JP2004116416A - Method for operating combined cycle power generation plant - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for operating a combined cycle power generation plant starting operation of the plant for a short time according to operation restraints without generating excessive thermal stress of the plant. <P>SOLUTION: This method includes a stage of establishing operation parameters of the plant as operation variables, a stage of establishing monitor parameters of the plant as restraints, a stage of establishing lead time PT for reaching designated power generation output as evaluation indexes, and a stage of solving an optimization problem for minimizing the evaluation indexes based on the operation variables and the restraints before start of operation and determining the operation parameters as the operation variables, in operation of the combined cycle power generating plant from start to the designated power generation output. <P>COPYRIGHT: (C)2004,JPO

Description

ここで、ｘ（ｔ）は状態ベクトル、ｕ（ｔ）は入力ベクトル、ｙ（ｔ）は出力ベクトル、ｔは時刻である。また、Ｆ，Ｇはプラントの動特性を表す式を、初期時刻ｔ０での初期状態ｘ（ｔ０）を初期値とした初期値問題を解くことにより求めることができる。
【００７９】
このような常微分方程式の解法では、初期状態ｘ（ｔ０）の値によりその後の結果が影響を受ける。したがって、方程式を記述する方法が正しく行われても、初期状態ｘ（ｔ０）の値が異なっていると、動的シミュレーションによる将来のプラントの運転状態は、実際のプラントの運転状態とは異なったものになる。
【００８０】
一方、初期状態の値は、プラントで実際に観測される量、すなわちプロセスデータから決定することができる。これにより、動的シミュレーションの結果の精度がよくなり、さらに、この結果を用いて最適問題を解いて得られるプラントの運転パラメータも、精度の良い値となる。
【００８１】
（第６の実施の形態）
本発明によるコンバインドサイクル発電プラントの運転方法に係る第６の実施の形態は、第１から５までの実施の形態における最適化計算を、図７に示すように、発電プラントの運転の進行に従って、所定の時間周期毎に繰り返し、運転パラメータを逐次更新するものである。この時、最適化計算に含まれる動特性シミュレーションのための初期値は、発電プラントの測定データが利用できるものについては、その計算時刻における発電プラントのプロセスデータを用いる。
【００８２】
第５の実施の形態でも説明したように、本発明では、発電プラントの将来の運転状態の予測計算の精度が、その予測計算を含む最適化計算により求める運転パラメータに影響を与える。
【００８３】
発電プラントの将来の振る舞いについての動特性シミュレーションによる予測計算では、計算のためのモデル式、計算に使うデータ、初期値、外部からの入力変数等に必ず誤差があるため、将来の予測値に誤差が生じる。
【００８４】
さらに、動特性シミュレーションの式には、モデル化されていない特性や、モデル式に含まれない外部からの影響等の要因によっても、将来の予測値に誤差が生じる。これらの要因による誤差は、予測期間が長くなればなるほど、誤差が累積して大きくなる。
【００８５】
本実施の形態は、実際のプラントの運転状態を取り込んで、予測計算と運転パラメータの計算を行い、これを運転の進行に従って逐次繰り返し行うことにより、予測誤差の累積を防ぐものである。これにより、実機の運転状況を反映した運転パラメータを算出することができる。
【００８６】
図８は、運転パラメータを決定する際の流れを示したものである。同図（ａ）は、通常の最適化計算の場合を示しており、同図（ｂ）は、本実施の形態における逐次最適化計算の場合を示している。本実施の形態においては、逐次計算を行うことがフィードバック効果を持たせているものと解釈することができる。
このフィードバックには、元来、制御対象のモデル化の誤差や、操作量以外の外部からの入力の影響に対して、感度を低減する効果がある。
【００８７】
要するに、実際のプラントの運転状態を取り込んで、予測計算と運転パラメータの計算を行い、これを運転の進行に従って逐次繰り返し行うことにより、実機の運転状態をよく反映した、誤差の小さい運転パラメータを算出することができる。
【００８８】
（第７の実施の形態）
本発明によるコンバインドサイクル発電プラントの運転方法に係る第７の実施の形態は、第１から第６の実施の形態で述べた運転パラメータの決め方に対して、蒸気タービンへの蒸気の通気以降の運転についての運転パラメータの決め方に係るものである。
【００８９】
本実施形態の最適化計算では、操作変数として、蒸気タービンの熱応力監視用の蒸気温度変化パターンを与え、制約条件として、蒸気温度の変化パターンから計算される、将来にわたる蒸気タービンの軸に発生する熱応力の予測値の最大値ＭＳＳ、および蒸気温度の変化は非減少であることを設定し、評価指標として、指定された蒸気温度に到達する時間ＰＴを設定し、その評価指標を最小にする最適問題を解いて蒸気温度のパターンを求め、得られた蒸気温度のパターンからコンバインド発電出力の運転パラメータを決定する。ここで、操作変数である温度変化パターンの与え方としては、予測計算の全時間にわたる一定の温度上昇率ＴＶ、所定の時間間隔毎に設定される，温度Ｔあるいは温度上昇率ＴＶのいずれかとする。
【００９０】
また、蒸気タービンの通気までの運転方法については，第１から６までの実施の形態の運転パラメータの決定方法を利用する。
【００９１】
最適化計算により運転パラメータを決定する場合には、操作変数が多くなったり、予測期間が長いと大きな計算量を必要とする。最適化計算を部分的にするので、計算量を抑えて、計算を実行する計算機の規模を低減することができる。
【００９２】
（第８の実施の形態）
本発明によるコンバインドサイクル発電プラントの運転方法に係る第９の実施の形態は、ガスタービン設備と蒸気タービン設備から構成されるコンバインドサイクル発電プラントの起動から所定の発電出力までの運転において、蒸気タービンへの蒸気の通気以降の運転を行う際に、操作変数として、蒸気タービンの熱応力監視用の蒸気温度変化パターンを与え、制約条件として、指定された蒸気温度に到達する時間ＰＴおよび蒸気温度の変化は非減少の条件を設定し、評価指標として、蒸気温度の変化パターンから計算される、将来にわたる蒸気タービンの軸に発生する熱応力の予測値の最大値ＭＳＳを設定し、評価指標を最小にする最適問題を解いて蒸気温度のパターンを求め、得られた蒸気温度のパターンから発電出力の運転パラメータを決定するものである。
【００９３】
ここで、操作変数である温度変化パターンの与え方としては、予測計算の全時間にわたる一定の温度上昇率ＴＶ、所定の時間間隔毎に設定される，温度あるいは温度上昇率ＰＶのいずれかとする。
【００９４】
また、蒸気タービンの通気までの運転方法については、第１から６までの実施の形態における運転パラメータの決定方法を利用する。
【００９５】
最適化計算による運転パラメータの決定においては、操作変数が多くなったり、予測期間が長いと大きな計算量を必要とする。第７の実施の形態と同様に、最適化計算を部分的にすることにより、計算量を抑えて、計算を実行する計算機の規模を低減することができる。
【００９６】
（第９の実施の形態）
本発明によるコンバインドサイクル発電プラントの運転方法に係る第９の実施の形態は、第７，８の実施の形態に述べたように、最適化計算を部分的とした場合に、第６の実施の形態で述べた、発電プラントの運転の進行に従って、所定の時間周期毎に繰り返し、運転パラメータを逐次更新するものである。
【００９７】
最適化計算に含まれる動特性シミュレーション計算のための初期値は、第６の実施の形態と同様に、発電プラントの測定データが利用できるものについては、その計算時刻における発電プラントのプロセスデータを用いる。
【００９８】
このように構成することにより、第７，８の実施の形態に述べたように、最適化計算を部分的とした場合には、第６の実施の形態で述べたモデル化誤差やプラント・パラメータの誤差、未知入力の影響を低減できる。
【００９９】
最適化計算による運転パラメータの決定を部分的にすることで、計算を実行する機器の規模を低減することができる。
【０１００】
さらに、最適化計算におけるモデル化誤差やプラント・パラメータの誤差、未知入力の影響を低減することができる。
【０１０１】
（第１０の実施の形態）
本発明によるコンバインドサイクル発電プラントの運転方法に係る第１０の実施の形態は、コンバインドサイクル発電プラントを起動から蒸気タービンへの通気まで運転するに当り、請求項１乃至６のいずれか１項記載の方法を利用し、蒸気タービンへの蒸気の通気以降の運転について請求項７から９までのいずれか１項の方法を利用して運転パラメータを決定するものである。
【０１０２】
このように構成することにより、蒸気タービン通気までの運転については通常の最適化計算による運転パラメータを利用でき、蒸気タービン通気以降は主要な制約である蒸気タービンの熱応力を考慮した運転を実現することができる。
【０１０３】
最適化計算による運転パラメータの決定を２分割することで、計算を実行する機器の規模を低減でき、制約条件を考慮した短時間での起動を実現することができる。
【０１０４】
（第１１の実施の形態）
本発明によるコンバインドサイクル発電プラントの運転方法に係る第１１の実施の形態は、第７乃至１０の実施の形態において、蒸気温度の変化パターンから将来にわたる蒸気タービンの軸に発生する熱応力の予測値を計算する際に、監視に用いられている線形差分方程式から求めた予測計算式を利用して、運転パラメータを決定するものである。
【０１０５】
一般に、蒸気タービンの軸は、大きな金属で作られているために、蒸気の注入による軸表面の温度上昇により、軸内部に温度分布を生じ、熱応力を発生する。
熱応力の予測計算式は、次ぎのように求めることができる。
【０１０６】
蒸気タービンの軸の熱応力は、軸の温度分布により発生する。図９に示すように、軸は表面が蒸気に曝された円板状の金属とみなされるので、軸に発生する温度分布は表面の蒸気温度との熱交換による熱の移動を境界条件として、熱伝導方程式によって表すことができる。円板における温度分布を求める熱伝導方程式は、時刻ｔ場所ｒの温度をＴ（ｔ，ｒ）と表すと、金属の性質と形状から決まる係数をａとおいて、次のような、極座標表示による偏微分方程式になる。

また、体積平均温度をＴａ（ｔ）と表すと、
Ｔａ（ｔ）＝２／（ｒ２ーｒ１）∫Ｔ（ｔ，ｒ）・ｒ・ｄｒ　　（２）
となる。ここで、ｒ２は軸の外径の２乗、ｒ１は軸の内径の２乗、積分は内径から外径まで行うものとする。
【０１０７】
熱応力は、体積平均温度Ｔａ（ｔ）と円板上の温度Ｔ（ｔ，ｒ）の差に比例した値になる。
【０１０８】
発電プラントの運転においては、これらの式を時間と場所について差分化した差分方程式が用いられている。式（２）は線形の式なので、差分化した熱応力の式も、以下に示すように線形の差分方程式になる。
【０１０９】
なお、これ以降は、温度変数としてＴの代わりにＸを用いる。離散化したサンプル時刻ｋでの温度分布ベクトルは、Ｘ（ｋ）と表される。この温度分布ベクトルＸ（ｋ）の転置ベクトルを［Ｘ１（ｋ），Ｘ２（ｋ），・・・，Ｘｎ（ｋ）］^Ｔと表す。
【０１１０】
熱応力をｙ（ｋ）、蒸気温度をｕ（ｋ）とすると、熱応力は次の差分方程式となる、
Ｘ（ｋ＋１）＝Ａ・Ｘ（ｋ）＋Ｂ・ｕ（ｋ）　　（３）
ｙ（ｋ）＝Ｃ・Ｘ（ｋ）　　　　　　　　　　　（４）
ここで，定数行列Ａ，Ｂ，Ｃは熱伝導方程式，体積平均温度の式を差分化して求められる、分割数，軸の形状，材質等から決まる定数行列である。
これにより、現時刻をｋとして、ｍ期間にわたる熱応力の推定値ベクトルＹの転置ベクトルを、
［ｙ（ｋ＋１），ｙ（ｋ＋２），・・・，ｙ（ｋ＋ｍ）］^Ｔ、
将来のｐ期間先にわたる操作変数の蒸気温度ベクトルＵの転置ベクトルを、
［ｕ（ｋ＋１），ｕ（ｋ＋２），・・・，ｕ（ｋ＋ｐ）］^Ｔ、
