Steam turbine circulating water pump characteristic real-time correction method based on support vector machine

Abstract

The invention relates to the technical field of valve steam turbines, in particular to a steam turbine circulating water pump characteristic real-time correction method based on a support vector machine, and the method comprises the steps: collecting at least one piece of operation data in a steam turbine circulating water pump; constructing a characteristic curve correction model of the steam turbine circulating water pump according to the at least one operation data; establishing a turbine circulating water pump feature library according to the circulating water pump power features and the circulating water pump flow features; based on the feature library, the circulating water pump current, the circulating water pump rotating speed and the circulating water pump power, classifying first training data and second training data of the steam turbine circulating water pump by using a support vector machine, and establishing a turbine circulating water pump characteristic curve dynamic correction model for correcting the turbine circulating water pump characteristic curve according to the classified training data. Therefore, the problems that the state of the steam turbine circulating water pump cannot be comprehensively known from different dimensions and different angles in the prior art, and the accuracy of operation adjustment of the circulating water pump cannot be improved are solved.

Description

Technical Field

[0001] This invention relates to the field of valve turbine technology, and in particular to a method for real-time correction of turbine circulating water pump characteristics based on support vector machine. Background Technology

[0002] Steam turbine operation requires adjusting the power of the circulating water pumps, which in turn regulates the circulating water flow rate, to cool the turbine condenser, reduce the turbine back pressure, and improve the unit's economic efficiency. The characteristic curves provided by hardware manufacturers often differ significantly from the actual unit; therefore, real-time correction of the turbine circulating water pump characteristic curves is crucial for guiding the operation of steam turbines in power plants.

[0003] In related technologies, the characteristic curve of the turbine circulating water pump is usually obtained based on theoretical calculations or experimental tests. The current turbine power generation is calculated based on the correction curve of exhaust pressure on turbine power generation, and the net power gain of the unit is calculated by changing the circulating water volume.

[0004] However, the relevant technologies cannot fully understand the status of the turbine circulating water pump from different dimensions and angles, and they cannot improve the accuracy of the operation and regulation of the circulating water pump, which urgently needs to be improved. Summary of the Invention

[0005] This invention provides a real-time correction method for the characteristics of a steam turbine circulating water pump based on support vector machines, in order to solve the problems that related technologies cannot fully understand the state of the steam turbine circulating water pump from different dimensions and angles, and cannot improve the accuracy of the operation and regulation of the circulating water pump.

[0006] A first aspect of this invention provides a method for real-time correction of the characteristics of a steam turbine circulating water pump based on a support vector machine (SVM), applied in the model building stage, comprising the following steps: collecting at least one type of operating data from the steam turbine circulating water pump; constructing a characteristic curve correction model for the steam turbine circulating water pump based on the at least one type of operating data; extracting the circulating water pump power characteristics and circulating water pump flow characteristics from the steam turbine circulating water pump based on the characteristic curve correction model, and establishing a feature library for the steam turbine circulating water pump based on the circulating water pump power characteristics and circulating water pump flow characteristics; classifying the first training data and the second training data of the steam turbine circulating water pump using a support vector machine based on the feature library and the circulating water pump current, circulating water pump speed, and circulating water pump power to generate classified data, and establishing a dynamic correction model for the characteristic curve of the steam turbine circulating water pump for correcting the characteristic curve of the steam turbine circulating water pump based on the classified training data.

[0007] Optionally, in one embodiment of the present invention, the acquisition of at least one type of operating data of the turbine circulating water pump includes: acquiring at least one type of operating data of the turbine circulating water pump, circulating water pump current, circulating water pump inlet and outlet water temperature, condenser back pressure, circulating water pump design curve parameters, and condenser design parameters.

[0008] Optionally, in one embodiment of the present invention, the step of constructing the characteristic curve correction model of the turbine circulating water pump based on the at least one piece of data includes: constructing the curve equation coefficients of the turbine circulating water pump based on the characteristic curve of the turbine circulating water pump; appending timestamps to the at least one piece of operating data to generate target operating data of the turbine circulating water pump; classifying the target operating data to determine the data group corresponding to the curve equation coefficients, and establishing the characteristic curve correction model of the turbine circulating water pump based on the data group.

[0009] Optionally, in one embodiment of the present invention, the step of extracting the power characteristics and flow characteristics of the circulating water pump in the turbine circulating water pump, and establishing a feature library of the turbine circulating water pump based on the power characteristics and flow characteristics, includes: extracting parameter data of the turbine circulating water pump from a preset equation coefficient database; performing a turbine cold-end thermodynamic analysis based on the parameter data to generate at least one operating data feature including the power characteristics and flow characteristics of the circulating water pump; and appending the timestamp to the at least one operating data feature to establish the feature library of the turbine circulating water pump.

[0010] Optionally, in one embodiment of the present invention, before classifying the first training data and second training data of the turbine circulating water pump using a support vector machine based on the feature library and the circulating water pump current, circulating water pump speed, and circulating water pump power, the method further includes: extracting corresponding data from the feature library respectively, and establishing a first training database and a second training database based on the corresponding data; analyzing the first training database using linear and nonlinear classifiers to obtain the first training data, and analyzing the second training database using linear regression and nonlinear regression methods to obtain the second training data.

