1294514 九、發明說明： 【發明所屬之技術領域】 本發明係關於音響感測裝置、音響解析診斷裝置及音響 感測益之製造方法，例如係關於利用對應聲音之振動而產 生振動之諧振器的音響感應裝置、音響解析診斷裝置及音 響感測器之製造方法。 曰 【先前技術】 過去’為騎機n或設備有無異常，已有種種設備診斷 方法之提案。此類設備診斷方法係在造成重大事故之前， 針對目前雖可充分繼續發揮功用，但將來可能導致重大故 障之異常，作為檢測對象，例如旋轉機械之軸承有所損 傷’或可動部分已發生損耗等。 例如，在可羅納公司（CORONA PUBLISHING c〇，ltd ) 所出版之「音、診斷之診斷工學」日本音響學界編（非特 許文獻1)當中，便記載以指向性麥克風收音，進行其信號 音之診斷，用以診斷、解析引擎之異常打音。 特開2004-20584號公報（特許文獻υ中揭示了異常監視裝 置及異常監視程式，該異常監視裝置及異常監視程式係以 感測器取得顯示物理量之訊號，該物理量反映診斷對象之 狀態，將上述訊號分割為複數個頻帶之頻帶訊號，對所得 到之複數個頻帶訊號分別設定有各判定基準，根據此基 準，依各頻帶判定珍斷對象有無異常。 [非特許文獻1]「音、診斷之診斷工學」曰本音響學界 編（可羅納公司出版）。 113171.doc 1294514 [特許文獻1]特開2004_20584號公報 [發明所欲解決之問題] ”非特許文獻1所記載之使用麥克風之診斷方法，於測定 J哀境中，有目的音以外之雜訊時，若雜訊之音壓大，使麥 克風之輸出達飽和狀態，則無法捕捉目的音。此乃因麥克 風之設計，係對可檢測之所有頻帶具有相同感度之故。 特許文獻1所揭示之裝置，係使用可編程帶通滤波器將 鲁感測器之輸出訊號分割為複數個頻帶的頻帶訊號。使用可 編程帶通濾、波器，若頻帶之分割數為n，首先進行第i頻帶 之抽出，針對所抽出之第丨頻帶信號的異常判定處理結束 之後，再針對第2個頻帶進行抽出處理，如此必須進 抽出及異常判定處理。該處理係基於程式之軟體處理，其 問題在於進行複雜之演算需要處理時間。 ^ 因此，本發明之目的為提供可以較簡單的結構，即時解 析診斷自測定對象所發出之聲音的音響感測裝置、音響解 • 冑診斷裝置及音響感測器之製造方法，以利用較簡單^結 構即時解析測定對象所發出之聲音。 ^ 【發明内容】 本發明為—種音響感測器裝置，其係檢測自測定對象發 出之寬廣頻帶之聲音中至少特定頻帶之聲音者，且具有: 第1音響檢測元件，其係包含複數個對應不同聲音產生窄 振之第1振子，檢測頻帶内第1頻帶之聲音，並抽出測定： 2第1特徵音者；及第2音響檢測元件，其係包含複數個 子應不同聲音產生諧振之第2振子，檢測頻帶内與第味帶 113171.doc 1294514 不同之第2頻帶之聲音’並抽出測定對象之第2特徵音者。 要製造出以1個元件檢測寬廣頻帶之聲音所需的元件慨 有困難’但分為第i音響檢測元件與第2音響檢測元件，二 測由測定對象發出之第丨及第2特定頻帶之聲音，因此二 除製造元件上之困難性。 ^ 7胃 理想的是’第i音響檢測元件係檢測自測定對象 之寬廣的全頻帶聲音中之特定頻帶之聲音，而第2音：二 測元件係檢測自測定對象所發出之寬廣的全頻帶聲音；-剩餘頻帶之聲音。藉此彳晷以拾、、目彳人a姚 之 對象之狀態。聲音，解析測定 :，弟1音響檢測元件係檢測自測定對象所發出 之寬廣的全頻帶聲音中之特定頻帶之聲音，而第2立 測元件係檢測自測定對象所發出之宽廣的 ：：欢 ，理想的是’第⑽徵音係顯示測定對象為正常 顯示測定對象為異常。藉此可診斷測定對象之正常 特定頻帶内部分頻帶之聲音及剩餘頻帶之聲音。曰之 本發明之另一態 測、解析並診斷自 特定頻帶之聲音者 包含對應各個不同 廣之頻帶内且各個 係解析由複數個音 響訊號，診斷聲音[Technical Field] The present invention relates to an acoustic sensing device, an acoustic analysis and diagnostic device, and a manufacturing method of acoustic sensing, for example, a resonator that generates vibration using vibration corresponding to sound. An acoustic sensing device, an acoustic analysis diagnostic device, and a method of manufacturing an acoustic sensor.曰 [Prior Art] In the past, there was a proposal for various equipment diagnostic methods for the rider n or equipment. This type of equipment diagnostic method is based on the abnormality that may continue to function in the future, but may cause major failures in the future, such as the bearing of the rotating machinery, or the loss of the movable part. . For example, in the "Acoustic and Diagnostic Diagnostic Engineering" published by CORONA PUBLISHING c〇, Ltd., the Japanese Acoustics Society (non-patent literature 1), it is described that the directional microphone is used to receive the signal and perform its signal. The diagnosis of sound is used to diagnose and analyze the abnormal sound of the engine. Japanese Laid-Open Patent Publication No. 2004-20584 (Available Document No.) discloses an abnormality monitoring device and an abnormality monitoring program that acquires a signal indicating a physical quantity by a sensor, and the physical quantity reflects the state of the diagnostic object. The signal is divided into frequency bands of a plurality of frequency bands, and each of the plurality of frequency band signals is set with each determination criterion, and based on the reference, the presence or absence of an abnormality is determined for each frequency band. [Non-licensed document 1] "Sound, diagnosis" "Diagnostic Engineering", edited by Sakamoto Acoustics (published by Rhone Corporation). 113171.doc 1294514 [Patent Document 1] JP-A-2004-20584 [Problems to be Solved by the Invention] "Microphones as described in Non-Patent Document 1" In the measurement method, when there is noise other than the target sound in the J sorrow, if the sound pressure of the noise is large and the output of the microphone is saturated, the target sound cannot be captured. This is due to the design of the microphone. The same sensitivity is applied to all the detectable frequency bands. The device disclosed in Patent Document 1 uses a programmable band pass filter to measure the Lu sensor. The output signal is divided into frequency band signals of a plurality of frequency bands. If the number of divisions of the frequency band is n, the extraction of the ith frequency band is first performed, and the abnormality determination processing for the extracted third frequency band signal is used. After the end, the extraction process is performed for the second frequency band, and thus the extraction and abnormality determination processing