JP4196518B2 - Valve positioner - Google Patents

Description

【００５３】
▲２▼上記式１におけるΨ（Ｐo／Ｐs）が０．５２８≦Ｐo／Ｐs≦０．９の時の
Ψ（Ｐo／Ｐs）は次の式３で表すことができる。
【００５４】
Ψ（Ｐｏ／Ｐｓ）＝｛２ｇＫ／（ＲＴ（Ｋ−１）｝^1/2
・｛（Ｐｏ／Ｐｓ）^2/K−（Ｐｏ／Ｐｓ）^(K+1)/K｝^1/2・・・・・式３
【００５５】
▲３▼上記式１におけるΨ（Ｐo／Ｐs）が０．９＜Ｐo＜Ｐsの時のΨ（Ｐo／Ｐs）は次の式４で表すことができる。
【００５６】
Ψ（Ｐｏ／Ｐｓ）＝｛２ｇ／ＲＴ｝^1/2
・｛（Ｐｏ／Ｐｓ）−（Ｐｏ／Ｐｓ）²｝^1/2・・・・・式４
【００５７】
但し、ｇ＝９８０ｃｍ／ｓ２、ｋ＝比熱比（空気＝１．４）、Ｒ＝気体定数２９２７ｃｍ／°ｋ、Ｔ＝絶対温度°ｋ、Ｃｒ＝流量係数、Ａ＝面積（ｃｍ²）、Ｐ１、Ｐ２＝絶対圧力ｋｇｆ／ｃｍ²である。
【００５８】
従って、Ψ（Ｐo／Ｐs）がＰo／Ｐs／０．５２８の状態で測定すれば、流量Ｑは一定となるが、Ψ（Ｐo／Ｐs）がＰo／Ｐs≦０．５２８の状態で測定した場合、出力圧Ｐoの変化により、流量Ｑが変化してしまう。しかし、これも、供給圧Ｐｓに対し、出力圧Ｐｏが小さく、圧力変化の少ない区間で測定することにより、誤差を小さくすることができる。
【００５９】
又、この方法の場合、電空変換機構部１３の応答速度が考慮されていないが、電空変換機構部１３の応答速度は、圧力増幅器１４とバルブ部１５の空気アクチュエーター部１５の応答速度に対して十分早いので、考慮しなくとも測定誤差の中に埋もれてしまう。又、電空変換機構部１３の応答速度をある一定値で考慮することにより、制御対象部位１７の応答速度が求められる。
【００６０】
次に、具体的な応答速度を測定するための手順をフローチャートを参照して説明する。
【００６１】
第１の測定方法の手順について、図６に示すフローチャートを参照して説明する。
【００６２】
先ず、設定信号を受けて測定が開始すると、制御信号ＭＶ値の初期化が行われるステップＳＴ１０）。弁開度信号ＰＶ値が予め設定されている値Ｙ１より低くなることを確認し、低くなると、次に、制御信号ＭＶ＝初期値＋ΔＭＶとして、圧力増幅器１４の不感帯の不感帯に排気弁座３４を入れる（ステップＳＴ１１、ＳＴ１２）。この状態で、ポペット弁３５と給気弁座３２と排気弁座３４は閉まった状態となり、ブリード孔３９だけは、バルブ部１５の空気アクチュエーター部２０に空気の流量を与える。
【００６３】
次に、弁開度信号ＰＶ値が予め設定されている値Ｙ１を超えるのを確認する（ステップＳＴ１３）。Ｙ１を超えるのを確認した後に、タイマーをスタートさせ、弁開度信号ＰＶ値がＹ１の値よりも大きな予め設定されている値Ｙ２を超えるのを確認する。Ｙ２を超えるのを確認した後にタイマーをストップさせる（ステップＳＴ１４、ＳＴ１５、ＳＴ１６）。
【００６４】
そして、応答速度を移動したＹ２−Ｙ１の値とそれに要した時間で割ることにより求める。そして、チューニングパラメータであるループゲインを応答速度から求めて測定は終了する（ステップＳＴ１７、ＳＴ１８）。
【００６５】
次に、第２の測定方法の手順について、図７に示すフローチャートを参照して説明する。
【００６６】
先ず、設定信号を受け測定が開始されると、制御信号ＭＶ値を初期化する（ステップＳＴ２０）。そして、弁開度信号ＰＶ値が予め設定されている値Ｙ１より低くなることを確認する。Ｙ１より低くなることを確認した後に、制御信号ＭＶ＝初期値＋ΔＭＶとして、圧力増幅器１４の不感帯に排気弁座３４を入れる（ステップＳＴ２１、ＳＴ２２）。この状態でポペット弁３５と給気弁座３２と排気弁座３４は閉まった状態となり、ブリード孔３９だけで、バルブ部１５の空気アクチュエーター部２０に空気の流量を与える。そして、弁開度信号ＰＶ値がＹ１を超えるのを確認する（ステップＳＴ２３）。Ｙ１を超えたことを確認した後にタイマーをスタートさせる。そして、タイマーをスタートさせた後に単位時間待つ（ステップＳＴ２４、ＳＴ２５）。
【００６７】
単位時間経った時の弁開度信号ＰＶ値をＹ２とする（ステップＳＴ２６）。応答速度を移動したＹ２−Ｙ１の値とそれに要した単位時間で割ることにより求める。そして、チューニングパラメータであるループゲインを応答速度から求めて測定は終了する（ステップＳＴ２７）。
【００６８】
［２］制御対象部位１７のヒステリシス測定
制御対象部位１７のヒステリシスとは、制御演算装置１２が出力する制御信号ＭＶから、位置センサー１６が出力する弁開度信号ＰＶまでの入出力関係に存在するヒステリシスのことである。このヒステリシスは、制御対象部位１７の特性の非線形性の代表格であり、その特性が制御に与える影響は大きい。制御対象部位１７がヒステリシスを抜けるまで、弁開度信号ＰＶは変化しないので、制御信号ＭＶを与えた時から、弁開度信号ＰＶが変化するまで遅れ時間が生じる。この遅れ時間は無駄時間となり、大きな位相遅れ要因となる。従って、制御演算装置１２が例えばＰＩＤ制御アルゴリズムを持っているとすると、位相補償を行っている微分時間や、積分時間のチューニングパラメータに影響する。
【００６９】
このヒステリシスの測定手法は、図１に示す電空変換機構部１３、圧力増幅器１４のヒステリシスを予めわかっているとすると、バルブ部１５のヒステリシスを直接測定する手法である。
【００７０】
このバルブ部１５のヒステリシスの測定は、図８に示すように、制御信号ＭＶをバルブ部１５から得られる弁開度信号ＰＶが変化するまで増加させ、弁開度信号ＰＶが変化した後に、制御信号ＭＶをバルブ部１５から得られた弁開度信号ＰＶが逆方向に変化するまで減少させる。この一連の動作のバルブ部１５の駆動信号である圧力信号と弁開度信号ＰＶを記憶することにより、バルブ部１５のヒステリシスは計算できる。このようにして、一連の圧力信号と弁開度信号ＰＶのデータを測定することにより、ヒステリシスを測定できるが、データ量が多くなるため、メモリなどのハードウエアのリソースが大量に必要になる。従って、ステムの動きを監視し、図８の▲３▼と▲４▼の圧力のデータのみを記憶することにより、ヒステリシスを計算することもできる。
【００７１】
しかしながら、この方法では、バルブ部１５の駆動信号である圧力を測定するセンサーが必要になるので、コスト、消費電力の点から不利になる。又、この方法の場合、バルブ部１５のヒステリシスしか測定できないというデメリットもある。
【００７２】
バルブ部１５の駆動信号を検出するセンサーがなくとも、ヒステリシスを測定する方法もある。その方法とは、制御信号ＭＶと弁開度信号ＰＶの入出力関係を測定することによりヒステリシスを測定する方法である。
【００７３】
即ち、制御信号ＭＶをゆっくり変化させ、弁開度信号ＰＶの変化を検出し、弁開度信号ＰＶに変化が現れたときの制御信号ＭＶ値であるＭＶ１を記録し、制御信号ＭＶを今まで変化した方向と逆方向に変化させ、弁開度信号ＰＶ値が変化した時の制御信号ＭＶ値であるＭＶ２を記録し、ＭＶ１−ＭＶ２によりヒステリシスは求められる。
【００７４】
しかし、この方法は、バルブポジショナ１０とバルブ部１５をオープンループにして測定するため、制御信号ＭＶ値の変化に対して、電空変換機構部１３＋圧力増幅器１４のゲインが高く設定されている場合は測定が困難である。従って、本発明では、バルブポジショナ１０とバルブ部１５をクローズトループにして、測定する方法を提案するものである。
【００７５】
制御演算装置１２の制御アルゴリズムを、少なくとも比例制御を持つ制御アルゴリズムとする。例えば、比例制御、或いは比例、微分制御とする。この制御アルゴリズムに、ある入力値を与え、弁開度信号ＰＶの値を整定したことを確認して入力信号ＳＰをゆっくりと変化させていく。弁開度信号ＰＶの値が入力の変化に反応した時点の入力信号ＳＰのＳＰ１を記録する。次に、入力信号ＳＰをそれまで変化してきた方向と逆方向にゆっくりと変化させる。弁開度信号ＰＶの値が入力の変化に対応した時点の入力信号ＳＰ２を記録する。
【００７６】
このアルゴリズムは比例制御である場合、制御ループの外乱（この場合ヒステリシス）は、制御ループのループゲイン分の１となり、定常偏差が残る。
【００７７】
ここで、図９に示すように、ユニティフィードバックのシステムを考えた場合、下記の式５で表すことができる。
【００７８】
Ｙ（Ｓ）＝ＳＰ（Ｓ）
＊Ｇ（Ｓ）／（１＋Ｇ（Ｓ））＋Ｄ（Ｓ）／（１＋Ｇ（Ｓ））・・・・・式５
ここで、Ｇ（Ｓ）：システムの伝達関数、ＳＰ（Ｓ）：入力信号、Ｄ（Ｓ）：システム全体の流れ、Ｙ（Ｓ）：ステム変位である。
【００７９】
今、ＳＰ１（Ｓ）からＳＰ２（Ｓ）までＹ（Ｓ）に変化がなかったとすると、次の式６、式７を得ることができる。
【００８０】
Ｙ１（Ｓ）＝ＳＰ１（Ｓ）＊
Ｇ（Ｓ）／（１＋Ｇ（Ｓ））＋Ｄ１（Ｓ）／（１＋Ｇ（Ｓ））・・・・・式６
【００８１】
Ｙ２（Ｓ）＝ＳＰ２（Ｓ）＊
Ｇ（Ｓ）／（１＋Ｇ（Ｓ））＋Ｄ２（Ｓ）／（１＋Ｇ（Ｓ））・・・・・式７
【００８２】
Ｄ（Ｓ）の変化量がヒステリシスに当たる。ここで、Ｙ１（Ｓ）＝Ｙ２（Ｓ）なので上式をまとめ、Ｄ（Ｓ）の変化量の式に直すと次の式８になる。
【００８３】
Ｄ２（Ｓ）−Ｄ１（Ｓ）＝（ＳＰ１（Ｓ）−ＳＰ２（Ｓ））＊Ｇ（Ｓ）・・・式８
このように、ヒステリシスは入力の変化にループゲインを掛けた式になる。
【００８４】
ヒステリシスには符号がないので、
Ｈｙｓ＝｜ＳＰ１−ＳＰ２｜＊ループゲイン
で求めることができる。
【００８５】
具体的なヒステリシスの測定方法について、図１０に示すフローチャートを参照して説明する。
【００８６】
