THE INVENTION Therefore, the inventors have in view of the current methods and devices for detecting tool cutters, and the problems and problems of identifying the ability to resist noise and the difference in stability, and through continuous research and experimentation, finally developed an improvement that can be improved. The invention is missing.
本發明之主要目的係在於提供一種多重感應之工具4 :具狀態偵測方法及其裝置’其係透過整合多重感應及t 號轉換與辨識的融合方式，即時地且準輕得知刀具斷$ :刀具磨耗之狀態，進而提供一方便安裝、即時檢測及表 高偵測可靠度之目的者。 為達到上述目的’本發明係提供一種多重感應… 機刀具狀態㈣方法’其操作流程係包含有： 、儀器設置：準備—偵測裝置，該㈣裝置設有-多！ 感應組、一訊號處理組及一監控組，該多重感應組設有劣 數個可分別對工具機刀具振動、聲音及電流變化進輸 的感應器’該訊號處理組與該多重感應組相電性連接且言; 貝料擷取卡及一電腦，該資料擷取卡與各感應器相驾 ::接，該電腦與該資料掏取卡相電性連接且設有一訊韻 、模組、-特徵選取模組及—辨識模組，該訊號轉換相 組與該資料擷取卡相連接，該特 ,, Λ将徵選取槟組與該訊號轉抬 連接，該辨識模組與該特徵選取模組相連接，該監 二广訊號處理組相電性連接且設有_與該電 的警報器； 夂奴 201226101 μ線上偵測與訊號處理··將該多重感應組的各感應器裝 :於工具機上’藉以對於該工具機刀具進行加工訊號的 肩測，將各感應器所取得的振動、聲音及電流訊號，經由 5亥貝訊擷取卡操取並傳送至該電腦中進行訊號的處理盘呈 現，在訊號處理過程中，藉由該訊號轉換模组將操取的原 。έ號進行號轉換，於訊號轉換後經由經特徵選取模組 選出”該刀具狀態相關的訊號，再經由該辨識模組進行 辨4藉以產生一與刀具狀態相關的判斷訊號； 偵測結果：將該判斷訊號輸出至該監控組中，即可讓 使用者即時得知該刀具的狀態，且可透過傳送訊號至該警 報器的方式，告知操作者該刀具的狀態。 進步，在儀器設置的操作步驟中，該多重感應組係 設有-加速規感應器、一聲射感應器、一麥克風陣列感應 器及一功率感應器。 "The main object of the present invention is to provide a multi-sensing tool 4: a method for detecting a state and a device thereof, which integrates multiple sensing and t-number conversion and recognition, and instantly and accurately knows that the tool is broken. : The state of tool wear, which provides a convenient installation, instant detection and high detection reliability. In order to achieve the above object, the present invention provides a multi-induction...machine tool state (four) method'. The operation process includes:, instrument setting: preparation-detection device, and (4) device is provided with - more! The sensing group, the signal processing group and the monitoring group, the multi-sensor group is provided with a plurality of sensors for separately inputting vibration, sound and current changes of the tool machine tool. The signal processing group is electrically connected to the multi-sensor group Sexual connection and speaking; a material pick-up card and a computer, the data capture card and each sensor::, the computer is electrically connected with the data capture card and has a signal, module, a feature selection module and an identification module, wherein the signal conversion phase group is connected to the data capture card, and the special selection is selected to be connected to the signal and the identification module and the feature are selected. The modules are connected, and the monitoring and processing unit of the second broadcasting signal is electrically connected and has an alarm device for the electrician; the slave slave 201226101 μ line detection and signal processing··installing the sensors of the multi-sensor group: On the machine tool, the vibration, sound and current signals obtained by each sensor are taken and transmitted to the computer for signal measurement. Processing disk presentation, processed in the signal In the signal converting module by taking the original operation. The nickname performs the number conversion, and after the signal conversion, the signal related to the tool state is selected by the feature selection module, and then the identification module is used to generate a judgment signal related to the tool state; the detection result: The judgment signal is outputted to the monitoring group, so that the user can immediately know the state of the tool, and can inform the operator of the state of the tool by transmitting a signal to the alarm. Progress, operation in the instrument setting In the step, the multiple sensing group is provided with an acceleration gauge sensor, an acoustic sensor, a microphone array sensor and a power sensor.