とすると、熱応力の推定値ベクトルＹは、操作変数ベクトルＵを使って、
Ｙ＝Ｆ・Ｘ（ｋ）＋Ｇ・Ｕ　　　（５）
と表すことができる。このように、熱応力の推定値Ｙが、蒸気温度Ｕを与えることにより決定できる。
【０１１１】
逆に、熱応力の推定値Ｙを与えれば、将来の蒸気温度Ｕの値を決定することができる。例えば、行列Ｇの転置行列をＧ^Ｔと書くと、（５）式から、最小二乗法を用いて、
Ｕ＝Ｈ・Ｇ^Ｔ・（Ｙ−Ｆ・Ｘ（ｋ））　　　（６）
として、求めることができる。ここで、行列Ｈは、Ｇ^Ｔ・Ｇの逆行列である。
【０１１２】
蒸気温度のパターンから発電出力のパターンへの変換は、図１０に示すように、
事前に求めておいた蒸気温度と発電出力の関係から決定できる。
【０１１３】
このように熱応力の予測式を用いることにより、計算量の多い動特性シミュレーションを実施することなく、最適な運転パラメータを求めることができる。
【０１１４】
すなわち、最適な運転パラメータを短時間で求めることができ、計算能力や記憶容量などについて大規模な制御装置に用いることができる。
【０１１５】
（第１２の実施の形態）
本発明によるコンバインドサイクル発電プラントの運転方法に係る第１２の実施の形態は、第１１の実施の形態で求めたように、第７乃至１０の実施の形態において、蒸気温度の変化パターンから将来にわたる蒸気タービンの軸に発生する熱応力の予測値ＳＳを計算する場合、熱応力の監視に用いられている線形差分方程式から求めた予測計算式によって求め、その結果を用いて運転パラメータを決定するときに、線形の最大値・最小値問題を利用するものである。
【０１１６】
この線形の最大値・最小値問題を利用するものには、２つの方法があるので、それぞれについて、図１１および図１２を用いて説明する。
【０１１７】
先ず、第１番目の方法は、熱応力を制約条件として、最短時間で起動するために温度パターンを計算する方法である。図１１は、その計算のフローチャートであり、その手順は、以下のとおりである。
（１）蒸気温度の操作回数Ｎの初期値を設定する。
（２）軸の温度分布の初期値Ｘ（０）を設定する。
（３）操作変数として、蒸気タービンの熱応力監視用の蒸気温度変化パターンＵを与え、制約条件として、指定された蒸気温度に到達する時間Ｍ、蒸気温度の変化は非減少の条件を設定し、評価指標として、蒸気温度の変化パターンから計算される、将来にわたる蒸気タービンの軸に発生する熱応力の予測値Ｙの最大値Ｊを与えて、最大値・最小値問題を解く。
（４）熱応力が制限値以下であれば、操作回数を減らして（１）に戻る。
制限値以上であれば、次に行く。
（５）（３）の計算で制限値以下を満たしたものがなければ、蒸気温度の操作回数Ｎを増やして（１）に戻る。（３）の計算で制限値以下を満たしたものがあれば、次に行く。
（６）前回の結果を蒸気温度の運転パターンとして採用する。
【０１１８】
次に、第２番目の方法は、起動時間を制約条件として、熱応力を最小として起動できるような蒸気温度パターンを計算する方法である。図１２は、このときの蒸気温度パターンを計算するためのフローチャートである。この場合は、一回だけ最適化問題を解くことにより解を得ることができる。
【０１１９】
このように、線形最大値・最小値問題を解くという形に、最適化計算が構成されており、高速で最適解を得ることができる。第９の実施の形態で述べた逐次最適化計算は、決められた制御周期毎に最適化計算を行うことが必要であるが、本実施の形態では、一回だけで最適化問題を解くことができるので、繰り返しの必要がない。したがって、計算能力の高くない制御システムにおいても、このような計算を容易に実現することができる。
【０１２０】
（第１３の実施の形態）
本発明によるコンバインドサイクル発電プラントの運転方法に係る第１３の実施の形態は、第１２の実施の形態の線形の最大値・最小値問題を利用する最適化計算の代わりに、線形計画法を用いるものである。
【０１２１】
線形の最大値・最小値問題と線形計画法は、前述の参考文献にも示されているように、図１３に示すような関係で相互に置き換えることができる。したがって、第１２の実施の形態の１番目の方法および２番目の方法において、最大値・最小値問題を使う部分を線形計画法で置き換えることができる。
【０１２２】
線形計画法を用いることにより、有限の計算回数で最適解を得ることができるので、計算時間の短縮を図ることができ、制御システムに高い計算能力を要求する必要がなくなる。
【０１２３】
（第１４の実施の形態）
本発明によるコンバインドサイクル発電プラントの運転方法に係る第１４の実施の形態は、第７から第１３までの実施の形態において、最適な運転を与えるように求めた温度パターンと、実際に熱応力を管理するために用いられる温度とが、一致するように制御を行うものである。
【０１２４】
図１４は、このような実施の形態による制御アルゴリズムを示したものである。最適化計算で求めた蒸気温度のパターンを目標値とし、熱応力の管理のために計測されている蒸気温度の計測値と比較して、偏差がなくなるようにフィードフォワード制御信号を作る。一方、予め求めた蒸気温度と発電出力との関係から発電出力信号を設定しておき、この発電出力信号と先のフィードフォワード制御信号を加え合わせて、発電出力の目標値を作成する。
【０１２５】
蒸気温度と発電出力の関係が容易に求められない場合には、フィードフォワード制御信号だけで発電出力の目標値とすることも可能である。
【０１２６】
このような制御アルゴリズムを構成することにより、最適化計算で求めた蒸気温度のパターンを目標値とし、予め求めた蒸気温度と発電出力との関係から設定した発電出力の目標値を修正することができる。
【０１２７】
【発明の効果】
本発明のコンバインド発電プラントの運転方法によれば、プラントの運転制約条件を考慮に入れて、起動時間を最小にする最適化計算により運転パラメータを求め、この運転パラメータに従ってプラントを運転することにより、過大な熱応力を発生させずに、短時間でプラントの運転を起動することができる。
【図面の簡単な説明】
【図１】第１の実施の形態の構成図。
【図２】第１の実施の形態の作用の流れを示すフローチャート。
【図３】第３の実施の形態において出力上昇区分化による最適計算モデル１を示す図。
【図４】第３の実施の形態において出力上昇区分化による最適計算モデル２を示す図。
【図５】第５の実施の形態において最適化計算による運転パラメータを決定するフローチャート。
【図６】第５の実施の形態において運転パラメータ計算システムを有する制御システムの構成図。
【図７】第６の実施の形態において運転パラメータを決定する方法１を示す図。
【図８】第６の実施の形態において運転パラメータを決定する方法２を示す図であり、（ａ）は開ループの場合、（ｂ）は本実施の形態の閉ループの場合を示す。
【図９】第１１の実施の形態において熱応力の計算方法を示す図。
【図１０】第１１の実施の形態において目標値変換を示す図。
【図１１】第１２の実施の形態において蒸気温度のパターンの計算の流れ１を示すフローチャート。
【図１２】第１２の実施の形態において蒸気温度のパターンの計算の流れ２を示すフローチャート。
【図１３】第１３の実施の形態において等価性を説明する図。
【図１４】第１４の実施の形態において発電出力目標値を作る制御アルゴリズムを示す図。
【図１５】従来のコンバインドサイクル発電プラントの構成図。
【図１６】従来のコンバインドサイクル発電プラントにおける燃料流量と、発電出力、排ガス流量、排ガス温度、および蒸気温度の関係を示す図。
【図１７】従来のコンバインドサイクル発電プラントの起動時の運転の一例を示す図。
【図１８】従来のコンバインドサイクル発電プラントの制御システムの構成図。
【符号の説明】
１…ガスタービン設備、１ａ…圧縮機、１ｂ…ガスタービン、２…廃熱回収ボイラ、３…蒸気タービン設備、３ａ…蒸気タービン、４…発電機（ガスタービン軸）、４ａ…発電機（蒸気タービン軸）、５…復水器、６…ポンプ、
７…燃焼器、８…燃料弁、９…蒸気弁（１）、１０…蒸気弁（２）、１１…従来の制御装置、１１ａ…第１の実施形態の制御装置、１１ｂ…第５の実施形態の制御装置、１２…上位制御装置、１２ａ…ガスタービン用下位制御装置、１２ｂ…蒸気タービン用下位制御装置、１３ａ…ガスタービン用入出力装置、１３ｂ…蒸気タービン用入出力装置、１４…発電プラント、１５…運転パラメータ、１５ａ…ガスタービン用運転パラメータ、１５ｂ…蒸気タービン用運転パラメータ、１６…操作信号、１６ａ…ガスタービン用操作信号、１６ｂ…蒸気タービン用操作信号、１７ａ…ガスタービン用プロセスデータ、１７ｂ…蒸気タービン用プロセスデータ、１８…運転パラメータ計算システム、１９…運転データ処理システム、２０…未知の入力、２１…運転目標値、２２…初期値データ、２３ａ…ガスタービン用操作器、２３ｂ…蒸気タービン用操作器、２４ａ…ガスタービン用検出器、２４ｂ…蒸気タービン用検出器、１８１…最適計算、１８２…予測計算、１９１…データ選択、１９２…プラントデータ収集システム[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an operation method of a combined cycle power plant, and more particularly to an operation method of starting the plant.
[0002]
[Prior art]
FIG. 15 is a schematic configuration diagram of a conventional combined cycle power plant. (For example, see Non-Patent Documents 1 and 2).
[0003]
The combined cycle power plant includes a gas turbine facility 1, a waste heat recovery boiler 2, a steam turbine facility 3, generators 4, 4 ', a condenser 5, a pump 6, a combustor 7, a fuel valve 8, and a steam valve (1). 9, a steam valve (2) 10 and the like.
[0004]
In the gas turbine facility 1, the intake air is compressed by the compressor 1 a to become high-temperature and high-pressure air, which is guided to the combustor 7. On the other hand, the fuel supplied from the fuel valve 8 is also guided to the combustor 7, where it is mixed with the high-temperature and high-pressure air from the compressor 1a and burned. The high-pressure and high-temperature combustion gas is guided to the gas turbine 1b, drives the gas turbine 1b, and rotates the generator 4 coaxially coupled. On the other hand, the flue gas after driving the gas turbine 1b is guided to the waste heat recovery boiler 2, generates steam in a heat exchanger (not shown), drives the steam turbine 3a, and coaxially couples the generator 4a. To rotate. The steam worked in the steam turbine 3a is condensed in the condenser 5 to return to water, and the pressure is increased by the pump 6 and sent to the waste heat recovery boiler 2 again. The steam (or water) used in the steam turbine 3a circulates in this manner.
[0005]
As valves used for main control, a fuel valve 8 for controlling the fuel flow rate of the gas turbine equipment 1 and a steam valve (1) for controlling the amount of steam generated in the waste heat recovery boiler 2 and flowing to the steam turbine 3a 9 and a steam valve (2) 10 for controlling the amount of bypass to the condenser 5 directly without passing through the steam turbine.
