TWI605322B - Intelligent machine condition monitoring and diagnostic method - Google Patents
Intelligent machine condition monitoring and diagnostic method Download PDFInfo
- Publication number
- TWI605322B TWI605322B TW105105342A TW105105342A TWI605322B TW I605322 B TWI605322 B TW I605322B TW 105105342 A TW105105342 A TW 105105342A TW 105105342 A TW105105342 A TW 105105342A TW I605322 B TWI605322 B TW I605322B
- Authority
- TW
- Taiwan
- Prior art keywords
- signal
- value
- vibration
- speed
- reference line
- Prior art date
Links
Landscapes
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Description
本發明係關於一種設備監診方法，特別是關於一種智慧設備監診方法。 The present invention relates to a method for monitoring a device, and more particularly to a method for monitoring a smart device.
設備監診技術是指利用儀器對於運轉中的設備進行整體或部分監測，以取得代表設備運轉狀況之訊號，並對訊號作分析以判斷設備運轉是否異常，以及運轉異常的種類，並進行維修方法的判斷。設備監診技術可對設備零件之老化、損壞、歪斜情形有精確的把握，並且，在此前提下將設備的保養從定期維修改為預防維修，一方面減少對於設備零件狀況不了解就進行拆卸所可能帶來的對設備的損傷，一方面減少設備停機所帶來的成本損失，進一步因提升了設備管理以及維護之效率而提升產業競爭力。 Equipment supervision technology refers to the use of the instrument to monitor the running equipment in whole or in part to obtain the signal representing the operating condition of the equipment, and to analyze the signal to determine whether the equipment is operating abnormally, and the type of abnormal operation, and to carry out maintenance methods. Judgment. The equipment supervision technology can accurately grasp the aging, damage and skew of the equipment parts, and under this premise, the maintenance of the equipment is changed from regular maintenance to preventive maintenance. On the one hand, the disassembly of the equipment parts is not understood. The damage to the equipment that may be brought about, on the one hand, reduces the cost loss caused by equipment downtime, and further enhances the industrial competitiveness by improving the efficiency of equipment management and maintenance.
習知的【0001】設備監診技術存在以下問題。習知的設備監診技術須在設備為固定轉速之運轉模式之設備，始能有效對於從設備擷取之訊號進行分析，故可應用之設備有限。並且，習知設備監診技術使用離散小波轉換將擷取之訊號進行分頻，而使用離散小波轉換無法將訊號分頻成任意頻寬範圍之訊號，如此將降低對於設備細部零件狀況的掌握。另外，習知的設備監診技術使用均方根值、波高率作為擷取之訊號的特徵指標，其中均方根值代表設備零件的振動程度，陡峭值代表零件之間的敲擊程度，然而設備運轉之異常非全靠均 方根植及波高率所能表現，故對於均方根植及波高率所無法表現之異常狀況，習知的設備監診技術無法偵測該異常而無法達到對設備損壞之預防。 The conventional [0001] equipment supervision technology has the following problems. The conventional equipment monitoring technology must be able to effectively analyze the signals extracted from the equipment when the equipment is in a fixed-speed operation mode, so the applicable equipment is limited. Moreover, conventional device monitoring techniques use discrete wavelet transform to divide the captured signal, and discrete wavelet transform cannot divide the signal into signals of any bandwidth, which will reduce the mastery of the detailed parts of the device. In addition, the conventional device monitoring technique uses the root mean square value and the wave height ratio as characteristic indicators of the captured signal, wherein the root mean square value represents the vibration degree of the device parts, and the steep value represents the degree of tapping between the parts, however The abnormal operation of the equipment is not all The square roots and the wave height rate can be expressed. Therefore, the conventional equipment monitoring technology cannot detect the abnormality and cannot prevent the damage of the equipment for the abnormal condition that the root mean rooting and the wave height rate cannot be expressed.
鑒於上述，習知的設備監診方法只能被應用於定轉速設備，且使用的分頻方式未能精確掌握所有設備零件的狀況，此外，習知的設備監診方法表現設備運轉之異常之特徵指標並不夠充分，設備監診對於設備異常的辨識能力較低。 In view of the above, the conventional device monitoring method can only be applied to a fixed-speed device, and the frequency division method used does not accurately grasp the condition of all the device parts. Moreover, the conventional device monitoring method shows that the device operates abnormally. The characteristic indicators are not sufficient, and the ability of equipment supervision to identify equipment anomalies is low.
