JP3792760B2 - Active silencer - Google Patents

Description

【００３２】
制御係数算出部２２の作用により、信号Ｓ₁ とＳ₂ との和である誤差信号ｅが零に近づき、式(4) により制御係数積和演算部２３の位相ψおよび騒音に対する体積速度比Ａ₂ ／Ａ₁ は式(5) となる。
【００３３】
ｅ＝Ｓ₁ ＋Ｓ₂ ＝０より
Ａ₁ cos φ_f ＋Ａ₂ cos ψ＝０ …(4)
∴ ψ＝φ_f ＋π，Ａ₂ ／Ａ₁ ＝１ …(5)
ここで、式(5) に示す位相ψを制御係数積和演算部２３で正しく算出するには、誤差信号ｅが因果性を満たす必要がある。
【００３４】
したがって、因果性を満たすためには、式(6) に示す制御係数積和演算部２３のデジタルフィルタＨが遅れ位相特性を有する必要があり、遅延回路部２５によりφ_f （＞０）の位相遅れを持たせればよいことになる。その結果、付加音源２４は式(5) で示す位相の付加音を出力することができる。
【００３５】
【数２】

【００３６】
次に、条件２について考える。
条件１により付加音を生成することができても、たとえば、極端な例として、遅延時間を１時間に設定した場合、電気的な因果性の式(6) は満たしていることから、逆位相を生成することは可能であるが、騒音に対して逆位相の音は１時間遅れで出力されることから消音は不可能である。周期的な騒音であれば、１時間遅れでもくり返し同じ振幅の音が伝わるので消音できるが、ランダム音の場合は消音できなくなる。
【００３７】
そこで、電気的な因果性以外に音響的な因果性も考慮する必要がある。
生成した付加音でランダム音波（音源）を打ち消すには、騒音源Ｘと付加音源２４との間の距離による空間音波の因果性の確保が条件となる。
【００３８】
遅延時間分だけ遅れた信号が付加音源２４から出力されても、騒音源Ｘからの騒音が付加音源２４に伝わる時間の方がさらに遅ければ消音が可能となる。
したがって、騒音源Ｘと付加音源２４との間の距離をｄとすると、式(7) が条件２となる。
【００３９】
ｋｄ＞φ_f （条件２） …(7)
条件１は音を創る条件を、条件２は物理的に空間で音を打ち消すことが可能な条件を示している。
【００４０】
最後に、音を打ち消すために音響パワーを低下させる条件３について考える。音響パワーを最小にするための付加音の位相ψ（制御係数積和演算部２３の位相）は、騒音に付加音を加えたときの全音響パワーをＷ₀ とすると、式(8) 、式(9) より、ψ＝πとなる。これは、式(5) とは一致せずに、φ_f だけ位相のずれた音を付加音源２４が出力することになる。
【００４１】
【数３】

【００４２】
ここで、φ_f ＝０とすればψ＝πとなり、パワーが一番低下する最適位相になるが、式(6) により電気的な因果性は満たさなくなるから、実際には騒音に対して逆位相を生成できなくなる。
【００４３】
したがって、最適条件を満たすことはできないが、φ_f だけ位相のずれた付加音を付加音源２４から出力することでパワーを低下させることを考える。
式(10)に音響パワーの低下量を表す式を示す。
騒音に対する付加音の位相を逆位相にできなくても、逆位相から僅かなずれならばパワーは低下する。これが本発明のねらいである。
【００４４】
【数４】

【００４５】
本手法により得られた式(5) の位相、振幅特性を式(10)に代入し、パワーが低下する条件について考えると、式(11)、式(12)となる。

したがって、条件１、２、３の全てを満たすように、騒音源Ｘと付加音源２４との間の距離ｄと、遅延時間φ_f を設定すれば、式(10)にしたがってパワーを低下させることができ、消音効果を得ることができる。
【００４６】
付加音源２４であるスピーカの構造上の遅れ特性を無視できた場合、消音効果としては、騒音源Ｘと付加音源２４との間の距離をｄとし、φ_f ＝π／４としたとき、
５００Ｈｚまで消音させるには３４ｃｍ＞ｄ＞８．５ｃｍ、
３００Ｈｚまで消音させるには５６ｃｍ＞ｄ＞１４ｃｍ、
となる。したがって、騒音源Ｘと付加音源２４との間の距離を上記範囲に設定すれば消音が可能となる。
【００４７】
なお、上記例では付加音源２４から騒音に対して逆位相の付加音を生成するために遅延時間を調整する手法を採用しているが、図２に示すように適応制御により付加音源２４からセンシング部２１までの空間伝達の係数を同定し、この係数を遅延回路２５の代わりに用いるようにしてもよい。
【００４８】
すなわち、信号発生器３２を内蔵させておき、この信号発生器３２でＭ系列信号を発生させ、これで付加音源２４を駆動し、このとき付加音源２４からからセンシング部２１までの空間伝達の係数を時々刻々適応的に制御係数算出部（Ａ−ＦＩＲ）３３で同定する。このとき、同定されたフィルタ係数Ｈ_d は、以下の式により得られる。
【００４９】
ｅ＝Ｍ・Ｇ_AD・Ｃｒ・Ｇ_AD−Ｍ・Ｈ_d ＝０
Ｈ_d ＝Ｇ_AD・Ｇ_DA・Ｃｒ …(13)
Ｇ_AD，Ｇ_DA：ＡＤ変換，ＤＡ変換の遅れ特性
これを図１に示す遅延回路部２５の代りに固定フィルタ係数（積和演算回路）として使用すると、式(6) より、式(14)となる。
【００５０】
【数５】

したがって、φ_AD＋φ_DA＋ψ_cr＞０より因果性は満足するが、誤差信号検出部２６に入る信号Ｓ₁ ，Ｓ₂ はそれぞれ式(15)となる。
【００５１】
【数６】

【００５２】
ｅ＝Ｓ₁ ＋Ｓ₂ ＝０からＡ₁ cos （φ_AD＋φ_DA＋ψ_cr）＋Ａ₂ cos ψ＝０
したがって、騒音に対する付加音の位相（制御係数積和演算部の位相）ψおよび体積速度比Ａ₂ ／Ａ₁ は、
ψ＝φ_AD＋φ_DA＋ψ_cr＋π，Ａ₂ ／Ａ₁ ＝１ …(16)
となり、ψ＝π＋２ｎπ（ｎ＝０，１，２，…）の最適位相からずれる量が大きくなる。φ_AD＋φ_DA＋ψ_crの項は、制御系の特性および空間音響系の特性で決まり、外部からの調整は容易にはできないことから、２ｎπに近づけることはできない。
【００５３】
したがって、同定したＨ_d フィルタ特性をそのまま固定ＦＩＲ係数として、適応制御システム中の遅延回路部に代って用いても、消音効果を向上させることは難しい。
【００５４】
そこで、図３に示すように、遅延要素としてＤＡ変換器３４，ＡＤ変換器３５を付加するとよい。なお、図３中、３６は先にＭ系列信号を使って同定したフィルタである。
【００５５】
ＡＤ変換，ＤＡ変換の遅れ位相に相当した時間だけ遅延回路部を調整してもよいが、システム構成面から考えるとＤＡ変換器３４，ＡＤ変換器３５を一段追加した図３の構成の方がＡＤ変換，ＤＡ変換の位相遅れ分を完全に除去できるので好ましい。
図３に示す構成において、誤差信号検出部２６に入る信号Ｓ₁ ，Ｓ₂ はそれぞれ式(17)となる。
【００５６】
【数７】

