JP4440499B2 - Centrifugal pump drive - Google Patents

Description

【００３４】
実験に用いた循環回路の模式図を図１０に示す。図１０中、符号１は遠心ポンプ、１１はモータを含む駆動部、５０は電磁血流計（ＭＦＶ−３２００；日本光電社製）、５１は圧力計（ＡＰ−６１１Ｇ；日本光電社製）、５２は液体（グリセリン水溶液）を貯留するためのリザーバ、５３は流量調節用のクランプである。
各濃度のグリセリン水溶液を実験用循環回路に充填し、温度を３０℃に保った。循環実験中は、回転数を一定にした状態で流量をクランプの調整により変化させた。
流量推定式算出のために行った流量−電流特性実験の手順を示す。
▲１▼粘性係数２．０[cP]、３．０[cP]、４．０[cP]に調整したグリセリン水溶液を実験回路に充填した。
▲２▼回転数を１０００〜３５００ｒｐｍの範囲において５００ｒｐｍ刻みで変化させ、各回転数で一定に保った。
▲３▼各回転数で流量を１[L/min]刻みに０[L/min]から最大流量まで変化させ、このときの電流値を計測した。
粘性係数２．０[cP]における流量−電流特性図を図１１に示す。
粘性係数３．０[cP]における流量−電流特性図を図１２に示す。
粘性係数４．０[cP]における流量−電流特性図を図１３に示す。
【００３５】
図１１〜図１３の流量−電流特性図から、モータの特性は各粘性係数の実験においても同様な傾きを示したが、グラフの切片である流量０[L/min]の電流値は粘性係数が大きくなれば同じ回転数でも上昇することが確認できた。よって、流量圧力推定には正確な粘性係数を考慮する必要があるものと考えられた。
以上を考慮し角速度を用いて、流量軸方向と電流軸方向に対し軸変換処理を行った。この軸変換により、測定データを一直線上に表すことが可能であり、流量推定式を算出することが可能である。
一例として、粘性係数３．０[cP]の流量−電流特性図（図１２）を用いて流量推定式の算出過程について記す。
電流軸の変換は、図１２の流量０[L/min]における電流値に着目し、回転数（角速度）に対する電流値の特性を調べた。粘性係数３．０[cP]の流量０[L/min]における角速度−損失電流特性図を図１４に示す。
【００３６】
図１４より粘性係数３．０[cP]の流量０[L/min]における角速度−損失電流特性図を次式（３）に示す。
Ｉ＝2.0×10^-06×ω^2.308 （３）
Ｒ２＝0.9998
（式中、Ｉは遠心ポンプ損失電流[A]、ωは角速度[rad/s]を示す）
【００３７】
図１４と式（３）から、粘性係数３．０[cP]の流量０[L/min]におけるポンプ損失電流は、回転角速度ωの２．３０８乗に上昇していることが示されている。そのため、流量０[L/min]における損失電流値の各切片を２．３０８乗で割ることにより、軸上の一点に集中させることができる。この角速度の２．３０８乗で割る操作を全てのデータに対して行った結果を図１５に示す。
図１５において、回転数の上昇とともに傾きが小さくなっていることが示されている。そこで各データの流量値を角速度の累乗で割ることにより、全ての傾きを同じにすることが可能である。各角速度と傾きの特性図を図１６に示す。
【００３８】
図１６から、角速度と傾きの関係式はｙを切片とすると角速度ωを用いて次式（４）のように示される。
ｙ＝63.139ω^-0.7079 （４）
この式（４）より、角速度の上昇により傾きが角速度の０．７０７９乗で減少していることが示される。これにより、流量値を角速度の０．７０７９乗で割ることにより、傾き全てが一致することになる。
流量方向と電流方向への軸変換の操作により全てのデータが一直線上に表現できる。粘性係数３．０[cP]の流量−電流特性図の全てのデータに軸変換を行った結果を図１７に示す。
【００３９】
図１７で一直線上に表現された特性図を線形近似式により特性式を求めると次式（５）のように示される。この式が粘性係数３．０[cP]における流量推定式となる。
Ｉ’＝0.3062Ｑ’＋17.445 （５）
同様の軸変換操作を各粘性係数の流量−電流特性図について行うことで、各粘性係数における流量推定式を求めることが可能である。
この時の各軸方向軸変換の累乗値を表２に示す。
【表２】

【００４０】
表２から、各粘性係数の累乗値の平均値を用いて、流量、電流方向への軸変換式は次のように示し、再度各粘性係数の流量−電流特性図に軸変換を行った結果を図１８に示す。
電流方向の軸変換式 Ｉ’＝Ｉ／ω^2.19×１０⁷
流量方向の軸変換式 Ｑ’＝Ｑ／ω^0.63×200
【００４１】
図１８から各粘性係数における流量推定式は以下のように算出した。
粘性係数２．０[cP]の流量推定式
Ｉ’_2.0＝0.415Ｑ’＋30.4
粘性係数３．０[cP]の流量推定式
Ｉ’_3.0＝0.415Ｑ’＋33.0
粘性係数４．０[cP]の流量推定式
Ｉ’_4.0＝0.4234Ｑ’＋35.1
【００４２】
以上より、粘性係数の違いによって各流量推定式の傾きは殆ど変化せず、切片のみが30.4、33.0、35.1と傾きに比例した時、大きく変化していることが確認できた。このため流量推定式の傾きには各粘性係数の推定式の傾き平均値を用いることとした。また、切片は粘性係数の変化により変化するため、粘性係数を考慮する必要があると考えられた。粘性係数と切片の関係を図１９に示す。
【００４３】
図１９より、切片と粘性係数の関係式は次式（６）のように示される。
Ｍ＝0.0032ｎ²−0.511ｎ＋3.6599 （６）
（式中、Ｍはグリセリン水溶液濃度[%]、ｎは粘性係数[cP]を示す）
以上より、粘性係数を考慮した一般化流量と一般化電流の関係式を次式に示す。
Ｉ’＝0.4181Ｑ’＋（2.2895ｎ＋26.013）
（流量推定のための実験式）
Ｑ’＝（Ｉ’−(2.2895ｎ＋26.013））／0.4181 （７）
（式中、Ｉは電流[A]、Ｑは流量[L/min]、ｎは粘性係数[cP]である）
【００４４】
以上の通り流量推定式が流量−電流特性図の実験結果（図１１〜図１３）より算出された。
図１８より粘性係数の変化により推定式の傾きは殆ど変化せず、切片のみが変化することが示された。そのため流量推定式の傾きは各粘性係数の推定式の平均値を用いることとし、切片は式（６）を用いて推定式の算出を行った。これにより、遠心ポンプ駆動装置の流量測定を回転数と電流値のみから行うよりも、液体の粘性係数も計測し考慮することが好ましいことが示された。
【００４５】
４．圧力推定式の算出
流量と回転数より圧力推定式の算出を試みた。圧力推定式算出において流量は前述した通り推定した流量値を用いた。また、実験に用いたグリセリン水溶液は、前述した式（１）を用いて粘性係数の調整を行った。
【００４６】
各濃度のグリセリン水溶液を実験用循環回路に充填し温度を３０℃に保った。実験循環回路は図１０に示す。
循環実験中は、回転数を一定にした状態で流量をクランプの調整により変化させた。次に実験手順を示す。
▲１▼粘性係数２．０[cP]、３．０[cP]、４．０[cP]に調整したグリセリン水溶液を実験回路に充填した。
▲２▼回転数を１０００〜３５００[rpm]において５００[rpm]刻みで変化させ、各回転数で一定に保った。
▲３▼各回転数で流量を１[L/min]刻みに最大圧力まで変化させ、この時の遠心ポンプの入口圧と出口圧を計測し、圧力差とした。
粘性係数２．０[cP]の時の流量−圧力特性図を図２０に示す。
粘性係数３．０[cP]の時の流量−圧力特性図を図２１に示す。
粘性係数４．０[cP]の時の流量−圧力特性図を図２２に示す。
【００４７】
図２０〜図２２の流量−圧力特性図から、回転数より求められる角速度を用いて、流量軸方向と電流軸方向に対し、軸変換処理を行った。これにより、測定値を一直線上に表すことができ、圧力推定式を算出することが可能となった。
粘性係数３．０[cP]の流量−圧力差特性図（図２１）を用いて圧力推定式の算出過程について述べる。
圧力軸の変換は、図２１の流量０[L/min]における圧力差に着目し、回転数（角速度）に対する圧力差の特性を調べた。粘性係数３．０[cP]の流量０[L/min]における角速度−圧力差特性図を図２３に示す。
【００４８】
図２３より粘性係数３．０[cP]の流量０[L/min]における角速度−圧力差特性図を次式（８）に示す。
Ｐ＝0.0021×ω^2.1439 （８）
Ｒ２＝1.0
（式中、Ｐは圧力差[mmHg]、ωは角速度[rad/s]を示す。）
【００４９】
この式（８）から、粘性係数３．０[cP]の流量０[L/min]における圧力差は、回転角速度ωの２．１４３９乗に上昇していることが示されている。そのため、流量０[L/min]における圧力差の各切片を２．１４３９乗で割ることにより、切片は一点に集中させることができる。この角速度の２．１４３９乗で割る操作を全てのデータに対して行った結果を図２４に示す。これが、圧力方向の軸変換となり次式のように示すことができる。
Ｐ’＝Ｐ／ω^2.05
（式中、Ｐ’は軸変換後の圧力差、ωは角速度を示す）
【００５０】
図２４において流量方向に軸変換を行う。しかし、圧力の場合、流量の変換時のように傾きがないため、軸変換の累乗値は最も相関の高いものを用いることとした。その結果、流量方向に１．１乗で割ることで、同一曲線上に表すことが示された。粘性係数３．０[cP]の流量−圧力差特性図の全てのデータに対し軸変換を行った結果を図２５に示す。
図２５で同一曲線上に表現された特性図から、多項式近似式により特性式を求めると次式のように表される。この式は粘性係数３．０[cP]における圧力推定式となる。
Ｐ’＝0.00007ω²−0.00076ω＋20.568
Ｒ²＝0.9581
【００５１】
同様の軸変換操作を各粘性係数の流量−圧力差特性図に行うことで、各粘性係数における圧力推定式を求めることが可能である。
この時の、各軸方向の軸変換の累乗値を表３に示す。この各粘性係数の累乗値の平均値を用いて、再度各粘性係数の流量−圧力差特性図に軸変換を行った結果を図２６に示す。
【表３】