must be performed. This processing is based on the software processing of the program, and the problem is that the processing time is required to perform complicated calculation. ^ Therefore, the object of the present invention is Providing an acoustic sensing device, an acoustic solution, a diagnostic device, and an acoustic sensor manufacturing method capable of analyzing the sound emitted from the measurement object in a simple configuration, for real-time analysis of the measurement object by using a simpler structure [Invention] The present invention relates to an acoustic sensor device that detects a sound of at least a specific frequency band among sounds of a wide frequency band emitted from a measurement target, and has: a first acoustic detecting element, a plurality of first vibrators corresponding to different sounds to generate a narrow vibration, and detecting the sound of the first frequency band in the frequency band, and Measurement: 2, the first characteristic sound; and the second acoustic detecting element, which includes a plurality of second transducers that generate resonances with different sounds, and sounds of the second frequency band different from the first taste band 113171.doc 1294514 in the detection frequency band 'And extracting the second characteristic sound of the measurement target. It is difficult to manufacture a component required to detect a wide-band sound with one component, but it is divided into an i-th acoustic detecting element and a second acoustic detecting element. Since the sounds of the third and second specific frequency bands emitted by the measurement target are measured, it is difficult to manufacture the components. ^7 The stomach is ideally that the 'i-th acoustic detection element detects the specificity of the wide-band sound from the measurement target. The sound of the band, and the second sound: the second measuring element detects the wide full-band sound emitted from the measuring object; the sound of the remaining band. In order to pick up and see the state of the object of Yao. Sound, analysis and measurement: The first sound detecting component detects the sound of a specific frequency band in the wide-band sound emitted from the measurement target, and the second vertical measuring component detects the wide sound emitted from the measurement target: Huan, it is desirable that the '10th sound system indicates that the measurement target is normal and the measurement target is abnormal. Thereby, the sound of the partial frequency band in the normal specific frequency band of the measurement target and the sound of the remaining frequency band can be diagnosed. Another aspect of the present invention is to measure, analyze, and diagnose sounds from a specific frequency band.
樣係為-種音響診斷解析ϋ，其係檢 測定對象發出之寬廣頻帶之H ’且具備：複數個音響檢測元件，立係 聲音產生譜振之振子，僅檢測包含於寬 不同頻帶之聲音；以及解析診斷部，其 響檢測元件所輸出之各個不同頻帶的音 之为佈，解析測定對象之特徵。 H3171.doc 1294514 本&月於測定對象為具有极數台有複數故障模式 之攻置設備之工廠時，可診斷各故障模式。 理想的是’解析診斷部係以測定對象之正常、異常作為 特徵進行解析。 理〜、的疋’解析診斷部依照解析異常檢測，判斷是否要 保養檢查測定對象。依照此解析結果，可確實判斷是否需 要保養。 本發明之又另一態樣，係一種音響感測器之製造方法， 该音響感測器係用於測定對象之音響解析者，其製造方法 具備：準備複數個音響檢測元件之步驟，上述複數個音響 才双測元件係包含對應各個不同之聲音產生諧振之振子；於 測疋對象附近設置複數個音響檢測元件，依照目的收集測 定對象資料之步驟；以及根據所收集之資料，決定複數個 音響檢測元件中與特定頻帶之聲音產生譜振之音響檢測元 件的組合之步驟。藉由此組合，可進行對應測定對象之狀 態的監視與診斷。 【實施方式】 圖1為一根據本發明之一實施態樣之音響感測裝置例之 平面圖。圖1中，作為音響檢測元件之音響應測裝置丨，形 成於半導體矽基板10上，包含隔膜2、橫隔片3、終止板 4，以及複數個諧振片5。 隔膜2由接受輸入音時會產生振動之薄板形成。橫隔片3 之形成結合隔膜2與終止板4，於隔膜2 一側的寬度較寬， 越向終止板4一側越窄，於終止板4端最窄。 113171.doc 1294514 二数個諧振片5長度經調整，各與特定頻率產生諧振， 3單面切並向兩側延伸。此音響感測裝置！係由 ，同二振頻率之譜振片5，一對對形成魚骨狀構造。 反4之°又置，係為將自隔膜2經由橫隔片3傳導而來之 ::音所產生振動加以吸收，使其不至於回傳至隔膜2。 此外’隔膜2、橫隔片3、終止板4,以及複數個諸振片5下 方形成空間，複數個諧振片5周圍則有開口。The sample system is an acoustic diagnosis analysis method, which detects a wide frequency band H′ emitted by a fixed object, and includes: a plurality of acoustic detecting elements, and a vibrator that generates a spectral sound, and detects only sounds included in different frequency bands; And an analysis diagnosing unit that analyzes the characteristics of the measurement target by the sound of each of the different frequency bands outputted by the detecting element. H3171.doc 1294514 This & month can diagnose each failure mode when the measurement target is a factory with a large number of attack devices with multiple failure modes. It is preferable that the 'analytical diagnosis unit analyzes the normal and abnormal characteristics of the measurement target. The analysis unit of the 