先ず、測定信号を受け測定が開始されると、制御演算装置１２の制御アルゴリズムを比較制御する（ステップＳＴ３０）。ループゲインをヒステリシス測定用設定パラメータに置き換えると共に、入力信号ＳＰを初期値に設定する（ステップＳＴ３１）。弁開度信号ＰＶの値が静止するのを待つ（ステップＳＴ３２）。弁開度信号ＰＶが静止している場合は、入力信号ＳＰ＝ＳＰ＋ΔＳＰとし、入力信号ＳＰを少しずつ増加させる方向に変化させる（ステップＳＴ３３、ＳＴ３４）。弁開度信号ＰＶが動き出したら、その時の入力信号ＳＰの値をＳＰ１として記憶する（ステップＳＴ３３、ＳＴ３５）。弁開度信号ＰＶが静止するのを待つ（ステップＳＴ３６）。弁開度信号ＰＶが静止している場合は、入力信号ＳＰ＝ＳＰ−ΔＳＰとし、入力信号ＳＰを少しずつ減少させる方向に変化させる（ステップＳＴ３７、ＳＴ３８）。弁開度信号ＰＶが動き出したら、その時の入力信号ＳＰの値をＳＰ２として記憶する（ステップＳＴ３９）。Ｈｙｓ＝｜ＳＰ１−ＳＰ２｜＊ループゲインの式でヒステリシスを求める（ステップＳＴ４０）。Ｈｙｓの値をもとに、制御パラメータの計算を行う。微分時間＝ｆ（Ｈｙｓ）、積分時間＝ｆ（ＨｙS）である。
【００８７】
［３］スリップ現象の測定
スリップ現象を起こした場合のステムの動きは、図１１に示すように、バルブの駆動信号である圧力信号を増加させていき、バルブ部１５がヒステリシスを抜けた瞬間、バルブの弁開度信号は素早く動き、その後、圧力の変化に応じた変化速度に落ち着く。このように、ヒステリシスを抜けた瞬間、弁開度信号が滑るような動きをする現象をスリップ現象と呼び、この現象が大きく制御性に影響する。
【００８８】
具体的には、スリップ現象を起こしている区間は、制御不能状態である。従って、例えば、０．１パーセントなどの微少な弁開度を制御しようとした場合、スリップ現象を起こすと、０．１パーセントの位置で弁開度が静止できないため、行き過ぎが生じてしまう。行き過ぎが生じた時、行き過ぎを戻そうとして、制御演算装置１２が逆側に制御する。この時、又、行き過ぎが生じ、この繰り返しでリミットサイクルが起こる。リミットサイクルを止めるためには、スリップ現象が起こる区間では、積極的に制御を行わないようにすればよい。従って、スリップ現象を起こす区間は制御アルゴリズムを変更するなどの処理が必要になってくる。スリップ現象を、バルブのヒステリシスで説明したが、この現象は、位置センサーとステムとのリンク機構のガタが原因で起こる。従って、制御対象全体の特性で測定する必要がある。
【００８９】
測定方法としては、制御演算装置１２が出力する制御信号ＭＶをゆっくり変化させ、弁開度信号ＰＶの変化のようすを測定する。現象としては、制御対象がヒステリシスや不感帯などを抜けた瞬間、弁開度信号がある幅だけ素早く動き、その後は正常な変化をするので、弁開度信号の変化速度を測定すればよい。弁開度信号ＰＶが静止している状態から、弁開度信号ＰＶが動き出す時の変化速度のようすを示したのが図１２である。図１２において、弁開度信号ＰＶの変化速度の変極点を検出すれば、スリップ現象の幅が測定できる。
【００９０】
つまり、図１１において、弁開度信号ＰＶが静止している時の弁開度信号ＰＶ信号の値であるＰＶ３を記憶し、弁開度信号ＰＶの変化速度の変極点の時の弁開度信号ＰＶをＰＶ４とすることにより、スリップ幅は、スリップ幅＝｜ＰＶ４−ＰＶ３｜で求める。
【００９１】
又、単に、弁開度信号ＰＶが静止している初期値をＰＶ３とし、制御信号ＭＶの値を変化させてやり、弁開度信号ＰＶが動き出した後、短いある一定時間後の弁開度信号ＰＶをＰＶ４とすることにより、スリップ現象のデータとしてもよい。但し、この場合、測定誤差が大きくなる可能性はある。
【００９２】
具体的なスリップ現象の測定方法について、図１３のフローチャートを参照して説明する。
【００９３】
先ず、測定信号を受け測定がスタートする。制御信号ＭＶ＝初期値に設定する（ステップＳＴ５０）。弁開度信号ＰＶが静止するまで待つ（ステップＳＴ５１）。制御信号ＭＶ＝ＭＶ＋ΔＭＶとし、ゆっくりと制御信号を、弁開度信号ＰＶ値が動くまで変化させる（ステップＳＴ５２、ＳＴ５３）。この作業により、いったん制御対象のヒステリシスをリセットさせる。次に、弁開度信号ＰＶが静止するまで待つ。弁開度信号ＰＶが静止した時のＰＶ値をＰＶ３として記憶する。即ち、ＰＶ３＝ＰＶとする（ステップＳＴ５４、ＳＴ５５）。
【００９４】
次に、制御信号ＭＶ＝ＭＶ−ΔＭＶとし、ゆっくりと制御信号を今までとは逆方向に変化させる。今回の弁開度信号ＰＶの変化速度を測定する。弁開度信号ＰＶの変化速度の変極点を検出するまで繰り返す（ステップＳＴ５６、ＳＴ５７、ＳＴ５８、ＳＴ５９）。
【００９５】
そして、弁開度信号ＰＶの変化速度の変極点を検出した時の弁開度信号をＰＶ４として記憶する。即ち、ＰＶ４＝ＰＶ、スリップ幅＝｜ＰＶ４−ＰＶ３｜、制御アルゴリズムの切り替え条件＝ｆ（スリップ幅）で決定する（ステップＳＴ６０、ＳＴ６１）。
【００９６】
［４］電空変換機構部１３の動作点測定
電空変換機構部１３は、ノズルフラッパー機構を電磁アクチュエータで駆動し、ノズル背圧を変化させることにより、電空変換を行っている。この電空変換機構部１３はバルブポジショナ１０の場合、外部から供給される電気的エネルギーが制限されているため、ローパワーで働くことが求められ、電磁アクチュエータに供給できるエネルギーが限られている。従って、電空変換機構部１３に求められる変換ゲインを稼ぐために、ノズルフラッパ機構の前段の空気流量を絞り、ノズルフラッパー機構のゲインを上げている。その結果、ノズルとフラッパーの僅かな間隙で、電空変換を行うことになり、外乱に対して敏感になっていることと、バルブポジショナ１０に供給される供給圧の範囲は広いため、実際の動作点に対し、電空変換機構部１３の駆動信号は大幅に広く設計し、外乱や供給圧変動による動作点ズレを吸収できるような設計になっている。
【００９７】
制御演算装置１２は、内部に積分器を積んでおり、外乱に応じて制御信号ＭＶを変化させ、外乱を吸収している。従って、電源立ち上げ時などのリセット後、積分器の値がリセットされた場合、電空変換機構部１３の動作点がずれてしまい、積分器がワインドアップした状態のように、バルブ部１５からの弁開度信号ＰＶが定常値に戻るまでには長い時間を要し、素早く立ち上がれない。
【００９８】
この問題を解決するために、積分器の出力を不揮発性メモリ等に記憶させておけば、電源立ち上げ時などのリセット時でも電空変換機構部１３の動作点ズレは少なくてすむが、積分器の出力を不揮発性メモリなどに記憶することは実質的にできない。何故ならば、不揮発性メモリには更新回数の限界値があるため、データを定期更新した場合、いずれ劣化して故障してしまうためである。
【００９９】
この問題の解決方法としては、電空変換機構部１３の動作点を測定し、動作点をオフセット（Ｏｆｆｓｅｔ）として、不揮発性メモリに記憶しておくことにより、電源立ち上げ時等のリセット時にデータをロードすることにより、動作点が補正できる。
【０１００】
ＰＩＤ制御アルゴリズムを例にすると、例えば、電空変換機構部１３の動作点が制御信号ＭＶ＝５０である場合、積分器のリセット時の初期値がゼロである場合、入力信号ＳＰと弁開度信号ＰＶの偏差をＥとすると、積分器出力＝１／Ｔｉ∫Ｅｄｔ＝５０になるまで、定常偏差が残る（Ｔｉ；積分時間）。特に、積分時間が長い場合は、定常偏差がなくなるまで時間がかかる。従って、予め電空変換機構部１３の動作点を測定してＯｆｆｓｅｔ値として記憶しておき、ＰＩＤ演算部にたし込んでやれば、動作点補正ができ、たとえ積分器がリセットされても、立ち上がり時間は大幅に改善できる。即ち、制御信号ＭＶ＝Ｋｐ＊（Ｐ＋Ｉ＋Ｄ）＋Ｏｆｆｓｅｔ とすればよい。
【０１０１】
測定方法は、制御演算装置１２の制御アルゴリズムを少なくとも積分器がある制御アルゴリズムに設定し、入力信号ＳＰを５０パーセントに設定し、偏差が設定値、例えば±１パーセントに入るまで待つ。この時の制御信号ＭＶの値をＯｆｆｓｅｔ値として記憶する。このようにすることによって、次からのリセット時からの立ち上がり時間を短縮できる。
【０１０２】
具体的な電空変換機構部１３の動作点の測定方法について、図１４に示すフローチャートを参照して説明する。
【０１０３】
先ず、設定信号を受け測定が開始される。制御演算装置の制御アルゴリズムを比例積分（ＰＩ）制御にする（ステップＳＴ７０）。入力信号ＳＰに初期値を与える制御が始まり、弁開度信号ＰＶが入力信号ＳＰに近づく（ステップＳＴ７１）。入力信号ＳＰと弁開度信号ＰＶの偏差がある設定値以下になるまで制御を続ける。偏差がある設定値以下になった時点での制御信号ＭＶ値をＯｆｆｓｅｔとして記憶する。そして測定は終了する（ステップＳＴ７２、ＳＴ７３）。
【０１０４】
［５］診断機能について
制御対象部位１７の特性を、測定し、自動チューニングをする際、測定した制御対象部位１７の特性が一般的な制御対象部位と比べて、大きく逸脱した特性が得られた場合、それは、バルブポジショナ１０のインスタレーションエラーとも考えられるので、エラーメッセージ又はワーニングメッセージを出すことができるようにする。例えば、制御対象部位１７の応答速度測定結果が通常の値より、桁違いに遅い結果が出た場合は、バルブ部１５の空気アクチュエーター部２０（図２参照）の漏れや、供給空気圧の設定ミスなどが考えられる。又、制御対象部位１７のヒステリシスの測定結果が、通常の値より大きい場合は、バルブ部１５のカジリや、変位センサーのステムへのリンクの異常等が考えられる。また、制御対象部位１７のスリップ幅の測定結果が、通常の値より大きい場合は、位置センサー１６の故障やバルブ部１５の故障が考えられる。更に、電空変換機構部１３の動作点の測定結果が、通常の値より大きくずれている場合は、供給空気圧の設定ミスや、電空変換機構部１３の故障などが考えられる。このような場合、自動チューニングが終わった後、作業者に知らせるようなメッセージを出力するようにする。
【０１０５】
【発明の効果】
上記説明したように、本発明に係るバルブポジショナは次に示すような効果を有する。
【０１０６】
（１）バルブポジショナに自動チューニング機能を備えたことにより、作業者を選ばず、誰でも容易にバルブポジショナのチューニングができるようになるという効果がある。
【０１０７】
