Steam turbine control valve flow characteristic optimization method and system based on data analysis

Abstract

The invention discloses a method and a system for optimizing flow characteristics of a steam turbine control valve based on data analysis. The method comprises the following steps: acquiring operation data of a steam turbine set through twice tests; the method comprises the following steps: preprocessing steam turbine set operation data, and establishing an adjusting valve flow characteristic expert database; performing feature analysis on the data by using a regulating valve flow feature expert database, extracting feature independent variables, and inputting the feature independent variables into a neural network for learning; according to the predicted value of neural network learning and the true value of the feature independent variable, the neural network parameters are continuously optimized to achieve expected accuracy and stability, and an optimized neural network model is obtained; and based on the neural network model, optimizing the flow control characteristics to obtain the optimal flow characteristics of the regulating valve. The system comprises a data collection module, a data preprocessing module, a data analysis and modeling module, a model verification and adjustment module and a flow characteristic optimization module. The optimal flow characteristic of the regulating valve can be obtained.

Description

Technical Field

[0001] The invention belongs to the technical field of thermal power generation, and in particular relates to a method and system for optimizing flow characteristics of a steam turbine regulating valve based on data analysis. Background Art

[0002] Most steam turbines of thermal power units use digital electro-hydraulic control systems (DEH), which are divided into two operating modes: single valve and sequence valve, and have valve switching function. In the single valve mode, each high-pressure regulating valve maintains the same opening, and the thermal stress of the steam turbine is relatively uniform, but the throttling loss is large, especially at low loads; in the sequence valve mode, each regulating valve is opened continuously in the set order, which can reduce throttling losses and help improve the economy of the unit. With the development of large power grid interconnection technology, grid-connected units are required to have frequency regulation and peak regulation capabilities, and automatic generation control (AGC) and primary frequency regulation have gradually become indicators for power grid assessment. Whether it is AGC or primary frequency regulation, it is required that the opening of the steam turbine regulating valve has a good linear relationship with the actual power generation to ensure the control accuracy and response rate of the load. In actual production, the flow characteristic function of the regulating valve set by DEH of most thermal power units is the theoretical characteristic function calculated at the factory. Due to installation and commissioning or long-term operation of the unit, the actual flow characteristics of the regulating valve vary greatly, which is manifested in slow AGC rate change, poor primary frequency regulation capability, large load fluctuation when switching steam distribution mode, poor coordinated response capability of the unit, etc. Therefore, it is objectively necessary to optimize the flow characteristics of the steam turbine regulating valve.

[0003] In recent years, the growth of energy demand and the requirements of environmental protection have made the operation efficiency and reliability of steam turbines particularly important. The steam turbine control valve is one of the key components in the steam turbine, which can adjust the load of the steam turbine and provide a stable steam flow. In order to optimize the flow characteristics of the steam turbine control valve, analysis based on test data has become an effective method. Summary of the invention

[0004] In view of the above problems, the object of the present invention is to provide a method and system for optimizing the flow characteristics of a steam turbine regulating valve based on test data analysis.

[0005] In order to achieve the above object, the present invention adopts the following technical solution: A method for optimizing flow characteristics of a steam turbine regulating valve based on data analysis, comprising: The steam turbine unit operation data was collected through two tests; By pre-processing the steam turbine unit operation data, an expert database of valve flow characteristics is established; Use the valve flow characteristic expert database to perform data feature analysis, extract feature independent variables, and input the feature independent variables into the neural network for learning; According to the predicted value learned by the neural network and the true value of the feature independent variable, the expected accuracy and stability are achieved through continuous optimization of the neural network parameters, and the optimized neural network model is obtained; Based on the neural network model, the flow control characteristics are optimized to obtain the best flow characteristics of the regulating valve.

[0006] A further improvement of the present invention is that the steam turbine unit operation data is obtained through two test collections, including: The test was carried out in two stages. In the first test, the corresponding relationship between the total valve position command, load and the opening of each regulating valve was obtained. In the second test, the curves of the high regulating valves CV2 and CV3 were revised online. The operation data of the steam turbine unit was then collected through the two tests.

[0007] A further improvement of the present invention is that the steam turbine unit operating data is acquired through two test collections, wherein the steam turbine unit operating data includes load, steam flow and turbine speed.