【００５７】
ｅ＝Ｓ₁ ＋Ｓ₂ ＝０より、Ａ₁ cos ψ_cr＋Ａ₂ cos ψ＝０
騒音に対する付加音の位相および体積速度比は式(18)となる。
ψ＝ψ_cr＋π，Ａ₂ ／Ａ₁ ＝１ …(18)
式(16)と比べて、ＡＤ，ＤＡ変換時の位相遅れが除去できた形となる。
なお、このときのＨフィルタ特性は
Ｓ・Ｇ_AD・Ｈ_d ＋Ｓ・Ｇ_AD・Ｇ_DA・Ｇ_AD・Ｈ＝０より、
【００５８】
【数８】

となり、ψ_cr＞０より因果性も満足する。
【００５９】
式(18)より、付加音の位相はψ_cr＋πとなり、逆位相とはならない。しかし、ψ_crは、空間経路が自由音場とすると、ψ_cr＝ｋｄとなることから、騒音源Ｘと付加音源２４との間隔ｄを小さくするほど逆位相に近づき、パワーは低下し出す。
このときのパワー低下量（ｄＢ）は、式(10)より、
【００６０】
【数９】

【００６１】
評価用マイクを使用した従来の手法においても、式(10)において、ｋｄ＝０．９９（周波数ｆ＝３００Ｈｚ；騒音源と付加音源との間隔ｄ＝０．１８ｍ）でも十分消音効果を確認していることから、もしψ_crの空間伝達関数が反射成分を含み、直接位相（位相ｋｄ）が実現されなくても、この位相遅れは間隔ｄを小さくすることで対応でき、消音効果の劣化にはつながらない。
【００６２】
ちなみに、ｋｄ＝０．３６（周波数＝２００Ｈｚ；騒音源と付加音源との間隔ｄ＝０．１ｍ）のときのψ_crの有無によるパワー低下量の差（ｄＢ）を見積もると以下となる。
【００６３】
(a) ψ_crを無視できた場合

(b) ψ_crを考慮した場合

したがって、ｆｄ＝１９．４８（ｋｄ＝０．３６より）のとき、
つまり、間隔ｄ＝０．０５ｍでは、約３９０Ｈｚまで、
ｄ＝０．１ｍでは 約１９５Ｈｚまで、
ｄ＝０．２ｍでは 約９７Ｈｚまでは、
約７．７ｄＢパワーが低下する。
【００６４】
ψ_crを無視できる場合と比べて、約６ｄＢ消音効果が劣化するが、それでも低減効果があることから、騒音源の性質によっては応用可能である。
図４には本発明のさらに別の実施形態に係る能動消音装置の概略構成が示されている。
【００６５】
先に説明した例では、付加音源を１個使用して装置を構成しているが、この例では付加音源を２個使用し、消音効果の向上を図っている。
この能動消音装置も、適応制御方式を採用しており、大きく別けて、開放空間に置かれた騒音源Ｘ（ダクト出口または筐体開口部でもよい）から放射された騒音と相関のある信号を検出するセンシング部４１と、制御系の制御係数を時々刻々適応的に算出する制御係数算出部（Ａ−ＦＩＲ）４２と、この制御係数算出部４２で算出された制御係数とセンシング部４１の出力信号との積和演算を行う制御係数積和演算部（ＦＩＲ）４３と、この制御係数積和演算部４３の出力を入力信号として対象とする騒音に音響エネルギを供給する付加音源４４ａ，４４ｂと、センシング部４１から得た信号の位相を遅らせる遅延回路部４５と、騒音と同じ体積速度の付加音を出力できるように制御係数積和演算部４３の出力を調整する演算部４６と、この演算部４６の出力と遅延回路部２５を通過した信号との和を誤差信号ｅとみなして制御係数算出部４２に与える誤差信号検出部４７とで構成されている。
【００６６】
この例の場合も、センシング部４１はマイクを主体に構成されており、また付加音４４ａ，４４ｂはスピーカによって構成されている。センシング部４１と付加音４４ａ，４４ｂとは筐体４８内に設けられている。すなわち、筐体４８には騒音源Ｘからの騒音をセンシング部４１に導く開口部４９が形成されており、また付加音源４４ａ，４４ｂで発生した付加音を外部へ導くための開口部５０ａ，５０ｂが形成されている。さらに、付加音源４４ａ，４４ｂで発生した付加音がセンシング部４１に直接入射してハウリングを起こさないようにするための遮音材５１が両者間に設けてあり、また反射成分の影響を取り除いて騒音源Ｘからセンシング部４１へ直接入射する音の寄与率を高くするために吸音材５２が設けられている。
【００６７】
なお、図中Ｇ_AD、Ｇ_DAは、ＡＤ変換およびＤＡ変化による遅れ要素の存在を示している。
このように、付加音源を２個使用した場合は、付加音源の配置法が問題になる。
【００６８】
２個の付加音源で発生した付加音に騒音を加えたときの全音響パワーＷ₀ は、

となる。なお、θ_i ：各音源ｉの位相、Ｒは放射抵抗、Ｖｉは振動速度である。
【００６９】
ここで、２つの付加音源は同位相θで制御されるものとすると、
θ＝θ₂ −θ₁ ＝θ₃ −θ₁ …(22)
となる。
【００７０】
２つの付加音源４４ａ，４４ｂは、騒音源Ｘから等距離ｄに配置されることを条件とすると、
Ｒ₁₁＝４πｒ₁ ² ρｃｋ² ｒ₁ ²
Ｒ₁₂＝４πｒ₁ ² ρｃｋ² ｒ₂ ² ・sin(ｋｄ）／ｋｄ
Ｒ₁₃＝Ｒ₁₂（∴ｒ₂ ＝ｒ₃ ）
Ｒ₂₃＝４πｒ₂ ² ρｃｋ² ｒ₃ ² ・sin(ｋＬ）／ｋＬ
｜Ｖ₁ ｜＝Ａ₁ ／４πｒ₁ ²
｜Ｖ₂ ｜＝Ａ₂ ／４πｒ₂ ²
｜Ｖ₃ ｜＝｜Ｖ₂ ｜（∴ｒ₂ ＝ｒ₃ ） …(23)
となる。なお、ｒ_i ：振動半径、Ｒ：放射抵抗、Ａ_i ：体積速度、Ｖ_i ：振動速度である。
【００７１】
制御前の騒音のみの音響パワーをＷ₁ とすると、制御後のパワー低下量（ｄＢ）は、