[0006]
This figure shows a configuration in which the gas turbine equipment 1 and the steam turbine equipment 3 have separate rotating shafts and have separate generators 4 and 4a, respectively. As a configuration of combined cycle power generation in other cases, there is also a configuration in which the gas turbine 1, the steam turbine 3, and the generators 4, 4a are connected by the same rotating shaft.
[0007]
Further, there is also a configuration in which the gas turbine equipment 1 and the waste heat recovery boiler 2 form a plurality of pairs and supply steam to one steam turbine. Since the present invention is not related to the difference between these configurations, the configuration in FIG. 15 will be described below.
[0008]
FIG. 16 shows the relationship between the fuel flow rate of the gas turbine equipment 1 and the power generation output by the gas turbine equipment 1, the exhaust gas flow rate, the exhaust gas temperature, and the steam temperature. As the fuel flow rate of the gas turbine equipment 1 increases, the power generation output of the gas turbine equipment 1 increases, and the exhaust gas temperature also increases while the exhaust gas flow rate is constant.
[0009]
Since the temperature of the steam generated in the waste heat recovery boiler 2 changes according to the exhaust gas temperature, the steam temperature rises while the gas turbine equipment 1 is started and the output is increased.
[0010]
Since the steam turbine equipment 3 and the waste heat recovery boiler 2 flow high-temperature and high-pressure steam and water, the metal parts are thick to maintain strength. When a temperature distribution occurs in a thick metal portion, thermal stress is generated. Since a large thermal stress consumes the life of the metal, an operation in which the thermal stress is controlled is required to avoid this.
[0011]
Further, such a temperature distribution is likely to occur at the start of a plant where the temperature difference between metal and steam is large. Therefore, it can be said that the thermal stress of the metal of the plant mainly occurs at the time of starting the plant.
[0012]
FIG. 17 shows a procedure of an operation operation when starting up a general combined cycle power plant.
(1) At the start of operation, the gas turbine equipment starts to rotate by auxiliary power such as a motor and the like, and after reaching a predetermined rotation speed, keeps the rotation speed for a predetermined time.
(2) After a lapse of a predetermined time, the output of the auxiliary power is reduced, the number of revolutions is reduced, and fuel is supplied to the gas turbine equipment and ignited.
(3) When the combustion starts, the supply amount of fuel is increased in accordance with a predetermined rate of change of the rotation speed, and the rotation speed is increased.
(4) When the rotation speed reaches a predetermined value, that is, 3000 rpm or 3600 rpm, which is the rotation speed corresponding to the frequency of the power system, the rotation speed is kept constant, and the power line of the generator is connected to the power system to generate power. Start.
(5) The fuel to the gas turbine equipment is increased at a predetermined change rate, and the output is increased to a predetermined value.
(6) When the pressure of the steam generated by the waste heat recovery boiler by the high-temperature exhaust gas of the gas turbine equipment rises to a predetermined value, the steam is bypassed to the steam turbine and directly flows to the condenser.
(7) When the amount of steam exceeds a predetermined amount, start flowing steam to the steam turbine, start reducing the amount of steam flowing from the bypass to the condenser, increase the amount of steam flowing to the steam turbine, and further reduce the entire amount of steam to the steam turbine. Pour
(8) Increase the output of the gas turbine until the target power generation output is reached. Naturally, the output of the steam turbine also increases.
In these driving operations, (1) to (4) are performed by a substantially fixed program. The operation after (5) is determined by the state of the plant at the time of startup.
[0013]
Among these operating parameters, the main ones are the output GO of the initial load holding, the pressure GSP of the steam during the load holding, the switching time CT for the bypass steam and the steam injection into the steam turbine, and after the steam switching. Is the rising rate CV of the combined power generation output.
[0014]
In the conventional control, at the start of operation, these operation parameters take a value specified in advance based on the process amount of the plant when the corresponding device is operated, or use a function value based on the process amount. Thus, the control value is determined. These values and functions were determined at the time of plant planning and test operation.
[0015]
In order to control the thermal stress generated in the steam turbine, first, the output GO of the first load holding described above and the load holding are determined by the temperature of the metal in the portion where the thermal stress of the steam turbine is controlled at the start of the plant. The pressure GSP of the medium vapor is determined. Next, the switching time CT for the above-mentioned bypass steam and steam injection into the steam turbine, and the rate of increase in the power generation output after the steam switching, based on the temperature of the steam and the temperature of the metal of the steam turbine when the steam is switched. The SV is determined.