（２）又、バルブポジショナに自動チューニング機能を備えることにより、バルブポジショナの立ち上がり工数を減らすことができるという効果がある。
【０１０８】
（３）更に、制御対象部位の特性から演算によりチューニングパラメータを求めることにより、正確なチューニングができるという効果がある。
【０１０９】
（４）制御対象部位の応答速度を測定することにより、バルブの特性のみならず、より正確なチューニング情報が得られるという効果がある。
【０１１０】
（５）制御対象部位のヒステリシスを測定することにより、より正確なチューニング情報が得られるという効果がある。
【０１１１】
（６）制御対象部位のスリップ幅を測定することにより、より正確なチューニング情報が得られるという効果がある。
【０１１２】
（７）電空変換機構部の動作点を測定することにより、バルブポジショナのリセット時からの立ち上がり時間を短くできるという効果がある。
【０１１３】
（８）制御対象部位を測定し、測定結果によってはエラーメッセージ又はワーニングメッセージを出すことにより、ポジショナのインスタレーションエラーを未然に妨げることができるという効果がある。
【図面の簡単な説明】
【図１】本発明に係るバルブポジショナの全体構成を略示的に示したブロック図である。
【図２】同図１におけるバルブ部を略示的に示した説明図である。
【図３】同図１における圧力増幅器の構造を示した平面図である。
【図４】同図３における圧力増幅器におけるノズル背圧と空気流量特性による不感帯を示したグラフである。
【図５】同図３における圧力増幅器のブリード孔の構成を示した説明図である。
【図６】同制御対象部位における応答速度を測定する手法を示したフローチャートである。
【図７】同制御対象部位における応答速度を測定する手法を示したフローチャートである。
【図８】同バルブ部のヒステリシスを計算するための手法を示した説明図である。
【図９】同ユニティフィードバックのシステムを示した概念図である。
【図１０】同ヒステリシスを求めるためのフローチャートである。
【図１１】同スリップ現象を起こした場合のステムの動きを示した説明図である。
【図１２】同ステム速度の変極点を示したグラフである。
【図１３】同スリップ幅を測定するための具体的な測定方法を示したフローチャートである。
【図１４】同積分器を搭載した制御演算装置における制御信号を測定する手法を示したフローチャートである。
【符号の説明】
１０；バルブポジショナ、１１；信号受信装置、１２；制御演算装置、１３；電空変換機構部、１４；圧力増幅器、１５；バルブ部、１６；位置センサー、１７；制御対象部位、２０；空気アクチュエーター部、２１；ダイヤフラム、２２；スプリング、２３；ステム、２４；弁体、３０；供給圧室、３１；出力圧室、３２；給気弁座、３３；大気圧室、３４；排気弁座、３５；ポペット弁、３６；出力圧ダイアフラム、３７；入力圧ダイアフラム、３８；入力圧室、３９；ブリード孔[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a valve positioner. More specifically, the present invention relates to a valve positioner that receives an input signal and controls the valve opening degree of the valve to match the received input signal. A position sensor that converts the signal into a signal, a control calculation unit that performs a control calculation so that the signal from the position sensor matches the input signal, and an electropneumatic device that converts the control signal calculated by the control calculation unit into a valve drive signal It has a conversion mechanism part, and automatically measures the characteristics of the control target part composed of these, and has a function for automatically tuning by obtaining the tuning parameter of the control calculation part from the characteristic of the control target part. Relates to the valve positioner.
[0002]
[Prior art]
Since the valve positioner in the prior art is equipped with an arithmetic function (CPU or the like), a control algorithm for controlling the valve opening degree of the valve can be realized by software. With this technology, complicated control calculations can be performed, and the controllability of the valve is significantly improved compared to the mechanical valve positioner of the previous generation. On the other hand, with regard to the control algorithm, as the number of control parameters used increases, the tuning becomes complicated and the installation takes time. To solve this problem, some models are equipped with an automatic tuning function. In addition to the control calculation, the use of the calculation function has extended to self-diagnosis, control valve diagnosis, and the like.
[0003]
However, one of the reasons why tuning of the valve positioner is difficult is that the types of valves that the valve positioner must control cannot be specified because there are many types of valves. That is, since it is impossible to specify what kind of characteristic valve is combined with the valve positioner, tuning has to be repeated by cutting and trying. Accordingly, a valve positioner having a function for assisting tuning work or a valve positioner having an automatic tuning function has been developed.
[0004]
As an automatic tuning method for realizing this automatic tuning function, there are (1) a tuning assist type valve positioner and (2) an automatic tuning type valve positioner.
[0005]
(1) The tuning assist type valve positioner is suitable for the valve positioner based on the data when the operator inputs the valve model name and its type and characteristics to be combined with the valve positioner to the valve positioner. This is a so-called semi-auto tuning method that selects a suitable tuning parameter.
[0006]
(2) The automatic tuning type valve positioner is a method that measures characteristics such as valve size and hysteresis and selects a tuning set of control parameters from a predetermined parameter table when an automatic setting signal is given to the valve positioner. .