[0008] A further improvement of the present invention is that, by pre-processing the steam turbine unit operation data, an expert database of regulating valve flow characteristics is established, including: After preprocessing the steam turbine unit operation data, an expert database of regulating valve flow characteristics is formed according to the flow characteristics of the test regulating valve of the thermal power unit, including: system normal operation data, system fault operation data and system core parameter data.

[0009] A further improvement of the present invention is that a valve regulating flow characteristic expert library is used to perform feature analysis of data, extract feature independent variables, and input the feature independent variables into a neural network for learning, wherein a ResNet method is used to extract the feature independent variables.

[0010] A further improvement of the present invention is that, according to the predicted value learned by the neural network and the true value of the feature independent variable, the expected accuracy and stability are achieved by continuously optimizing the neural network parameters, and an optimized neural network model is obtained, wherein the neural network adopts the ResNet-LSTM neural network.

[0011] A steam turbine valve flow characteristic optimization system based on data analysis, comprising: The data collection module collects the steam turbine unit operation data through two tests; The data preprocessing module preprocesses the steam turbine unit operation data to establish an expert database of valve flow characteristics; The data analysis and modeling module uses the valve flow characteristic expert library to perform data feature analysis, extract feature independent variables, and input the feature independent variables into the neural network for learning; The model verification and adjustment module achieves the expected accuracy and stability based on the predicted values ​​learned by the neural network and the true values ​​of the feature independent variables, and obtains the optimized neural network model by continuously optimizing the neural network parameters; The flow characteristic optimization module optimizes the flow control characteristics based on the neural network model to obtain the best flow characteristics of the regulating valve.

[0012] A further improvement of the present invention is that, in the data collection module, the steam turbine unit operation data is collected through two tests, including: The test was carried out in two stages. In the first test, the corresponding relationship between the total valve position command, load and the opening of each regulating valve was obtained. In the second test, the curves of the high regulating valves CV2 and CV3 were revised online. The operation data of the steam turbine unit was then collected through the two tests.

[0013] A further improvement of the present invention is that, in the data collection module, the steam turbine unit operating data is collected through two tests, wherein the steam turbine unit operating data includes load, steam flow and turbine speed.

[0014] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the steps of a method for optimizing a steam turbine regulating valve flow characteristic based on data analysis.

[0015] Compared with the prior art, the present invention has at least the following beneficial technical effects: The present invention provides a method and system for optimizing the flow characteristics of a steam turbine regulating valve based on test data analysis. First, an expert database of regulating valve flow characteristics based on steam turbine operating parameters is established to ensure the integrity of the database. Secondly, the predicted value learned by the neural network and the real value of the characteristic independent variable are combined with the expert database of regulating valve flow characteristics to continuously adjust and update the neural network model online to find the best engineering application range. Finally, the best regulating valve flow characteristics that meet the actual on-site needs are obtained, and the optimized operation of the steam turbine is further improved. The present invention can obtain the best regulating valve flow characteristics and improve the operating efficiency and reliability of the steam turbine through data collection and preparation, data analysis and modeling, model verification and adjustment, flow characteristics optimization, and real-time monitoring and maintenance. This will make an important contribution to the development of the energy industry and environmental protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The present invention is a flow chart of a method for optimizing flow characteristics of a steam turbine regulating valve based on data analysis.

[0017] Figure 2 The present invention is a structural block diagram of a steam turbine regulating valve flow characteristic optimization system based on data analysis. DETAILED DESCRIPTION

[0018] In the following, only some exemplary embodiments are briefly described. As those skilled in the art will appreciate, the described embodiments may be modified in various ways without departing from the spirit or scope of the present invention. Therefore, the drawings and descriptions are considered to be exemplary and non-restrictive in nature.

[0019] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0020] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0021] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0022] Various structural schematic diagrams of the embodiments disclosed in the present invention are shown in the accompanying drawings. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0023] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0024] Example 1 like Figure 1 As shown, this embodiment provides a method for optimizing flow characteristics of a steam turbine regulating valve based on data analysis, comprising: The steam turbine unit operation data was collected through two tests; By pre-processing the steam turbine unit operation data, an expert database of valve flow characteristics is established; Use the valve flow characteristic expert database to perform data feature analysis, extract feature independent variables, and input the feature independent variables into the neural network for learning; According to the predicted value learned by the neural network and the true value of the feature independent variable, the expected accuracy and stability are achieved through continuous optimization of the neural network parameters, and the optimized neural network model is obtained; Based on the neural network model, the flow control characteristics are optimized to obtain the best flow characteristics of the regulating valve.