[0016]
At the time of plant planning, as shown in FIG. 16, a change in steam temperature when the power output of the gas turbine is increased is assumed based on the relationship between the fuel flow rate of the gas turbine and the steam temperature with respect to the power output of the gas turbine. Estimate and calculate the change in temperature distribution of the metal part and the thermal stress generated by it. The value of the operating parameter is determined so that the thermal stress by this estimation calculation satisfies the limit value.
[0017]
The values of these operating parameters may be adjusted to reflect the actual operating conditions at the time of test operation, but should be adjusted to reflect the actual operating conditions of the subsequent plants each time startup operation is performed. There is no.
[0018]
Therefore, in consideration of changes in the characteristics of the plant and equipment, changes in operating conditions, and the like, an operation method with an allowance, that is, an operation method with an allowance for generation of stress is set. Therefore, it takes time for the start-up operation.
[0019]
Regarding the metal part of the waste heat recovery boiler, at the time of plant planning, it is evaluated whether the generation of thermal stress can be tolerated against the change in steam temperature determined by the steam turbine.
[0020]
FIG. 18 is a schematic configuration diagram of a control system of a conventional combined cycle power plant.
[0021]
The host controller 12 creates operation parameters and the like for start-up as target values for the operation, and sends them to the gas turbine lower controller 12a and the steam turbine lower controller 12b. The lower control devices 12a and 12b generate operation signals for operating the on-site equipment, and send the operation signals to the power generation plant via the gas turbine input / output device 13a and the steam turbine input / output device 13b. The gas turbine process data 17a from the power plant is sent to the lower control device 12a and the higher control device 12 via the input / output device 13a. Similarly, the steam turbine process data 17b from the power plant is sent to the lower control device 12b and the upper control device 12 via the input / output device 13b.
[0022]
[Non-patent document 1]
Tsutomu Michigami "Power Generation / Transformation" The Institute of Electrical Engineers of Japan, July 1, 1997, p185-188
[Non-patent document 2]
Chiba Yuki "Thermal Power Station" Denki Shoin, July 10, 1994, p469-503
[0023]
[Problems to be solved by the invention]
As described above, in the conventional method, the value of the operating parameter at the time of startup is determined at the time of design or test operation, and does not reflect the actual operating state of the plant thereafter. That is, an operation method with a margin is used in consideration of changes in characteristics of a plant and equipment, changes in operation conditions, and the like. In particular, regarding the generation of stress, there is a problem that it takes too much time to start up the plant because an operation method with a margin is used.
[0024]
In order to solve such a problem, the present invention determines an operating parameter that can be started in a short time without generating excessive thermal stress in the plant, and operates the combined cycle power plant that operates based on the operating parameter. It is intended to provide.
[0025]
[Means for Solving the Problems]
The method for operating a combined cycle power plant according to claim 1 is a method for operating a combined cycle power plant including a gas turbine facility and a steam turbine facility from start-up to a predetermined power generation output. Speed GV of the gas turbine, the output GO of the first load holding of the gas turbine, the pressure GSP of the steam while the load of the gas turbine is held, the switching time CT for bypass steam and steam injection into the steam turbine, and the combination after the switching of the steam A step of setting an operation parameter including a rate of increase CV of the power generation output and, as a constraint, a prediction calculation based on a dynamic characteristic simulation of a combined cycle power plant, and the calculation is performed on a shaft of the steam turbine, which is predicted to be generated by the operation. Thermal stress value SS, generated in high temperature part of steam generator Of a plant monitoring parameter consisting of a thermal stress value HS and a nitrogen oxide emission value NO in a flue gas, and a prediction calculation based on a dynamic characteristic simulation of a combined cycle power plant as an evaluation index. Setting an arrival time PT, which is predicted to be caused by the driving operation, at which the power generation output reaches the specified power generation output; and, based on the operation variables and the constraints, optimize the optimization problem to minimize the evaluation index before starting the operation. Solving and determining an operating parameter that is an operation variable.
[0026]
According to the operating method of the combined cycle power plant according to the first aspect, an operating parameter that satisfies the constraint conditions and minimizes the time to reach the specified generated power is obtained, and the combined cycle power plant is started according to the operating parameter. Can be operated.
[0027]
A method for operating a combined cycle power plant according to claim 2 is the method for operating a combined cycle power plant according to claim 1, wherein the increase rate CV of the combined power generation output after the switching of steam is used as an operation variable. The power generation output of the gas turbine is replaced by a rise rate GV.
[0028]
According to the operation method of the combined cycle power plant according to the second aspect, similarly to the operation method of the combined cycle power plant according to the first aspect, the constraint condition is satisfied and the time required to reach the specified generated power is minimized. The operating parameters are determined, and the combined cycle power plant can be operated at the time of startup according to the operating parameters.
[0029]
A method for operating a combined cycle power plant according to claim 3 is the method for operating a combined cycle power plant according to claim 1 or 2, wherein the operating variable is an increase rate CV of the combined power generation output after switching of steam or a gas turbine output. Is divided into a plurality of time units, and a rise rate CVn or GVn is taken for each section.
[0030]
According to the operation method of the combined cycle power plant according to the third aspect, the degree of freedom of the optimization calculation is increased by partitioning the rate of increase of the combined power generation output or the gas turbine output after the switching of the steam. As compared with the operation methods of the first and second combined cycle power plants, it is possible to obtain the operation parameters for shortening the start-up time.
[0031]
A method of operating a combined cycle power plant according to claim 4 is the method of operating a combined power plant according to any one of claims 1 to 3, wherein the time PT to reach the specified power generation output, which is an evaluation index, is set. , Replaced by the constraint conditions, the value of the thermal stress SS generated on the shaft of the steam turbine, the value of the thermal stress HS generated in the high temperature part of the steam generator, the amount of nitrogen oxides in the flue gas Is replaced with an evaluation index, and the other conditions are left as they are as constraints.
[0032]
According to the operation method of the combined cycle power plant according to claim 4, of the operation methods that reach the target output by the designated time, generation of thermal stress SS in the shaft of the steam turbine, It is possible to determine and operate an operation method that minimizes the generation of thermal stress HS or the generation of nitrogen oxide NO in flue gas.
[0033]
A method of operating a combined cycle power plant according to claim 5 is the method of operating a combined power plant according to any one of claims 1 to 5, wherein the process data of the power plant is used as an initial value for a dynamic characteristic simulation calculation. Is used.
[0034]
According to the operation method of the combined cycle power plant according to claim 5, by setting the value of the initial state by using the process data actually observed from the plant performing the dynamic characteristic simulation, the future value of the actual plant is set. The trend can be estimated more accurately. Therefore, a more accurate value can be obtained for the operation parameter that is the result of the optimization calculation using the estimated value.
[0035]
The method of operating a combined cycle power plant according to claim 6 is the method of operating a combined cycle power plant according to claim 5, wherein the optimization calculation is repeated every predetermined time period in accordance with the progress of the operation of the power plant, and the operation parameters are changed. It is characterized in that it is updated sequentially.
[0036]
According to the operation method of the combined cycle power plant according to the sixth aspect, the actual operation state of the plant is taken in, the prediction calculation and the calculation of the operation parameters are performed, and these are sequentially and repeatedly performed according to the progress of the operation. It is possible to calculate operation parameters reflecting the operation state.
[0037]
The operation method of the combined cycle power plant according to the invention of claim 7 is a method of operating a combined cycle power plant including gas turbine equipment and steam turbine equipment from start-up to a predetermined combined power generation output of steam to the steam turbine. In performing the operation after the ventilation, a steam temperature change pattern for monitoring the thermal stress of the steam turbine is set as an operation variable at a constant temperature increase rate CT over the entire time of the prediction calculation or at a predetermined time interval. And setting the temperature rise rate Tn as one of the steam temperature change patterns, and as a constraint, thermal stress generated on the shaft of the steam turbine in the future calculated by dynamic simulation from the steam temperature change pattern The maximum MSS of the predicted value of Setting the time PT to reach a specified steam temperature as an evaluation index, and solving the optimization problem to minimize the evaluation index based on the manipulated variables and the constraint conditions. Determining a temperature pattern and determining an operation parameter of the power generation output from the obtained steam temperature pattern.
[0038]
According to the operation method of the combined cycle power plant according to the seventh aspect, by partially determining the operation parameters by the optimization calculation, the scale of the equipment to be calculated can be reduced.
[0039]
The operation method of the combined cycle power plant according to claim 8 is characterized in that, in the operation from the start of the combined cycle power plant including the gas turbine equipment and the steam turbine equipment to a predetermined combined power generation output, after the steam is passed to the steam turbine. In performing the operation, a steam temperature change pattern for monitoring the thermal stress of the steam turbine is used as an operation variable, a constant temperature rise rate CT over the entire time of the prediction calculation, or a temperature set at predetermined time intervals. Setting one of the steam temperature change patterns of the rate of increase PVn, and setting, as constraints, a time PT at which the designated steam temperature is reached, and that the change in steam temperature is non-decreasing; As an evaluation index, it is calculated by dynamic simulation from the change pattern of steam temperature. Setting the maximum value MSS of the predicted value of the thermal stress generated in the shaft of the steam turbine, and solving an optimization problem for minimizing the evaluation index based on the manipulated variables and the constraints to determine the steam temperature. Determining a pattern and determining an operation parameter of the power generation output from the obtained steam temperature pattern.
[0040]
According to the operation method of the combined cycle power plant according to the eighth aspect, as in the preceding paragraph, by partially determining the operation parameters by the optimization calculation, the scale of the equipment to be calculated can be reduced.