[0007]
[Problems to be solved by the invention]
However, as explained in the above prior art, since tuning for controlling the valve is difficult for the valve positioner, (1) a tuning assist type valve positioner, which is a function for supporting control tuning of the valve described above, (2) There are automatic tuning type valve positioners, but each has the following problems.
[0008]
The problem with the automatic tuning method (1) is that the operator needs a certain level of expertise because he needs information and knowledge of the valve. Therefore, since it is necessary to input valve information and characteristics to the valve positioner, it takes time for input, and there is a risk of causing human errors such as input errors.
[0009]
The problem with the automatic tuning method (2) is that the valve table positioner must have an enormous amount of parameter table data when trying to tune the valve characteristics that can be used in countless combinations. This requires hardware and costs. In order to solve this problem, if an attempt is made to use a small number of parameter tables, there is a problem that the tuning becomes coarse and the tuning lacks accuracy.
[0010]
Thus, some of the elements of the valve positioner itself have non-linear characteristics consisting of unique characteristics, and there is a problem that practical tuning reflecting these effects cannot be performed. This is because, depending on the control device of the valve positioner, the controlled object is not only the valve. Specifically, depending on the method of measuring the characteristics of the valve, accurate measurement may not be possible, or measurement may take time. There is a problem that the time required for tuning becomes long.
[0011]
Therefore, there is a problem that has to be solved by solving such a problem and performing automatic tuning of the valve more easily, more accurately, and more quickly.
[0012]
[Means for Solving the Problems]
In order to solve the above problems, the valve positioner according to the present invention is configured as follows.
[0013]
(1) An input signal for setting the valve opening of the valve, a position sensor for detecting the valve opening of the valve, and a deviation between the valve opening signal obtained by the position sensor and the input signal. A control calculation unit that generates a control signal by performing control calculation so that the valve opening coincides with the input signal, an electropneumatic conversion mechanism unit that generates an air flow rate based on the control signal, and the electropneumatic conversion mechanism unit A valve positioner comprising a pressure amplifier for supplying to the valve air pressure based on the air flow rate generated in
The control object part composed of the electropneumatic conversion mechanism part, the pressure amplifier, the position sensor, and the valve receives an automatic setting signal,
Measure a slip width of a slip phenomenon that occurs when the valve opening degree starts moving from a stationary state, and calculate a tuning parameter of a control signal generated by the control calculation unit according to a parameter composed of the measured slip width
A valve positioner characterized by
[0016]
(2) When measuring the slip width of the slip phenomenon that occurs when the valve opening degree of the valve starts moving from a stationary state, it is confirmed that the valve opening signal of the valve is stationary, and the valve opening degree at that time PV1 is stored, and when the control signal is changed, the change rate of the valve opening signal when the valve opening signal reacts is measured, and the valve opening at the inflection point of the change rate of the valve opening signal The valve positioner according to (1), wherein a signal PV2 is stored, and a difference between the valve opening signals PV1 and PV2 is used as a slip width of the valve and a value of the slip width is stored.
[0017]
(3) When measuring the slip width of the slip phenomenon that occurs when the valve opening degree of the valve starts moving from a stationary state, it is confirmed that the valve opening signal is stationary, and the valve opening PV1 at that time is determined. Memorize | stored and memorize | stores the valve opening signal PV2 after a preset short time after the valve opening signal of the valve reacts next when the control signal is changed, and stores the valve opening signals PV1 and PV2 The valve positioner according to (1), wherein the difference is set as a slip width of the valve and a value of the slip width is stored.
[0018]
(4) The control target part composed of the electropneumatic conversion mechanism, the pressure amplifier, the position sensor, and the valve measures the characteristics of the control target part when receiving the automatic setting signal, and the measured control (2) to (3), wherein when the value of the characteristic of the target part deviates from the value of the allowable range information including the characteristic of the control target part, an abnormal signal is output to the outside through the communication means. Valve positioner.
[0019]
In this way, tuning parameters can be automatically adjusted and parameters that measure the characteristics of control valves, etc., can be used as tuning parameters. Will be able to.
[0020]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of a valve positioner according to the present invention will be described with reference to the drawings.
[0021]
First, the valve positioner is fate that it is not possible to select a valve to be controlled, and it is impossible to specify what kind of characteristic valve is attached. That is, there are an infinite number of combinations such as the capacity of the air actuator of the valve and the spring range that determines the input / output relationship of the air actuator, and the response characteristics differ depending on the combination.
[0022]
Further, the mechanical frictional force of the valve promotes the dead time of the response characteristic and has an action of making the response characteristic oscillating. Further, when the valve starts to move out of the mechanical frictional force or play of the controlled object, a so-called slip phenomenon occurs in which the valve opens quickly and moves quickly and slides. This slip phenomenon causes a vibration phenomenon such as a limit cycle. Accordingly, the valve positioner is required to absorb the non-linearity of such a valve and to make the characteristic linear and stable with respect to the input signal and easy to control.
[0023]
Thus, there is a need for a valve positioner that absorbs the non-linearity of the valve and is linear and stable with respect to the input signal, making it easy to control, but in developing a valve positioner that requires this function, There are two problems.
[0024]
The first problem is the design of a control algorithm to absorb valve nonlinearities. The control algorithm design is considered to determine the controllability of the valve positioner.Therefore, careful attention is paid to the design, but the control algorithm must be complicated to absorb the nonlinearity of the controlled object. I don't get it. As a result, the tuning parameters of the control algorithm are also increasing.
[0025]
The second problem is a tuning method for applying the designed control algorithm to the controlled object. For example, taking the PID control algorithm as an example, there are three types of tuning parameters, proportional gain, integration time, and derivative time, and the control parameters increase as the control algorithm becomes more complex, making tuning more complicated Tend to. When tuning becomes complicated, a great amount of man-hours are required for tuning, and usability deteriorates.
[0026]
Therefore, as a function of the valve positioner, a function of automatically determining a tuning parameter has been required. In order to automatically determine the tuning parameter, it is necessary to grasp typical characteristics of the controlled object. Even if the characteristics of the controlled object are grasped, there are parameters that have a large influence on the control, including (1) response speed of the controlled object, (2) hysteresis of the controlled object, and (3) controlled object. This is a slip phenomenon resulting from hysteresis.
[0027]
Based on such a premise, the valve positioner according to the present invention is a pneumatic valve positioner 10 having a valve driven by air pressure, which is the most general one, as shown in FIG. The configuration of the pneumatic valve positioner includes an input signal based on a signal receiver 11 that receives an input signal, and a deviation between the valve opening signal PV obtained by converting the valve opening of the valve into an electrical signal and the input signal SP. A control operation unit 12 that is a control operation unit that performs control operation so as to match SP and outputs a control signal MV, an electropneumatic conversion mechanism unit 13 that generates an air flow rate based on the control signal MV, and the air flow rate The pressure amplifier 14 supplies the air pressure based on the pressure to the valve portion 15, and a valve that adjusts the valve body by changing the stem based on the air pressure, that is, the valve portion 15, and detects the stem displacement of the valve portion 15 to detect the valve And a position sensor 16 that generates an opening signal PV. Among them, the electropneumatic conversion mechanism part 13, the pressure amplifier 14, the position sensor 16, and the valve part 15 constitute a control target part 17.
[0028]
As shown in FIG. 2, the valve unit 15 includes an air actuator unit 20 that receives the air pressure signal Po from the pressure amplifier 14, a stem 23 that is connected to the air actuator unit 20, and a movement of the stem 23. And a valve body 24 for opening and closing the valve. The actuator unit 20 includes an air chamber therein, and a diaphragm 21 disposed so as to divide the air chamber into two parts, and a spring 22 that elastically maintains the diaphragm 21 at a predetermined position. The movement of the diaphragm 21 is a stem. 23 is linked to the structure.
[0029]
In the valve positioner 10 having such a configuration, first, when the signal receiving device 11 receives an input signal SP for setting the valve opening degree of the valve portion 15, the control arithmetic device 12 receives the received input signal SP and the valve portion. The control signal MV is generated by performing control calculation so as to coincide with the actual valve opening of the valve unit 15 to be opened by the input signal SP based on the deviation from the valve opening signal PV fed back from 15. This control signal MV is supplied to the electropneumatic conversion mechanism unit 13, outputs an air flow rate corresponding to the control signal MV, and gives a signal to the pressure amplifier 14. The pressure amplifier 14 outputs an air flow rate corresponding to the signal obtained by the electropneumatic conversion mechanism unit 14 and an air pressure signal Po, and supplies them to the valve unit 15. The valve unit 15 receives the air pressure signal Po output from the pressure amplifier 14, receives pressure from the diaphragm 21 in the air actuator unit 20, converts the pressure into a physical force, and transmits the converted force to the stem 23. Move up and down. The stem 23 is connected to a valve body 24 in a pipe through which the process fluid flows, and adjusts the flow rate of the process fluid by opening and closing the valve according to the movement of the stem 23. Here, the position sensor 16 built in the valve positioner 10 is connected to the stem 23 through a link mechanism, and a control opening / closing signal PV comprising an electric signal corresponding to the position where the stem 23 moves up and down is obtained. Give feedback to. Thus, the valve positioner 10 and the valve unit 15 form a closed loop, and the valve positioner 10 is configured to control the valve opening.