再進步在線上偵測與訊號處理的操作步驟中該 加速規感應器與該聲射感應器係裝設於該工具機的夾= 上，該麥克風陣列感應器係設置於該工具機平台周圍的任 一位置上，而該功率感應器係與該工具機的主軸相連接。Further, in the operation step of detecting and signal processing on the line, the accelerometer sensor and the sound sensor are mounted on the clamp of the machine tool, and the microphone array sensor is disposed around the machine tool platform. In either position, the power sensor is connected to the main shaft of the machine tool.
較佳地’在儀器設置的操作步驟中 與該訊號轉換模組、該特徵選取模組或該辨識模組相連接 的融合模組，藉以對於轉換訊號、特徵訊號或辨識訊號進 行融合，進而提供與刀具狀態相符合之訊號。 較佳地，在儀器設置的操作步驟中，該融合模組為一 與該訊號轉換模組相連接的訊號融合組’且在線上该測與 訊號處理的操作步驟中’將該訊號轉換模組轉換後的可 5 201226101 ，並經該特徵選取模組選 再經由該辨識模組進行辨 號《，經由S亥訊5虎融合組進行融合 取出與該刀具狀態相關的訊號， 識，藉以產生一與刀具狀態相關的判斷訊號。 ，該融合模組為一 9特徵融合組，且在線上偵測與Preferably, the fusion module connected to the signal conversion module, the feature selection module or the identification module in the operation step of the instrument is configured to fuse the conversion signal, the characteristic signal or the identification signal, thereby providing A signal that matches the state of the tool. Preferably, in the operation step of the instrument setting, the fusion module is a signal fusion group connected to the signal conversion module and in the operation step of the measurement and signal processing on the line, the signal conversion module The converted can be 201226101, and the feature selection module is selected and then identified by the identification module, and the signal related to the tool state is extracted through the S-Xun 5 tiger fusion group, thereby generating a The judgment signal related to the tool status. The fusion module is a 9-feature fusion group, and is detected on the line.
與刀具狀態相關的判斷訊號。 較佳地，在儀器設置的操作步驟中， 與該特徵選取模組相連接的特徵融合組， 較佳地，在儀器設置的操作步驟中，該融合模組為一 與該辨識模組相連接的決策融合組，且在線上情測與訊號 處理的操作步驟中，將該訊號轉換模組轉換後的訊號，經 该特徵選取模組選取出與該刀具狀態相關之訊號，並由該 辨識模組進行特徵訊號的辨識，最後由該決策融合組進行 融合的方式，產生一與刀具狀態相關的判斷訊號β 車父佳地’在儀器設置的操作步驟中，該訊號處理组於 為料擷取卡與该電腦間設有一與該資料擷取卡相電性連 接的類比/數位轉換器，藉以將類比訊號轉換成一數位訊號。 較佳地’在線上偵測與訊號處理的操作步驟中，該訊 破轉換模組係藉由一小波轉換方程式進行訊號的轉換。 摩父佳地，在線上偵測與訊號處理的操作步驟中，該訊 號轉換模組係藉由一傅立葉轉換方程式進行訊號的轉換。 較佳地’在線上偵測與訊號處理的操作步驟中，該訊 號轉換模組係藉由一快速傅立葉轉換方程式進行訊號的轉r 201226101 換。 較佳地，在線上偵測與訊號處理的操作步驟中，該辨 識模組係以一費雪線性辨識函數作為訊號辨識的方法。 杈佳地，在線上偵測與訊號處理的操作步驟中，該辨 識模組係以一類神經網路作為訊號辨識的方法。 較佳地，在線上偵測與訊號處理的操作步驟中，該辨 識模組係以一模糊邏輯辨識方程式作為訊號辨識的方法。 本發明另提供一種多重感應之工具機刀具狀態偵測裝 鲁置，其係包含有-多重感應組、一訊號處理組及一監控組， 其中： 該多重感應組係設有複數個可對工具機刀具振動、3 音及電流變化進行偵測的感應器； 該訊號處理組係與該多重感應組相電性連接且設有-資料擷取卡及-電腦，該資料擷取卡與各感應器相電心 接’而該電腦與該資料#貞取卡相電性連接且設有—訊號車 換模組、-特徵選取模組及一辨識模組，該訊號轉換模自 與該資料擷取卡相連接，該特徵選取模組與該訊號轉換右 組相連接，該辨識模組與該特徵選取模組相連接；以及 該監控組與該訊號處理組相電性連接且設有一與該售 腦相連接的警報器。 進步°亥夕重感應組係設有—加速規感應器、一聋 射感應n、-麥克風陣列感應器及_功率感應器。 再進-步’該電腦係設有一與該訊號轉換模組'該特 徵選取模組或該辨識模組相連接的融合模組，藉以對於轉 換訊號、特徵訊號或辨識訊號進行融合，進而提供與刀具 201226101 狀態相符合之訊號。 車父佳地’該融合模組為一與該訊號轉換模組相連接的 訊號融合組。 較佳地’該融合模組為一與該特徵選取模組相連接的 特徵融合組。 較佳地，该融合模組為一與該辨識模組相連接的決策 融合組。 較佳地，在儀器設置的操作步驟中，該訊號處理組於 鲁該資料擷取卡與該電腦間設有一與該資料操取卡相電性連 接的類比/數位轉換器，藉以將類比訊號轉換成一數位訊號。 藉由上述之技術手段，本發明多重感應之工具機刀具 狀態情測方法及其裝置，主要係透過整合不同感應特性的 感應器的方式，藉以分別對於刀具加工所產生的振動、聲 音及電流進行訊號的偵測，不僅可透過交又比對的方式降 低雜訊的干擾程度，且可透過該融合模組的訊號整合方 式，對於轉換讯號、特徵訊號或辨識訊號進行融合，提供 籲-可靠度高的判斷訊號，進而大幅提升線上刀具狀態（斷裂 或磨耗）的監測準確性，並以警報器警示操作者，或者透過 與該工具機的控制器相連接的方式，對於該工具機即時停 機或實施刀具的更換； 因此，本發明之多重感應 之工具機刀具狀態偵測裝 器裝設於該工 置，於使用時僅需將該多重感應組的各感應 具機的平台上，即可分別對於加工的刀具進行聲音的偵 測，不僅安裝上相當方便，且可在不同之需求下，任意地 應用於不同軸數之銑削、鑽削與攻牙工具機等等的工具機 201226101 上，藉以提升系統使用之稼動率，而達到節省投資成本之 目的，且可同時減少操作人員的數量，並大大提升以振動、 聲音及電流訊號為基礎之線上刀具斷裂或磨耗狀態監測之 正確性，提供一方便安裝、即時檢測及提高偵測可靠度之 偵測方法及裝置者。 