Steam turbine operation optimization control method and system based on dynamic sliding pressure curve

Abstract

The invention relates to a steam turbine operation optimization control method and system based on a dynamic sliding pressure curve. The optimization control method comprises the steps that real-time operation data of a steam turbine unit are collected; based on the real-time operation data and a pre-stored test database, a backpressure-load-main steam flow three-dimensional mapping model is established, and an initial sliding pressure curve parameter set is output; correcting the initial sliding pressure curve parameter set based on the environment temperature compensation factor to generate a compensated sliding pressure curve parameter set; generating an optimized sliding pressure curve function in combination with a predefined sliding pressure curve equation and a unit design constraint condition; target main steam pressure is calculated, and a steam turbine control valve control instruction is generated; and based on a comparison result of the actual power supply coal consumption data and a preset coal consumption threshold value, optimizing and updating the sliding pressure curve parameter set, and applying the sliding pressure curve parameter set to reconstruction of the three-dimensional mapping model. The main steam pressure of the steam turbine can be accurately adjusted, and the purposes of reducing coal consumption and improving efficiency are achieved.

Description

Technical Field

[0001] This application relates to the fields of automation control and energy-saving optimization technology, and in particular to a method and system for optimizing the operation of steam turbines based on dynamic sliding pressure curves. Background Technology

[0002] With the increasing demands of power systems for the flexibility and economy of thermal power units, how to achieve efficient and stable operation of units while meeting the needs of grid dispatch has become a research hotspot. As one of the three main components of a thermal power plant, the operating efficiency of the steam turbine directly affects the energy consumption level and economic benefits of the entire power plant.

[0003] Currently, traditional sliding pressure operation of steam turbines often uses a fixed sliding pressure curve to set the main steam pressure. While this method can adapt to the effects of load changes to some extent, it often struggles to balance the safety and economy of the unit under complex operating conditions, especially when faced with drastic fluctuations in the external environment. Particularly in high-temperature summer or low-temperature winter environments, changes in cooling water temperature can cause significant shifts in condenser back pressure, thereby affecting the work capacity of the low-pressure cylinder and the overall thermal efficiency of the turbine. Continuing to use the original static sliding pressure strategy in such situations can easily lead to increased inlet steam throttling losses, frequent operation of high-pressure regulating valves, and even the risk of surge, severely restricting the stability and economy of the unit's operation. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a method and system for optimizing turbine operation control based on dynamic sliding pressure curves.

[0005] Firstly, this application provides a turbine operation optimization control method based on dynamic sliding pressure curves, employing the following technical solution: A turbine operation optimization control method based on dynamic sliding pressure curve, the optimization control method comprising: Collect real-time operating data of the steam turbine unit, including current unit load, ambient temperature, and main steam flow; Based on the real-time operating data and the pre-stored test database, a three-dimensional mapping model of back pressure-load-main steam flow is established, and the initial sliding pressure curve parameter set is output. An ambient temperature compensation factor is obtained based on the ambient temperature, and the initial sliding pressure curve parameter set is corrected based on the ambient temperature compensation factor to generate a compensated sliding pressure curve parameter set. Based on the compensated sliding pressure curve parameter set, the predefined sliding pressure curve equation, and the unit design constraints, an optimized sliding pressure curve function is generated. The target main steam pressure is calculated based on the current unit load and the optimized sliding pressure curve function, and a turbine control valve command is generated based on the target main steam pressure. Obtain actual coal consumption data for power supply, and based on the comparison results between the actual coal consumption data for power supply and the preset coal consumption threshold, optimize and update the sliding pressure curve parameter set, output the updated sliding pressure curve parameter set, and apply it to the reconstruction of the three-dimensional mapping model.

[0006] By adopting the above technical solution, a sliding pressure curve model that reflects the current operating environment and equipment characteristics is constructed and continuously updated, thereby achieving precise regulation of the main steam pressure of the steam turbine, reducing coal consumption and improving efficiency. This technical solution can not only capture internal and external disturbance information of the unit at any time and respond promptly, but also continuously improve its own control level by relying on advanced data analysis methods. It effectively overcomes the problem that traditional fixed sliding pressure strategies cannot balance safety and economy, greatly improving the precision and technological advancement of thermal power plant operation and management.

[0007] Optionally, the steps of establishing a three-dimensional mapping model of back pressure-load-main steam flow based on the real-time operating data and the pre-stored test database, and outputting the initial sliding pressure curve parameter set, include: Analyze the real-time operating data of the steam turbine unit to obtain the current unit load, ambient temperature, and main steam flow rate; The current operating condition range is determined based on the current unit load and the ambient temperature. Extract historical test data that matches the current operating condition range from the pre-stored test database; A three-dimensional coordinate system is constructed with the current unit load as the first coordinate axis, back pressure as the second coordinate axis, and main steam flow as the third coordinate axis; In the three-dimensional coordinate system, a continuous surface model of back pressure-load-main steam flow is generated based on the historical test data; the continuous surface model is parametrically calculated, and the initial sliding pressure curve parameter set is output.