[0041]
The method of operating a combined cycle power plant according to claim 9 is included in the optimization calculation when performing the optimization calculation for solving the optimization problem in the operation method of the combined power plant according to claim 7 or 8. The initial value for the dynamic characteristic simulation calculation uses the process data of the power plant at the calculation time, and the optimization calculation is repeated every predetermined time period according to the progress of the operation of the power plant, and the operating parameters are sequentially updated. It is characterized by the following.
[0042]
According to the operating method of the combined cycle power plant according to the ninth aspect, similarly to the seventh and eighth aspects, by partially determining the operating parameters by the optimization calculation, it is possible to reduce the scale of the equipment to be calculated. . Further, the effects of modeling errors, plant parameter errors, and unknown inputs in the optimization calculation can be reduced.
[0043]
The method for operating a combined cycle power plant according to claim 10 is a method for operating a combined cycle power plant including a gas turbine facility and a steam turbine facility from start-up to a predetermined combined power generation output. The operation before ventilation of the steam turbine uses the method according to any one of claims 1 to 6, and the operation after ventilation of steam to the steam turbine uses the method according to any one of claims 7 to 9. A method for operating a combined power plant, comprising:
[0044]
According to the operation method of the combined cycle power plant according to claim 10, the operation parameters by the optimization calculation can be used for the determination of the operation parameters before the steam turbine ventilation. An operation can be performed in consideration of the thermal stress of a certain steam turbine.
[0045]
An operation method for a combined cycle power plant according to claim 11 is the method for operating a combined power plant according to any one of claims 7 to 10, wherein the change occurs in the steam turbine shaft in the future from the change pattern of the steam temperature. When calculating the predicted value SS of the thermal stress, a prediction calculation formula obtained from a linear difference equation is used instead of the dynamic simulation.
[0046]
According to the operation method of the combined cycle power plant according to the eleventh aspect, it is possible to obtain an optimal operation parameter in a short time without performing a dynamic characteristic simulation with a large amount of calculation. It can be used for large-scale control devices.
[0047]
A method of operating a combined cycle power plant according to claim 12 is the method of operating a combined power plant according to any one of claims 7 to 10, wherein heat generated on a shaft of a steam turbine in the future from a change pattern of steam temperature. When calculating the predicted value SS of the stress, instead of using the dynamic simulation, a prediction calculation formula obtained from a linear difference equation is used, and a linear maximum value / minimum value problem is used for the optimization calculation. Features.
[0048]
According to the operation method of the combined cycle power plant according to the twelfth aspect, the calculation load can be reduced by shortening the control cycle and performing the optimization calculation. Therefore, even in a control system having a low calculation capability, such optimization calculation can be easily realized.
[0049]
Further, by shortening the control cycle for performing the sequential optimization and frequently correcting the prediction error, it is possible to reduce the effects of modeling errors, plant parameter errors, and unknown inputs.
[0050]
According to a thirteenth aspect of the present invention, there is provided the combined cycle power plant operating method according to any one of the seventh to tenth aspects, wherein the combined cycle power generation plant is generated in a future steam turbine shaft from a steam temperature change pattern. When calculating the predicted value SS of the thermal stress, instead of the dynamic simulation, a prediction calculation formula obtained from a linear difference equation is used, and the optimization calculation uses a linear programming method. .
[0051]
According to the operation method of the combined cycle power plant according to the thirteenth aspect, the portion using the maximum value / minimum value problem in the preceding item can be replaced with a linear programming method. By using the linear programming, an optimal solution can be obtained with a finite number of calculations, so that the calculation time can be reduced.
[0052]
The method for operating a combined cycle power plant according to claim 14 is the method for operating a combined power plant according to any one of claims 7 to 13, wherein a steam temperature pattern obtained by an optimization calculation is set as a target value, Feedforward control is performed on the target value so that the target value and the measured value of the steam temperature measured for the management of thermal stress match, and the power generation output obtained in advance from the relationship between the steam temperature and the power generation output. A target value of a power generation output is set by adding a signal and the feedforward control signal.
[0053]
According to the operation method of the combined cycle power plant according to the fourteenth aspect, by performing such feedforward control, it is possible to accurately control the target value of the power generation output obtained from the target steam temperature pattern.
[0054]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0055]
(First Embodiment)
FIG. 1 shows a configuration example of a plant control system having an operation parameter calculation system for determining an operation parameter by an optimization calculation including a prediction calculation based on a dynamic characteristic simulation according to an operation method of a combined cycle power plant according to the present invention. It is. As shown in FIG. 1, an operation parameter 15 obtained by the optimization calculation is given as a control target of the lower-level control devices 12a and 12b, and the lower-level control devices 12a and 12b generate an operation signal of the device according to the target value. I do.
[0056]
FIG. 2 is a flowchart showing the flow of the above-described operation parameter calculation system. The optimization calculation in the operating parameter calculation system of FIG. 2 includes, as operating variables, a gas turbine acceleration rate GV, an output GO of the first load holding of the gas turbine, a pressure GSP of steam while the load of the gas turbine is held, and bypass steam. A switching time CT for injecting steam into the steam turbine, an operating parameter at the time of operating operation including a rising rate CV of the combined power generation output after the switching of steam is set, and as a constraint, a dynamic characteristic simulation of the combined cycle power plant is performed. The value of the thermal stress SS generated in the shaft of the steam turbine, the value of the thermal stress HS generated in the high temperature portion of the steam generator, and the nitrogen oxides in the flue gas, which are predicted to be generated by the operation based on the prediction calculation Monitoring parameters of the plant consisting of the NO value of the emissions of A time PT to reach a specified power generation output, which is predicted to be caused by a driving operation by a prediction calculation by a dynamic characteristic simulation of a cycle power plant, is set, an evaluation index is used as an evaluation function, and based on the operation variables and constraints, Compute the manipulated variables that minimize the cost function.
[0057]
By solving the optimization problem in this way, it is possible to obtain an operation variable that satisfies the constraint conditions and minimizes the time required to reach the specified power generation output, that is, an operation parameter.
[0058]
As an optimization calculation method, for example, a nonlinear optimization calculation method with constraints such as SQP (sequential quadratic programming) may be used.
[0059]
Details of SQP are described in, for example, G. L. Nemhauser, A .; H. G. FIG. Rinnooy Kan, M .; J. Tood, Optimization, North-Holland (1989).
[0060]
According to the configuration and operation of the present embodiment, an operation parameter that satisfies the constraint conditions and minimizes the time required to reach the specified combined power generation output is obtained, and the combined cycle power plant is operated at startup according to the operation parameter. Can be.
[0061]
(Second embodiment)
The second embodiment according to the operation method of the combined cycle power plant of the present invention is characterized in that, in the first embodiment described above, the increase rate CV of the combined power generation output after the switching of steam, which is given as an operation variable, This is replaced with the rate of increase GV of the power output by the gas turbine.
[0062]
Since the rise rate GV of the power generation output by the gas turbine has a first-order correlation with the rise rate CV of the combined power generation output after switching of steam, the former can substitute for the latter.
[0063]
Therefore, also in the second embodiment, an operation parameter that satisfies the constraints and minimizes the time required to reach the specified combined power generation output is determined, and the combined cycle power plant is operated according to this operation parameter at the time of startup. Can be.
[0064]
(Third embodiment)
The third embodiment according to the operation method of the combined cycle power plant according to the present invention is the same as the first embodiment or the second embodiment, except that the combined power generation output is increased after the switching of the steam set as the operation variable. The increase GV of the gas turbine power generation output after the CV or steam switching is divided into a plurality of sections from a single increase pattern as shown in FIG. 3, and the output change for each section is used as an operation variable. is there.
[0065]
By increasing the number of operation variables for optimization calculation in this way, the degree of freedom of optimization calculation is increased, and the fluctuation of the value of optimization can be further reduced.
[0066]
FIGS. 4A and 4B show the effect of performing the optimal calculation by segmenting the rise in the output of the gas turbine.
[0067]
FIG. 4A shows a case where the output rise of the gas turbine is fixed for comparison. The upper diagram of FIG. 4A shows the change of the output of the gas turbine and the steam turbine and the output of the combined power generation in that case, and the lower diagram shows the change of the thermal stress in that case. is there. Since the change in the thermal stress has a steep peak, it is necessary to suppress the output rise rate so as not to exceed the limit value of the thermal stress. By suppressing the output rise rate, it takes time to start.
[0068]
On the other hand, FIG. 4B shows a case in which the increase in the output of the gas turbine is divided and the optimal calculation is performed. The upper diagram of FIG. 4 (b) shows the change of the output of the segmented gas turbine and the steam turbine and the output of the combined power generation in that case, and the lower diagram shows the change of the thermal stress in that case. It is shown. The change in the thermal stress increases the degree of freedom in the optimization calculation, so that even if the peak of the thermal stress rises quickly, it can be kept below the limit value of the thermal stress. Therefore, the output can be increased in a short time.
[0069]
That is, when the optimal calculation is performed by partitioning the output rise of the gas turbine, the plant can be started in a shorter time than when the optimal calculation is performed with the output increase constant under the same operating constraints.
[0070]
Therefore, it is possible to obtain operating parameters that can shorten the start-up time as compared with the first and second embodiments.
