Deep peak regulation thermal power generating unit speed regulation system parameter identification and evaluation method

Abstract

The invention relates to a deep peak regulation thermal power generating unit speed regulation system parameter identification and evaluation method, which comprises the following steps: establishing a simulation model of a steam turbine and a speed regulation system thereof, including a speed regulator, an execution mechanism and a steam turbine model; obtaining static and dynamic test data of the unit, and preprocessing and normalizing the static and dynamic test data; on the basis of unit static test data, partial parameters in the speed regulator and execution mechanism model and a hydraulic servo-motor on-off time constant are calculated, and PID parameters of the speed regulator and execution mechanism model are identified by adopting an intelligent identification method; valve flow characteristics are obtained based on unit dynamic test data under different depth peak regulation operation conditions, and turbine model parameters are identified by adopting an intelligent identification method; and the identification parameters are substituted into the simulation model for operation, the goodness of fit of simulation and actual measurement output data is used as an evaluation index to evaluate an identification result, and a deep peak regulation thermal power generating unit model suitable for power system simulation analysis is established. According to the method, the simulation precision and adaptability of the unit model under the deep peak regulation working condition are effectively improved.

Description

Technical Field

[0001] This invention belongs to the field of power system operation and control technology, specifically relating to a method for identifying and evaluating parameters of deep peak-shaving thermal power units based on measured data. Background Technology

[0002] Under the guidance of the "dual-carbon" strategy, my country's power system development goal is to build a new type of power system with new energy sources as the mainstay. With the expansion of new energy grid connection, its randomness and volatility have led to problems such as low frequency regulation quality and large output fluctuations in hybrid systems. To improve the system's insufficient flexible peak-shaving capacity, my country is vigorously promoting the thermal power flexibility enhancement project. The typical model parameters of thermal power units obtained based on rated operating conditions are no longer applicable to deep peak-shaving conditions, and their simulation calculation results deviate significantly from actual operating conditions. This is because the typical model does not fully consider the nonlinear elements in the turbine and its speed regulation system, but instead performs simple linear fitting. As a result, when the peak-shaving depth of the unit varies widely, the influence of the ignored nonlinear elements becomes increasingly apparent.

[0003] Currently, my country is committed to building a new power system with new energy sources as the mainstay. With the large-scale grid connection of new energy sources such as wind power and photovoltaics, their inherent randomness and volatility pose severe challenges to the frequency regulation quality and power stability of the power grid. To improve the flexible adjustment capability of the power system, deep peak-shaving operation of thermal power units has become an inevitable choice. However, the simulation results of typical thermal power unit models obtained based on rated operating conditions show significant deviations from actual operating conditions when the units are under low load and large-scale deep peak-shaving conditions.

[0004] The main reason for this deviation is that existing typical models fail to fully consider the nonlinear elements (such as valve flow characteristics) in the turbine and its speed regulation system, often employing simple linearization. When the peak-shaving depth of the unit varies over a wide range, these neglected nonlinear effects become increasingly significant, leading to a decrease in the accuracy of model-based simulation analysis, primary frequency regulation characteristic evaluation, and control strategy optimization.

[0005] Therefore, by combining measured data of the unit under actual deep peak shaving conditions, it is of great significance to refine the simulation model of the turbine and its speed regulation system that can accurately reflect the dynamic characteristics under different peak shaving depths, and to quickly and accurately identify its key parameters, so as to improve the accuracy of power grid simulation and optimize the unit operation and control strategy. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method for identifying and evaluating the parameters of the speed regulation system of a deep peak-shaving thermal power unit. This method can effectively utilize field test data to identify the parameters of each module of the speed regulation system step by step and in stages, thereby quickly and accurately obtaining the unit model parameters adapted to different deep peak-shaving conditions.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: a method for parameter identification and evaluation of speed regulation system of deep peak-shaving thermal power unit, comprising the following steps:

[0008] S1: Establish a simulation model of the steam turbine and its speed control system on the simulation platform. The simulation model includes a governor model, an actuator model, and a steam turbine model.

[0009] S2: Obtain the static and speed disturbance dynamic test data of the unit under the set operating conditions, perform preprocessing on the test data including denoising and smoothing, and normalize the preprocessed data to obtain the measured input and output data of the governor model, actuator model and turbine model.

[0010] S3: Based on the static test data of the unit, calculate and obtain the parameters of the limiting, dead zone, and delay links in the governor model and the actuator model, as well as the start and stop time constant of the hydraulic actuator in the actuator model;

[0011] S4: Based on the static test data of the unit, the PID parameters of the governor model and the actuator model are identified and obtained using an intelligent identification method.

[0012] S5: Based on the dynamic test data of unit speed disturbance under different peak shaving operation conditions, piecewise linear fitting is performed to obtain valve flow characteristics and establish the nonlinear relationship between valve opening, main steam pressure and steam flow.

