[go: up one dir, main page]

JP2009257933A - Magnetic field measuring device, nondestructive inspection device, and magnetic field measurement signal processing method - Google Patents

Magnetic field measuring device, nondestructive inspection device, and magnetic field measurement signal processing method Download PDF

Info

Publication number
JP2009257933A
JP2009257933A JP2008107460A JP2008107460A JP2009257933A JP 2009257933 A JP2009257933 A JP 2009257933A JP 2008107460 A JP2008107460 A JP 2008107460A JP 2008107460 A JP2008107460 A JP 2008107460A JP 2009257933 A JP2009257933 A JP 2009257933A
Authority
JP
Japan
Prior art keywords
magnetic field
signal
measurement
magnetic sensor
magnetic
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
JP2008107460A
Other languages
Japanese (ja)
Other versions
JP5403940B2 (en
Inventor
Yoshito Fukumoto
吉人 福本
Takayuki Hiekata
孝之 稗方
Toshiaki Shimoda
敏章 下田
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Kobe Steel Ltd
Original Assignee
Kobe Steel Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Kobe Steel Ltd filed Critical Kobe Steel Ltd
Priority to JP2008107460A priority Critical patent/JP5403940B2/en
Publication of JP2009257933A publication Critical patent/JP2009257933A/en
Application granted granted Critical
Publication of JP5403940B2 publication Critical patent/JP5403940B2/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Measuring Magnetic Variables (AREA)
  • Investigating Or Analyzing Materials By The Use Of Magnetic Means (AREA)

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a measurement signal of a magnetic field with a high S/N ratio without requiring troublesome and elaborate adjustment when measuring an AC magnetic field in a test object using a high-sensitivity magnetic sensor. <P>SOLUTION: The magnetic field measuring device includes: a main SQUID magnetic sensor Am having a pickup coil A1m disposed opposite the test object 1; a sub SQUID magnetic sensor As including another pickup coil A1s juxtaposed with the pickup coil A1m of the main SQUID magnetic sensor Am; and a magnetic field measurement signal processing device B performing blind signal source separation processing based on an independent component analysis method on the basis of detection signals Vm, Vs of both the sensors and deriving a measurement signal of an AC magnetic field in the test object 1 on the basis of the separation signal thus obtained. <P>COPYRIGHT: (C)2010,JPO&amp;INPIT

Description

本発明は,交流磁場を検出する磁気センサを備えた磁場測定装置及びそれを含む非破壊検査装置,並びに磁場測定信号処理方法に関するものである。   The present invention relates to a magnetic field measuring apparatus including a magnetic sensor for detecting an alternating magnetic field, a nondestructive inspection apparatus including the same, and a magnetic field measurement signal processing method.

近年,発電・製造・輸送・建築等の多くの産業分野で,非破壊検査の重要性が高まっている。例えば,発電所の構造物や橋梁設備等の被検体について,その表面や内部に存在する亀裂や損傷部,ボイド,組成が他と特に異なっている部分等の特異部を,それが未だ微小である段階において外部から非破壊で検出することは,安全性確保のために非常に重要である。従来,構造物の非破壊検査では,X線探傷や超音波探傷,渦電流探傷等の探傷法が採用されることが多い。
また,人体等の被検体内における微小な特異部を高感度で検査する場合に,SQUID磁気センサが用いられることがある。SQUID磁気センサは,超伝導量子干渉素子(SQUID:Superconducting Quantum Interference Device)を備えた磁気センサである。このSQUID磁気センサは,超伝導体による量子干渉効果を利用し,例えば,地磁気の1億分の1(1ピコテスラ)以下の磁場変化を検出できる超高感度の磁気センサである。そのため,SQUID磁気センサは,被検体内のごく微小な特異部を検出したい場合の非破壊検査に採用され得る。なお,磁気センサが被検体の検査に用いられる場合,一般に,被検体の測定部位ごとに得られる磁気センサの検出信号の相対評価により被検体の検査(欠陥有無の検査等)が行われる。従って,非破壊検査等における磁気センサの検出信号は,検査対象の種類や測定環境等に応じてその絶対値が異なることについては特に問題とならない。
In recent years, the importance of non-destructive inspection has increased in many industrial fields such as power generation, manufacturing, transportation, and construction. For example, for specimens such as power plant structures and bridge facilities, the surface and interior of cracks, damaged parts, voids, and parts that are particularly different in composition, etc., are still very small. Non-destructive detection from the outside at a certain stage is very important for ensuring safety. Conventionally, in non-destructive inspection of structures, flaw detection methods such as X-ray flaw detection, ultrasonic flaw detection, and eddy current flaw detection are often employed.
In addition, a SQUID magnetic sensor may be used when a minute singular part in a subject such as a human body is examined with high sensitivity. The SQUID magnetic sensor is a magnetic sensor including a superconducting quantum interference device (SQUID). This SQUID magnetic sensor is a super-sensitive magnetic sensor that can detect a change in magnetic field of, for example, one hundred millionth of the geomagnetism (one picotela) or less using the quantum interference effect of a superconductor. Therefore, the SQUID magnetic sensor can be employed for nondestructive inspection when it is desired to detect a very small singular part in a subject. When the magnetic sensor is used for the inspection of the subject, the inspection of the subject (inspection for the presence or absence of defects, etc.) is generally performed by relative evaluation of the detection signal of the magnetic sensor obtained for each measurement site of the subject. Therefore, the detection signal of the magnetic sensor in the non-destructive inspection or the like does not particularly pose a problem in that the absolute value varies depending on the type of inspection object and the measurement environment.

図5は,SQUID磁気センサの概略構成図である。なお,特許文献1に,図5に示されるものと同様のSQUID磁気センサが示されている。
図5に示されるように,SQUID磁気センサは,例えば,ピックアップコイルA1と,インプットコイルA2と,超伝導量子干渉素子A3(SQUID)と,磁束固定ループ(FLL:Flux Lock Loop)回路A4とを備えている。
前記SQUID(A3)は,ジョセフソン接合により接合されて微小なループ(超伝導ループ)を形成する超伝導体を含むデバイスである。このSQUID(A3)は,液体窒素等によって超伝導特性が得られる温度に冷却される。
前記ピックアップコイルA1が,被検体に生じている磁場を受けて誘導電流を生じさせ,そのピックアップコイルA1に接続された前記インプットコイルA2が,前記SQUID(A3)の超伝導ループに磁場を誘導する。これにより,前記ピックアップコイルA1付近の磁場の強さに応じて,前記超伝導ループに磁場が誘導される。
そして,前記SQUID(A3)においてジョセフソン接合された第1の超伝導体から第2の超伝導体へバイアス電流Ibが供給されると,2つの超伝導体の間に,前記インプットコイルA2によって誘導された磁場の強さに応じた電圧Vsが生じる。即ち,電圧Vsは,前記ピックアップコイルA1付近の磁場の強さに応じたレベルとなる。
前記FLL回路A4は,前記SQUID(A3)における電圧Vsの信号をアンプにより増幅し,さらに積分器により積分して得られる磁場の検出信号を出力する。同時に,前記FLL回路A4は,前記超伝導ループに磁場を誘導するフィードバックコイルに対し,前記検出信号のレベルに応じたフィードバック電流Ifを供給する。このFLL回路A4の作用により,前記SQUID(A3)が一定の磁場動作点で動作し,前記SQUID(A3)において誘導される磁場の強さに対して線形な前記検出信号のレベル(電圧)が得られる。
FIG. 5 is a schematic configuration diagram of the SQUID magnetic sensor. Patent Document 1 discloses a SQUID magnetic sensor similar to that shown in FIG.
As shown in FIG. 5, the SQUID magnetic sensor includes, for example, a pickup coil A1, an input coil A2, a superconducting quantum interference device A3 (SQUID), and a flux lock loop (FLL) circuit A4. I have.
The SQUID (A3) is a device including a superconductor that is joined by a Josephson junction to form a minute loop (superconducting loop). The SQUID (A3) is cooled to a temperature at which superconducting characteristics can be obtained with liquid nitrogen or the like.
The pickup coil A1 receives the magnetic field generated in the subject to generate an induced current, and the input coil A2 connected to the pickup coil A1 induces a magnetic field in the superconducting loop of the SQUID (A3). . Thereby, a magnetic field is induced in the superconducting loop according to the strength of the magnetic field in the vicinity of the pickup coil A1.
Then, when the bias current Ib is supplied from the first superconductor Josephson junction in the SQUID (A3) to the second superconductor, the input coil A2 causes the bias current Ib to pass between the two superconductors. A voltage Vs corresponding to the strength of the induced magnetic field is generated. That is, the voltage Vs has a level corresponding to the strength of the magnetic field near the pickup coil A1.
The FLL circuit A4 amplifies the signal of the voltage Vs in the SQUID (A3) with an amplifier, and further outputs a magnetic field detection signal obtained by integrating with an integrator. At the same time, the FLL circuit A4 supplies a feedback current If corresponding to the level of the detection signal to a feedback coil that induces a magnetic field in the superconducting loop. Due to the action of the FLL circuit A4, the SQUID (A3) operates at a constant magnetic field operating point, and the level (voltage) of the detection signal linear with respect to the strength of the magnetic field induced in the SQUID (A3) is can get.

ところで,SQUID磁気センサは,超高感度であるがゆえに,磁気シールドがなされない環境下で使用された場合には,磁気ノイズの影響を強く受けるという問題点があった。例えば,SQUID磁気センサは,屋外で使用された場合,前記FLL回路A4が正常に動作しなくなったり,前記検出信号が環境ノイズに埋没したりすることにより,正常な検出信号が得られないことが多い。
そのような問題点に対応するため,特許文献2には,測定用のSQUID磁気センサの検出信号と,ノイズキャンセル用のSQUID磁気センサの検出信号との差分信号を,測定対象の磁場の測定信号とする非破壊検査装置が示されている。
特開2005−265780号公報 特開2007−132923号公報 猿渡洋,「アレー信号処理を用いたブラインド音源分離の基礎」,電子情報通信学会技術報告,vol.EA2001-7,pp.49-56,April 2001. 高谷智哉他,「SIMOモデルに基づくICAを用いた高忠実度なブラインド音源分離」,電子情報通信学会技術報告,vol.US2002-87,EA2002-108,January 2003.
By the way, the SQUID magnetic sensor has a problem that it is strongly affected by magnetic noise when used in an environment where a magnetic shield is not used because of its ultra-high sensitivity. For example, when the SQUID magnetic sensor is used outdoors, a normal detection signal may not be obtained because the FLL circuit A4 does not operate normally or the detection signal is buried in environmental noise. Many.
In order to deal with such problems, Patent Document 2 discloses a difference signal between a detection signal of a SQUID magnetic sensor for measurement and a detection signal of a SQUID magnetic sensor for noise cancellation, as a measurement signal of a magnetic field to be measured. A nondestructive inspection device is shown.
JP 2005-265780 A JP 2007-132923 A Hiroshi Saruwatari, “Basics of Blind Sound Source Separation Using Array Signal Processing”, IEICE Technical Report, vol.EA2001-7, pp.49-56, April 2001. Tomoya Takatani et al., “High fidelity blind source separation using ICA based on SIMO model”, IEICE technical report, vol.US2002-87, EA2002-108, January 2003.