[0011] A second aspect of the present invention provides a method for real-time correction of the characteristics of a steam turbine circulating water pump based on a support vector machine, applied in the model application stage, comprising the following steps: acquiring at least one operating data feature of the steam turbine circulating water pump, and inputting the at least one operating data feature into a pre-established dynamic correction model of the characteristic curve of the steam turbine circulating water pump, so as to correct the characteristic curve of the circulating water pump using the pre-established dynamic correction model of the characteristic curve of the steam turbine circulating water pump, wherein the dynamic correction model of the characteristic curve of the steam turbine circulating water pump is constructed from data in a feature database.

[0012] Optionally, in one embodiment of the present invention, the step of correcting the circulating water pump characteristic curve using the pre-established dynamic correction model of the turbine circulating water pump characteristic curve includes: using parameters in a first training database and a second training database to correct equation coefficients to generate corresponding correction values; based on the correction values, connecting the dynamic correction model of the turbine circulating water pump characteristic curve to a real-time data interface, and dynamically updating the first training database and the second training database using data input from the real-time data interface to generate updated first training database and second training database; adjusting correction parameters according to the updated first training database and second training database to correct the circulating water pump characteristic curve according to the adjusted correction parameters.

[0013] A third aspect of this invention provides a real-time characteristic correction device for a steam turbine circulating water pump based on a support vector machine, applied in the model building stage, comprising: a data acquisition module for acquiring at least one type of operating data from the steam turbine circulating water pump; a data construction module for constructing a characteristic curve correction model for the steam turbine circulating water pump based on the at least one type of operating data; an extraction module for extracting circulating water pump power characteristics and circulating water pump flow characteristics from the steam turbine circulating water pump based on the characteristic curve correction model, and establishing a feature library for the steam turbine circulating water pump based on the circulating water pump power characteristics and the circulating water pump flow characteristics; and a data establishment module for classifying first training data and second training data of the steam turbine circulating water pump using a support vector machine based on the feature library and circulating water pump current, circulating water pump speed, and circulating water pump power, to generate classified data, and establishing a dynamic correction model for the characteristic curve of the steam turbine circulating water pump for correcting the characteristic curve of the steam turbine circulating water pump based on the classified training data.

[0014] Optionally, in one embodiment of the present invention, the acquisition module includes: an acquisition unit, used to acquire at least one of the following operating data of the turbine circulating water pump: circulating water pump flow rate, circulating water pump current, circulating water pump inlet and outlet water temperature, condenser back pressure, circulating water pump design curve parameters, and condenser design parameters.

[0015] Optionally, in one embodiment of the present invention, the construction module includes: a construction unit, configured to construct curve equation coefficients of the turbine circulating water pump based on the characteristic curve of the turbine circulating water pump; a generation unit, configured to attach timestamps to the at least one type of operating data to generate target operating data of the turbine circulating water pump; and a determination unit, configured to classify the target operating data to determine the data group corresponding to the curve equation coefficients, and establish a characteristic curve correction model of the turbine circulating water pump based on the data group.

[0016] Optionally, in one embodiment of the present invention, the extraction module includes: an extraction unit for extracting parameter data of the turbine circulating water pump from a preset equation coefficient database; an analysis unit for performing a turbine cold-end thermodynamic analysis based on the parameter data to generate at least one operating data feature including the power characteristic and the flow characteristic of the circulating water pump; and an appending unit for appending the timestamp to the at least one operating data feature to establish a feature library of the turbine circulating water pump.

[0017] Optionally, in one embodiment of the present invention, it further includes: a data extraction module, used to extract corresponding data from the feature library before classifying the first training data and the second training data of the turbine circulating water pump using a support vector machine based on the feature library and the circulating water pump current, circulating water pump speed and circulating water pump power, and to establish a first training database and a second training database based on the corresponding data; and a training data generation module, used to analyze the first training database using linear and nonlinear classifiers to obtain the first training data, and to analyze the second training database using linear regression and nonlinear regression to obtain the second training data.

[0018] A fourth aspect of the present invention provides a real-time correction device for the characteristics of a steam turbine circulating water pump based on a support vector machine, applied in the model application stage, comprising: an acquisition module for acquiring at least one operating data feature of the steam turbine circulating water pump; and a correction module for inputting the at least one operating data feature into a pre-established dynamic correction model for the characteristic curve of the steam turbine circulating water pump, so as to correct the characteristic curve of the circulating water pump using the pre-established dynamic correction model for the characteristic curve of the steam turbine circulating water pump, wherein the dynamic correction model for the characteristic curve of the steam turbine circulating water pump is constructed from data in a feature database.

[0019] Optionally, in one embodiment of the present invention, the correction module includes: a numerical generation unit, configured to use parameters in a first training database and a second training database to correct equation coefficients and generate corresponding corrected values; an update unit, configured to connect the dynamic correction model of the turbine circulating water pump characteristic curve to a real-time data interface based on the corrected values, and dynamically update the first training database and the second training database using data input from the real-time data interface to generate updated first training database and second training database; and a correction unit, configured to adjust correction parameters according to the updated first training database and second training database, so as to correct the circulating water pump characteristic curve according to the adjusted correction parameters.