〜~, 解析 analysis unit determines whether or not to check the measurement target in accordance with the analysis of the abnormality. According to the results of this analysis, it is possible to determine whether maintenance is required. Still another aspect of the present invention provides a method of manufacturing an acoustic sensor for measuring an acoustic analyzer of a subject, the method of manufacturing comprising: a step of preparing a plurality of acoustic detecting elements, the plural The acoustic dual-measurement component includes a vibrator that generates resonance corresponding to each different sound; a plurality of acoustic detecting components are disposed in the vicinity of the measuring object, and the measuring object data is collected according to the purpose; and a plurality of sounds are determined according to the collected data. The step of detecting a combination of an acoustic detecting element that produces a spectral sound with a sound of a particular frequency band in the component. By this combination, monitoring and diagnosis of the state of the measurement target can be performed. [Embodiment] Fig. 1 is a plan view showing an example of an acoustic sensing device according to an embodiment of the present invention. In Fig. 1, a sound response measuring device 音响 as an acoustic detecting element is formed on a semiconductor substrate 10, and includes a diaphragm 2, a lateral spacer 3, a termination plate 4, and a plurality of resonator plates 5. The diaphragm 2 is formed of a thin plate that generates vibration when receiving an input sound. The formation of the diaphragm 3 is combined with the diaphragm 2 and the terminating plate 4, and the width on the side of the diaphragm 2 is wider, the narrower toward the end of the terminating plate 4, and the narrowest at the end of the terminating plate 4. 113171.doc 1294514 Two of the resonator plates 5 are adjusted in length, each resonating with a specific frequency, and 3 are cut on one side and extended to both sides. This sound sensing device! The pair of pairs of the same vibrational frequency of the diaphragm 5, a pair of pairs form a fishbone structure. The reverse 4 is again set to absorb the vibration generated by the sound from the diaphragm 2 via the diaphragm 3 so as not to be transmitted back to the diaphragm 2. Further, a space is formed under the diaphragm 2, the diaphragm 3, the stopper plate 4, and the plurality of diaphragms 5, and an opening is formed around the plurality of resonator sheets 5.
輸出音所造成的振動使隔膜2垂直振動，該振動以水平 方向傳導至橫隔片3,複數個諸振片5中所對應的振子便垂 讯，動1以改變谐振片5之長度或厚度，將各譜振頻率 …；所希望之數值。各諧振片5中，於橫隔片3側設有未 圖示之壓阻’壓阻之阻力大小變化可加以顯示，例如以惠 斯頓電橋的輸出來顯示。 圖2係一顯不圖1所示之音響感測裝置中複數個譜振片之 頻率應答性之圖’圖3係顯示輸出、輸入特性之圖，圖4係The vibration caused by the output sound causes the diaphragm 2 to vibrate vertically, and the vibration is transmitted to the diaphragm 3 in the horizontal direction, and the vibrator corresponding to the plurality of diaphragms 5 is transmitted, and the movement 1 changes the length or thickness of the resonator sheet 5. , will be the frequency of each spectrum ...; the desired value. In each of the resonator plates 5, a change in the magnitude of the resistance of the piezoresistive piezoresistive (not shown) provided on the side of the diaphragm 3 can be displayed, for example, as an output of a Wheatstone bridge. Fig. 2 is a diagram showing the frequency responsiveness of a plurality of spectral patches in the acoustic sensing device shown in Fig. 1. Fig. 3 is a diagram showing output and input characteristics, and Fig. 4 is a diagram
顯示環境噪音耐性之圖。 圖1所揭示之音響感測裝置1之複數個諧振片5，係如圖2 所顯示’如調整長度使其與12他〜2 5服之諸振頻率產 生-皆振，可使其與各諧振頻率進行選擇性應答。而且，諧 振片5間之相互作用，亦可使各諧振頻率内之頻率進行應 隔膜2以圖3所顯不之橫轴所示數值產生振幅時，複 數個諧振片5如圖3之縱軸所示之值產生振幅，因此輸出約 可放大至約14倍。 113171.doc -10- J294514 此外’圖1所顯示之立鄕 之曰響感測裝置1，即使受到目的音以 外之妨礙音輸出，供可从、， 一 仏測出目的音之訊號。圖4所顯示 之波瓜a顯不目的音（2_ Ηζ，叫輸入時的輸出波形， 波形b顯示妨礙音⑽Hz，1()pa)輸入時的輸出波形，波 心顯示目的音與㈣音同時輪入時的輸出波形。自本圖4 中卩使又到开乂成雜讯的妨礙音輸入，仍可檢測出音壓為 妨礙音的1/100之目的音。A graph showing the environmental noise tolerance. The plurality of resonator plates 5 of the acoustic sensing device 1 disclosed in FIG. 1 are as shown in FIG. 2, and if the length is adjusted to be generated by the vibration frequencies of 12 to 25, the vibration and the vibration are generated. The resonant frequency is selectively responsive. Moreover, the interaction between the resonator plates 5 can also cause the frequencies in the respective resonance frequencies to be generated when the diaphragm 2 generates amplitudes by the values shown by the horizontal axis shown in FIG. 3, and the plurality of resonator plates 5 are as shown in the vertical axis of FIG. The values shown produce amplitude, so the output can be scaled up to about 14 times. 