緣此，本發明之一目的即在提供一種智慧設備監診方法，其能夠不受設備的轉速類型限制，且提高設備監診的精確性與對於設備異常的辨識能力。 Accordingly, it is an object of the present invention to provide a smart device monitoring method that is not limited by the type of rotational speed of the device, and that improves the accuracy of device monitoring and the ability to identify abnormalities of the device.
本創作為解決習知技術之問題所採用之技術手段係提供一種智慧設備監診方法，用以對一軸轉設備之運轉狀況進行檢測，該智慧設備檢測方法包含下列步驟：(a)在該軸轉設備於一軸轉狀態下時偵測該軸轉設備而得出一運轉動態訊號，該運轉動態訊號包括一振動訊號且選擇性地包括一軸轉速訊號；(b)對於該運轉動態訊號予以分頻為預定的複數個頻寬範圍而計算出每個該頻寬範圍所個別對應的複數個特徵值，其中該複數個特徵值分別為一均方根值、一陡峭值及一調變值；以及(c)就每個特徵值而將每個該頻寬範圍所個別對應的複數個特徵值分別與對應於每個該頻寬範圍的複數個特徵值的個別基準線比對，而判斷該設備之運轉狀況，其中，當該設備為定轉速時，該個別基準線係由對於該軸轉設備於一正常軸轉狀態下所預先取得的一正常運轉振動訊號的每個該頻寬範圍所個別對應的複數個基準特徵值利用統計分析之信賴區間定義而計算所得，以及當該設備為變轉速時，該個別基準線係利用迴歸分析建立該軸轉設備 於正常軸轉狀態下所預先取得的一正常運轉軸轉速訊號與該正常運轉振動訊號的每個該頻寬範圍所個別對應的複數個基準特徵值的回歸模型，再利用統計分析之信賴區間定義而計算所得。 The technical means adopted by the present invention for solving the problems of the prior art provides a smart device monitoring method for detecting the running condition of the one-axis rotating device, and the smart device detecting method comprises the following steps: (a) in the axis When the rotating device detects the shaft rotating device in the one-axis rotating state, an operating dynamic signal is obtained, the running dynamic signal includes a vibration signal and optionally includes a shaft speed signal; (b) the operating dynamic signal is divided. Calculating, for a predetermined plurality of bandwidth ranges, a plurality of eigenvalues corresponding to each of the bandwidth ranges, wherein the plurality of eigenvalues are respectively a root mean square value, a steep value, and a modulation value; (c) judging the device by comparing each of the plurality of feature values corresponding to each of the bandwidth ranges to an individual reference line corresponding to a plurality of feature values of each of the bandwidth ranges for each feature value The operating condition, wherein, when the device is at a fixed speed, the individual reference line is each of a normal operating vibration signal pre-acquired in the normal axis rotation state of the shaft rotating device Wherein the plurality of reference bandwidth range corresponding to the individual values calculated using the true statistical analysis of the obtained interval is defined, and when the apparatus is a variable speed, the individual reference line is the regression analysis using the apparatus rotation axis a regression model of a plurality of reference eigenvalues corresponding to each of the bandwidth ranges of the normal running shaft speed signal and the normal running vibration signal obtained in the normal axis rotation state, and then using the confidence interval definition of the statistical analysis And calculate the income.
在本發明的一實施例中係提供一種智慧設備監診方法，其中步驟(a)中，該振動訊號係為一振動加速度訊號。 In an embodiment of the invention, a smart device monitoring method is provided, wherein in step (a), the vibration signal is a vibration acceleration signal.
在本發明的一實施例中係提供一種智慧設備監診方法，其中步驟(b)中，該不同類別的特徵值的計算包括下列步驟：(b1)利用積分方法將該振動加速度訊號積分為一振動速度訊號；(b2)對於該振動加速度訊號及該振動速度訊號進行諧波小波濾波，使該振動加速度訊號及該振動速度訊號根據該複數個頻寬範圍而被分頻；以及(b3)計算出該振動加速度訊號及該振動速度訊號對應於每個該頻寬範圍的速度均方根值、速度陡峭值、速度調變值、加速度均方根值、加速度陡峭值及加速度調變值。 In an embodiment of the present invention, a smart device monitoring method is provided, wherein in step (b), the calculation of the different category of feature values includes the following steps: (b1) integrating the vibration acceleration signal into one by using an integration method. a vibration velocity signal; (b2) performing harmonic wavelet filtering on the vibration acceleration signal and the vibration velocity signal, so that the vibration acceleration signal and the vibration velocity signal are divided according to the plurality of bandwidth ranges; and (b3) calculating The vibration acceleration signal and the vibration speed signal correspond to a speed root mean square value, a speed steepness value, a speed modulation value, an acceleration root mean square value, an acceleration steepness value, and an acceleration modulation value for each of the bandwidth ranges.