[0030]
In such a system of the valve positioner 10 and the valve unit 15, the control target portion 17 to be controlled by the control arithmetic device 12 is the valve opening sensed by the position sensor 16 from the control signal MV value output from the control arithmetic device. It refers to all of the signal conversion elements up to the degree signal PV value. That is, the control target part 17 is configured by the electropneumatic conversion mechanism part 13, the pressure amplifier 14, the valve part 15, and the position sensor 16.
[0031]
In the control arithmetic device 12, in order to efficiently absorb the saturation and non-linear characteristics of the control target part 17, the tuning can be performed by measuring the characteristic of the control target part 17 when tuning is performed. In the present invention, such characteristics of the control target part 17 are automatically measured and the tuning parameters of the control arithmetic unit 12 are automatically tuned.
[0032]
Hereinafter, [1] response speed of the control target part 17 necessary for automatic tuning, [2] hysteresis measurement of the control target part 17, [3] measurement of slip phenomenon, [4] operation of the electropneumatic conversion mechanism unit 13 The point measurement and [5] diagnostic function will be described in this order.
[0033]
[1] Response speed of the control target part 17
The response speed of the control target portion 17 indicates the speed of the control signal MV generated by the control calculation device 12 performing control calculation. By knowing the response speed of the control target region 17, it is possible to calculate how much the deviation between the input signal SP and the valve opening signal PV should be amplified and output as the control signal MV. Therefore, the loop gain of the system of the valve positioner 10 and the valve unit 15 can be determined.
[0034]
In the case of the valve positioner 10 that is driven by the air actuator unit 20 to control the valve, most of the response speed depends on the air processing capability that the pressure amplifier 14 can process, the size of the air actuator unit 20 of the valve unit 15, and the range of the spring 22. It will be decided. The air actuator unit 20 includes a diaphragm type as shown in FIG. 2 and a type having a piston in a cylinder (not shown).
[0035]
The air pressure generated in the factory is usually 10kgf / cm from the viewpoint of safety and cost.²Less pressure. 10kgf / cm²Is not supplied to the valve positioner 10 as it is, but is 1.4 kgf / cm throttled by a pressure reducing valve.²To 4kgf / cm²Is supplied to the valve positioner 10. On the other hand, since the air actuator unit 20 of the valve unit 15 must generate a force that can overcome the fluid pressure flowing through the valve body 24, the area of the diaphragm 21 or the cross-sectional area of the cylinder is designed to be large. Therefore, in order to drive the air actuator unit 20, a large volume of air is required.
[0036]
Here, since the amount of air that can be processed by the pressure amplifier 14 of the valve positioner 10 is limited, the response speed depends on the flow rate that can be processed by the pressure amplifier 14. In other words, the response speed of the control object of the valve positioner 10 is mostly governed by the amount of air required by the air actuator unit 20 and the amount of air processed by the pressure amplifier 14. In this way, the response speed of the valve positioner 10 tends to be determined by the size of the air actuator unit 20 of the valve unit 15, but the air flow rate of the pressure amplifier 14 that drives the air actuator unit 20 by applying an air flow rate to the air actuator unit 20. It depends on. This is because, even in the valve unit 15 adopting the same size of the air actuator unit 20, the response speed is increased if the pressure processing capacity of the pressure amplifier 14 is large, but the response speed is decreased if the air processing capacity is small. Because. The air processing capacity of the pressure amplifier 14 varies depending on the supply pressure supplied to the valve positioner 10 and the diameter and length of the air piping to the valve unit 15, and the valve positioner 10 can be used under any condition. You can not choose whether to be attached to.
[0037]
In this way, when measuring the response speed of the control target portion 17, it is obtained by measuring the response speed of the valve valve opening with respect to the air processing amount processed by the pressure amplifier 14. Specifically, a constant air flow rate within the range of the air flow rate that can be processed by the pressure amplifier 14 is given to the air actuator unit 20 of the valve unit 15, and the change rate of the valve opening signal PV is measured. It can be measured.
[0038]
Next, a method for measuring the response speed of the control target portion 17 will be specifically described below.
[0039]
The first measurement method for measuring the response speed of the control target portion 17 is the simplest method, and maximizes the input of the pressure amplifier 14 to saturate the intake air processing amount of the pressure amplifier 14. Or, conversely, the flow rate of air supplied to the air actuator unit 20 of the valve unit 15 is determined by saturating the exhaust air processing amount of the pressure amplifier 14 while minimizing the input of the pressure amplifier 14. In this state, the change rate of the valve opening signal PV, which is the output of the position sensor 16, may be measured. However, this method has a large error in measurement values for the following reasons, and cannot be measured well.
[0040]
(1) Since the maximum processing flow rate of the pressure amplifier 14 is supplied to the valve unit 15, the influence of pressure loss when flowing through the air circuit inside the valve positioner 10 becomes large. Similarly, the pressure loss of the diameter of the air pipe connecting the valve positioner 10 and the valve portion 15 is increased. (2) In the case of the valve unit 15 having a relatively small air actuator unit 20, the change rate of the valve opening degree is too fast, and the measurement variation becomes large. (3) Since the change in pressure is large, the air flow rate cannot be kept constant, resulting in a large measurement error. (4) Since the amount of change in the air flow rate when the actual valve unit 15 is controlled is not the limit value of the processing flow rate of the pressure amplifier 14, there is a difference between the data measured at the limit value and the characteristics during actual control. Because.
[0041]
The measurement technique in the pressure amplifier 14 according to the present invention has solved such a problem. Hereinafter, a second measurement method for measuring the response speed of the control target portion 17 will be described.
[0042]
The second measurement method for measuring the response speed of the control target portion 17 relates to the measurement method in the pressure amplifier 14 shown in FIG. 3, and this pressure amplifier 14 supplies the supply pressure chamber 30 and the output pressure chamber 31. The air valve seat 32 communicates, the output pressure chamber 31 and the atmospheric pressure chamber 33 communicate with each other through the exhaust valve seat 34, and the opening area of the air supply valve seat 32 and the exhaust valve seat 34 are adjusted. The poppet valve 35 with which the valve body was united is equipped. An output pressure diaphragm 36 and an input pressure diaphragm 37 are attached to the exhaust valve seat 34, and an atmospheric pressure chamber 33 is provided between the output pressure diaphragm 36 and the input pressure diaphragm 37. The air flow in the atmospheric pressure chamber 33, the atmospheric pressure chamber 33, and the input pressure chamber 38 is blocked. The exhaust valve seat 34 has a structure that moves in the left-right direction with respect to FIG. 3 by receiving the difference between the force received by the input pressure diaphragm 37 by the input pressure and the force received by the output pressure diaphragm 36 by the output pressure. Yes. By moving the exhaust valve seat 34, the opening areas of the poppet valve 35 and the exhaust valve seat 34 can be changed, and the exhaust flow rate of the air amount in the output pressure chamber 31 can be adjusted. Further, when the exhaust valve seat 34 comes into contact with the poppet valve 35 and moves further to the left with respect to FIG. 3, the exhaust valve seat 34 pushes the poppet valve 35 to the left, so that the air supply valve seat 32 and the poppet valve 35 are moved. , And the supply of air flow rate from the supply pressure chamber 30 to the output pressure chamber 31 can be adjusted. In this way, air supply / exhaust adjustment of the air amount in the output pressure chamber 31 is performed, the exhaust valve seat 34 is balanced in a certain state, and the output pressure can be changed according to the input pressure. In the case of FIG. 3, a bleed hole 39 is provided between the supply pressure chamber 30 and the output pressure chamber 31, and the supply pressure chamber 30 and the output pressure chamber 31 communicate with each other with a certain area. A pressure amplifier having a structure in which the bleed hole 39 is provided in the output pressure chamber 31 and the atmospheric pressure chamber 33 also exists.
[0043]
Next, the operation of the pressure amplifier 14 having such a structure will be described.
[0044]
First, in the case of (1) steady state, when the input pressure is constant and the exhaust valve seat 34 is balanced, the poppet valve 35 receives a force from the supply pressure and is pressed against the air supply valve seat 32. . Accordingly, the air supply valve seat 32 is shut off, but the air in the supply pressure chamber 30 flows into the output pressure chamber 31 from the bleed hole 39. In a steady state, the relationship between the output pressure and the input pressure does not collapse, so the flow rate that flows into the output pressure chamber 31 through the bleed hole 39 is exhausted through a gap created by the exhaust valve seat 34 and the poppet valve 35. In other words, the position of the exhaust valve seat 34 is determined by the balance of force so as to maintain an opening area for exhausting the flow rate flowing from the bleed hole 39. Therefore, in a steady state, the exhaust valve seat 34 and the poppet valve 35 are not in contact.