【實施方式】 為能詳細瞭解本發明的技術特徵及實用功效，並可依 照說明書的内容來實施，玆進一步以圖式（如圖i至5所示） 所示的較佳實施例，詳細說明如后： 本發明之目的在於提供一多重感應之工具機刀具狀態 偵測方法及其裝置’其係安裝於一工具機5〇上，且透過= 重感應的方式進行偵測，即時地並準確地得知刀具5彳狀 態，進而提供一方便安裝、即時檢測及提高偵測可靠度之 偵測方法及裝置者。 本發明之多重感應之工具機刀具狀態债測方法，其操 作流程係包含有： A '儀器設置：準備一偵測裝置，該偵測裝置係設有一 夕重感應組1 0、一 sfl號處理組2 0及一監控組3 0，其中1 多重感應組1 0係設有複數個不同類型的感應器，藉以分別 對於工具機50的刀具51進行振動、聲音及電流變化的價 測，較佳地，如圖1所示該多重感應組係包含有一裝設於 工具機52夾具52上的加速規感應器11或聲射感應器12 或一設置於工具機50平台53周圍的任一位置上的麥克風 陣列感應器13及一與邊工具機50主軸54相連接的功率感 應器14 ; 201226101 該訊號處理組20係與該多重感應組1 〇相電性連接且 具有訊號處理運算能力及人機控制介面的功能，其中該訊 號處理組20係設有一資料擷取卡21及一電腦22，該資料 擷取卡21係分別與該多重感應組1〇的各感應器川，， 13, 14相電性連接，用以擷取各感應器w，12, 13, ^ 所偵測到的振動、聲音及電流訊號； 該電腦22係與該資料擷取卡21相電性連接且設有一 訊號轉換模組23 ' —特徵選取模組24、一辨識模組25及 一融合模組26,其中該訊號轉換模組23係與該資料擷取卡 21相連接，藉以將經該資料擷取卡22所擷取到的振動、聲 音及電流訊號進行轉換，較佳地，該訊號轉換模組23係可 藉由一小波轉換方程式231(Wavelet Transform)、一傅立 葉轉換方程式232(Fourier Transform)或一快速傅立葉轉 換方程式233(Fast Fourier Transform; FFT)進行訊號的轉 換； 該特徵選取模組24係與該訊號轉換模組23相連接， 用以將經该訊號轉換模組2 3處理的振動、聲音及電流訊號 進行特徵值的選取（該特徵值的選取數量可為彳個以上），該 辨識模組25係與該特徵選取模組24相連接，藉以透過各 選取的特徵值進行振動、聲音及電流訊號的辨識處理，進 而得到刀具51的狀態，較佳地，該辨識模組25係以一費 雪線性辨識函數 251 (Linear Discriminant Function)、一類 神經網路252(Neur〇 Network)或一模糊邏輯辨識方程式 253(Hidden MorKov Model)作為訊號辨識的方法； °亥融合模組26係設於該電腦22中與該訊號轉換模組[q 201226101 23 6亥特徵選取模組24或該辨識模組25相連接，藉以對 於轉換訊號、特徵訊號或辨識訊號進行融合，進而提供與 -51狀&'相4合之訊號’較佳地，如圖5所示該融合模 . 組26可為一與該訊號轉換模組23相連接的訊號融合組 261、-與該特徵選取模組24相連接的特徵融合組262或 一與該辨識模組25相連接的決策融合組263,較佳地，該 況5虎處理組20於該資料擷取+ 2 ]與該電腦22間係設有一 與該資料掘取卡21相電性連接的類比/數位轉換器27，藉 謇以將類比訊號轉換成一數位訊號；以及 該β控組3 0係與该訊號處理組2 〇相電性連接且設有 -警報器31，該警報器31係與該電腦22相連接，經由該 警報器31提醒操作者該刀具5彳的狀態（斷裂或磨耗）； Β、線上偵測與訊號處理：如圖彳所示將該多重感應組 1〇的各感應器11，12，13，14裝設於一工具機50的夾具 52、平台53周圍任一位置與工具機5〇的主軸54上藉以 _ 對於該工具機50刀具51進行加工訊號的偵測，將各感應 器11，12，13，14所取得的振動、聲音及電流訊號，經由 忒 > 訊擷取卡21擷取並經該類比/數位轉換器2 7的轉換 後，傳送至該電腦22中進行訊號的處理與呈現，其中在訊 號處理過程中，可如圖2所示將擷取的原始訊號透過小波 轉換方程式方程式231、傅立葉轉換方程式232或快速傅 立葉轉換方程式233進行訊號的轉換，於訊號轉換後經由 該訊號融合組261進行融合，其係將各感應器1彳，1 2，1 3， 1 4之原始訊號組成一組新的訊號，再進行特徵之擷取以及 辨識步驟，其中如X(t)、Υ⑴及z(t)的訊號係分別不同感應『 201226101 器 11 ’ 12，13，14 邮％ π 所取侍，則新的訊號0(t)係可由以下之 方程式獲得： ⑴Λ號之相加：外)=aX(t)+bY(t卜功）；或 (2)、訊號之組合：料χ(松(t)cz(t)]; 再·.呈A特镟選取模組24選取出與該刀具5 ]狀態相關 的訊號再、-垔由4辨識模組25進行辨識，藉以產生一與刀 具51狀態相關的判斷訊號；The judgment signal related to the tool status. Preferably, in the operating step of the instrument setting, the feature fusion group is connected to the feature selection module. Preferably, in the operation step of the instrument setting, the fusion module is connected to the identification module. The decision fusion group, and in the operation step of the online situation measurement and signal processing, the signal converted by the signal conversion module is selected by the feature selection module to select a signal related to the tool state, and the identification mode is The group performs the identification of the characteristic signal, and finally the fusion mode of the decision fusion group generates a judgment signal β related to the state of the tool. In the operation step of the instrument setting, the signal processing group draws the material. An analog/digital converter electrically connected to the data capture card is disposed between the card and the computer to convert the analog signal into a digital signal. Preferably, in the operation step of detecting and signal processing on the line, the burst conversion module performs signal conversion by a wavelet conversion equation. In the operation step of online detection and signal processing, the signal conversion module performs signal conversion by a Fourier transform equation. Preferably, in the operation step of on-line detection and signal processing, the