A method, device, equipment, medium and product for optimizing operation of thermal power units

Abstract

The present application discloses an operation optimization method, device, equipment, medium and product for a thermal power unit, which relates to the field of electrical digital data processing technology. The method includes obtaining a target digital twin model of a flue gas waste heat cascade utilization system of a thermal power unit and boundary parameters of the target digital twin model; obtaining multiple adjustment parameters in the target digital twin model; determining the model key parameters from the multiple adjustment parameters; optimizing the model key parameters based on the boundary parameters to obtain the optimal key parameters corresponding to the model key parameters; applying the optimal key parameters to the target equipment corresponding to the optimal key parameters, so that the flue gas waste heat cascade utilization system of the thermal power unit operates based on the optimal key parameters; wherein the target equipment is the equipment in the flue gas waste heat cascade utilization system of the thermal power unit, and the present application can realize real-time and accurate optimization operation of the equipment of the thermal power unit.

Description

Technical Field

[0001] The present application relates to the technical field of electrical digital data processing, and in particular to an operation optimization method, device, equipment, medium and product for a thermal power unit. Background Art

[0002] In the daily operation and management of thermal power plants, improving energy efficiency, reducing emissions, and lowering equipment downtime and operating costs have become key goals for the industry. Improving energy efficiency helps alleviate energy shortages and reduce reliance on limited energy resources; while reducing equipment downtime and lowering operating costs are directly related to the economic benefits and market competitiveness of power plants.

[0003] However, traditional optimization methods currently used for thermal power plants have numerous limitations. These methods largely rely on manual experience and scheduled maintenance strategies, lacking precise control over the equipment's real-time operating status. In actual operation, equipment operating conditions are complex and volatile, and relying solely on experience makes it difficult to adapt to these dynamic changes. Consequently, adjustments often lag behind actual needs, making it impossible to achieve real-time and precise optimization of thermal power plant equipment. Summary of the Invention

[0004] The purpose of this application is to provide a method, device, equipment, medium and product for optimizing the operation of a thermal power unit, which can realize real-time and accurate optimization operation of the equipment of the thermal power unit.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides an operation optimization method for a thermal power unit, comprising:

[0007] Obtaining a target digital twin model of a flue gas waste heat cascade utilization system of a thermal power unit and boundary parameters of the target digital twin model;

[0008] Obtaining multiple adjustment parameters in the target digital twin model;

[0009] determining a model key parameter from the plurality of adjustment parameters;

[0010] Optimizing the model key parameters based on the boundary parameters to obtain optimal key parameters corresponding to the model key parameters;

[0011] The optimal key parameters are applied to the target equipment corresponding to the optimal key parameters so that the flue gas waste heat cascade utilization system of the thermal power unit operates based on the optimal key parameters; wherein the target equipment is the equipment in the flue gas waste heat cascade utilization system of the thermal power unit.

[0012] Optionally, the operation optimization method of the thermal power unit further includes:

[0013] A digital twin model is constructed based on the flue gas waste heat cascade utilization system of the thermal power unit; wherein the digital twin model includes at least a boiler sub-model, a waste heat sub-model, and a turbine sub-model;

[0014] At preset time intervals, real-time data is collected from the flue gas waste heat cascade utilization system, and the real-time data is injected into the digital twin model, so that the digital twin model performs real-time dynamic simulation based on the real-time data;

[0015] Furthermore, the method of obtaining the target digital twin model of the flue gas waste heat cascade utilization system of the thermal power unit and the boundary parameters of the target digital twin model is specifically as follows:

[0016] The digital twin model of the flue gas waste heat cascade utilization system of the thermal power unit into which the latest real-time data is injected is determined as the target digital twin model;

[0017] Obtain boundary parameters corresponding to the target digital twin model.

[0018] Optionally, the adjustment parameters include: the opening of the first air preheater bypass flue gas damper, the opening of the second air preheater bypass flue gas damper, the circulating water pump frequency, the opening of the main valve from the water supply to the high-pressure economizer, the opening of the first branch valve from the water supply to the high-pressure economizer, the opening of the second branch valve from the water supply to the high-pressure economizer, the opening of the booster pump from the condensate to the low-pressure economizer, the opening of the bypass valve from the condensate to the low-pressure economizer, the opening of the valve from the condensate to the first low-pressure economizer, and the opening of the valve from the condensate to the second low-pressure economizer; and determining the key model parameters from the multiple adjustment parameters specifically includes:

[0019] Obtain multiple sets of training adjustment parameter groups; wherein each set of training adjustment parameter groups includes all adjustment parameters, and the value of each adjustment parameter is a value randomly obtained within the value range corresponding to the current adjustment parameter;

[0020] Obtain the actual coal consumption of each set of training adjustment parameters in actual operation;

[0021] Performing simulation calculations on each set of training adjustment parameter groups using the target digital twin model to obtain simulated coal consumption for each set of training adjustment parameter groups;

[0022] Using the actual coal consumption and the simulated coal consumption, the key parameters of the model are determined from the multiple adjustment parameters; wherein, the key parameters of the model are the opening of the first air preheater bypass flue gas damper, the opening of the second air preheater bypass flue gas damper, the circulating water pump frequency, and the opening of the first sub-valve of the feed water to the high-pressure economizer.

[0023] Optionally, the using the actual coal consumption and the simulated coal consumption to determine the key model parameters from the multiple adjustment parameters specifically includes:

[0024] Compare the actual coal consumption of each training adjustment parameter group with the simulated coal consumption to obtain the coal consumption residual value of each training adjustment parameter group;

[0025] Construct a residual table based on the actual coal consumption, simulated coal consumption and coal consumption residual values ​​of each training adjustment parameter group;

[0026] Analyzing the residual table to determine the parameter sensitivity of each adjustment parameter;

[0027] The adjustment parameters whose parameter sensitivity is greater than a preset threshold are determined as the key parameters of the model.

[0028] Optionally, optimizing the model key parameters based on the boundary parameters to obtain optimal key parameters corresponding to the model key parameters specifically includes:

[0029] determining key boundary parameters corresponding to the model key parameters from the boundary parameters;

[0030] Determining an initial key parameter corresponding to the model key parameter based on the key boundary parameter; wherein the initial key parameter is less than or equal to the key boundary parameter;

[0031] Calculating based on the initial key parameters and the preset initial Lagrangian multiplier and initial penalty factor to obtain the current minimum value of the augmented Lagrangian function;

[0032] Updating the initial key parameter based on the current minimum value to obtain a target key parameter;

[0033] Based on the initial Lagrangian multiplier, the initial penalty factor, and the key boundary parameter, the initial Lagrangian multiplier is updated to obtain a target Lagrangian multiplier;

[0034] Updating the initial penalty factor based on preset parameters and the initial penalty factor to obtain a target penalty factor;

[0035] The optimization operation is repeatedly performed based on the target key parameter, the target Lagrange multiplier, and the target penalty factor until the difference between the current minimum value and the previous minimum value corresponding to the current minimum value is less than or equal to a preset difference, or the target key parameter is the same as the key boundary parameter.