[0071]
(Fourth embodiment)
The fourth embodiment according to the combined cycle power plant operation method according to the present invention is different from the first, second and third embodiments in that the evaluation index “time to reach the specified power generation output” is used. Is changed to the constraint condition, and the constraint condition is the value of the thermal stress SS generated in the shaft of the steam turbine, the value of the thermal stress HS generated in the high temperature part of the steam generator, and the value of the nitrogen oxide in the flue gas. One of the values of the emission amount NO is selected as an evaluation index, and the rest is used as a constraint as it is to solve an optimal problem that minimizes the evaluation index and determine an operation parameter that is an operation variable.
[0072]
As a method of operating a power plant, there is a case where a power plant is started and operated so as to generate target power at a certain designated time.
[0073]
In order to cope with this, as in the present embodiment, it is assumed that starting at a given time is a constraint condition, any one of the above conditions selected as an evaluation index is minimized, and an item not selected as an evaluation index is If the operating parameters for starting are determined by using the operating conditions as the constraint conditions, the power plant can be started by using those operating parameters, and the target power can be generated at the designated time.
[0074]
(Fifth embodiment)
The sixth embodiment according to the method for operating a combined cycle power plant according to the present invention is the same as the first to fifth embodiments, except that the initial value of the dynamic characteristic simulation calculation for predicting the future operation state of the power plant is obtained. In this method, the optimal problem for minimizing the evaluation index is solved before starting the operation using the process data of the power plant, and the operation parameters are determined.
[0075]
FIG. 5 is a flowchart for determining operation parameters by optimization calculation in the operation parameter calculation system configured as described above. This flowchart shows a calculation flow for predicting a future operation state of the plant by incorporating available data from the plant into initial values of the dynamic characteristic simulation.
[0076]
FIG. 6 shows a configuration diagram of a control system having the above-described operation parameter calculation system. As shown in the figure, process data 17a and 17b are input from input / output devices 13a and 13b to an operation parameter calculation system 18 and the obtained operation parameters 15 are given as target values of the lower control devices 12a and 12b.
[0077]
The optimization performed in the first to fifth embodiments of the present invention is based on the estimated value of the future operation state of the power plant calculated by the dynamic characteristic simulation. For this reason, the prediction accuracy by the dynamic characteristic simulation affects the calculation result of the optimization, and affects the operation accuracy of the plant.
[0078]
The dynamic characteristic simulation performed in the first to fifth embodiments of the present invention is a nonlinear ordinary differential equation system and can be expressed as follows.

Here, x (t) is a state vector, u (t) is an input vector, y (t) is an output vector, and t is time. Further, F and G can be obtained by solving equations representing the dynamic characteristics of the plant by solving an initial value problem using the initial state x (t0) at the initial time t0 as an initial value.
[0079]
In the solution of such an ODE, the subsequent result is affected by the value of the initial state x (t0). Therefore, even if the method of describing the equation is correctly performed, if the value of the initial state x (t0) is different, the future operating state of the plant according to the dynamic simulation will be different from the actual operating state of the plant. Become something.
[0080]
On the other hand, the value in the initial state can be determined from the amount actually observed in the plant, that is, from the process data. As a result, the accuracy of the result of the dynamic simulation is improved, and the operating parameters of the plant obtained by solving the optimal problem using the result are also accurate values.
[0081]
(Sixth embodiment)
The sixth embodiment according to the combined cycle power plant operating method according to the present invention performs the optimization calculation in the first to fifth embodiments according to the progress of the operation of the power plant as shown in FIG. The operation parameters are repeatedly updated at predetermined time intervals to sequentially update the operation parameters. At this time, as for the initial value for the dynamic characteristic simulation included in the optimization calculation, the process data of the power plant at the calculation time is used for the measurement data of the power plant that can be used.
[0082]
As described in the fifth embodiment, in the present invention, the accuracy of the prediction calculation of the future operating state of the power plant affects the operation parameters obtained by the optimization calculation including the prediction calculation.
[0083]
In the prediction calculation based on dynamic characteristics simulation of the future behavior of the power plant, there is always an error in the model formula for calculation, data used in the calculation, initial values, external input variables, etc. Occurs.
[0084]
Further, in the dynamic characteristic simulation formula, an error occurs in the future predicted value due to factors such as unmodeled characteristics and external influences not included in the model formula. The errors due to these factors accumulate and increase as the prediction period becomes longer.
[0085]
The present embodiment is to prevent the accumulation of the prediction error by taking in the actual operation state of the plant, performing the prediction calculation and the calculation of the operation parameters, and repeatedly performing the calculation in accordance with the progress of the operation. As a result, it is possible to calculate an operation parameter reflecting the operation state of the actual machine.
[0086]
FIG. 8 shows a flow when the operation parameters are determined. FIG. 11A shows a case of normal optimization calculation, and FIG. 10B shows a case of sequential optimization calculation in the present embodiment. In the present embodiment, it can be interpreted that performing the sequential calculation has a feedback effect.
Originally, this feedback has the effect of reducing the sensitivity to modeling errors of the controlled object and to the effects of external inputs other than the manipulated variables.
[0087]
In short, the actual operating condition of the plant is taken in, the prediction calculation and the calculation of the operating parameters are performed, and these are sequentially repeated according to the progress of the operation. can do.
[0088]
(Seventh embodiment)
The seventh embodiment according to the method for operating the combined cycle power plant according to the present invention is different from the first embodiment to the sixth embodiment in that the operation parameters are determined in accordance with the operation after the aeration of steam to the steam turbine. This is related to how to determine the operating parameters for.
[0089]
In the optimization calculation of the present embodiment, a steam temperature change pattern for monitoring the thermal stress of the steam turbine is given as an operation variable, and as a constraint, a steam temperature change generated in the future steam turbine axis calculated from the steam temperature change pattern is used. The maximum value MSS of the predicted value of the thermal stress to be performed and the change in the steam temperature are set to be non-decreasing, the time PT to reach the specified steam temperature is set as the evaluation index, and the evaluation index is minimized. Then, a steam temperature pattern is obtained by solving the optimal problem to be performed, and the operating parameters of the combined power generation output are determined from the obtained steam temperature pattern. Here, the method of giving the temperature change pattern, which is an operation variable, is any of the constant temperature rise rate TV over the entire time of the prediction calculation, the temperature T or the temperature rise rate TV set at predetermined time intervals. .
[0090]
As for the operation method up to the ventilation of the steam turbine, the method for determining the operation parameters of the first to sixth embodiments is used.
[0091]
When the operation parameters are determined by the optimization calculation, a large amount of calculation is required if the number of operation variables is large or the prediction period is long. Since the optimization calculation is partially performed, the amount of calculation can be suppressed, and the scale of a computer that executes the calculation can be reduced.
[0092]
(Eighth embodiment)
The ninth embodiment according to the combined cycle power plant operation method according to the present invention relates to a combined cycle power plant including a gas turbine facility and a steam turbine facility. When the operation after the steam aeration is performed, a steam temperature change pattern for monitoring the thermal stress of the steam turbine is given as an operation variable, and a time PT at which the steam temperature reaches the designated steam temperature and a change in the steam temperature are set as constraints. Sets the condition of non-reduction, sets the maximum value MSS of the predicted value of the thermal stress generated in the future axis of the steam turbine calculated from the change pattern of the steam temperature as the evaluation index, and minimizes the evaluation index Solves the optimal problem to determine the steam temperature pattern, and from the obtained steam temperature pattern, the operating parameters of the power generation output It is those determined.
[0093]
Here, the method of giving the temperature change pattern, which is an operation variable, is any one of a constant temperature rise rate TV over the entire time of the prediction calculation, a temperature set at predetermined time intervals, and a temperature rise rate PV.
[0094]
As for the operation method up to the ventilation of the steam turbine, the method for determining the operation parameters in the first to sixth embodiments is used.
[0095]
In determining the operation parameters by the optimization calculation, a large amount of calculation is required if the number of operation variables is large or the prediction period is long. As in the seventh embodiment, by partially optimizing the calculation, the amount of calculation can be suppressed, and the scale of the computer that executes the calculation can be reduced.
[0096]
(Ninth embodiment)
The ninth embodiment according to the combined cycle power plant operating method according to the present invention, as described in the seventh and eighth embodiments, performs the sixth embodiment when the optimization calculation is partially performed. As described in the embodiment, the operating parameters are repeatedly updated at predetermined time intervals in accordance with the progress of the operation of the power plant, and the operating parameters are sequentially updated.
[0097]
As for the initial value for the dynamic characteristic simulation calculation included in the optimization calculation, the process data of the power plant at the calculation time is used for those for which the measurement data of the power plant can be used, as in the sixth embodiment. .
[0098]
With such a configuration, as described in the seventh and eighth embodiments, when the optimization calculation is partially performed, the modeling error and the plant parameter described in the sixth embodiment are reduced. And the effects of unknown inputs can be reduced.
[0099]
By partially determining the operation parameters by the optimization calculation, it is possible to reduce the scale of a device that executes the calculation.
[0100]
Further, the effects of modeling errors, plant parameter errors, and unknown inputs in the optimization calculation can be reduced.
[0101]
(Tenth embodiment)
A tenth embodiment according to a method for operating a combined cycle power plant according to the present invention relates to operating the combined cycle power plant from startup to ventilation to a steam turbine, according to any one of claims 1 to 6. A method is used to determine operating parameters for the operation after the aeration of steam to the steam turbine by using the method according to any one of claims 7 to 9.