[0008] By adopting the above technical solution, integrating the actual operating status information of the steam turbine unit with historical test data in the standard performance test database, and combining advanced mathematical modeling methods, the set of basic parameters required for the optimal sliding pressure control strategy suitable for the current operating conditions is automatically generated.

[0009] Optionally, the step of obtaining an ambient temperature compensation factor based on the ambient temperature, and modifying the initial sliding pressure curve parameter set based on the ambient temperature compensation factor to generate a compensated sliding pressure curve parameter set includes: Obtain the real-time ambient temperature of the steam turbine unit; The ambient temperature compensation factor is obtained by querying the predefined temperature-compensation factor mapping table based on the real-time ambient temperature. Receive the initial sliding pressure curve parameter set; The initial sliding pressure curve parameter set is corrected and calculated using the ambient temperature compensation factor to generate a compensated sliding pressure curve parameter set.

[0010] By adopting the above technical solutions, not only is a highly sensitive response capability to external climate change achieved, but a closed-loop feedback system architecture is also established, effectively overcoming many drawbacks of traditional static setting modes that cannot cope with complex and ever-changing operating scenarios. By precisely matching the coupling relationship between ambient temperature and thermal performance, the originally rigid control parameters can be dynamically adjusted while maintaining the original control direction, thereby significantly enhancing the unit's overall energy efficiency and long service life throughout its entire life cycle.

[0011] Optionally, the step of generating an optimized sliding pressure curve function based on the compensated sliding pressure curve parameter set, the predefined sliding pressure curve equation, and the unit design constraints includes: Receive the parameter set of the sliding pressure curve after compensation; Call the pre-stored sliding pressure curve equation; Substitute the compensated sliding pressure curve parameter set into the sliding pressure curve equation to calculate the original sliding pressure curve function; Loading unit design constraints; The original sliding pressure curve function is subjected to boundary processing using the unit design constraints to generate an optimized sliding pressure curve function.

[0012] By adopting the above technical solution, the effective transformation from the original parameters to the final optimized function is realized. The final optimized sliding pressure curve can not only maximize the efficiency potential of the unit while meeting various stringent process requirements, but also greatly enhance the overall anti-disturbance capability and operational stability performance in the face of uncertain external disturbances.

[0013] Optionally, the step of calculating the target main steam pressure based on the current unit load and the optimized sliding pressure curve function includes: Obtain the current unit load value; Call the optimized sliding pressure curve function; Input the current unit load value into the optimized sliding pressure curve function to calculate the initial value of the main steam pressure; Based on the unit design constraints in the optimized sliding pressure curve function, the initial value of the main steam pressure is boundary-corrected, and the target main steam pressure is output.

[0014] By adopting the above technical solution, not only is the function of intelligently selecting appropriate pressure setpoints based on different load conditions realized, but also multiple safety assurance mechanisms are nested at each level of the entire decision-making chain, fundamentally eliminating the possibility of major safety accidents caused by misoperation. Through precise control of multiple key technical elements, the final adjustment commands can both maximize energy-saving potential and firmly safeguard the inviolable lifeline of safe production, achieving the goal of maximizing operational efficiency under ideal conditions.

[0015] Optionally, the step of obtaining actual power supply coal consumption data and optimizing and updating the sliding pressure curve parameter set based on the comparison results of the actual power supply coal consumption data and the preset coal consumption threshold includes: Real-time collection of coal consumption data for unit power supply; Read the preset coal consumption threshold; Calculate the absolute value of the deviation between the power supply coal consumption data and the preset coal consumption threshold; When the absolute value of the deviation exceeds the preset tolerance, a parameter optimization command is triggered; In response to the parameter optimization command, based on the absolute value of the deviation, the optimization algorithm is used to perform reverse iterative calculation on the sliding pressure curve parameter set to generate an updated sliding pressure curve parameter set; The updated sliding pressure curve parameter set is stored in the model parameter library.

[0016] By adopting the above technical solution, and through dynamic monitoring and feedback adjustment of coal consumption for power generation during unit operation, automated optimization of key parameters in the sliding pressure control strategy has been achieved. This technical solution improves the adaptability and energy-saving potential of thermal power units under complex variable load environments, while reducing the frequency of human intervention and lowering operation and maintenance costs.

[0017] Optionally, after generating the turbine control valve command, the following steps are also included: Collect actual main steam pressure values; Calculate the real-time deviation between the actual main steam pressure value and the target main steam pressure; When the real-time deviation value continuously exceeds the preset deviation threshold for a preset duration, the emergency correction mode of the sliding pressure curve parameter is activated; in the emergency correction mode, the current sliding pressure curve parameter set is overwritten based on the historical optimal sliding pressure curve parameter set. The covered sliding pressure curve parameter set is input into the three-dimensional mapping model for real-time reconstruction.

[0018] By adopting the above technical solution, an effective leap has been achieved from the conventional PID control level to the model-driven adaptive compensation level. This solution can autonomously complete a series of complex linked operations such as fault location, parameter rollback, and model reconstruction with almost no impact on normal power generation. It can bring the unit back to near its optimal efficiency operating point in a short time, significantly reducing the frequency of unplanned shutdowns while also curbing unnecessary energy waste.