特許文献2に示される技術は,2つのSQUID磁気センサの個体差や位置関係が,磁場の測定信号である前記差分信号に大きく影響する。
例えば,2つのSQUID磁気センサにおける磁場の検出ゲイン等の特性に個体差がある場合,その個体差の影響が前記差分信号にそのまま反映される。また,2つのSQUID磁気センサの位置関係が,測定ごとに異なる場合や,測定途中に変化した(位置ずれが生じた)場合も,それらば前記差分信号にそのまま反映される。
しかしながら,複数のSQUID磁気センサの個体差や位置関係を緻密に調整する作業は手間であり,また,その調整が難しいという問題点があった。
In the technique disclosed in Patent Document 2, individual differences and positional relationships between two SQUID magnetic sensors greatly affect the difference signal that is a magnetic field measurement signal.
For example, when there is an individual difference in characteristics such as the detection gain of the magnetic field in two SQUID magnetic sensors, the influence of the individual difference is reflected as it is in the difference signal. In addition, even when the positional relationship between the two SQUID magnetic sensors is different for each measurement, or when the positional relationship is changed during the measurement (positional deviation occurs), the difference signal is reflected as it is.
However, the work of precisely adjusting individual differences and positional relationships among a plurality of SQUID magnetic sensors is troublesome and has the problem that adjustment is difficult.

一方,磁場測定の技術分野とは全く異なる音声信号処理の技術分野において,複数の音源が存在する音響環境下で収音される音響信号から,特定の音源から出力される音響に対応する目的音信号を抽出する技術が開発されている。
所定の音響空間に複数の音源と複数のマイクロホンとが存在する場合,その複数のマイクロホンごとに,複数の音源各々からの個別の音響信号(以下,音源信号という)が重畳された混合信号が入力される。このようにして入力された複数の混合信号のみに基づいて,前記音源信号各々を同定(分離)する信号源分離処理は,ブラインド信号源分離処理(以下,BSS処理という,BSS:Blind Source Separation)と呼ばれる。
さらに,BSS処理の1つに,独立成分分析法(Independent Component Analysis,以下,ICA法という)に基づくBSS処理がある。このICA法に基づくBSS処理は,前記混合信号に含まれる複数の信号どうしが統計的に独立であることを利用して所定の分離行列(逆混合行列)を最適化し,入力された複数の前記混合信号に対して最適化された分離行列によるフィルタ処理を施すことにより,各信号源に対応する信号の同定(信号源分離)を行う処理である。その際,分離行列の最適化は,ある時点で設定されている分離行列を用いたフィルタ処理により同定(分離)された信号(分離信号)に基づいて,逐次計算(学習計算)により以降に用いる分離行列を算出することによって行われる。
ここで,ICA法に基づくBSS処理によれば,分離信号各々は,混合信号の入力数(例えば,マイクロホンの数)と同じ数の出力端(出力チャンネルといってもよい)各々を通じて出力される。このようなICA法に基づくBSS処理は,例えば,非特許文献1や非特許文献2等に詳説されている。
On the other hand, in the audio signal processing technical field, which is completely different from the magnetic field measurement technical field, the target sound corresponding to the sound output from a specific sound source from the acoustic signal collected in the acoustic environment where there are multiple sound sources exists. Techniques for extracting signals have been developed.
When a plurality of sound sources and a plurality of microphones exist in a predetermined acoustic space, a mixed signal in which individual sound signals (hereinafter referred to as sound source signals) from each of the plurality of sound sources are superimposed is input to each of the plurality of microphones. Is done. A signal source separation process for identifying (separating) each of the sound source signals based only on the plurality of mixed signals input in this manner is a blind signal source separation process (hereinafter referred to as a BSS process, BSS: Blind Source Separation). Called.
Further, as one of the BSS processes, there is a BSS process based on an independent component analysis method (hereinafter referred to as ICA method). The BSS processing based on the ICA method optimizes a predetermined separation matrix (inverse mixing matrix) by using the fact that a plurality of signals included in the mixed signal are statistically independent, This is a process for identifying a signal corresponding to each signal source (signal source separation) by performing a filtering process using an optimized separation matrix on the mixed signal. At that time, the optimization of the separation matrix is used later by sequential calculation (learning calculation) based on the signal (separated signal) identified (separated) by the filter processing using the separation matrix set at a certain time. This is done by calculating a separation matrix.
Here, according to the BSS processing based on the ICA method, each separated signal is output through each output terminal (which may be called an output channel) as many as the number of mixed signals (for example, the number of microphones). . Such BSS processing based on the ICA method is described in detail in, for example, Non-Patent Document 1, Non-Patent Document 2, and the like.

従って,本発明は上記事情に鑑みてなされたものであり,その目的とするところは,SQUID磁気センサに代表される高感度の磁気センサを用いて被検体における磁場(ここでは,交流磁場)を測定する場合に,手間のかかる緻密な調整を要することなく,磁気ノイズの影響の小さな(S/N比の高い)磁場の測定信号を得ることができる磁場測定装置及びそれを備えた非破壊検査装置,並びに磁場測定信号処理方法を提供することにある。   Therefore, the present invention has been made in view of the above circumstances, and the object of the present invention is to generate a magnetic field (here, an alternating magnetic field) in a subject using a highly sensitive magnetic sensor represented by a SQUID magnetic sensor. A magnetic field measuring apparatus capable of obtaining a measurement signal of a magnetic field (high S / N ratio) with a small influence of magnetic noise without requiring elaborate and precise adjustment in measurement, and a nondestructive inspection provided with the same An apparatus and a magnetic field measurement signal processing method are provided.

上記目的を達成するために本発明に係る磁場測定装置は,磁場ピックアップ部を通じて入力される信号に基づいてその磁場ピックアップ部付近の交流磁場(強度が振動する磁場)を検出する磁気センサを複数備える。そして,本発明に係る磁場測定装置は,次の(1)〜(4)に示される各構成要素を備えている。
(1)測定対象に対向配置される前記磁場ピックアップ部を有する前記磁気センサである主磁気センサ。
(2)前記主磁気センサの前記磁場ピックアップ部と並設された他の前記磁場ピックアップ部を有する前記磁気センサである副磁気センサ。なお,この副磁気センサは,1つである場合の他,複数設けられる場合も考えられる。
(3)前記主磁気センサの検出信号と前記副磁気センサの検出信号とに基づいて,独立成分分析法に基づくブラインド信号源分離処理を行う信号源分離手段。
(4)前記信号源分離手段により得られる分離信号に基づいて前記測定対象の磁場の測定信号を導出する測定信号導出手段。
In order to achieve the above object, a magnetic field measurement apparatus according to the present invention includes a plurality of magnetic sensors that detect an alternating magnetic field (a magnetic field whose intensity oscillates) in the vicinity of the magnetic field pickup unit based on a signal input through the magnetic field pickup unit. . And the magnetic field measuring apparatus which concerns on this invention is provided with each component shown by following (1)-(4).
(1) The main magnetic sensor which is the magnetic sensor having the magnetic field pickup unit arranged to face the measurement target.
(2) A secondary magnetic sensor which is the magnetic sensor having the other magnetic field pickup unit provided in parallel with the magnetic field pickup unit of the main magnetic sensor. Note that there may be a case where a plurality of sub magnetic sensors are provided in addition to the case where there is one.
(3) Signal source separation means for performing blind signal source separation processing based on an independent component analysis method based on the detection signal of the main magnetic sensor and the detection signal of the sub magnetic sensor.
(4) Measurement signal deriving means for deriving a measurement signal of the magnetic field to be measured based on the separation signal obtained by the signal source separation means.

本発明において,前記主磁気センサは,主として測定対象における交流磁場を検出するものであるが,その検出信号には,測定対象周辺のノイズ磁場の信号成分も含まれ得る。一方,前記副磁気センサは,主として測定対象周辺のノイズ磁場(背景ノイズ)を検出するものである。なお,前記副磁気センサの検出信号にも,前記測定対象における交流磁場の信号成分が含まれ得る。ここで,測定対象における交流磁場の振動成分と,測定対象周辺のノイズ磁場の振動成分とは,統計的に独立であると想定できる。
そして,前記信号源分離手段は,主/副それぞれの検出信号に基く最適化計算(学習計算)によって分離行列(逆混合行列)を最適化し,最適化された分離行列を用いて,主/副それぞれの検出信号に対してフィルタ処理を実行する。これにより,前記主磁気センサの検出信号から,主として測定対象における交流磁場の振動成分が抽出された分離信号(以下,主分離信号という)が得られる。さらに,前記副磁気センサの検出信号から,主として測定対象周辺のノイズ磁場(背景ノイズ)の振動成分が抽出された分離信号(以下,副分離信号という)が得られる。
従って,前記測定信号導出手段は,前記分離信号に基いて,磁気ノイズの影響の小さな(S/N比の高い)測定対象における磁場の測定信号を導出することができる。
また,本発明に係る磁場測定装置は,ノイズ磁場の影響を受けやすい高感度の磁気センサが用いられる場合,例えば,前記主磁気センサ及び前記副磁気センサが,超伝導量子干渉素子を備えた磁気センサである場合に特に好適である。
In the present invention, the main magnetic sensor mainly detects an alternating magnetic field in a measurement target, but the detection signal may include a signal component of a noise magnetic field around the measurement target. On the other hand, the sub magnetic sensor mainly detects a noise magnetic field (background noise) around the measurement target. Note that the detection signal of the sub magnetic sensor may also include a signal component of the alternating magnetic field in the measurement target. Here, it can be assumed that the vibration component of the alternating magnetic field in the measurement target and the vibration component of the noise magnetic field around the measurement target are statistically independent.
The signal source separation means optimizes the separation matrix (inverse mixing matrix) by optimization calculation (learning calculation) based on the main / sub detection signals, and uses the optimized separation matrix. Filter processing is performed on each detection signal. As a result, a separation signal (hereinafter referred to as a main separation signal) in which the vibration component of the AC magnetic field in the measurement target is extracted is obtained from the detection signal of the main magnetic sensor. Furthermore, a separation signal (hereinafter referred to as a sub-separation signal) from which a vibration component of a noise magnetic field (background noise) around the measurement target is extracted is obtained from the detection signal of the sub-magnetic sensor.
Therefore, the measurement signal deriving unit can derive the measurement signal of the magnetic field in the measurement object with a small influence of magnetic noise (high S / N ratio) based on the separated signal.
In addition, the magnetic field measurement apparatus according to the present invention, when a high-sensitivity magnetic sensor that is easily affected by a noise magnetic field is used, for example, the main magnetic sensor and the sub magnetic sensor are magnetic sensors each including a superconducting quantum interference element. It is particularly suitable when it is a sensor.