[0036] Optionally, the optimization operation specifically includes:

[0037] Performing calculation based on the target key parameter, the target Lagrangian multiplier, and the target penalty factor to obtain a current minimum value of the augmented Lagrangian function;

[0038] If the difference between the current minimum value and the previous minimum value corresponding to the current minimum value is greater than a preset difference, the target key parameter is updated based on the current minimum value to obtain an updated target key parameter; wherein the target key parameter used in subsequent calculations is the updated target key parameter;

[0039] Based on the target Lagrangian multiplier, the target penalty factor, and the key boundary parameter, the target Lagrangian multiplier is updated to obtain an updated target Lagrangian multiplier; wherein the target Lagrangian multiplier used in subsequent calculations is the updated target Lagrangian multiplier;

[0040] The target penalty factor is updated based on the preset parameter and the target penalty factor to obtain an updated target penalty factor; wherein the target penalty factor used in subsequent calculations is the updated target penalty factor.

[0041] In a second aspect, the present application provides an operation optimization device for a thermal power unit, comprising:

[0042] A first acquisition unit is used to acquire a target digital twin model of a flue gas waste heat cascade utilization system of a thermal power unit and boundary parameters of the target digital twin model;

[0043] A second acquisition unit, configured to acquire a plurality of adjustment parameters in the target digital twin model;

[0044] a determining unit, configured to determine a key model parameter from the plurality of adjustment parameters;

[0045] an optimization unit, configured to optimize the key parameters of the model based on the boundary parameters to obtain optimal key parameters corresponding to the key parameters of the model;

[0046] An application unit is used to apply the optimal key parameters to the target equipment corresponding to the optimal key parameters, so that the flue gas waste heat cascade utilization system of the thermal power unit operates based on the optimal key parameters; wherein, the target equipment is the equipment in the flue gas waste heat cascade utilization system of the thermal power unit.

[0047] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the operation optimization method of a thermal power unit described in any one of the above.

[0048] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the operation optimization method of any one of the above-mentioned thermal power units.

[0049] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the operation optimization method of any one of the thermal power units described above.

[0050] In a sixth aspect, the present application provides a chip, which includes a processor and a communication interface, the communication interface and the processor are coupled, the processor is used to run a program or instruction, and when the processor executes the program or instruction, it implements the steps of the operation optimization method of any one of the above-mentioned thermal power units.

[0051] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0052] The present application provides an operation optimization method, device, equipment, medium and product for a thermal power unit. By building a target digital twin model of the flue gas waste heat cascade utilization system of the thermal power unit, real-time simulation processing can be performed on the data collected from the flue gas waste heat cascade utilization system of the thermal power unit, so that the target digital twin model can reflect the operating condition changes of the equipment in the flue gas waste heat cascade utilization system in real time. Then, the target digital twin model can be used to perform optimization calculations based on the collected data to obtain the optimal key parameters for adjusting the operation of the flue gas waste heat cascade utilization system, thereby realizing real-time and accurate optimization operations on the equipment of the thermal power unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0054] Figure 1 This is a flow chart of an operation optimization method for a thermal power unit according to an embodiment of the present application;

[0055] Figure 2 This is a schematic diagram of the functional modules of a target digital twin model of a flue gas waste heat cascade utilization system for a thermal power unit provided in one embodiment of the present application;

[0056] Figure 3 A schematic diagram of a tornado with adjustment parameters provided in an embodiment of the present application;

[0057] Figure 4A schematic diagram of an AUGLAG algorithm optimization result provided in one embodiment of the present application;

[0058] Figure 5 A schematic diagram of a PSO algorithm optimization result provided in one embodiment of the present application;

[0059] Figure 6 A schematic diagram of COBYLA algorithm optimization results provided in one embodiment of the present application;

[0060] Figure 7 A schematic diagram of a GA algorithm optimization result provided in one embodiment of the present application;

[0061] Figure 8 A schematic diagram of the functional modules of an operation optimization device for a thermal power unit provided in one embodiment of the present application;

[0062] Figure 9 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0064] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0065] In an exemplary embodiment, Figure 1 As shown, a method for optimizing the operation of a thermal power unit is provided. The method is executed by a computer device, specifically, a computer device such as a terminal or a server, or a terminal and a server. In the embodiment of the present application, the method includes the following steps 101 to 105. Among them:

[0066] Step 101: Obtain a target digital twin model of a flue gas waste heat cascade utilization system of a thermal power unit and boundary parameters of the target digital twin model.

[0067] As an optional implementation, before step 101, the following steps may be further performed:

[0068] A digital twin model is constructed based on the flue gas waste heat cascade utilization system of the thermal power unit; wherein the digital twin model includes at least a boiler sub-model, a waste heat sub-model, and a steam turbine sub-model;

[0069] At preset time intervals, real-time data is collected from the flue gas waste heat cascade utilization system, and the real-time data is injected into the digital twin model, so that the digital twin model performs real-time dynamic simulation based on the real-time data;

[0070] Furthermore, the method of obtaining the target digital twin model of the flue gas waste heat cascade utilization system of the thermal power unit and the boundary parameters of the target digital twin model in step 101 may specifically be:

[0071] The digital twin model of the flue gas waste heat cascade utilization system of the thermal power unit into which the latest real-time data is injected is determined as the target digital twin model;

[0072] Get the boundary parameters corresponding to the target digital twin model.

[0073] Among them, this implementation method is to build a digital twin model containing boiler, waste heat and turbine sub-models based on the flue gas waste heat cascade utilization system of the thermal power unit, providing an accurate virtual mapping for the operation of power plant equipment. By collecting and injecting real-time data at preset time intervals for real-time dynamic simulation, the digital twin model can closely follow the changes in actual equipment operating conditions, greatly improving the real-time and accuracy of the model. The model injected with the latest real-time data is determined as the target digital twin model and the corresponding boundary parameters are obtained to ensure that subsequent optimization operations are based on the most realistic equipment status. This not only helps to timely discover potential equipment problems, but also can formulate more effective optimization strategies based on accurate data, thereby significantly improving the energy utilization efficiency of thermal power units, reducing energy consumption and operating costs, while reducing pollutant emissions, enhancing equipment operation stability, and effectively promoting the green, efficient and sustainable development of thermal power units.

[0074] In the embodiment of the present application, the digital twin model of the flue gas waste heat cascade utilization system of the thermal power unit can be constructed based on Sysplorer software and using the Modelica modeling language.

[0075] In addition, the boiler subsystem can also include equipment models such as furnace, water wall, superheater, and reheater;

[0076] The waste heat subsystem can also include equipment models such as air preheater, low-temperature economizer, air heater, and expansion tank;

[0077] The steam turbine subsystem can also include equipment models such as steam turbine, condenser, high and low pressure heaters, and steam cooler.

[0078] Furthermore, the boiler sub-model, waste heat sub-model and turbine sub-model may also include equipment-level models such as furnace, water-cooled wall, steam turbine, economizer, air preheater, superheater and reheater.