[0019] Secondly, this application provides a turbine operation optimization control system based on a dynamic sliding pressure curve, which adopts the following technical solution: A turbine operation optimization control system based on dynamic sliding pressure curve is characterized in that the operation optimization control system includes: a data acquisition module for acquiring real-time operating data of the turbine unit, including current unit load, ambient temperature and main steam flow; The initial parameter set generation module is used to establish a three-dimensional mapping model of back pressure-load-main steam flow based on the real-time operating data and the pre-stored test database, and output the initial sliding pressure curve parameter set; The temperature compensation module is used to obtain an ambient temperature compensation factor based on the ambient temperature, and to correct the initial sliding pressure curve parameter set based on the ambient temperature compensation factor to generate a compensated sliding pressure curve parameter set. The optimization module is used to generate an optimized sliding pressure curve function based on the compensated sliding pressure curve parameter set, the predefined sliding pressure curve equation, and the unit design constraints. The regulating valve control command generation module is used to calculate the target main steam pressure based on the current unit load and the optimized sliding pressure curve function, and generate turbine regulating valve control commands based on the target main steam pressure; The optimization and update module is used to acquire actual power supply coal consumption data, optimize and update the sliding pressure curve parameter set based on the comparison results between the actual power supply coal consumption data and the preset coal consumption threshold, output the updated sliding pressure curve parameter set and apply it to the reconstruction of the three-dimensional mapping model.

[0020] Thirdly, this application provides a computer device, which adopts the following technical solution: A computer device includes a memory, a processor, and a computer program stored in the memory, the processor executing the computer program to perform the steps of the method as described in the first aspect.

[0021] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium storing a computer program that can be loaded by a processor and executed as in any of the methods in the first aspect. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the first process of a turbine operation optimization control method based on dynamic sliding pressure curve, which is one embodiment of this application.

[0023] Figure 2 This is a second flowchart of a turbine operation optimization control method based on dynamic sliding pressure curve, according to one embodiment of this application.

[0024] Figure 3 This is a schematic diagram of the third process of a turbine operation optimization control method based on dynamic sliding pressure curve, according to one embodiment of this application.

[0025] Figure 4 This is a schematic diagram of the fourth process of a turbine operation optimization control method based on dynamic sliding pressure curve according to one embodiment of this application.

[0026] Figure 5 This is a fifth flowchart of a turbine operation optimization control method based on dynamic sliding pressure curve, according to one embodiment of this application.

[0027] Figure 6 This is a schematic diagram of the sixth process of a turbine operation optimization control method based on dynamic sliding pressure curve, according to one embodiment of this application.

[0028] Figure 7 This is a schematic diagram of the seventh process of a turbine operation optimization control method based on dynamic sliding pressure curve according to one embodiment of this application. Detailed Implementation

[0029] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figure 1-7 The present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the application.

[0030] This application discloses a turbine operation optimization control method based on dynamic sliding pressure curves.

[0031] Reference Figure 1A method for optimizing turbine operation based on dynamic sliding pressure curves is proposed. The optimization control method includes: Step S101, collecting real-time operating data of the turbine unit, including current unit load, ambient temperature, and main steam flow rate. The unit load refers to the electrical power output by the generator per unit time, usually expressed in megawatts (MW), reflecting the unit's workload. Ambient temperature affects the heat exchange performance of the cooling system, thus indirectly changing the condenser back pressure. The main steam flow rate is the mass flow rate of steam entering the turbine's high-pressure cylinder, directly affecting work capacity and thermal efficiency. These three variables have complex nonlinear coupling relationships, especially since they jointly determine the energy conversion efficiency inside the turbine and the back pressure level on the exhaust side.

[0032] Step S102: Based on real-time operating data and pre-stored test database, establish a three-dimensional mapping model of back pressure-load-main steam flow and output the initial sliding pressure curve parameter set; Back pressure is the pressure of steam discharged from the low-pressure cylinder of the steam turbine, which is also the saturation pressure value corresponding to the vacuum level in the condenser. It is an important indicator for measuring the flow capacity of the last stage of the steam turbine and the characteristics of the cold source. Since back pressure is significantly affected by factors such as seasonal changes and fluctuations in circulating water temperature, incorporating it into the multidimensional modeling system helps improve prediction accuracy.

[0033] In this embodiment, the ASME PTC6 test database can be used. This database is a large collection of historical operating condition data accumulated according to the performance testing standards published by the American Society of Mechanical Engineers. It covers records of various key parameters under different load points. By performing similarity matching or interpolation on the current operating state within the database, the closest actual operating condition sample can be quickly located. Then, methods such as least squares, radial basis function fitting, or multilayer neural network regression are used to construct a three-dimensional surface plot describing the combined trend of back pressure and steam flow. This surface is essentially a visual representation of the initial sliding pressure law, and the set of coefficients it contains constitutes the initial sliding pressure curve parameter set, which can be used for further correction.