前記測定信号導出手段による測定信号の導出方法としては,例えば,前記主磁気センサと前記副磁気センサとが1つずつである場合,前記信号源分離手段により得られる前記主磁気センサに対応する1つの分離信号(前記主分離信号)を測定信号とすることが考えられる。
また,前記主磁気センサの磁場ピックアップ部には,複数のノイズ源からのノイズ磁場が混入することも考えられる。この場合,前記副磁気センサが1つであると,前記信号源分離手段によって複数のノイズ源それぞれに対応する分離信号を分離生成できない。
そこで,前記信号源分離処理手段が,1つの前記主磁気センサの検出信号と複数の前記副磁気センサの検出信号それぞれとの組み合わせごとに前記ブラインド信号源分離処理を行うことが考えられる。この場合,前記測定信号導出手段が,次の(1−1)〜(1−3)のいずれかの処理を実行することが考えられる。
(1−1)前記測定信号導出手段が,前記信号源分離手段により得られる前記主磁気センサに対応する複数の分離信号それぞれのスペクトルから,周波数帯域ごとの信号成分が所定の近似条件を満たすものを抽出することによって前記測定対象の磁場の測定信号を導出する。
(1−2) 前記測定信号導出手段が,前記主磁気センサの検出信号と,前記信号源分離手段により得られる複数の前記副磁気センサに対応する複数の分離信号と,の間でスペクトル減算処理を行うことによって前記測定対象の磁場の測定信号を導出する。
(1−3)前記測定信号導出手段が,前記信号源分離手段により得られる前記主磁気センサに対応する複数の分離信号を合成した信号と,前記信号源分離手段により得られる複数の前記副磁気センサに対応する複数の分離信号と,の間でスペクトル減算処理を行うことによって前記測定対象の磁場の測定信号を導出する。
これら(1−1)〜(1−3)のいずれかに示される測定信号導出手段によれば,複数のノイズ源からのノイズ磁場の振動成分が除かれた測定信号,即ち,S/N比の高い測定信号が得られる。
As a method of deriving a measurement signal by the measurement signal deriving unit, for example, when there is one main magnetic sensor and one sub magnetic sensor, the measurement signal deriving unit corresponds to the main magnetic sensor obtained by the signal source separating unit. One separated signal (the main separated signal) can be considered as a measurement signal.
Further, it is conceivable that noise magnetic fields from a plurality of noise sources are mixed in the magnetic field pickup section of the main magnetic sensor. In this case, if there is only one sub magnetic sensor, the signal source separating means cannot separate and generate separated signals corresponding to a plurality of noise sources.
Accordingly, it is conceivable that the signal source separation processing means performs the blind signal source separation processing for each combination of the detection signal of one main magnetic sensor and each of the detection signals of the plurality of sub magnetic sensors. In this case, it is conceivable that the measurement signal deriving means executes any one of the following processes (1-1) to (1-3).
(1-1) The measurement signal deriving unit has a signal component for each frequency band satisfying a predetermined approximation condition from the spectrum of each of a plurality of separated signals corresponding to the main magnetic sensor obtained by the signal source separating unit. The measurement signal of the magnetic field to be measured is derived by extracting.
(1-2) The measurement signal deriving unit performs spectral subtraction processing between the detection signal of the main magnetic sensor and a plurality of separated signals corresponding to the plurality of sub magnetic sensors obtained by the signal source separating unit. To derive a measurement signal of the magnetic field to be measured.
(1-3) The measurement signal deriving means combines a plurality of separation signals corresponding to the main magnetic sensor obtained by the signal source separation means and a plurality of the sub-magnets obtained by the signal source separation means. A measurement signal of the magnetic field to be measured is derived by performing a spectral subtraction process with a plurality of separated signals corresponding to the sensor.
According to the measurement signal deriving means shown in any one of (1-1) to (1-3), the measurement signal from which the vibration component of the noise magnetic field from a plurality of noise sources is removed, that is, the S / N ratio. A high measurement signal can be obtained.

また,本発明は,測定対象における磁場を測定することにより,その測定対象における特異な部分(亀裂や損傷部,ボイド,組成が他と特に異なっている部分等)の有無を検出する非破壊検査装置として捉えることもできる。
即ち,本発明に係る非破壊検査装置は,次の(2−1)及び(2−2)に示される各構成要素を備えている。
(2−1)前述した本発明に係る磁場測定装置。
(2−2)前記磁場測定装置における磁場ピックアップ部が測定対象に対し走査されることによって得られるその測定対象における磁場の測定信号の分布に基いて,前記測定対象における特異な部分の有無を検出する特異部検出手段。
本発明に係る非破壊検査装置によれば,屋外に存在する測定対象等,磁気シールドがなされていない測定対象における微小な特異部の有無を,ノイズ磁場の影響を受けずに高感度で検査することができる。
また,本発明は,磁場ピックアップ部付近の交流磁場を検出する複数の磁気センサからその検出信号を取得し,その検出信号に基づいて測定対象の磁場の測定信号を導出する処理を,コンピュータによって実行する磁場測定信号処理方法として捉えることもできる。
即ち,本発明に係る磁場測定信号処理方法は,次の(3−1)〜(3−4)に示される各手順をコンピュータにより実行する方法である。
(3−1)前記測定対象に対向配置される前記磁場ピックアップ部を有する前記磁気センサである主磁気センサから検出信号を取得する主検出信号取得手順。
(3−2)前記主磁気センサの前記磁場ピックアップ部と並設された他の前記磁場ピックアップ部を有する前記磁気センサである副磁気センサから検出信号を取得する副検出信号取得手順。
(3−3)前記主磁気センサの検出信号と前記副磁気センサの検出信号とに基づいて,独立成分分析法に基づくブラインド信号源分離処理を行う信号源分離手順。
(3−4)前記信号源分離手順により得られる分離信号に基づいて前記測定対象の磁場の測定信号を導出する測定信号導出手順。
In addition, the present invention is a nondestructive inspection that detects the presence or absence of a specific part (a crack, a damaged part, a void, a part having a different composition from others) by measuring the magnetic field in the measurement object. It can also be understood as a device.
That is, the nondestructive inspection apparatus according to the present invention includes the components shown in the following (2-1) and (2-2).
(2-1) The magnetic field measuring apparatus according to the present invention described above.
(2-2) Based on the distribution of the measurement signal of the magnetic field in the measurement object obtained by scanning the measurement object with the magnetic field pickup unit in the magnetic field measurement device, the presence or absence of a specific portion in the measurement object is detected. A singular part detecting means.
According to the nondestructive inspection apparatus according to the present invention, the presence or absence of a minute singular part in a measurement target that is not magnetically shielded such as a measurement target that exists outdoors is inspected with high sensitivity without being affected by a noise magnetic field. be able to.
In the present invention, the computer acquires a detection signal from a plurality of magnetic sensors that detect an alternating magnetic field near the magnetic pickup unit, and derives a measurement signal of the magnetic field to be measured based on the detection signal by a computer. It can also be understood as a magnetic field measurement signal processing method.
That is, the magnetic field measurement signal processing method according to the present invention is a method in which each procedure shown in the following (3-1) to (3-4) is executed by a computer.
(3-1) A main detection signal acquisition procedure for acquiring a detection signal from a main magnetic sensor which is the magnetic sensor having the magnetic field pickup unit disposed to face the measurement target.
(3-2) A sub-detection signal acquisition procedure for acquiring a detection signal from a sub-magnetic sensor which is the magnetic sensor having the other magnetic field pickup unit arranged in parallel with the magnetic field pickup unit of the main magnetic sensor.
(3-3) A signal source separation procedure for performing blind signal source separation processing based on an independent component analysis method based on the detection signal of the main magnetic sensor and the detection signal of the sub magnetic sensor.
(3-4) A measurement signal deriving procedure for deriving a measurement signal of the magnetic field to be measured based on the separation signal obtained by the signal source separation procedure.

本発明によれば,SQUID磁気センサに代表される高感度の磁気センサを用いて被検体における交流磁場を測定する場合に,手間のかかる緻密な調整を要することなく,磁気ノイズの影響の小さな(S/N比の高い)磁場の測定信号を得ることができる。   According to the present invention, when measuring an alternating magnetic field in a subject using a high-sensitivity magnetic sensor represented by a SQUID magnetic sensor, the influence of magnetic noise is small without requiring elaborate and precise adjustment ( A measurement signal of a magnetic field having a high S / N ratio can be obtained.

以下添付図面を参照しながら,本発明の実施の形態について説明し,本発明の理解に供する。尚,以下の実施の形態は,本発明を具体化した一例であって,本発明の技術的範囲を限定する性格のものではない。
ここに,図1は本発明の実施形態に係る非破壊検査装置Xの概略構成図,図2は本発明の第1実施形態に係る磁場測定装置W1の概略構成を表すブロック図,図3は本発明の第2実施形態に係る磁場測定装置W2の概略構成を表すブロック図,図4は本発明の第3実施形態に係る磁場測定装置W3の概略構成を表すブロック図,図5はSQUID磁気センサの概略構成図,図6はICA法に基づくBSS処理を行う信号源分離装置Zの概略構成を表すブロック図である。
Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings so that the present invention can be understood. The following embodiment is an example embodying the present invention, and does not limit the technical scope of the present invention.
1 is a schematic configuration diagram of a nondestructive inspection apparatus X according to an embodiment of the present invention, FIG. 2 is a block diagram showing a schematic configuration of a magnetic field measuring apparatus W1 according to the first embodiment of the present invention, and FIG. 4 is a block diagram showing a schematic configuration of a magnetic field measurement apparatus W2 according to the second embodiment of the present invention, FIG. 4 is a block diagram showing a schematic configuration of a magnetic field measurement apparatus W3 according to the third embodiment of the present invention, and FIG. 5 is a SQUID magnetism. FIG. 6 is a block diagram showing a schematic configuration of a signal source separation device Z that performs BSS processing based on the ICA method.

まず,図1に示される概略構成図を参照しつつ,本発明の実施形態に係る非破壊検査装置Xの構成について説明する。
非破壊検査装置Xは,磁場測定装置Wを備え,その磁場測定装置Wの磁場ピックアップ部を被検体1に対し走査させて得られる磁場の測定信号の分布から,被検体1の表面や内部における特異な部分(特異部)の有無を検出する装置である。前記被検体1は,屋外に存在する構造物等,磁気シールドがなされていない環境下に存在する測定対象である。また,前記特異部は,前記被検体1の表面や内分に存在する亀裂や損傷部,ボイド,或いは組成(材質)が他と特に異なっている部分等である。
前記被検体1は,その表面や内部において交流磁場が生じているものである。例えば,前記被検体1は,微弱な電流が流れる金属製の配管等であって,その電流の変動や物理的な振動等によって交流磁場が生じているもの等が考えられる。また,前記被検体1が,それ自体が交流磁場を発するものではない場合,前記被検体1に対して交流電流を印加すること,或いは一様な交流磁場を放射すること等により,強制的に前記被検体1内に磁場を生じさせることも考えられる。
交流磁場が生じている被検体1において,その表面や内部に前記特異部が存在すると,その特異部において磁場の乱れ(他の部分と異なる磁場)が生じる。
非破壊検査装置Xは,被検体1における交流磁場の検出信号から,前記特異部で生じる磁場の乱れの有無を検出することにより,前記特異部の有無を検出する装置である。
First, the configuration of the nondestructive inspection apparatus X according to the embodiment of the present invention will be described with reference to the schematic configuration diagram shown in FIG.
The nondestructive inspection apparatus X includes a magnetic field measurement device W, and the distribution of the measurement signal of the magnetic field obtained by scanning the magnetic field pickup unit of the magnetic field measurement device W with respect to the subject 1 It is a device that detects the presence or absence of a specific part (singular part). The subject 1 is a measurement object that exists in an environment where a magnetic shield is not provided, such as a structure existing outdoors. The singular part is a crack, a damaged part, a void, or a part in which the composition (material) is particularly different from the others, etc. existing on the surface or internal part of the subject 1.
The subject 1 has an alternating magnetic field generated on the surface or inside thereof. For example, the subject 1 may be a metal pipe or the like through which a weak current flows, and an AC magnetic field is generated due to fluctuations in the current or physical vibration. Further, when the subject 1 itself does not generate an alternating magnetic field, it is forcibly applied by applying an alternating current to the subject 1 or emitting a uniform alternating magnetic field. It is also conceivable to generate a magnetic field in the subject 1.
In the subject 1 in which an alternating magnetic field is generated, if the singular part is present on the surface or inside thereof, magnetic field disturbance (magnetic field different from other parts) is generated in the singular part.
The nondestructive inspection apparatus X is an apparatus that detects the presence / absence of the singular part by detecting the presence / absence of magnetic field disturbance generated in the singular part from the detection signal of the alternating magnetic field in the subject 1.