在本發明的一實施例中係提供一種智慧設備監診方法，其中步驟(c)中，該個別基準線係根據一標準差方法而從該複數個基準特徵值所得到，該標準差方法係包括下列步驟：計算各個基準特徵值分佈之標準差及平均值；以及以每個基準特徵值之統計樣本之平均值為一圓心座標，以該種基準特徵值之統計樣本之六個標準差為二弦，以該二弦之一斜邊為半徑畫圓形成該基準線。 In an embodiment of the present invention, a smart device monitoring method is provided, wherein in step (c), the individual reference line is obtained from the plurality of reference feature values according to a standard deviation method, the standard deviation method system The method comprises the following steps: calculating a standard deviation and an average value of each of the reference feature value distributions; and calculating an average value of the statistical samples of each of the reference feature values as a center coordinate, wherein the six standard deviations of the statistical samples of the reference feature values are A two-string, the circle is formed by drawing a circle with a hypotenuse of one of the two strings.
在本發明的一實施例中係提供一種智慧設備監診方法，其中步驟(c)中，該特徵值與個別基準線之比對係判斷該複數個特徵值相對於該個別基準線的位置，當該複數特徵值相對該圓心座標的位置在該個別基準線範圍之內，判斷該設備之運轉狀況為正常，當該複數個特徵值相對該圓心座標的位置在該基準線範圍之外，判斷該設備之運轉狀況為異常。 In an embodiment of the present invention, a smart device monitoring method is provided, wherein in step (c), the ratio of the feature value to an individual reference line determines a position of the plurality of feature values relative to the individual reference line, When the position of the complex feature value relative to the center of the circle is within the range of the individual reference line, determining that the operating condition of the device is normal, and determining that the position of the plurality of feature values relative to the center of the circle is outside the range of the reference line The operating condition of the device is abnormal.
在本發明的一實施例中係提供一種智慧設備監診方法，其中在步驟(c)之後更包括一步驟(d)：對於該設備之運轉狀況為異常下的設備之異常模式建立對應的故障編碼，而記憶該設備之異常模式，以及步驟(c)中，更包括根據該故障編碼判斷該設備之異常模式。 In an embodiment of the present invention, a smart device monitoring method is provided, wherein after step (c), a step (d) is further included: establishing a corresponding fault for an abnormal mode of the device whose operating condition is abnormal. Encoding, while remembering the abnormal mode of the device, and in step (c), further comprising determining an abnormal mode of the device according to the fault code.
在本發明的一實施例中係提供一種智慧設備監診方法，其中在步驟(d)中係利用專家系統記憶該設備之異常模式，以及步驟(c)中係利用該專家系統判斷該設備之異常模式。 In an embodiment of the present invention, a smart device monitoring method is provided, wherein in step (d), an expert system is used to memorize an abnormal mode of the device, and in step (c), the expert system is used to determine the device. Abnormal mode.
經由本發明的智慧設備監診方法所採用之技術手段，本發明的智慧設備監診方法能應用在定轉速設備及變轉速設備，不受設備的轉速類型，故能夠廣泛應用於各類型設備。此外，在對於擷取的訊號進行分析時，能任意設定擷取訊號之分頻頻寬，提高設備監診的精確性。最後，本發明提供的設備監診方法增加了能夠顯示設備零件間交互關係的特徵值，提高了對於設備異常的辨識能力。 According to the technical means adopted by the smart device monitoring method of the present invention, the smart device monitoring method of the present invention can be applied to a fixed speed device and a variable speed device, and is not applicable to various types of devices, regardless of the type of rotation speed of the device. In addition, when analyzing the captured signals, the frequency division bandwidth of the captured signals can be arbitrarily set to improve the accuracy of the equipment supervision. Finally, the device monitoring method provided by the present invention increases the feature value capable of displaying the interaction relationship between the device parts, and improves the recognition ability for the device abnormality.