[0034] Step S103: Obtain the ambient temperature compensation factor based on the ambient temperature, and modify the initial sliding pressure curve parameter set based on the ambient temperature compensation factor to generate the compensated sliding pressure curve parameter set. The temperature-compensation factor mapping table is a pre-trained lookup table structure, where each row corresponds to a specific range of ambient temperature and its corresponding correction coefficient. For example, when the outdoor temperature rises, the cooling tower's cooling effect weakens, causing the condenser back pressure to increase. If the original sliding pressure curve is still followed, it may lead to increased steam throttling losses and reduced economic efficiency. Therefore, the recommended target main steam pressure setpoint needs to be appropriately lowered, and vice versa. This correction mechanism can be completed through simple multiplication operations, ensuring both response speed and reduced computational complexity. The final compensated sliding pressure curve parameter set incorporates real-time climate information, making it closer to the needs of real-world working scenarios.

[0035] Step S104: Based on the compensated sliding pressure curve parameter set, the predefined sliding pressure curve equation, and the unit design constraints, generate the optimized sliding pressure curve function; This process involves substituting the environmentally corrected parameters into a predefined sliding pressure curve equation, while considering the unit's own safety boundary limitations, thereby deriving an optimized sliding pressure function expression suitable for practical control. The sliding pressure curve equation here typically takes the form of a quadratic polynomial: P = a·L 2 +b·L+c, where P represents the target main steam pressure, L represents the current load percentage, and a, b, c are compensation parameters from the previous step.

[0036] In addition, certain engineering constraints must be imposed, such as stipulating that the maximum / minimum allowable main steam pressure shall not exceed 24MPa and be lower than 8MPa, respectively. This is to prevent problems such as erosion and wear caused by excessive valve opening or surge and instability caused by excessive valve closing. By setting upper and lower limits and trimming the excess portion, the resulting sliding pressure function is ensured to always be within a safe and controllable range, demonstrating good robustness and practicality.

[0037] Step S105: Calculate the target main steam pressure based on the current unit load and the optimized sliding pressure curve function, and generate turbine control valve commands based on the target main steam pressure; This step can be accomplished using a PID controller module to transform a given target into an output action. The PID controller, with its proportional, integral, and derivative terms, achieves functions such as real-time error correction, cumulative offset elimination, and overshoot suppression, and is widely used in various automation systems.

[0038] In this embodiment, the controller receives the ideal pressure signal calculated by the sliding pressure function as a reference input, then compares it with the current measured pressure value to generate a deviation signal. After internal algorithm calculation, it outputs a standard 4-20mA current signal to drive the servo mechanism to move and push the regulating valve stem to move until the two tend to be consistent.

[0039] Step S106: Obtain actual power supply coal consumption data; based on the comparison results between the actual power supply coal consumption data and the preset coal consumption threshold, optimize and update the sliding pressure curve parameter set; output the updated sliding pressure curve parameter set and apply it to the reconstruction of the three-dimensional mapping model.

[0040] Specifically, the actual coal consumption data for power supply is read on an average basis over a period of time and compared with the theoretical optimal value. Once the difference between the two is found to exceed a preset threshold (such as 0.5 g / kWh), it is determined that the current sliding pressure strategy may no longer be applicable to the current operating condition combination, and it is necessary to recalibrate it.

[0041] To address this, the gradient descent algorithm, commonly used in machine learning, can be introduced for backpropagation. By continuously adjusting the parameter direction, the objective function gradually converges to a local minimum, thus obtaining a new set of sliding pressure coefficients. This updated set of parameters is then written back to a dedicated model library for the corresponding capacity level (typically 300MW to 1000MW) for future use, thus completing the evolution from an empirical model to an online adaptive model.

[0042] It should be noted that although the adjustment range is limited each time, the cumulative effect over a long period of time can largely offset the inherent deviations caused by equipment aging, scaling and deposits, and maintain the effectiveness and timeliness of the entire control logic.

[0043] In the above embodiments, a sliding pressure curve model that reflects the current operating environment and equipment characteristics is constructed and continuously updated, thereby achieving precise regulation of the main steam pressure of the steam turbine, and realizing the goal of reducing coal consumption and improving efficiency. This technical solution can not only capture internal and external disturbance information of the unit at any time and respond in a timely manner, but also continuously improve its own regulation level by relying on advanced data analysis methods. It effectively overcomes the problem that traditional fixed sliding pressure strategies cannot balance safety and economy, and greatly improves the precision and technological advancement of thermal power plant operation and management.

[0044] Reference Figure 2 As one implementation of step S102, the steps of establishing a three-dimensional mapping model of back pressure-load-main steam flow based on real-time operating data and a pre-stored test database, and outputting the initial sliding pressure curve parameter set, include: Step S201 involves analyzing the real-time operating data of the steam turbine unit to obtain the current unit load, ambient temperature, and main steam flow rate. These three key parameters together constitute the core vector describing the turbine's operating state, and they have complex coupling relationships. Therefore, they must be acquired synchronously to ensure the accuracy of subsequent analysis. For example, in high-temperature environments during summer, even with a low load, the vacuum may deteriorate due to the rise in cold source temperature, resulting in higher main steam consumption under the same load. In such cases, relying solely on a single-dimensional reference value for adjustment is insufficient to adapt to the complex and ever-changing real-world operating scenarios. To ensure that these physical quantities are fed back to the control system in a timely and accurate manner, industrial communication protocols (such as Modbus / TCP) are generally used to connect the field sensor network and the central processing unit to achieve real-time online monitoring.