図1に示されるように,前記非破壊検査装置Xは,SQUID磁気センサAと,磁場測定信号処理装置Bと,信号解析装置Cと,冷却装置Dと,可動式支持装置Eとを備えている。
前記SQUID磁気センサAには,複数のSQUID磁気センサが含まれ,その1つを主SQUID磁気センサAmと称し,残りの2つを副SQUID磁気センサAsと称する。これら主SQUID磁気センサAm及び副SQUID磁気センサAsは,SQUID(超伝導量子干渉素子)を備えた磁気センサであり,それぞれ図5に示されたものと同じものである。なお,前記SQUID磁気センサAにおけるFLL回路A4(主:A4m,副:A4s)の積分器(図5参照)は,被検体1における交流磁場の交流成分を通過し,その交流成分が前記検出信号に表れるよう,十分に長い時定数が設定されている。
ここで,前記主SQUID磁気センサAmに設けられた磁場ピックアップ部であるピックアップコイルA1mは,被検体1の表面から僅かに離間した位置でその被検体1に対向して配置される。以下,前記主SQUID磁気センサAmにおけるピックアップコイルA1mのことを,主ピックアップコイルA1mと称する。
また,前記副SQUID磁気センサAsに設けられたピックアップコイルA1sは,前記主SQUID磁気センサAmのピックアップコイルA1mと並設されている。以下,前記副SQUID磁気センサAsにおけるピックアップコイルA1sのことを,副ピックアップコイルA1sと称する。
例えば,2つの前記副ピックアップコイルA1sは,前記主ピックアップコイルA1mに対し,その両側の位置に並設される。なお,前記主ピックアップコイルA1mと前記副ピックアップコイルA1sとの相対的な位置関係は固定されている。
前記主ピックアップコイルA1mは,主として被検体1の対向する部分(以下,測定部位という)における交流磁場をピックアップするものであるが,前記測定部位の周辺にノイズ磁場が生じている場合,そのノイズ磁場もピックアップしてしまう。
一方,前記副ピックアップコイルA1sは,主として前記測定部位周辺のノイズ磁場(背景ノイズ)をピックアップするものである。
以下,前記主SQUID磁気センサAmによる被検体1における磁場の検出信号を主検出信号Vm,前記副SQUID磁気センサAsによる磁場の検出信号を副検出信号Vsと称する。これら主検出信号Vm及び副検出信号Vsは,前記磁場測定信号処理装置Bへ伝送される。
As shown in FIG. 1, the nondestructive inspection device X includes a SQUID magnetic sensor A, a magnetic field measurement signal processing device B, a signal analysis device C, a cooling device D, and a movable support device E. Yes.
The SQUID magnetic sensor A includes a plurality of SQUID magnetic sensors, one of which is called a main SQUID magnetic sensor Am, and the other two are called sub-SQUID magnetic sensors As. The main SQUID magnetic sensor Am and the sub SQUID magnetic sensor As are magnetic sensors each having a SQUID (superconducting quantum interference device), and are the same as those shown in FIG. The integrator (see FIG. 5) of the FLL circuit A4 (main: A4m, sub: A4s) in the SQUID magnetic sensor A passes the AC component of the AC magnetic field in the subject 1, and the AC component is the detection signal. A sufficiently long time constant is set so that
Here, a pickup coil A1m, which is a magnetic field pickup section provided in the main SQUID magnetic sensor Am, is arranged to face the subject 1 at a position slightly separated from the surface of the subject 1. Hereinafter, the pickup coil A1m in the main SQUID magnetic sensor Am is referred to as a main pickup coil A1m.
The pickup coil A1s provided in the sub SQUID magnetic sensor As is provided in parallel with the pickup coil A1m of the main SQUID magnetic sensor Am. Hereinafter, the pickup coil A1s in the sub SQUID magnetic sensor As is referred to as a sub pickup coil A1s.
For example, the two sub-pickup coils A1s are juxtaposed at positions on both sides of the main pickup coil A1m. The relative positional relationship between the main pickup coil A1m and the sub pickup coil A1s is fixed.
The main pick-up coil A1m mainly picks up an alternating magnetic field in a portion facing the subject 1 (hereinafter referred to as a measurement site). When a noise magnetic field is generated around the measurement site, the noise magnetic field is generated. Also pick up.
On the other hand, the sub-pickup coil A1s mainly picks up a noise magnetic field (background noise) around the measurement site.
Hereinafter, the detection signal of the magnetic field in the subject 1 by the main SQUID magnetic sensor Am is referred to as a main detection signal Vm, and the detection signal of the magnetic field by the sub SQUID magnetic sensor As is referred to as a sub detection signal Vs. The main detection signal Vm and the sub detection signal Vs are transmitted to the magnetic field measurement signal processing device B.

前記冷却装置Dは,液体窒素が収容された冷却容器D1を備えている。この冷却容器D1の外面の一部に,前記主SQUID磁気センサAm及び前記副SQUID磁気センサAsそれぞれのデバイス部Am3,As3(図5におけるSQUID(A3)に相当)が取り付けられている。
前記冷却容器D1の壁は,内外の断熱性を高めるため,真空層が内包された壁となっている。但し,前記冷却容器D1の壁における,前記デバイス部Am3,As3と接する部分には,そのデバイス部Am3,As3と液体窒素との間で効率的な熱交換が行われるように,熱伝導率の高い伝熱プレートD2が設けられている。
The cooling device D includes a cooling container D1 in which liquid nitrogen is accommodated. Device parts Am3 and As3 (corresponding to SQUID (A3) in FIG. 5) of the main SQUID magnetic sensor Am and the sub SQUID magnetic sensor As are attached to a part of the outer surface of the cooling container D1.
The wall of the cooling container D1 is a wall in which a vacuum layer is included in order to improve the heat insulation inside and outside. However, the portion of the wall of the cooling container D1 in contact with the device portions Am3, As3 has a thermal conductivity so that efficient heat exchange is performed between the device portions Am3, As3 and liquid nitrogen. A high heat transfer plate D2 is provided.

また,前記可動式支持装置Eは,前記冷却装置D及び前記SQUID磁気センサAを支持するとともに,その支持位置を被検体1の表面に沿って移動させるものである。
図1に示される前記可動式支持装置Eは,被検体1の表面に対し一定の間隔で形成されたレール3と,そのレール3に支持された移動装置2と,その移動装置2の前記レール3上の位置を検出する位置センサ4とを備えている。
前記移動装置2は,前記冷却装置D及び前記SQUID磁気センサAを保持しつつ前記レール3に沿って移動することにより,前記SQUID磁気センサAにおける前記主ピックアップコイルA1mを被検体1に対して走査させる。
また,前記位置センサ4は,例えば,前記移動装置2の移動に応じてその移動距離をカウントアップするポジショナ等である。この位置センサ4の検出結果は,前記信号解析装置Cへ伝送される。
The movable support device E supports the cooling device D and the SQUID magnetic sensor A and moves the support position along the surface of the subject 1.
The movable support device E shown in FIG. 1 includes a rail 3 formed at a fixed interval with respect to the surface of the subject 1, a moving device 2 supported by the rail 3, and the rail of the moving device 2. 3 is provided with a position sensor 4 for detecting a position on the position 3.
The moving device 2 scans the subject 1 with the main pickup coil A1m in the SQUID magnetic sensor A by moving along the rail 3 while holding the cooling device D and the SQUID magnetic sensor A. Let
The position sensor 4 is, for example, a positioner that counts up the moving distance according to the movement of the moving device 2. The detection result of the position sensor 4 is transmitted to the signal analysis device C.

前記磁場測定信号処理装置Bは,前記ピックアップコイルA1m,A1s付近の交流磁場を検出する複数のSQUID磁気センサAm,Asそれぞれからその検出信号を取得し,その検出信号に基づいて被検体1における磁場の測定信号を導出する処理を実行する装置である。前記磁場測定信号処理装置Bは,例えば,コンピュータの一例であるDSP(Digital Signal Processor)及びそのDSPにより実行されるプログラムが記憶されたROM,或いはASIC等により具現化される。この場合,そのROMには,前記DSPが,後述する信号源分離処理部10,前記主分離信号合成処理部20及びスペクトル減算処理部31等として機能するために実行するプログラムが予め記憶されている。
前記信号解析装置Cは,前記磁場測定信号処理装置Bから被検体1における磁場の測定信号(以下,測定部磁場信号Sgと称する)と,前記位置センサ4の検出信号(位置情報)とを取得し,それらの信号に基づいて特異部検出処理を実行するコンピュータである。
前記特異部検出処理は,前記磁場測定装置Wにおける前記ピックアップコイルA1mが被検体1に対し走査されることによって得られる前記測定部磁場信号Sgの分布に基いて,被検体1の各位置における特異部の有無を検出する処理である。
例えば,前記信号解析装置Cは,前記測定部位ごとに所定時間内に得られる前記測定部磁場信号Sgの平均値や標準偏差を算出する。そして,前記信号解析装置Cは,ある測定部位について得られた前記測定部磁場信号Sgの平均値又は標準偏差が,その前後の測定部位について得られた同平均値や標準偏差に対し,予め設定された差分以上の差がある場合に,当該測定部位に前記特異部が存在すると判別する。なお,前記特異部検出処理は,本発明における特徴をなす部分ではない。
The magnetic field measurement signal processing device B acquires the detection signals from each of the plurality of SQUID magnetic sensors Am and As that detect the alternating magnetic field in the vicinity of the pickup coils A1m and A1s, and based on the detection signals, the magnetic field in the subject 1 It is an apparatus which performs the process which derives | leads-out the measurement signal. The magnetic field measurement signal processing device B is embodied by, for example, a DSP (Digital Signal Processor) which is an example of a computer, a ROM storing a program executed by the DSP, or an ASIC. In this case, the ROM stores in advance a program executed by the DSP to function as a signal source separation processing unit 10, a main separation signal synthesis processing unit 20, a spectral subtraction processing unit 31, and the like, which will be described later. .
The signal analyzer C acquires a magnetic field measurement signal (hereinafter referred to as a measurement unit magnetic field signal Sg) in the subject 1 and a detection signal (position information) of the position sensor 4 from the magnetic field measurement signal processor B. The computer executes the singular part detection process based on these signals.
The singular part detection process is performed based on a distribution of the measurement part magnetic field signal Sg obtained by scanning the subject 1 with the pickup coil A1m in the magnetic field measuring device W. This is processing for detecting the presence or absence of a part.
For example, the signal analyzer C calculates an average value and a standard deviation of the measurement unit magnetic field signal Sg obtained within a predetermined time for each measurement site. Then, the signal analyzing apparatus C sets the average value or standard deviation of the measurement unit magnetic field signal Sg obtained for a certain measurement site in advance with respect to the same average value or standard deviation obtained for the measurement site before and after that. If there is a difference that is greater than or equal to the calculated difference, it is determined that the singular part is present at the measurement site. Note that the singular part detection process is not a characteristic part of the present invention.