[0013] S6: Based on the dynamic test data of unit speed disturbance and valve flow characteristics under different peak shaving operation conditions, an intelligent identification method is used to identify and obtain the relevant parameters of the turbine model;

[0014] S7: Substitute the identified parameters into the simulation model, run the simulation model to obtain the model output data; use the goodness of fit between the model output data and the measured output data as the evaluation index to evaluate the model parameter identification results, and establish a deep peak-shaving thermal power unit model suitable for power system simulation analysis.

[0015] Further, in step S1, the inputs of the speed governor model are the power setpoint, the speed setpoint, the active power of the unit, and the unit speed. After passing through the power feedforward and controller control links, the output is the valve opening command.

[0016] The input to the actuator model is the valve opening command, which, after passing through the controller and hydraulic actuator, outputs the valve opening.

[0017] The input to the turbine model is the main steam flow rate calculated based on the valve opening and main steam pressure, and the output is mechanical power. The turbine model takes into account the valve flow characteristics.

[0018] Further, in step S3, based on the input and output data of the large-step fully open and large-step fully closed disturbances of the actuator valve opening in the unit static test data, the start-up and closing time constant of the hydraulic actuator in the actuator model is calculated; the formula for calculating the start-up and closing time constant of the hydraulic actuator is:

[0019]

[0020] In the formula, T is the start-stop time constant of the hydraulic actuator; P max and P min These represent the maximum and minimum values ​​of the valve opening, respectively; Δt is the time required for the valve to fully open or fully close during a large step; ΔP y This refers to the change in valve opening during a large step transition between full opening and full closing; when the hydraulic actuator is activated, ΔP y This represents the increase in valve opening, where T corresponds to the opening time constant of the hydraulic actuator; when the hydraulic actuator is closed, ΔP y This represents the amount of decrease in the valve opening, and T corresponds to the closing time constant of the hydraulic actuator.

[0021] Further, in step S4, based on the input and output data of the small step disturbance of the actuator valve opening in the static test data of the unit, an intelligent identification method is used to identify the PID controller parameters of the governor model and the actuator model; the objective function f of the intelligent identification method is to obtain the minimum mean square error R. MSE With the objective function f as the goal, the expression for the objective function f is:

[0022]

[0023] In the formula, n is the total number of data points used in the objective function calculation; y i and y represents the i-th measured output data and the model output data, respectively; for the speed governor model, y refers to the gate opening command output by the speed governor model; for the actuator model, y refers to the gate opening output by the actuator model.

[0024] Furthermore, in step S5, based on the dynamic test data of unit speed disturbance under different peak-shaving operating conditions, the valve flow characteristics are obtained, specifically including:

[0025] For the turbine model, let the unit output at different peak-shaving operating points be P. g1P g2 ... P gn Select the unit output range [P] gi , P gj Establish a linear fitting relationship; steam flow rate G s The expression is:

[0026]

[0027] In the formula, G sn and P Tn These are the main steam flow rate and main steam pressure under rated operating conditions, respectively; F() is the valve flow characteristic function;

[0028] For peak shaving operation points at different depths, piecewise linear fitting is performed to generate the valve opening P. GV Main steam pressure P T With steam flow rate G s The nonlinear correspondence between them is used to achieve nonlinear fitting; in the unit output range [P] gi , P gj Within [the specified range], the linear fitting expression for the valve flow characteristic function is:

[0029]

[0030] In the formula, F i( P GV ) represents the unit's output range [P] gi , P gj The valve flow characteristic function is defined within the range ], where a0 and a1 are fitting coefficients.

[0031] Furthermore, in step S6, based on the dynamic test data of unit speed disturbance under different peak-shaving operating conditions, and according to the regulating stage steam pressure and reheat steam pressure data measured by the dynamic test of unit speed disturbance under different peak-shaving operating conditions, the first-order inertial element is identified. The reheater volume time constant T rh Where s is a complex variable, P zr To regulate the steam pressure, P t The reheat steam pressure is then determined. Based on the valve opening, main steam pressure, and unit output mechanical power data obtained from dynamic tests of unit speed disturbance under different peak-shaving operating conditions, the high-pressure steam chamber volume time constant T is identified. ch And the natural over-adjustment coefficient λ of the high-pressure cylinder power.

[0032] Furthermore, in step S7, the goodness of fit R... 2 The expression is:

[0033]

[0034] In the formula, This is the average value of all measured output data.

[0035] Furthermore, the intelligent identification method is the Quantum Particle Swarm Optimization (QPSO) algorithm.

[0036] The present invention also provides a computer device, comprising: at least one processor, at least one memory, and computer program instructions stored in the memory, which implement the above-described method when executed by the processor.