[0079] In the embodiment of the present application, the latest real-time data can be the real-time operating data of the flue gas waste heat cascade utilization system of the thermal power unit; Python can be used to write a communication component, and the COM (Component Object Model) interface provided by the power plant industrial data management platform (Plant Information System, PI) can be called through the communication component to realize the reading and writing of the real-time operating data in the PI database.

[0080] Among them, real-time operating data may include the amount of coal entering the furnace, air volume, turbine side feed water flow, valve opening, pump frequency, etc.

[0081] In an embodiment of the present application, the target digital twin model can perform simulation operations based on the latest injected real-time data, and obtain simulation operation results, and then the simulation operation results can be output to a display so that the staff can intuitively see the simulation operation results of the target digital twin model on the real flue gas waste heat cascade utilization system. The simulation operation results can be text-type information, image-type information, or statistical chart-type information, which is not limited in this embodiment of the present application. The simulation operation results can include current coal consumption information, energy utilization rate information, etc., which is not limited in this embodiment of the present application.

[0082] In an embodiment of the present application, the boundary parameters corresponding to the target digital twin model may be constraints on the operating parameters in the target digital twin model.

[0083] For example, see Table 1:

[0084] Table 1 Boundary parameters corresponding to the target digital twin model

[0085]

[0086] Please also refer to Figure 2 , Figure 2 A schematic diagram of the functional modules of a target digital twin model of a flue gas waste heat cascade utilization system for a thermal power unit provided in one embodiment of the present application.

[0087] Step 102: Obtain multiple adjustment parameters in the target digital twin model.

[0088] In the embodiment of the present application, the adjustment parameters include: the opening of the first air preheater bypass flue gas damper, the opening of the second air preheater bypass flue gas damper, the circulating water pump frequency, the opening of the main valve from water supply to high-pressure economizer, the opening of the first branch valve from water supply to high-pressure economizer, the opening of the second branch valve from water supply to high-pressure economizer, the opening of the booster pump from condensate to low-pressure economizer, the opening of the bypass valve from condensate to low-pressure economizer, the opening of the valve from condensate to the first low-pressure economizer, and the opening of the valve from condensate to the second low-pressure economizer.

[0089] Step 103: Determine key model parameters from the multiple adjustment parameters.

[0090] In an embodiment of the present application, the key parameters of the model can be determined based on the degree of influence of each adjustment parameter on the coal consumption of the target digital twin model. The key parameters of the model are adjustment parameters that have a greater degree of influence on the coal consumption of the target digital twin model.

[0091] As an optional implementation, the method of determining the key model parameters from the multiple adjustment parameters in step 103 may be:

[0092] Obtain multiple sets of training adjustment parameter groups; wherein each set of training adjustment parameter groups includes all adjustment parameters, and the value of each adjustment parameter is a value randomly obtained within the value range corresponding to the current adjustment parameter;

[0093] Obtain the actual coal consumption of each set of training adjustment parameters in actual operation;

[0094] Through the target digital twin model, each set of training adjustment parameter groups is simulated and calculated to obtain the simulated coal consumption of each set of training adjustment parameter groups;

[0095] Using the actual coal consumption and the simulated coal consumption, the key parameters of the model are determined from the multiple adjustment parameters; among them, the key parameters of the model are the opening of the first air preheater bypass flue gas damper, the opening of the second air preheater bypass flue gas damper, the circulating water pump frequency and the opening of the first sub-valve of the water supply to the high-pressure economizer.

[0096] Among them, implementing this implementation method, by obtaining multiple sets of training adjustment parameter groups randomly generated within the adjustment parameter value range, and respectively obtaining their actual operating coal consumption and simulated coal consumption obtained by simulation calculation using the target digital twin model, provides rich data support for a comprehensive and objective analysis of the impact of adjustment parameters on coal consumption. Based on the actual coal consumption and simulated coal consumption, key model parameters such as the opening of the first air preheater bypass flue gas damper, the opening of the second air preheater bypass flue gas damper, the circulating water pump frequency, and the opening of the first branch valve of the water supply to the high-pressure economizer are determined, which can accurately locate the parameters with the greatest impact on coal consumption. This enables the thermal power unit to make targeted adjustments to these key model parameters during the subsequent operation optimization process, avoiding blind attempts, greatly improving optimization efficiency, and helping to quickly achieve the goals of reducing coal consumption and improving energy utilization efficiency. At the same time, it reduces unnecessary resource waste and enhances the economy and stability of the thermal power unit operation.

[0097] In an embodiment of the present application, multiple groups of training adjustment parameter groups can be generated through a Monte Carlo algorithm.

[0098] Furthermore, the Monte Carlo algorithm can be used to analyze the impact of parameter changes on model output.

[0099] In Monte Carlo, we usually focus on the statistical properties of a target variable (such as expected value, variance, etc.) obtained through simulation. The basic evaluation formula is as follows:

[0100] Monte Carlo simulation is a numerical computation method based on random sampling, which is used to estimate the performance, results or certain characteristics of a complex system or model by simulating a large number of random samples.

[0101] Suppose we have a function Calculate the simulated coal consumption, the input variables are the training adjustment parameters , the input variable is from a probability distribution The goal of the Monte Carlo method is to estimate the expected value of the function, that is:

[0102]

[0103] in, is the expected value, is the input variable The probability density function of .

[0104] Monte Carlo estimation formula: If it is difficult or impossible to solve this integral directly, the expected value can be approximated by random sampling:

[0105]

[0106] in: is the number of random samples, From the distribution Randomly selected sample points.

[0107] Variance estimation: Assume that the function Perform Monte Carlo simulation to obtain a set of estimated values , the variance formula of Monte Carlo estimation is:

[0108]

[0109] Specifically, the derivation process of the variance formula of Monte Carlo estimation is:

[0110] Known independent and identically distributed samples , the Monte Carlo estimator is the sample mean:

[0111]

[0112] The Monte Carlo estimator is the sample mean, and its variance formula is:

[0113]

[0114] The single sample variance can be decomposed into the difference between the second moment and the square of the first moment:

[0115]

[0116] Use the sample mean instead of the theoretical expectation:

[0117] Second moment estimate:

[0118]

[0119] First moment squared estimate:

[0120]

[0121] Substitute the two into the variance decomposition formula:

[0122]

[0123] therefore,

[0124]

[0125] In the embodiment of the present application, if the variance of the samples is relatively large, the accuracy of the estimation is low, and more samples are needed to improve the accuracy of the estimation.

[0126] Error estimation: A key issue in Monte Carlo methods is determining the error of the estimate. is an estimate of the expected value, then the error can be calculated by the following formula:

[0127]

[0128] Since the error of Monte Carlo estimation is usually proportional to the number of samples The standard deviation of the error is inversely proportional to the square root of , so it is:

[0129]

[0130] In order to achieve a lower error at a given accuracy, the number of simulations needs to be increased. .

[0131] Convergence in Monte Carlo simulation: As the number of samples increases As the error increases, the Monte Carlo estimation result will get closer and closer to the true expected value. It has the following convergence characteristics:

[0132]

[0133] Monte Carlo simulation confidence interval: Monte Carlo can also be used to construct the confidence interval of the expected value. Assume that the simulation obtains indivual The confidence interval of the expected value can be estimated based on the sample mean and standard deviation. For example, the 95% confidence interval can be given by the following formula:

[0134]

[0135] Among them, 1.96 is the critical value of the normal distribution at the 95% confidence level.