S110、S12、S121、S122、S123、S130、S140、S141、S142、S140、S210、S220、S221、S222、S223、S230、S240、S241、S242‧‧‧步驟 Steps S110, S12, S121, S122, S123, S130, S140, S141, S142, S140, S210, S220, S221, S222, S223, S230, S240, S241, S242‧‧
〔第1圖〕為顯示根據本發明的第一實施例的智慧設備監診方法應用於一定轉速設備的流程圖；〔第2圖〕為顯示根據本發明的第一實施例的智慧設備監診方法應用於一變轉速設備的流程圖；〔第3圖〕為顯示根據本發明的實施例的智慧設備監診方法對於一運轉狀態為正常的設備進行監診所得到的柱狀圖； 〔第4圖〕為顯示根據本發明的實施例的智慧設備監診方法對於一運轉狀態為異常的設備進行監診所得到的柱狀圖；〔第5圖〕為顯示根據本發明的實施例的智慧設備監診方法對於一運轉狀態為異常的另一設備進行監診所得到的柱狀圖；〔第6圖〕為顯示根據本發明的實施例的智慧設備監診方法對於一變轉速設備進行監診時建立轉速與振動訊號之回歸模型圖。 [Fig. 1] is a flowchart showing the application of the smart device monitoring method to a certain rotational speed device according to the first embodiment of the present invention; [Fig. 2] is a view showing the smart device monitoring according to the first embodiment of the present invention. The method is applied to a flow chart of a variable speed device; [Fig. 3] is a histogram showing a smart device monitoring method according to an embodiment of the present invention for a device in which the operating state is normal; [Fig. 4] is a bar graph showing a smart device monitoring method according to an embodiment of the present invention for a device having an abnormal operating state; [Fig. 5] is a view showing an embodiment according to the present invention. The smart device monitoring method performs a histogram obtained by the monitoring clinic for another device having an abnormal operating state; [Fig. 6] is a view showing a smart device monitoring method according to an embodiment of the present invention for monitoring a variable speed device A regression model of the rotational speed and vibration signal is established at the time of diagnosis.
以下根據第1圖至第6圖，而說明本發明的實施方式。該說明並非為限制本發明的實施方式，而為本發明之實施例的一種。 Hereinafter, embodiments of the present invention will be described with reference to Figs. 1 to 6 . This description is not intended to limit the embodiments of the invention, but is an embodiment of the invention.
第1圖係顯示根據本發明之一實施例的智慧設備監診方法應用於一定轉速設備之流程圖，包含下列步驟：在該軸轉設備於一軸轉狀態下時偵測該軸轉設備而得出一運轉動態訊號，該運轉動態訊號包括一振動訊號(步驟S110)；對於該運轉動態訊號予以分頻為預定的複數個頻寬範圍而計算出每個該頻寬範圍所個別對應的複數個特徵值，其中該複數個特徵值分別為一均方根值、一陡峭值及一調變值(步驟S120)；就每個特徵值而將每個該頻寬範圍所個別對應的複數個特徵值分別與對應於每個該頻寬範圍的複數個特徵值的個別基準線比對(步驟S130)；對於比對結果進行故障編碼並記憶該異常模式(步驟S140)。上述各步驟的詳細說明如下。 1 is a flow chart showing a smart device monitoring method applied to a certain speed device according to an embodiment of the present invention, which includes the following steps: detecting the axis rotating device when the axis rotating device is in a pivot state An operation dynamic signal is generated, the operation dynamic signal includes a vibration signal (step S110), and the operation dynamic signal is divided into a predetermined plurality of bandwidth ranges to calculate a plurality of individual corresponding to each of the bandwidth ranges. An eigenvalue, wherein the plurality of eigenvalues are respectively a root mean square value, a steep value, and a modulation value (step S120); and each of the plurality of features corresponding to the bandwidth range for each eigenvalue The values are respectively aligned with individual reference lines corresponding to a plurality of feature values of each of the bandwidth ranges (step S130); the comparison result is fault coded and the abnormal mode is memorized (step S140). The detailed description of each of the above steps is as follows.