[0102]
With this configuration, it is possible to use the operation parameters obtained by normal optimization calculation for the operation up to the steam turbine ventilation, and realize the operation in consideration of the thermal stress of the steam turbine, which is a major constraint after the steam turbine ventilation. be able to.
[0103]
By dividing the determination of the operation parameters by the optimization calculation into two, the scale of the device that executes the calculation can be reduced, and the startup in a short time in consideration of the constraint condition can be realized.
[0104]
(Eleventh embodiment)
An eleventh embodiment according to the method of operating the combined cycle power plant according to the present invention is directed to the seventh embodiment to the tenth embodiment, in which the predicted value of the thermal stress generated on the shaft of the steam turbine in the future from the change pattern of the steam temperature. Is calculated, the operating parameters are determined using a prediction calculation formula obtained from the linear difference equation used for monitoring.
[0105]
In general, since the shaft of a steam turbine is made of a large metal, a temperature distribution occurs inside the shaft due to a rise in the temperature of the shaft surface due to the injection of steam, thereby generating thermal stress.
The thermal stress prediction formula can be obtained as follows.
[0106]
The thermal stress of the shaft of the steam turbine is generated by the temperature distribution of the shaft. As shown in FIG. 9, since the shaft is regarded as a disc-shaped metal whose surface is exposed to steam, the temperature distribution generated on the shaft is based on heat transfer due to heat exchange with the surface steam temperature as a boundary condition. It can be represented by the heat conduction equation. The heat conduction equation for obtaining the temperature distribution in the disk is represented by the following polar coordinate expression, where a is a coefficient determined from the property and shape of the metal, where T is the temperature at time t and location r is T (t, r). It becomes a PDE.

When the volume average temperature is expressed as Ta (t),
Ta (t) = 2 / (r2−r1) ∫T (t, r) · r · dr (2)
It becomes. Here, r2 is the square of the outer diameter of the shaft, r1 is the square of the inner diameter of the shaft, and integration is performed from the inner diameter to the outer diameter.
[0107]
The thermal stress is a value proportional to the difference between the volume average temperature Ta (t) and the temperature T (t, r) on the disk.
[0108]
In the operation of the power plant, a difference equation obtained by differentiating these equations with respect to time and place is used. Since the equation (2) is a linear equation, the equation for the differentiated thermal stress is also a linear difference equation as shown below.
[0109]
In the following, X is used instead of T as a temperature variable. The temperature distribution vector at the discrete sample time k is represented as X (k). The transposed vector of the temperature distribution vector X (k) is represented by [X1 (k), X2 (k),..., Xn (k)] ^T It expresses.
[0110]
Assuming that the thermal stress is y (k) and the steam temperature is u (k), the thermal stress becomes the following difference equation:
X (k + 1) = A.X (k) + B.u (k) (3)
y (k) = C ・ X (k) (4)
Here, the constant matrices A, B, and C are constant matrices determined by differentiating the heat conduction equation and the volume average temperature equation and determined by the number of divisions, the shape of the shaft, the material, and the like.
Thus, the transposed vector of the estimated value vector Y of the thermal stress over the m period, where k is the current time,
[Y (k + 1), y (k + 2), ..., y (k + m)] ^T ,
The transposed vector of the steam temperature vector U of the manipulated variable over p future periods is
[U (k + 1), u (k + 2), ..., u (k + p)] ^T ,
Then, the estimated value vector Y of the thermal stress is obtained by using the manipulated variable vector U,
Y = FX × (k) + GU (5)
It can be expressed as. Thus, the estimated value Y of the thermal stress can be determined by giving the steam temperature U.
[0111]
Conversely, if the estimated value Y of the thermal stress is given, the value of the future steam temperature U can be determined. For example, the transpose of matrix G is given by G ^T From equation (5), using the least squares method,
U = HG ^T ・ (YF ・ X (k)) (6)
Can be sought. Here, the matrix H is G ^T The inverse matrix of G
[0112]
The conversion from the steam temperature pattern to the power generation output pattern is performed as shown in FIG.
It can be determined from the relationship between the previously determined steam temperature and the power generation output.
[0113]
By using the thermal stress prediction formula as described above, it is possible to obtain the optimum operation parameters without performing a dynamic characteristic simulation requiring a large amount of calculation.
[0114]
That is, the optimum operation parameters can be obtained in a short time, and the calculation performance and the storage capacity can be used for a large-scale control device.
[0115]
(Twelfth embodiment)
The twelfth embodiment according to the combined cycle power plant operating method according to the present invention, as determined in the eleventh embodiment, includes the seventh to tenth embodiments in which the steam temperature change pattern When calculating the predicted value SS of the thermal stress generated on the shaft of the steam turbine, when determining the operating parameter using the prediction calculation formula obtained from the linear difference equation used for monitoring the thermal stress and using the result First, a linear maximum / minimum problem is used.
[0116]
There are two methods for utilizing the linear maximum / minimum problem, and each method will be described with reference to FIGS.
[0117]
First, the first method is a method of calculating a temperature pattern in order to start up in the shortest time with thermal stress as a constraint. FIG. 11 is a flowchart of the calculation, and the procedure is as follows.
(1) Set the initial value of the number N of operations of the steam temperature.
(2) Set an initial value X (0) of the temperature distribution of the axis.
(3) A steam temperature change pattern U for monitoring the thermal stress of the steam turbine is given as an operation variable, and as a constraint, a time M at which the steam temperature reaches the designated steam temperature and a condition that the change in the steam temperature is not reduced are set. The maximum value / minimum value problem is solved by giving, as an evaluation index, the maximum value J of the predicted value Y of the thermal stress generated in the future axis of the steam turbine calculated from the change pattern of the steam temperature.
(4) If the thermal stress is equal to or less than the limit value, reduce the number of operations and return to (1).
If it is more than the limit value, go to the next.
(5) If none of the calculations satisfy the limit value or less in the calculation of (3), the operation number N of the steam temperature is increased, and the process returns to (1). If there is one satisfying the limit value or less in the calculation of (3), the process goes to the next step.
(6) The previous result is adopted as the steam temperature operation pattern.
[0118]
Next, the second method is a method of calculating a steam temperature pattern that can be started with a minimum thermal stress, with the starting time as a constraint. FIG. 12 is a flowchart for calculating the steam temperature pattern at this time. In this case, a solution can be obtained by solving the optimization problem only once.
[0119]
As described above, the optimization calculation is configured to solve the linear maximum value / minimum value problem, and an optimum solution can be obtained at high speed. In the sequential optimization calculation described in the ninth embodiment, it is necessary to perform the optimization calculation for each determined control cycle, but in the present embodiment, it is necessary to solve the optimization problem only once. So there is no need to repeat. Therefore, such a calculation can be easily realized even in a control system having a low calculation capability.
[0120]
(Thirteenth embodiment)
The thirteenth embodiment of the combined cycle power plant operation method according to the present invention uses a linear programming method instead of the optimization calculation using the linear maximum / minimum problem of the twelfth embodiment. Things.
[0121]
The linear maximum / minimum problem and the linear programming can be interchanged with each other in a relationship as shown in FIG. Therefore, in the first method and the second method of the twelfth embodiment, the part using the maximum value / minimum value problem can be replaced by the linear programming.
[0122]
By using the linear programming, an optimal solution can be obtained with a finite number of calculations, so that the calculation time can be reduced and it is not necessary to require a high calculation capability for the control system.
[0123]
(14th embodiment)
The fourteenth embodiment according to the combined cycle power plant operating method according to the present invention relates to the seventh to thirteenth embodiments in which the temperature pattern obtained so as to provide the optimum operation and the thermal stress are actually calculated. The control is performed so that the temperature used for management matches.
[0124]
FIG. 14 shows a control algorithm according to such an embodiment. The steam temperature pattern obtained by the optimization calculation is set as a target value, and a feedforward control signal is generated so that there is no deviation as compared with a measured value of the steam temperature measured for thermal stress management. On the other hand, a power generation output signal is set based on the relationship between the previously determined steam temperature and the power generation output, and a target value of the power generation output is created by adding the power generation output signal and the previous feedforward control signal.
[0125]
When the relationship between the steam temperature and the power generation output cannot be easily obtained, the target value of the power generation output can be obtained only by the feedforward control signal.
[0126]
By configuring such a control algorithm, the steam temperature pattern obtained by the optimization calculation can be used as the target value, and the target value of the power generation output set from the relationship between the steam temperature and the power generation output obtained in advance can be corrected. it can.
[0127]
【The invention's effect】
According to the operation method of the combined power plant of the present invention, by taking into account the operation constraints of the plant, the operation parameters are obtained by an optimization calculation that minimizes the startup time, and the plant is operated according to the operation parameters. The operation of the plant can be started in a short time without generating excessive thermal stress.
[Brief description of the drawings]
FIG. 1 is a configuration diagram of a first embodiment.
FIG. 2 is a flowchart showing a flow of operation according to the first embodiment;
FIG. 3 is a diagram showing an optimal calculation model 1 based on output rise partitioning in a third embodiment.
FIG. 4 is a diagram showing an optimal calculation model 2 based on output rise partitioning in a third embodiment.
FIG. 5 is a flowchart for determining an operation parameter by an optimization calculation in the fifth embodiment.