[0025] In this embodiment, the steam turbine unit operation data collected through two tests include: In the first test, the corresponding relationship between the total valve position command and the load and the opening of each regulating valve was obtained. The curves of CV2, CV3 and CV4 were revised based on the test data to improve the total valve position command and flow rate characterization. In the second test, the curves of the high-regulating valves CV2 and CV3 were revised online. Through the test, it was checked whether the modified valve position curve is energy-saving and what the energy-saving efficiency is, and the shaft system vibration and bearing temperature conditions were observed.

[0026] Through the linearization test and research of the high-pressure speed regulating valve flow curve, the points with poor linearity in the curve of the correspondence between the flow and the total valve position instruction are found. Under the premise of ensuring the safety performance of the unit, the valve function is optimized and corrected to improve the efficiency of the turbine and reduce coal consumption.

[0027] First trial project: In single-valve mode, the unit operates at a constant pressure of 16MPa, and the load is increased or decreased from 50% to 100%. After the load changes by 20MW and stabilizes for a period of time, the total valve position command, load, regulating valve opening and other related parameters are recorded. At the end of the test, the relevant curve is drawn.

[0028] Second test project: In single-valve mode, the curves of high-pressure regulating valves CV2 and CV3 are revised online, and the operating personnel adjust the unit parameters to try to maintain the stability of parameters such as vacuum, so that the unit can operate at a constant pressure of 16MPa, and increase or decrease the load from 50% to 100%. After each load change of 20MW and stabilization for a period of time, the total valve position command, load, regulating valve opening and other related parameters are recorded, and the relevant curves are drawn after the test.

[0029] The CV4 valve curve was revised based on the data from the first test project: In the single-valve mode, the operating personnel adjusted the unit parameters, tried to maintain the vacuum and other parameters stable, so that the unit operated at a constant pressure of 16MPa, and increased and decreased the load from 50% to 100%. After each load change of 20MW and stabilization for a period of time, the total valve position command, load, regulating valve opening and other related parameters were recorded, and the relevant curves were drawn after the test.

[0030] Finally, the collected steam turbine unit operating data includes load, steam flow and turbine speed.

[0031] In this embodiment, by preprocessing the steam turbine unit operation data, an expert database of valve flow characteristics is established, including: First, based on the collected operation data of the steam turbine unit, including load, steam flow, turbine speed, etc. Secondly, archive and organize these data to ensure the accuracy and completeness of the data. Finally, according to actual needs, select the appropriate time range and time interval to generate a data set for analysis. According to the flow characteristics of the test regulating valve of the thermal power unit, an expert library of the flow characteristics of the regulating valve is formed. The second step is to eliminate and screen the abnormal data, thereby forming the abnormal working conditions of the internal leakage data of the high-pressure anti-fuel oil system, and screen and eliminate them. The expert library of the valve flow characteristics includes: system normal operation data, system fault operation data and system core parameter data.

[0032] In this embodiment, the valve flow characteristic expert library is used to perform feature analysis on the data, extract feature independent variables, and input the feature independent variables into the neural network for learning, including: Through the analysis of test data, the flow characteristics of steam turbine control valve and its influencing factors are understood; The piecewise linearization with dead zone projection algorithm can be used to find out the factors related to flow characteristics and establish a mathematical model to describe the relationship between them. The load, steam flow and control valve opening are used as independent variables. The ResNet method is first used to extract the features of the data. Then the time series algorithm LSTM is built to establish the order between the data. The flow is used as the dependent variable for neural network learning.

[0033] A piecewise linear test data modeling method with a dead zone projection algorithm is adopted, and a piecewise linear test data modeling object description method includes: Consider the following n-input n-output nonlinear discrete system:

[0034] Where: u, y are the system input and output vectors; f{.} is the input and output nonlinear vector function; n a , n b is the system order.