[0037] The present invention also provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the above-described method.

[0038] Compared with existing technologies, this invention has the following advantages: First, it fully considers the nonlinearity of turbine valve flow characteristics under deep peak-shaving conditions, establishing a model that more closely reflects the actual operating state of the unit, overcoming the limitations of traditional linearized models over a wide load range. Second, it proposes a step-by-step, modular parameter identification process, combining direct calculation with intelligent optimization algorithms, starting with easier steps and gradually acquiring key parameters of the governor, actuator, and turbine body, improving the efficiency and systematic nature of the identification. Third, it uses the goodness of fit between simulation results and measured data as the core evaluation index to verify and validate the identified parameters, ensuring the accuracy and reliability of the final model parameters. Fourth, by performing parameter identification under different deep peak-shaving conditions, this invention can obtain multiple parameter sets applicable to typical and deep peak-shaving conditions, enabling the model to maintain high simulation accuracy across the entire load range, providing a reliable tool for refined power grid simulation analysis. Attached Figure Description

[0039] Figure 1 is a flowchart of the parameter identification and evaluation method for the speed regulation system of deep peak-shaving thermal power units provided in an embodiment of the present invention;

[0040] Figure 2 This is a structural diagram of the speed governor model in an embodiment of the present invention;

[0041] Figure 3 This is a structural diagram of the actuator model in an embodiment of the present invention;

[0042] Figure 4 This is a structural diagram of the steam turbine model in an embodiment of the present invention;

[0043] Figure 5 This is a flowchart of the QPSO algorithm in an embodiment of the present invention;

[0044] Figure 6 This is a flowchart illustrating the parameter identification process of the simulation model of the steam turbine and its speed control system in an embodiment of the present invention.

[0045] Figure 7 This is a comparison chart of the actual data smoothing and denoising results in the embodiments of the present invention;

[0046] Figure 8 The graph shows the comparison between simulation and measured data of the actuator model in the embodiment of the present invention; where (a) is the upper step of GV2, (b) is the lower step of GV2, (c) is the upper step of GV3, (d) is the lower step of GV3, (e) is the upper step of GV4, and (f) is the lower step of GV4.

[0047] Figure 9 As described in the embodiments of the present invention Comparison curves of simulation and measured data after identification; where (a) represents 90% power positive disturbance, (b) represents 90% power negative disturbance, (c) represents 60% power positive disturbance, (d) represents 60% power negative disturbance, (e) represents 40% power positive disturbance, and (f) represents 40% power negative disturbance.

[0048] Figure 10 As described in the embodiments of the present invention and The comparison curves of simulation and measured data after identification are shown; where (a) represents 90% power negative perturbation, (b) represents 90% power positive perturbation, (c) represents 60% power negative perturbation, (d) represents 60% power positive perturbation, (e) represents 40% power negative perturbation, and (f) represents 40% power positive perturbation. Detailed Implementation

[0049] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0050] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0051] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0052] This embodiment provides a method for parameter identification and evaluation of the speed control system of a deep peak-shaving thermal power unit based on measured data, including the following steps:

[0053] S1: Establish a simulation model of the steam turbine and its speed control system on the simulation platform. The simulation model includes a governor model, an actuator model, and a steam turbine model.

[0054] S2: Obtain the static and speed disturbance dynamic test data of the unit under the set operating conditions, perform preprocessing on the test data including denoising and smoothing, and normalize the preprocessed data to obtain the measured input and output data of the governor model, actuator model and turbine model.

[0055] S3: Based on the static test data of the unit, calculate and obtain the parameters of the limiting, dead zone, and delay links in the governor model and the actuator model, as well as the start and stop time constant of the hydraulic actuator in the actuator model;

[0056] S4: Based on the static test data of the unit, the PID parameters of the governor model and the actuator model are identified and obtained using an intelligent identification method.

[0057] S5: Based on the dynamic test data of unit speed disturbance under different peak shaving operation conditions, piecewise linear fitting is performed to obtain valve flow characteristics and establish the nonlinear relationship between valve opening, main steam pressure and steam flow.

[0058] S6: Based on the dynamic test data of unit speed disturbance and valve flow characteristics under different peak shaving operation conditions, an intelligent identification method is used to identify and obtain the relevant parameters of the turbine model;

[0059] S7: Substitute the identified parameters into the simulation model, run the simulation model to obtain the model output data; use the goodness of fit between the model output data and the measured output data as the evaluation index to evaluate the model parameter identification results, and establish a deep peak-shaving thermal power unit model suitable for power system simulation analysis.