[0136] As an optional implementation manner, using the actual coal consumption and the simulated coal consumption, a method for determining the key model parameters from the multiple adjustment parameters may specifically be:

[0137] Compare the actual coal consumption of each training adjustment parameter group with the simulated coal consumption to obtain the coal consumption residual value of each training adjustment parameter group;

[0138] Construct a residual table based on the actual coal consumption, simulated coal consumption and coal consumption residual values ​​of each training adjustment parameter group;

[0139] Analyze the residual table to determine the parameter sensitivity of each adjustment parameter;

[0140] The adjustment parameter whose sensitivity is greater than a preset threshold is determined as the key parameter of the model.

[0141] Among them, by implementing this implementation method, the residual value is obtained by comparing the actual and simulated coal consumption, a clear residual table is constructed, the sensitivity of each adjustment parameter is accurately analyzed, and the parameters with sensitivity greater than the preset threshold are determined as key parameters, which helps operating personnel to accurately control and avoid blind operation, improve optimization efficiency, optimize the digital twin model, and enhance model reliability and practicality.

[0142] For example, the residual table can be shown in Table 2 below:

[0143] Table 2 Residual table

[0144]

[0145] In addition, a scatter plot can be constructed based on the actual coal consumption, simulated coal consumption and coal consumption residual values ​​of each training adjustment parameter group. The horizontal axis of the scatter plot can be the measured value and the vertical axis can be the simulated value. Ideally, the data points should be distributed near the 45° line.

[0146] It is also possible to construct a residual distribution graph based on the actual coal consumption, simulated coal consumption and coal consumption residual values ​​of each training adjustment parameter group. The horizontal axis of the residual distribution graph can be the training adjustment parameter value (such as the baffle opening), and the vertical axis can be the residual, which is used to analyze the impact of the training adjustment parameters on the error.

[0147] Optionally, parameter sensitivity can also include multiple sensitivity indicators, and a tornado diagram of the adjustment parameters can be generated based on the multiple sensitivity indicators. Figure 3 , Figure 3 A tornado diagram of adjustment parameters provided for an embodiment of the present application; wherein an adjustment parameter includes seven sensitivity indicators: correlation-linear, correlation-ranking, correlation-Kendall, standardized regression-linear, standardized regression-ranking, partial correlation-linear, and partial correlation-ranking.

[0148] Specifically, Correlation-Linear is used to show the linear correlation between the adjustment parameters and the output.

[0149] Correlation-ranking is used to show the degree of correlation between the adjustment parameters and the output and sort them by the degree of influence.

[0150] Correlation-Kendall refers to measuring the correlation between adjustment parameters using Kendall's τ coefficient, and displaying the analysis results in a tornado chart. Kendall's τ coefficient is a statistical indicator used to measure the rank correlation between two adjustment parameters.

[0151] Standardized regression-linear involves the application of standardized regression coefficients in linear regression models. The use of standardized regression coefficients in tornado plots primarily allows for comparison of the relative impact of different independent variables on the dependent variable. Because different independent variables may have different units and magnitudes, directly comparing raw regression coefficients can be misleading. Standardization brings all independent variables into the same scale, and the absolute value of their standardized regression coefficients directly reflects the independent variable's impact on the dependent variable. Larger absolute values ​​indicate a greater impact.

[0152] Standardized regression-sorting is used to analyze and display the degree of influence of multiple independent variables on the dependent variable. The independent variables obtained through standardized regression are arranged in descending order according to the degree of their influence on the dependent variable.

[0153] Partial correlation-linearity refers to the analysis of the impact of multiple adjustment parameters on a single outcome, taking into account the partial correlations between the adjustment parameters and performing the analysis based on the linear assumption. Under this assumption, the strength of the linear relationship between the adjustment parameters is measured by calculating the partial correlation coefficient. The partial correlation coefficient has the same range as the ordinary correlation coefficient, between -1 and 1. The closer the absolute value of the coefficient is to 1, the stronger the linear relationship between the two adjustment parameters; the closer it is to 0, the weaker the linear relationship.

[0154] Partial correlation-sorting is used to analyze the relationship between multiple independent variables and dependent variables, and arrange the partial correlation coefficients of each independent variable and dependent variable in descending order according to the size of their absolute values.

[0155] Therefore, it can be obtained from Figure 3 It can be seen that the opening of the first air preheater bypass flue gas damper, the opening of the second air preheater bypass flue gas damper, the circulating water pump frequency and the opening of the first sub-valve from water supply to high-pressure economizer have a greater impact on coal consumption. Therefore, the opening of the first air preheater bypass flue gas damper, the opening of the second air preheater bypass flue gas damper, the circulating water pump frequency and the opening of the first sub-valve from water supply to high-pressure economizer can be determined as the key parameters of the model.

[0156] Specifically, corresponding index values ​​may be determined for each of the seven sensitivity indexes, and an adjustment parameter for each sensitivity index that is greater than a preset threshold value of the sensitivity index may be determined as a key parameter of the model.

[0157] Step 104 : Optimize the model key parameters based on the boundary parameters to obtain optimal key parameters corresponding to the model key parameters.

[0158] In the embodiment of the present application, the augmented Lagrangian algorithm (AUGLAG) may be used to optimize key parameters of the model.

[0159] As an optional implementation manner, step 104 optimizes the key parameters of the model based on the boundary parameters to obtain the optimal key parameters corresponding to the key parameters of the model by:

[0160] determining, from the boundary parameters, key boundary parameters corresponding to the key parameters of the model;

[0161] Determining an initial key parameter corresponding to the model key parameter based on the key boundary parameter; wherein the initial key parameter is less than or equal to the key boundary parameter;

[0162] Calculating based on the initial key parameters and the preset initial Lagrangian multiplier and initial penalty factor to obtain the current minimum value of the augmented Lagrangian function;

[0163] The initial key parameter is updated based on the current minimum value to obtain the target key parameter;

[0164] Based on the initial Lagrangian multiplier, the initial penalty factor, and the key boundary parameter, the initial Lagrangian multiplier is updated to obtain a target Lagrangian multiplier;

[0165] The initial penalty factor is updated based on the preset parameters and the initial penalty factor to obtain a target penalty factor;

[0166] The optimization operation is repeatedly performed based on the target key parameter, the target Lagrange multiplier, and the target penalty factor until the difference between the current minimum value and the previous minimum value corresponding to the current minimum value is less than or equal to a preset difference, or the target key parameter is the same as the key boundary parameter.

[0167] Among them, implementing this implementation method, determining the key boundary parameters from the boundary parameters and obtaining the initial key parameters based on them, provides a reasonable starting point for the optimization process and ensures that the optimization operation is carried out within the equipment safety and operating limits. By calculating the current minimum value of the augmented Lagrangian function based on the initial key parameters, initial Lagrangian multipliers and initial penalty factors, and continuously updating the key parameters, Lagrangian multipliers and penalty factors, the preset optimization target can be gradually approached. In the process of multiple iterative optimizations, until the difference between the current minimum value and the previous minimum value is less than or equal to the preset difference or the target key parameters are the same as the key boundary parameters, this makes the optimization results more accurate and more in line with actual needs.

