A modeling method for steam turbine heat consumption curve considering coupling effect of power and extraction steam quantity

Abstract

This invention discloses a method for modeling the heat consumption curve of a steam turbine considering the coupling effect of power and steam extraction, belonging to the field of thermal power unit heat consumption performance analysis and scheduling optimization technology. Based on the principle of energy balance, this method constructs a coupling functional relationship between power generation, steam extraction for heating, and heat consumption rate, breaking through the limitations of traditional methods that rely solely on the heat consumption curve under pure condensing conditions. By analyzing historical operating data, typical operating points are adaptively identified on the power-extraction plane. Combined with clustering algorithms and distribution homogenization strategies, the model's coverage is improved. Simultaneously, prior knowledge of the steam turbine's thermodynamic mechanism is integrated to physically constrain parameters and compensate for sparse regions, ensuring the model's rationality and high accuracy under unsampled operating conditions. Verified by actual units, this method exhibits small fitting errors, good stability, and online updates, effectively supporting integrated heat and power scheduling and energy-saving operation, meeting the needs of smart power plants for refined performance modeling.

Description

Technical Field

[0001] This invention relates to the field of thermal power unit heat consumption performance analysis and scheduling optimization technology, and in particular to a method for modeling turbine heat consumption curves that considers the coupling effect of power and steam extraction. Background Technology

[0002] The heat consumption curve of a steam turbine describes the relationship between the unit's power generation and fuel consumption (or heat rate), which differs significantly between pure condensing operation (without steam extraction for heating) and extraction heating operation. Traditional dispatching and load allocation often assume that the unit's heat consumption curve only changes with the power generation load, neglecting the coupled impact of extraction steam volume on unit efficiency. This simplification may lead to inaccurate fuel consumption estimates in actual optimization calculations, thus affecting the economic operation of the unit and the effectiveness of load optimization.

[0003] Currently, there are two main approaches to handling the heat consumption characteristics of steam turbines under heating conditions: The first is a mechanism-based modeling method, which simulates the fuel consumption changes of the turbine unit under different extraction rates by constructing a thermodynamic model and performing heat balance calculations. This physical modeling method is logically sound, with clearly defined parameters, but its drawback lies in its heavy reliance on equipment design parameters and empirical formulas. The model is complex and struggles to accurately reflect the actual operating state of the unit. Due to the use of idealized assumptions and empirical corrections in the modeling, the calculated heat consumption curves often deviate from actual measurements. The second approach employs simplified empirical methods in engineering, such as using standard heat consumption curves for pure condensing conditions, along with a fixed heating penalty factor to approximate the impact of extraction on heat consumption. However, the fixed factor cannot cover all operating conditions and fails to reflect the nonlinear trend of heat rate changes under the coupling effect of power and extraction. Therefore, current heat consumption modeling or load optimization lacks a universal modeling method that can both reflect the power-extraction coupling effect and possess sufficient accuracy and adaptability, making it difficult to meet the requirements of refined optimization scheduling. Summary of the Invention

[0004] The present invention aims to at least partially solve one of the technical problems in the related art.

[0005] Therefore, the first objective of this invention is to propose a method for modeling the heat consumption curve of a steam turbine that considers the coupling effect of power and steam extraction rate.

[0006] Another objective of this invention is to provide a modeling device for turbine heat consumption curves that considers the coupling effect of power and extraction steam volume.

[0007] The third objective of this invention is to provide a computer device.

[0008] A fourth objective of this invention is to provide a non-transitory computer-readable storage medium.