となる。ｄ：音源と付加音源との間隔、Ｌ：２つの付加音源の間隔である。
【００７２】
【数１０】

より、パワーを最小にさせることができる。
【００７３】
付加音源４４ａ，４４ｂは騒音源Ｘを中心とした半径ｄの円周上にあることから、この中で２つの付加音源間隔Ｌが最小となる配置は、図５(a) のように付加音源が隣接する場合である。したがって、この例に係る能動消音装置では、実際には図５(b) 示すように、付加音源４４ａ，４４ｂはセンシング部４１から等距離で隣接して配置されている。
図１に示した構成のように遅延回路部４５を使用したとき、誤差信号検出部４７に入る信号Ｓ₁ ，Ｓ₂ は式(26)となる。
【００７４】
【数１１】

【００７５】
ｅ＝Ｓ₁ ＋Ｓ₂ ＝０より、Ａ₁ cos φ_f ＋２Ａ₂ cos ψ＝０
したがって、騒音に対する付加音の位相および体積速度比は式(27)となる。
ψ＝φ_f ＋π，Ａ₂ ／Ａ₁ ＝１／２ …(27)
ｅ＝Ｓ・Ｇ_AD・Ｆ＋Ｓ・Ｇ_AD・Ｈ＝０より、
制御系フィルタ係数Ｈは式(28)となる。
【００７６】
【数１２】

【００７７】
遅延回路部４５において、φ_f ＝２ｎπ（ｎ＝０，１，２，…）を満たせば、式(27)より付加音は逆位相に近づく。このときの低下量は式(27)を式(24)に代入することで得られる。
【００７８】
なお、上記例では付加音源４４ａ，４４ｂから逆位相の付加音を生成するために遅延時間を調整する手法を採用しているが、図６に示すように適応制御により付加音源４４ａ，４４ｂからセンシング部４１までの空間伝達の係数を同定し、この係数を遅延回路部４５の代わりに用いるようにしてもよい。
【００７９】
すなわち、内蔵された信号発生器３２でＭ系列信号を発生させ、これで付加音源４４ａ，４４ｂを駆動し、このとき付加音源４４ａ，４４ｂからからセンシング部４１までの空間伝達の係数を時々刻々適応的に制御係数算出部（Ａ−ＦＩＲ）６２で同定する。このとき、同定されたフィルタ係数Ｈ_d は、以下の式により得られる。
【００８０】
【数１３】

【００８１】
このＨ_d フィルタを、図４に示す遅延回路部４５の代わりに積和演算回路としてそのまま使用すると、付加音源１個の場合と同じように消音効果が劣化する。したがって、図７に示すように付加音源１個の場合と同様に、ＡＤ変換回路６３，ＤＡ変換回路６４と体積速度係数を調整する演算回路６５とを使用する。
【００８２】
誤差信号検出部４７に入る信号をＳ₁ ，Ｓ₂ とすると、最適な体積速度を出力するためには、Ｓ₂ 信号の前段には４倍の係数を掛けなければいけないことから、Ｓ₁ ，Ｓ₂ は式(30)となる。
【００８３】
【数１４】

【００８４】
ψ＝ψ_cr1 ＋π，Ａ₂ ／Ａ₁ ＝１／２ …(32)
ここで、誤差信号ｅ＝Ｓ・Ｇ_AD・Ｈ_d ＋Ｓ・Ｇ_AD・Ｇ_DA・Ｇ_AD・Ｈ＝０より、
制御系フィルタＨは式(33)となる。
【００８５】
【数１５】

となり、因果性も満たす（∴ψ_cr1 ＞０）。
【００８６】
したがって、付加音源４４ａ，４４ｂから式(32)の位相で出力させることができる。
付加音源４４ａ，４４ｂは逆位相からψ_cr1 ずれた位相を出力することになるが、Ａ₂ ／Ａ₁ ＝１／２となり、また、ψ_cr1 ，ψ_cr2 は空間経路を自由音場とすると、
【００８７】
【数１６】