[0168] In the embodiment of the present application, a constrained optimization problem can be constructed based on the key boundary parameters in the form of:

[0169]

[0170]

[0171] Among them, x is the initial key parameter, is the objective function of the simulated coal consumption, is an inequality constraint. It is constructed by key boundary parameters, namely =x-key boundary parameter, it can be seen that Should be less than or equal to 0. m represents the number of model key parameters included in the initial key parameters.

[0172] In the embodiment of the present application, the augmented Lagrangian function The form is:

[0173]

[0174] in: is the initial or target penalty factor (usually a positive number). is the initial Lagrange multiplier or the target Lagrange multiplier.

[0175] In the embodiment of the present application, the optimization operation specifically includes:

[0176] Calculating based on the target key parameter, the target Lagrangian multiplier, and the target penalty factor to obtain a current minimum value of the augmented Lagrangian function;

[0177] If the difference between the current minimum value and the previous minimum value corresponding to the current minimum value is greater than a preset difference, the target key parameter is updated based on the current minimum value to obtain an updated target key parameter; wherein the target key parameter used in subsequent calculations is the updated target key parameter;

[0178] Based on the target Lagrangian multiplier, the target penalty factor, and the key boundary parameter, the target Lagrangian multiplier is updated to obtain an updated target Lagrangian multiplier; wherein the target Lagrangian multiplier used in subsequent calculations is the updated target Lagrangian multiplier;

[0179] The target penalty factor is updated based on the preset parameter and the target penalty factor to obtain an updated target penalty factor; wherein the target penalty factor used in subsequent calculations is the updated target penalty factor.

[0180] This implementation method uses the target key parameters, target Lagrangian multipliers, and target penalty factors as the basis for calculations, continuously searching for the current minimum of the augmented Lagrangian function. At each iteration, the difference between the current minimum and the previous minimum is compared to determine whether to continue optimization. If the difference is greater than a preset value, the target key parameters, target Lagrangian multipliers, and target penalty factors are updated, ensuring that each iteration progresses towards a more optimal solution. This step-by-step iterative optimization process precisely adjusts the model's key parameters to better meet actual operational requirements.

[0181] In the embodiment of the present application, in each iteration, after solving the minimum value of the augmented Lagrangian function, the target key parameter is updated in the form of:

[0182]

[0183] The updated Lagrange multiplier is:

[0184]

[0185] The updated penalty factor is in the form of:

[0186]

[0187] In the embodiment of the present application, the parameter comparison before and after optimization and the optimization strategy recommendation can also be displayed on the Web.

[0188] In addition, coal consumption can also be optimized using constrained optimization by linear approximations (COBYLA), particle swarm optimization (PSO), and genetic algorithms (GA).

[0189] The COBYLA algorithm handles constraints through linear approximation and gradually approaches the optimal solution through iterative optimization. At each iteration, the algorithm constructs a linear model based on the current solution, searching for a better solution, ultimately approaching the global optimal solution. It is an effective constrained optimization algorithm applicable to a wide range of complex problems. By properly setting constraints, initial values, and parameters, it can efficiently solve practical problems.

[0190] The PSO algorithm is an optimization technique that simulates the foraging behavior of bird flocks. In PSO, each solution is considered a "particle" with a position and velocity. Particles update their velocity and position by sharing information to find the optimal solution. PSO is applicable to a variety of complex optimization problems. By properly setting parameters, designing fitness functions, and handling constraints, it can effectively solve practical problems.

[0191] The GA algorithm is a global optimization algorithm based on natural selection and genetics. Potential solutions to a problem are represented as individuals in a population, encoded in genes. The initial population is typically randomly generated, and the population is iteratively updated to approximate the optimal solution. Each individual's fitness, typically calculated by an objective function, measures its ability to solve the problem. Parent individuals are selected based on their fitness, and crossover and mutation operations are performed to generate new offspring. By continuously iterating these operations, the GA algorithm can effectively solve complex optimization problems.

[0192] The AUGLAG algorithm is an effective method for solving constrained optimization problems. It combines the advantages of the Lagrange multiplier method and the penalty function method. By introducing Lagrange multipliers and penalty parameters, it transforms the constrained optimization problem into a series of unconstrained optimization problems. Its basic concept is to construct an augmented Lagrangian function, combining the objective function, a linear combination of constraints, and a penalty term. The optimal solution of this function is then iterated to gradually approach the optimal solution of the original problem. With appropriate parameter settings and convergence conditions, the AUGLAG algorithm has been widely used to solve various complex optimization problems.

[0193] Please also refer to Figures 4 to 7 , Figure 4 A schematic diagram of an AUGLAG algorithm optimization result provided in one embodiment of the present application; Figure 5 A schematic diagram of a PSO algorithm optimization result provided in one embodiment of the present application; Figure 6 A schematic diagram of COBYLA algorithm optimization results provided in one embodiment of the present application; Figure 7 A schematic diagram of a GA algorithm optimization result provided in one embodiment of the present application.

[0194] Specifically, after optimization using the COBYLA algorithm, the coal consumption value was 275.28g, an improvement of approximately 0.5g. The optimization process was completed in 15 steps and convergence was achieved. The key control parameters after optimization were: 75% air preheater flue gas side profile opening, 94% total feedwater to high-pressure economizer total regulating valve opening, 88% condensate to low-pressure economizer booster pump frequency, and 48.63Hz circulating water pump frequency for the heater and low-temperature economizer.

[0195] After final optimization using the PSO algorithm, the coal consumption reached 275.3g, an optimized coal consumption of approximately 0.5g. Convergence was achieved after 10 optimization steps. The optimized control parameters included a 91% profile opening on the flue gas side of the air preheater, an 82.5% total throttle valve opening on the feedwater to high-pressure economizer, a 55.2% booster pump frequency on the condensate to low-pressure economizer, and a 48.5Hz circulating water pump frequency on the heater and low-temperature economizer.

[0196] After final optimization using the GA algorithm, the coal consumption value reached 275.3g, an optimized coal consumption of approximately 0.5g. Convergence was achieved after six optimization steps. The optimized control parameters were: the air preheater flue gas side profile opening was 72.4%, the feedwater to high-pressure economizer total valve opening was 82.3%, the condensate to low-pressure economizer booster pump frequency was 68%, and the heater and low-temperature economizer circulating water pump frequency was 47Hz.

[0197] After final optimization using the AUGLAG algorithm, the coal consumption reached 275.34g, with an optimized coal consumption of approximately 0.46g. Convergence was achieved after 20 optimization steps. The optimized control parameters included a 49% air preheater flue gas side profile opening, a 76% total feedwater to high-pressure economizer valve opening, a 77.5% booster pump frequency for the condensate to low-pressure economizer, and a 47Hz circulating water pump frequency for the heater and low-temperature economizer.