[0009] To achieve the above objectives, a first aspect of the present invention proposes a method for modeling the heat consumption curve of a steam turbine considering the coupling effect of power and steam extraction rate, comprising:

[0010] S1, acquire historical operating data of the steam turbine under various operating conditions, the historical operating data including power generation, steam extraction for heating and corresponding heat consumption rate; S2, Establish a physical information neural network model; S3, the physical information neural network model is trained based on the historical operating data, wherein the loss function of the training process includes a data error term and a physical information error term, and the physical information error term is used to reflect the coupling influence characteristics between power generation and heating steam extraction. S4 inputs the real-time operating data of the steam turbine into the trained physical information neural network model to obtain the heat consumption curve of the steam turbine heat consumption rate as a function of power generation and steam extraction for heating.

[0011] In one embodiment of the present invention, training the physical information neural network model based on the historical operational data includes: The historical operating data is analyzed for distribution characteristics to identify dense and sparse regions in the parameter space formed by power generation and steam extraction for heating. Representative operating points are selected from the data-dense region, and virtual data points are generated in the sparse region based on the steam turbine thermodynamic mechanism to obtain the training dataset; The physical information neural network model is trained based on the training dataset.

[0012] In one embodiment of the present invention, after obtaining the training dataset, the method further includes: Based on the training dataset, establish the correspondence between main steam flow rate and power generation, heating steam extraction rate, and equivalent power generation under pure condensing conditions. Establish the functional relationship between the equivalent heat rate under pure condensing conditions and the equivalent power generation under pure condensing conditions.

[0013] In one embodiment of the present invention, the physical constraint error term is constructed based on the thermodynamic equilibrium relationship of the steam turbine; The thermodynamic equilibrium relationship is related to the actual power generation, actual heat rate, equivalent power generation under pure condensing conditions, equivalent heat rate under pure condensing conditions, and steam extraction for heating.

[0014] In one embodiment of the present invention, before inputting the real-time operating data of the steam turbine into the trained neural network model, the method further includes: The validity of the real-time operating data is verified, and abnormal data that exceeds the preset reasonable range is removed; The real-time running data that passes the verification is standardized to make the data format consistent with the input requirements of the neural network model.

[0015] In one embodiment of the present invention, the step of inputting the real-time operating data of the steam turbine into the trained physical information neural network model to obtain the heat consumption curve of the steam turbine heat consumption rate as a function of power generation and heating extraction steam volume further includes: The neural network model matches the corresponding model calculation parameters based on the numerical combination of power generation and steam extraction for heating in the real-time operation data. Based on the matched model calculation parameters, the heat rate corresponding to the numerical combination is output, thereby generating a continuous heat rate curve of the turbine heat rate as a function of power generation and heat extraction steam volume.

[0016] To achieve the above objectives, a second aspect of the present invention provides a device for modeling the heat consumption curve of a steam turbine that considers the coupling effect of power and steam extraction rate, comprising: The data acquisition module is used to acquire historical operating data of the steam turbine under various operating conditions. The historical operating data includes power generation, steam extraction for heating, and corresponding heat consumption rate. The model building module is used to build physical information neural network models; The model training module is used to train the physical information neural network model based on the historical operating data. The loss function of the training process includes a data error term and a physical information error term. The physical information error term is used to reflect the coupling influence characteristics between power generation and heating steam extraction. The curve determination module is used to input the real-time operating data of the steam turbine into the trained physical information neural network model to obtain the heat consumption curve of the steam turbine heat consumption rate as a function of power generation and heating extraction steam volume.

[0017] This invention provides a method and apparatus for modeling the heat consumption curve of a steam turbine that considers the coupling effect of power and steam extraction rate. This method effectively solves the problem of insufficient modeling accuracy of the heat consumption curve of a steam turbine under the influence of power and steam extraction rate coupling, improves the coverage of operating conditions and the physical consistency of the model, and achieves high-precision heat rate prediction under limited operating data conditions.

[0018] To achieve the above objectives, a third aspect of this application provides a computer device, including a processor and a memory; wherein the processor runs a program corresponding to the executable program code by reading executable program code stored in the memory, for implementing a method for modeling a turbine heat consumption curve considering the coupling effect of power and extraction steam volume as described in the first aspect embodiment.

