Method and apparatus for estimating internal state of thermal component

1. A method of estimating an internal state of a thermal component including at least one of a turbine and a boiler, the method comprising:

a measurement step of measuring a state quantity of the thermal component, the state quantity including at least one of a temperature of a working fluid, a pressure of the working fluid, an outside-air temperature of the turbine, a metal temperature of a turbine rotor, a metal temperature of a casing, an expansion of the casing, and a clearance between the turbine rotor and the casing;

an analysis-condition creation step of creating an analysis condition which is to be used to analyze an internal state of the thermal component;

a contraction-model construction step of obtaining a subspace characterizing a linear space defined by a vector set of each of the temperature distribution and the displacement distribution of a finite element model and constructing a contraction model from the finite element model using the subspace;

a data assimilation step of obtaining a probability distribution of the temperature distribution within the thermal component and the displacement distribution within the thermal component through data assimilation calculation using the contraction model, a measured value of the state quantity in the measurement step, and the analysis condition, wherein the contraction model is calculated by dimensionally lowering the finite element model;

a post-processing step of obtaining the internal state of the thermal component from an expected value of the probability distribution obtained in the data assimilation step;

an analysis step of obtaining solution vectors of each of the temperature distribution and the displacement distribution of the finite element model,

wherein, in the data assimilation step, the probability distribution is calculated using the contraction model,

wherein, in the analysis step and the contraction-model construction step, the contraction model is constructed using a result of the analysis-condition creation step, and

wherein, in the data assimilation step, a metal-surface heat-transfer-rate condition on a metal surface of a metal in the thermal component and a model variable of the contraction model are identified as identification parameters for which the probability distribution is calculated.

2. The method according to claim 1 ,

wherein, in the data assimilation step, a probability distribution of a corrected value of the analysis condition is obtained,

wherein, in the post-processing step, the corrected value of the analysis condition is obtained, and

wherein, in the analysis step and the contraction-model construction step, the contraction model is reconstructed using the corrected value from the post-processing step.

3. The method according to claim 1 ,

wherein, in the contraction-model construction step, a heat-transfer contraction model showing heat transfer on the metal in the thermal component and a structure contraction model showing a clearance in the thermal component are used as the contraction model given by dimensionally lowering the finite element model for the internal state analysis of the thermal component.

4. The method according to claim 3 ,

wherein the finite element model is dimensionally lowered with a heat-transfer coefficient parameter remaining when the heat-transfer contraction model regarding heat transfer on the metal in the thermal component is constructed as the contraction model.

5. The method according to claim 3 ,

wherein the finite element model is dimensionally lowered with a pressure parameter remaining when the structure contraction model regarding the clearance in the thermal component is constructed as the contraction model given by dimensionally lowering the finite element model.

6. The method according to claim 1 ,

wherein a model parameter is obtained from an analysis-condition creation model used in the analysis-condition creation step, and

wherein, in the data assimilation step, the probability distribution of each of the temperature distribution and the displacement distribution in the thermal component is obtained through data assimilation calculation using the contraction model, the measured value of the measurement step, and the model parameter, wherein the contraction model is given by dimensionally lowering the finite element model for the internal state analysis of the thermal component.

7. An apparatus of estimating an internal state of a thermal component including at least one of a turbine and a boiler,

wherein the apparatus is configured to perform:

a measurement step of measuring a state quantity of the thermal component, the state quantity including at least one of a temperature of a working fluid, a pressure of the working fluid, an outside-air temperature of the turbine, a metal temperature of a turbine rotor, a metal temperature of a casing, an expansion of the casing, and a clearance between the turbine rotor and the casing;

an analysis-condition creation step of creating an analysis condition which is to be used to analyze an internal state of the thermal component;

a contraction-model construction step of obtaining a subspace characterizing a linear space defined by a vector set of each of the temperature distribution and the displacement distribution of a finite element model and constructing a contraction model from the finite element model using the subspace;

a data assimilation step of obtaining a probability distribution of the temperature distribution within the thermal component and the displacement distribution within the thermal component through data assimilation calculation using the contraction model, a measured value of the state quantity in the measurement step, and the analysis condition, wherein the contraction model is calculated by dimensionally lowering a finite element model;

a post-processing step of obtaining the internal state of the thermal component from an expected value of the probability distribution obtained in the data assimilation step; and

an analysis step of obtaining solution vectors of each of the temperature distribution and the displacement distribution of the finite element model,

wherein, in the data assimilation step, the apparatus calculates the probability distribution using the contraction model in the contraction-model construction step and a metal-surface heat-transfer-rate condition on a metal surface in the thermal component and a model variable of the contraction model are identified as identification parameters,

wherein, in the analysis step and the contraction-model construction step, the apparatus constructs the contraction model using a result of the analysis-condition creation step,

wherein the measurement step, the analysis-condition creation step, the data assimilation step, and the post-processing step are performed on a thermal component side,

wherein the analysis step and the contraction-model construction step are performed at a data management center which can manage data in a plurality of power plants,

wherein communication is performed between the thermal component and the data management center, and

wherein, in the data assimilation step, a metal-surface heat-transfer-rate condition on a metal surface in the thermal component and a model variable of the contraction model are identified as identification parameters for which the probability distribution is calculated.