[0198] By applying four algorithms under certain constraints, with minimizing coal consumption as the optimization objective, the results showed that the optimized coal consumption remained around 0.5g. The booster pump for condensate to the low-pressure economizer had a high frequency, while the circulating water pump had a low frequency. During the optimization process, although the PSO and GA algorithms had fewer iterations, they took the longest time. The AUGLAG algorithm, on the other hand, performed best in terms of time consumption and had a more moderate number of steps. Therefore, the AUGLAG algorithm is more suitable for implementing online optimization of the later-stage twin model.

[0199] Step 105 : Apply the optimal key parameters to the target equipment corresponding to the optimal key parameters, so that the flue gas waste heat cascade utilization system of the thermal power unit operates based on the optimal key parameters.

[0200] In the embodiment of the present application, the target device is a device in the flue gas waste heat cascade utilization system of the thermal power unit.

[0201] Optionally, a device performance degradation curve may be constructed based on the latest real-time data collected.

[0202] Real-time acquisition of various operating parameters of the system, such as temperature, pressure, and flow.

[0203] Next, the collected operating parameters are cleaned to ensure their accuracy and consistency.

[0204] One of the three algorithms, COBYLA, PSO, and GA, is selected for parameter estimation to ensure that the simulation values ​​are highly consistent with the measured values.

[0205] During system operation, data is continuously collected and model parameters are updated to achieve online calibration of the model. Table 3 below shows the calibration parameter variables and corresponding parameter names corresponding to the calibration and performance degradation modules.

[0206] Table 3 Performance degradation module parameters

[0207]

[0208] In addition, the digital twin model will call parameter estimation and add the most recent optimization result as a new record to the MySQL database. At the same time, the device performance degradation curve will also be automatically added to represent the current calibration result. The device performance degradation curve can be output to the web for display.

[0209] Implementing the above-mentioned steps 101 to 105 can achieve real-time and accurate optimization operations on the equipment of the thermal power unit. In addition, the present application can also ensure that subsequent optimization operations are based on the most realistic equipment status. In addition, the present application can also make targeted adjustments to these key parameters of the model to avoid blind attempts, greatly improving the optimization efficiency, helping to quickly achieve the goals of reducing coal consumption and improving energy utilization efficiency, while reducing unnecessary waste of resources and enhancing the economy and stability of the operation of the thermal power unit. In addition, the present application can also help operating personnel to accurately control, avoid blind operations, improve optimization efficiency, optimize the digital twin model, and enhance model reliability and practicality. In addition, the present application can also make the optimization results more accurate and more in line with actual needs. In addition, the present application can also accurately adjust the key parameters of the model to make it more in line with actual operation needs.

[0210] Based on the same inventive concept, embodiments of the present application also provide a thermal power unit operation optimization device for implementing the aforementioned thermal power unit operation optimization method. The implementation solution provided by this device is similar to the implementation solution described in the aforementioned method. Therefore, the specific limitations of the one or more thermal power unit operation optimization device embodiments provided below can be found in the above-mentioned limitations of the thermal power unit operation optimization method and will not be further elaborated here.

[0211] In an exemplary embodiment, Figure 8 As shown, an operation optimization device for a thermal power unit is provided, comprising:

[0212] The first acquisition unit 801 is used to obtain a target digital twin model of the flue gas waste heat cascade utilization system of the thermal power unit and boundary parameters of the target digital twin model;

[0213] A second acquisition unit 802 is used to acquire multiple adjustment parameters in the target digital twin model;

[0214] In the embodiment of the present application, the adjustment parameters include: the opening of the first air preheater bypass flue gas damper, the opening of the second air preheater bypass flue gas damper, the circulating water pump frequency, the opening of the main valve from water supply to high-pressure economizer, the opening of the first branch valve from water supply to high-pressure economizer, the opening of the second branch valve from water supply to high-pressure economizer, the opening of the booster pump from condensate to low-pressure economizer, the opening of the bypass valve from condensate to low-pressure economizer, the opening of the valve from condensate to the first low-pressure economizer, and the opening of the valve from condensate to the second low-pressure economizer.

[0215] A determining unit 803 is configured to determine a key model parameter from the plurality of adjustment parameters;

[0216] An optimization unit 804 is configured to optimize the key parameters of the model based on the boundary parameters to obtain optimal key parameters corresponding to the key parameters of the model;

[0217] Application unit 805 is used to apply the optimal key parameters to the target device corresponding to the optimal key parameters, so that the flue gas waste heat cascade utilization system of the thermal power unit operates based on the optimal key parameters; wherein, the target device is the device in the flue gas waste heat cascade utilization system of the thermal power unit.

[0218] As an optional implementation manner, the first acquiring unit 801 is further configured to:

[0219] A digital twin model is constructed based on the flue gas waste heat cascade utilization system of the thermal power unit; wherein the digital twin model includes at least a boiler sub-model, a waste heat sub-model, and a steam turbine sub-model;

[0220] At preset time intervals, real-time data is collected from the flue gas waste heat cascade utilization system, and the real-time data is injected into the digital twin model, so that the digital twin model performs real-time dynamic simulation based on the real-time data;

[0221] Furthermore, the first acquisition unit 801 acquires the target digital twin model of the flue gas waste heat cascade utilization system of the thermal power unit and the boundary parameters of the target digital twin model in the following manner:

[0222] The digital twin model of the flue gas waste heat cascade utilization system of the thermal power unit into which the latest real-time data is injected is determined as the target digital twin model;

[0223] Get the boundary parameters corresponding to the target digital twin model.

[0224] Among them, this implementation method is to build a digital twin model containing boiler, waste heat and turbine sub-models based on the flue gas waste heat cascade utilization system of the thermal power unit, providing an accurate virtual mapping for the operation of power plant equipment. By collecting and injecting real-time data at preset time intervals for real-time dynamic simulation, the digital twin model can closely follow the changes in actual equipment operating conditions, greatly improving the real-time and accuracy of the model. The model injected with the latest real-time data is determined as the target digital twin model and the corresponding boundary parameters are obtained to ensure that subsequent optimization operations are based on the most realistic equipment status. This not only helps to timely discover potential equipment problems, but also can formulate more effective optimization strategies based on accurate data, thereby significantly improving the energy utilization efficiency of thermal power units, reducing energy consumption and operating costs, while reducing pollutant emissions, enhancing equipment operation stability, and effectively promoting the green, efficient and sustainable development of thermal power units.

[0225] As an optional implementation manner, the determining unit 803 may determine the key model parameters from the multiple adjustment parameters in the following manner:

[0226] Obtain multiple sets of training adjustment parameter groups; wherein each set of training adjustment parameter groups includes all adjustment parameters, and the value of each adjustment parameter is a value randomly obtained within the value range corresponding to the current adjustment parameter;

[0227] Obtain the actual coal consumption of each set of training adjustment parameters in actual operation;

[0228] Through the target digital twin model, each set of training adjustment parameter groups is simulated and calculated to obtain the simulated coal consumption of each set of training adjustment parameter groups;

[0229] Using the actual coal consumption and the simulated coal consumption, the key parameters of the model are determined from the multiple adjustment parameters; among them, the key parameters of the model are the opening of the first air preheater bypass flue gas damper, the opening of the second air preheater bypass flue gas damper, the circulating water pump frequency and the opening of the first sub-valve of the water supply to the high-pressure economizer.