詳細而言，步驟S110中的該振動訊號為一振動加速度訊號。步驟S120具體包含下列步驟：利用積分方法將該振動加速度訊號積分為一振動速度訊號(步驟S121)；對於該振動加速度訊號及該振動速度訊號進行諧波小波濾波，使該振動加速度訊號及該振動速度訊號根據該複數個頻寬範圍而被分頻(步驟 S122)；以及計算出該振動加速度訊號及該振動速度訊號對應於每個該頻寬範圍的速度均方根值、速度陡峭值、速度調變值、加速度均方根值、加速度陡峭值及加速度調變值(步驟S123)。 In detail, the vibration signal in step S110 is a vibration acceleration signal. Step S120 specifically includes the following steps: integrating the vibration acceleration signal into a vibration speed signal by using an integration method (step S121); performing harmonic wavelet filtering on the vibration acceleration signal and the vibration speed signal to make the vibration acceleration signal and the vibration The speed signal is divided according to the plurality of bandwidth ranges (step S122); and calculating the vibration acceleration signal and the vibration velocity signal corresponding to a speed root mean square value, a speed steepness value, a speed modulation value, an acceleration root mean square value, an acceleration steepness value, and an acceleration of each of the bandwidth ranges The modulation value is obtained (step S123).
當設備出現異常時，其徵兆出現順序為：振動、聲音、熱、冒煙等，因此本發明的智慧監診方法選擇擷取振動訊號來分析設備運轉狀態，以在設備出現異常的初始階段即能測得異常而進行及時維修，避免設備進一步損壞。振動訊號加速度訊號、速度訊號以及位移訊號呈現，其中加速度迅速主要呈現高頻振動的設備異常，例如：軸承異常，但無法呈現對中頻振動的設備異常，例如：齒輪齧合差，以及低頻振動的設備異常，例如：軸偏心。而速度訊號則主要呈現中頻振動的設備異常，位移訊號，主要呈現低頻振動的設備異常。本發明使用的振動感測器偵測加速度訊號，再將加速度訊號積分成速度訊號，能全面檢測各種故障現象。當然，本發明並不以此實施例為限，在其它實施例中，該振動訊號可為振動速度訊號，而步驟S121中係以微分方法將該速度訊號維分成加速度訊號。 When the device is abnormal, the symptoms appear in the order of vibration, sound, heat, smoke, etc. Therefore, the smart monitoring method of the present invention selects the vibration signal to analyze the operating state of the device, so that the initial stage of the abnormality of the device It can be detected abnormally and repaired in time to avoid further damage to the equipment. The vibration signal acceleration signal, the speed signal and the displacement signal are presented, wherein the acceleration rapidly presents the abnormality of the device with high frequency vibration, for example, the bearing is abnormal, but the device abnormality for the intermediate frequency vibration cannot be presented, for example, the gear mesh difference, and the low frequency vibration The device is abnormal, for example: axis eccentricity. The speed signal mainly presents the abnormality of the medium frequency vibration equipment, and the displacement signal mainly presents the equipment abnormality of the low frequency vibration. The vibration sensor used in the invention detects the acceleration signal, and then integrates the acceleration signal into a speed signal, which can comprehensively detect various fault phenomena. Of course, the present invention is not limited to this embodiment. In other embodiments, the vibration signal may be a vibration speed signal, and in step S121, the speed signal is divided into acceleration signals by a differential method.
設備轉動時其零件損壞所呈現的振動訊號分三類型：第一類型為弦波及寬頻白噪音，其中弦波可表示例如齒輪偏心、齧合不良等徵兆，而寬頻白噪音則可表現零件間潤滑不良；第二類型為衝擊波，可顯示傳動軸系存在間隙、齒輪斷齒等設備異常訊息；第三類型為調變波，可表示齒輪偏斜、軸承損壞等現象。本發明分別使用均方根值來檢視弦波、使用陡峭值檢視衝擊波、使用調變值檢視調變波，可分別檢測設備是否出現上述各類型的故障現象。 The vibration signals exhibited by the parts when the equipment is rotated are divided into three types: the first type is sine wave and wide-band white noise, wherein the sine wave can indicate signs such as gear eccentricity and poor meshing, while wide-band white noise can express lubrication between parts. Bad; the second type is shock wave, which can display equipment abnormalities such as gaps in the drive shaft and gear breakage; the third type is modulated wave, which can indicate gear deflection and bearing damage. The present invention uses the rms value to examine the sine wave, uses the steep value to view the shock wave, and uses the modulation value to view the modulated wave, and can detect whether the above-mentioned various types of fault phenomena occur in the device.