【００５２】
粘性係数２．０[cP]の流量推定式
Ｐ’_2.0＝−0.0003Ｑ''²＋0.0036Ｑ''＋33.68
粘性係数３．０[cP]の流量推定式
Ｐ’_3.0＝−0.0002Ｑ''²＋0.0128Ｑ''＋33.79
粘性係数４．０[cP]の流量推定式
Ｐ’_4.0＝−0.0002Ｑ''²＋0.0048Ｑ''＋34.18
【００５３】
図２６に示されるように、各粘性係数の軸変換後の特性は、ほぼ同一曲線状に表現できるため、各粘性係数における特性の違いを推定式に考慮する必要はないと考えられる。そこで、各粘性係数における特性式の平均を圧力推定式とした。
圧力推定のための実験式（１０）
Ｐ’＝−0.0002Ｑ''²＋0.007Ｑ''＋33.88 （１０）
【００５４】
以上の通り、圧力推定式（１０）が流量−圧力差特性図の実験結果より算出された。圧力推定式には、前記「３．流量推定式の算出」において推定した流量を用いた。そのため、流量に大幅な推定誤差が生じれば、圧力推定にも影響を及ぼすこととなり、推定誤差が生じるものと考えられる。
【００５５】
５．牛血液を用いた流量圧力推定式の評価実験
流量圧力推定式の評価実験として、グリセリン水溶液を用いた時と同じ実験回路において、実際に牛血液を用いて推定式の評価実験を行った。
次に実験手順を示す。
▲１▼牛血液を３０℃でヘマトクリット値４０％に調整し、実験回路に充填する。
▲２▼回転数を１０００〜３５００[rpm]において推定式の評価実験を行う。
【００５６】
流量推定式の評価方法として計測流量値に対する推定流量値をグラフ上にプロットし、ｙ＝ｘの値からのずれを比較した。流量推定式の精度図を図２７に示し、これまでの流量推定式の精度図を図２８に示す。圧力推定式の精度図を図２９に示す。また、従来の推定式との流量及び圧力推定精度の比較表を表４、表５に示す。
【表４】

【表５】

【００５７】
牛血液を用いた流量推定式評価実験において、平均流量精度２．０％、最大推定流量誤差０．５６[L/min]と高い精度を得ることが示された。流量推定誤差の要因としては、血液はグリセリン水溶液と比較した場合、循環中の温度等の要因による粘性の変化が生じると考えられる。しかし、粘性係数を考慮することで推定式の精度が従来に比較し大幅に向上したことから、流量推定における粘性係数の重要性が示された。
現在のシステムでは、循環中に変化する血液の粘性係数を測定することは困難であるために、粘性係数情報を入力又は切替により与えて測定精度を上げたが、今後何らかの方法によりリアルタイムな粘性係数を知ることができることが望まれる。
また、牛血液を用いた圧力推定式評価実験において、平均圧力精度５．８％、最大推定圧力誤差３０．７[mmHg]となった。圧力推定式には、前記「３．流量推定式の算出」において推定した流量を用いている。そのため、流量に推定誤差が生じれば、圧力推定にも推定誤差が生じるものと考えられる。
【００５８】
［実施例２：小型コントロールシステム］
１．推定流量圧力表示システム
８ビットマイクロコンピュータを用いて、推定パラメータである電流値と回転数をリアルタイムにマイコンに取り込むことで、前記実施例１で算出・評価を行った流量推定式により、血液循環中に流量表示が可能な小型システムの試作を行った。実験溶液の粘性係数は、予め粘度計を用いて計測を行い入力した。
推定システムのブロック図を図３０に示す。
【００５９】
駆動部に用いられているＤＣブラシレスモータには、ロータの位置検出用にホールセンサが内蔵されている。このホールセンサからの出力は方形波であるため、その出力回数は周波数計を用いることにより計測が可能である。このモータでは、１回転する間にホールセンサから２回出力があるため、周波数計に表示される周波数を３０倍することによって一分間あたりの回転数[rpm]に換算することができる。周波数計には秋月電子通商製「ＰＩＣ１６Ｃ７１ 使用液晶表示周波数カウンタキットVer.2」を用いた。
【００６０】
前記の通り８ビットマイクロコンピュータを用いた流量推定システムの試作を行った。試作した表示部により、実質的に流量計として使用し得る小型遠心ポンプ駆動装置コントローラの実現が可能となった。また、循環中に変化する粘性係数をリアルタイムに検知し、推定式に組み込むことが可能になればより高精度な推定が行えると考えられる。
【００６１】
２．オンライン粘度計の検討
実施例１において、流量の推定には血液の粘性係数の影響を受けることが示された。そのため、磁歪素子を用いた超音波振動子により、小型でリアルタイムに粘性係数が計測可能なオンライン粘度計について検討を行った。
磁歪振動子は、磁性体の磁歪現象を利用した電気音響変換器として、超音波の発生及び検出のために古くから用いられており、音響測深機、魚群探知機、ソナー、或いは超音波を利用した洗浄、乳化、分散、殺菌、機械加工、溶接などに広く利用されている。
消磁された強磁性体が磁化されると外形が僅かに変化する。この現象を磁歪という。鉄の丸棒を長さｌに沿って磁化するとδｌだけ伸びる。この比δｌ／ｌを磁歪λで表す。
λ＝δｌ／ｌ
このλは鉄では１０^-5程度であり、符号は正を伸びとする。磁界の方向の磁歪による効果を縦効果といい、磁界に垂直な磁歪を横効果というが、体積効果である体積磁歪はあまり大きくなく、したがって普通縦効果と横効果とは逆符号である。磁歪は磁界に対しては履歴を伴うが、磁化に対しては１価関数である。磁歪現象は最初Ｊｏｕｌｅによって発見され、ジュール効果とも言う。この逆効果はビラリ効果という。
磁歪振動子には角形振動子と環状振動子がある。金属磁歪材料を用いる場合には渦電流を減らすために薄板にし、表面絶縁を施して積層固着するが、フェライト磁歪材料の場合にはその必要がない。
磁性体の静磁歪量は磁化の方向を反対にしても変化せず、消磁状態では小振幅の振動に対する動的な磁歪変換の機能はない。したがって、磁歪振動子を使用する場合には、偏奇磁化を与える必要がある。フェライト振動子の場合は磁石偏奇磁化方式が多く用いられる。これは磁気回路に磁石を入れて偏奇磁化を与え、直流偏奇電流を不要にしたものである。磁石の中に交流磁束を通す必要があるから、金属の磁石は使用できず、バリウムフェライトが使用される。
磁歪振動子は殆どの場合、機械的共振周波数で使用される。したがって、振動子の振動方向の寸法は超音波の周波数によってほぼ決まる。このような寸法上の制約もあって、磁歪振動子の使用される周波数は主として数ｋＨｚ〜１００ｋＨｚの範囲である。
静磁歪による振動子の伸縮率は非常に小さいが、機械的共振の状態ではそれよりはるかに大きな振動が生ずる。零進を強くしていくとその振動による応力が破壊限度を超えて振動子が壊れてしまうこともある。
【００６２】
強磁性体に磁界を加えるとその磁化にともなって磁界方向に歪みを生ずる。磁気歪みには幾つかの型があるが、粘度の測定に重要なのは、磁気歪み効果またはジュール効果と呼ばれる縦方向の長さが変化する現象および、磁化された材料に応力を加えると透磁率が変化するビラリ効果または逆磁気ひずみ効果と呼ばれる縦方向の長さが変化する現象および、磁化された材料に応力を加えると透磁率が変化するビラリ効果または逆磁気歪み効果と呼ばれる現象である。
液体中で振動子を発振させると、素子の伸び縮みを妨げる力が生じる。この力を受けることにより素子の透磁率が変化する。透磁率が変化することによりコイルのインダクタンスが変化し、インピーダンスが変化する。この変化を測定する。この振動片粘度計の特徴は透明、不透明を問わず管内を流れる流体の粘性係数を連続的に測定できる点である。
【００６３】
実験に用いたＴＤＫ（株）製、超音波発振用磁歪振動子フェライトの主な特性について表６に示す。また、超音波振動子の外形を図３１に、また各部の寸法を表７に示す。
【表６】