113171.doc -10- J294514 In addition, the squeaking sensing device 1 shown in Fig. 1 receives the obstruction sound output from the target sound, and can detect the signal of the destination sound. The Bogagua a shown in Fig. 4 shows the target sound (2_ Ηζ, which is called the output waveform at the time of input, and the waveform b shows the output waveform when the input sound (10) Hz, 1 () pa) is input, and the wave center displays the target sound and the (4) sound simultaneously. Output waveform when wheeled. From the interference input of this figure 4 to the noise, it is still possible to detect the sound of the sound that is 1/100 of the obstruction sound.
圖5為本發明之—之實施態樣音響解析診斷裝置之方塊 圖’（A)顯示資料收集、傳輸電路，（B)顯示資料接收、解 析診斷電路。 圖5(A)中，資料收集、傳輸電路3〇，包含音響感測器 11 15、貝料轉換電路21〜25、控制…丨、資料傳輸電路 、及天線3 3。曰響感測器11〜15，係檢測測定對象發 出之頻帶寬廣之聲音的所有頻帶者，冑用圖」顯示之音響 感測裝置1。 音響感測器11，例如可與5〇〇 kHz之聲音產生諧 振，音響感測器12與900 Hz〜L8 kHz，音響感測器13與17 kHZ 3·4 kHz，音響感測器14與3.3 kHz〜6.8 kHz，音響感 測15與6.7 kHz〜1〇 kHz之聲音產生諧振，音響感測器n〜l5 各自由所包含的諧振片構成。因此，音響感測，如 此例’可檢測出頻帶為6〇〇 HZ〜10 kHz之聲音。 要以1個音響感測裝置1，檢測頻帶寬廣之聲音，則橫隔 片將變長’使言皆振片5之製造更為困難，但若以5個音響感 測器11〜1 5分別檢測各頻帶，諧振片5較容易製造，可省略 H3l7l.doc 1294514 不必要的頻帶的振子。 頻：者:,可視測定對象任意設定頻帶’測定對象所輪出的 、：继例如要檢測_ Hz〜1〇〇咖之全頻帶之聲 择 加印響感測器的數量，式奂 9 者亦可自測定對象所輸出之全頻 、曰，僅針對部分頻帶進行檢測。此外，亦可 個音響感測器檢測所有葙德由甘y ^ 1 J所有頻帶中某個頻帶的聲音，以其他立 曰感測器檢測其餘頻帶之聲音。 曰 各音：感測器㈣之輸出，將傳達至資料轉換電路 5。貝料轉換電路21〜25 ’將音響感測器"〜 :頻帶之聲音的訊號進行A/D轉換，以音壓強度分佈』 科輪出至控制部31。控制部31將資料轉換電路21〜25所傳 達之資料依時間相依序輸^資料傳輸電路32,資料傳 輸電路32自天線33將各資料以無線電或光等傳輸。 圖5(B)所顯示之資料接收、解析診斷電路4〇,包含天線 4卜資料接收電路42、解析診斷電㈣，以及顯示部料。 資料接收電路42將資料收隹、# μ $ & 竹貝科收集、傳輸電路3〇所傳輸來之各音 響感測器n〜15的各資料，經由天線41接收，傳達至解析 診斷電路43。 解析診斷電路43視診斷之目的，事先記憶音壓強度分佈 作為資料’針對該音壓強度分佈與資料所傳達之音壓強度 加以比較，進行診斷目的之特徵抽出。即，若診斷目的係 測定對象為正常狀態之檢測或異常狀態之檢測，將記憶測 定對象正常動作時或異常時之聲音的音壓分佈，作為特徵 加以記憶。當測定對象異㈣，所呈現的聲音之音壓分佈 Π3171.doc -12- 1294514 將與正常狀態時不同，因此將各自相對之聲音的音壓強度 • 作為特徵音事先記憶。又，異常狀態係包含測定對象故障 之狀態，以及預知測定對象的故障之狀態。 解析診斷電路43，一旦接收對應於各音響感測器u〜15 之資料，便將各自之音壓強度與事先記憶之音壓強度進行 比較，自音響感測器11〜15中選擇最適當的組合。顯示部 44將顯示被選擇出的音響感測器組合。 φ 圖6係為說明本發明之一實施態樣中音響解析診斷裝置 之動作的流程圖，（A)顯示資料收集、傳輸電路3〇的動 作，（B)顯示資料接收、解析診斷電路4〇的動作。 其次，將參照圖6，針對本發明之一之實施態樣音響解 析診斷裝置所診斷測定的對象之正常或異常的動作加以說 明。音響感測器11〜15係配置於未圖示之測定對象附近， 測定對象一產生動作，音響感測器n〜15*，圖以斤示之隔 膜2便應測定對象所發出之聲音產生振動，該振動經由橫 • 隔片3傳遞至複數個諧振片5。複數個諧振片5中對應於測 定對象之聲音的振子將垂直振動。壓阻元件之阻力數值因 該振動發生變化，以電壓變化的形式得出。 自各音響感測器11〜15輸出各自與其對應之聲音振動相 應之訊號，傳達至資料轉換電路21〜25, a/d轉換後作為顯 不音壓強度之資料，傳達至控制部31。控制部”於圖心） 所顯示之步驟（圖示中簡稱為sp)spi中，判別顯示來自音響 感測器11〜15之顯示音壓度的資料是否己輸入，若己輸 入，則於SP2中，僮白咨刺& 1更自貝料傳輸電路32經由天線33，依次 113171.doc 1294514 傳輸各資料。 圖5(B)顯示之資料接收、解析診斷電路4〇之資料接收電 路42，於圖6(B)所顯示之步驟spllt，判別是否己接收音 響感測器11〜15之所有資料。若已接收，則解析診斷電路 43於步驟SP12中，根據各音響感測器u〜15之資料，自各 聲音之音壓強度中抽出作為診斷目的的特徵訊號，於步驟 13中，決定至少1個以上來自複數個組合之音響感測器來 檢測其特徵訊號，作為診斷對象設備之訊號檢測裝置。接 著，於步驟SP14中，顯示出該裝置之音響檢測器u〜i5* $亥當者。 破決定用來作為訊號檢測裝置之複數個音響感測器 11〜15之組合，依診斷目的進行選擇。例如，診斷之目的 ,檢測敎對象為正常狀態或異常狀態之故障預測，亦或 是檢測出故障’依診斷之目的來決定。由於測定對象所發 出的聲音的分佈，視正常狀態或預知故障歧障狀態有戶; 不同，因此音響感測器之組合，以能夠檢測出各聲音之分 佈來決定。例如’要檢測出正常狀態，選擇音響感測： 12、13 ’及15的組合；要檢測出故障預知，則選擇音響檢 :器11、U，及13的組合；要檢測出故障則選擇音響二 器13、14，及1 5的組合。 如上所述，根據此實施態樣’由複數個音響感" 11〜15各自對不同聲音產生諧振，僅檢測出全頻帶中各個 不同頻帶之聲音，解析複數個音響感測器u〜_輸出之 各個不同頻帶之音響訊號，診斷聲音之分佈，以應^特衩 113171.doc •14- 1294514 選擇音響感測器的組合，不僅構造較為簡$， 解析診斷測定對象所發出之聲音。 ° 、 而且，曰響感測器11〜15採用諧振片5，一對對形成魚骨 狀構造之感測器，不需要以軟體處理來分解頻率，可即時 進打頻率分解。此外，可減少分解各頻帶的計算時間，因 此可縮短處理時間。 圖7係顯示採用本發明之另—實施樣態之音響解析診斷 裝置’於全頻帶中解析診斷測定對象音之一例之圖，'圖8 係圖7所顯示之例中，以音響解析診斷裝置配置最適之音 響感測器之一例之圖。圖7及圖8之各（A)係顯示資料收 集、傳輸電路，（B)係顯示資料接收、解析診斷電路。 如圖7(A)所示，音響感測器11〜15配置於測定對象附 近。資料收集、傳輸電路3〇a為圖5(A)所顯示之資料收 集、傳輸電路3〇中，包含除音響感測n〜15n之部分，資 料接收、解析診斷電路40以圖6(Β)所顯示之流程圖為^ 準，於全頻帶中解析診斷測定對象5〇所發出之聲音。而例 如’為檢測出正常狀態，選擇音響感測器12、 υ，及15的 組合。將所選擇之音響感測器12、13，及15加以配置即為 圖8(A)。 … 以圖8(B)所顯示之資料接收、解析診斷電路4〇，收集測 定對象50所發出之聲音的資料，解析診斷測定對象處於正 常狀態或異常狀態，進行測定對象5〇是否需要維修之判斷 等狀態監視。此外，如圖5之說明，進行故障預知時，配 置音響感測11、12，及13 ;進行故障檢測時，酉己置音響感 113171.doc 1294514 測器13、14，及15即可。 圖9係顯不本發明之另一實施態樣之音響解析診斷裝 置，（A)為顯示工場内音響感測器配置之一例之概念圖， (B)則顯不資料接收、解析診斷電路。圖9(A)之工廠6〇 中，雖未圖示，但設置有複數台具有複數個故障模式之裝 置設備。因此工廠内發生了複數個異常聲音之雜訊。因 此，在本實施態樣中，工廠6〇内配置有複數個音響感測器 11〜15，以檢測出各異常聲音。 此外’圖9⑷中’圖示雖經省略，但圖7中所顯示之資 料收集、傳輸電路30a已收到各音響感冑器"七之輸出。 圖8(B)所顯示之資料接收、解析診斷電路姆配置於工廠 60之外貝料接收、解析診斷電路40係由音響感測器 1卜15對各個聲音進行檢測，因此根據各音響感測器⑴心 之貝料，進仃裝置設備為正常或異常、是否需要維修之判 斷等狀態監視。 圖10係顯示本發日月> 7 2 ^ 、 月之又另一態樣之音響解析診斷裴置， (A)為顯示於固有奘罟由 口有裒置中配置音響感測器之一例之圖， 則顯示資料接收、解析診斷電路。 :圖1〇(A)所示，工廒等地配置有固有裝置η，，各. 表置71〜74發出雜訊。