signal conversion module performs signal switching by a fast Fourier transform equation. Preferably, in the operation step of online detection and signal processing, the identification module uses a Fisher linear identification function as a method for signal identification. In the operation step of online detection and signal processing, the identification module uses a type of neural network as a method of signal identification. Preferably, in the operation step of online detection and signal processing, the identification module uses a fuzzy logic identification equation as a method for signal identification. The present invention further provides a multi-induction tool machine tool state detection device, which comprises a multi-sensor group, a signal processing group and a monitoring group, wherein: the multi-sensor group is provided with a plurality of tools. a sensor for detecting vibration, 3-tone and current change of the machine; the signal processing group is electrically connected to the multi-sensor group and has a data capture card and a computer, the data capture card and each sensor The computer is electrically connected to the device and the computer is electrically connected to the data card and is provided with a signal car replacement module, a feature selection module and an identification module, and the signal conversion module is connected to the data. The card is connected, the feature selection module is connected to the signal conversion right group, the identification module is connected to the feature selection module, and the monitoring group is electrically connected to the signal processing group and is provided with Soldiers connected to the brain. The Progress ° Haixi Heavy Sensing Group is equipped with an accelerometer sensor, an infrared sensor, a microphone array sensor and a power sensor. Further, the computer system is provided with a fusion module connected to the signal conversion module 'the feature selection module or the identification module, so as to fuse the conversion signal, the characteristic signal or the identification signal, thereby providing The signal that the tool 201226101 is in compliance with. The driver's module is a signal fusion group connected to the signal conversion module. Preferably, the fusion module is a feature fusion group connected to the feature selection module. Preferably, the fusion module is a decision fusion group connected to the identification module. Preferably, in the operation step of the instrument setting, the signal processing group is provided with an analog/digital converter electrically connected to the data acquisition card between the data acquisition card and the computer, so as to analog signal Convert to a digital signal. According to the above technical means, the multi-inductive tool machine tool condition sensing method and device thereof are mainly used for integrating the vibration, sound and current generated by the tool processing by integrating sensors with different inductive characteristics. The detection of the signal can not only reduce the interference level of the noise through the way of matching and matching, but also integrate the conversion signal, the characteristic signal or the identification signal through the signal integration mode of the fusion module to provide a call-reliable High-level judgment signal, which greatly improves the monitoring accuracy of the on-line tool status (fracture or wear), and alerts the operator with an alarm or connects to the