[0045] Step S202: Determine the current operating condition range based on the current unit load and ambient temperature; The current operating condition range refers to several representative regions formed by discretizing and partitioning continuously changing operating parameters. In this embodiment, a dual partitioning strategy is specifically adopted: on the one hand, it is divided into multiple sub-ranges of fixed width according to the unit load (e.g., each 10% of the rated load is a unit); on the other hand, a similar grading scale is set for the ambient temperature (e.g., each 5°C is a level).

[0046] Specifically, once the specific load value and temperature reading are measured at a certain moment, the corresponding two-dimensional combination range can be quickly located through simple table lookup calculations. For example, if a 600MW unit has a load of 380MW (approximately 63.3% load) and the outdoor dry-bulb temperature is 27℃, then the corresponding operating condition label might be "60%-70% load + 25℃~30℃".

[0047] Step S203: Extract historical test data that matches the current operating condition range from the pre-stored test database; The adopted standard, ASME PTC6, is an internationally recognized performance acceptance test standard for thermal power generation equipment, covering detailed measurement procedures, uncertainty assessment criteria, and other content. The results of formal tests conducted based on this standard are widely considered the most authoritative and traceable official data. However, due to the high cost and long cycle of on-site experiments, frequent repetition is not feasible. Therefore, it is necessary to solidify and store the results obtained within a limited number of trials, and support subsequent queries and reuse through a reasonable indexing mechanism. In this context, a pre-compiled database acts like a "knowledge graph," which can be viewed as a high-dimensional spatial structure composed of numerous discrete observation points. Each node carries the system's response function expression under specific boundary conditions. When the current operating state is identified as belonging to a known category, the closest set of one or more past records can be retrieved as reference objects, laying the theoretical foundation for further deducing the expected behavior pattern under the current state. This step is particularly important for improving system robustness and shortening model training time.

[0048] Step S204: Construct a three-dimensional coordinate system with the current unit load as the first coordinate axis, back pressure as the second coordinate axis, and main steam flow as the third coordinate axis; Back pressure refers to the pressure at the turbine exhaust end, a crucial indicator for evaluating vacuum system performance, and is often significantly affected by seasonal changes and cooling tower performance fluctuations. The reason for choosing these three variables to form a three-dimensional Cartesian coordinate framework is that they exhibit strong nonlinear correlations yet are independently controllable, making them highly suitable for characterizing the essential features of the energy conversion process within the turbine. Furthermore, using a unified spatial representation facilitates subsequent visualization and numerical approximation operations.

[0049] Step S205: In the three-dimensional coordinate system, a continuous surface model of back pressure-load-main steam flow is generated based on historical test data; This step employs a non-uniform rational B-splines (NURBS) interpolation / approximation algorithm. Compared to traditional polynomial interpolation or piecewise linear approximation, NURBS offers stronger local control and higher smoothness, making it particularly suitable for constructing the distribution patterns of phenomena with singularities or drastic gradient changes.

[0050] In this embodiment, the sample set selected in the aforementioned steps can be used as input point cloud data, and then a mature numerical software package can be called to perform global optimization fitting to obtain a complete implicit function surface. Further constraints, such as least-squares error minimization, can be applied to ensure that the resulting graph closely approximates real physical laws while maintaining sufficient flexibility to handle various disturbances.

[0051] Step S206: Perform parametric calculations on the continuous surface model and output the initial sliding pressure curve parameter set.

[0052] The sliding pressure curve refers to the optimal back pressure setting trajectory that the steam turbine should maintain under different load conditions. By utilizing the aforementioned three-dimensional continuous surface model, the intrinsic relationships between various factors can be captured more accurately, thus reversing a target path that more closely reflects reality. Parameterization is the process of transforming abstract geometric shapes into a set of mathematical expressions that can be directly used by the controller. A typical approach involves slicing along a given direction to extract a series of profile segments, then applying Fourier series expansions or other types of function approximation to each segment, finally extracting the coefficients as the final output. The advantage of this approach is that it preserves the overall structural properties of the original model while facilitating its integration into existing DCS platforms for closed-loop control operation.

[0053] In the above implementation, the actual operating status information of the steam turbine unit is integrated with historical test data in the standard performance test database. Combined with advanced mathematical modeling methods, the set of basic parameters required for the optimal sliding pressure control strategy applicable to the current operating conditions is automatically generated.