FIG. 6 is a configuration diagram of a control system having an operation parameter calculation system in a fifth embodiment.
FIG. 7 is a diagram showing a method 1 for determining an operation parameter in the sixth embodiment.
FIGS. 8A and 8B are diagrams illustrating a method 2 for determining an operation parameter in the sixth embodiment, wherein FIG. 8A illustrates an open-loop case, and FIG. 8B illustrates a closed-loop case of the present embodiment.
FIG. 9 is a diagram showing a method for calculating thermal stress in the eleventh embodiment.
FIG. 10 is a diagram showing target value conversion in the eleventh embodiment.
FIG. 11 is a flowchart showing a flow 1 of calculating a steam temperature pattern in the twelfth embodiment.
FIG. 12 is a flowchart showing a flow 2 of calculating a steam temperature pattern in the twelfth embodiment.
FIG. 13 is a view for explaining equivalence in the thirteenth embodiment.
FIG. 14 is a diagram showing a control algorithm for generating a power generation output target value in a fourteenth embodiment.
FIG. 15 is a configuration diagram of a conventional combined cycle power plant.
FIG. 16 is a diagram showing a relationship between a fuel flow rate, a power generation output, an exhaust gas flow rate, an exhaust gas temperature, and a steam temperature in a conventional combined cycle power plant.
FIG. 17 is a diagram showing an example of a conventional combined cycle power plant operation at startup.
FIG. 18 is a configuration diagram of a control system of a conventional combined cycle power plant.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... gas turbine equipment, 1a ... compressor, 1b ... gas turbine, 2 ... waste heat recovery boiler, 3 ... steam turbine equipment, 3a ... steam turbine, 4 ... generator (gas turbine shaft), 4a ... generator (steam) Turbine shaft), 5 ... condenser, 6 ... pump,
7 ... combustor, 8 ... fuel valve, 9 ... steam valve (1), 10 ... steam valve (2), 11 ... conventional control device, 11a ... control device of the first embodiment, 11b ... fifth embodiment Control device of the form, 12: Upper control device, 12a: Lower control device for gas turbine, 12b: Lower control device for steam turbine, 13a: Input / output device for gas turbine, 13b: Input / output device for steam turbine, 14: Power generation Plant, 15: Operating parameters, 15a: Operating parameters for gas turbine, 15b: Operating parameters for steam turbine, 16: Operating signal, 16a: Operating signal for gas turbine, 16b: Operating signal for steam turbine, 17a: Process for gas turbine Data 17b Process data for steam turbine 18 Operating parameter calculation system 19 Operating data processing system 20 Unknown input Reference numeral 21: operation target value, 22: initial value data, 23a: gas turbine operating device, 23b: steam turbine operating device, 24a: gas turbine detector, 24b: steam turbine detector, 181: optimal calculation, 182 ... Prediction calculation, 191 ... Data selection, 192 ... Plant data collection system

Claims (14)

In the operation from the start of the combined cycle power plant composed of the gas turbine equipment and the steam turbine equipment to the predetermined power generation output, the operation variables include the gas turbine speed-up rate GV, the output of the gas turbine at the initial load retention. A step of setting operation parameters including GO, a pressure GSP of steam during load holding of the gas turbine, a switching time CT for injecting steam into the bypass steam and the steam turbine, and a rising rate CV of the combined power generation output after the switching of steam. When,
As constraint conditions, a prediction calculation is performed by a dynamic characteristic simulation of a combined cycle power plant, and a heat stress value SS generated in a steam turbine shaft, which is predicted to be caused by an operation operation, and a heat generated in a high temperature portion of a steam generator. Stress value HS,
Setting a monitoring parameter of the plant consisting of the value NO of the emission of nitrogen oxides in the flue gas;
Performing a prediction calculation by a dynamic characteristic simulation of the combined cycle power plant as an evaluation index, and setting a reaching time PT to reach a specified power generation output, which is predicted to be caused by the driving operation;
Solving an optimization problem that minimizes the evaluation index before starting operation based on the manipulated variables and the constraint conditions, and determining an operating parameter that is the manipulated variable. Driving method. 2. The operation of the combined power plant according to claim 1, wherein the increase rate CV of the combined power generation output after the switching of the steam is replaced with the increase rate GV of the power generation output of the gas turbine after the switching of the steam as the operation variable. Method. The operation variable, the increase rate CV of the combined power generation output after the switching of steam or the increase rate GV of the gas turbine power generation output after the switching of the steam, is divided into a plurality of time units, and the increase rate CVn for each of the divisions. Alternatively, the method of operating a combined power generation plant according to claim 1 or 2, wherein GVn is adopted. The time PT to reach the specified power generation output, which is an evaluation index, is replaced with a constraint, and the constraint is the value of the thermal stress SS generated on the shaft of the steam turbine, the heat generated in the high-temperature portion of the steam generator. 4. The method according to claim 1, wherein one of the stress value HS and the value NO of the nitrogen oxide emission amount in the flue gas is replaced with an evaluation index, and the other conditions are left as constraint conditions. An operation method of the combined power generation plant according to any one of the preceding claims. The method for operating a combined power plant according to any one of claims 1 to 5, wherein process data of the power plant is used as an initial value for the dynamic characteristic simulation calculation. The method of operating a combined power plant according to claim 5, wherein the optimization calculation is repeated at predetermined time intervals in accordance with the progress of the operation of the power plant, and the operating parameters are sequentially updated. In the operation from the start of the combined cycle power plant including the gas turbine equipment and the steam turbine equipment to the predetermined power generation output, in performing the operation after the ventilation of the steam to the steam turbine,
As an operation variable, a steam temperature change pattern for monitoring thermal stress of the steam turbine is calculated by using either a constant temperature rise rate CT over the entire time of the prediction calculation or a temperature rise rate Tn set at predetermined time intervals. And setting the steam temperature change pattern,
As the constraints, the maximum value MSS of the predicted value of the thermal stress generated in the shaft of the steam turbine in the future calculated by the dynamic simulation from the change pattern of the steam temperature and the change in the steam temperature are set to be non-decreasing. Stage to
Setting a time PT at which a designated steam temperature is reached as an evaluation index, and solving an optimization problem for minimizing the evaluation index based on the manipulated variables and the constraints to obtain a steam temperature pattern. Determining an operating parameter of the combined power generation output from the obtained steam temperature pattern. In the operation from the start of the combined cycle power plant including the gas turbine equipment and the steam turbine equipment to the predetermined power generation output, in performing the operation after the ventilation of the steam to the steam turbine,
As a manipulated variable, a steam temperature change pattern for monitoring thermal stress of a steam turbine is calculated by using either a constant temperature increase rate CT over the entire time of the prediction calculation or a temperature increase rate PVn set at predetermined time intervals. Setting a steam temperature change pattern of
Setting, as constraints, the time PT to reach the specified steam temperature, and that the change in steam temperature is non-decreasing;
Setting, as an evaluation index, a maximum value MSS of a predicted value of a thermal stress generated in a shaft of the steam turbine in the future, which is calculated by a dynamic simulation from a change pattern of the steam temperature;
Solving the optimization problem that minimizes the evaluation index based on the manipulated variables and the constraints to obtain a steam temperature pattern, and determining an operation parameter of the power generation output from the obtained steam temperature pattern. A method for operating a combined power plant, comprising: In the optimization calculation for solving the optimization problem, the initial value for the dynamic characteristic simulation calculation included in the optimization calculation uses process data of the power plant at the calculation time, and the optimization calculation includes power generation. 9. The operating method of a combined power plant according to claim 7, wherein the operating parameters are repeatedly updated at predetermined time intervals as the operation of the plant progresses. In performing the operation from the start of the combined cycle power generation plant including the gas turbine equipment and the steam turbine equipment to the predetermined combined power generation output, the operation before aeration of the steam to the steam turbine is performed according to any one of claims 1 to 6. 10. A method of operating a combined power plant, wherein the method according to claim 1 is used, and the operation after the aeration of steam to the steam turbine is performed using the method according to any one of claims 7 to 9. In calculating the predicted value SS of the thermal stress generated on the shaft of the steam turbine in the future from the change pattern of the steam temperature, a prediction calculation formula obtained from a linear difference equation is used instead of the dynamic simulation. The method for operating a combined power plant according to any one of claims 7 to 10. In calculating the predicted value of the thermal stress generated in the shaft of the steam turbine in the future from the change pattern of the steam temperature, instead of the dynamic simulation, a prediction calculation formula obtained from a linear difference equation is used. The method for operating a combined power plant according to any one of claims 7 to 10, wherein a linear maximum / minimum value problem is used for the conversion calculation. In calculating the predicted value of the thermal stress generated in the shaft of the steam turbine in the future from the change pattern of the steam temperature, instead of the dynamic simulation, a prediction calculation formula obtained from a linear difference equation is used. The method for operating a combined power generation plant according to any one of claims 7 to 10, wherein the linearization method is used for the conversion calculation. The steam temperature pattern obtained by the optimization calculation is set as a target value, and the feedforward control is performed on the target value so that the target value matches the measured value of the steam temperature measured for managing the thermal stress. And generating a target value of the power generation output by adding the power generation output signal previously obtained from the relationship between the steam temperature and the power generation output and the feedforward control signal. The operation method of the combined power plant according to the above item.

Images