となる。
【００８８】
ｆ・ｄ＝１９．４８（ｋｄ＝０．３６より）、ｋＬ＝０．３６のとき、

となり、良好な消音効果が得られることになる。
【００８９】
【発明の効果】
以上説明したように、本発明によれば、ダクトのような音響遅延空間を持たない騒音に対しても、騒音源に付加音源を近付け、ランダム騒音に対しても逆位相、同振幅の付加音を生成させることが可能となり、音響パワーの最小化が図れ、騒音源の周囲全空間領域を消音させることができる。
【図面の簡単な説明】
【図１】本発明の第１の実施形態に係る能動消音装置の概略構成図
【図２】本発明の第２の実施形態に係る能動消音装置において適応制御によって遅延時間を同定する場合の構成図
【図３】本発明の第３の実施形態に係る能動消音装置の概略構成図
【図４】本発明の第４の実施形態に係る能動消音装置の概略構成図
【図５】同装置における付加音源の配置を説明するための図
【図６】本発明の第５の実施形態に係る能動消音装置において適応制御によって遅延時間を同定する場合の構成図
【図７】本発明の第６の実施形態に係る能動消音装置の概略構成図
【図８】ダクトの出口から漏れ出す騒音を消音する場合の例を説明するための図
【図９】筐体の開口部から漏れ出す騒音を消音する場合の例を説明するための図
【図１０】従来の手法では消音が困難な例を説明するための図
【図１１】付加音源を２個使用して音響パワーを最小とするための最適配置の説明図
【符号の説明】
２１，４１…センシング部
２２，４２…制御係数算出部（Ａ−ＦＩＲ）
２３，４３…制御係数積和演算部（ＦＩＲ）
２４，４４ａ，４４ｂ…付加音源
２５，４５…遅延回路部
２６，４７…誤差信号検出部
２７，４８…筐体
２８，２９，４９，５０ａ，５０ｂ…開口部
３０，５１…遮音材
３１，５２…吸音材
３２，６１…Ｍ系列信号を発生する信号発生器
３３，６２…制御係数算出部
３６，６６…フィルタ
ｘ…騒音源[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an active silencer used to reduce noise radiated from an opening of a casing, a duct opening, or equipment.
[0002]
[Prior art]
As is well known, various methods for reducing noise using acoustic technology and electrical control technology have been proposed recently.
For example, as shown in FIG. 8 (a), when the noise emitted from the noise source 2 in the duct 1 is prevented from leaking outside from the outlet 3 of the duct 1, normally, as shown in FIG. The method shown in FIG. 8C is adopted.
[0003]
In the method shown in FIG. 8 (b), a sound having an opposite phase to the sound propagated from the noise source 2 is output by the speaker 4, and this sound and the sound propagated from the noise source 2 are caused to interfere with each other. The sound pressure near the outlet 3 is set to zero. When a point with zero sound pressure is created, the sound is reflected at that point due to the difference in acoustic impedance, and the sound does not propagate to the outlet 3 side. The detection microphone 5 is used for detecting the sound of the noise source 2, and the evaluation microphone 6 is used for detecting whether or not a point of zero sound pressure has been formed. The adaptive control system 7 that automatically determines the control coefficient is used for the control to bring the sound pressure close to zero.
[0004]
In the method shown in FIG. 8 (C), a speaker 4 is installed outside the duct 1 and the speaker 2 sucks in the same amount (same volume velocity) as the fluid that has flowed out from the outlet 3 of the duct 1, and the acoustic power of the duct noise. The ambient sound pressure is reduced by minimizing the noise. The phase and amplitude (volume velocity) of the additional sound source for efficiently sucking the sound radiated from the duct 1 into the speaker 4 depends on the positions of the evaluation microphone 6 and the speaker 4 installed near the outlet 3 of the duct 1. It is determined. Therefore, in this method, the evaluation microphone 6 or the like is installed in an optimal arrangement obtained by calculation in advance, and adaptive control is performed by the adaptive control system 7a, so that not only the sound pressure at the installation position of the evaluation microphone 6 but also the surroundings The entire space area is also muted. The feature of this method is that the silencing effect varies depending on the number of speakers and the arrangement method.
[0005]
Further, for example, as shown in FIG. 9 (a), when the noise emitted from the noise source 2 in the housing 11 is prevented from leaking outside through the opening 12 provided in the housing 11. Usually, the method shown in FIG. 9 (b) or FIG. 9 (C) is employed.
[0006]
In the method shown in FIG. 9 (b), the duct 13 is installed inside the casing 11 so that the sound emitted from the noise source 2 propagates one-dimensionally through the duct 13, and this structure is shown in FIG. 8 (b). The sound radiated from the opening 12 is muted by applying the method shown in FIG.
[0007]
In the method shown in FIG. 9 (c), the speaker 4 and the evaluation microphone 6 are installed near the opening 12 in the casing 11, and the method shown in FIG. 8 (c) is applied to mute the sound. .
[0008]
However, the conventional silencing method described above has the following problems.
That is, as shown in FIGS. 10 (a), 10 (b), and 10 (C), when there is no space for installing a duct inside the casing 11, or as shown in FIG. 10 (d), it is covered with a casing or the like. If the noise source 2 is present in the open space, the method of sucking sound and minimizing the sound power is suitable as shown in FIG. 8 (c). However, even if this method is applied, a certain degree of silencing effect can be obtained for periodic sounds, but only a slight silencing effect can be obtained for random noise in the low frequency range.
[0009]
The reason for this will be described below.
In general, when a random noise source is to be silenced using an adaptive control system, the following points must be noted. That is, the sound cannot be silenced if the sound emitted from the speaker as the additional sound source propagates to the evaluation microphone later than the sound emitted from the noise source. This state is called not satisfying the causality, and the control filter coefficient at this time has a leading phase characteristic and cannot be realized by the control circuit. For this reason, it is impossible to lower the sound pressure at the evaluation microphone position. Therefore, in order to reduce the sound pressure at the evaluation microphone position, the sound from the noise source must be delayed and reach the evaluation microphone for the time required to generate the additional sound by the control circuit. A duct as shown in (b) is required. A sound delay is applied to the sound emitted from the noise source by propagating through the duct. This is the same in the adaptive control system that sucks in the sound shown in FIG. 8 (c). If there is no duct acting as an acoustic delay, random noise cannot be silenced.
[0010]
Here, let us consider four types of arrangements for minimizing the sound power when a silencer system that absorbs sound is adopted for a noise source without a duct.
Now, as shown in FIG. 11, the noise source is A, the evaluation microphone is B, the speakers for sucking sound are Za and Zb, the distance between the noise source A and the evaluation microphone B is L, and the speaker Za. , Zb and the evaluation microphone B are R, and the distance between the noise source A and the speakers Za, Zb is D.
[0011]
If the acoustic propagation distance corresponding to the time until the antiphase is generated by the control circuit is S, the following conditional expression must be satisfied in order to satisfy the causality.
LR> S (= 0.4 m) (1)
That is, it is necessary to set LR to 40 cm or more.
[0012]
Therefore, consider the four arrangement patterns shown in FIGS. 11 (a), (b), (c), and (d).
(1) In the case of the arrangement pattern shown in Fig. 11 (a)
From the optimal placement condition L = R = D,
LR = 0
Therefore, conditional expression (1) is not satisfied and random noise cannot be silenced.
[0013]
(2) In the case of the arrangement pattern shown in Fig. 11 (b)
Optimal placement condition R = (L² + D² )^0.5
Since R> L, conditional expression (1) is not satisfied and random noise cannot be silenced.
[0014]
(3) In the case of the arrangement pattern shown in Fig. 11 (c)
Optimal placement condition L = 0.6D
R = DL = 0.4D
From conditional expression (1), LR = 0.2D
In this case, if D> 2 m is satisfied, the causality is satisfied, and random noise can be silenced at the position of the evaluation microphone B. However, the sound power is determined by the distance D between the noise source A and the speakers Za and Zb. Since it gets worse as it gets larger, it can't be expected to produce a significant silencing effect over a wide frequency band when it is 2 m away.
[0015]
(4) In the case of the arrangement pattern shown in Fig. 11 (d)
The arrangement of the noise source A and the speakers Za and Zb is the same as in (3), but the evaluation mask B is installed outside the speaker Zb.
[0016]
From the conditional expression, L−R = D. In this case, if the distance D between the noise source A and the speakers Za and Zb is set to 0.4 m or more, the causality is satisfied and the evaluation microphone B is located. Sound pressure can be reduced. However, the position of the evaluation microphone B at this time deviates from the optimum arrangement, and a significant silencing effect cannot be obtained in a wide frequency band. In addition, if it is limited to a low frequency band, the acoustic power can be reduced.
[0017]
Although the silencing effect was estimated based on the conditions satisfying the causality for the four types of optimal arrangements above, the silencing effect of any arrangement compared to the case where the sound radiated from the duct shown in FIG. Will deteriorate significantly.
[0018]