[0054] Reference Figure 3 As one implementation of step S103, the step of obtaining an ambient temperature compensation factor based on the ambient temperature, and correcting the initial sliding pressure curve parameter set based on the ambient temperature compensation factor to generate a compensated sliding pressure curve parameter set includes: Step S301: Obtain the real-time ambient temperature of the steam turbine unit; Step S302: Based on the real-time ambient temperature, a predefined temperature-compensation factor mapping table is queried to obtain the ambient temperature compensation factor. This mapping table is a solidified product derived from the analysis and refinement of a large amount of historical operating records or simulation test data during the early R&D phase, rather than being obtained through online learning. Its construction process involves multiple complex steps, including the collection of multi-dimensional performance evaluation indicators, quantitative modeling of thermodynamic cycle efficiency deviations, and fitting of function approximation algorithms.

[0055] Specifically, technicians will manually set multiple typical ambient temperature test nodes (such as 5℃, 15℃, 25℃, 35℃, etc.) in different seasonal periods, and record the maximum net output power that the turbine generator set can achieve under the corresponding conditions and the corresponding main steam pressure set value at each node. Then, the theoretical optimal operating pressure under each temperature condition is derived using the heat balance equation, and the difference ratio is calculated by comparing it with the measured recommended pressure, which is the so-called "thermal efficiency deviation rate". Based on this, a continuous and smooth nonlinear compensation curve is derived using regression tools such as the least squares method or multilayer perceptron neural network. Finally, the curve is discretized and truncated according to the piecewise linear interpolation idea commonly used in engineering practice, forming several constant coefficient table items in fixed intervals for the controller to access and call at high speed.

[0056] In this embodiment, once a specific ambient temperature T_env is detected in real time, the target interval index i to which it belongs can be quickly located using the classic binary search method, and the corresponding proportional gain K_i can be extracted as the weighting factor required for the next parameter correction. It is worth noting that, to prevent over- or under-compensation due to extreme weather changes, the mapping table also sets reasonable upper and lower limit threshold constraints (e.g., 0.92 to 1.08), and the slope increment between two adjacent temperature levels is strictly limited to ±0.01 / ℃, ensuring that any small disturbance will not cause drastic jumps, thus improving the robustness and security of the system.

[0057] Step S303: Receive the initial sliding pressure curve parameter set; The initial sliding pressure curve parameter set is a combination of parameters to be adjusted generated by basic control logic or expert experience rules before any environmental correction. It may include a series of important variables that directly affect the turbine's work capacity, such as the percentage of rated load, the desired exhaust vacuum, and the ideal nozzle opening command. These parameters are usually formulated based on static design drawings or reference manuals provided by the manufacturer and are applicable to a certain standard meteorological condition (such as the 15°C dry-bulb temperature specified in ISO 19475). However, in real-world environments, it is almost impossible to find a situation that perfectly conforms to this benchmark. Without correction, it is easy to cause the equipment to deviate from the high-efficiency range or even induce vibration instability and other problems.

[0058] Step S304: The initial sliding pressure curve parameter set is corrected and calculated using the ambient temperature compensation factor to generate the compensated sliding pressure curve parameter set.

[0059] The system iterates through each element P_j_old in the initial parameter set and multiplies it by the previously obtained global gain coefficient K to obtain a new target value P_j_new = K * P_j_old. Because many thermodynamic parameters are essentially extensive masses, they exhibit an approximately proportional growth trend with changes in the external environment. For example, as the temperature rises, the cylinder wall heat dissipation rate decreases, leading to an increase in the required new steam flow rate at the same rotational speed, and vice versa.

[0060] However, simply scaling up or down cannot completely solve the problem, especially under certain extreme boundary conditions, which may lead to risks exceeding the equipment's capacity. Therefore, an additional normalization process is needed, mapping all the initially transformed results back to their respective valid domain intervals (e.g., maximum allowable opening must not exceed 100%, minimum exhaust back pressure must not be lower than the local saturated steam pressure, etc.) to avoid cascading failures and shutdowns caused by violations of process specifications. This final parameter sequence forms the main basis for the next stage of sliding pressure control strategy updates and iterations, and can be directly applied to various downstream application scenarios such as PID controller setpoint refresh and DEH valve position servo drive signal generation.

[0061] The above implementation not only achieves a highly sensitive response to external climate change, but also establishes a closed-loop feedback system architecture, effectively overcoming the many drawbacks of traditional static setting modes that cannot cope with complex and ever-changing operating scenarios. By precisely matching the coupling relationship between ambient temperature and thermal performance, the originally rigid control parameters can be dynamically adjusted while maintaining the original control direction, thereby significantly enhancing the unit's overall energy efficiency and long-term service life throughout its entire life cycle.

[0062] Reference Figure 4 As one implementation of step S104, the step of generating an optimized sliding pressure curve function based on the compensated sliding pressure curve parameter set, the predefined sliding pressure curve equation, and the unit design constraints includes: Step S401: Receive the set of parameters for the compensated sliding pressure curve; The data involved in the parameter set of the compensated sliding pressure curve are key input quantities that have been modeled, identified, or corrected through feedback in the early stages. These parameters are usually derived from historical operating data, real-time operating condition feedback, and theoretical model fitting, and may include dynamic adjustment terms such as temperature correction factors and load fluctuation adaptation coefficients.