具體而言，步驟S123中的均方根值可視作振動程度之參考指標，可表示為下列數學式1：
陡峭值(Kurtosis)則是作為零件間敲擊狀況的參考指標，正常訊號多屬於高斯分佈，其陡峭值其值約為3，若發生衝擊波時，表示傳動軸系存在間隙或齒輪斷齒等設備異常，其陡峭值統計參數會大於3。調變值則作為零件間交互關係的參考指標，調變值的計算係先將訊號經過Hilbert轉換後，取絕對值的曲線，經統計判斷別是否屬於弦波，其中Hilbert轉換可以下列數學式2表示：
步驟S130中，該個別基準線係根據一標準差方法而從該複數個基準特徵值所得到，該標準差方法係包括下列步驟：計算各個基準特徵值分佈之標準差及平均值；以及以每個基準特徵值之統計樣本之平均值為一圓心座標，以該種基準特徵值之統計樣本之六個標準差為二弦，以該二弦之一斜邊為半徑畫圓形成該基準線。另外，該特徵值與個別基準線之比對係判斷該複數個特徵值相對於該個別基準線的位置，當該複數特徵值相對該圓心座標的位置在該個別基準線範圍之內，判斷該設備之運轉狀況為正常，當該複數個特徵值相對該圓心座標的位置在該基準線範圍之外，判斷該設備之運轉狀況為異常。 In step S130, the individual reference line is obtained from the plurality of reference feature values according to a standard deviation method, the standard deviation method comprising the steps of: calculating a standard deviation and an average value of each of the reference feature value distributions; The average value of the statistical samples of the reference feature values is a centroid coordinate, and the six standard deviations of the statistical samples of the reference feature values are two chords, and the reference line is formed by drawing a circle with one of the two chords as a radius. In addition, the ratio of the feature value to the individual reference line determines the position of the plurality of feature values relative to the individual reference line, and when the position of the complex feature value relative to the center of the circle is within the range of the individual reference line, determining the The operating condition of the device is normal. When the position of the plurality of characteristic values relative to the center of the circle is outside the range of the reference line, it is determined that the operating condition of the device is abnormal.
步驟S140中，該故障編碼係利用對於該振動加速度訊號及該振動速度訊號對應於每個該頻寬範圍所計算出的該些特徵值相對於該個別基準線的位置所得。 In step S140, the fault code is obtained by using the vibration acceleration signal and the vibration velocity signal corresponding to the position of the characteristic values calculated for each of the bandwidth ranges with respect to the individual reference line.
詳細而言，步驟S140中係利用專家系統記憶該設備之異常模式，以及步驟S130中係利用該專家系統判斷該設備之異常模式，以供日後設備監診之判斷。如此，本發明之智慧設備監診方法可針對不同類型的設備的失效模式建立對應的檢測標準，故可廣泛應用於各種設備，提高了本發明的應用價值。 In detail, in step S140, the abnormal mode of the device is memorized by the expert system, and in step S130, the abnormal mode of the device is judged by the expert system for the judgment of the device supervision in the future. In this way, the smart device monitoring method of the present invention can establish corresponding detection standards for failure modes of different types of devices, and thus can be widely applied to various devices, thereby improving the application value of the present invention.
第2圖係顯示根據本發明之一實施例的智慧設備監診方法應用於一變轉速設備之流程圖，包含下列步驟：在該軸轉設備於一軸轉狀態下時偵測該軸轉設備而得到一振動訊號及一軸轉速訊號(步驟S210)；對於該運轉動態訊號予以分頻為預定的複數個頻寬範圍而計算出每個該頻寬範圍所個別對應的複數個特徵值，其中該複數個特徵值分別為一均方根值、一陡峭值及一調變值(步驟S220)；建立轉速與特徵值的回歸模型，再將各轉速下的各特徵值與對應轉速之對應各特徵值的個別基準線比對(步驟S230)；對於該比對結果進行故障編碼，並記憶該設備之異常模式(步驟S240)。 2 is a flow chart showing the application of the smart device monitoring method to a variable speed device according to an embodiment of the present invention, comprising the steps of: detecting the axis rotating device when the axis rotating device is in a pivoting state; Obtaining a vibration signal and a one-axis rotation speed signal (step S210); dividing the operation dynamic signal into a predetermined plurality of bandwidth ranges to calculate a plurality of eigenvalues corresponding to each of the bandwidth ranges, wherein the complex number The eigenvalues are respectively a root mean square value, a steep value and a modulation value (step S220); establishing a regression model of the rotational speed and the eigenvalue, and then corresponding characteristic values of each eigenvalue at each rotational speed and the corresponding rotational speed The individual reference lines are aligned (step S230); the comparison result is fault coded, and the abnormal mode of the device is memorized (step S240).