As described above, the conventional active silencer that absorbs sound and minimizes the sound power has a problem that only the silence of the periodic sound can be realized for the noise source in which the sound source exists in the open space. .
[0019]
[Problems to be solved by the invention]
As described above, in the conventional active silencer that absorbs sound and minimizes the sound power, the closer the speaker, which is an additional sound source, is closer to the noise source, the greater the silencing effect can be obtained. In the case of an object that cannot be brought close to and a noise source exists in an open space due to this, there is a problem that only the silence of the periodic sound can be realized.
[0020]
Therefore, the present invention activates acoustic power not only for noise radiated from the opening of the casing and the opening of the duct, but also for random noise from noise sources existing in the open space. It aims at providing the active silencer which can be reduced.
[0021]
[Means for Solving the Problems]
  In order to achieve the above object, the present invention is an active silencer for reducing the acoustic power of noise radiated from a noise source placed in an open space, and is correlated with the noise radiated from the noise source. Sensing means for detecting a signal, control coefficient calculating means for adaptively calculating the control coefficient of the control system from moment to moment, and predetermined calculation of the control coefficient calculated by this means and the detection signal obtained by the sensing means The control coefficient calculation means for performing the above and the sensing means are fixedly disposed via a sound insulating material, and the output of the control coefficient calculation means is introduced to have the same volume velocity (in the opposite phase to the noise of the noise source) (Additional sound source that emits acoustic energy), delay means that delays the phase of the detection signal obtained by the sensing means, and the sum of the signal that has passed through the delay means and the output of the control coefficient calculation means And a error signal detecting means for introducing into the control coefficient calculating means as a difference signal,The phase delay of the delay means is φ _f Where kd> φ as a condition for canceling noise emitted from the noise source, where d is the distance between the noise source and the additional sound source. _f (Where k is a wavelength) and π / 2> φ as a condition for reducing the acoustic power in order to cancel the noise _f > −π / 2 and π> k · dSaid to meetPhase delay φ _f Is set in the delay means, and the noise source and the additional sound source are arranged,Absorbing the radiant energy of the noise source, and locating the sensing means in the vicinity of the noise source to absorb the radiant energy of the noise source, thereby reducing the acoustic power in the entire spatial region around the noise source. It is said.
[0022]
Two or more additional sound sources may be provided.
In addition, the delay means may transmit a spatial transmission from the additional sound source identified by adaptive control using the random noise signal detected by the sensing means when the additional sound source is driven by an M-sequence random signal to the sensing means. The phase lag may be realized using the coefficient of.
[0023]
Furthermore, it is preferable that the sensing means and the additional sound source are substantially disposed in one housing.
In the present invention, the sensing means is arranged in the vicinity of the noise source, and the causality is satisfied by delaying the phase of the detection signal obtained by the sensing means by the delay means. By controlling the additional sound source with the control system so as to have (amplitude), an active silencer that does not require an acoustic delay space is realized.
[0024]
That is, in the conventional active silencer, an acoustic delay space for satisfying the causality is provided, and the starting point is that the sound must be silenced at the evaluation microphone position. In contrast, the device of the present invention focuses only on the fact that the additional sound source can be controlled with the opposite phase and the same volume velocity in order to reduce the acoustic power regardless of the process in the middle. And in order to implement | achieve this idea, causality is ensured by inserting a delay means in an electric signal path | route, and the additional sound of an antiphase with respect to a random sound is produced | generated.
[0025]
Therefore, in the case of the device of the present invention, the additional sound source can be brought close to the housing opening, the duct outlet, or the device that generates noise, and the acoustic power can be minimized even for random noise. A significant silencing effect can be achieved in the entire surrounding space.
[0026]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the invention will be described with reference to the drawings.
FIG. 1 shows a configuration example of an active silencer according to an embodiment of the present invention.
This active silencer employs an adaptive control system, and is roughly divided into signals correlated with noise radiated from a noise source X (which may be a duct outlet or a housing opening) placed in an open space. The sensing unit 21 to detect, the control coefficient calculation unit (A-FIR) 22 that adaptively calculates the control coefficient of the control system from time to time, the control coefficient calculated by the control coefficient calculation unit 22 and the output of the sensing unit 21 A control coefficient product-sum operation unit (FIR) 23 that performs a product-sum operation on the signal, an additional sound source 24 that supplies energy to the target noise using the output of the control coefficient product-sum operation unit 23 as an input signal, and a sensing unit 21. The delay circuit unit 25 that delays the phase of the signal obtained from the signal 21 and the sum of the signal that has passed through the delay circuit unit 25 and the output of the control coefficient product-sum operation unit 23 are given to the control coefficient calculation unit 22 as an error signal. It is constituted by the difference signal detector 26.
[0027]
In this example, the sensing unit 21 is configured mainly with a microphone, and the additional sound source 24 is configured with a speaker. The sensing unit 21 and the additional sound source 24 are provided in the housing 27. That is, the housing 27 is formed with an opening 28 for guiding the noise from the noise source X to the sensing unit 21, and an opening 29 for guiding the additional sound generated by the additional sound source 24 to the outside. Yes. Furthermore, a sound insulating material 30 is provided between the two to prevent the additional sound generated by the additional sound source 24 from directly entering the sensing unit 21 and causing howling, and the noise source X is removed by removing the influence of the reflection component. A sound absorbing material 31 is provided in order to increase the contribution ratio of the sound that directly enters the sensing unit 21 from the sound source.
[0028]
This active silencer uses an adaptive control system, but unlike conventional devices, it does not have an evaluation microphone that creates a point of zero sound pressure in the space, and an error signal is transmitted to the electric circuit in the control circuit. Treated as a signal.
[0029]
In this active silencer, the following conditions are necessary to mute the spatial region.
Condition 1:
In order to calculate the phase and amplitude of the additional sound generated by the additional sound source 24, the causality of the electric signal is ensured by the delay circuit unit 25.
Condition 2:
In order to cancel random sound waves (sound source) with the additional sound generated by the additional sound source 24, the causality of the spatial sound wave is ensured by the distance between the noise source X and the additional sound source 24.
Condition 3:
Examination of the allowable range of the phase of the additional sound generated by the additional sound source 24 in order to reduce the acoustic power of the noise
Next, the conditions 1 will be described in order.
[0030]
Signal S that has passed through delay circuit section 25₁ And the output signal S of the control coefficient product-sum operation unit 23₂ Are the volume velocities of noise and additional sound, respectively.₁ , A₂ , Phase lag ψ of control coefficient product-sum operation unit 23, phase lag φ of delay circuit unit 25_f , Wavelength k, angular frequency ω, phase delay at control system input φ_ADThen, it is expressed by equations (2) and (3).
[0031]
[Expression 1]