controller of the machine tool to stop the machine immediately. Or implementing the replacement of the tool; therefore, the multi-induction tool machine tool state detecting device of the present invention is installed in the working device, and only needs to be used on the platform of each sensing device of the multiple sensing group. The sound detection of the processed tool is not only easy to install, but also can be used under different requirements. The tool machine 201226101, which is used for milling, drilling and tapping machine tools of different axes, is used to increase the utilization rate of the system, thereby saving the investment cost and reducing the number of operators at the same time. It greatly improves the correctness of online tool breakage or wear state monitoring based on vibration, sound and current signals, and provides a convenient method for detecting, detecting and improving detection reliability. [Embodiment] In order to understand the technical features and practical effects of the present invention in detail, and in accordance with the contents of the specification, the detailed description of the preferred embodiment shown in the drawings (shown in Figures i to 5) will be described in detail. For example, the object of the present invention is to provide a multi-induction tool tool tool state detecting method and device thereof, which are mounted on a machine tool 5 , and detected by means of re-sensing, instantaneously and Accurately know the state of the tool 5 ,, and then provide a convenient method for detecting, detecting and improving the detection reliability and detection. The multi-induction tool machine tool state debt testing method of the present invention comprises the following steps: A 'instrument setting: preparing a detecting device, the detecting device is provided with an evening sensing group 10, a sfl number processing The group 20 and the monitoring group 30, wherein the 1 multi-sensor group 10 is provided with a plurality of different types of sensors, so that the vibration, sound and current changes of the tool 51 of the machine tool 50 are respectively measured. The multi-sensor group includes an accelerometer sensor 11 or an acoustic sensor 12 mounted on the fixture 52 of the machine tool 52 or a position disposed around the platform 53 of the machine tool 50. The microphone array sensor 13 and a power sensor 14 connected to the spindle 54 of the side tool machine 50; 201226101 The signal processing group 20 is electrically connected to the multi-sensor group 1 and has signal processing capability and man-machine The function of the control interface, wherein the signal processing group 20 is provided with a data capture card 21 and a computer 22, and the data capture card 21 is respectively associated with the sensors of the multi-sensor group 1 , Chuan, 13, 14 Electrical connection The vibration, sound and current signals detected by the sensors w, 12, 13, ^ are captured; the computer 22 is electrically connected to the data capture card 21 and is provided with a signal conversion module 23'. The module 24, an identification module 25 and a fusion module 26 are selected, wherein the signal conversion module 23 is connected to the data capture card 21, so as to capture the vibration captured by the data capture card 22. The sound and current signals are converted. Preferably, the signal conversion module 23 can be implemented by a wavelet transform equation 231 (Wavelet Transform), a Fourier transform equation 232 (Fourier