【表７】

【００６４】
図３２に示す実験装置を用いて、サンプル溶液（グリセリン１００％、グリセリン５０％水溶液及び水）をビーカに入れ、液体中に振動子を入れた時の電圧波形および電圧値を測定した。
図３３〜図３５にそれぞれのサンプル溶液でオシロスコープを用いて測定した電圧波形を示す。図３３はグリセリン１００％の時の電圧波形を、図３４はグリセリン５０％水溶液の時の電圧波形を、図３５は水の時の電圧波形をそれぞれ示す。
図３３〜図３５に示した実験結果より、グリセリンの濃度の違いによる電圧波形の変化が示された。
【００６５】
本実施例１，２では、二点支持式遠心ポンプ駆動装置（Ｃ１Ｅ３；京セラ製）を用いたＰＣＰＳシステムの小型化について検討を行った。
モータの特性実験を行い、粘性係数を考慮し流量圧力推定式を算出した。推定式の評価実験として、実際に牛血液を用いて評価実験を行った。血液における流量の推定精度は２．０％と高精度となり、各流量において安定した流量推定結果が得られた。また、市販されている遠心ポンプシステムに用いられている超音波流量計の誤差が±１０％程度、電磁流量計の誤差が±５％程度と報告されていることから、本実施例における流量推定精度の高さが確認できた。
圧力推定精度は５．８％と圧力についても高精度な推定結果が得られた。
流量圧力推定において流体の粘性係数も重要な推定要素であり、新たに粘性係数を推定式に代入することで流量圧力推定精度が大幅に改善されることが示された。また、マイクロコンピュータを用いてモータ駆動部を制御し、推定流量値と推定圧力値を表示させることで流量計と圧力計を必要としない小型遠心ポンプ循環システムの実現の可能性が示唆された。
【００６６】
【発明の効果】
本発明によれば、従来の超音波流量計や電磁流量計よりも安価な流量測定手段を備えた遠心ポンプ駆動装置を提供することができる。
また、遠心ポンプの経時劣化による交換時期を示唆できる、より安全性の高い遠心ポンプ駆動装置を提供することができる。
さらに切り替え操作によってクリアプライム液と血液の両方の流量測定ができる流量計を備えた遠心ポンプ駆動装置を提供することができる。
また、切り替え操作によって損失電流の異なる各種の遠心ポンプ駆動装置に応じて流量測定を正確に実施可能な流量計を備えた遠心ポンプ駆動装置を提供することができる。
【図面の簡単な説明】
【図１】 遠心ポンプを例示する縦断面図である。
【図２】 本発明による遠心ポンプ駆動装置の参考例を示す構成図である。
【図３】 本発明による遠心ポンプ駆動装置の第１の実施形態を示す構成図である。
【図４】 本発明による遠心ポンプ駆動装置の第２の実施形態を示す構成図である。
【図５】 本発明による遠心ポンプ駆動装置の第３の実施形態を示す構成図である。
【図６】 本発明に係る実施例の結果を示し、使用した遠心ポンプの流量圧力特性を示すグラフである。
【図７】 各ヘマトクリット値と粘性係数及び温度の関係を示すグラフである。
【図８】 グリセリン水溶液の濃度と粘性係数の関係を示すグラフである。
【図９】 モータ損失電流と角速度の関係を示すグラフである。
【図１０】 実験で用いた循環回路の構成図である。
【図１１】 粘性係数２．０[cP]における流量−電流特性図である。
【図１２】 粘性係数３．０[cP]における流量−電流特性図である。
【図１３】 粘性係数４．０[cP]における流量−電流特性図である。
【図１４】 粘性係数３．０[cP]の流量０[L/min]における角速度−損失電流特性図である。
【図１５】 流量一般化電流特性図である。
【図１６】 角速度−傾き特性図である。
【図１７】 一般化流量−一般化電流特性図である。
【図１８】 各粘性係数の一般化流量−一般化電流特性図である。
【図１９】 切片−粘性係数特性図である。
【図２０】 粘性係数２．０[cP]における流量−圧力差特性図である。
【図２１】 粘性係数３．０[cP]における流量−圧力差特性図である。
【図２２】 粘性係数４．０[cP]における流量−圧力差特性図である。
【図２３】 角速度−圧力特性図である。
【図２４】 流量−一般化圧力特性図である。
【図２５】 一般化流量−一般化圧力差特性図である。
【図２６】 各粘性係数の一般化流量−一般化圧力特性図である。
【図２７】 粘性を考慮した流量推定式の精度図である。
【図２８】 従来の流量推定式の精度図である。
【図２９】 粘性を考慮した圧力推定式の精度図である。
【図３０】 小型コントロールシステムの推定システムブロック図である。
【図３１】 実験に用いた超音波振動子の外形を示す斜視図である。
【図３２】 電圧波形測定用実験装置の構成図である。
【図３３】 グリセリン１００％で測定した電圧波形を示す図である。
【図３４】 グリセリン５０％水溶液で測定した電圧波形を示す図である。
【図３５】 水で測定した電圧波形を示す図である。
【符号の説明】
１ 遠心ポンプ
２ 血液入口
３ 血液出口
４ 磁石
５ インペラ
１０，２０，３０，４０ 遠心ポンプ駆動装置
１１ 駆動部
１２ 電流計
１３，２４，３２，４２ 流量計
２１，２２ 圧力計
３１ 第１の補正手段
４１ 第２の補正手段[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a centrifugal pump drive device used for extracorporeal circulation and the like, and more particularly to a centrifugal pump drive device equipped with an inexpensive flow meter.
[0002]
[Prior art]
The blood pump is a pump for an open heart cardiopulmonary device that is used to return blood to the patient once it has been removed in the extracorporeal circulation. In the auxiliary artificial heart device, the pumping action of the patient's heart itself is used as an auxiliary agent. It is used for etc.
Conventionally, most of the blood pumps used for open heart surgery for these purposes have been roller pumps, but recently, the use of centrifugal pumps has increased due to the strong demand for safety. Centrifugal pumps were originally used for auxiliary circulation and extracorporeal circulation, but (1) no dangerous high pressure was generated, (2) high safety against air contamination, and (3) no excessive negative pressure. , (4) There is no need to adjust the degree of closure before use like a roller pump, (5) Blood cell damage is less than a roller pump, (6) Abrasion powder due to rubbing of the inner surface of the pump tube, seen in a roller pump (7) Many models with small drive units and built-in batteries are excellent in portability, and there are many facilities in Japan that are also used for general open heart surgery and extracorporeal circulation. It has increased.
[0003]
FIG. 1 is a diagram illustrating an example of a centrifugal pump, and this centrifugal pump includes a centrifugal pump 1 and a drive unit (not shown). The centrifugal pump 1 is hermetically disposed in a space having a blood inlet 2 and a blood outlet 3, and includes a magnet 4 that is attracted by a magnet of a driving unit, and rotates by rotation of the magnet of the driving unit to rotate the blood inlet 2. An impeller 5 for sending blood toward the blood outlet 3 is provided. The impeller 5 is rotatably supported by two-point support by a top pivot bearing 7 that is in contact with the tip of a shaft 6 that is formed to protrude from the top, and a bottom pivot bearing 8 that is in contact with a bottom protrusion. A first vane 5a is provided on the upper surface of the impeller, and a second vane 9 is provided on the lower surface. When the impeller 5 rotates in a certain direction, blood, a clear prime solution, or the like is directed from the blood inlet 2 to the blood outlet 3. The liquid is supposed to be sent.
The drive unit for driving the centrifugal pump 1 includes a magnet for strongly attracting the magnet 4 of the centrifugal pump 1 attached to the upper portion, and a motor for rotating the magnet. When the motor of this drive unit is driven, the impeller rotates at the same rotational speed, the impeller 5 of the centrifugal pump 1 attached to the upper portion rotates, and the liquid is sent from the blood inlet 2 toward the blood outlet 3. The magnet 4 may be replaced with a low-cost magnetic material as long as the rotational speed of the magnet is the same as that of the drive unit.
[0004]
Even if this centrifugal pump is operated at the same rotation speed, the flow rate changes when the pressure load fluctuates. Therefore, it is necessary to provide a separate flow meter to know the pump flow rate. It is necessary to monitor the meter and operate the rotation speed appropriately. 2. Description of the Related Art Conventionally, there has been provided a centrifugal pump drive device including a controller for measuring the flow rate of a centrifugal pump and appropriately adjusting the number of rotations of a motor so as to obtain an arbitrary flow rate. Furthermore, in this type of centrifugal pump drive device, an ultrasonic flow meter and an electromagnetic flow meter are known as flow meters for measuring the flow rate.
[0005]
[Problems to be solved by the invention]
However, these conventional centrifugal pump drive device flow meters have the following problems.
The ultrasonic flowmeter cannot measure the flow rate of the clear prime liquid due to its measurement principle, and the accuracy of the low flow rate region of 1 L / min or less is poor. Also, currently available devices are expensive. Furthermore, a flow sensor is required for each tube size. In addition, there are problems such as a large measurement error of about 5 to 10%.
Electromagnetic flow meters are also more expensive than ultrasonic flow meters. Furthermore, there is a problem that a flow sensor is required for each tube size. The measurement error is about 5%, which is better than the ultrasonic flowmeter.
[0006]
The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a centrifugal pump driving device provided with a flow rate measuring means that is less expensive than conventional ultrasonic flowmeters and electromagnetic flowmeters.
Another object of the present invention is to provide a centrifugal pump driving device capable of suggesting replacement timing due to deterioration of the centrifugal pump over time.
Still another object of the present invention is to provide a centrifugal pump driving device including a flow meter that can accurately measure the flow rates of both the clear prime solution and blood by a switching operation.
Still another object of the present invention is to provide a centrifugal pump driving device including a flow meter capable of accurately performing flow measurement according to various centrifugal pumps having different loss currents by switching operation.
[0007]
[Means for Solving the Problems]
  In order to achieve the above-mentioned object, the present invention encloses and arranges in a space having a liquid inlet and a liquid outlet, and rotates at the same rotation speed as that of the magnet arranged adjacent to the outside. A centrifugal pump provided with an impeller connected to or embedded in a magnet or a magnetic body for sending blood from the liquid inlet to the liquid outlet, and a magnet and a motor for rotationally driving the impeller of the centrifugal pump A centrifugal pump drive device having a drive unit, having an ammeter for measuring a current value of a motor of the drive unit and a tachometer for measuring a rotation speed;And saidA pressure gauge for measuring the inlet pressure and the outlet pressure of the centrifugal pump;Calculate the calculated differential pressure value from the flow rate, the inlet pressure and the outlet pressure of the centrifugal pump from the current value and the number of rotations, measure the measured differential pressure value from the measured values of the inlet pressure and the outlet pressure of the centrifugal pump, The centrifugal pump drive device according to claim 1, wherein a difference between the pressure difference and the calculated pressure difference can be calculated.I will provide a.
[0009]
Further, the centrifugal pump driving device of the present invention may include a first correction unit that corrects the flow rate calculation value by inputting or switching the viscosity coefficient information of the liquid flowing through the centrifugal pump.
[0010]
Furthermore, the centrifugal pump drive device of the present invention can include a second correction unit that corrects the flow rate calculation value by inputting or switching the loss current value of the centrifugal pump.
[0011]
DETAILED DESCRIPTION OF THE INVENTION
  FIG. 2 shows a centrifugal pump drive device according to the present invention.Reference exampleFIG. The centrifugal pump driving device 10 is hermetically disposed in a space having a blood inlet 2 and a blood outlet 3, and is rotated by rotation of a magnet (not shown) in the driving unit 11 to move from the blood inlet 2 to the blood outlet 3. A centrifugal pump 1 having an impeller 5 for sending blood, a drive unit 11 having a magnet and a motor (not shown) for rotationally driving the impeller 5 of the centrifugal pump 1, and a current value of the motor of the drive unit 11 An ammeter 12 for measuring and a tachometer (not shown) for measuring the rotation speed are provided, and a flow meter 13 for calculating a flow rate, and a differential pressure between the inlet pressure and the outlet pressure of the centrifugal pump is provided from the current value and the rotation speed It has a configuration.
[0012]
The flow meter 13 of the centrifugal pump driving device 10 is configured to determine the flow rate and the centrifugal pump 1 based on the current value of the motor of the driving unit 11 (proportional to torque: excluding the loss current of the motor including the centrifugal pump 1) and the rotational speed (rpm). It is possible to calculate the differential pressure between before and after (that is, blood inlet 2 and blood outlet 3). Further, in addition to the flow meter 13, a rotation setting unit (not shown) for controlling the rotation speed of the motor is provided so that the rotation speed of the motor can be controlled so that the calculated flow rate and / or differential pressure becomes arbitrary. Yes. If the accuracy and reliability of the flow rate and other additional functions to be described later are not required, the purpose of the flow rate measurement can be sufficiently achieved with this simple configuration.
[0013]
This flow meter 13 can be configured to display the rotational speed of the measured current value and rotational speed, and to easily calculate the flow rate or the flow rate and the differential pressure based on the data, and further to the measured current value and rotational speed. It is possible to provide a display function to arbitrarily display the flow rate and differential pressure based on the number, and automatically control the motor rotation speed so that the calculated flow rate and / or differential pressure is within a certain range. An automatic control mechanism can also be provided.
[0014]
  FIG. 3 shows a centrifugal pump drive device according to the present invention.First embodimentFIG. The centrifugal pump drive device 20 is configured to include the same components as the centrifugal pump drive device 10 shown in FIG. 2, and pressure gauges 21 and 22 that measure the pressure at the blood inlet 2 and blood outlet 3 of the centrifugal pump 1. The flow meter 24 includes an ammeter 12 that measures the current value of the motor of the drive unit 11, a tachometer (not shown) that measures the rotation speed, and pressure measurement data of the pressure gauges 21 and 22, or The pressure display unit 23 displays the differential pressure on the blood inlet 2 side and the blood outlet 3 side and / or the blood inlet pressure and the blood outlet pressure.
[0015]