[0063] Step S402: Call the pre-stored sliding pressure curve equation; The sliding pressure curve equation is typically derived from manufacturer-provided design specifications or empirical formulas accumulated through long-term operational experience. It reflects the mapping relationship between the main steam pressure (Pmain) and the unit load (Pload), serving as a crucial bridge between operational objectives and physical behavior. Choosing an appropriate functional form is essential for improving control accuracy; in this scheme, a quadratic polynomial structure is used, i.e., P = a·L. 2 +b·L+c, where P represents the target main steam pressure, L represents the current load percentage, and a, b, and c are derived from the compensated sliding pressure curve parameter set in the preceding steps. The purpose of this stage is to establish a generalized mathematical framework, enabling pressure control strategies under different operating conditions to be implemented under a standardized paradigm.

[0064] Step S403: Substitute the compensated sliding pressure curve parameter set into the sliding pressure curve equation to calculate the original sliding pressure curve function; In this process, a known set of compensated parameters are precisely assigned to the coefficients of the predetermined equation, thereby completing a full numerical mapping operation. This process is essentially a typical function instantiation operation, obtaining an executable pressure output sequence under a specific situation by concretely assigning values ​​to an abstract formula. The resulting original sliding pressure curve function actually represents the pressure distribution pattern that should exist under an ideal state without any restrictions or interventions.

[0065] Step S404: Apply unit design constraints; The unit design constraints cover multiple dimensions of safety threshold settings, including but not limited to the upper limit of maximum operating pressure, the lower limit of minimum starting pressure, and other engineering indicators related to material tolerance and thermal stress distribution.

[0066] In this embodiment, special emphasis is placed on reading the main steam pressure threshold range, that is, 8MPa to 24MPa is introduced into the system's judgment logic as a rigid interval boundary. Such constraint information is generally derived from the unit's manufacturer's manual, industry standards, or even best practice guidelines summarized from years of operation and maintenance practice, to prevent the controller from issuing operating commands that exceed the equipment's tolerance limits.

[0067] Step S405: Apply the unit design constraints to perform boundary processing on the original sliding pressure curve function to generate an optimized sliding pressure curve function.

[0068] In this process, the pressure function under the aforementioned ideal state undergoes rigorous pruning and reconstruction to ensure it fluctuates within a legally reasonable range. Specifically, piecewise linear interpolation is used for numerical normalization: if the calculated pressure at a certain point in time is found to be below 8 MPa, it is forcibly increased to 8 MPa; conversely, if it exceeds 24 MPa, it is truncated to 24 MPa; while in the normal region between the two, it remains unchanged. This scheme effectively balances the requirements of smooth transition and hard blocking, avoiding the risk of mechanical damage caused by sudden shocks while ensuring a continuous output of the expected target signal throughout the entire operating cycle. More importantly, by using a continuous function approximation method instead of a discrete lookup table method, the resulting optimized sliding pressure curve has higher resolution and stronger robustness.

[0069] In the above implementation, an effective transformation from the original parameters to the final optimized function is achieved. The resulting optimized sliding pressure curve not only maximizes the unit's efficiency potential while meeting various stringent process requirements, but also greatly enhances the overall anti-disturbance capability and operational stability when facing uncertain external disturbances.

[0070] Reference Figure 5 As one implementation of step S105, the step of calculating the target main steam pressure based on the current unit load and the optimized sliding pressure curve function includes: Step S501: Obtain the current unit load value; In thermal power plants, unit load refers to the ratio of the generator's actual output power to its rated capacity, usually expressed as a percentage (e.g., 50% represents half-load operation). This parameter reflects changes in user-side electricity demand and directly affects the coordinated operation of multiple subsystems, such as boiler combustion intensity and feedwater flow regulation. Relevant electrical signals are collected by sensors such as current transformers and voltage transformers installed at the generator outlet or the low-voltage side of the step-up transformer. These signals are then transmitted via a Supervisory Data Acquisition and Monitoring System (SCADA) to a Distributed Control System (DCS), ultimately forming digital load feedback information usable by algorithmic models.

[0071] Step S502: Call the function to optimize the sliding pressure curve; Step S503: Input the current unit load value into the optimized sliding pressure curve function to calculate the initial value of the main steam pressure; In this process, actual load data is substituted into the established mathematical model as independent variables. After a series of addition, subtraction, multiplication, division, and exponentiation operations, a preliminary recommended pressure value is obtained. For example, assuming the current load is measured to be 70% at a certain moment, substituting it into the aforementioned quadratic equation will yield the corresponding theoretical main steam pressure of approximately 16 MPa (the specific value depends on the coefficients a, b, and c).

[0072] Step S504: Based on the unit design constraints in the optimized sliding pressure curve function, the initial value of the main steam pressure is boundary-corrected, and the target main steam pressure is output.

[0073] Specifically, the automatic judgment task is completed based on the conditional branch statements of the unit design constraints: if the initial value is less than the minimum permissible level, the latter is forcibly assigned as the alternative; otherwise, if it exceeds the maximum tolerance limit, it is truncated to the corresponding extreme value; in other cases, it is allowed to continue using the original value.