以下,前記非破壊検査装置Xにおいて前記磁場測定装置Wとして採用され得る3つの実施形態である磁場測定装置W1〜W3それぞれについて説明する。
なお,前記磁場測定装置W1〜W3は,前記磁場測定信号処理装置Bの部分のみが異なり,その他の構成は同じである。以下,前記磁場測定装置W1〜W3それぞれが備える前記磁場測定信号処理装置Bを,前記磁場測定信号処理装置B1,B2,B3として区別する。
[磁場測定信号処理装置B1]
まず,図2に示されるブロック図を参照しつつ,第1実施形態に係る磁場測定装置W1が備える磁場測定信号処理装置B1について説明する。
図2に示されるように,前記磁場測定信号処理装置B1は,信号源分離処理部10−1,10−2と,主分離信号合成処理部20と,スペクトル減算処理部31とを備えている。
前記信号源分離処理部10−1,10−2は,前記主SQUID磁気センサAmを通じて得られる前記主検出信号Vmと,複数の前記副SQUID磁気センサAsを通じて得られる前記副検出信号Vsそれぞれとの組み合わせごとに設けられている。
そして,前記信号源分離処理部10−1,10−2は,前記主検出信号Vm及び前記副検出信号Vcに基づいて,ICA法に基づくBSS処理(信号源分離処理)を行う。これにより,前記主SQUID磁気センサAmに対応する第1の分離信号と,前記主SQUID磁気センサAmに対応する第2の分離信号とが得られる。以下,前記第1の分離信号を主分離信号Vm’,前記第2の分離信号を副分離信号Vs’と称する。
なお,各SQUID磁気センサAm,Asと前記信号源分離処理部10−1,10−2との間には,不図示のA/Dコンバータが設けられている。そのA/Dコンバータによってデジタル信号に変換された検出信号が,前記信号源分離処理部10−1,10−2に伝送される。なお,前記A/Dコンバータのサンプリング周期は,測定対象とする交流磁場の周波数に応じて設定される。
ここで,前記信号源分離処理部10−1,10−1は,非特許文献1や非特許文献2に示される独立成分分析法に基づくブラインド信号源分離処理(ICA−BSS処理)を実行するものである。
Hereinafter, each of the magnetic field measuring apparatuses W1 to W3, which are three embodiments that can be employed as the magnetic field measuring apparatus W in the nondestructive inspection apparatus X, will be described.
The magnetic field measurement devices W1 to W3 are different only in the magnetic field measurement signal processing device B, and the other configurations are the same. Hereinafter, the magnetic field measurement signal processing device B included in each of the magnetic field measurement devices W1 to W3 is distinguished as the magnetic field measurement signal processing devices B1, B2, and B3.
[Magnetic field measurement signal processor B1]
First, the magnetic field measurement signal processing device B1 provided in the magnetic field measurement device W1 according to the first embodiment will be described with reference to the block diagram shown in FIG.
As shown in FIG. 2, the magnetic field measurement signal processing device B <b> 1 includes signal source separation processing units 10-1 and 10-2, a main separation signal synthesis processing unit 20, and a spectrum subtraction processing unit 31. .
The signal source separation processing units 10-1 and 10-2 include the main detection signal Vm obtained through the main SQUID magnetic sensor Am and the sub detection signals Vs obtained through the plurality of sub SQUID magnetic sensors As. It is provided for each combination.
The signal source separation processing units 10-1 and 10-2 perform BSS processing (signal source separation processing) based on the ICA method based on the main detection signal Vm and the sub detection signal Vc. As a result, a first separation signal corresponding to the main SQUID magnetic sensor Am and a second separation signal corresponding to the main SQUID magnetic sensor Am are obtained. Hereinafter, the first separation signal is referred to as a main separation signal Vm ′, and the second separation signal is referred to as a sub separation signal Vs ′.
An A / D converter (not shown) is provided between each SQUID magnetic sensor Am, As and the signal source separation processing units 10-1, 10-2. The detection signals converted into digital signals by the A / D converter are transmitted to the signal source separation processing units 10-1 and 10-2. The sampling period of the A / D converter is set according to the frequency of the alternating magnetic field to be measured.
Here, the signal source separation processing units 10-1 and 10-1 execute blind signal source separation processing (ICA-BSS processing) based on the independent component analysis method shown in Non-Patent Document 1 and Non-Patent Document 2. Is.

以下,図6に示すブロック図を参照しつつ,前記信号源分離処理部10−1,10−2として採用可能な信号源分離装置Zについて説明する。図6に示される前記信号源分離装置Zは,ICA−BSS処理の一例であるFDICA方式(Frequency-Domain ICA)の信号源分離処理を行うものである。
FDICA方式の信号源分離処理では,まず,時系列信号である入力信号x(t)について,ST−DFT処理部13によって所定の周期ごとに区分された信号であるフレーム毎に短時間離散フーリエ変換(Short Time Discrete Fourier Transform,以下,ST−DFT処理という)を行い,観測信号の短時間分析を行う。なお,前記入力信号x(t)は,前記主検出信号Vm及び前記副検出信号Vsに相当する。
そして,そのST−DFT処理後の各チャンネルの信号(各周波数成分の信号)について,分離演算処理部11fにより分離行列W(f)に基づくフィルタ処理(分離演算処理)を施すことによって信号源分離(音源信号の同定)を行う。ここでfを周波数ビン,mを分析フレーム番号とすると,分離信号(同定信号)y(f,m)は,次の(1)式のように表すことができる。

Figure 2009257933
この(1)式からわかるように,前記フィルタ処理は,周波数ビンごとに行われる。
ここで,分離行列W(f)を学習して更新する処理を表す式は,例えば次の(2)式のように表すことができる。
Figure 2009257933
このFDICA方式の信号源分離処理によれば,信号源分離処理が各狭帯域における瞬時混合問題として取り扱われ,比較的簡単かつ安定に分離行列W(f)を更新することができる。
図6において,前記主検出信号Vmに対応する分離信号y1(f)が前記主分離信号Vm’である。また,前記副検出信号Vsに対応する分離信号y2(f)が前記副分離信号Vs’である。但し,分離信号y1(f),y2(f),即ち,前記主分離信号Vm’及び前記副分離信号Vs’は,周波数領域の信号である。
また,(1)式に基づく分離演算処理(フィルタ処理)は,前記検出信号Vm,Vsが所定周期で区分されたフレーム信号ごとに行われる。このフィルタ処理は,演算負荷の小さな処理であり,実用的なプロセッサによってリアルタイムでの処理を実現できる。
また,前記分離行列W(f)を求めるための(2)式に基づく学習計算(逐次計算)も,前記フィルタ処理と並行して行われる。
前記信号源分離処理部10−1,10−2を構成するプロセッサは,前記分離演算処理をリアルタイムで行いつつ,空き時間において前記学習計算を随時行う。これにより,リアルタイムでの信号源分離処理が行われる。
なお,図6においては,入力信号x1,x2のチャンネル数(即ち,SQUID磁気センサの数)が2つである例について示しているが,(チャンネル数n)≧(信号源の数m)であれば,3チャンネル以上であっても同様の構成により実現できる。 Hereinafter, the signal source separation device Z that can be employed as the signal source separation processing units 10-1 and 10-2 will be described with reference to the block diagram shown in FIG. The signal source separation device Z shown in FIG. 6 performs FDICA (Frequency-Domain ICA) signal source separation processing, which is an example of ICA-BSS processing.
In the signal source separation process of the FDICA method, first, a short-time discrete Fourier transform is performed for each frame that is a signal divided by the ST-DFT processing unit 13 with respect to an input signal x (t) that is a time-series signal. (Short Time Discrete Fourier Transform, hereinafter referred to as ST-DFT processing) and short-time analysis of the observation signal. The input signal x (t) corresponds to the main detection signal Vm and the sub detection signal Vs.
Then, the signal processing of each channel (signal of each frequency component) after the ST-DFT processing is performed by performing filtering processing (separation calculation processing) based on the separation matrix W (f) by the separation calculation processing unit 11f. (Sound source signal identification) is performed. Here, if f is a frequency bin and m is an analysis frame number, the separated signal (identification signal) y (f, m) can be expressed as the following equation (1).
Figure 2009257933
As can be seen from the equation (1), the filtering process is performed for each frequency bin.
Here, an expression representing the process of learning and updating the separation matrix W (f) can be expressed as, for example, the following expression (2).
Figure 2009257933
According to the FDICA signal source separation process, the signal source separation process is handled as an instantaneous mixing problem in each narrow band, and the separation matrix W (f) can be updated relatively easily and stably.
In FIG. 6, the separation signal y1 (f) corresponding to the main detection signal Vm is the main separation signal Vm ′. Further, the separation signal y2 (f) corresponding to the sub detection signal Vs is the sub separation signal Vs ′. However, the separation signals y1 (f) and y2 (f), that is, the main separation signal Vm ′ and the sub-separation signal Vs ′ are signals in the frequency domain.
Further, the separation calculation process (filter process) based on the equation (1) is performed for each frame signal obtained by dividing the detection signals Vm and Vs at a predetermined period. This filter processing is a processing with a small calculation load, and real-time processing can be realized by a practical processor.
Further, learning calculation (sequential calculation) based on equation (2) for obtaining the separation matrix W (f) is also performed in parallel with the filter processing.
The processors constituting the signal source separation processing units 10-1 and 10-2 perform the learning calculation as needed in the idle time while performing the separation calculation processing in real time. Thereby, signal source separation processing is performed in real time.
FIG. 6 shows an example in which the number of channels of the input signals x1 and x2 (that is, the number of SQUID magnetic sensors) is two, but (number of channels n) ≧ (number of signal sources m). If there are three or more channels, the same configuration can be realized.

また,前記磁場測定信号処理装置W1において,前記主分離信号合成処理部20は,前記信号源分離処理部10−1,10−2それぞれにより得られる複数の前記主分離信号Vm’の合成処理を実行し,それにより得られる合成信号を出力するものである。
例えば,前記主分離信号合成処理部20は,複数の前記主分離信号Vm’について,複数に区分された周波数成分(周波数ビン)ごとに平均処理や加重平均処理を実行すること等により,それら主分離信号Vm’を合成する。
また,前記スペクトル減算処理部31は,前記主分離信号合成処理部20により得られた合成信号と,前記信号源分離処理部10−1,10−2それぞれにより分離生成された複数の前記副分離信号Vs’との間でスペクトル減算処理を行う。さらに,前記磁場測定信号処理装置B1は,不図示の逆フーリエ変換処理部により,前記スペクトル減算処理によって得られた周波数領域の信号を時間領域の信号に変換し,変換後の信号を前記測定部磁場信号Sgとして出力する。
このように,前記主分離信号合成処理部20,前記スペクトル減算処理部31及び不図示の前記逆フーリエ変換処理部は,前記信号源分離処理部10−1,10−2により得られる複数の前記主分離信号Vm’を合成した信号と,同じく前記信号源分離処理部10−1,10−2により得られる複数の前記副分離信号Vs’との間でスペクトル減算処理を行うことにより,前記測定部磁場信号Sgを導出する測定信号導出手段の一例である。
In the magnetic field measurement signal processing device W1, the main separation signal synthesis processing unit 20 performs a synthesis process of the plurality of main separation signals Vm ′ obtained by the signal source separation processing units 10-1 and 10-2. This is executed, and the resultant synthesized signal is output.
For example, the main separated signal synthesis processing unit 20 performs an average process or a weighted average process for each of a plurality of divided frequency components (frequency bins) for the plurality of main separated signals Vm ′. The separation signal Vm ′ is synthesized.
In addition, the spectrum subtraction processing unit 31 includes the composite signal obtained by the main separation signal synthesis processing unit 20 and a plurality of the sub-separations generated by the signal source separation processing units 10-1 and 10-2. Spectral subtraction processing is performed with the signal Vs ′. Further, the magnetic field measurement signal processing device B1 converts a frequency domain signal obtained by the spectral subtraction process into a time domain signal by an inverse Fourier transform processing unit (not shown), and converts the converted signal into the measurement unit. Output as magnetic field signal Sg.
As described above, the main separation signal synthesis processing unit 20, the spectrum subtraction processing unit 31, and the inverse Fourier transform processing unit (not shown) include a plurality of the signal source separation processing units 10-1 and 10-2. By performing spectral subtraction between the signal obtained by synthesizing the main separation signal Vm ′ and the plurality of sub-separation signals Vs ′ obtained by the signal source separation processing units 10-1 and 10-2, the measurement is performed. It is an example of the measurement signal deriving means for deriving the partial magnetic field signal Sg.