[0019] To achieve the above objectives, the fourth aspect of this application proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a method for modeling a turbine heat consumption curve considering the coupling effect of power and extraction steam volume as described in the first aspect embodiment.

[0020] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0021] Figure 1 This is a flowchart of a method for modeling a steam turbine heat consumption curve considering the coupling effect of power and steam extraction volume according to an embodiment of the present invention; Figure 2 This is a diagram of a heat dissipation estimation model based on a physical information neural network according to an embodiment of the present invention; Figure 3 This is a diagram showing the effect of the heating steam extraction rate on the power generation heat consumption according to an embodiment of the present invention. Figure 4 This is a timing diagram of the actual output power data of (pure condensation) according to an embodiment of the present invention; Figure 5 This is a time series diagram showing the fitting effect of the (pure condensing) main steam flow rate according to an embodiment of the present invention; Figure 6 This is a fitting graph of the (pure condensation) heat consumption-power relationship according to an embodiment of the present invention; Figure 7 This is a comparison chart of the fitting effect of the (pure condensing) main steam flow rate according to an embodiment of the present invention; Figure 8 This is a timing diagram of the actual output power (steam extraction) according to an embodiment of the present invention; Figure 9 This is a time-series diagram of actual steam extraction volume according to an embodiment of the present invention; Figure 10 This is a time series diagram showing the fitting effect of the (extraction) main steam flow rate according to an embodiment of the present invention; Figure 11 This is a comparison chart of the fitting effect of the (extraction) main steam flow rate according to an embodiment of the present invention; Figure 12 This is a heat loss-steam extraction-power fitting effect diagram according to an embodiment of the present invention; Figure 13 This is a fitted heat dissipation diagram according to an embodiment of the present invention; Figure 14 This is a heat loss-steam extraction-power fitting effect diagram according to an embodiment of the present invention; Figure 15 This is a fitted heat dissipation diagram according to an embodiment of the present invention; Figure 16 This is a structural diagram of a turbine heat consumption curve modeling device that considers the coupling effect of power and steam extraction volume according to an embodiment of the present invention; Figure 17 It is a computer device according to an embodiment of the present invention. Detailed Implementation

[0022] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0024] The following describes, with reference to the accompanying drawings, a method and apparatus for modeling the heat consumption curve of a steam turbine that considers the coupling effect of power and steam extraction volume, according to an embodiment of the present invention.

[0025] Figure 1 This is a flowchart illustrating a method for modeling a steam turbine heat consumption curve considering the coupling effect of power and steam extraction rate according to an embodiment of the present invention, such as... Figure 1 As shown, it includes: S1, acquire historical operating data of the steam turbine under various operating conditions, the historical operating data including power generation, steam extraction for heating and corresponding heat consumption rate; S2, Establish a physical information neural network model; S3, the physical information neural network model is trained based on the historical operating data, wherein the loss function of the training process includes a data error term and a physical information error term, and the physical information error term is used to reflect the coupling influence characteristics between power generation and heating steam extraction. S4 inputs the real-time operating data of the steam turbine into the trained physical information neural network model to obtain the heat consumption curve of the steam turbine heat consumption rate as a function of power generation and steam extraction for heating.

[0026] This invention, based on a physical information neural network, establishes a functional relationship R=f(P,G) between the turbine heat rate (R), power generation (P), and steam extraction rate (G) by integrating physical mechanism relationships and data-driven modeling. Figure 2 As shown. Figure 3This is a schematic diagram illustrating the impact of heating steam extraction on power generation heat consumption. The model identifies parameters using actual power and steam extraction operation data, and introduces the physical constraint relationship between steam extraction and power as a supplementary loss function to further improve the accuracy of heat consumption fitting. Its core is to introduce the influence term of heating steam extraction on power on the traditional pure condensing heat consumption curve, forming a heat consumption calculation formula that couples power and steam extraction.