[0230] Among them, the implementation of this embodiment provides rich data support for a comprehensive and objective analysis of the impact of the adjustment parameters on coal consumption by obtaining multiple sets of training adjustment parameter groups randomly generated within the adjustment parameter value range, and obtaining their actual operating coal consumption and simulated coal consumption obtained by simulation calculation using the target digital twin model. Based on the actual coal consumption and simulated coal consumption, key model parameters such as the opening of the first air preheater bypass flue gas damper, the opening of the second air preheater bypass flue gas damper, the circulating water pump frequency, and the opening of the first branch valve of the water supply to the high-pressure economizer are determined, which can accurately locate the parameters with the greatest impact on coal consumption. This enables the thermal power unit to make targeted adjustments to these key parameters during the subsequent operation optimization process, avoiding blind attempts, greatly improving optimization efficiency, and helping to quickly achieve the goals of reducing coal consumption and improving energy utilization efficiency. At the same time, it reduces unnecessary resource waste and enhances the economy and stability of the thermal power unit operation.

[0231] As an optional implementation manner, the determining unit 803 uses the actual coal consumption and the simulated coal consumption to determine the key model parameters from the multiple adjustment parameters in the following manner:

[0232] Compare the actual coal consumption of each training adjustment parameter group with the simulated coal consumption to obtain the coal consumption residual value of each training adjustment parameter group;

[0233] Construct a residual table based on the actual coal consumption, simulated coal consumption and coal consumption residual values ​​of each training adjustment parameter group;

[0234] Analyze the residual table to determine the parameter sensitivity of each adjustment parameter;

[0235] The adjustment parameter whose sensitivity is greater than a preset threshold is determined as the key parameter of the model.

[0236] Among them, by implementing this implementation method, the residual value is obtained by comparing the actual and simulated coal consumption, a clear residual table is constructed, the sensitivity of each adjustment parameter is accurately analyzed, and the parameters with sensitivity greater than the preset threshold are determined as key parameters, which helps operating personnel to accurately control and avoid blind operation, improve optimization efficiency, optimize the digital twin model, and enhance model reliability and practicality.

[0237] As an optional implementation manner, the optimization unit 804 optimizes the model key parameters based on the boundary parameters to obtain the optimal key parameters corresponding to the model key parameters in the following manner:

[0238] determining, from the boundary parameters, key boundary parameters corresponding to the key parameters of the model;

[0239] Determining an initial key parameter corresponding to the model key parameter based on the key boundary parameter; wherein the initial key parameter is less than or equal to the key boundary parameter;

[0240] Calculating based on the initial key parameters and the preset initial Lagrangian multiplier and initial penalty factor to obtain the current minimum value of the augmented Lagrangian function;

[0241] The initial key parameter is updated based on the current minimum value to obtain the target key parameter;

[0242] Based on the initial Lagrangian multiplier, the initial penalty factor, and the key boundary parameter, the initial Lagrangian multiplier is updated to obtain a target Lagrangian multiplier;

[0243] The initial penalty factor is updated based on the preset parameters and the initial penalty factor to obtain a target penalty factor;

[0244] The optimization operation is repeatedly performed based on the target key parameter, the target Lagrange multiplier, and the target penalty factor until the difference between the current minimum value and the previous minimum value corresponding to the current minimum value is less than or equal to a preset difference, or the target key parameter is the same as the key boundary parameter.

[0245] Among them, implementing this implementation method, determining the key boundary parameters from the boundary parameters and obtaining the initial key parameters based on them, provides a reasonable starting point for the optimization process and ensures that the optimization operation is carried out within the equipment safety and operating limits. By calculating the current minimum value of the augmented Lagrangian function based on the initial key parameters, initial Lagrangian multipliers and initial penalty factors, and continuously updating the key parameters, Lagrangian multipliers and penalty factors, the preset optimization target can be gradually approached. In the process of multiple iterative optimizations, until the difference between the current minimum value and the previous minimum value is less than or equal to the preset difference or the target key parameters are the same as the key boundary parameters, this makes the optimization results more accurate and more in line with actual needs.

[0246] In the embodiment of the present application, the optimization operation may specifically be:

[0247] Calculating based on the target key parameter, the target Lagrangian multiplier, and the target penalty factor to obtain a current minimum value of the augmented Lagrangian function;

[0248] If the difference between the current minimum value and the previous minimum value corresponding to the current minimum value is greater than a preset difference, the target key parameter is updated based on the current minimum value to obtain an updated target key parameter; wherein the target key parameter used in subsequent calculations is the updated target key parameter;

[0249] Based on the target Lagrangian multiplier, the target penalty factor, and the key boundary parameter, the target Lagrangian multiplier is updated to obtain an updated target Lagrangian multiplier; wherein the target Lagrangian multiplier used in subsequent calculations is the updated target Lagrangian multiplier;

[0250] The target penalty factor is updated based on the preset parameter and the target penalty factor to obtain an updated target penalty factor; wherein the target penalty factor used in subsequent calculations is the updated target penalty factor.

[0251] This implementation method uses the target key parameters, target Lagrangian multipliers, and target penalty factors as the basis for calculations, continuously searching for the current minimum of the augmented Lagrangian function. At each iteration, the difference between the current minimum and the previous minimum is compared to determine whether to continue optimization. If the difference is greater than a preset value, the target key parameters, target Lagrangian multipliers, and target penalty factors are updated, ensuring that each iteration progresses towards a more optimal solution. This step-by-step iterative optimization process precisely adjusts the model's key parameters to better meet actual operational requirements.

[0252] The implementation of the above-mentioned embodiment can realize real-time and accurate optimization operation of the equipment of the thermal power unit. In addition, the present application can also ensure that subsequent optimization operations are based on the most realistic equipment status. In addition, the present application can also make targeted adjustments to these key parameters of the model to avoid blind attempts, greatly improving the optimization efficiency, helping to quickly achieve the goals of reducing coal consumption and improving energy utilization efficiency, while reducing unnecessary waste of resources and enhancing the economy and stability of the operation of thermal power units. In addition, the present application can also help operating personnel to accurately control, avoid blind operations, improve optimization efficiency, optimize digital twin models, and enhance model reliability and practicality. In addition, the present application can also make the optimization results more accurate and more in line with actual needs. In addition, the present application can also accurately adjust the key parameters of the model to make it more in line with actual operation needs.

[0253] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 9As shown. The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, memory and I / O interface are connected via a system bus, and the communication interface is connected to the system bus via the I / O interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store operation optimization data of a thermal power unit. The I / O interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method for optimizing the operation of a thermal power unit is implemented.

[0254] Those skilled in the art will understand that Figure 9 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0255] In an exemplary embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0256] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0257] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0258] In an exemplary embodiment, a chip is provided, which includes a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the steps in the above-mentioned method embodiments and achieve the same technical effects. To avoid repetition, they will not be described here.