[0045]
{Circle around (2)} When the exhaust operation is performed, if the input pressure decreases or the output pressure increases due to some disturbance, the exhaust valve seat 34 moves to the left with respect to FIG. 3, and the opening area with the poppet valve 35 decreases. At this time, the flow rate at which the air in the output pressure chamber 31 is exhausted to the atmospheric pressure with respect to the flow rate flowing into the output pressure chamber 31 from the bleed hole 39 is reduced, so that the pressure in the output pressure chamber 31 increases. Further, when the input pressure is increased, the exhaust valve seat 34 contacts the poppet valve 35 and blocks the flow path between the output pressure chamber 31 and the atmospheric pressure chamber 33. At this time, the air supply valve seat 32 and the poppet valve 35 are not immediately opened. This is because the poppet valve 35 receives a force corresponding to its area from the supply pressure, and the force is supported by the supply valve seat 32. Further, when the input pressure is increased or the output pressure is increased due to some disturbance, the exhaust valve seat 34 is only popped after all the force received by the poppet valve 35 from the supply pressure chamber 30 is transferred to the exhaust valve seat 34. The valve 35 is pushed leftward with respect to FIG. 3, the opening area can be adjusted with respect to the air supply valve seat 32, and the flow rate flowing from the supply pressure chamber 30 into the output pressure chamber 31 can be adjusted. . Therefore, the poppet valve 35 and the air supply valve seat 32 remain closed until the exhaust valve seat 34 comes into contact with the poppet valve 35 and is pushed open from the air supply valve seat 32, so that the supply pressure chamber 30 changes to the output pressure chamber 31. Since the flow rate of flow is determined by the area of the bleed hole 39, there is no change in the flow rate with respect to the change amount of the input pressure or the increase amount of the output pressure due to the disturbance. FIG. 4 shows the characteristics of the input air pressure and the output air flow rate. When the bleed hole 39 is between the output pressure chamber 31 and the atmospheric pressure chamber 33, a dead zone occurs in the exhaust direction.
[0046]
As described above, if the opening area of the poppet valve 35 and the air supply valve seat 32 or the poppet valve 35 and the exhaust valve seat 34 is kept constant according to the operating principle of the pressure amplifier 14, the air actuator portion 20 ( The flow rate of air fed into (see FIG. 2) can be made constant. This method only needs to add a certain change in the control signal MV value in the electropneumatic conversion mechanism unit 13, but the displacement of the exhaust valve seat 34 changes due to a slight change in the output pressure or the input pressure. It is difficult to keep the opening areas of the poppet valve 35 and the supply valve seat 32 or the poppet valve 35 and the exhaust valve seat 34 constant. Therefore, paying attention to the structure and characteristics of the pressure amplifier 14, the supply pressure chamber 30 and the output pressure chamber 31, or the output pressure chamber 31 with a certain opening area are utilized by using a dead zone peculiar to the pressure amplifier 14 of the system shown in FIG. A method of maintaining the atmospheric pressure chamber 33 can be considered.
[0047]
As described in the operation principle of the pressure amplifier 14, this type of pressure amplifier 14 has a dead zone in the relationship between the input pressure and the output air flow rate. By using the dead zone, it is relatively easy to create a state in which both the poppet valve 35 and the supply valve seat 32 and the exhaust valve seat 34 are closed. This is because the dead band width of the pressure amplifier 14 is relatively wide. That is, by keeping the control signal MV value within a certain range, the output of the electropneumatic conversion mechanism unit 13 is within a certain range, and the input pressure of the pressure amplifier 14 can be within the dead zone of the pressure amplifier 14. In this state, air flows only through the bleed hole 39. The area of the bleed hole 39 is fixed, and when the bleed hole 39 is between the supply pressure chamber 30 and the output pressure chamber 31, the air flow rate flowing into the air actuator portion 20 of the valve portion 15 is constant. Further, when the bleed hole 39 is between the output pressure chamber 31 and the atmospheric pressure chamber 33, the flow rate of air flowing from the air actuator unit 20 of the valve unit 15 is constant.
[0048]
In such a state, the change rate of the valve opening signal PV of the valve unit 15 is measured. Since the area of the bleed hole 39 is known and constant, the pressure amplifier 14 is driven by replacing the area of the bleed hole 39 with the relationship between the poppet valve 35 and the supply valve seat 32 or the relationship between the poppet valve 35 and the exhaust valve seat 34. The response speed of the valve unit 15 that can be obtained can be obtained by calculation. Strictly speaking, even if such a device is used, the air flow rate supplied to the air actuator unit 20 is not constant.
[0049]
By the way, the model when the pressure amplifier 14 enters the dead zone can be expressed as shown in FIG. In such a case, the air flow rate Q flowing through the bleed hole 39 can be expressed by the following formula 1 from the throttle flow rate formula.
[0050]
Q = Cr * A * Ps * Ψ (Po / Ps) Equation 1
Here, Cr: flow coefficient, A: bleed hole area, Ps: supply pressure, Po: output pressure,
Ψ (): Characteristic function.
[0051]
(1) Ψ (Po / Ps) when Ψ (Po / Ps) in the above expression 1 is P1 / Ps <0.528 can be expressed by the following expression 2.
[0052]

[0053]
(2) When Ψ (Po / Ps) in Equation 1 is 0.528 ≦ Po / Ps ≦ 0.9
Ψ (Po / Ps) can be expressed by the following expression 3.
[0054]
Ψ (Po / Ps) = {2 gK / (RT (K−1)}^1/2
・ {(Po / Ps)^{2 / K}-(Po / Ps)^{(K + 1) / K}}^1/2・ ・ ・ ・ ・ Formula 3
[0055]
(3) Ψ (Po / Ps) when Ψ (Po / Ps) in the above expression 1 is 0.9 <Po <Ps can be expressed by the following expression 4.
[0056]
Ψ (Po / Ps) = {2 g / RT}^1/2
・ {(Po / Ps)-(Po / Ps)²}^1/2・ ・ ・ ・ ・ Formula 4
[0057]
Where g = 980 cm / s2, k = specific heat ratio (air = 1.4), R = gas constant 2927 cm / ° k, T = absolute temperature ° k, Cr = flow coefficient, A = area (cm²), P1, P2 = absolute pressure kgf / cm²It is.
[0058]
Therefore, if Ψ (Po / Ps) is measured in the state of Po / Ps / 0.528, the flow rate Q is constant, but Ψ (Po / Ps) is measured in the state of Po / Ps ≦ 0.528. In this case, the flow rate Q changes due to the change in the output pressure Po. However, the error can also be reduced by measuring in a section where the output pressure Po is smaller than the supply pressure Ps and the pressure change is small.
[0059]
In this method, the response speed of the electropneumatic conversion mechanism section 13 is not taken into consideration, but the response speed of the electropneumatic conversion mechanism section 13 is the response speed of the pressure actuator 14 and the air actuator section 15 of the valve section 15. On the other hand, since it is sufficiently fast, it is buried in the measurement error without considering it. In addition, by considering the response speed of the electropneumatic conversion mechanism unit 13 with a certain constant value, the response speed of the control target portion 17 is obtained.
[0060]
Next, a procedure for measuring a specific response speed will be described with reference to a flowchart.
[0061]
The procedure of the first measurement method will be described with reference to the flowchart shown in FIG.
[0062]
First, when measurement is started in response to a setting signal, the control signal MV value is initialized (step ST10). After confirming that the valve opening signal PV value is lower than the preset value Y1, the exhaust valve seat 34 is set to the dead zone of the dead zone of the pressure amplifier 14 as the control signal MV = initial value + ΔMV. (Steps ST11 and ST12). In this state, the poppet valve 35, the supply valve seat 32, and the exhaust valve seat 34 are closed, and only the bleed hole 39 gives an air flow rate to the air actuator portion 20 of the valve portion 15.
[0063]
Next, it is confirmed that the valve opening signal PV value exceeds a preset value Y1 (step ST13). After confirming that Y1 is exceeded, the timer is started and it is confirmed that the valve opening signal PV value exceeds a preset value Y2 that is larger than the value of Y1. After confirming that Y2 is exceeded, the timer is stopped (steps ST14, ST15, ST16).
[0064]
Then, the response speed is obtained by dividing by the value of Y2-Y1 moved and the time required for it. Then, the loop gain, which is a tuning parameter, is obtained from the response speed, and the measurement ends (steps ST17 and ST18).
[0065]
Next, the procedure of the second measurement method will be described with reference to the flowchart shown in FIG.
[0066]
First, when the measurement is started upon receiving the setting signal, the control signal MV value is initialized (step ST20). Then, it is confirmed that the valve opening signal PV value is lower than a preset value Y1. After confirming that it is lower than Y1, the exhaust valve seat 34 is inserted into the dead zone of the pressure amplifier 14 as control signal MV = initial value + ΔMV (steps ST21 and ST22). In this state, the poppet valve 35, the supply valve seat 32 and the exhaust valve seat 34 are closed, and the air flow rate is given to the air actuator portion 20 of the valve portion 15 only by the bleed hole 39. Then, it is confirmed that the valve opening signal PV value exceeds Y1 (step ST23). After confirming that Y1 has been exceeded, the timer is started. Then, after starting the timer, it waits for a unit time (steps ST24 and ST25).
[0067]
The valve opening signal PV value when the unit time has elapsed is set to Y2 (step ST26). The response speed is obtained by dividing the value of Y2−Y1 moved and the unit time required for it. Then, the loop gain, which is a tuning parameter, is obtained from the response speed, and the measurement ends (step ST27).
[0068]
[2] Hysteresis measurement of control target part 17
The hysteresis of the control target portion 17 is a hysteresis that exists in an input / output relationship from the control signal MV output from the control arithmetic device 12 to the valve opening signal PV output from the position sensor 16. This hysteresis is a representative characteristic of the non-linearity of the characteristics of the control target portion 17 and has a great influence on the control. Since the valve opening signal PV does not change until the control target portion 17 passes through the hysteresis, a delay time occurs from when the control signal MV is given until the valve opening signal PV changes. This delay time becomes a dead time and becomes a large phase delay factor. Therefore, if the control arithmetic unit 12 has, for example, a PID control algorithm, it affects the differential time during phase compensation and the tuning parameter of the integration time.