步驟S220可詳細分為下列步驟：利用積分方法將該振動加速度訊號積分為一振動速度訊號，以及根據該軸轉速訊號計算轉速(步驟S221)；對於該振動加速度訊號及該振動速度訊號進行諧波小波濾波，使該振動加速度訊號及該振動速度訊號根據該複數個頻寬範圍而被分頻(步驟S222)；以及計算出該振動加速度訊號及該振動速度訊號對應於每個該頻寬範圍的速度均方根值、速度陡峭值、速度調變值、加速度均方根值、加速度陡峭值及加速度調變值(步驟S223)。 Step S220 can be further divided into the following steps: integrating the vibration acceleration signal into a vibration speed signal by using an integration method, and calculating the rotation speed according to the shaft rotation speed signal (step S221); performing harmonics on the vibration acceleration signal and the vibration speed signal Wavelet filtering, the vibration acceleration signal and the vibration speed signal are divided according to the plurality of bandwidth ranges (step S222); and calculating the vibration acceleration signal and the vibration speed signal corresponding to each of the bandwidth ranges The speed rms value, the speed steepness value, the speed modulation value, the acceleration rms value, the acceleration steepness value, and the acceleration modulation value (step S223).
相較於對定轉速設備的監診，本發明的智慧設備監診方法對於變轉速設備進行監診時增加了擷取軸轉速訊號與計算轉速的步驟，並，如第6圖所示，於各特徵值與對應的個別基準線進行比對之前建立一轉速與特徵值的回歸模型，以取得各轉速下對應的特徵值。 Compared with the monitoring of the fixed speed device, the intelligent device monitoring method of the present invention increases the steps of capturing the shaft speed signal and calculating the speed when monitoring the variable speed device, and as shown in Fig. 6, A regression model of the rotational speed and the eigenvalue is established before the eigenvalues are compared with the corresponding individual reference lines to obtain corresponding eigenvalues at the respective rotational speeds.
第3圖至第5圖為本發明的智慧設備監診方法對於一設備進行監診的示意圖，分別將各頻段的速度均方根值、速度調變值、速度陡峭值、加速度軍方根值、加速度調變值以及加速度陡峭值以柱狀圖顯示，其中淺色長條柱代表各個別基準線，深色長條柱代表擷取訊號之對應速度與加速度之各頻段的特徵值。如第3圖所示，當設備運轉正常，擷取訊號對應各頻段的各特徵值低於各對應的個別基準線。如第4圖所示，當設備軸承損壞，零件間出現撞擊，代表振動情形的均方根值以及代表敲擊狀況的陡峭值均出現擷取訊號之特徵值高於個別基準線的情形。如第5圖所示，當對第4圖的情況施予維修，即更換軸承，但設備存在齒輪偏斜的情形，均方根值及陡峭值的柱狀圖中各頻段特徵值均下降，代表零件交互關係的調變值則出現異常。 Fig. 3 to Fig. 5 are schematic diagrams showing the method for monitoring the wisdom of a device according to the present invention, respectively, the speed rms value, the speed modulation value, the speed steep value, and the acceleration military root value of each frequency band. The acceleration modulation value and the acceleration steepness value are displayed in a histogram, wherein the light-length long bar represents each individual reference line, and the dark long bar represents the characteristic value of each frequency band of the corresponding speed and acceleration of the captured signal. As shown in Fig. 3, when the device is operating normally, the characteristic values of the respective signals corresponding to the captured signals are lower than the corresponding individual reference lines. As shown in Fig. 4, when the equipment bearing is damaged and there is an impact between the parts, the root mean square value representing the vibration situation and the steep value representing the tapping condition all have the case where the characteristic value of the extraction signal is higher than the individual reference line. As shown in Fig. 5, when the maintenance of Fig. 4 is carried out, that is, the bearing is replaced, but the gear deflection occurs in the device, and the characteristic values of the respective frequency bands in the histogram of the root mean square value and the steep value are decreased. An exception occurs in the modulation value that represents the interaction of the part.