  The aboveReference exampleThe centrifugal pump drive apparatus 10 according to the above does not directly monitor the blood flow. ThereforeFirstIn the embodiment, the blood flow monitor is measured by the versatile pressure gauges 21 and 22, and the differential pressure between the blood inlet 2 side and the blood outlet 3 side can be detected. As the pressure gauges 21 and 22, a pressure sensor having a small and simple structure can be used. As described above, the differential pressure calculated from the measurement results of the ammeter 12 and the tachometer(Calculated differential pressure)And the differential pressure measured on the blood inlet 2 side and blood outlet 3 side separately.(Measured differential pressure)Can be verified and calibrated. In addition, reliability can be improved by performing actual measurement of blood flow independently. Furthermore, since the difference between the measured differential pressure value and the calculated differential pressure value increases due to an increase in the rotational resistance value due to deterioration over time of the bearing that supports the impeller 5 of the centrifugal pump 1, it is possible to suggest the replacement time of the centrifugal pump 1. .
[0016]
  FIG. 4 shows a centrifugal pump drive device according to the present invention.Second embodimentFIG. The centrifugal pump drive device 30 is configured to include the same components as the centrifugal pump drive device 20 shown in FIG. 3, and the flow meter 32 rotates with the ammeter 12 that measures the current value of the motor of the drive unit 11. A tachometer (not shown) for measuring the number, pressure display data 23 for displaying pressure measurement data of the pressure gauges 21 and 22 or a differential pressure between the blood inlet 2 side and the blood outlet 3 side, and a liquid flowing in the centrifugal pump 1 The first correction means 31 for correcting the flow rate value by inputting or switching the viscosity coefficient information is provided.
[0017]
The first correction means 31 compares the viscosity coefficient information input by using an input means such as a keyboard attached to the flow meter 32 or connected to the flow meter 32 with previously recorded viscosity coefficient information, Correction can be made so that an optimum flow rate is displayed for the viscosity. As a result, the centrifugal pump driving device 30 of the present embodiment can accurately measure the flow rate of not only blood but also a liquid such as a clear prime liquid having a viscosity different from that of the blood.
The viscosity coefficient information includes Ht (hematocrit value), Hb (hemoglobin amount), TP (total protein amount) and the like in addition to the viscosity coefficient value of blood or other liquids.
In the present embodiment, as a simple switching method by the first correction unit 31, for example, a method of switching by distinguishing about three stages (0 to 10%, 10 to 25%, 25% to) of the Ht value is adopted. it can.
[0018]
  FIG. 5 shows a centrifugal pump drive device according to the present invention.Third embodimentFIG. The centrifugal pump drive device 40 is configured to include the same components as the centrifugal pump drive device 30 shown in FIG. 4, and the flow meter 42 rotates with the ammeter 12 that measures the current value of the motor of the drive unit 11. A tachometer (not shown) for measuring the number, pressure display data 23 for displaying pressure measurement data of the pressure gauges 21 and 22 or a differential pressure between the blood inlet 2 side and the blood outlet 3 side, and a liquid flowing in the centrifugal pump 1 The first correction unit 31 corrects the flow rate value by inputting or switching the viscosity coefficient information, and the second correction unit 41 correcting the flow rate value by inputting or switching the loss current value of the centrifugal pump drive device. ing.
[0019]
As the current value, the current value excluding the loss current of the motor including the centrifugal pump 1 is used. This current value is a fixed value based on a combination of various driving units 11 and various centrifugal pumps 1. In order to calibrate the difference between the fixed values, the fixed value by the centrifugal pump 1 from each manufacturer is memorized, and the second value is used for new use and when the centrifugal pump 1 is replaced with another type. By inputting a new fixed value of the centrifugal pump 1 corresponding to the correcting means 31 or switching the displayed product number display of the centrifugal pump 1, etc., it can be used for various centrifugal pumps 1 and is calibrated by the fixed value. Accurate flow rate values can be calculated and displayed. The drive unit 11 can cope with the use of an adapter, the exchange of an external motor, etc., as in the present situation.
[0020]
As described above, according to the present invention, it is possible to provide a centrifugal pump driving device including a flow rate measuring unit that is less expensive than conventional ultrasonic flowmeters and electromagnetic flowmeters.
Moreover, according to this invention, the centrifugal pump drive device which can suggest the replacement time by the time-dependent deterioration of a centrifugal pump can be provided.
Furthermore, it is possible to provide a centrifugal pump driving device including a flow meter that can measure the flow rates of both the clear prime solution and blood by a switching operation.
Furthermore, it is possible to provide a centrifugal pump driving device including a flow meter that can accurately measure the flow rate according to various centrifugal pump driving devices having different loss currents by switching operation.
[0021]
In addition, when the flow rate of the centrifugal pump decreases, the conventional flow meter only gives a flow rate lower limit alarm display, but in the present invention, since the differential pressure is continuously measured, is the cause of the decrease in blood flow caused by poor blood removal? It is easy to guess whether blood pressure has risen, and quick response is possible.
In particular, in PCPS (percutaneous cardiopulmonary assist system), blood pressure and blood pressure are continuously measured. Therefore, it is possible to determine whether the cause of the decrease in the flow rate is poor blood removal or an increase in blood pressure, which is extremely effective.
[0022]
【Example】
[System Overview]
A centrifugal pump drive device was manufactured using a commercially available centrifugal pump, a drive unit, and a controller (flow meter) prepared for this example, and its performance was examined.
[0023]
(Centrifugal pump)
A two-point supported small centrifugal pump (C1E3) employing a pivot bearing manufactured by Kyocera Corporation was used. The flow rate and pressure characteristics of this centrifugal pump are shown in FIG. 6 and the specifications are described below.
・ Specification of centrifugal pump (C1E3)
Outer diameter [mm]: 86
Height [mm]: 48
Weight [g]: 124
Internal capacity [ml]: 40
Rotation speed [rpm]: Max4000
Discharge pressure [mmHg]: 500
Discharge flow rate [L / min]: 10
Inflow and discharge port size [inch]: 3/8
[0024]
(Drive part)
A DC brushless motor manufactured by Softronics was used as the drive motor and the dedicated driver. The specifications of this motor are as follows.
・ Motor specifications
Power supply voltage: DC24V
Motor type: M028 type
Rated output: 45W
Rated speed: 3000rpm
Rated torque: 1.5kgfcm
Rated current: 3A
Maximum current: 5A
Motor type: 3-phase DC brushless motor
Motor structure: outer rotor, star connection
Number of magnetic poles: 4 poles
Number of slots: 6 slots
Rotor position detection: Hall IC
Insulation class: Class E (120 ° C)
Protection function: Overload protection (thermal protector)
Specification environment: Temperature 0-40 ° C (no freezing)
Humidity 85% RH or less (no condensation)
Insulation resistance: 100MΩ or more at 500 VDC megger (between case and coil)
Dielectric strength: AC1000V for over 1 minute (between case and coil)
[0025]
(Controller (flow meter))
As shown in FIG. 2, the controller (flow meter) used in this embodiment can measure the current value and the rotational speed of the motor of the drive unit, and can change the rotational speed of the motor by adjusting the knob. Possible structure. The controller used is 320 g by itself, and the total weight of the centrifugal pump and drive unit is 1290 g, which is very light and compact, so it can be easily carried.
[0026]
[Example 1: Flow pressure estimation]
1. Relationship between concentration of glycerin aqueous solution and viscosity coefficient
FIG. 7 shows a relationship diagram of the hematocrit value indicating the ratio of the red blood cell component in the blood, the viscosity coefficient, and the temperature. From FIG. 7, it was confirmed that the viscosity coefficient of blood changed significantly as the hematocrit value and temperature changed.
Next, an experiment was conducted on the relationship between the concentration of the glycerin aqueous solution and the viscosity coefficient in order to strictly adjust the viscosity coefficient of the glycerin aqueous solution and improve the flow rate pressure estimation accuracy.
A B-type viscometer (4840; Tokimec) was used for measuring the viscosity coefficient.
[0027]
The procedure of the measurement experiment of the viscosity coefficient at the concentration of each glycerin aqueous solution is shown.
(1) The density and weight of water and glycerin were taken into consideration to adjust the concentration of each glycerin aqueous solution, and 10 sample glycerin aqueous solutions were prepared.
(2) The temperature of the glycerin aqueous solution was adjusted to 30 ° C.
(3) The viscosity coefficient was measured using a viscometer.
(4) A graph of the concentration-viscosity coefficient of each glycerin aqueous solution was prepared.
(5) A relational expression was calculated from the graph of concentration-viscosity coefficient of each glycerin aqueous solution by a polynomial approximation.
[0028]
FIG. 8 shows a relationship diagram of the concentration-viscosity coefficient of each glycerol aqueous solution at 30 ° C.
The relational expression calculated from this graph is shown in the following expression (1).
M = 0.0032n²-0.511n + 3.6599 (1)
(In the formula, M is the glycerin aqueous solution concentration (%), and n is the viscosity coefficient (cP)).
By considering the value of the viscosity coefficient in calculating the estimation formula, it is possible to improve the accuracy of the flow pressure estimation. For this reason, the glycerin aqueous solution close to the blood change was used in the experiment, and as a result of conducting an experiment on the relationship between the concentration of the glycerin aqueous solution and the viscosity coefficient, the viscosity coefficient was calculated from the calculated concentration-viscosity coefficient relationship of each glycerin aqueous solution. Experiments that take this into account became possible.
[0029]
2. Measurement of loss current of motor alone
In this embodiment, the motor rotation speed and torque are used to calculate the estimation formula. In general, torque generated in a DC brushless motor is proportional to the value of current flowing through the motor due to the characteristics of the motor. Therefore, we decided to use the current value instead of the torque value to calculate the estimation formula in this study. The current that flows when the motor is driven is the sum of the current value that drives the motor itself and the current value that operates the centrifugal pump. Therefore, it is possible to calculate a current value for operating the centrifugal pump, that is, a torque, by subtracting the loss current of the motor alone from the current value at the time of driving the motor. In this experiment, the loss current of the motor alone is measured.
[0030]
An experimental procedure for calculating the loss current of the motor alone from the angular velocity is shown.
(1) The motor was arbitrarily rotated in the range of 0 to 3500 [rpm] in a no-load state.
(2) The current value at each rotational speed was measured.
(3) A graph of angular velocity versus loss current was prepared.
(4) A relational expression of angular velocity-loss current was calculated.
[0031]
The measurement result of the motor loss current value at no load is shown in FIG.
The relational expression between the loss current and the angular velocity from the measurement result of the motor loss current at no load is shown in the following expression (2).
I = 0.0008ω + 0.0416 (2)
(In the formula, ω represents the angular velocity [rad / s] of the motor, and I represents the loss current [A] of the motor alone.)
[0032]
From this experimental result, a relational expression between the motor angular velocity and the loss current of the motor alone was obtained. By using this equation (2), it is possible to determine the loss current value of the motor at each rotational speed. Thus, by subtracting this loss current value from the current value at the time of the experiment using the glycerin aqueous solution, it became possible to take out the current change due to the increase or decrease of the torque.
[0033]
3. Calculation of flow rate estimation formula
In order to realize a simple circuit configuration, an attempt was made to estimate the flow rate from the motor current value and the rotation speed when the centrifugal pump was driven. In the estimation formula calculation, the current value used for obtaining the torque necessary for the operation of the centrifugal pump obtained by subtracting the loss current of the motor alone described above was used. In addition, the viscosity coefficient of the aqueous glycerin solution used in the experiment was adjusted using the above-described equation (1). Table 1 shows an example of a conversion table for the viscosity coefficient to the concentration of the glycerin aqueous solution.
[Table 1]