[0020] A fifth aspect of the present invention provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a real-time correction method for the characteristics of a steam turbine circulating water pump based on a support vector machine as described in the above embodiments.

[0021] This invention utilizes a support vector machine to simultaneously acquire data from different circulating water pumps. By combining the detection results of multiple features and a combined model, it comprehensively understands the status of the turbine circulating water pumps from different dimensions and perspectives, thereby improving the accuracy of monitoring the operating characteristics and adjusting the operation of the turbine circulating water pumps. This solves the problems of related technologies that cannot comprehensively understand the status of the turbine circulating water pumps from different dimensions and perspectives, and that they cannot improve the accuracy of circulating water pump operation adjustment.

[0022] 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

[0023] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart illustrating the application of a real-time correction method for turbine circulating water pump characteristics based on support vector machines in the model building stage, according to an embodiment of the present invention. Figure 2 This is an overall flowchart of a method for real-time correction of turbine circulating water pump characteristics based on support vector machine according to an embodiment of the present invention. Figure 3 A flowchart illustrating the application of a real-time correction method for turbine circulating water pump characteristics based on support vector machines in the model application stage, according to an embodiment of the present invention. Figure 4A schematic diagram of a real-time correction device for the characteristics of a steam turbine circulating water pump based on a support vector machine, provided in an embodiment of the present invention, applied in the model building stage; Figure 5 A schematic diagram of a real-time correction device for the characteristics of a steam turbine circulating water pump based on a support vector machine, according to an embodiment of the present invention, applied in the model application stage; Figure 6 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present invention.

[0024] Among them, 10-A real-time correction device for the characteristics of a steam turbine circulating water pump based on support vector machine; 100-Acquisition module, 200-Construction module, 300-Extraction module, 400-Establishment module; 20-A real-time correction device for the characteristics of a steam turbine circulating water pump based on support vector machine, 500-Acquisition module, 600-Correction module; 601-Memory, 602-Processor, 603-Communication interface. Detailed Implementation

[0025] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0026] The following describes a real-time correction method for turbine circulating water pump characteristics based on support vector machines (SVM) according to an embodiment of the present invention, with reference to the accompanying drawings. Addressing the problem mentioned in the background art that related technologies cannot comprehensively understand the state of turbine circulating water pumps from different dimensions and angles, and cannot improve the accuracy of circulating water pump operation regulation, the present invention provides a real-time correction method for turbine circulating water pump characteristics based on support vector machines. In this method, data from different circulating water pumps are simultaneously acquired using support vector machines, and the state of the turbine circulating water pump is comprehensively understood from different dimensions and angles by combining the detection results of multiple features and models. This improves the accuracy of monitoring the operating characteristics of the turbine circulating water pump and the accuracy of turbine circulating water pump operation regulation. Therefore, this solves the problems of related technologies being unable to comprehensively understand the state of turbine circulating water pumps from different dimensions and angles, and their inability to improve the accuracy of circulating water pump operation regulation.

[0027] Specifically, Figure 1 This is a flowchart illustrating the application of a real-time correction method for turbine circulating water pump characteristics based on support vector machines in the model building stage, as provided in an embodiment of the present invention.

[0028] like Figure 1As shown, the real-time correction method for turbine circulating water pump characteristics based on support vector machines includes the following steps: In step S101, at least one piece of operating data from the turbine circulating water pump is collected.

[0029] It is understood that the data collected by the circulating water pump in this embodiment of the invention includes real-time operation data and historical operation data.

[0030] In actual implementation, such as Figure 2 As shown, the embodiments of the present invention can collect at least one type of operating data from the turbine circulating water pump, thereby providing support for the subsequent construction of a characteristic curve correction model of the turbine circulating water pump, and thus improving the accuracy of turbine circulating water pump operation monitoring.

[0031] Optionally, in one embodiment of the present invention, at least one type of operating data of the turbine circulating water pump is collected, including: the circulating water pump flow rate, circulating water pump current, circulating water pump inlet and outlet water temperature, condenser back pressure, circulating water pump design curve parameters, and condenser design parameters.

[0032] It is understood that the circulating water pump flow rate in the embodiments of the present invention can be the amount of water passing through the circulating water pump per unit time; the circulating water pump current can reflect the working status and load of the pump; the circulating water pump inlet and outlet water temperatures can be monitored; the condenser back pressure can be the pressure after steam condensation; the circulating water pump design curve parameters include various performance indicators preset during the design of the circulating water pump; the condenser design parameters involve the design specifications of the condenser.

[0033] Specifically, embodiments of the present invention can collect at least one of the following operating data: circulating water pump flow rate, circulating water pump current, circulating water pump inlet and outlet water temperatures, condenser back pressure, circulating water pump design curve parameters, and condenser design parameters. The circulating water pump flow rate measurement error is required to be no more than 1%. The condenser design parameters include: condenser circulating water flow rate design value, condenser area, design cleanliness coefficient, condenser water resistance curve, and condenser heat load-back pressure characteristic curve.