[0027] The main variables involved in the modeling process include: power generation P (unit: MW), steam extraction rate for heating G (unit: t / h), unit heat rate R (unit: kJ / kWh, referring to the heat consumed per kWh of electricity generated), and the corresponding power generation P0 and heat rate R0 under pure condensing conditions. Main steam flow rate g (unit: t / h) is also introduced as an intermediate variable to relate the combined effects of power generation and steam extraction on unit operating conditions.

[0028] Specifically, a physical information network (PIN) approach is used to train the heat consumption curve by inputting typical operating conditions related to power and heat consumption. The loss function component, building upon the existing accuracy, incorporates constraints related to extraction steam rate, power, and main steam flow rate into the PIN constraints, combining mechanism and data to improve heat consumption accuracy.

[0029] The loss function is The first term is the accuracy error term, and the second term is the physical information error term.

[0030] Based on this, it is necessary to fit the functional relationship between heat consumption, steam extraction rate, and power generation to serve as the physical information error term in the loss function of the physical information neural network. From prior knowledge, P... R=P0 R0-k G; where P is the actual power value, R is the actual heat consumption value, P0 is the pure condensing power value, R0 is the pure condensing heat consumption value, and G is the steam extraction rate for heating. The following is the parameter flow for solving this function.

[0031] In one embodiment of the present invention, taking Unit 1 as an example, under pure condensing conditions: The main steam flow rate is directly proportional to the output power. The original output power data is as follows: Figure 4 As shown, g=k0 P0, the fitted curve yields k0=3.304, and the fitted curve is as follows: Figure 5 As shown, the fitting effect is compared to... Figure 7 As shown, g = 3.304P0.

[0032] The next step is to fit the heat dissipation-power relationship, R0=k1 P0 3 +k2 P0 2 +k3 P0+b, the fitted curve is as follows Figure 5 As shown, R0 = -0.0002P0 3 +0.1311P0 2 -38.451P0+12547.

[0033] Under extraction steam conditions, the output power timing diagram is as follows: Figure 8 As shown, the steam extraction rate is as follows Figure 9 As shown, the relationship between the main steam flow rate, power, and extraction steam rate is fitted. g=k4 P+b1 G, the fitting result is as follows Figure 10 As shown, the fitting effect is as follows: Figure 11 As shown, we get g = 3.304P + 0.176G.

[0034] In conclusion, P R=P0 R0-k G; g=3.304P 0， g = 3.304P + 0.176G; R0=f(P,G)=-0.0002P0 3 +0.1311P0 2 -38.451P0+12547;

[0035] Solve for the problem, and you will get the following results.

[0036] Thus, the relationship between power and steam extraction rate on heat consumption is obtained.

[0037] The next step is parameter identification for k. This invention employs an adaptive typical operating condition selection method based on the distribution of actual operating data in its operating point design. This method differs from theoretical sampling (such as Latin hypercube or orthogonal experimental design). Through multidimensional statistical analysis of historical unit operating data, it identifies dense and sparse regions in the power-extraction steam volume space. The algorithm first uses cluster analysis (such as K-means or density clustering) to divide the operating data into clusters, and then, combined with the principle of uniform distribution, selects representative operating points from each power-extraction steam volume interval to ensure uniform coverage of test points throughout the entire operable domain. For sparse or uncovered data areas, prior knowledge of turbine thermodynamic mechanisms (such as extraction steam heat balance and efficiency constraints) is introduced, and virtual interpolation points are used to supplement the model training set, thereby improving the comprehensiveness of curve identification and prediction accuracy. This strategy achieves maximum information utilization under limited data conditions, providing a solid foundation for stable modeling of heat consumption curves. As shown in Table 1, the data provided by the power plant is substituted into the solution.

[0038] Table 1

[0039]

[0040] Heat loss fitting error as Figure 12 As shown, the fitted heat dissipation is as follows Figure 13 As shown.