[0034]
A schematic diagram of the circulation circuit used in the experiment is shown in FIG. In FIG. 10, reference numeral 1 is a centrifugal pump, 11 is a drive unit including a motor, 50 is an electromagnetic blood flow meter (MFV-3200; manufactured by Nihon Kohden), 51 is a pressure gauge (AP-611G; manufactured by Nihon Kohden), 52 is a reservoir for storing a liquid (glycerin aqueous solution), and 53 is a clamp for adjusting the flow rate.
The glycerin aqueous solution of each concentration was filled in the experimental circulation circuit, and the temperature was kept at 30 ° C. During the circulation experiment, the flow rate was changed by adjusting the clamp while keeping the rotation speed constant.
The flow-current characteristic experiment procedure performed for calculating the flow rate estimation formula is shown.
(1) A glycerin aqueous solution adjusted to a viscosity coefficient of 2.0 [cP], 3.0 [cP], and 4.0 [cP] was filled in an experimental circuit.
(2) The number of revolutions was changed in increments of 500 rpm in the range of 1000 to 3500 rpm, and kept constant at each number of revolutions.
(3) The flow rate was changed in increments of 1 [L / min] from 0 [L / min] to the maximum flow rate at each rotation speed, and the current value at this time was measured.
FIG. 11 shows a flow rate-current characteristic diagram at a viscosity coefficient of 2.0 [cP].
A flow rate-current characteristic diagram at a viscosity coefficient of 3.0 [cP] is shown in FIG.
A flow rate-current characteristic diagram at a viscosity coefficient of 4.0 [cP] is shown in FIG.
[0035]
From the flow-current characteristic diagrams of FIGS. 11 to 13, the motor characteristics showed the same slope in the experiments for each viscosity coefficient, but the current value at a flow rate of 0 [L / min], which is an intercept of the graph, is the viscosity coefficient. It can be confirmed that the value increases even at the same rotational speed if becomes larger. Therefore, it was considered that an accurate viscosity coefficient should be taken into account for the flow pressure estimation.
Considering the above, an axial conversion process was performed for the flow axis direction and the current axis direction using the angular velocity. By this axis conversion, the measurement data can be expressed on a straight line, and a flow rate estimation formula can be calculated.
As an example, a calculation process of a flow rate estimation formula will be described using a flow rate-current characteristic diagram (FIG. 12) having a viscosity coefficient of 3.0 [cP].
For the conversion of the current axis, focusing on the current value at a flow rate of 0 [L / min] in FIG. 12, the characteristics of the current value with respect to the rotational speed (angular velocity) were examined. FIG. 14 shows an angular velocity-loss current characteristic diagram when the viscosity coefficient is 3.0 [cP] and the flow rate is 0 [L / min].
[0036]
FIG. 14 shows the angular velocity-loss current characteristic diagram at the flow rate 0 [L / min] with a viscosity coefficient of 3.0 [cP] in the following equation (3).
I = 2.0 × 10^-06× ω^2.308    (3)
R2 = 0.9998
(Where I is the centrifugal pump loss current [A], and ω is the angular velocity [rad / s])
[0037]
FIG. 14 and equation (3) show that the pump loss current at a flow rate of 0 [L / min] with a viscosity coefficient of 3.0 [cP] increases to the rotational angular velocity ω to the power of 2.308. . Therefore, by dividing each intercept of the loss current value at a flow rate of 0 [L / min] by the power of 2.308, it can be concentrated on one point on the axis. FIG. 15 shows the result of performing the operation of dividing the angular velocity by the power of 2.308 on all data.
FIG. 15 shows that the inclination becomes smaller as the rotational speed increases. Therefore, it is possible to make all the slopes the same by dividing the flow value of each data by the power of the angular velocity. A characteristic diagram of each angular velocity and inclination is shown in FIG.
[0038]
From FIG. 16, the relational expression between the angular velocity and the inclination is expressed as the following equation (4) using the angular velocity ω, where y is an intercept.
y = 63.139ω^-0.7079    (4)
From this equation (4), it is shown that the inclination is reduced to the 0.7079th power of the angular velocity due to the increase of the angular velocity. Thus, by dividing the flow rate value by the 0.7079th power of the angular velocity, all the slopes are matched.
All data can be expressed in a straight line by the axis conversion operation in the flow direction and the current direction. FIG. 17 shows the result of axial conversion performed on all data in the flow rate-current characteristic diagram having a viscosity coefficient of 3.0 [cP].
[0039]
When the characteristic diagram expressed on a straight line in FIG. 17 is obtained by a linear approximation formula, the following formula (5) is obtained. This formula is a flow rate estimation formula at a viscosity coefficient of 3.0 [cP].
I '= 0.3062Q' + 17.445 (5)
By performing the same axis conversion operation on the flow rate-current characteristic diagram of each viscosity coefficient, it is possible to obtain a flow rate estimation formula for each viscosity coefficient.
Table 2 shows the power values of the axial transformations at this time.
[Table 2]