以上之敘述以及說明僅為本發明之較佳實施例之說明，對於此項技術具有通常知識者當可依據以下所界定申請專利範圍以及上述之說明而作其他之修改，惟此些修改仍應是為本發明之發明精神而在本發明之權利範圍中。 The above description and description are only illustrative of the preferred embodiments of the present invention, and those of ordinary skill in the art can make other modifications in accordance with the scope of the invention as defined below and the description above, but such modifications should still be It is within the scope of the invention to the invention of the invention.
S210、S220、S221、S222、S223、S230、 S240、S241、S242‧‧‧步驟 S210, S220, S221, S222, S223, S230, S240, S241, S242‧‧‧ steps
Claims (6)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW105105342A TWI605322B (en) | 2016-02-24 | 2016-02-24 | Intelligent machine condition monitoring and diagnostic method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW105105342A TWI605322B (en) | 2016-02-24 | 2016-02-24 | Intelligent machine condition monitoring and diagnostic method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| TW201730700A TW201730700A (en) | 2017-09-01 |
| TWI605322B true TWI605322B (en) | 2017-11-11 |
Family
ID=60479782
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| TW105105342A TWI605322B (en) | 2016-02-24 | 2016-02-24 | Intelligent machine condition monitoring and diagnostic method |
Country Status (1)
| Country | Link |
|---|---|
| TW (1) | TWI605322B (en) |
-
2016
- 2016-02-24 TW TW105105342A patent/TWI605322B/en active
Also Published As
| Publication number | Publication date |
|---|---|
| TW201730700A (en) | 2017-09-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9989439B2 (en) | Method and data processing device for severity assessment of bearing defects using vibration energy | |
| Mechefske et al. | Fault detection and diagnosis in low speed rolling element bearings Part I: The use of parametric spectra | |
| US11105712B2 (en) | Integrated vibe/ODM fusion and trending analysis for prognostic health management of engine bearing | |
| EP3788328B1 (en) | System and process for pattern matching bearing vibration diagnostics | |
| US20150322924A1 (en) | Method of monitoring the condition of a wind turbine | |
| CN116304848B (en) | Rolling bearing fault diagnosis system and method | |
| JPH11271181A (en) | Method and device for diagnosing failure in rolling bearing | |
| CN116873156B (en) | Intelligent monitoring method for power abnormality of natural gas ship based on big data | |
| KR20190081933A (en) | Method for sensing and diagnosing abnormality of manufacture equipment | |
| JPH07159231A (en) | Deterioration diagnostic system for pump | |
| CN113757093B (en) | Flash steam compressor unit fault diagnosis method | |
| US20190332102A1 (en) | Machine health monitoring of rotating machinery | |
| WO2022147684A1 (en) | Method and apparatus for identifying abnormalities in mechanical apparatus or mechanical component | |
| CN109416299A (en) | The method for identifying the failure in the gear train in wind turbine | |
| CN104931262A (en) | Bearing fault diagnosis device used in cooperation with wheel set bearing running-in machine and diagnosis method thereof | |
| JP7002417B2 (en) | Equipment abnormality diagnosis device and abnormality diagnosis method | |
| TWI605322B (en) | Intelligent machine condition monitoring and diagnostic method | |
| TWI607811B (en) | Status detection method of rolling apparatus | |
| CN114294183A (en) | Fan blade fault monitoring method and device and fan | |
| CN115931399A (en) | Method for real-time online detection of polygonal fault of high-speed train wheel | |
| JP4312477B2 (en) | Diagnosis method, diagnosis apparatus and program for rotating machine | |
| TWI398629B (en) | Equipment supervision method | |
| JP2003344150A (en) | Condition diagnosing method and apparatus therefor by analyzing shaft vibration of rotating machine | |
| CN115524118A (en) | Bearing fault diagnosis method and system | |
| TWI819824B (en) | Method for applying different rotation-speed-vibration thresholds to diagnose abnormality of equipment with variable rotation speeds |