[0041] In one embodiment of the present invention, for Unit #2: P R=P0 R0-k G; g=3.305P 0， g = 3.305P + 0.175G; R0=f(P,G)=-0.000225P0 3 +0.1943P0 2 -56.441P0+13785; Heat loss fitting error as Figure 14 As shown, the fitted heat dissipation is as follows Figure 15 As shown.

[0042] The heat consumption curve modeling method considering the influence of power-extraction coupling proposed in this invention has achieved good results in actual units. The model parameter identification process is simple and feasible, and the fitted heat consumption curve has high accuracy and physical rationality, which can provide strong support for the economic output optimization of thermal power units under heating conditions.

[0043] This invention achieves high-precision heat consumption curve identification under limited data by adaptive operating point selection and mechanism-integrated modeling, incorporating the physical constraint relationship between power and extraction steam into the loss function. The method automatically identifies representative power-extraction steam combinations based on operating data, improving operating condition coverage; it combines prior knowledge of turbine mechanisms to constrain the model, ensuring physical consistency and improving prediction accuracy. The algorithm is simple, requires few parameters, can be updated online, and can be corrected in real time for changes in random group states. Compared with traditional methods, the model calculation accuracy is significantly improved, directly serving the optimization of thermal power load and energy-saving operation, providing effective support for unit economic assessment and system peak shaving.

[0044] To achieve the above embodiments, such as Figure 16 As shown, this embodiment also provides a turbine heat consumption curve modeling device 10 that considers the coupling effect of power and steam extraction, including: The data acquisition module 100 is used to acquire historical operating data of the steam turbine under various operating conditions. The historical operating data includes power generation, steam extraction for heating and the corresponding heat consumption rate. Model building module 200 is used to build a physical information neural network model; The model training module 300 is used to train the physical information neural network model based on the historical operating data. The loss function of the training process includes a data error term and a physical information error term. The physical information error term is used to reflect the coupling influence characteristics between power generation and heating steam extraction. The curve determination module 400 is used to input the real-time operating data of the steam turbine into the trained physical information neural network model to obtain the heat consumption curve of the steam turbine heat consumption rate as a function of power generation and heating extraction steam volume.

[0045] This invention provides a turbine heat consumption curve modeling device that considers the coupling effect of power and steam extraction rate. This device effectively solves the problem of insufficient modeling accuracy of turbine heat consumption curve under the influence of power and steam extraction rate coupling, improves the coverage of operating conditions and the physical consistency of the model, and achieves high-precision heat rate prediction under limited operating data conditions.

[0046] Furthermore, the model training module 300 is also used for: The historical operating data is analyzed for distribution characteristics to identify dense and sparse regions in the parameter space formed by power generation and steam extraction for heating. Representative operating points are selected from the data-dense region, and virtual data points are generated in the sparse region based on the steam turbine thermodynamic mechanism to obtain the training dataset; The physical information neural network model is trained based on the training dataset.

[0047] To implement the methods of the above embodiments, the present invention also provides a computer device, such as... Figure 17 As shown, the computer device 600 includes a memory 601 and a processor 602; wherein, the processor 602 reads the executable program code stored in the memory 601 to run a program corresponding to the executable program code, so as to implement the various steps of the above-described method for modeling the heat consumption curve of a steam turbine considering the coupling effect of power and steam extraction.

[0048] To implement the above embodiments, this application also proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a method for modeling a steam turbine heat consumption curve considering the coupling effect of power and extraction steam volume as described in the foregoing embodiments.

[0049] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0050] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for modeling the heat consumption curve of a steam turbine considering the coupling effect of power and steam extraction rate, characterized in that, include: S1, acquire historical operating data of the steam turbine under various operating conditions, the historical operating data including power generation, steam extraction for heating and corresponding heat consumption rate; S2, Establish a physical information neural network model; S3, the physical information neural network model is trained based on the historical operating data, wherein the loss function of the training process includes a data error term and a physical information error term, and the physical information error term is used to reflect the coupling influence characteristics between power generation and heating steam extraction. S4 inputs the real-time operating data of the steam turbine into the trained physical information neural network model to obtain the heat consumption curve of the steam turbine heat consumption rate as a function of power generation and steam extraction for heating.