[0034] The embodiments of the present invention can collect at least one type of operational data to improve the real-time performance and accuracy of the turbine circulating water pump operation status assessment: without increasing the number of measuring point sensors, the real-time performance of the circulating water pump operation status assessment is improved, the result deviation caused by the time difference between the test correction time point and the actual operation process is reduced, and the accuracy of the circulating water pump assessment results is improved.

[0035] In step S102, a characteristic curve correction model of the turbine circulating water pump is constructed based on at least one set of operating data.

[0036] It is understood that the embodiments of the present invention can construct a characteristic curve correction model of the turbine circulating water pump based on at least one set of operating data. The present invention can serve as a guide for the adjustment of the turbine cold end system. The online characteristic curve correction algorithm can accurately assess the current operating status of the turbine circulating water pump without relying on additional measuring point sensors, and improve the optimization of the influencing factors of the circulating water pump during the economic operation of the turbine cold end system.

[0037] Optionally, in one embodiment of the present invention, constructing a characteristic curve correction model for a turbine circulating water pump based on at least one set of data includes: constructing curve equation coefficients for the turbine circulating water pump based on its characteristic curve; attaching timestamps to at least one set of operating data to generate target operating data for the turbine circulating water pump; classifying the target operating data to determine data groups corresponding to the curve equation coefficients; and establishing a characteristic curve correction model for the turbine circulating water pump based on the data groups.

[0038] It is understood that the coefficients of the curve equation in the embodiments of the present invention can be represented by A0, A1, and A2.

[0039] In practical implementation, embodiments of the present invention can fit curves to various curve equations and construct parametric equations based on the characteristic curves of the circulating water pump design, with A0, A1, A2, etc. representing the coefficients of the curve equations. At least one piece of operational data obtained in the above steps is timestamped to generate target operational data for the turbine circulating water pump. This target operational data is divided into 5 categories, with labels D1, D2, D3, D4, and D5 representing the data groups corresponding to the 5 coefficient equations. A characteristic curve correction model for the turbine circulating water pump is then established based on these data groups.

[0040] Furthermore, the data classification is based on the mathematical properties of each curve equation, including but not limited to the coefficient relationships of quadratic equations, linear equations, circles, ellipses, hyperbolas, etc. Taking a standard quadratic polynomial as an example, label D1 can represent the data set of the quadratic coefficient A0, the linear term A1, and the constant term A2. Taking a standard linear equation as an example, label D2 represents the data set of the two coefficients of the slope A1 and the intercept A2 of the line.

[0041] Furthermore, the classification criteria for the labels include: 1) Classification criteria for long-term historical data feature database: The database covers a time period of at least one year and all operating conditions. The features of this part of the database can represent the average operating level of the unit and will not be affected by individual factors.

[0042] 2) Classification criteria for wide-range operating condition data feature library: the most recent time is given priority, the range of data covers the upper and lower limits of certain restrictions, and at the same time, the wide range intervals are represented by only a certain finite number of data.

[0043] 3) Feature library of operating conditions with the most frequent occurrences: taking the most recent time as the priority factor, and according to the normal distribution probability, taking the standard deviation width of the probability distribution above a certain limit value as the classification basis.

[0044] 4) Operating data feature library near the maximum value: Considering both actual machine values ​​and design values, the data range is defined by a certain limit value below the maximum value and the design maximum value.

[0045] 5) Minimum value appendix operating condition data feature library: simultaneously considering actual machine values ​​and design values, the data range with a certain limit value below the minimum value and design minimum value.

[0046] The embodiments of the present invention can effectively construct a characteristic curve correction model of a steam turbine circulating water pump, enabling precise monitoring and optimization of its performance. This not only improves monitoring accuracy but also enhances the economy and reliability of the system without adding additional sensors.

[0047] In step S103, based on the characteristic curve correction model, the power characteristics and flow characteristics of the circulating water pump in the turbine circulating water pump are extracted, and a feature library of the turbine circulating water pump is established according to the power characteristics and flow characteristics of the circulating water pump.

[0048] It is understood that the feature library in the embodiments of the present invention can be established according to different classification criteria to adapt to different needs.

[0049] In actual implementation, the embodiments of the present invention can extract features from the circulating water pump operating data based on the characteristic curve correction model and the thermodynamic principle of the cold end of the steam turbine, extract the circulating water pump power features and circulating water pump flow features from the steam turbine circulating water pump, and establish a feature library of the steam turbine circulating water pump based on the circulating water pump power features and circulating water pump flow features.

[0050] The embodiments of the present invention can comprehensively understand the status of the turbine circulating water pump from different dimensions and angles by combining the detection results of various features and models, thereby improving the accuracy of circulating water pump operation and regulation.

[0051] Optionally, in one embodiment of the present invention, extracting the power characteristics and flow characteristics of the circulating water pump in the turbine circulating water pump, and establishing a feature library of the turbine circulating water pump based on the power characteristics and flow characteristics, includes: extracting parameter data of the turbine circulating water pump from a preset equation coefficient database; performing a turbine cold-end thermodynamic analysis based on the parameter data to generate at least one operating data feature including the power characteristics and flow characteristics of the circulating water pump; and attaching a timestamp to the at least one operating data feature to establish the feature library of the turbine circulating water pump.