[0060] 1. Simulation Model of Steam Turbine and its Speed ​​Control System

[0061] The turbine control system has gone through several stages: mechanical control system, hydraulic control system, and digital electro-hydraulic control system. The mechanical and hydraulic control systems input speed change signals caused by load disturbances into the governor, which are amplified by the spool valve actuator to control the opening of the regulating valve. The mathematical model of the turbine and speed control system consists of three parts: the governor, the actuator, and the turbine model.

[0062] 1.1 Speed ​​Regulator Model

[0063] The governor model's input signals are the power setpoint, speed setpoint, unit active power, and unit speed. These signals are processed by a PID (Proportional-Integral-Differential) controller as a correction unit, outputting the governor valve opening command. The governor model is as follows:Figure 2 As shown. Figure 2 In the diagram, control modes 1 and 2 represent valve control and power control, respectively; K1 is the speed deviation amplification factor. For load control feedforward coefficients, , , These represent the proportional, integral, and derivative factors of the PID controller, respectively. CV This is the valve opening command. For unit speed, P is the given value for the rotational speed. e P is the active power output of the generator unit. ref T1~T4 are the power setpoints and the time constants of the delay elements.

[0064] 1.2 Actuator Model

[0065] The actuator includes an electro-hydraulic converter and a hydraulic actuator. The actuator amplifies the control signal from the regulator and converts it into the valve opening degree, controlling the steam flow into the turbine. The input signal to the actuator model is the valve opening command, which, after passing through the controller and hydraulic actuator stages, outputs the valve opening degree. The actuator model is as follows: Figure 3 As shown.

[0066] Figure 3 middle, , , These represent the proportional, integral, and differential factors of the integrated amplification module. The integrated amplification module typically consists of a proportional or proportional-integral component. , To provide the upper and lower limits for the comprehensive amplification stage, , These are the overspeed opening and overspeed closing coefficients, respectively. , These are the opening and closing time constants of the hydraulic actuator, respectively. This is the time constant of the LVDT (Low Volume Flow Time Detector) for the hydraulic motor, typically taken as 0.02s.

[0067] 1.3 Steam Turbine Model

[0068] For intermediate reheat units, when the speed control system activates, such as when the control valve suddenly opens wide, the steam inlet pressure to the high-pressure cylinder rises rapidly. Due to the large intermediate volume, the exhaust pressure to the high-pressure cylinder can only rise slowly, causing a change in the steam pressure difference between the inlet and outlet of the high-pressure cylinder. This results in the output proportional coefficient of the high-pressure cylinder during the dynamic process being greater than the proportional coefficient during steady-state operation. To reflect this physical phenomenon, a natural overshoot coefficient for the high-pressure cylinder power is introduced. Steam turbine model, such as Figure 4As shown. The input to the turbine model is the main steam flow rate calculated based on the valve opening and main steam pressure, and the output is mechanical power. The turbine model considers the valve flow characteristics. Figure 4 middle, , , The power coefficients of the high, medium, and low-pressure cylinders of the steam turbine are given. ;P T Main steam pressure; G s T represents the steam flow rate; ch T rh T co These are the volumetric time constants for high-pressure steam, reheat steam, and low-pressure steam, respectively.

[0069] In the formula, G sn and P Tn These are the main steam flow rate and main steam pressure values ​​under rated operating conditions, respectively.

[0070] Based on dynamic test data of unit speed disturbance under different peak-shaving operating conditions, valve flow characteristics were obtained. Figure 4 The turbine model shown is given, with the unit output at different peak-shaving operating points being P. g1 P g2 ... P gn Select the unit output range [P] gi , P gj Establish a linear fitting relationship; steam flow rate G s The expression is:

[0071]

[0072] In the formula, G sn and P Tn These are the main steam flow rate and main steam pressure under rated operating conditions, respectively; F() is the valve flow characteristic function.

[0073] For peak shaving operation points at different depths, piecewise linear fitting is performed to generate the valve opening P. GV Main steam pressure P T With steam flow rate G s The nonlinear correspondence between them is used to achieve nonlinear fitting. Within the unit output range [P] gi , P gj Within [the specified range], the linear fitting expression for the valve flow characteristic function is:

[0074]

[0075] In the formula, F i( P GV) represents the unit's output range [P] gi , P gj The valve flow characteristic function is defined within the range ], where a0 and a1 are fitting coefficients.

[0076] 2 Intelligent Identification Method

[0077] 2.1 Quantum Particle Swarm Optimization Algorithm

[0078] The Quantum Behaved Particle Swarm Optimization (QPSO) algorithm incorporates quantum computing theory, breaking through the dependence of traditional particle swarm optimization models on particle velocity and position information. It uses wave functions to describe particle states and probability density functions to describe particle positions, which can appear at any position in the solution space with a certain probability. This gives the algorithm better global search capabilities and solves the problem that traditional particle swarm optimization algorithms are prone to getting trapped in local optima in high-dimensional searches. It is suitable for parameter identification of multi-dimensional nonlinear models such as steam turbines and their speed control systems.