[0032]
Due to the action of the control coefficient calculator 22, the signal S₁ And S₂ Error signal e, which is the sum of and the volume velocity ratio A with respect to the phase ψ and noise of the control coefficient product-sum operation unit 23 according to equation (4)₂ / A₁ Becomes Equation (5).
[0033]
e = S₁ + S₂ = 0
A₁ cos φ_f + A₂ cos ψ = 0 (4)
Ψ ψ = φ_f + Π, A₂ / A₁ = 1 (5)
Here, in order for the control coefficient product-sum calculation unit 23 to correctly calculate the phase ψ shown in Equation (5), the error signal e needs to satisfy causality.
[0034]
Therefore, in order to satisfy the causality, the digital filter H of the control coefficient product-sum operation unit 23 shown in the equation (6) needs to have a delay phase characteristic._f It suffices to have a phase delay of (> 0). As a result, the additional sound source 24 can output the additional sound having the phase represented by the equation (5).
[0035]
[Expression 2]

[0036]
Next, condition 2 is considered.
Even if the additional sound can be generated under condition 1, for example, when the delay time is set to 1 hour as an extreme example, the electrical causality equation (6) is satisfied, However, it is impossible to mute the sound because the sound with the opposite phase to the noise is output with a delay of 1 hour. If it is periodic noise, it can be silenced because the sound of the same amplitude is transmitted repeatedly even with a delay of one hour, but in the case of random sound, it cannot be silenced.
[0037]
Therefore, it is necessary to consider acoustic causality in addition to electrical causality.
In order to cancel the random sound wave (sound source) with the generated additional sound, it is necessary to ensure the causality of the spatial sound wave according to the distance between the noise source X and the additional sound source 24.
[0038]
Even if a signal delayed by the delay time is output from the additional sound source 24, the sound can be silenced if the time during which the noise from the noise source X is transmitted to the additional sound source 24 is further delayed.
Therefore, if the distance between the noise source X and the additional sound source 24 is d, Equation (7) is Condition 2.
[0039]
kd> φ_f                     (Condition 2) (7)
Condition 1 indicates a condition for creating a sound, and condition 2 indicates a condition for physically canceling the sound in space.
[0040]
Finally, consider condition 3 for reducing the acoustic power in order to cancel the sound. The phase ψ of the additional sound for minimizing the acoustic power (the phase of the control coefficient product-sum calculation unit 23) is the total acoustic power when the additional sound is added to the noise.₀ Then, ψ = π from Equation (8) and Equation (9). This does not agree with equation (5), and φ_f The additional sound source 24 outputs a sound whose phase is shifted by as much as possible.
[0041]
[Equation 3]

[0042]
Where φ_f If = 0, it becomes ψ = π, and the optimum phase where the power decreases the most is obtained. However, since the electrical causality is not satisfied by Equation (6), it is actually impossible to generate an antiphase for noise. .
[0043]
Therefore, the optimum condition cannot be satisfied, but φ_f It is considered that the power is reduced by outputting the additional sound whose phase is shifted from the additional sound source 24.
Expression (10) shows an expression representing the reduction amount of the sound power.
Even if the phase of the additional sound with respect to the noise cannot be reversed, the power decreases if it is slightly deviated from the opposite phase. This is the aim of the present invention.
[0044]
[Expression 4]

[0045]
Substituting the phase and amplitude characteristics of equation (5) obtained by this method into equation (10) and considering the conditions for power reduction, equations (11) and (12) are obtained.

Therefore, the distance d between the noise source X and the additional sound source 24 and the delay time φ so as to satisfy all of the conditions 1, 2, 3_f Is set, the power can be reduced according to the equation (10), and a silencing effect can be obtained.
[0046]
When the structural delay characteristics of the speaker, which is the additional sound source 24, can be ignored, as a silencing effect, the distance between the noise source X and the additional sound source 24 is d, and φ_f = Π / 4,
34cm> d> 8.5cm to mute to 500Hz,
56cm> d> 14cm to mute up to 300Hz,
It becomes. Therefore, mute is possible if the distance between the noise source X and the additional sound source 24 is set within the above range.
[0047]
In the above example, the method of adjusting the delay time is used to generate the additional sound having the opposite phase to the noise from the additional sound source 24, but sensing from the additional sound source 24 by adaptive control as shown in FIG. It is also possible to identify the coefficient of spatial transmission to the unit 21 and use this coefficient instead of the delay circuit 25.
[0048]
That is, a signal generator 32 is built in, and an M-sequence signal is generated by the signal generator 32 to drive the additional sound source 24. At this time, the coefficient of spatial transmission from the additional sound source 24 to the sensing unit 21 Is identified adaptively by the control coefficient calculation unit (A-FIR) 33. At this time, the identified filter coefficient H_d Is obtained by the following equation.
[0049]
e = MG_AD・ Cr ・ G_AD-M ・ H_d = 0
H_d = G_AD・ G_DA・ Cr (13)
G_AD, G_DA: Delay characteristics of AD conversion and DA conversion
When this is used as a fixed filter coefficient (product-sum operation circuit) instead of the delay circuit unit 25 shown in FIG. 1, Equation (14) is obtained from Equation (6).
[0050]
[Equation 5]

Therefore, φ_AD+ Φ_DA+ Ψ_crAlthough the causality is satisfied from> 0, the signal S entering the error signal detection unit 26₁ , S₂ Respectively becomes Equation (15).
[0051]
[Formula 6]

[0052]
e = S₁ + S₂ = 0 to A₁ cos (φ_AD+ Φ_DA+ Ψ_cr) + A₂ cos ψ = 0
Therefore, the phase of the additional sound relative to the noise (the phase of the control coefficient product-sum operation unit) ψ and the volume velocity ratio A₂ / A₁ Is
ψ = φ_AD+ Φ_DA+ Ψ_cr+ Π, A₂ / A₁ = 1 (16)
Thus, the amount deviated from the optimum phase of ψ = π + 2nπ (n = 0, 1, 2,...) Increases. φ_AD+ Φ_DA+ Ψ_crThis term is determined by the characteristics of the control system and the characteristics of the spatial acoustic system, and cannot be adjusted close to 2nπ because it cannot be easily adjusted from the outside.
[0053]
Therefore, the identified H_d Even if the filter characteristic is used as it is as a fixed FIR coefficient instead of the delay circuit unit in the adaptive control system, it is difficult to improve the silencing effect.
[0054]
Therefore, as shown in FIG. 3, a DA converter 34 and an AD converter 35 may be added as delay elements. In FIG. 3, reference numeral 36 denotes a filter previously identified using an M-sequence signal.
[0055]
The delay circuit unit may be adjusted by a time corresponding to the delay phase of AD conversion and DA conversion. However, in view of the system configuration, the configuration of FIG. 3 in which a DA converter 34 and an AD converter 35 are added in one stage is better. This is preferable because the phase delay of AD conversion and DA conversion can be completely removed.
In the configuration shown in FIG. 3, the signal S entering the error signal detection unit 26.₁ , S₂ Respectively becomes the equation (17).
[0056]
[Expression 7]