[0040]
From Table 2, using the average value of the power of each viscosity coefficient, the axis conversion formula in the direction of flow and current is shown as follows, and the result of axis conversion to the flow-current characteristic diagram of each viscosity coefficient again: Is shown in FIG.
Current direction axis conversion formula I '= I / ω^2.19× 10⁷
Axis conversion formula in flow direction Q '= Q / ω^0.63× 200
[0041]
From FIG. 18, the flow rate estimation formula for each viscosity coefficient was calculated as follows.
Flow rate estimation formula with viscosity coefficient 2.0 [cP]
I ’_2.0= 0.415Q '+ 30.4
Flow rate estimation formula with viscosity coefficient 3.0 [cP]
I ’_3.0= 0.415Q '+ 33.0
Flow rate estimation formula with viscosity coefficient 4.0 [cP]
I ’_4.0= 0.4234Q '+ 35.1
[0042]
From the above, it was confirmed that the slope of each flow rate estimation formula hardly changed due to the difference in viscosity coefficient, and that when only the intercept was 30.4, 33.0, 35.1 proportional to the slope, it changed greatly. For this reason, the gradient average value of the estimation formula of each viscosity coefficient is used as the gradient of the flow rate estimation formula. In addition, since the intercept changes with changes in the viscosity coefficient, it was considered necessary to consider the viscosity coefficient. FIG. 19 shows the relationship between the viscosity coefficient and the intercept.
[0043]
From FIG. 19, the relational expression between the intercept and the viscosity coefficient is expressed as the following expression (6).
M = 0.0032n²-0.511n + 3.6599 (6)
(In the formula, M represents the glycerin aqueous solution concentration [%], and n represents the viscosity coefficient [cP])
From the above, the relational expression between the generalized flow rate and the generalized current considering the viscosity coefficient is shown in the following formula.
I '= 0.4181Q' + (2.2895n + 26.013)
(Empirical formula for flow estimation)
Q '= (I'-(2.2895n + 26.013)) / 0.4181 (7)
(Where I is current [A], Q is flow rate [L / min], and n is viscosity coefficient [cP])
[0044]
As described above, the flow rate estimation formula was calculated from the experimental results (FIGS. 11 to 13) of the flow rate-current characteristic diagrams.
FIG. 18 shows that the slope of the estimation equation hardly changes and only the intercept changes due to the change of the viscosity coefficient. For this reason, the gradient of the flow rate estimation formula was determined by using the average value of the estimation formulas of the respective viscosity coefficients, and the intercept was calculated using formula (6). This indicates that it is preferable to measure and take into account the viscosity coefficient of the liquid, rather than measuring the flow rate of the centrifugal pump drive device only from the rotational speed and current value.
[0045]
4). Calculation of pressure estimation formula
An attempt was made to calculate the pressure estimation formula from the flow rate and the rotational speed. In the pressure estimation formula calculation, the flow rate estimated as described above was used as the flow rate. In addition, the viscosity coefficient of the aqueous glycerin solution used in the experiment was adjusted using the above-described equation (1).
[0046]
The glycerin aqueous solution of each concentration was filled in the experimental circulation circuit and the temperature was kept at 30 ° C. The experimental circuit is shown in FIG.
During the circulation experiment, the flow rate was changed by adjusting the clamp while keeping the rotation speed constant. The experimental procedure is as follows.
(1) A glycerin aqueous solution adjusted to a viscosity coefficient of 2.0 [cP], 3.0 [cP], and 4.0 [cP] was filled in an experimental circuit.
(2) The rotational speed was changed in increments of 500 [rpm] from 1000 to 3500 [rpm] and kept constant at each rotational speed.
(3) The flow rate was changed to the maximum pressure in increments of 1 [L / min] at each rotation speed, and the inlet pressure and the outlet pressure of the centrifugal pump at this time were measured to obtain a pressure difference.
A flow rate-pressure characteristic diagram when the viscosity coefficient is 2.0 [cP] is shown in FIG.
FIG. 21 shows a flow rate-pressure characteristic diagram when the viscosity coefficient is 3.0 [cP].
FIG. 22 shows a flow rate-pressure characteristic diagram when the viscosity coefficient is 4.0 [cP].
[0047]
From the flow rate-pressure characteristic diagrams of FIGS. 20 to 22, axial conversion processing was performed for the flow rate axis direction and the current axis direction using the angular velocity obtained from the rotation speed. As a result, the measured values can be represented on a straight line, and the pressure estimation formula can be calculated.
The calculation process of the pressure estimation formula will be described using a flow rate-pressure difference characteristic diagram (FIG. 21) with a viscosity coefficient of 3.0 [cP].
For the conversion of the pressure axis, focusing on the pressure difference at a flow rate of 0 [L / min] in FIG. 21, the characteristics of the pressure difference with respect to the rotational speed (angular velocity) were examined. FIG. 23 shows an angular velocity-pressure difference characteristic diagram at a flow rate of 0 [L / min] with a viscosity coefficient of 3.0 [cP].
[0048]
From FIG. 23, the following equation (8) shows an angular velocity-pressure difference characteristic diagram at a flow rate of 0 [L / min] with a viscosity coefficient of 3.0 [cP].
P = 0.0021 × ω^2.1439    (8)
R2 = 1.0
(In the formula, P represents a pressure difference [mmHg], and ω represents an angular velocity [rad / s].)
[0049]
From this equation (8), it is shown that the pressure difference at a flow rate of 0 [L / min] with a viscosity coefficient of 3.0 [cP] increases to the 2.1439th power of the rotational angular velocity ω. Therefore, by dividing each intercept of the pressure difference at a flow rate of 0 [L / min] by the power of 2.439, the intercept can be concentrated at one point. FIG. 24 shows the result of performing the operation of dividing the angular velocity by the power of 2.439 on all data. This is an axis conversion in the pressure direction and can be expressed as the following equation.
P ′ = P / ω^2.05
(Where P ′ is the pressure difference after axis conversion, and ω is the angular velocity)
[0050]
In FIG. 24, axis conversion is performed in the flow direction. However, in the case of pressure, since there is no inclination as in the case of the flow rate conversion, the highest value of the axial conversion power is used. As a result, it was shown that it was represented on the same curve by dividing by 1.1 to the flow direction. FIG. 25 shows the result of axial conversion performed on all the data in the flow rate-pressure difference characteristic diagram having a viscosity coefficient of 3.0 [cP].
From the characteristic diagram expressed on the same curve in FIG. 25, when the characteristic equation is obtained by a polynomial approximation equation, it is expressed as the following equation. This formula is a pressure estimation formula at a viscosity coefficient of 3.0 [cP].
P ′ = 0.00007ω²-0.00076ω + 20.568
R²= 0.9581
[0051]
By performing a similar axis conversion operation on the flow rate-pressure difference characteristic diagram of each viscosity coefficient, it is possible to obtain a pressure estimation formula for each viscosity coefficient.
Table 3 shows the power values of the axis conversion in each axis direction at this time. FIG. 26 shows the result of axial conversion to the flow rate-pressure difference characteristic diagram of each viscosity coefficient again using the average value of the power values of the viscosity coefficients.
[Table 3]