[0079] The QPSO algorithm describes the state of a particle using a wave function, with the specific formula as follows:

[0080]

[0081] In the formula, For the first In the nth iteration, the 1st The 1st dimension in the search space The position of each particle; For the first The average optimal position of the population after several iterations; This is the attractor of the particle; The coefficient of contraction and expansion; for Random numbers between; For the first The optimal position of each individual particle; This represents the particle's globally optimal position. for Random numbers between; The total number of particles in the population; The dimension of the particle search space.

[0082] The specific implementation process of the QPSO algorithm is as follows: Figure 5 As shown.

[0083] 2.2 Parameter Identification Process Based on QPSO Algorithm

[0084] This method employs the QPSO algorithm as an intelligent identification method to identify relevant parameters of the simulation model of the steam turbine and its speed control system. The main steps include:

[0085] (1) Establish a simulation model of the steam turbine and its speed control system;

[0086] (2) Obtain static and speed disturbance dynamic test data of the unit under specific operating conditions, perform noise reduction, smoothing, resampling and other preprocessing on the test data, and normalize the preprocessed data to facilitate subsequent calculation and analysis;

[0087] (3) The parameters of the limiting, dead zone, and delay components in the governor model and actuator model are directly calculated from the static test data of the unit;

[0088] (4) Based on the large step input and output data of the static test of the actuator, the start-stop time constant of the hydraulic motor in the actuator model is directly calculated; based on the small step input and output data of the static test of the actuator, the PID parameters of the governor model and the actuator model are identified by the QPSO algorithm.

[0089] (5) Based on the dynamic test data of the unit under specific operating conditions, the nonlinear characteristic relationship of valve flow is obtained, and the high-pressure steam chamber volume time constant, reheater volume time constant, and high-pressure cylinder power natural over-adjustment coefficient are identified by the QPSO algorithm.

[0090] (6) Substitute all the identified parameters into the simulation model, and use the fit between the simulation curve and the measured data as the evaluation index to verify the effectiveness of the model and parameters.

[0091] The parameter identification process for steam turbines and their speed control systems is as follows: Figure 6 As shown.

[0092] For the input and output data of the governor model, actuator model, and turbine model in static and dynamic test data under specific operating conditions, a threshold denoising method is used to achieve data smoothing and denoising preprocessing. Since the quantization noise in the measured data is Gaussian white noise, a threshold denoising method is employed for data preprocessing. The effect of wavelet transform smoothing and denoising on the measured data is compared below. Figure 7 As shown.

[0093] 3. Parameter Identification Results

[0094] 3.1 Parameter Identification Results of the Actuator Model

[0095] (1) Calculation of the opening and closing time constant of the hydraulic actuator

[0096] The start-up and closing time constant of the hydraulic actuator is a key parameter for evaluating the dynamic characteristics of the turbine regulating system, directly affecting the speed and stability of the unit's load response. In this method, based on the input and output data of large-step full-open and large-step full-close disturbances of the actuator's regulating valve opening in the unit's static test data, the start-up and closing time constant of the hydraulic actuator in the actuator model is calculated. The formula for calculating the start-up and closing time constant of the hydraulic actuator is:

[0097]

[0098] In the formula, T is the start-stop time constant of the hydraulic actuator; P max and P min These represent the maximum and minimum values ​​of the valve opening, respectively; Δt is the time required for the valve to fully open or fully close during a large step; ΔP y This refers to the change in valve opening during a large step transition between full opening and full closing; when the hydraulic actuator is activated, ΔP y This represents the increase in valve opening, where T corresponds to the opening time constant of the hydraulic actuator; when the hydraulic actuator is closed, ΔP y This represents the reduction in valve opening, and T corresponds to the closing time constant of the hydraulic actuator. In this embodiment, the calculated opening time constant is taken from 25% to 75% of the opening, and the calculated closing time constant is taken from 75% to 25% of the opening.

[0099] The measured data comes from a 350MW thermal power unit, with the turbine being a supercritical single-stage reheat condensing turbine. Based on the static full-open and full-close test data of the unit's actuators, the opening and closing time constants of different control valves were calculated. Specifically, the opening time constant for GV1 is 0.4732s; for GV2, it is 0.4605s; for GV3, it is 0.4533s; and for GV4, it is 0.4457s. The average of these values ​​yields the unit's opening time constant. It is 0.4582s.

[0100] The shutdown time constants for GV1, GV2, GV3, and GV4 are 0.5074 s, 0.5098 s, 0.5060 s, and 0.4915 s, respectively. The average shutdown time constant is then calculated to obtain the unit's shutdown time constant. It is 0.5037s.