Transform) or a fast Fourier transform equation 233 (Fast Fourier Transform; FFT) performs signal conversion; the feature selection module 24 is connected to the signal conversion module 23 for performing eigenvalues on the vibration, sound and current signals processed by the signal conversion module 23. The selection module 25 is connected to the feature selection module 24 to perform vibration, sound and current signal identification processing through the selected feature values. The state of the tool 51 is obtained. Preferably, the identification module 25 is identified by a Linear Discriminant Function 251, a neural network 252 (Neur〇 Network), or a fuzzy logic identification equation 253 ( The Hidden MorKov Model is used as a signal identification method. The °H fusion module 26 is connected to the computer 22 and connected to the signal conversion module [q 201226101 23 6 Hai feature selection module 24 or the identification module 25, thereby For the conversion of the conversion signal, the characteristic signal or the identification signal, and then providing the signal with the -51 shape & '4', preferably, the fusion mode is shown in FIG. 5. The group 26 can be a conversion mode with the signal. The group 23 is connected to the signal fusion group 261, the feature fusion group 262 connected to the feature selection module 24, or a decision fusion group 263 connected to the identification module 25, preferably, the situation is 5 The group 20 is provided with an analog/digital converter 27 electrically connected to the data capture card 21 to the computer 22 to convert the analog signal into a digital signal; Beta control group 30 system and the signal processing 2 〇 phase is electrically connected and provided with an alarm 31, which is connected to the computer 22, via which the operator 31 is alerted to the state of the tool 5 (break or wear); Measurement and signal processing: As shown in FIG. 各, the sensors 11, 12, 13, 14 of the multi-sensor group 1 are mounted on the fixture 52 of a machine tool 50, at any position around the platform 53, and the machine tool 5 The main shaft 54 is _ for detecting the processing signal of the machine tool 50, and the vibration, sound and current signals obtained by the sensors 11, 12, 13, 14 are transmitted via the 忒> After being converted by the analog/digital converter 27, the signal is sent to the computer 22 for processing and presentation of the signal. During the signal processing, the captured original signal can be transmitted through the wavelet as shown in FIG. Converting the equation 231, the Fourier transform equation 232, or the fast Fourier transform equation 233 to convert the signal, and after the signal conversion, the fusion is performed via the signal fusion group 261, which is to connect the inductors 1彳, 1 2, 1 3, 1 4 The original signals form a group The signal, and then the feature extraction and identification steps, wherein the signals of X(t), Υ(1), and z(t) are differently sensed by "201226101 11' 12, 13, 14 mail % π, then The new signal 0(t) can be obtained by the following equation: (1) Addition of apostrophes: outer) = aX(t) + bY (t gong); or (2), combination of signals: χ (song (t Cz(t)]; again. The A feature selection module 24 selects the signal related to the state of the tool 5], and then identifies it by the 4 identification module 25, thereby generating a state related to the state of the tool 51. Judgment signal