於 ^ ^ 器^〜14。資料μ Α 有音響感須 貝枓接收、解析診斷電路40根據未圖示之 收集、傳輪雷败Μ 貝寸 n k達之對應於各音響感測器11〜14之f 枓，檢測出各裝置71〜74 之貝 ^ 口令工掌曰，得以進行正、 吊之判斷或特徵抽出等之診斷。 … Π3171 .doc • 16 · 1294514 如上所述，參照圖式飪 八針對本發明之實施態樣進行說明， 但本發明並不限於所圖 '、之實施態樣。針對已圖示之實施 悲樣，於與本發明之同一 把圍内，或於均等之範圍内，可 加以種種修正與變化。 [發明的效果] 根據本發明，具備Α φ、ΒΪ + 出測疋對象之第1特徵音的第1音響 檢測元件，以及抽出測定對^ ^ ^ ^ ^ ^ ^ 〜β豕之笫2特欲音的第2音響檢測 元件，可視測定對象之特徵 订饿g，以較間早之結構即時對測 定對象所發出之聲音進行解析。 同時’藉由與各個不同的聲音發生諧振，檢測出包含於 寬廣頻帶内且不同頻帶之聲音，解析複數個音響檢測元件 所輸出之各不相同之頻帶的音響訊號，診斷其聲音之分 佈，藉由解析敎對象之特徵，即時解析測定對象所發出 之聲音。 此外’準備與各個不同聲音產生諧振之複數個音響檢測 讀，將該複數個音響檢測元件設置於測定對象附近，依 目的收集測定對象之資料，絲據所收㈣之㈣，自複 數個音響檢測元件中’決定對特定頻帶之聲音產生諧振之 音響檢測元件的組合，由此可製造即時 衣1峙解析測定對象所產 生之聲音的音響檢測元件。 [産業上之利用可能性] 本發明之音響感測裝置、音響解析診斷裝置，及音響感 測器之製造方法，可利用於判斷工礙内之機器或裝=備 有無異常。 113171.doc -17- 1294514 【圖式簡單說明】 圖1為一根據本發明之一實施態樣之音響感測裝置例之 一平面圖； 圖2係一顯示圖1所示之音響感測裝置中複數個讀振片之 頻率應答性之圖； 圖3係一顯示音響感測裝置中輸出、輸入特性之圖； 圖4係一顯示音響感測裝置中環境噪音耐性之圖； φ 圖5(A)、（B)係二本發明之一實施態樣中音響解析診斷 裝置之方塊圖； 圖6(A)、（B)係二為說明本發明之一實施態樣中音響解 析診斷裝置之動作之流程圖； 圖7(A)、（B)係二顯示採用本發明之音響解析診斷裝置， 於全頻帶中解析診斷測定對象音之一例之圖； 圖8(A)、（B)係二顯示圖7所示之以該音響解析診斷裝置， 配置最適當之音響感測器之一例之圖； _ 圖9(A)、（B)係二顯示工廠内配置音響感測器之一例之 概念圖； 圖10(A)、（B)係一顯示於固有裝置中配置音響感測器之 一例之圖。 【主要元件符號說明】 1 音響感測裝置 2 隔膜 3 橫隔片 4 終止板 113171.doc -18- 1294514 5 譜振片 10 半導體矽基板 11 〜15 音響感測器 21 〜25 資料轉換電路 30, 30a 資料收集、傳輸電路 31 控制部 32 資料傳輸電路 33, 41 天線 40 資料接收、解析診斷電路 42 資料接收電路 43 解析診斷電路 44 顯示部 50 測定對象 60 工廒 71 〜74 裝置 113171.doc 19-Fig. 5 is a block diagram of the embodiment of the present invention for analyzing the sound diagnostic apparatus. (A) shows a data collecting and transmitting circuit, and (B) shows a data receiving and analyzing diagnostic circuit. In Fig. 5(A), the data collecting and transmitting circuit 3 includes an acoustic sensor 11 15 , a bedding conversion circuit 21 to 25, a control unit 丨, a data transmission circuit, and an antenna 33. The click sensors 11 to 15 detect all the frequency bands of the sound having a wide frequency band emitted from the measurement target, and use the sound sensing device 1 shown in the figure. The acoustic sensor 11, for example, can resonate with a sound of 5 kHz, the acoustic sensor 12 and 900 Hz to L8 kHz, the acoustic sensor 13 and 17 kHZ 3·4 kHz, the acoustic sensor 14 and 3.3 kHz ~ 6.8 kHz, the acoustic sensing 15 and the sound of 6.7 kHz ~ 1 〇 kHz resonate, the acoustic sensors n ~ l5 are composed of the resonators contained in each free. Therefore, the acoustic sensing, as in this example, can detect a sound having a frequency band of 6 〇〇 HZ 10 10 kHz. If one sound sensing device 1 is used to detect a sound with a wide frequency band, the diaphragm will become longer, making it more difficult to manufacture the sounding film 5, but if the five acoustic sensors 11 to 15 respectively The resonator piece 5 is easier to manufacture by detecting each frequency band, and the oscillator of the unnecessary frequency band of H3l7l.doc 1294514 can be omitted. Frequency: The visual measurement target is arbitrarily set to the frequency band 'measured by the measurement target: for example, the number of sound-sensing and printing sensors that detect the full frequency band of _ Hz~1 〇〇 coffee, the formula 奂9 It is also possible to detect the full frequency and 曰 output from the measurement object only for some frequency bands. In addition, an acoustic sensor can detect the sound of all the bands in all the bands of the y y ^ 1 J, and the sound of the remaining bands can be detected by other stereo sensors.曰 Each tone: The output of the sensor (4) will be transmitted to the data conversion circuit 5. The beating material conversion circuits 21 to 25' perform A/D conversion of the sound of the sound sensor "~: frequency band, and the sound pressure intensity distribution is output to the control unit 31. The control unit 31 outputs the data transmitted from the material conversion circuits 21 to 25 to the data transmission circuit 32 in order of time, and the data transmission circuit 32 transmits the data from the antenna 33 by radio or light. The data receiving and analyzing diagnostic circuit 4 shown in Fig. 5(B) includes an antenna 4 data receiving circuit 42, a diagnostic diagnostic power (4), and a display material. The data receiving circuit 42 receives the data, the data of each of the acoustic sensors n to 15 transmitted from the data collection and transmission circuit 3〇, and transmits it to the analysis diagnostic circuit 43 via the antenna 41. . The analysis diagnostic circuit 43 pre-stores the sound pressure intensity distribution as a