[0069]
This hysteresis measurement method is a method of directly measuring the hysteresis of the valve unit 15 assuming that the hysteresis of the electropneumatic conversion mechanism unit 13 and the pressure amplifier 14 shown in FIG.
[0070]
As shown in FIG. 8, the hysteresis of the valve unit 15 is measured by increasing the control signal MV until the valve opening signal PV obtained from the valve unit 15 changes, and after the valve opening signal PV changes, The signal MV is decreased until the valve opening signal PV obtained from the valve unit 15 changes in the reverse direction. The hysteresis of the valve unit 15 can be calculated by storing the pressure signal and the valve opening signal PV, which are drive signals for the valve unit 15 in this series of operations. Thus, the hysteresis can be measured by measuring a series of data of the pressure signal and the valve opening signal PV. However, since the amount of data increases, a large amount of hardware resources such as a memory are required. Therefore, the hysteresis can be calculated by monitoring the movement of the stem and storing only the pressure data of (3) and (4) in FIG.
[0071]
However, this method is disadvantageous in terms of cost and power consumption because it requires a sensor that measures the pressure that is the drive signal of the valve unit 15. In addition, this method has a demerit that only the hysteresis of the valve portion 15 can be measured.
[0072]
There is also a method for measuring hysteresis without a sensor for detecting the drive signal of the valve unit 15. The method is a method of measuring hysteresis by measuring the input / output relationship between the control signal MV and the valve opening signal PV.
[0073]
That is, the control signal MV is changed slowly, the change of the valve opening signal PV is detected, the control signal MV value MV1 when the change appears in the valve opening signal PV is recorded, and the control signal MV is recorded up to now. MV2 which is a control signal MV value when the valve opening degree signal PV value is changed is recorded in a direction opposite to the changed direction, and hysteresis is obtained by MV1-MV2.
[0074]
However, in this method, since the valve positioner 10 and the valve unit 15 are measured in an open loop, the gain of the electropneumatic conversion mechanism unit 13 and the pressure amplifier 14 is set high with respect to the change in the control signal MV value. Is difficult to measure. Therefore, the present invention proposes a measurement method in which the valve positioner 10 and the valve unit 15 are closed loop.
[0075]
The control algorithm of the control arithmetic device 12 is a control algorithm having at least proportional control. For example, proportional control or proportional / derivative control is used. A certain input value is given to this control algorithm, and after confirming that the value of the valve opening signal PV has been set, the input signal SP is slowly changed. The SP1 of the input signal SP at the time when the value of the valve opening signal PV reacts to the input change is recorded. Next, the input signal SP is slowly changed in the direction opposite to the direction that has been changed. The input signal SP2 at the time when the value of the valve opening signal PV corresponds to the input change is recorded.
[0076]
When this algorithm is proportional control, the disturbance of the control loop (in this case, hysteresis) becomes 1 / loop gain of the control loop, and a steady deviation remains.
[0077]
Here, as shown in FIG. 9, when a unity feedback system is considered, it can be expressed by the following equation (5).
[0078]
Y (S) = SP (S)
* G (S) / (1 + G (S)) + D (S) / (1 + G (S)) Equation 5
Here, G (S): system transfer function, SP (S): input signal, D (S): overall system flow, Y (S): stem displacement.
[0079]
Assuming that there is no change in Y (S) from SP1 (S) to SP2 (S), the following equations 6 and 7 can be obtained.
[0080]
Y1 (S) = SP1 (S) *
G (S) / (1 + G (S)) + D1 (S) / (1 + G (S)) Equation 6
[0081]
Y2 (S) = SP2 (S) *
G (S) / (1 + G (S)) + D2 (S) / (1 + G (S)) Equation 7
[0082]
The amount of change in D (S) corresponds to hysteresis. Here, since Y1 (S) = Y2 (S), the above equation is summarized, and the equation (8) below is obtained by converting the equation for the change amount of D (S).
[0083]
D2 (S) −D1 (S) = (SP1 (S) −SP2 (S)) * G (S) (Equation 8)
Thus, the hysteresis is an expression obtained by multiplying the change in input by the loop gain.
[0084]
Since hysteresis has no sign,
Hys = | SP1-SP2 | * Loop gain
Can be obtained.
[0085]
A specific method for measuring hysteresis will be described with reference to the flowchart shown in FIG.
[0086]
First, when a measurement signal is received and measurement is started, the control algorithm of the control arithmetic device 12 is compared and controlled (step ST30). The loop gain is replaced with a setting parameter for hysteresis measurement, and the input signal SP is set to an initial value (step ST31). It waits for the value of the valve opening signal PV to stop (step ST32). When the valve opening signal PV is at rest, the input signal SP is set to SP + ΔSP, and the input signal SP is gradually changed (steps ST33 and ST34). When the valve opening signal PV starts to move, the value of the input signal SP at that time is stored as SP1 (steps ST33 and ST35). Wait until the valve opening signal PV stops (step ST36). When the valve opening signal PV is stationary, the input signal SP = SP−ΔSP is set, and the input signal SP is gradually changed to decrease (steps ST37 and ST38). When the valve opening signal PV starts to move, the value of the input signal SP at that time is stored as SP2 (step ST39). Hys = | SP1-SP2 | * Hysteresis is obtained by the equation of loop gain (step ST40). Based on the value of Hys, control parameters are calculated. Differential time = f (Hys), integration time = f (HyS).
[0087]
[3] Measurement of slip phenomenon
As shown in FIG. 11, the movement of the stem when the slip phenomenon occurs increases the pressure signal, which is the valve drive signal, and the valve opening signal of the valve It moves quickly and then settles at a rate of change according to pressure changes. As described above, a phenomenon in which the valve opening signal slides at the moment of passing through the hysteresis is called a slip phenomenon, and this phenomenon greatly affects the controllability.
[0088]
Specifically, the section in which the slip phenomenon occurs is in an uncontrollable state. Therefore, for example, when trying to control a very small valve opening such as 0.1%, if the slip phenomenon occurs, the valve opening cannot be stopped at the position of 0.1%, resulting in overshoot. When an overshoot occurs, the control arithmetic device 12 controls the reverse side in an attempt to return the overshoot. At this time, too much overshoot occurs and a limit cycle occurs with this repetition. In order to stop the limit cycle, it is only necessary that the control is not actively performed in the section where the slip phenomenon occurs. Therefore, processing such as changing the control algorithm is required in the section where the slip phenomenon occurs. Although the slip phenomenon has been described with respect to the hysteresis of the valve, this phenomenon occurs due to the backlash of the link mechanism between the position sensor and the stem. Therefore, it is necessary to measure the characteristics of the entire control target.
[0089]
As a measuring method, the control signal MV output from the control arithmetic unit 12 is slowly changed, and the change of the valve opening signal PV is measured. As a phenomenon, since the valve opening signal moves quickly by a certain width at the moment when the control object goes out of hysteresis, dead zone, etc., and then changes normally, the change rate of the valve opening signal may be measured. FIG. 12 shows the rate of change when the valve opening signal PV starts to move from the state where the valve opening signal PV is stationary. In FIG. 12, the width of the slip phenomenon can be measured by detecting the inflection point of the change rate of the valve opening signal PV.
[0090]
That is, in FIG. 11, PV3 which is the value of the valve opening signal PV signal when the valve opening signal PV is stationary is stored, and the valve opening at the inflection point of the changing speed of the valve opening signal PV. By setting the signal PV to PV4, the slip width is obtained by slip width = | PV4-PV3 |.
[0091]
Further, the initial value at which the valve opening signal PV is stationary is set to PV3, the value of the control signal MV is changed, and the valve opening after a short period of time after the valve opening signal PV starts to move. It is good also as data of a slip phenomenon by making signal PV into PV4. However, in this case, there is a possibility that the measurement error becomes large.
[0092]
A specific method for measuring the slip phenomenon will be described with reference to the flowchart of FIG.
[0093]
First, a measurement signal is received and measurement starts. Control signal MV is set to an initial value (step ST50). Wait until the valve opening signal PV stops (step ST51). The control signal MV is set to MV + ΔMV, and the control signal is slowly changed until the valve opening signal PV value moves (steps ST52 and ST53). This operation once resets the hysteresis of the control target. Next, it waits until the valve opening signal PV stops. The PV value when the valve opening signal PV is stationary is stored as PV3. That is, PV3 = PV (steps ST54 and ST55).
[0094]
Next, the control signal MV = MV−ΔMV is set, and the control signal is slowly changed in the opposite direction. The rate of change of the current valve opening signal PV is measured. The process is repeated until an inflection point of the change rate of the valve opening signal PV is detected (steps ST56, ST57, ST58, ST59).
[0095]
Then, the valve opening signal when the inflection point of the changing speed of the valve opening signal PV is detected is stored as PV4. That is, PV4 = PV, slip width = | PV4-PV3 |, and control algorithm switching condition = f (slip width) are determined (steps ST60 and ST61).
[0096]
[4] Operating point measurement of electropneumatic conversion mechanism unit 13
The electropneumatic conversion mechanism unit 13 performs electropneumatic conversion by driving the nozzle flapper mechanism with an electromagnetic actuator and changing the nozzle back pressure. In the case of the valve positioner 10, the electropneumatic conversion mechanism unit 13 is required to work with low power because the electric energy supplied from the outside is limited, and the energy that can be supplied to the electromagnetic actuator is limited. Therefore, in order to increase the conversion gain required for the electropneumatic conversion mechanism unit 13, the air flow rate before the nozzle flapper mechanism is reduced to increase the gain of the nozzle flapper mechanism. As a result, electro-pneumatic conversion is performed in a slight gap between the nozzle and the flapper, which is sensitive to disturbance and the range of supply pressure supplied to the valve positioner 10 is wide. The drive signal of the electropneumatic conversion mechanism unit 13 is designed to be significantly wider than the operating point, and is designed to absorb the operating point shift due to disturbance and supply pressure fluctuation.