[0052] It is understood that the preset equation coefficient database in the embodiments of the present invention can be three.

[0053] In actual implementation, this embodiment of the invention can extract data from the coefficient database of three equation sets, and perform turbine cold-end thermodynamic analysis on the extracted data to complete feature value extraction, including circulating water pump power characteristics and circulating water pump flow characteristics. After the extracted feature values ​​are timestamped, they are stored in feature libraries labeled F1, F2, F3, F4 and F5, where F1 corresponds to the long-term historical data feature library, F2 corresponds to the wide-range operating condition data feature library, F3 corresponds to the operating condition data feature library with the most occurrences, F4 corresponds to the operating data feature library near the maximum value, and F5 corresponds to the operating data feature library near the minimum value.

[0054] In this embodiment of the invention, the physical characteristic values ​​include: the circulating water pump flow rate calculated based on the condenser back pressure, and the circulating water pump power calculated based on the theory of the relationship between circulating water pump current, voltage, and resistance; the time characteristic values ​​are the mean, root mean square, standard deviation, minimum, maximum, autocorrelation function, partial autocorrelation function, moving average, moving standard deviation, and exponentially weighted moving average of each corresponding characteristic in the physical characteristic values.

[0055] The embodiments of the present invention can establish a feature library of turbine circulating water pumps, and further comprehensively understand the status of turbine circulating water pumps from different dimensions and angles, thereby improving the accuracy of circulating water pump operation and regulation.

[0056] In step S104, based on the feature library and the circulating water pump current, circulating water pump speed and circulating water pump power, the first training data and the second training data of the turbine circulating water pump are classified using support vector machine to generate classified data, and a dynamic correction model of the turbine circulating water pump characteristic curve is established based on the classified training data to correct the characteristic curve of the turbine circulating water pump.

[0057] It is understood that embodiments of the present invention can acquire data from different circulating water pumps simultaneously using a support vector machine.

[0058] In this embodiment of the invention, based on a feature library and circulating water pump current, circulating water pump speed and circulating water pump power, the first training data and the second training data of the turbine circulating water pump are classified using a support vector machine to generate classified data, and a dynamic correction model of the turbine circulating water pump characteristic curve is established based on the classified training data to correct the characteristic curve of the turbine circulating water pump.

[0059] The embodiments of the present invention can improve the accuracy of turbine circulating water pump operation monitoring: by feature extraction and feature library classification, data from sensors of different circulating water pumps are obtained simultaneously, and by combining multiple features and the results of support vector machine models, the operating status of turbine circulating water pumps can be comprehensively understood from different dimensions and angles, thereby improving the accuracy of circulating water pump regulation.

[0060] Optionally, in one embodiment of the present invention, before classifying the first and second training data of the turbine circulating water pump using a support vector machine based on the feature library and the circulating water pump current, circulating water pump speed, and circulating water pump power, the method further includes: extracting corresponding data from the feature library respectively, and establishing a first training database and a second training database based on the corresponding data; analyzing the first training database using linear and nonlinear classifiers to obtain the first training data, and analyzing the second training database using linear regression and nonlinear regression methods to obtain the second training data.

[0061] It is understood that in the embodiments of the present invention, the first training database is the T1 training database and the second training database is the T2 training database.

[0062] In actual implementation, the embodiments of the present invention can extract data from feature libraries F1, F2, F3, F4 and F5 respectively, and establish two training databases, namely training databases T1 and T2. The data of T1 comes from F1 and F2, and weights are set for each feature library. For example, if the recent data is relatively stable and covers a wide range, a higher weight can be set for F2. If the recent data is poor, a higher weight can be set for F1. The data of T2 comes from F3, F4 and F5, and weights are set for each feature library. Similarly, the weights are set according to the time similarity of these data.

[0063] Furthermore, support vector machine (SVM) classification analysis is performed on the T1 training database. The SVM classification analysis construction methods include linear and nonlinear classifiers. Support vector machine (SVM) regression analysis is performed on the T2 training database. The regression model construction methods include linear regression and nonlinear regression.

[0064] Furthermore, the training data is classified using a support vector machine, and the classification criteria include, but are not limited to, the current of the circulating water pump, the rotational speed of the circulating water pump, and the power of the circulating water pump.

[0065] Furthermore, Figure 3 This is a flowchart illustrating the application of a real-time correction method for turbine circulating water pump characteristics based on support vector machines in the model application stage, as provided in an embodiment of the present invention.

[0066] like Figure 3 As shown, the real-time correction method for turbine circulating water pump characteristics based on support vector machines includes the following steps: Step S301: Obtain at least one operating data feature of the turbine circulating water pump.

[0067] In this embodiment of the invention, at least one operational data feature of the turbine circulating water pump can be obtained, thereby providing support for subsequent model construction and improving the accuracy of turbine circulating water pump operation monitoring.