前記信号源分離処理部10−1,10−2により得られる前記主分離信号Vm’は,被検体1の測定部位における磁場の信号成分が抽出されたものであるが,除去しきれずに残ったノイズ成分が含まれることも考えられる。そのノイズ成分は,複数の前記主分離信号Vmそれぞれにおいて共通性のない状態で存在することが多いと考えられる。
そこで,前記主分離信号合成処理部20が,共通性のないノイズ成分を含む複数の前記主分離信号Vm’を合成(例えば,スペクトル合成)することにより,その合成信号は,共通性のある測定部位の磁場の信号成分がノイズ成分よりも強調されたする信号となる。
さらに,前記スペクトル減算処理部31が,前記合成信号のスペクトルから,主としてノイズ成分が抽出された前記副分離信号Vs’のスペクトルを減算することにより,前記合成信号からさらにノイズ成分が除去される。
これにより,非常にS/N比の高い前記測定部磁場信号Sgが得られる。
なお,前記磁場測定信号処理装置B1が,前記主分離信号合成処理部20により得られる前記合成信号を,前記測定部磁場信号Sgとして出力することも考えられる。
The main separation signal Vm ′ obtained by the signal source separation processing units 10-1 and 10-2 is obtained by extracting the signal component of the magnetic field at the measurement site of the subject 1, but remains without being removed. It is also conceivable that a noise component is included. It is considered that the noise component often exists in a state having no commonness in each of the plurality of main separation signals Vm.
Therefore, the main separated signal synthesis processing unit 20 synthesizes a plurality of the main separated signals Vm ′ including noise components having no commonness (for example, spectrum synthesis), so that the synthesized signal has a common measurement. The signal component of the magnetic field at the part is a signal that is more emphasized than the noise component.
Further, the spectrum subtraction processing unit 31 subtracts the spectrum of the sub-separated signal Vs ′ from which the noise component is mainly extracted from the spectrum of the synthesized signal, thereby further removing the noise component from the synthesized signal.
Thereby, the measurement part magnetic field signal Sg having a very high S / N ratio is obtained.
It is also conceivable that the magnetic field measurement signal processing device B1 outputs the synthesized signal obtained by the main separation signal synthesis processing unit 20 as the measurement unit magnetic field signal Sg.

[磁場測定信号処理装置B2]
次に,図3に示されるブロック図を参照しつつ,第2実施形態に係る磁場測定装置W2が備える磁場測定信号処理装置B2について説明する。なお,図3において,磁場測定信号処理装置B2が備える構成要素のうち,前記磁場測定信号処理装置B1が備えるものと同じ処理を実行する構成要素については図2における符号と同じ符号が付されている。
図3に示されるように,前記磁場測定信号処理装置B2は,前記信号源分離処理部10−1,10−2と,スペクトル近似信号抽出処理部32とを備えている。以下,前記磁場測定信号処理装置B2について,前記磁場測定信号処理装置B1と異なる部分についてのみ説明する。
[Magnetic field measurement signal processor B2]
Next, the magnetic field measurement signal processing device B2 provided in the magnetic field measurement device W2 according to the second embodiment will be described with reference to the block diagram shown in FIG. In FIG. 3, among the constituent elements included in the magnetic field measurement signal processing device B <b> 2, constituent elements that perform the same processing as those included in the magnetic field measurement signal processing device B <b> 1 are denoted by the same reference numerals as in FIG. 2. Yes.
As shown in FIG. 3, the magnetic field measurement signal processing device B <b> 2 includes the signal source separation processing units 10-1 and 10-2 and a spectrum approximation signal extraction processing unit 32. Hereinafter, only portions of the magnetic field measurement signal processing device B2 that are different from the magnetic field measurement signal processing device B1 will be described.

前記スペクトル近似信号抽出処理部32は,前記信号源分離処理部10−1,10−2によって分離生成された複数の前記主分離信号Vm’について,複数に区分された周波数帯域(周波数ビン)ごとの信号成分のうち,その信号成分が前記主分離信号Vm’相互間で所定の近似条件を満たすものを抽出する。このように抽出された周波数領域の信号は,不図示の前記逆フーリエ変換処理部によって時間領域の信号に変換され,変換後の信号が,前記測定部磁場信号Sgとして出力される。
例えば,前記スペクトル近似信号抽出処理部32は,複数の前記主分離信号Vm’について,周波数ビンごとにそれらの信号成分のレベル(パワー)を比較し,そのレベルの比や差が予め定められた範囲内にあるという前記近似条件を満たす場合に,それらの信号成分のいずれか1つを選択する,又はそれらの信号成分を合成する(例えば平均値や最小値を算出する)ことによって前記測定部磁場信号Sgを抽出する。
前記信号源分離処理部10−1,10−2により得られる前記主分離信号Vm’は,被検体1の測定部位における磁場の信号成分が抽出されたものであるが,除去しきれずに残ったノイズ成分が残存していることも考えられる。そのノイズ成分は,複数の前記主分離信号Vmそれぞれにおいて共通性のない状態で存在することが多いと考えられる。
そこで,前記スペクトル近似信号抽出処理部32は,共通性のないノイズ成分を含む複数の前記主分離信号Vm’のスペクトルから,共通性のある(近似条件を満たす)信号成分を測定部位の磁場の信号成分であるとして抽出することにより,前記主分離信号Vm’からノイズ成分を除去する。
これにより,S/N比の高い前記測定部磁場信号Sgが得られる。
なお,前記スペクトル近似信号抽出処理部32及び不図示の前記逆フーリエ変換処理部が,複数の前記主分離信号Vm’それぞれのスペクトルから,周波数帯域ごとの信号成分が所定の近似条件を満たすものを抽出することによって前記測定部磁場信号Sgを導出する測定信号導出手段の一例である。
The spectrum approximate signal extraction processing unit 32 is configured for each of a plurality of divided frequency bands (frequency bins) for the plurality of main separation signals Vm ′ separated and generated by the signal source separation processing units 10-1 and 10-2. Are extracted from the main separated signals Vm ′ satisfying a predetermined approximation condition. The frequency domain signal thus extracted is converted into a time domain signal by the inverse Fourier transform processing unit (not shown), and the converted signal is output as the measurement unit magnetic field signal Sg.
For example, the spectrum approximation signal extraction processing unit 32 compares the levels (power) of the signal components for each of the frequency bins with respect to the plurality of main separation signals Vm ′, and the ratio and difference between the levels are determined in advance. When the approximate condition of being within a range is satisfied, the measurement unit is selected by selecting any one of those signal components or combining the signal components (for example, calculating an average value or a minimum value). The magnetic field signal Sg is extracted.
The main separation signal Vm ′ obtained by the signal source separation processing units 10-1 and 10-2 is obtained by extracting the signal component of the magnetic field at the measurement site of the subject 1, but remains without being removed. It is also conceivable that noise components remain. It is considered that the noise component often exists in a state having no commonness in each of the plurality of main separation signals Vm.
Therefore, the spectrum approximate signal extraction processing unit 32 extracts a common signal component (satisfying the approximate condition) from the spectrum of the plurality of main separation signals Vm ′ including a noise component having no commonness. By extracting as a signal component, a noise component is removed from the main separation signal Vm ′.
Thereby, the measurement part magnetic field signal Sg having a high S / N ratio is obtained.
Note that the spectrum approximate signal extraction processing unit 32 and the inverse Fourier transform processing unit (not shown) are such that the signal component for each frequency band satisfies a predetermined approximate condition from the spectrum of each of the plurality of main separation signals Vm ′. It is an example of a measurement signal deriving unit that derives the measurement unit magnetic field signal Sg by extracting.

[磁場測定信号処理装置B2]
次に,図4に示されるブロック図を参照しつつ,第3実施形態に係る磁場測定装置W3が備える磁場測定信号処理装置B3について説明する。なお,図4において,磁場測定信号処理装置B3が備える構成要素のうち,前記磁場測定信号処理装置B1,B2が備えるものと同じ処理を実行する構成要素については図2,3における符号と同じ符号が付されている。
図4に示されるように,前記磁場測定信号処理装置B3は,前記信号源分離処理部10−1,10−2と,スペクトル減算処理部31’とを備えている。以下,前記磁場測定信号処理装置B3について,前記磁場測定信号処理装置B1と異なる部分についてのみ説明する。
前記スペクトル減算処理部31’は,前記主SQUID磁気センサAmを通じて得られる前記主検出信号Vmと,前記信号源分離処理部10−1,10−2により分離生成された複数の前記副分離信号Vs’との間で前述したスペクトル減算処理を行う。このスペクトル減算処理部31’は,処理対象(観測信号)が前記合成信号から前記主検出信号Vmに置き換わったこと以外は,前記磁場測定信号処理装置B1における前記スペクトル減算処理部31と同じものである。
前記スペクトル減算処理部31’は,ノイズ成分を含む前記主検出信号Vmのスペクトルから,ノイズ成分が抽出された信号である前記副分離信号Vs’それぞれのスペクトルを減算することにより,前記主検出信号Vmからノイズ成分を除去する。
これにより,S/N比の高い前記測定部磁場信号Sgが得られる。
なお,前記スペクトル減算処理部31’及び不図示の前記逆フーリエ変換処理部が,前記主検出信号Vmと複数の前記副分離信号Vs’との間でスペクトル減算処理を行うことによって前記測定部磁場信号Sgを導出する測定信号導出手段の一例である。
[Magnetic field measurement signal processor B2]
Next, the magnetic field measurement signal processing device B3 provided in the magnetic field measurement device W3 according to the third embodiment will be described with reference to the block diagram shown in FIG. In FIG. 4, among the components included in the magnetic field measurement signal processing device B3, the components that execute the same processing as that included in the magnetic field measurement signal processing devices B1 and B2 are the same as those in FIGS. Is attached.
As shown in FIG. 4, the magnetic field measurement signal processing device B3 includes the signal source separation processing units 10-1 and 10-2 and a spectrum subtraction processing unit 31 ′. Hereinafter, only portions of the magnetic field measurement signal processing device B3 that are different from the magnetic field measurement signal processing device B1 will be described.
The spectrum subtraction processing unit 31 ′ includes the main detection signal Vm obtained through the main SQUID magnetic sensor Am and a plurality of the sub-separation signals Vs separated and generated by the signal source separation processing units 10-1 and 10-2. The above-mentioned spectrum subtraction process is performed between The spectrum subtraction processing unit 31 ′ is the same as the spectrum subtraction processing unit 31 in the magnetic field measurement signal processing device B1 except that the processing target (observation signal) is replaced by the main detection signal Vm from the synthesized signal. is there.
The spectrum subtraction processing unit 31 ′ subtracts the spectrum of the sub-separation signal Vs ′, which is a signal obtained by extracting a noise component, from the spectrum of the main detection signal Vm including a noise component, thereby obtaining the main detection signal. Remove noise components from Vm.
Thereby, the measurement part magnetic field signal Sg having a high S / N ratio is obtained.
The spectral subtraction processing unit 31 ′ and the unillustrated inverse Fourier transform processing unit perform spectral subtraction processing between the main detection signal Vm and the plurality of sub-separation signals Vs ′, thereby the measurement unit magnetic field. It is an example of the measurement signal deriving means for deriving the signal Sg.