[0097]
The control arithmetic device 12 has an integrator built therein, and changes the control signal MV according to the disturbance to absorb the disturbance. Therefore, when the value of the integrator is reset after resetting, such as when the power is turned on, the operating point of the electropneumatic conversion mechanism unit 13 is deviated, and the valve unit 15 is in a state where the integrator is wound up. It takes a long time for the valve opening signal PV to return to a steady value, and it cannot rise quickly.
[0098]
In order to solve this problem, if the output of the integrator is stored in a non-volatile memory or the like, the operating point shift of the electropneumatic conversion mechanism unit 13 can be reduced even at the time of reset such as when the power is turned on. It is virtually impossible to store the output of the storage device in a nonvolatile memory or the like. This is because the non-volatile memory has a limit value for the number of updates, and when data is regularly updated, it will eventually deteriorate and fail.
[0099]
As a solution to this problem, the operating point of the electropneumatic conversion mechanism unit 13 is measured, and the operating point is stored as an offset (Offset) in a non-volatile memory. The operating point can be corrected by loading.
[0100]
Taking the PID control algorithm as an example, for example, when the operating point of the electropneumatic conversion mechanism unit 13 is the control signal MV = 50, when the initial value when the integrator is reset is zero, the input signal SP and the valve opening Assuming that the deviation of the signal PV is E, a steady deviation remains until the integrator output = 1 / Ti∫Edt = 50 (Ti; integration time). In particular, when the integration time is long, it takes time until the steady deviation disappears. Therefore, if the operating point of the electropneumatic conversion mechanism unit 13 is measured in advance and stored as an offset value and applied to the PID calculation unit, the operating point can be corrected, even if the integrator is reset. Rise time can be greatly improved. That is, the control signal MV = Kp * (P + I + D) + Offset may be used.
[0101]
In the measurement method, the control algorithm of the control arithmetic unit 12 is set to at least a control algorithm with an integrator, the input signal SP is set to 50%, and the process waits until the deviation enters a set value, for example, ± 1%. The value of the control signal MV at this time is stored as an Offset value. In this way, the rise time from the next reset can be shortened.
[0102]
A specific method for measuring the operating point of the electropneumatic conversion mechanism unit 13 will be described with reference to the flowchart shown in FIG.
[0103]
First, measurement is started upon receiving a setting signal. The control algorithm of the control arithmetic device is set to proportional integral (PI) control (step ST70). Control for giving an initial value to the input signal SP starts, and the valve opening signal PV approaches the input signal SP (step ST71). Control is continued until the deviation between the input signal SP and the valve opening signal PV becomes a certain set value or less. The control signal MV value at the time when the deviation falls below a certain set value is stored as Offset. Then, the measurement ends (steps ST72 and ST73).
[0104]
[5] About the diagnostic function
When the characteristic of the control target part 17 is measured and automatically tuned, if the characteristic of the measured control target part 17 is greatly deviated from that of a general control target part, this is indicated by the valve positioner. Since 10 installation errors are also considered, an error message or warning message can be issued. For example, when the response speed measurement result of the control target portion 17 is orders of magnitude slower than a normal value, leakage of the air actuator unit 20 (see FIG. 2) of the valve unit 15 or a setting error of the supply air pressure Etc. are considered. If the measurement result of the hysteresis of the control target portion 17 is larger than a normal value, it is possible that the valve portion 15 is flawed, the link of the displacement sensor to the stem is abnormal, or the like. Further, when the measurement result of the slip width of the control target portion 17 is larger than a normal value, a failure of the position sensor 16 or a failure of the valve unit 15 can be considered. Furthermore, when the measurement result of the operating point of the electropneumatic conversion mechanism unit 13 is greatly deviated from a normal value, there may be a setting error of the supply air pressure, a failure of the electropneumatic conversion mechanism unit 13, or the like. In such a case, a message that informs the operator is output after the automatic tuning is completed.
[0105]
【The invention's effect】
As described above, the valve positioner according to the present invention has the following effects.
[0106]
(1) Since the valve positioner has an automatic tuning function, anyone can easily tune the valve positioner regardless of the operator.
[0107]
(2) Further, by providing the valve positioner with an automatic tuning function, there is an effect that the man-hours for starting up the valve positioner can be reduced.
[0108]
(3) Furthermore, there is an effect that accurate tuning can be performed by obtaining a tuning parameter by calculation from the characteristics of the control target part.
[0109]
(4) By measuring the response speed of the part to be controlled, there is an effect that more accurate tuning information can be obtained as well as the characteristics of the valve.
[0110]
(5) By measuring the hysteresis of the part to be controlled, there is an effect that more accurate tuning information can be obtained.
[0111]
(6) There is an effect that more accurate tuning information can be obtained by measuring the slip width of the part to be controlled.
[0112]
(7) By measuring the operating point of the electropneumatic conversion mechanism, there is an effect that the rise time from the reset of the valve positioner can be shortened.
[0113]
(8) By measuring the part to be controlled and issuing an error message or warning message depending on the measurement result, there is an effect that an installation error of the positioner can be prevented in advance.
[Brief description of the drawings]
FIG. 1 is a block diagram schematically showing the overall configuration of a valve positioner according to the present invention.
FIG. 2 is an explanatory view schematically showing a valve portion in FIG. 1;
3 is a plan view showing the structure of the pressure amplifier in FIG. 1. FIG.
4 is a graph showing a dead zone due to nozzle back pressure and air flow characteristics in the pressure amplifier in FIG. 3;
5 is an explanatory view showing a configuration of a bleed hole of the pressure amplifier in FIG. 3. FIG.
FIG. 6 is a flowchart showing a method for measuring a response speed in the control target part.
FIG. 7 is a flowchart showing a method of measuring a response speed in the control target part.
FIG. 8 is an explanatory diagram showing a method for calculating the hysteresis of the valve portion.
FIG. 9 is a conceptual diagram showing the unity feedback system.
FIG. 10 is a flowchart for obtaining the hysteresis.
FIG. 11 is an explanatory view showing the movement of the stem when the slip phenomenon occurs.
FIG. 12 is a graph showing an inflection point of the stem speed.
FIG. 13 is a flowchart showing a specific measuring method for measuring the slip width.
FIG. 14 is a flowchart showing a method of measuring a control signal in a control arithmetic device equipped with the integrator.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10; Valve positioner, 11; Signal receiving device, 12; Control arithmetic device, 13; Electropneumatic conversion mechanism part, 14; Pressure amplifier, 15; Valve part, 16; Position sensor, 17; Part, 21; diaphragm, 22; spring, 23; stem, 24; valve body, 30; supply pressure chamber, 31; output pressure chamber, 32; supply valve seat, 33; atmospheric pressure chamber, 34; exhaust valve seat, 35; Poppet valve, 36; Output pressure diaphragm, 37; Input pressure diaphragm, 38; Input pressure chamber, 39; Bleed hole

Claims (4)

An input signal for setting the valve opening of the valve, a position sensor for detecting the valve opening of the valve, and a valve opening of the valve from a deviation between the valve opening signal obtained by the position sensor and the input signal Generated in the electropneumatic conversion mechanism unit, a control calculation unit that generates a control signal by performing control calculation so as to match the input signal, an electropneumatic conversion mechanism unit that generates an air flow rate based on the control signal A valve positioner comprising a pressure amplifier for supplying air pressure based on the air flow rate to the valve,
The control object part composed of the electropneumatic conversion mechanism part, the pressure amplifier, the position sensor, and the valve receives an automatic setting signal,
Measuring a slip width of a slip phenomenon that occurs when the valve opening degree of the valve starts moving from a stationary state, and calculating a tuning parameter of a control signal generated by the control calculation unit based on a parameter including the measured slip width <br> A valve positioner characterized by When measuring the slip width of the slip phenomenon that occurs when the valve opening of the valve starts moving from a stationary state, it is confirmed that the valve opening signal of the valve is stationary, and the valve opening PV1 at that time is stored. Then, when the control signal is changed, the change rate of the valve opening signal when the valve opening signal reacts is measured, and the valve opening signal PV2 at the inflection point of the change rate of the valve opening signal is obtained. 2. The valve positioner according to claim 1, wherein the valve positioner stores the difference between the valve opening signals PV1 and PV2 as the slip width of the valve and stores the value of the slip width. When measuring the slip width of the slip phenomenon that occurs when the valve opening of the valve starts to move from a stationary state, confirm that the valve opening signal is stationary, store the valve opening PV1 at that time, Next, after the valve opening signal of the valve reacts when the control signal is changed, the valve opening signal PV2 after a short time set in advance is stored, and the difference between the valve opening signals PV1 and PV2 is The valve positioner according to claim 1, wherein the valve width is stored as a slip width of the valve. The control target part composed of the electropneumatic conversion mechanism part, the pressure amplifier, the position sensor, and the valve measures the characteristics of the control target part when receiving the automatic setting signal, and the measured control target part 4. The valve positioner according to claim 2, wherein when the value of the characteristic deviates from the value of the allowable range information including the characteristic of the part to be controlled, an abnormal signal is output to the outside through communication means.

Images