[0068] Step S302: Input at least one operating data feature into the pre-established dynamic correction model of the turbine circulating water pump characteristic curve, so as to correct the circulating water pump characteristic curve using the pre-established dynamic correction model of the turbine circulating water pump characteristic curve, wherein the dynamic correction model of the turbine circulating water pump characteristic curve is constructed from the data in the feature database.

[0069] In this embodiment of the invention, at least one operating data feature can be input into a pre-established dynamic correction model of the turbine circulating water pump characteristic curve. The dynamic correction model of the turbine circulating water pump characteristic curve is composed of data from a feature database. This invention can adapt to different types of turbine circulating water pumps: different types of turbine circulating water pumps have different operating conditions and measurement point types, but they all have their own characteristic curves. By using multiple models, online correction can be performed on the characteristic curves of different types of circulating water pumps, which has a wider range of applicability.

[0070] Optionally, in one embodiment of the present invention, the characteristic curve of the circulating water pump is corrected using a pre-established dynamic correction model of the turbine circulating water pump characteristic curve, including: using parameters in a first training database and a second training database to correct equation coefficients to generate corresponding correction values; based on the correction values, connecting the dynamic correction model of the turbine circulating water pump characteristic curve to a real-time data interface, and dynamically updating the first and second training databases using data input from the real-time data interface to generate updated first and second training databases; adjusting correction parameters according to the updated first and second training databases to correct the circulating water pump characteristic curve according to the adjusted correction parameters.

[0071] In this embodiment of the invention, parameters obtained from training databases T1 and T2 can be used to correct the equation coefficients established in the above steps, generating corresponding correction values. The curve polynomial expression constructed from the coefficients is used as the expression for the circulating water pump characteristic curve, and the upper and lower bounds of the circulating water pump characteristic curve are used as the domain of the polynomial expression. The dynamic correction model for the turbine circulating water pump characteristic curve is connected to a real-time data interface, dynamically updating the above training database based on the input data, calculating and adjusting correction parameters, thus realizing a real-time correction method for the turbine circulating water pump characteristic curve.

[0072] This invention proposes a real-time correction method for turbine circulating water pump characteristics based on support vector machines. By using support vector machines, it simultaneously acquires data from different circulating water pumps and comprehensively understands the state of the turbine circulating water pumps from different dimensions and perspectives through the detection results of multiple features and combined models. This improves the accuracy of monitoring the operating characteristics and adjusting the operation of the turbine circulating water pumps. Therefore, it solves the problem that related technologies cannot comprehensively understand the state of turbine circulating water pumps from different dimensions and perspectives, and cannot improve the accuracy of circulating water pump operation adjustment.

[0073] Next, referring to the accompanying drawings, a real-time correction device for the characteristics of a steam turbine circulating water pump based on a support vector machine is described according to an embodiment of the present invention.

[0074] Figure 4 This is a schematic diagram of a real-time correction device for the characteristics of a steam turbine circulating water pump based on a support vector machine, according to an embodiment of the present invention.

[0075] like Figure 4 As shown, the real-time correction device 10 for the characteristics of a steam turbine circulating water pump based on support vector machine includes: a data acquisition module 100, a data construction module 200, an extraction module 300, and a data establishment module 400.

[0076] Specifically, the acquisition module 100 is used to acquire at least one type of operating data from the turbine circulating water pump.

[0077] Module 200 is used to construct a characteristic curve correction model of the turbine circulating water pump based on at least one set of operating data.

[0078] The extraction module 300 is used to extract the power characteristics and flow characteristics of the circulating water pump in the turbine circulating water pump based on the characteristic curve correction model, and to establish a feature library of the turbine circulating water pump based on the power characteristics and flow characteristics.

[0079] Module 400 is established to classify the first and second training data of the turbine circulating water pump based on the feature library and the circulating water pump current, circulating water pump speed and circulating water pump power using support vector machine to generate classified data, and to establish a dynamic correction model of the turbine circulating water pump characteristic curve based on the classified training data to correct the characteristic curve of the turbine circulating water pump.

[0080] Optionally, in one embodiment of the present invention, the acquisition module 100 includes: an acquisition unit.

[0081] The data acquisition unit is used to acquire at least one of the following operating data: circulating water pump flow rate, circulating water pump current, circulating water pump inlet and outlet water temperature, condenser back pressure, circulating water pump design curve parameters, and condenser design parameters.

[0082] Optionally, in one embodiment of the present invention, the construction module 200 includes: a construction unit, a generation unit, and a determination unit.

[0083] Among them, the construction unit is used to construct the curve equation coefficients of the turbine circulating water pump based on the characteristic curve of the turbine circulating water pump.

[0084] The generation unit is used to append a timestamp to at least one piece of operating data to generate target operating data for the turbine circulating water pump.

[0085] The determination unit is used to classify the target operating data to determine the data group corresponding to the coefficients of the curve equation, and to establish a characteristic curve correction model of the turbine circulating water pump based on the data group.

[0086] Optionally, in one embodiment of the present invention, the extraction module 300 includes: an extraction unit, an analysis unit, and an additional unit.