[0101] (2) Identification of PID parameters for actuator model

[0102] In this method, based on the input and output data of small step disturbances in the actuator valve opening from the unit's static test data, an intelligent identification method is used to identify the PID controller parameters of the governor model and the actuator model. In this embodiment, using a PI control loop for the actuator model yields better control performance, i.e., setting K... d =0, meaning only identification is performed. , parameter.

[0103] In this method, the mean square error R is introduced. MSE with goodness of fit As an indicator to judge the reliability of parameter identification, the corresponding calculation formula is:

[0104]

[0105]

[0106] In the formula, This represents the number of data points. For the first The measured output value of each data point; For the first The model simulation output value of each data point; y represents the average value of all measured output data. For the governor model, y refers to the gate opening command output by the governor model; for the actuator model, y refers to the gate opening output by the actuator model.

[0107] The objective function f of the intelligent identification method is to obtain the minimum mean square error R. MSE With the objective function f as the goal, the expression for the objective function f is:

[0108]

[0109] The objective function is set as minimizing the mean square error between the simulated and measured valve opening curves, and the contraction / expansion coefficient is used. A linear decreasing strategy is adopted to enhance the algorithm's global exploration capability in the early stages and strengthen its local fine-grained search capability in the later stages. The value of the algorithm decreases linearly from approximately 1.0 to 0.5 with the number of iterations, and the dynamic adjustment formula is as follows:

[0110]

[0111] Set particle swarm size The maximum number of iterations is 50. 100, parameter dimension Set the boundary conditions to 2; based on the turbine manufacturer's report and experience values, set the boundary conditions. , The number of calculations per generation was set to 20 to balance computational efficiency and search capability. Parameter identification was performed by adjusting the gates GV2, GV3, and GV4 by small steps of 8% up and down. The identification results are shown in Table 1. The simulation results and measured data after parameter identification are compared as follows: Figure 8 As shown in Table 1, the error analysis between the simulation results and the measured data is presented.

[0112] Table 1. PID parameter identification results for the actuator model

[0113]

[0114] Because the simulated curve of the regulating valve under step conditions exhibits a steady-state process, only the data segment from 1.5 to 5 seconds is used to calculate R when calculating the error index between the simulated and measured curves of each regulating valve opening under step conditions. MSE .

[0115] like Figure 8 As shown, under static testing, the operating characteristics of each valve tend to be consistent when subjected to a step change in the same direction, but differ when subjected to step changes in different directions. Therefore, the obtained parameters are only applicable to single-valve operation and not to sequential valve operation.

[0116] 3.2 Turbine parameter identification results

[0117] In this method, based on the dynamic test data of unit speed disturbance under different peak-shaving operating conditions, an intelligent identification method is used to identify the reheater volume time constant T. rh The time constant T of the steam chamber before the high-pressure cylinder of the steam turbine ch And the natural over-adjustment coefficient λ of the high-pressure cylinder power.

[0118] (1) Based on the dynamic test data of regulating stage steam pressure and reheat steam pressure obtained from the unit speed disturbance under different peak shaving operation conditions, identify the first-order inertial element. The reheater volume time constant T rh Where s is a complex variable, P zr To regulate the steam pressure, P t This is the reheat steam pressure.

[0119] (2) Based on the valve opening, main steam pressure and unit output mechanical power data obtained from the dynamic test of unit speed disturbance under different peak shaving operation conditions, the high-pressure steam chamber volume time constant T is identified. ch And the natural overshoot coefficient λ of the high-pressure cylinder power.

[0120] Tests were conducted at 90%, 60%, and 40% of the unit's rated load, with primary frequency regulation engaged, and positive and negative speed step disturbance tests were performed. According to the unit's design parameters, the power ratio coefficients for the high-pressure, intermediate-pressure, and low-pressure cylinders were 0.3022, 0.3447, and 0.3531, respectively. Typically, the low-pressure connecting pipe volume time constant... With reheater volume time constant The difference is relatively small, so the model of medium and low pressure cylinders combined is adopted in the stability calculation, that is, the power ratio coefficient of the medium pressure cylinder is taken. Therefore, the required identification parameter for the steam turbine is the high-pressure steam chamber volume time constant. Reheater volume time constant High-pressure cylinder power natural overshoot coefficient Three parameters.

[0121] 1) Reheater volumetric time constant Identify

[0122] Set particle swarm size The maximum number of iterations is 50. 100, parameter dimension Set the boundary conditions to 1; based on the turbine manufacturer's report and experience values. The number of calculations per generation was set to 20. Based on the dynamic test data of the regulating stage steam pressure and reheat steam pressure, parameter identification was performed under operating conditions of 90%, 60%, and 40% rated power, with a speed disturbance of ±6 r / min. The results for each operating condition are shown in Table 2. In Table 2, "+6" indicates a positive speed step disturbance, and "-6" indicates a negative speed step disturbance. In Table 3, 1.37e-6 represents 1.37 × 10⁻⁶. -6 .