data for the purpose of diagnosis, and compares the sound pressure intensity distribution with the sound pressure intensity transmitted by the data to extract the features for the purpose of diagnosis. In other words, if the purpose of the diagnosis is the detection of the normal state or the detection of the abnormal state, the sound pressure distribution of the sound when the measurement target is normally operated or abnormal is stored as a feature. When the measurement object is different (4), the sound pressure distribution of the presented sound Π3171.doc -12-1294514 will be different from the normal state, so the sound pressure intensity of each relative sound is stored as a characteristic sound in advance. Further, the abnormal state includes the state of the measurement target failure and the state of the failure of the measurement target. The analysis diagnostic circuit 43 compares the respective sound pressure intensity with the previously stored sound pressure intensity once the data corresponding to each of the acoustic sensors u 15 is received, and selects the most appropriate one from the acoustic sensors 11 to 15 combination. The display unit 44 will display the selected acoustic sensor combination. Fig. 6 is a flow chart for explaining the operation of the acoustic analysis and diagnostic apparatus in an embodiment of the present invention, (A) showing the operation of the data collecting and transmitting circuit 3, and (B) displaying the data receiving and analyzing diagnostic circuit. Actions. Next, the normal or abnormal operation of the object diagnosed and measured by the acoustic analysis and diagnosis apparatus according to the embodiment of the present invention will be described with reference to Fig. 6 . The acoustic sensors 11 to 15 are disposed in the vicinity of the measurement target (not shown), and the sound sensor n to 15* is generated as soon as the measurement target is generated, and the diaphragm 2 shown in the figure is vibrated by the sound emitted by the measurement target. This vibration is transmitted to the plurality of resonator plates 5 via the horizontal spacer 3. The vibrator of the plurality of resonator plates 5 corresponding to the sound of the measuring object will vibrate vertically. The resistance value of the piezoresistive element is changed by the vibration and is obtained in the form of a voltage change. Each of the acoustic sensors 11 to 15 outputs a signal corresponding to the corresponding sound vibration, and transmits it to the data conversion circuits 21 to 25, and the data is converted to the sound pressure intensity and transmitted to the control unit 31 after the a/d conversion. In the step (referred to as sp) in the step displayed by the control unit (in the figure), it is determined whether or not the data indicating the display sound pressure from the acoustic sensors 11 to 15 has been input. If it is input, the SP2 is input. In the middle, the child white thorns & 1 further transmits the data from the bedding transmission circuit 32 via the antenna 33, and sequentially 113171.doc 1294514. The data receiving circuit 42 of the data receiving and analyzing diagnostic circuit 4 shown in Fig. 5(B), In step Spllt shown in FIG. 6(B), it is determined whether all the data of the acoustic sensors 11 to 15 have been received. If received, the analysis diagnostic circuit 43 proceeds to step SP12 according to each of the acoustic sensors u 15 In the data, the characteristic signal for diagnosis purpose is extracted from the sound pressure intensity of each sound, and in step 13, at least one or more acoustic sensors from a plurality of combinations are determined to detect the characteristic signal, and the signal detection of the device to be diagnosed is detected. Then, in step SP14, the acoustic detector u~i5*$Huier of the device is displayed. The combination of the plurality of acoustic sensors 11-15, which are used as the signal detecting device, is determined according to the purpose of diagnosis. get on For example, for the purpose of diagnosis, to detect the failure prediction of the target or the abnormal state, or to detect the fault 'determined according to the purpose of the diagnosis. Due to the distribution of the sound emitted by the measurement object, depending on the normal state or prediction The faulty fault state has households; different, so the combination of acoustic sensors is