[0087] The extraction unit is used to extract parameter data of the turbine circulating water pump from the preset equation coefficient database.

[0088] The analysis unit is used to perform thermodynamic analysis of the turbine cold end based on parameter data to generate at least one operating data feature, including circulating water pump power characteristics and circulating water pump flow characteristics. An additional unit is used to attach a timestamp to at least one operating data feature to establish a feature library for the turbine circulating water pump.

[0089] Optionally, in one embodiment of the present invention, a real-time correction device 10 for the characteristics of a steam turbine circulating water pump based on a support vector machine further includes: a data extraction module and a training data generation module.

[0090] The data extraction module is used to extract corresponding data from the feature library before classifying the turbine circulating water pump using support vector machines based on the feature library and the circulating water pump current, circulating water pump speed and circulating water pump power, and to establish the first training database and the second training database based on the corresponding data.

[0091] The training data generation module is used to analyze the first training database using linear and nonlinear classifiers to obtain the first training data, and to analyze the second training database using linear and nonlinear regression methods to obtain the second training data.

[0092] Figure 5 A real-time correction device 20 for the characteristics of a steam turbine circulating water pump based on a support vector machine is provided for application in the model application stage. It includes an acquisition module 500 and a correction module 600.

[0093] Specifically, the acquisition module 500 is used to acquire at least one operating data feature of the turbine circulating water pump.

[0094] The correction module 600 is used to input at least one operating data feature into a pre-established dynamic correction model of the turbine circulating water pump characteristic curve, so as to correct the circulating water pump characteristic curve using the pre-established dynamic correction model of the turbine circulating water pump characteristic curve, wherein the dynamic correction model of the turbine circulating water pump characteristic curve is constructed from data in the feature database.

[0095] Optionally, in one embodiment of the present invention, the correction module 600 includes: a value generation unit, an update unit, and a correction unit.

[0096] The numerical generation unit is used to modify the coefficients of the equations using parameters from the first training database and the second training database to generate corresponding modified values.

[0097] The update unit is used to connect the dynamic correction model of the turbine circulating water pump characteristic curve to the real-time data interface based on the correction values, and dynamically update the first training database and the second training database using the data input from the real-time data interface, thereby generating the updated first training database and the second training database.

[0098] The correction unit is used to adjust the correction parameters according to the updated first training database and second training database, so as to correct the characteristic curve of the circulating water pump according to the adjusted correction parameters.

[0099] It should be noted that the foregoing explanation of an embodiment of a real-time correction method for turbine circulating water pump characteristics based on support vector machines also applies to a real-time correction device for turbine circulating water pump characteristics based on support vector machines in this embodiment, and will not be repeated here.

[0100] This invention proposes a real-time correction device for turbine circulating water pump characteristics based on support vector machines. By using support vector machines, it simultaneously acquires data from different circulating water pumps and comprehensively understands the status of the turbine circulating water pumps from different dimensions and perspectives through the detection results of multiple features and combined models. This improves the accuracy of monitoring the operating characteristics and adjusting the operation of the turbine circulating water pumps. Therefore, it solves the problem that related technologies cannot comprehensively understand the status of turbine circulating water pumps from different dimensions and perspectives, and cannot improve the accuracy of circulating water pump operation adjustment.

[0101] Figure 6 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. The electronic device may include: The memory 601, the processor 602, and the computer program stored on the memory 601 and capable of running on the processor 602.

[0102] When the processor 602 executes the program, it implements the real-time correction method for the characteristics of a steam turbine circulating water pump based on a support vector machine provided in the above embodiments.

[0103] Furthermore, electronic devices also include: Communication interface 603 is used for communication between memory 601 and processor 602.

[0104] The memory 601 is used to store computer programs that can run on the processor 602.

[0105] The memory 601 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0106] If the memory 601, processor 602, and communication interface 603 are implemented independently, then the communication interface 603, memory 601, and processor 602 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 6 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0107] Optionally, in a specific implementation, if the memory 601, processor 602, and communication interface 603 are integrated on a single chip, then the memory 601, processor 602, and communication interface 603 can communicate with each other through an internal interface.

[0108] Processor 602 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0109] 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.

[0110] 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 indicated technical features. 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, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0111] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0112] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0113] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any of the following techniques known in the art, or a combination thereof: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0114] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments.

[0115] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0116] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for real-time correction of turbine circulating water pump characteristics based on support vector machine, characterized in that, Applied to the model building phase, it includes the following steps: Collect at least one type of operating data from the turbine circulating water pump; Construct a characteristic curve correction model for the turbine circulating water pump based on at least one of the aforementioned operating data; Based on the characteristic curve correction model, the power characteristics and flow characteristics of the circulating water pump in the turbine circulating water pump are extracted, and a feature library of the turbine circulating water pump is established according to the power characteristics and flow characteristics of the circulating water pump. Based on the feature library and the circulating water pump current, circulating water pump speed and circulating water pump power, the first training data and the second training data of the turbine circulating water pump are classified using support vector machine to generate classified data, and a dynamic correction model of the turbine circulating water pump characteristic curve is established based on the classified training data to correct the characteristic curve of the turbine circulating water pump.