[0057]
e = S₁ + S₂ = 0, A₁ cos ψ_cr+ A₂ cos ψ = 0
The phase of the additional sound with respect to the noise and the volume velocity ratio are expressed by Equation (18).
ψ = ψ_cr+ Π, A₂ / A₁ = 1 (18)
Compared with equation (16), the phase lag during AD and DA conversion can be eliminated.
The H filter characteristic at this time is
S ・ G_AD・ H_d + S ・ G_AD・ G_DA・ G_AD・ From H = 0
[0058]
[Equation 8]

And ψ_crCausality is also satisfied from> 0.
[0059]
From equation (18), the phase of the additional sound is ψ_crIt becomes + π and does not become an antiphase. But ψ_crIs ψ if the spatial path is a free sound field_cr= Kd, the closer the phase d between the noise source X and the additional sound source 24 is, the closer to the opposite phase and the lower the power.
The power reduction amount (dB) at this time is obtained from the equation (10).
[0060]
[Equation 9]

[0061]
Even in the conventional method using the evaluation microphone, in Formula (10), a sufficient silencing effect was confirmed even when kd = 0.99 (frequency f = 300 Hz; distance d = 0.18 m between the noise source and the additional sound source). So if ψ_crEven if the spatial transfer function includes a reflection component and the direct phase (phase kd) is not realized, this phase delay can be dealt with by reducing the interval d, and the silencing effect is not deteriorated.
[0062]
Incidentally, ψ when kd = 0.36 (frequency = 200 Hz; interval between noise source and additional sound source d = 0.1 m)_crWhen the difference (dB) in the amount of power decrease due to the presence or absence of is estimated, it is as follows.
[0063]
(a) ψ_crCan be ignored

  (b) ψ_crWhen considering

Therefore, when fd = 19.48 (from kd = 0.36),
In other words, at an interval d = 0.05 m, up to about 390 Hz,
up to about 195Hz at d = 0.1m
At d = 0.2m up to about 97Hz,
The power is reduced by about 7.7 dB.
[0064]
ψ_crAs compared with the case where the noise can be ignored, the noise reduction effect is reduced by about 6 dB. However, since it still has a reduction effect, it can be applied depending on the nature of the noise source.
FIG. 4 shows a schematic configuration of an active silencer according to still another embodiment of the present invention.
[0065]
In the example described above, the device is configured by using one additional sound source, but in this example, two additional sound sources are used to improve the silencing effect.
This active silencer also adopts an adaptive control system, and is largely divided into signals correlated with noise radiated from a noise source X (which may be a duct outlet or a housing opening) placed in an open space. A sensing unit 41 for detection, a control coefficient calculation unit (A-FIR) 42 for adaptively calculating control coefficients of the control system from time to time, a control coefficient calculated by the control coefficient calculation unit 42 and an output of the sensing unit 41 A control coefficient product-sum operation unit (FIR) 43 that performs a product-sum operation with the signal, and additional sound sources 44a and 44b that supply acoustic energy to target noise using the output of the control coefficient product-sum operation unit 43 as an input signal; A delay circuit unit 45 that delays the phase of the signal obtained from the sensing unit 41, a calculation unit 46 that adjusts the output of the control coefficient product-sum calculation unit 43 so that an additional sound having the same volume velocity as the noise can be output, And the sum of the signal passing through the output delay circuit section 25 of the calculation part 46 is composed of the error signal detection unit 47 applied to the control coefficient calculation unit 42 is regarded as an error signal e.
[0066]
Also in this example, the sensing unit 41 is mainly composed of a microphone, and the additional sounds 44a and 44b are composed of speakers. The sensing unit 41 and the additional sounds 44 a and 44 b are provided in the housing 48. That is, an opening 49 for guiding noise from the noise source X to the sensing unit 41 is formed in the casing 48, and openings 50a and 50b for guiding additional sound generated by the additional sound sources 44a and 44b to the outside. Is formed. Further, a sound insulating material 51 is provided between the two additional sound sources 44a and 44b so that the additional sound generated by the additional sound sources 44a and 44b is not directly incident on the sensing unit 41 to cause howling. A sound absorbing material 52 is provided to increase the contribution ratio of the sound that is directly incident on the sensing unit 41 from the source X.
[0067]
In the figure, G_AD, G_DAIndicates the presence of a delay element due to AD conversion and DA change.
As described above, when two additional sound sources are used, the placement method of the additional sound sources becomes a problem.
[0068]
Total sound power W when noise is added to additional sound generated by two additional sound sources₀ Is

It becomes. Θ_i : Phase of each sound source i, R is radiation resistance, and Vi is vibration speed.
[0069]
Here, if the two additional sound sources are controlled by the same phase θ,
θ = θ₂ −θ₁ = Θ_Three −θ₁                             …(twenty two)
It becomes.
[0070]
On condition that the two additional sound sources 44a and 44b are arranged at an equal distance d from the noise source X,
R₁₁= 4πr₁ ² ρck² r₁ ²
R₁₂= 4πr₁ ² ρck² r₂ ² ・ Sin (kd) / kd
R₁₃= R₁₂(∴r₂ = R_Three )
R_{twenty three}= 4πr₂ ² ρck² r_Three ² ・ Sin (kL) / kL
｜ V₁ | = A₁ / 4πr₁ ²
｜ V₂ | = A₂ / 4πr₂ ²
｜ V_Three | = | V₂ ｜ (∴r₂ = R_Three ) …(twenty three)
It becomes. R_i : Vibration radius, R: Radiation resistance, A_i : Volume velocity, V_i : Vibration speed.
[0071]
Sound power of only noise before control is W₁ Then, the power reduction amount (dB) after control is

It becomes. d: interval between the sound source and the additional sound source, L: interval between the two additional sound sources.
[0072]
[Expression 10]

Thus, the power can be minimized.
[0073]
Since the additional sound sources 44a and 44b are on the circumference of the radius d with the noise source X as the center, the arrangement in which the interval L between the two additional sound sources L is the minimum is the additional sound source as shown in FIG. Are adjacent to each other. Therefore, in the active silencer according to this example, the additional sound sources 44a and 44b are actually arranged adjacently at an equal distance from the sensing unit 41, as shown in FIG.
When the delay circuit unit 45 is used as in the configuration shown in FIG.₁ , S₂ Becomes Equation (26).
[0074]
[Expression 11]