另可如圖3所不’將擷取的原始訊號透過小波轉換方 私式方粒式231、傅立葉轉換方程式232或快速傅立葉轉 換方私式233進行訊號轉換後，經該特徵選取模組選取 出Ί刀具51狀$相關的訊號，並將所選取的特徵訊號透 過該特徵融合'组262的融合方式整合成一特徵訊號序列， f係將各感應器11 ’ 12，13 ’ 14之原始訊號經轉換後所取 ^特徵組成—組新的特徵’再進行特徵之辨識，其 中如Χ(η) γ(η)及z(n)特徵分別為由不同感應器11，π 所取彳于，而新的讯號特徵知)係可由以下之方程式獲 付· 12 201226101 訊號的辨識’最後由該決策融合組263進行融合的方式， 其係將各感應器11，1 2 ’ 1 3 ’ 14之原始訊號經轉換並辨識 後所取得之決策係數組成一組新的決策係數，以便決定最 後的狀態，其中如Χ(λ )、Υ(λ )及Ζ(λ )的決策係數分別由 不同感應器11 ’ 1 2，1 3 ’ 14所取得，而新的決策係數^⑷係 可由以下之方程式獲得： ^⑴=aX(A)+bY(A)+cz ⑴ 再依據係數之水準決定最後之刀具狀態，進而產生一 # 與刀具51狀態相關的判斷訊號；以及 C Μ貞測結果：當該電腦22將各感應器11，12，1 3， 14所擷取的訊號進行處理後’產生一與刀具51狀態相關的 判斷訊號’將該判斷訊號輸出至該監控組3〇中，即可讓使 用者即時得知該刀具51的狀態（磨耗），且可透過傳送訊號 至該警報器31的方式’告知操作者該刀具51的狀態（斷裂卜 藉由上述之技術手段’本發明多重感應之工具機刀具 狀態偵測方法及其裝置，主要係透過整合不同感應特性的 • 感應器11 ’12，13，14方式，藉以對於刀具51加工所產 生的振動、聲音及電流進行訊號的偵測，不僅可透過交又 比對的方式降低雜訊的干擾程度，且可透過該電腦22融合 模組26的訊號整合方式，對於轉換訊號、特徵訊號或辨識 訊號進行融合’提供一可靠度高的判斷訊號，進而大幅提 升線上刀具51狀態（斷裂或磨耗）的監測準確性，並以警報 器31警不操作者’或者透過與該工具機5〇的控制器相連 接的方式，對於該工具機5〇即時停機或實施刀具52的更 換； 13 201226101 因此’本發明之多重感應之工具機刀具狀態偵測裝 置，於使用時僅需將該多重感應組10的各感應器彳1，12， 14’ 14裝設於該工具機5〇的夾具52、平台53及主軸54 -上，即可分別對於加工的刀具51進行聲音的偵測，不僅安 裝上相當方便，且可在不同之需求下，任意地應用於不同 軸數之銑削、鑽削與攻牙工具機…等等的工具機上，藉以提 升系統使用之稼動率，而達到節省投資成本之目的，且可 同時減少操作人員的數量，並大大提升以振動、聲音及電 ^ 流訊號為基礎之線上刀具51斷裂或磨耗狀態監測之正確 性，提供一方便安裝、即時檢測及提高偵測可靠度之偵測 方法及裝置者。 以上所述’僅是本發明的較佳實施例，並非對本發明 作任何形式上的限制，任何所屬技術領域中具有通常知識 者，右在不脫離本發明所提技術方案的範圍内利用本發 明所揭不技術内容所作出局部更動或修飾的等效實施例， 並且未脫離本發明的技術方案内容，均仍屬於本發明技術 方案的範圍内。 【圖式簡單說明】 圖1係本發明多重感應之工具機刀具狀態偵測裝置設 置於一工具機上之立體外觀示意圖。 。圖2係本發明多重感應之工具機刀具狀態偵測方法之 操作流程方塊圖。 圖3係本發明多重感應之工具機刀具狀態積測方法之 另—操作流程方塊圖。 圖4係本發明多重感應之工具機刀具狀態偵測方法之f 201226101 再一操作流程方塊圖。 圖5係本發明融合模組之操作示意圖 【主要元件符號說明】 1 2聲射感應器 14功率感應器 1 〇多重感應組 11加速規感應器 1 3麥克風陣列感應器 20訊號處理組 21資料擷取卡 22電腦 23訊號轉換模組 231小波轉換方程式 232傅立葉轉換方程式233快速傅立葉轉換方程 24特徵選取模組 25辨識模組 252類神經網路 26融合模組 262特徵融合組 27類比/數位轉換器 30監控組 5〇工具機 51 刀具 53平台Alternatively, as shown in FIG. 3, the original signal captured by the wavelet transforming square 231, the Fourier transform equation 232, or the fast Fourier transform square 233 is used for signal conversion, and then selected by the feature selection module. The tool 51 has a related signal, and the selected feature signal is integrated into a characteristic signal sequence through the fusion of the feature fusion group 262, and f converts the original signals of the sensors 11 ' 12, 13 ' 14 After the feature is taken—the group of new features' is identified, and the features such as Χ(η) γ(η) and z(n) are respectively taken by different inductors 11, π, and new The signal characteristic can be obtained by the following equation. 12 201226101 Identification of the signal 'The final fusion method by the decision fusion group 263, which is the original signal of each sensor 11, 1 2 ' 1 3 ' 14 The decision coefficients obtained after conversion and identification constitute a new set of decision coefficients to determine the final state, where the decision coefficients such as Χ(λ), Υ(λ) and Ζ(λ) are respectively determined by different inductors 11 ' 1 2,1 3 '14 obtained, The new decision coefficient ^(4) can be obtained by the following equation: ^(1)=aX(A)+bY(A)+cz (1) Determine the final tool state according to the level of the coefficient, and then generate a # related to the state of the tool 51. Judging signal; and C guessing result: when the computer 22 processes the signals captured by the sensors 11, 12, 13, 3, and then 'generates a judgment signal related to the state of the tool 51' The output to the monitoring group 3〇 allows the user to instantly know the state (wearing) of the tool 51, and can inform the operator of the state of the tool 51 by transmitting a signal to the alarm device 31 (breaking) According to the above technical means, the multi-inductive tool