[0052]
Flow rate estimation formula with viscosity coefficient 2.0 [cP]
P ’_2.0= -0.0003Q ''²+ 0.0036Q '' + 33.68
Flow rate estimation formula with viscosity coefficient 3.0 [cP]
P ’_3.0= -0.0002Q ''²+ 0.0128Q '' + 33.79
Flow rate estimation formula with viscosity coefficient 4.0 [cP]
P ’_4.0= -0.0002Q ''²+ 0.0048Q '' + 34.18
[0053]
As shown in FIG. 26, the characteristics of each viscosity coefficient after axis conversion can be expressed in substantially the same curve, so it is considered unnecessary to consider the difference in characteristics in each viscosity coefficient in the estimation formula. Therefore, the average of the characteristic equations for each viscosity coefficient was used as the pressure estimation equation.
Empirical formula for pressure estimation (10)
P '=-0.0002Q' '²+ 0.007Q '' + 33.88 (10)
[0054]
As described above, the pressure estimation formula (10) was calculated from the experimental result of the flow rate-pressure difference characteristic diagram. As the pressure estimation formula, the flow rate estimated in “3. Calculation of flow rate estimation formula” was used. Therefore, if a large estimation error occurs in the flow rate, it will affect the pressure estimation, and it is considered that an estimation error occurs.
[0055]
5. Evaluation experiment of flow pressure estimation formula using bovine blood
As an evaluation experiment of the flow pressure estimation equation, an evaluation equation evaluation experiment was actually performed using bovine blood in the same experimental circuit as when the glycerin aqueous solution was used.
The experimental procedure is as follows.
(1) Adjust bovine blood to a hematocrit value of 40% at 30 ° C. and fill the experimental circuit.
(2) The estimation formula is evaluated at a rotational speed of 1000 to 3500 [rpm].
[0056]
As an evaluation method of the flow rate estimation formula, the estimated flow rate value with respect to the measured flow rate value was plotted on a graph, and the deviation from the value of y = x was compared. FIG. 27 shows an accuracy diagram of the flow rate estimation equation, and FIG. 28 shows an accuracy diagram of the flow rate estimation equation so far. FIG. 29 shows an accuracy diagram of the pressure estimation formula. Tables 4 and 5 show comparison tables of flow rate and pressure estimation accuracy with the conventional estimation formula.
[Table 4]

[Table 5]

[0057]
In the flow rate estimation formula evaluation experiment using bovine blood, it was shown that high accuracy was obtained with an average flow rate accuracy of 2.0% and a maximum estimated flow rate error of 0.56 [L / min]. As a factor of flow rate estimation error, it is considered that blood changes in viscosity due to factors such as circulating temperature when compared with a glycerin aqueous solution. However, considering the viscosity coefficient, the accuracy of the estimation formula has been greatly improved compared to the conventional formula, indicating the importance of the viscosity coefficient in the flow rate estimation.
In the current system, it is difficult to measure the viscosity coefficient of blood that changes during circulation. Therefore, the measurement accuracy has been improved by inputting or switching the viscosity coefficient information. It is desirable to be able to know.
In the pressure estimation formula evaluation experiment using bovine blood, the average pressure accuracy was 5.8%, and the maximum estimated pressure error was 30.7 [mmHg]. As the pressure estimation formula, the flow rate estimated in “3. Calculation of flow rate estimation formula” is used. Therefore, if an estimation error occurs in the flow rate, it is considered that an estimation error also occurs in the pressure estimation.
[0058]
[Example 2: Small control system]
1. Estimated flow pressure display system
Using an 8-bit microcomputer, the current value and rotation speed, which are estimation parameters, are taken into the microcomputer in real time, and the flow rate can be displayed during blood circulation using the flow rate estimation formula calculated and evaluated in Example 1 above. Prototype of a small system. The viscosity coefficient of the experimental solution was measured and input in advance using a viscometer.
A block diagram of the estimation system is shown in FIG.
[0059]
The DC brushless motor used in the drive unit has a built-in hall sensor for detecting the rotor position. Since the output from the Hall sensor is a square wave, the number of outputs can be measured by using a frequency meter. In this motor, since there is an output from the Hall sensor twice during one rotation, it can be converted into the number of revolutions per minute [rpm] by multiplying the frequency displayed on the frequency meter by 30. The frequency meter used was “PIC16C71 liquid crystal display frequency counter kit Ver.2” manufactured by Akizuki Dentsu.
[0060]
As described above, a prototype of a flow rate estimation system using an 8-bit microcomputer was made. The prototype display unit has made it possible to realize a small centrifugal pump drive controller that can be used substantially as a flow meter. In addition, if it becomes possible to detect the viscosity coefficient that changes during circulation in real time and incorporate it into the estimation equation, it is considered that more accurate estimation can be performed.
[0061]
2. Examination of online viscometer
In Example 1, it was shown that the estimation of the flow rate is affected by the viscosity coefficient of blood. Therefore, an online viscometer that can measure the viscosity coefficient in real time with an ultrasonic transducer using a magnetostrictive element was investigated.
Magnetostrictive transducers have long been used for the generation and detection of ultrasonic waves as electroacoustic transducers that use the magnetostriction phenomenon of magnetic materials, and use acoustic sounders, fish finders, sonars, or ultrasonic waves. Widely used for cleaning, emulsification, dispersion, sterilization, machining, welding, etc.
When the demagnetized ferromagnetic material is magnetized, the outer shape slightly changes. This phenomenon is called magnetostriction. When a round iron bar is magnetized along the length l, it extends by δl. This ratio δl / l is represented by magnetostriction λ.
λ = δl / l
This λ is 10 for iron.^-FiveThe sign is positive. The effect due to magnetostriction in the direction of the magnetic field is referred to as the longitudinal effect, and the magnetostriction perpendicular to the magnetic field is referred to as the transverse effect. Magnetostriction has a history for magnetic fields, but is a monovalent function for magnetization. The magnetostriction phenomenon was first discovered by Joule, also called the Joule effect. This adverse effect is called the “biliary effect”.
The magnetostrictive vibrator includes a square vibrator and an annular vibrator. In the case of using a metal magnetostrictive material, it is made thin to reduce eddy current and is laminated and fixed by surface insulation, but this is not necessary in the case of a ferrite magnetostrictive material.
The magnetostriction amount of the magnetic material does not change even when the direction of magnetization is reversed, and there is no function of dynamic magnetostriction conversion for small amplitude vibration in the demagnetized state. Therefore, when a magnetostrictive vibrator is used, it is necessary to give an eccentric magnetization. In the case of a ferrite vibrator, a magnet eccentric magnetization method is often used. This is one in which a magnet is inserted into a magnetic circuit to give an eccentric magnetization, and a DC eccentric current is unnecessary. Since it is necessary to pass an alternating magnetic flux through the magnet, a metal magnet cannot be used, and barium ferrite is used.
Magnetostrictive vibrators are most often used at mechanical resonance frequencies. Therefore, the dimension of the vibrator in the vibration direction is substantially determined by the frequency of the ultrasonic wave. Due to such dimensional restrictions, the frequency at which the magnetostrictive vibrator is used is mainly in the range of several kHz to 100 kHz.
Although the expansion / contraction rate of the vibrator due to magnetostriction is very small, a much larger vibration occurs in the state of mechanical resonance. When the zero advance is strengthened, the stress due to the vibration exceeds the fracture limit, and the vibrator may be broken.
[0062]
When a magnetic field is applied to a ferromagnetic material, distortion occurs in the magnetic field direction along with the magnetization. There are several types of magnetostriction, but what is important for measuring viscosity is a phenomenon called the magnetostrictive effect or Joule effect that changes the length in the longitudinal direction, and the permeability increases when stress is applied to a magnetized material. There are a phenomenon in which the length in the longitudinal direction changes called a changing billiary effect or inverse magnetostriction effect, and a phenomenon called a billiary effect or inverse magnetostriction effect in which the permeability changes when stress is applied to a magnetized material.
When the vibrator is oscillated in the liquid, a force that hinders expansion and contraction of the element is generated. By receiving this force, the magnetic permeability of the element changes. When the magnetic permeability changes, the inductance of the coil changes and the impedance changes. This change is measured. This vibrating piece viscometer is characterized in that the viscosity coefficient of the fluid flowing in the pipe can be continuously measured regardless of whether it is transparent or opaque.
[0063]
Table 6 shows the main characteristics of the magnetostrictive vibrator ferrite for ultrasonic oscillation, manufactured by TDK Corporation. Further, FIG. 31 shows the outer shape of the ultrasonic transducer, and Table 7 shows the dimensions of each part.
[Table 6]