[0123] Table 2 Identification results

[0124]

[0125] Table 3 Error Analysis of Simulation and Measured Data

[0126]

[0127] Depend on Figure 9 As shown in Table 2, the volumetric time constant of the turbine reheater is... There was no significant difference at different peak-shaving depths, so the average value of the identification results was uniformly selected in the subsequent simulation model parameter settings. s.

[0128] 2) and Identify

[0129] Set particle swarm size The maximum number of iterations is 100. 100, parameter dimension Set the value to 2; set the reheater volume time constant for the corresponding operating condition according to Table 2; set the boundary conditions based on the turbine manufacturer's report and empirical values. , The number of calculations per generation was set to 20. Based on the data of the main control valve opening, main steam pressure, and mechanical power obtained from dynamic tests, parameter identification was performed under the conditions of 90%, 60%, and 40% rated power, and speed disturbance ±6 r / min. The identification results are shown in Table 4. The simulation results after parameter identification were compared with the measured data. Figure 10 As shown in Table 5, the error analysis between the simulation results and the measured data is presented.

[0130] Table 4 and Identification results

[0131]

[0132] Table 5 and Error Analysis of Simulation and Measured Data

[0133]

[0134] Depend on Figure 10 As shown in Table 4, the identified parameters are basically consistent under the operating conditions of 90% and 60% rated power; however, the identified parameters under the operating condition of 40% rated power are significantly different from those under the first two conditions. The value increased significantly. The values ​​decreased slightly. Parameters obtained under different disturbance directions at the same unit output exhibit slight differences within the error range. In practical engineering, operating conditions with 40% or less of the unit's rated power are typically referred to as deep peak-shaving conditions. The identification results indicate that under deep peak-shaving conditions, the high-pressure steam chamber volume time constant... and the natural overshoot coefficient of high-pressure cylinder power There is a certain deviation from typical operating conditions. Therefore, in the subsequent simulation model parameter settings, the average value of the identification results was selected under operating conditions of 90% rated power and 60% rated power. s、 As the setpoint under typical operating conditions; the average value of the identification results was selected under 40% rated power operating conditions. s、 This is the set value under deep peak shaving conditions.

[0135] The parameter identification method based on QPSO yields some differences in parameter results under different peak-shaving conditions. (The text also mentions the high-pressure steam chamber volume time constant in the turbine stage.) There are slight differences within the allowable range under 90% and 60% output conditions, and the value increases significantly under 40% output conditions; the high-pressure cylinder power natural over-adjustment coefficient There was no significant difference between the 90% and 60% output conditions, but the value decreased slightly at the 40% output condition.

[0136] This embodiment also provides a computer device, including: at least one processor, at least one memory, and computer program instructions stored in the memory, which implement the above-described method when executed by the processor.

[0137] This embodiment also provides a computer-readable storage medium storing computer program instructions that, when executed by a processor, implement the above-described method.

[0138] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0139] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0140] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0141] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0142] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for parameter identification and evaluation of speed control system of deep peak-shaving thermal power unit, characterized in that, Includes the following steps: S1: Establish a simulation model of the steam turbine and its speed control system on the simulation platform. The simulation model includes a governor model, an actuator model, and a steam turbine model. S2: Obtain the static and speed disturbance dynamic test data of the unit under the set operating conditions, perform preprocessing on the test data including denoising and smoothing, and normalize the preprocessed data to obtain the measured input and output data of the governor model, actuator model and turbine model. S3: Based on the static test data of the unit, calculate and obtain the parameters of the limiting, dead zone, and delay links in the governor model and the actuator model, as well as the start and stop time constant of the hydraulic actuator in the actuator model; S4: Based on the static test data of the unit, the PID parameters of the governor model and the actuator model are identified and obtained using an intelligent identification method. S5: Based on the dynamic test data of unit speed disturbance under different peak shaving operation conditions, piecewise linear fitting is performed to obtain valve flow characteristics and establish the nonlinear relationship between valve opening, main steam pressure and steam flow. S6: Based on the dynamic test data of unit speed disturbance and valve flow characteristics under different peak shaving operation conditions, an intelligent identification method is used to identify and obtain the relevant parameters of the turbine model; S7: Substitute the identified parameters into the simulation model, run the simulation model to obtain the model output data; The goodness of fit between the model output data and the measured output data is used as the evaluation index to evaluate the model parameter identification results, and a deep peak-shaving thermal power unit model suitable for power system simulation analysis is established.