determined by being able to detect the distribution of each sound. For example, 'to detect the normal state, select the acoustic sensing: 12, 13 ' and 15 combinations; To detect the fault prediction, select the combination of the sound detectors: 11, U, and 13; to detect the fault, select the combination of the sound two devices 13, 14, and 15. As described above, according to this embodiment' Each of the plurality of acoustic sensations "11~15" resonates with different sounds, detects only the sounds of different frequency bands in the entire frequency band, and analyzes the acoustic signals of the different frequency bands of the plurality of acoustic sensors u~_, and diagnoses the sound. The distribution of the sound sensor is selected by the combination of 113171.doc •14-1294514, which not only has a simple structure of $, but also analyzes the sound of the measurement object. Moreover, the squeaking sensors 11 to 15 employ the resonator piece 5, and a pair of sensors forming a fishbone structure do not need to be processed by the soft body to decompose the frequency, and the frequency can be decomposed immediately. Further, the decomposition can be reduced. The calculation time of the frequency band can shorten the processing time. Fig. 7 is a view showing an example of analyzing the sound of the diagnosis target in the entire frequency band by the acoustic analysis diagnostic apparatus of the other embodiment of the present invention, and Fig. 8 is a diagram of Fig. 7 In the example shown, an example of an optimum acoustic sensor is arranged in the acoustic analysis diagnostic apparatus. Each of (A) of FIG. 7 and FIG. 8 displays data collection and transmission circuits, and (B) shows data reception and analysis. Diagnostic circuit. As shown in Fig. 7(A), the acoustic sensors 11 to 15 are disposed in the vicinity of the measurement target. The data collection and transmission circuit 3A is the data collection and transmission circuit 3 shown in FIG. 5(A), and includes the portion except the acoustic sensing n 1515, and the data receiving and analyzing diagnostic circuit 40 is shown in FIG. 6 (Β). The flow chart shown is correct, and the sound emitted by the diagnostic measurement object 5 is analyzed in the entire frequency band. For example, to detect a normal state, a combination of acoustic sensors 12, υ, and 15 is selected. The selected acoustic sensors 12, 13, and 15 are arranged as shown in Fig. 8(A). The data is received and analyzed by the data shown in Fig. 8(B), and the data of the sound emitted by the measurement object 50 is collected, and the diagnostic measurement object is in a normal state or an abnormal state, and the measurement target 5 is required to be repaired. Judging state monitoring. Further, as illustrated in Fig. 5, when the fault is predicted, the acoustic sensing 11, 12, and 13 are arranged; when the fault is detected, the acoustic sense 113171.doc 1294514, 13, 14, and 15 can be set. Fig. 9 is a diagram showing an acoustic analysis diagnostic apparatus according to another embodiment of the present invention. (A) is a conceptual diagram showing an example of an acoustic sensor configuration in a factory, and (B) is a data receiving and analyzing diagnostic circuit. In the factory 6A of Fig. 9(A), although not shown, a plurality of devices having a plurality of failure modes are provided. Therefore, there are a number of abnormal noises in the factory. Therefore, in the present embodiment, a plurality of acoustic sensors 11 to 15 are disposed in the factory 6 to detect abnormal sounds. Further, although the illustration in Fig. 9(4) is omitted, the data collecting and transmitting circuit 30a shown in Fig. 7 has received the output of each of the acoustic sensors. The data receiving and analyzing diagnostic circuit shown in FIG. 8(B) is disposed outside the factory 60. The bedding receiving and analyzing diagnostic circuit 40 detects each sound by the acoustic sensor 1 and thus detects each sound. (1) The heart of the material, the equipment of the device is normal or abnormal, and it is necessary to monitor the status. Fig. 10 is a view showing an acoustic analysis diagnostic device of another aspect of the present day and the month; (A) is an example of the arrangement of an acoustic sensor displayed in the intrinsic port. In the figure, the data receiving and analyzing diagnostic circuit is displayed. : As shown in Fig. 1A (A), the intrinsic device η is arranged in the work area, and each of the tables 71 to 74 emits noise. In ^ ^ device ^ ~ 14. The data μ Α has an acoustic sensation, and the diagnosing and diagnosing circuit 40 detects each device 71 according to the 收集 寸 nk nk nk nk 对应 对应 corresponding to each of the acoustic sensors 11 to 14 ~74 of the shell ^ password work palm, can be used for the diagnosis of positive, hanging or feature extraction. Π3171 .doc • 16 · 1294514 As described above, the embodiment of the present invention will be described with reference to the drawings, but the present invention is not limited to the embodiment shown in the drawings. Modifications and variations are possible within the scope of the invention or the equivalents of the invention. [Effects of the Invention] According to the present invention, the first acoustic detecting element including the first characteristic sound of Α φ, ΒΪ + and the measurement target is extracted and the measurement is performed on ^ ^ ^ ^ ^ ^ ^ β β The second acoustic detecting element of the sound can be visually recognized by the characteristics of the measurement target, and the sound emitted by the measurement target can be immediately analyzed with an earlier structure. At the same time, by resonating with different sounds, sounds included in a wide frequency band and different frequency bands are detected, and audio signals of different frequency bands output by a plurality of acoustic detecting elements are analyzed, and the distribution of sounds is diagnosed. By analyzing the characteristics of the object, the sound emitted by the measurement object is instantly analyzed. In addition, 'a plurality of acoustic detection readings are prepared to generate resonances with different sounds, and the plurality of acoustic detecting elements are placed in the vicinity of the measurement target, and the data of the measurement target is collected according to the purpose, and according to the received (four) (four), the plurality of acoustic detections In the element, a combination of acoustic detecting elements that determine the resonance of sound in a specific frequency band is used, whereby an acoustic detecting element that analyzes the sound generated by the measuring object can be manufactured. [Industrial Applicability] The acoustic sensing device, the acoustic analysis and diagnostic device, and the method of manufacturing the acoustic sensor of the present invention can be used to determine whether or not there is an abnormality in the device or equipment in the obstacle. 113171.doc -17- 1294514 BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a plan view showing an example of an acoustic sensing device according to an embodiment of the present invention; FIG. 2 is a view showing the acoustic sensing device shown in FIG. Figure 3 is a diagram showing the output and input characteristics of the acoustic sensing device; Figure 4 is a diagram showing the environmental noise tolerance in the acoustic sensing device; φ Figure 5 (A (B) is a block diagram of an acoustic analysis diagnostic apparatus in an embodiment of the present invention; and FIGS. 6(A) and (B) are diagrams for explaining the operation of the acoustic analysis diagnostic apparatus in an embodiment of the present invention; FIG. 7(A) and FIG. 7(B) are diagrams showing an example of analyzing the sound of the diagnosis target in the entire frequency band by using the acoustic analysis diagnostic apparatus of the present invention; FIG. 8(A) and FIG. A diagram showing an example of the most appropriate acoustic sensor in the acoustic analysis diagnostic apparatus shown in FIG. 7 is shown in FIG. 7; FIG. 9(A) and (B) show the concept of an example in which an acoustic sensor is disposed in the factory. Figure 10 (A), (B) is a diagram showing an example of arranging an acoustic sensor in an intrinsic device.[Main component symbol description] 1 Acoustic sensing device 2 Diaphragm 3 Transverse plate 4 Terminating plate 113171.doc -18- 1294514 5 Spectral plate 10 Semiconductor 矽 substrate 11 ～ 15 Acoustic sensor 21 ～ 25 Data conversion circuit 30, 30a Data collection and transmission circuit 31 Control unit 32 Data transmission circuit 33, 41 Antenna 40 Data reception and analysis diagnostic circuit 42 Data reception circuit 43 Analysis diagnostic circuit 44 Display unit 50 Measurement target 60 Process 71 to 74 Device 113171.doc 19-