2. The method as described in claim 1, characterized in that, Training the physical information neural network model based on the historical operational data includes: The historical operating data is analyzed for distribution characteristics to identify dense and sparse regions in the parameter space formed by power generation and steam extraction for heating. Representative operating points are selected from the data-dense region, and virtual data points are generated in the sparse region based on the steam turbine thermodynamic mechanism to obtain the training dataset; The physical information neural network model is trained based on the training dataset.

3. The method as described in claim 2, characterized in that, After obtaining the training dataset, it also includes: Based on the training dataset, establish the correspondence between main steam flow rate and power generation, heating steam extraction rate, and equivalent power generation under pure condensing conditions. Establish the functional relationship between the equivalent heat rate under pure condensing conditions and the equivalent power generation under pure condensing conditions.

4. The method as described in claim 1, characterized in that, The physical constraint error term is constructed based on the thermodynamic equilibrium relationship of the steam turbine. The thermodynamic equilibrium relationship is related to the actual power generation, actual heat rate, equivalent power generation under pure condensing conditions, equivalent heat rate under pure condensing conditions, and steam extraction for heating.

5. The method as described in claim 1, characterized in that, Before inputting the real-time operating data of the steam turbine into the trained neural network model, the following steps are also included: The validity of the real-time operating data is verified, and abnormal data that exceeds the preset reasonable range is removed; The real-time running data that passes the verification is standardized to make the data format consistent with the input requirements of the neural network model.

6. The method as described in claim 5, characterized in that, The step of inputting real-time operating data of the steam turbine into the trained physical information neural network model to obtain the heat consumption curve of the steam turbine heat rate as a function of power generation and heating extraction steam volume also includes: The neural network model matches the corresponding model calculation parameters based on the numerical combination of power generation and steam extraction for heating in the real-time operation data. Based on the matched model calculation parameters, the heat rate corresponding to the numerical combination is output, thereby generating a continuous heat rate curve of the turbine heat rate as a function of power generation and heat extraction steam volume.

7. A device for modeling the heat consumption curve of a steam turbine considering the coupling effect of power and steam extraction rate, characterized in that, include: The data acquisition module is used to acquire historical operating data of the steam turbine under various operating conditions. The historical operating data includes power generation, steam extraction for heating, and corresponding heat consumption rate. The model building module is used to build physical information neural network models; The model training module is used to train the physical information neural network model based on the historical operating data. The loss function of the training process includes a data error term and a physical information error term. The physical information error term is used to reflect the coupling influence characteristics between power generation and heating steam extraction. The curve determination module is used to input the real-time operating data of the steam turbine into the trained physical information neural network model to obtain the heat consumption curve of the steam turbine heat consumption rate as a function of power generation and heating extraction steam volume.

8. The apparatus as claimed in claim 7, characterized in that, The model training module: The historical operating data is analyzed for distribution characteristics to identify dense and sparse regions in the parameter space formed by power generation and steam extraction for heating. Representative operating points are selected from the data-dense region, and virtual data points are generated in the sparse region based on the steam turbine thermodynamic mechanism to obtain the training dataset; The physical information neural network model is trained based on the training dataset.

9. A computer device, characterized in that, Including processor and memory; The processor reads the executable program code stored in the memory to run the program corresponding to the executable program code, so as to implement the turbine heat consumption curve modeling method considering the coupling effect of power and steam extraction as described in any one of claims 1-6.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements a method for modeling the heat consumption curve of a steam turbine that considers the coupling effect of power and extraction steam volume as described in any one of claims 1-6.

Images