また,ICA法に基づくBSS処理は,入力信号に混在する複数の信号成分の統計的な独立性(相関の低さ)に基づいて信号源分離を行う処理である。そのため,複数の前記ピックアップコイルA1m,A1sの位置関係のずれや,複数の前記SQUID磁気センサAm,Asそれぞれの検出ゲインの個体差等に起因して,入力信号に混在する各信号成分の振幅等に多少のばらつきが生じても,信号源の分離性能への影響は小さい。
その結果,前記磁場測定装置W1〜W3は,複数のSQUID磁気センサAm,Asの感度や前記ピックアップコイルA1m,A1sの位置等について,手間のかかる緻密な調整を要しないロバスト性を有している。
従って,前記磁場測定信号処理装置B1〜B3によれば,高感度のSQUID磁気センサを用いて被検体1における交流磁場を測定する場合に,手間のかかる緻密な調整を要することなく,磁気ノイズの影響の小さな(S/N比の高い)磁場の測定信号Sgを得ることができる。
The BSS process based on the ICA method is a process for performing signal source separation based on statistical independence (low correlation) of a plurality of signal components mixed in an input signal. For this reason, the amplitude of each signal component mixed in the input signal due to the positional relationship between the plurality of pickup coils A1m, A1s, individual differences in the detection gains of the plurality of SQUID magnetic sensors Am, As, etc. Even if some variation occurs, the influence on the separation performance of the signal source is small.
As a result, the magnetic field measuring devices W1 to W3 have robustness that does not require laborious and precise adjustment with respect to the sensitivity of the plurality of SQUID magnetic sensors Am and As, the positions of the pickup coils A1m and A1s, and the like. .
Therefore, according to the magnetic field measurement signal processing devices B1 to B3, when measuring an alternating magnetic field in the subject 1 using a highly sensitive SQUID magnetic sensor, it is possible to reduce magnetic noise without requiring precise adjustment. A measurement signal Sg of a magnetic field having a small influence (high S / N ratio) can be obtained.

また,以上に示した実施形態は,前記副SQUID磁気センサAsが複数設けられた例を示したが,前記主SQUID磁気センサAm及び前記副SQUID磁気センサAsが1つずつ設けられた装置も考えられる。この場合,例えば,前記磁場測定信号処理装置B1(図2参照)が,前記信号源分離処理部10−1により得られる前記主分離信号Vm’を前記測定部磁場信号Sgとして出力することが考えられる。
また,前記磁場測定信号処理装置B1〜B3が,3つ以上の前記副SQUID磁気センサAsを備えた構成も考えられる。この場合,前記信号源分離処理部10−1,10−2の数が増え,前記主分離信号合成処理部20や前記スペクトル減算処理部31,31’,前記スペクトル近似信号抽出処理部32の信号入力数が増えるだけである。
また,本発明には,前記SQUID磁気センサの他,フラックスゲートセンサやピックアップコイル等,他の磁気センサも採用可能である。
なお,以上に示したように,前記信号源分離処理部10−1,10−2や前記スペクトル減算処理部31等が実行する処理は,入力信号が振動信号(交流信号)であることが前提である。このため,本発明の実施形態に係る磁場測定装置W1〜W3は,被検体1において交流磁場が生じていることが前提である。
Moreover, although the embodiment shown above showed the example in which the said sub SQUID magnetic sensor As was provided with two or more, the apparatus provided with the said main SQUID magnetic sensor Am and the said sub SQUID magnetic sensor As one by one is also considered. It is done. In this case, for example, it is considered that the magnetic field measurement signal processing device B1 (see FIG. 2) outputs the main separation signal Vm ′ obtained by the signal source separation processing unit 10-1 as the measurement unit magnetic field signal Sg. It is done.
Further, a configuration in which the magnetic field measurement signal processing devices B1 to B3 include three or more sub-SQUID magnetic sensors As is also conceivable. In this case, the number of the signal source separation processing units 10-1 and 10-2 increases, and the signals of the main separation signal synthesis processing unit 20, the spectrum subtraction processing units 31 and 31 ′, and the spectrum approximate signal extraction processing unit 32 are displayed. It only increases the number of inputs.
In addition to the SQUID magnetic sensor, other magnetic sensors such as a fluxgate sensor and a pickup coil can be used in the present invention.
As described above, the processing executed by the signal source separation processing units 10-1 and 10-2, the spectral subtraction processing unit 31 and the like is based on the premise that the input signal is a vibration signal (AC signal). It is. For this reason, the magnetic field measuring apparatuses W1 to W3 according to the embodiments of the present invention are based on the premise that an alternating magnetic field is generated in the subject 1.

本発明は,交流磁場を検出する磁気センサを備えた磁場測定装置及びそれを含む非破壊検査装置等に利用可能である。   INDUSTRIAL APPLICABILITY The present invention can be used for a magnetic field measuring device including a magnetic sensor that detects an alternating magnetic field, a nondestructive inspection device including the same, and the like.

本発明の実施形態に係る非破壊検査装置Xの概略構成図。The schematic block diagram of the nondestructive inspection apparatus X which concerns on embodiment of this invention. 本発明の第1実施形態に係る磁場測定装置W1の概略構成を表すブロック図。1 is a block diagram illustrating a schematic configuration of a magnetic field measurement apparatus W1 according to a first embodiment of the present invention. 本発明の第2実施形態に係る磁場測定装置W2の概略構成を表すブロック図。The block diagram showing schematic structure of the magnetic field measuring apparatus W2 which concerns on 2nd Embodiment of this invention. 本発明の第3実施形態に係る磁場測定装置W3の概略構成を表すブロック図。The block diagram showing schematic structure of the magnetic field measuring apparatus W3 which concerns on 3rd Embodiment of this invention. SQUID磁気センサの概略構成図。The schematic block diagram of a SQUID magnetic sensor. ICA法に基づくBSS処理を行う信号源分離装置Zの概略構成を表すブロック図。The block diagram showing the schematic structure of the signal source separation apparatus Z which performs the BSS process based on ICA method.

符号の説明Explanation of symbols

X :非破壊検査装置
A :SQUID磁気センサ
Am:主SQUID磁気センサ
As:副SQUID磁気センサ
B :磁場測定信号処理装置
C :信号解析装置
D :冷却装置
E :可動式支持装置
W,W1〜W3:磁場測定装置
10−1,10−2:信号源分離処理部
20:主分離信号合成処理部
31,31’:スペクトル減算処理部
32:スペクトル近似信号抽出処理部
X: Nondestructive inspection device A: SQUID magnetic sensor Am: Main SQUID magnetic sensor As: Sub SQUID magnetic sensor B: Magnetic field measurement signal processing device C: Signal analysis device D: Cooling device E: Movable support devices W, W1 to W3 : Magnetic field measuring apparatus 10-1, 10-2: Signal source separation processing unit 20: Main separation signal synthesis processing unit 31, 31 ′: Spectrum subtraction processing unit 32: Spectrum approximate signal extraction processing unit

Claims (7)