2. The method for real-time correction of turbine circulating water pump characteristics based on support vector machine according to claim 1, characterized in that, The collection of at least one piece of operating data from the turbine circulating water pump includes: Collect at least one of the following operating data from the turbine circulating water pump: circulating water pump flow rate, circulating water pump current, circulating water pump inlet and outlet water temperature, condenser back pressure, circulating water pump design curve parameters, and condenser design parameters.

3. The method for real-time correction of turbine circulating water pump characteristics based on support vector machine according to claim 1, characterized in that, The step of constructing the characteristic curve correction model of the turbine circulating water pump based on the at least one piece of data includes: Based on the characteristic curve of the turbine circulating water pump, the curve equation coefficients of the turbine circulating water pump are constructed. The at least one piece of operating data is appended with a timestamp to generate the target operating data for the turbine circulating water pump; The target operating data is classified to determine the data group corresponding to the coefficients of the curve equation, and a characteristic curve correction model of the turbine circulating water pump is established based on the data group.

4. The method for real-time correction of turbine circulating water pump characteristics based on support vector machine according to claim 3, characterized in that, The step of extracting the power characteristics and flow characteristics of the circulating water pumps in the turbine circulating water pumps, and establishing a feature library of the turbine circulating water pumps based on the power characteristics and flow characteristics, includes: Extract the parameter data of the turbine circulating water pump from the preset equation coefficient database; Perform a steam turbine cold-end thermodynamic analysis based on the parameter data to generate at least one operating data feature including the circulating water pump power characteristic and the circulating water pump flow characteristic; The timestamp is appended to the at least one operational data feature to establish a feature library for the turbine circulating water pump.

5. The method for real-time correction of turbine circulating water pump characteristics based on support vector machine according to claim 1, characterized in that, Before classifying the first and second training data of the turbine circulating water pump using a support vector machine based on the feature library and the circulating water pump current, circulating water pump speed, and circulating water pump power, the process further includes: Extract the corresponding data from the feature library respectively, and establish a first training database and a second training database based on the corresponding data; The first training database is analyzed using linear and nonlinear classifiers to obtain the first training data, and the second training database is analyzed using linear and nonlinear regression methods to obtain the second training data.

6. A method for real-time correction of turbine circulating water pump characteristics based on support vector machine, characterized in that, When applied to the model application phase, the following steps are included: Obtain at least one operating data feature of the turbine circulating water pump; The at least one operating data feature is input into a pre-established dynamic correction model of the turbine circulating water pump characteristic curve, so as to correct the circulating water pump characteristic curve using the pre-established dynamic correction model of the turbine circulating water pump characteristic curve, wherein the dynamic correction model of the turbine circulating water pump characteristic curve is constructed from data in the feature database.

7. The method for real-time correction of turbine circulating water pump characteristics based on support vector machine according to claim 6, characterized in that, The step of correcting the circulating water pump characteristic curve using the pre-established dynamic correction model of the turbine circulating water pump characteristic curve includes: The coefficients of the equation are corrected using parameters from the first and second training databases to generate corresponding corrected values. Based on the corrected values, the dynamic correction model of the turbine circulating water pump characteristic curve is connected to the real-time data interface, and the first training database and the second training database are dynamically updated using the data input from the real-time data interface to generate the updated first training database and the second training database. The correction parameters are adjusted based on the updated first and second training databases to correct the characteristic curve of the circulating water pump according to the adjusted correction parameters.

8. A real-time correction device for the characteristics of a steam turbine circulating water pump based on a support vector machine, characterized in that, Applied to the model building phase, including: The data acquisition module is used to collect at least one type of operating data from the turbine circulating water pump; A construction module is used to construct a characteristic curve correction model of the turbine circulating water pump based on the at least one type of operating data; The extraction module is used to extract the power characteristics and flow characteristics of the circulating water pump in the turbine circulating water pump based on the characteristic curve correction model, and to establish a feature library of the turbine circulating water pump based on the power characteristics and flow characteristics of the circulating water pump. A module is established to classify the first and second training data of the turbine circulating water pump based on the feature library and the circulating water pump current, circulating water pump speed and circulating water pump power using a support vector machine to generate classified data, and to establish a dynamic correction model of the turbine circulating water pump characteristic curve for correcting the characteristic curve of the turbine circulating water pump based on the classified training data.

9. A real-time correction device for the characteristics of a steam turbine circulating water pump based on a support vector machine, characterized in that, Applied to the model application phase, including: The acquisition module is used to acquire at least one operating data feature of the turbine circulating water pump; The correction module is used to input the at least one operating data feature into a pre-established dynamic correction model of the turbine circulating water pump characteristic curve, so as to correct the circulating water pump characteristic curve using the pre-established dynamic correction model of the turbine circulating water pump characteristic curve, wherein the dynamic correction model of the turbine circulating water pump characteristic curve is constructed from data in the feature database.

10. An electronic device, characterized in that, include: The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a real-time correction method for turbine circulating water pump characteristics based on support vector machines as described in any one of claims 1-5 or 6-7.

Images