[0035] Near the equilibrium point, it can be approximated as follows:

[0036] Where: A(q-1) and B(q-1) are n×n matrix polynomials with orders n and a and n b ; v[.] is a high-order nonlinear term.

[0037] The method for estimating the parameters of the piecewise linear test data modeling model includes:

[0038] Using dead zone projection algorithm:

[0039] Where: c>0 is a constant that is added to prevent the denominator from being zero, and θ(t) is the estimated value of θ(t) at time t. By finding θ, the object model can be obtained.

[0040] Mathematical analysis of steam inlet volume of turbine valve using piecewise linear test data modeling, including:

[0041] Where: Q1(k) represents the steam inlet volume of the single valve mode at time k; Q2(k) represents the steam inlet volume of the sequential valve mode at time k; α1 and α2 are the selection factors of the single valve and sequential valve modes respectively, and they satisfy:

[0042] From the above formula, we can get:

[0043] Where n is the number of high-speed valves; Fi(x) is the flow characteristic function of high-speed valve i in single valve mode; is the flow characteristic function of high-speed valve i in sequential valve mode; D(k)∈[0,100] is the real-time flow command variable at time k.

[0044] Piecewise linearization test data modeling deep learning, including: The theoretical predicted flow value of the valve is obtained, and the process parameters are obtained from a distributed control system, wherein the distributed control system is a control system used in the field of industrial automation, and the process parameters are used to analyze the actual flow value of the valve.

[0045] In this embodiment, according to the predicted value learned by the neural network and the true value of the feature independent variable, the expected accuracy and stability are achieved by continuously optimizing the neural network parameters, and an optimized neural network model is obtained, including: ResNet-LSTM neural network training is used, including: The process parameters are input into a neural network prediction model so as to process the process parameters using the process simulation model to obtain an actual flow value of the valve, wherein the process simulation model is trained using multiple sets of training data by machine learning, and each of the multiple sets of training data includes: sample process parameters and sample actual flow values ​​corresponding to the sample process parameters.

[0046] In addition, a subset of the test data can be used for validation to compare the difference between the actual flow and the flow predicted by the model. If there is a difference, the model can be adjusted, such as adding or removing independent variables, adjusting model parameters, etc., to improve the prediction accuracy.

[0047] Optimization algorithms, such as genetic algorithms and simulated annealing algorithms, can be used to optimize the model to maximize the stability and accuracy of the flow. At the same time, the relevant parameters of the control valve can also be optimized to improve the response speed and control accuracy of the control valve.

[0048] In this embodiment, once the optimized control valve flow characteristics are implemented, real-time monitoring and maintenance are required to ensure the normal operation of the steam turbine. A real-time data acquisition system and a monitoring system can be used to monitor the status of the steam turbine in real time and perform regular maintenance and inspection on the control valve.

[0049] The operation efficiency and reliability of the steam turbine can be improved by optimizing the flow characteristics of the steam turbine control valve based on the test data analysis. The optimized control valve flow characteristics can provide a more stable steam flow and reduce the impact of load fluctuations on the steam turbine. At the same time, the optimized control valve can also improve the response speed and control accuracy, and improve the operation flexibility and stability of the steam turbine. This will help reduce energy consumption and emissions, while improving the economic benefits and environmental friendliness of the steam turbine.

[0050] Example 2 like Figure 1 As shown, this embodiment provides a steam turbine valve flow characteristic optimization system based on data analysis, including: The data collection module collects the steam turbine unit operation data through two tests; The data preprocessing module preprocesses the steam turbine unit operation data to establish an expert database of valve flow characteristics; The data analysis and modeling module uses the valve flow characteristic expert library to perform data feature analysis, extract feature independent variables, and input the feature independent variables into the neural network for learning; The model verification and adjustment module achieves the expected accuracy and stability based on the predicted values ​​learned by the neural network and the true values ​​of the feature independent variables, and obtains the optimized neural network model by continuously optimizing the neural network parameters; The flow characteristic optimization module optimizes the flow control characteristics based on the neural network model to obtain the best flow characteristics of the regulating valve.

[0051] Example 3 This embodiment provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the steps of a method for optimizing the flow characteristics of a steam turbine regulating valve based on data analysis.