磁場ピックアップ部を通じて入力される信号に基づいて前記磁場ピックアップ部付近の交流磁場を検出する磁気センサを複数備えた磁場測定装置であって,
測定対象に対向配置される前記磁場ピックアップ部を有する前記磁気センサである主磁気センサと,
前記主磁気センサの前記磁場ピックアップ部と並設された他の前記磁場ピックアップ部を有する前記磁気センサである副磁気センサと,
前記主磁気センサの検出信号と前記副磁気センサの検出信号とに基づいて,独立成分分析法に基づくブラインド信号源分離処理を行う信号源分離手段と,
前記信号源分離手段により得られる分離信号に基づいて前記測定対象の磁場の測定信号を導出する測定信号導出手段と,
を具備してなることを特徴とする磁場測定装置。
A magnetic field measuring apparatus comprising a plurality of magnetic sensors for detecting an alternating magnetic field in the vicinity of the magnetic field pickup unit based on a signal input through the magnetic field pickup unit,
A main magnetic sensor which is the magnetic sensor having the magnetic field pick-up unit arranged opposite to a measurement object;
A sub-magnetic sensor that is the magnetic sensor having the other magnetic field pickup unit arranged in parallel with the magnetic field pickup unit of the main magnetic sensor;
Signal source separation means for performing blind signal source separation processing based on an independent component analysis method based on the detection signal of the main magnetic sensor and the detection signal of the sub magnetic sensor;
Measurement signal deriving means for deriving a measurement signal of the magnetic field to be measured based on the separation signal obtained by the signal source separation means;
A magnetic field measuring apparatus comprising:
前記信号源分離処理手段が,1つの前記主磁気センサの検出信号と複数の前記副磁気センサの検出信号それぞれとの組み合わせごとに前記ブラインド信号源分離処理を行い,
前記測定信号導出手段が,前記信号源分離手段により得られる前記主磁気センサに対応する複数の分離信号それぞれのスペクトルから,周波数帯域ごとの信号成分が所定の近似条件を満たすものを抽出することによって前記測定対象の磁場の測定信号を導出してなる請求項1に記載の磁場測定装置。
The signal source separation processing means performs the blind signal source separation processing for each combination of a detection signal of one main magnetic sensor and each of detection signals of the plurality of sub magnetic sensors,
The measurement signal deriving means extracts, from the spectrum of each of the plurality of separated signals corresponding to the main magnetic sensor obtained by the signal source separating means, a signal component for each frequency band that satisfies a predetermined approximation condition; The magnetic field measurement apparatus according to claim 1, wherein a magnetic field measurement signal is derived.
前記信号源分離処理手段が,1つの前記主磁気センサの検出信号と複数の前記副磁気センサの検出信号それぞれとの組み合わせごとに前記ブラインド信号源分離処理を行い,
前記測定信号導出手段が,前記主磁気センサの検出信号と,前記信号源分離手段により得られる複数の前記副磁気センサに対応する複数の分離信号と,の間でスペクトル減算処理を行うことによって前記測定対象の磁場の測定信号を導出してなる請求項1に記載の磁場測定装置。
The signal source separation processing means performs the blind signal source separation processing for each combination of a detection signal of one main magnetic sensor and each of detection signals of the plurality of sub magnetic sensors,
The measurement signal deriving means performs spectral subtraction processing between the detection signal of the main magnetic sensor and a plurality of separated signals corresponding to the plurality of sub magnetic sensors obtained by the signal source separating means. The magnetic field measurement apparatus according to claim 1, wherein a magnetic field measurement signal is derived.
前記信号源分離処理手段が,1つの前記主磁気センサの検出信号と複数の前記副磁気センサの検出信号それぞれとの組み合わせごとに前記ブラインド信号源分離処理を行い,
前記測定信号導出手段が,前記信号源分離手段により得られる前記主磁気センサに対応する複数の分離信号を合成した信号と,前記信号源分離手段により得られる複数の前記副磁気センサに対応する複数の分離信号と,の間でスペクトル減算処理を行うことによって前記測定対象の磁場の測定信号を導出してなる請求項1に記載の磁場測定装置。
The signal source separation processing means performs the blind signal source separation processing for each combination of a detection signal of one main magnetic sensor and each of detection signals of the plurality of sub magnetic sensors,
The measurement signal deriving means combines a plurality of separation signals corresponding to the main magnetic sensor obtained by the signal source separation means, and a plurality of signals corresponding to the plurality of sub magnetic sensors obtained by the signal source separation means. The magnetic field measurement apparatus according to claim 1, wherein a measurement signal of the magnetic field to be measured is derived by performing a spectral subtraction process between the separated signal and the separated signal.
前記主磁気センサ及び前記副磁気センサが,超伝導量子干渉素子を備えた磁気センサである請求項1〜4のいずれかに記載の磁場測定装置。   The magnetic field measurement apparatus according to claim 1, wherein the main magnetic sensor and the sub magnetic sensor are magnetic sensors each including a superconducting quantum interference element. 磁気センサを有する磁場測定装置と,該磁場測定装置における磁場ピックアップ部が測定対象に対し走査されることによって得られる該測定対象における磁場の測定信号の分布に基いて,前記測定対象における特異な部分の有無を検出する特異部検出手段と,を備えた非破壊検査装置であって,
前記磁場測定装置が,請求項1〜5のいずれかに記載の磁場測定装置であることを特徴とする非破壊検査装置。
Based on the distribution of the measurement signal of the magnetic field in the measurement object obtained by scanning the measurement object with the magnetic field pickup unit having the magnetic sensor and the magnetic field pickup unit in the magnetic field measurement apparatus, the unique part in the measurement object A nondestructive inspection device comprising singular part detection means for detecting the presence or absence of
A nondestructive inspection apparatus, wherein the magnetic field measuring apparatus is the magnetic field measuring apparatus according to claim 1.
磁場ピックアップ部を通じて入力される信号に基づいて前記磁場ピックアップ部付近の交流磁場を検出する複数の磁気センサからその検出信号を取得し,該検出信号に基づいて測定対象の磁場の測定信号を導出する処理を,コンピュータによって実行する磁場測定信号処理方法であって,
複数の前記磁場ピックアップ部のうちの前記測定対象に対向配置される第1の磁場ピックアップ部を有する前記磁気センサである主磁気センサから,その検出信号を取得する主検出信号取得手順と,
複数の前記磁場ピックアップ部のうちの前記第1の記磁場ピックアップ部と並設された第2の磁場ピックアップ部を有する前記磁気センサである副磁気センサから検出信号を取得する副検出信号取得手順と,
前記主磁気センサの検出信号と前記副磁気センサの検出信号とに基づいて,独立成分分析法に基づくブラインド信号源分離処理を行う信号源分離手順と,
前記信号源分離手順により得られる分離信号に基づいて前記測定対象の磁場の測定信号を導出する測定信号導出手順と,
をコンピュータにより実行してなることを特徴とする磁場測定信号処理方法。
The detection signals are obtained from a plurality of magnetic sensors that detect an alternating magnetic field in the vicinity of the magnetic field pickup unit based on a signal input through the magnetic field pickup unit, and a measurement signal of the magnetic field to be measured is derived based on the detection signal. A magnetic field measurement signal processing method in which processing is executed by a computer,
A main detection signal acquisition procedure for acquiring a detection signal from a main magnetic sensor, which is the magnetic sensor having the first magnetic field pickup unit arranged to face the measurement object among the plurality of magnetic field pickup units;
A sub-detection signal acquisition procedure for acquiring a detection signal from a sub-magnetic sensor, which is the magnetic sensor, having a second magnetic field pickup unit arranged in parallel with the first magnetic field pickup unit among the plurality of magnetic field pickup units; ,
A signal source separation procedure for performing blind signal source separation processing based on an independent component analysis method based on the detection signal of the main magnetic sensor and the detection signal of the sub magnetic sensor;
A measurement signal derivation procedure for deriving a measurement signal of the magnetic field to be measured based on the separation signal obtained by the signal source separation procedure;
Is executed by a computer. A magnetic field measurement signal processing method comprising:
JP2008107460A 2008-04-17 2008-04-17 Magnetic field measuring device, nondestructive inspection device Expired - Fee Related JP5403940B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2008107460A JP5403940B2 (en) 2008-04-17 2008-04-17 Magnetic field measuring device, nondestructive inspection device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2008107460A JP5403940B2 (en) 2008-04-17 2008-04-17 Magnetic field measuring device, nondestructive inspection device

Publications (2)

Publication Number Publication Date
JP2009257933A true JP2009257933A (en) 2009-11-05
JP5403940B2 JP5403940B2 (en) 2014-01-29

Family

ID=41385539

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2008107460A Expired - Fee Related JP5403940B2 (en) 2008-04-17 2008-04-17 Magnetic field measuring device, nondestructive inspection device

Country Status (1)

Country Link
JP (1) JP5403940B2 (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102445650A (en) * 2011-09-22 2012-05-09 重庆大学 Circuit fault diagnosis method based on blind signal separation algorithm
JP2014006059A (en) * 2012-06-21 2014-01-16 Asahi Kasei Electronics Co Ltd Sensor signal processing device and sensor signal processing method
CN103595468A (en) * 2013-11-25 2014-02-19 国家电网公司 Quantum signal test method of high-voltage overhead power line
KR101799046B1 (en) 2016-04-04 2017-11-17 한국표준과학연구원 Ultra Low Magnetic Field Nuclear Magnetic Resonance Apparatus
CN113705050A (en) * 2021-08-26 2021-11-26 哈尔滨工程大学 Virtual experimental platform for superconducting quantum interferometer teaching
JP2022003322A (en) * 2020-06-23 2022-01-11 株式会社テイエルブイ Wall thickness measurement system, wall thickness measurement method, wall thickness measurement program and wall thickness measurement device
WO2023152949A1 (en) * 2022-02-14 2023-08-17 Tdk株式会社 Signal processing method, signal processing device, and program

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001000407A (en) * 1999-06-21 2001-01-09 Shimadzu Corp Biological signal measurement device
JP2001104269A (en) * 1999-09-14 2001-04-17 Hitachi Ltd Magnetic field measurement device
JP2005091560A (en) * 2003-09-16 2005-04-07 Nissan Motor Co Ltd Method and apparatus for signal separation
JP2005284519A (en) * 2004-03-29 2005-10-13 Koyo Seiko Co Ltd Abnormality diagnosis apparatus
JP2007033825A (en) * 2005-07-26 2007-02-08 Kobe Steel Ltd Device, program, and method for sound source separation

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001000407A (en) * 1999-06-21 2001-01-09 Shimadzu Corp Biological signal measurement device
JP2001104269A (en) * 1999-09-14 2001-04-17 Hitachi Ltd Magnetic field measurement device
JP2005091560A (en) * 2003-09-16 2005-04-07 Nissan Motor Co Ltd Method and apparatus for signal separation
JP2005284519A (en) * 2004-03-29 2005-10-13 Koyo Seiko Co Ltd Abnormality diagnosis apparatus
JP2007033825A (en) * 2005-07-26 2007-02-08 Kobe Steel Ltd Device, program, and method for sound source separation

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102445650A (en) * 2011-09-22 2012-05-09 重庆大学 Circuit fault diagnosis method based on blind signal separation algorithm
JP2014006059A (en) * 2012-06-21 2014-01-16 Asahi Kasei Electronics Co Ltd Sensor signal processing device and sensor signal processing method
CN103595468A (en) * 2013-11-25 2014-02-19 国家电网公司 Quantum signal test method of high-voltage overhead power line
CN103595468B (en) * 2013-11-25 2016-08-17 国家电网公司 A kind of quantum signal method of testing of high ammonia syrings
KR101799046B1 (en) 2016-04-04 2017-11-17 한국표준과학연구원 Ultra Low Magnetic Field Nuclear Magnetic Resonance Apparatus
US11064900B2 (en) 2016-04-04 2021-07-20 Korea Research Institute Of Standards And Science Ultra-low field nuclear magnetic resonance device
JP2022003322A (en) * 2020-06-23 2022-01-11 株式会社テイエルブイ Wall thickness measurement system, wall thickness measurement method, wall thickness measurement program and wall thickness measurement device
JP7403826B2 (en) 2020-06-23 2023-12-25 株式会社テイエルブイ Wall thickness measurement system, wall thickness measurement method, wall thickness measurement program, and wall thickness measurement device
CN113705050A (en) * 2021-08-26 2021-11-26 哈尔滨工程大学 Virtual experimental platform for superconducting quantum interferometer teaching
CN113705050B (en) * 2021-08-26 2023-10-27 哈尔滨工程大学 A virtual experimental platform for superconducting quantum interferometer teaching
WO2023152949A1 (en) * 2022-02-14 2023-08-17 Tdk株式会社 Signal processing method, signal processing device, and program

Also Published As

Publication number Publication date
JP5403940B2 (en) 2014-01-29

Similar Documents

Publication Publication Date Title
JP5403940B2 (en) Magnetic field measuring device, nondestructive inspection device
US7759931B2 (en) Device for measuring magnetic impedance
Amineh et al. Characterization of surface-breaking cracks using one tangential component of magnetic leakage field measurements
US9453817B2 (en) Nondestructive inspection device using alternating magnetic field, and nondestructive inspection method
RU2013154019A (en) DEVICE AND METHOD FOR MAGNETIC DEFECTOSCOPY
EP3441753B1 (en) Inspection device and inspection method
Zhang et al. A new signal processing method for the nondestructive testing of a steel wire rope using a small device
EP3376216B1 (en) Method for eddy-current testing of electrically conductive objects and device for realizing said method
Zhou et al. Imaging damage in steel using a diamond magnetometer
Zhang et al. Mechanism study for directivity of TR probe when applying Eddy current testing to ferro-magnetic structural materials
JP6551885B2 (en) Nondestructive inspection device and nondestructive inspection method
Tsukada et al. Magnetic property mapping system for analyzing three-dimensional magnetic components
JP2009103534A (en) Magnetic measuring device
CN111289606B (en) A magnetic effect stress detection system and method based on the existing magnetism of structural steel
Kreutzbruck et al. Defect detection in thick aircraft samples using HTS SQUID magnetometers
Ricci et al. Magnetic imaging and machine vision NDT for the on-line inspection of stainless steel strips
Gontarz et al. Identification of magnetomechanical phenomena in a degradation process of loaded steel elements
Hatsukade et al. All-round inspection of a pipe based on ultrasonic guided wave testing utilizing magnetostrictive method and HTS-SQUID gradiometer
Chady et al. Fusion of electromagnetic inspection methods for evaluation of stress-loaded steel samples
Psuj et al. Observation of material degradation under fatigue and static loading condition using selected electromagnetic NDT methods
JP2005265780A (en) Non-destructive inspection device using SQUID magnetic sensor
Zhang et al. A stress defect state measurement method based on low-frequency ACMFL excitation and Hall sensor array collection
JP2021001814A (en) Non-destructive inspection method and non-destructive inspection equipment
Xing et al. Design of A Multifunctional Micro-Magnetic Testing Instrument
KR101173760B1 (en) Detection method of eddy current signal of small amplitude

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20110204

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20121227

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20130226

A521 Written amendment

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20130319

A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20131029

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20131029

LAPS Cancellation because of no payment of annual fees