[0075]
e = S₁ + S₂ = 0, A₁ cos φ_f + 2A₂ cos ψ = 0
Therefore, the phase and volume velocity ratio of the additional sound with respect to the noise are expressed by Equation (27).
ψ = φ_f + Π, A₂ / A₁ = 1/2 (27)
e = S ・ G_AD・ F + S ・ G_AD・ From H = 0
The control system filter coefficient H is expressed by equation (28).
[0076]
[Expression 12]

[0077]
In the delay circuit unit 45, φ_f If = 2nπ (n = 0, 1, 2,...) Is satisfied, the additional sound approaches an opposite phase according to the equation (27). The amount of decrease at this time can be obtained by substituting equation (27) into equation (24).
[0078]
In the above example, the method of adjusting the delay time is employed to generate the additional sound with the opposite phase from the additional sound sources 44a and 44b. However, sensing from the additional sound sources 44a and 44b by adaptive control as shown in FIG. The coefficient of spatial transmission to the unit 41 may be identified and used in place of the delay circuit unit 45.
[0079]
That is, the built-in signal generator 32 generates an M-sequence signal to drive the additional sound sources 44a and 44b. At this time, the spatial transmission coefficient from the additional sound sources 44a and 44b to the sensing unit 41 is adapted from time to time. Thus, the control coefficient calculation unit (A-FIR) 62 identifies them. At this time, the identified filter coefficient H_d Is obtained by the following equation.
[0080]
[Formula 13]

[0081]
This H_d When the filter is used as it is as a product-sum operation circuit instead of the delay circuit unit 45 shown in FIG. 4, the silencing effect is deteriorated as in the case of one additional sound source. Therefore, as shown in FIG. 7, as in the case of one additional sound source, an AD conversion circuit 63, a DA conversion circuit 64, and an arithmetic circuit 65 for adjusting the volume velocity coefficient are used.
[0082]
The signal that enters the error signal detection unit 47 is S₁ , S₂ Then, in order to output the optimum volume velocity, S₂ Since the previous stage of the signal must be multiplied by a factor of four, S₁ , S₂ Becomes Equation (30).
[0083]
[Expression 14]

[0084]
ψ = ψ_cr1 + Π, A₂ / A₁ = 1/2 ... (32)
Here, the error signal e = S · G_AD・ H_d + S ・ G_AD・ G_DA・ G_AD・ From H = 0
The control system filter H is expressed by Equation (33).
[0085]
[Expression 15]

And satisfies the causality (∴ψ_cr1 > 0).
[0086]
Therefore, it is possible to output from the additional sound sources 44a and 44b with the phase of Expression (32).
The additional sound sources 44a and 44b_cr1 A shifted phase will be output.₂ / A₁ = 1/2 and ψ_cr1 , Ψ_cr2 If the spatial path is a free sound field,
[0087]
[Expression 16]

It becomes.
[0088]
When f · d = 19.48 (from kd = 0.36), kL = 0.36,

Thus, a good silencing effect can be obtained.
[0089]
【The invention's effect】
As described above, according to the present invention, even for noise that does not have an acoustic delay space such as a duct, the additional sound source is brought close to the noise source, and the additional sound having the opposite phase and the same amplitude is also obtained for random noise. Can be generated, the acoustic power can be minimized, and the entire space around the noise source can be silenced.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram of an active silencer according to a first embodiment of the present invention.
FIG. 2 is a configuration diagram for identifying a delay time by adaptive control in an active silencer according to a second embodiment of the present invention.
FIG. 3 is a schematic configuration diagram of an active silencer according to a third embodiment of the present invention.
FIG. 4 is a schematic configuration diagram of an active silencer according to a fourth embodiment of the present invention.
FIG. 5 is a diagram for explaining the arrangement of additional sound sources in the apparatus;
FIG. 6 is a configuration diagram in the case of identifying a delay time by adaptive control in an active silencer according to a fifth embodiment of the present invention.
FIG. 7 is a schematic configuration diagram of an active silencer according to a sixth embodiment of the present invention.
FIG. 8 is a diagram for explaining an example in which noise leaking from the duct outlet is muted
FIG. 9 is a diagram for explaining an example in the case of silencing noise leaking from an opening of a housing
FIG. 10 is a diagram for explaining an example in which it is difficult to mute with a conventional method;
FIG. 11 is an explanatory diagram of an optimal arrangement for using two additional sound sources to minimize the sound power.
[Explanation of symbols]
21, 41 ... Sensing part
22, 42 ... Control coefficient calculation unit (A-FIR)
23, 43 ... Control coefficient product-sum operation unit (FIR)
24, 44a, 44b ... Additional sound source
25, 45 ... delay circuit section
26, 47 ... error signal detector
27, 48 ... casing
28, 29, 49, 50a, 50b ... opening
30, 51 ... Sound insulation material
31, 52 ... Sound absorbing material
32, 61 ... Signal generator for generating M-sequence signal
33, 62 ... control coefficient calculation unit
36, 66 filter
x ... Noise source

Claims (4)

An active silencer for reducing the acoustic power of noise radiated from a noise source placed in an open space, the sensing means for detecting a signal correlated with the noise radiated from the noise source, and control of the control system Control coefficient calculating means for adaptively calculating the coefficient from time to time, control coefficient calculating means for performing a predetermined calculation of the control coefficient calculated by this means and the detection signal obtained by the sensing means, and the sensing means An additional sound source that is fixedly disposed through a sound insulating material and emits acoustic energy having the same volume velocity (amplitude) in phase opposite to that of the noise of the noise source by introducing the output of the control coefficient calculation means; Delay means for delaying the phase of the detection signal obtained by the sensing means, and the sum of the signal that has passed through the delay means and the output of the control coefficient calculation means as an error signal to the control coefficient calculation means And a error signal detection means for entering,
When the phase delay of the delay means is φ _f and the distance between the noise source and the additional sound source is d,
The conditions for canceling the noise emitted from the noise source are:
kd> φ _f (where k is the wavelength)
As a condition for reducing the sound power to satisfy the above and to cancel the noise:
satisfies π / 2> φ _f > −π / 2, and
π> k · d
The phase delay φ _f is set in the delay means so as to satisfy the above condition, and the noise source and the additional sound source are arranged, thereby positioning the sensing means in the vicinity of the noise source. An active silencer characterized in that it absorbs radiant energy and reduces the acoustic power of the entire space around the noise source.   The active silencer according to claim 1, wherein two or more additional sound sources are provided.   The delay means is a coefficient of spatial transmission from the additional sound source to the sensing means identified by adaptive control using a random noise signal detected by the sensing means when the additional sound source is driven by an M-sequence random signal. The active silencer according to claim 1, wherein phase delay is realized using   3. The active silencer according to claim 1, wherein the sensing unit and the additional sound source are substantially disposed in one casing. 4.

Images