tool tool state detecting method and device thereof are mainly produced by processing the tool 51 by integrating the sensors 11 '12, 13, 14 of different inductive characteristics. The vibration, sound and current signal detection can not only reduce the interference level of the noise through the matching and matching, but also through the signal integration mode of the computer 22 fusion module 26, for the conversion signal The feature signal or the identification signal is fused to provide a highly reliable judgment signal, thereby greatly improving the monitoring accuracy of the on-line tool 51 state (fracture or wear), and the alarm 31 is not used by the operator' or through the machine tool 5〇 of the controller is connected, for the machine tool 5〇 immediately stop or implement the replacement of the tool 52; 13 201226101 Therefore, the multi-induction tool machine tool state detecting device of the invention only needs to be used The sensors 彳1, 12, 14' 14 of the multi-sensor group 10 are mounted on the clamp 52, the platform 53 and the spindle 54- of the machine tool 5, so that the processed tool 51 can be detected separately. It is not only easy to install, but also can be applied to the machine tools of different milling, drilling and tapping machine tools, etc. under different requirements, so as to improve the utilization rate of the system and save. The purpose of investment cost, and can reduce the number of operators at the same time, and greatly improve the online tool 51 fracture or wear state monitoring based on vibration, sound and electric signal. Right, providing an easy installation, testing and real time detection method and apparatus to improve the reliability of detection by. The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any one of ordinary skill in the art will be able to utilize the present invention without departing from the scope of the present invention. The equivalent embodiments of the present invention are not limited to the details of the technical solutions of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a perspective view showing the three-dimensional appearance of a multi-induction tool machine tool state detecting device of the present invention placed on a machine tool. . Fig. 2 is a block diagram showing the operation flow of the multi-induction tool machine tool state detecting method of the present invention. Fig. 3 is a block diagram showing another operation flow of the multi-induction tool machine tool state integration method of the present invention. 4 is a block diagram of another operation flow of the multi-induction tool machine tool state detecting method of the present invention. FIG. 5 is a schematic diagram of the operation of the fusion module of the present invention. [Main component symbol description] 1 2 sound sensor 14 power sensor 1 〇 multiple sensor group 11 acceleration gauge sensor 1 3 microphone array sensor 20 signal processing group 21 data Capture Card 22 Computer 23 Signal Conversion Module 231 Wavelet Conversion Equation 232 Fourier Transform Equation 233 Fast Fourier Transform Equation 24 Feature Selection Module 25 Identification Module 252 Neural Network 26 Fusion Module 262 Feature Fusion Group 27 Analog/Digital Conversion 30 monitoring group 5 〇 machine tool 51 tool 53 platform
251費雪線性辨識函數 253模糊邏輯辨識方程式 2 61 號融合組 263決策融合組 31警報器 52夾具 54主軸251 Fisher's linear identification function 253 fuzzy logic identification equation 2 61 fusion group 263 decision fusion group 31 alarm 52 fixture 54 spindle