[0052] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0053] The present application is described with reference to the flowcharts and / or block diagrams of the methods, systems and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A system that specifies the functions of a box or multiple boxes.

[0054] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0055] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0056] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

[0057] In addition, it should be understood that although this specification is described in accordance with the implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation modes that can be understood by those skilled in the art. The above content is only to illustrate the technical idea of ​​the present invention, and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution according to the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A method for optimizing flow characteristics of a steam turbine valve based on data analysis, characterized in that: include: The steam turbine unit operation data was collected through two tests; By pre-processing the steam turbine unit operation data, an expert database of valve flow characteristics is established; Use the valve flow characteristic expert database to perform data feature analysis, extract feature independent variables, and input the feature independent variables into the neural network for learning; According to the predicted value learned by the neural network and the true value of the feature independent variable, the expected accuracy and stability are achieved through continuous optimization of the neural network parameters, and the optimized neural network model is obtained; Based on the neural network model, the flow control characteristics are optimized to obtain the best flow characteristics of the regulating valve.

2. The method for optimizing flow characteristics of a steam turbine regulating valve based on data analysis according to claim 1, characterized in that: The steam turbine unit operation data was collected through two tests, including: The test was carried out in two stages. In the first test, the corresponding relationship between the total valve position command, load and the opening of each regulating valve was obtained. In the second test, the curves of the high regulating valves CV2 and CV3 were revised online. The operation data of the steam turbine unit was then collected through the two tests.

3. The method for optimizing flow characteristics of a steam turbine valve based on data analysis according to claim 1, characterized in that: The steam turbine unit operating data was collected through two tests, including load, steam flow and turbine speed.

4. The method for optimizing the flow characteristics of a steam turbine valve based on data analysis according to claim 3 is characterized in that: By preprocessing the steam turbine unit operation data, an expert database of valve flow characteristics is established, including: After preprocessing the steam turbine unit operation data, an expert database of regulating valve flow characteristics is formed according to the flow characteristics of the test regulating valve of the thermal power unit, including: system normal operation data, system fault operation data and system core parameter data.

5. The method for optimizing flow characteristics of a steam turbine valve based on data analysis according to claim 1, characterized in that: The expert database of valve flow characteristics is used to perform feature analysis on the data, extract feature independent variables, and input the feature independent variables into the neural network for learning, in which the ResNet method is used to extract feature independent variables.

6. The method for optimizing flow characteristics of a steam turbine valve based on data analysis according to claim 1, characterized in that: According to the predicted values ​​learned by the neural network and the true values ​​of the feature independent variables, the expected accuracy and stability are achieved by continuously optimizing the neural network parameters, and an optimized neural network model is obtained, in which the neural network adopts the ResNet-LSTM neural network.

7. A steam turbine valve flow characteristic optimization system based on data analysis, characterized in that: include: The data collection module collects the steam turbine unit operation data through two tests; The data preprocessing module preprocesses the steam turbine unit operation data to establish an expert database of valve flow characteristics; The data analysis and modeling module uses the valve flow characteristic expert library to perform data feature analysis, extract feature independent variables, and input the feature independent variables into the neural network for learning; The model verification and adjustment module achieves the expected accuracy and stability based on the predicted values ​​learned by the neural network and the true values ​​of the feature independent variables, and obtains the optimized neural network model by continuously optimizing the neural network parameters; The flow characteristic optimization module optimizes the flow control characteristics based on the neural network model to obtain the best flow characteristics of the regulating valve.

8. The steam turbine valve flow characteristic optimization system based on data analysis according to claim 7, characterized in that: In the data collection module, the steam turbine unit operation data is collected through two tests, including: The test was carried out in two stages. In the first test, the corresponding relationship between the total valve position command, load and the opening of each regulating valve was obtained. In the second test, the curves of the high regulating valves CV2 and CV3 were revised online. The operation data of the steam turbine unit was then collected through the two tests.

9. The steam turbine valve flow characteristic optimization system based on data analysis according to claim 7, characterized in that: In the data collection module, the steam turbine unit operating data is collected through two tests, where the steam turbine unit operating data includes load, steam flow and turbine speed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the computer program implements the steps of a method for optimizing the flow characteristics of a steam turbine regulating valve based on data analysis according to any one of claims 1 to 6.

Images