[0259] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0260] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0261] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0262] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0263] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0264] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of this application. In summary, the content of this specification should not be construed as limiting this application.

Claims

1. A method for optimizing the operation of a thermal power unit, characterized in that: The operation optimization method of the thermal power unit comprises: Obtaining a target digital twin model of a flue gas waste heat cascade utilization system of a thermal power unit and boundary parameters of the target digital twin model; Obtaining multiple adjustment parameters in the target digital twin model; determining a model key parameter from the plurality of adjustment parameters; Optimizing the model key parameters based on the boundary parameters to obtain optimal key parameters corresponding to the model key parameters; Applying the optimal key parameters to the target equipment corresponding to the optimal key parameters, so that the flue gas waste heat cascade utilization system of the thermal power unit operates based on the optimal key parameters; wherein the target equipment is equipment in the flue gas waste heat cascade utilization system of the thermal power unit; The adjustment parameters include: the opening of the first air preheater bypass flue gas damper, the opening of the second air preheater bypass flue gas damper, the circulating water pump frequency, the opening of the main valve from the water supply to the high-pressure economizer, the opening of the first branch valve from the water supply to the high-pressure economizer, the opening of the second branch valve from the water supply to the high-pressure economizer, the opening of the booster pump from the condensate to the low-pressure economizer, the opening of the bypass valve from the condensate to the low-pressure economizer, the opening of the valve from the condensate to the first low-pressure economizer, and the opening of the valve from the condensate to the second low-pressure economizer; Determining the key model parameters from the multiple adjustment parameters specifically includes: Obtain multiple sets of training adjustment parameter groups; wherein each set of training adjustment parameter groups contains all the adjustment parameters, and the value of each adjustment parameter is a value randomly obtained within the value range corresponding to the current adjustment parameter; obtain the actual coal consumption of each set of training adjustment parameter groups in actual operation; perform simulation calculations on each set of training adjustment parameter groups through the target digital twin model to obtain the simulated coal consumption of each set of training adjustment parameter groups; use the actual coal consumption and the simulated coal consumption to determine the model key parameters from the multiple adjustment parameters; wherein the model key parameters are the opening of the first air preheater bypass flue gas damper, the opening of the second air preheater bypass flue gas damper, the circulating water pump frequency, and the opening of the first sub-valve of the feed water to the high-pressure economizer; The use of the actual coal consumption and the simulated coal consumption to determine the key parameters of the model from the multiple adjustment parameters specifically includes: comparing the actual coal consumption of each training adjustment parameter group with the simulated coal consumption to obtain the coal consumption residual value of each training adjustment parameter group; constructing a residual table based on the actual coal consumption, simulated coal consumption and coal consumption residual value of each training adjustment parameter group; analyzing the residual table to determine the parameter sensitivity of each adjustment parameter; and determining the adjustment parameter whose parameter sensitivity is greater than a preset threshold as the key parameter of the model.

2. The operation optimization method of a thermal power unit according to claim 1, characterized in that: The operation optimization method of the thermal power unit further includes: A digital twin model is constructed based on the flue gas waste heat cascade utilization system of the thermal power unit; wherein the digital twin model includes at least a boiler sub-model, a waste heat sub-model, and a turbine sub-model; At preset time intervals, real-time data is collected from the flue gas waste heat cascade utilization system, and the real-time data is injected into the digital twin model, so that the digital twin model performs real-time dynamic simulation based on the real-time data; Furthermore, the method of obtaining the target digital twin model of the flue gas waste heat cascade utilization system of the thermal power unit and the boundary parameters of the target digital twin model is specifically as follows: The digital twin model of the flue gas waste heat cascade utilization system of the thermal power unit into which the latest real-time data is injected is determined as the target digital twin model; Obtain boundary parameters corresponding to the target digital twin model.

3. The operation optimization method of a thermal power unit according to claim 1, characterized in that: Optimizing the model key parameters based on the boundary parameters to obtain optimal key parameters corresponding to the model key parameters specifically includes: determining key boundary parameters corresponding to the model key parameters from the boundary parameters; Determining an initial key parameter corresponding to the model key parameter based on the key boundary parameter; wherein the initial key parameter is less than or equal to the key boundary parameter; Calculating based on the initial key parameters and the preset initial Lagrangian multiplier and initial penalty factor to obtain the current minimum value of the augmented Lagrangian function; Updating the initial key parameter based on the current minimum value to obtain a target key parameter; Based on the initial Lagrangian multiplier, the initial penalty factor, and the key boundary parameter, the initial Lagrangian multiplier is updated to obtain a target Lagrangian multiplier; Updating the initial penalty factor based on preset parameters and the initial penalty factor to obtain a target penalty factor; The optimization operation is repeatedly performed based on the target key parameter, the target Lagrange multiplier, and the target penalty factor until the difference between the current minimum value and the previous minimum value corresponding to the current minimum value is less than or equal to a preset difference, or the target key parameter is the same as the key boundary parameter.

4. The operation optimization method of a thermal power unit according to claim 3, characterized in that: The optimization operation specifically includes: Performing calculation based on the target key parameter, the target Lagrangian multiplier, and the target penalty factor to obtain a current minimum value of the augmented Lagrangian function; If the difference between the current minimum value and the previous minimum value corresponding to the current minimum value is greater than a preset difference, the target key parameter is updated based on the current minimum value to obtain an updated target key parameter; wherein the target key parameter used in subsequent calculations is the updated target key parameter; Based on the target Lagrangian multiplier, the target penalty factor, and the key boundary parameter, the target Lagrangian multiplier is updated to obtain an updated target Lagrangian multiplier; wherein the target Lagrangian multiplier used in subsequent calculations is the updated target Lagrangian multiplier; The target penalty factor is updated based on the preset parameter and the target penalty factor to obtain an updated target penalty factor; wherein the target penalty factor used in subsequent calculations is the updated target penalty factor.

5. An operation optimization device for a thermal power unit, characterized in that: For executing the operation optimization method according to any one of claims 1 to 4, the operation optimization device of the thermal power unit comprises: A first acquisition unit is used to acquire a target digital twin model of a flue gas waste heat cascade utilization system of a thermal power unit and boundary parameters of the target digital twin model; A second acquisition unit, configured to acquire a plurality of adjustment parameters in the target digital twin model; a determining unit, configured to determine a key model parameter from the plurality of adjustment parameters; an optimization unit, configured to optimize the key parameters of the model based on the boundary parameters to obtain optimal key parameters corresponding to the key parameters of the model; An application unit is used to apply the optimal key parameters to the target equipment corresponding to the optimal key parameters, so that the flue gas waste heat cascade utilization system of the thermal power unit operates based on the optimal key parameters; wherein, the target equipment is the equipment in the flue gas waste heat cascade utilization system of the thermal power unit.

6. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for optimizing the operation of a thermal power unit according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the operation optimization method of a thermal power unit according to any one of claims 1 to 4 are implemented.

8. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the operation optimization method of a thermal power unit according to any one of claims 1 to 4 are implemented.

Images