[Table 7]

[0064]
Using the experimental apparatus shown in FIG. 32, the sample solution (glycerin 100%, glycerin 50% aqueous solution and water) was placed in a beaker, and the voltage waveform and voltage value when the vibrator was placed in the liquid were measured.
33 to 35 show voltage waveforms measured using an oscilloscope for each sample solution. FIG. 33 shows the voltage waveform when glycerin is 100%, FIG. 34 shows the voltage waveform when glycerin is 50% aqueous solution, and FIG. 35 shows the voltage waveform when water is water.
From the experimental results shown in FIGS. 33 to 35, changes in the voltage waveform due to the difference in the concentration of glycerin were shown.
[0065]
In Examples 1 and 2, the miniaturization of a PCPS system using a two-point support centrifugal pump drive device (C1E3; manufactured by Kyocera) was examined.
A motor characteristic experiment was conducted, and a flow pressure estimation formula was calculated in consideration of the viscosity coefficient. As the estimation expression evaluation experiment, an evaluation experiment was actually performed using bovine blood. The estimation accuracy of the flow rate in blood was as high as 2.0%, and a stable flow rate estimation result was obtained at each flow rate. In addition, it is reported that an error of an ultrasonic flowmeter used in a commercially available centrifugal pump system is about ± 10% and an error of an electromagnetic flowmeter is about ± 5%. The high accuracy was confirmed.
The pressure estimation accuracy was 5.8%, and a highly accurate estimation result was obtained for the pressure.
The fluid viscosity coefficient is also an important estimation factor in the flow pressure estimation, and it has been shown that the flow pressure estimation accuracy can be greatly improved by newly substituting the viscosity coefficient into the estimation equation. In addition, it was suggested that a small centrifugal pump circulation system that does not require a flow meter and pressure gauge could be realized by controlling the motor drive unit using a microcomputer and displaying the estimated flow value and estimated pressure value.
[0066]
【The invention's effect】
ADVANTAGE OF THE INVENTION According to this invention, the centrifugal pump drive device provided with the flow volume measurement means cheaper than the conventional ultrasonic flowmeter and electromagnetic flowmeter can be provided.
In addition, it is possible to provide a centrifugal pump driving device with higher safety that can suggest a replacement time due to deterioration of the centrifugal pump over time.
Furthermore, it is possible to provide a centrifugal pump driving device provided with a flow meter that can measure the flow rates of both the clear prime solution and blood by a switching operation.
In addition, it is possible to provide a centrifugal pump driving device including a flow meter that can accurately measure the flow rate according to various centrifugal pump driving devices having different loss currents by switching operation.
[Brief description of the drawings]
FIG. 1 is a longitudinal sectional view illustrating a centrifugal pump.
FIG. 2 shows a centrifugal pump drive device according to the present invention.Reference exampleFIG.
FIG. 3 shows a centrifugal pump drive device according to the present invention.First embodimentFIG.
FIG. 4 shows a centrifugal pump driving device according to the present invention.Second embodimentFIG.
FIG. 5 shows a centrifugal pump drive device according to the present invention.Third embodimentFIG.
FIG. 6 is a graph showing the results of the examples according to the present invention and showing the flow rate and pressure characteristics of the used centrifugal pump.
FIG. 7 is a graph showing the relationship between each hematocrit value, viscosity coefficient, and temperature.
FIG. 8 is a graph showing the relationship between the concentration of a glycerin aqueous solution and the viscosity coefficient.
FIG. 9 is a graph showing the relationship between motor loss current and angular velocity.
FIG. 10 is a configuration diagram of a circulation circuit used in the experiment.
FIG. 11 is a flow rate-current characteristic diagram at a viscosity coefficient of 2.0 [cP].
FIG. 12 is a flow rate-current characteristic diagram at a viscosity coefficient of 3.0 [cP].
FIG. 13 is a flow rate-current characteristic diagram at a viscosity coefficient of 4.0 [cP].
FIG. 14 is an angular velocity-loss current characteristic diagram at a flow rate of 0 [L / min] with a viscosity coefficient of 3.0 [cP].
FIG. 15 is a flow rate generalized current characteristic diagram.
FIG. 16 is an angular velocity-tilt characteristic diagram.
FIG. 17 is a generalized flow rate-generalized current characteristic diagram.
FIG. 18 is a generalized flow rate-generalized current characteristic diagram of each viscosity coefficient.
FIG. 19 is an intercept-viscosity coefficient characteristic diagram.
FIG. 20 is a flow rate-pressure difference characteristic diagram at a viscosity coefficient of 2.0 [cP].
FIG. 21 is a flow rate-pressure difference characteristic diagram at a viscosity coefficient of 3.0 [cP].
FIG. 22 is a flow rate-pressure difference characteristic diagram at a viscosity coefficient of 4.0 [cP].
FIG. 23 is an angular velocity-pressure characteristic diagram.
FIG. 24 is a flow rate-generalized pressure characteristic diagram.
FIG. 25 is a generalized flow rate-generalized pressure difference characteristic diagram.
FIG. 26 is a generalized flow rate-generalized pressure characteristic diagram of each viscosity coefficient.
FIG. 27 is an accuracy diagram of a flow rate estimation formula in consideration of viscosity.
FIG. 28 is an accuracy diagram of a conventional flow rate estimation formula.
FIG. 29 is an accuracy diagram of a pressure estimation formula in consideration of viscosity.
FIG. 30 is an estimation system block diagram of a small control system.
FIG. 31 is a perspective view showing the outer shape of an ultrasonic transducer used in the experiment.
FIG. 32 is a configuration diagram of an experimental apparatus for voltage waveform measurement.
FIG. 33 shows voltage waveforms measured with glycerin 100%.
FIG. 34 is a diagram showing voltage waveforms measured with a 50% aqueous solution of glycerin.
FIG. 35 is a diagram showing voltage waveforms measured with water.
[Explanation of symbols]
  1 Centrifugal pump
  2 Blood inlet
  3 Blood outlet
  4 Magnet
  5 Impeller
  10, 20, 30, 40 Centrifugal pump drive
  11 Drive unit
  12 Ammeter
  13, 24, 32, 42 Flow meter
  21, 22 Pressure gauge
  31 First correction means
  41 Second correction means

Claims (3)

A magnet enclosed in a space having a liquid inlet and a liquid outlet, and rotated at the same rotational speed as that of a magnet disposed adjacent to the outside to send blood from the liquid inlet to the liquid outlet; In a centrifugal pump drive device having a centrifugal pump including an impeller coupled to a magnetic body or a magnet or an impeller embedded with a magnetic body, and a drive unit having a magnet and a motor for rotationally driving the impeller of the centrifugal pump,
As well as have a tachometer which measures the speed and current meter for measuring the current value of the motor of the driving unit, it has a pressure gauge for measuring the inlet pressure and the outlet pressure of the centrifugal pump,
Calculate the calculated differential pressure value from the flow rate and the inlet pressure and outlet pressure of the centrifugal pump from the current value and the rotational speed,
Measure the actual differential pressure value from the measured values of the inlet pressure and outlet pressure of the centrifugal pump,
A centrifugal pump driving device configured to be able to calculate a difference between the measured pressure difference and the calculated pressure difference by comparing the measured differential pressure value with the calculated differential pressure value. The centrifugal pump drive device according to claim 1, further comprising first correction means for correcting the flow rate calculation value by inputting or switching the viscosity coefficient information of the liquid flowing in the centrifugal pump. 3. The centrifugal pump drive device according to claim 1, further comprising: a second correction unit that corrects the flow rate calculation value by inputting or switching a loss current value of the centrifugal pump.

Images