2. The method for parameter identification and evaluation of a deep peak-shaving thermal power unit speed control system according to claim 1, characterized in that, In step S1, the inputs to the speed governor model are the power setpoint, the speed setpoint, the active power of the unit, and the unit speed. After passing through the power feedforward and controller control loops, the output is the valve opening command. The input to the actuator model is the valve opening command, which, after passing through the controller and hydraulic actuator, outputs the valve opening. The input to the turbine model is the main steam flow rate calculated based on the valve opening and main steam pressure, and the output is mechanical power. The turbine model takes into account the valve flow characteristics.

3. The method for parameter identification and evaluation of a deep peak-shaving thermal power unit speed control system according to claim 1, characterized in that, In step S3, based on the input and output data of the large-step fully open and large-step fully closed disturbances of the actuator valve opening in the unit static test data, the start-up and closing time constant of the hydraulic actuator in the actuator model is calculated; the formula for calculating the start-up and closing time constant of the hydraulic actuator is: In the formula, T is the start-stop time constant of the hydraulic actuator; P max and P min These represent the maximum and minimum values ​​of the valve opening, respectively; Δt is the time required for the valve to fully open or fully close during a large step; ΔP y This refers to the change in valve opening during a large step transition between full opening and full closing; when the hydraulic actuator is activated, ΔP y This represents the increase in valve opening, where T corresponds to the opening time constant of the hydraulic actuator; when the hydraulic actuator is closed, ΔP y This represents the amount of decrease in the valve opening, and T corresponds to the closing time constant of the hydraulic actuator.

4. The method for parameter identification and evaluation of a deep peak-shaving thermal power unit speed control system according to claim 1, characterized in that, In step S4, based on the input and output data of the small step disturbance of the actuator valve opening in the static test data of the unit, an intelligent identification method is used to identify the PID controller parameters of the governor model and the actuator model; the objective function f of the intelligent identification method is to obtain the minimum mean square error R. MSE With the objective function f as the goal, the expression for the objective function f is: In the formula, n is the total number of data points used in the objective function calculation; y i and y represents the i-th measured output data and the model output data, respectively; for the speed governor model, y refers to the gate opening command output by the speed governor model; for the actuator model, y refers to the gate opening output by the actuator model.

5. The method for parameter identification and evaluation of a deep peak-shaving thermal power unit speed control system according to claim 1, characterized in that, In step S5, based on the dynamic test data of unit speed disturbance under different peak-shaving operation conditions, the valve flow characteristics are obtained, specifically including: For the turbine model, let the unit output at different peak-shaving operating points be P. g1 P g2 ... P gn Select the unit output range [P] gi ,P gj Establish a linear fitting relationship; steam flow rate G s The expression is: In the formula, G sn and P Tn These are the main steam flow rate and main steam pressure under rated operating conditions, respectively; F() is the valve flow characteristic function; For peak shaving operation points at different depths, piecewise linear fitting is performed to generate the valve opening P. GV Main steam pressure P T With steam flow rate G s The nonlinear correspondence between them is used to achieve nonlinear fitting; in the unit output range [P] gi ,P gj Within [the specified range], the linear fitting expression for the valve flow characteristic function is: In the formula, F i( P GV ) represents the unit's output range [P] gi ,P gj The valve flow characteristic function is defined within the range ], where a0 and a1 are fitting coefficients.

6. The method for parameter identification and evaluation of a deep peak-shaving thermal power unit speed control system according to claim 1, characterized in that, In step S6, based on the dynamic test data of unit speed disturbance under different peak-shaving operating conditions, and according to the regulating stage steam pressure and reheat steam pressure data measured by the dynamic test of unit speed disturbance under different peak-shaving operating conditions, the first-order inertial element is identified. The reheater volume time constant T rh Where s is a complex variable, P zr To regulate the steam pressure, P t The reheat steam pressure is then determined. Based on the valve opening, main steam pressure, and unit output mechanical power data obtained from dynamic tests of unit speed disturbance under different peak-shaving operating conditions, the high-pressure steam chamber volume time constant T is identified. ch And the natural over-adjustment coefficient λ of the high-pressure cylinder power.

7. The method for parameter identification and evaluation of a deep peak-shaving thermal power unit speed control system according to claim 1, characterized in that, In step S7, the goodness of fit R 2 The expression is: In the formula, This is the average value of all measured output data.

8. The method for parameter identification and evaluation of a deep peak-shaving thermal power unit speed control system according to claim 1, characterized in that, The intelligent identification method is the Quantum Particle Swarm Optimization (QPSO) algorithm.

9. A computer device, characterized in that, include: At least one processor, at least one memory, and computer program instructions stored in the memory, which, when executed by the processor, implement the method as described in any one of claims 1-8.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by a processor, the method described in any one of claims 1-8 is implemented.

Images