[0074] The above implementation not only achieves the function of intelligently selecting appropriate pressure setpoints according to different load conditions, but also incorporates multiple safety assurance mechanisms nested throughout the entire decision-making chain, fundamentally eliminating the possibility of major safety accidents caused by misoperation. Through precise control of multiple key technical elements, the final adjustment commands can both maximize energy-saving potential and firmly safeguard the inviolable lifeline of safe production, achieving the goal of maximizing operational efficiency under ideal conditions.

[0075] Reference Figure 6 As one implementation of step S106, the step of obtaining actual power supply coal consumption data and optimizing and updating the sliding pressure curve parameter set based on the comparison results of the actual power supply coal consumption data and the preset coal consumption threshold includes: Step S601: Real-time collection of unit power supply coal consumption data; Specifically, the mass of standard coal consumed by the boiler per unit time is continuously recorded using a coal metering device (such as an electronic belt scale or a weighing coal feeder); simultaneously, the actual active power value at the generator output port is collected using an electric power sensor. These two variables constitute the basic input factors for calculating coal consumption. According to the internationally accepted definition, coal consumption for power generation is the number of grams of raw coal required to generate one kilowatt-hour of electricity, expressed as: Coal consumption for power generation = Coal consumption (kg) / Power generation (kWh). Based on this, the system obtains a set of timely measured coal consumption values ​​as the data basis for subsequent comparative analysis. This method of direct measurement using online instruments avoids the lag and error risks associated with traditional manual meter reading, improves the accuracy and timeliness of data acquisition, and thus enhances the responsiveness of the entire optimization mechanism.

[0076] Step S602: Read the preset coal consumption threshold; The preset coal consumption threshold refers to the optimal energy consumption reference level derived from the turbine's thermodynamic characteristic model under specific load conditions. It is usually provided by the manufacturer and pre-configured in the DCS (Distributed Control System) database. It not only reflects the ideal economic operating range achievable during the equipment design phase but also carries the target constraints at the power plant operation and management level.

[0077] It should be noted that this threshold is not fixed, but fluctuates regularly with changes in load. Therefore, it is necessary to dynamically apply the corresponding baseline value according to different operating conditions.

[0078] Step S603: Calculate the absolute value of the deviation between the power supply coal consumption data and the preset coal consumption threshold; The deviation between the current actual coal consumption and its expected optimal level is mathematically characterized. Since deviations in both positive and negative directions may reflect potential operational problems (e.g., excessively low coal consumption may be accompanied by incomplete combustion), absolute value calculations are used to eliminate the influence of sign and ensure that the deviation index remains within the positive range. This deviation Δ is the absolute value of the actual coal consumption minus the preset coal consumption threshold. From a control theory perspective, this deviation constitutes the error signal source in the feedback loop, and its magnitude determines whether subsequent optimization processes are initiated and the adjustment range.

[0079] Furthermore, historical data analysis shows that an increase in Δ is often accompanied by an increased probability of problems such as unstable steam parameters, valve leakage, or decreased combustion efficiency. Therefore, it can be regarded as an effective carrier of an early warning mechanism.

[0080] Step S604: When the absolute value of the deviation exceeds the preset tolerance, a parameter optimization command is triggered; The tolerance range is the key boundary that distinguishes normal fluctuations from abnormal disturbances. In this embodiment, it is set to 0.5 g / (kW·h), meaning that the next optimization action will only be activated when the coal consumption deviation exceeds this allowable fluctuation limit. The reason for setting such a threshold is to prevent frequent false triggering of unnecessary parameter changes due to minor noise interference, which could cause oscillations in the control system or even disrupt the original stable operating state.

[0081] Step S605: In response to the parameter optimization command, based on the absolute value of the deviation, the optimization algorithm is used to perform reverse iterative calculation on the sliding pressure curve parameter set to generate an updated sliding pressure curve parameter set; The sliding pressure curve parameter set refers to a set of adjustable coefficients that determine the relationship between the main steam pressure and load variation. These parameters directly affect the working efficiency of the high-pressure cylinder flow path and the overall matching status of the reheat cycle. In practice, they may be weight terms in a polynomial fitting function or pressure setpoints at various nodes in a lookup table method.

[0082] In this embodiment, the gradient descent optimization algorithm can be used. This algorithm is a classic local search optimization tool, particularly suitable for solving high-dimensional nonlinear objective function minimization problems. The basic idea of ​​gradient descent is to move a certain step size in the opposite direction of the gradient of the loss function in each iteration, gradually approaching the global minimum. The squared coal consumption deviation is constructed as the objective function (Loss Function), i.e.: L(θ) = (Actual coal consumption - Theoretical coal consumption) 2 , where θ represents the parameter vector of the sliding pressure curve to be optimized. The update rule for each iteration is as follows: Where η is the learning rate hyperparameter, used to regulate the balance between convergence speed and stability. The entire process essentially involves continuous trial and error and calibration to gradually bring the coal consumption under the current operating mode closer to the theoretical optimal solution, ultimately outputting a new set of parameter configurations to more effectively guide the steam pressure regulation behavior in the next operating cycle.

[0083] Step S606: Store the updated sliding pressure curve parameter set into the model parameter library.

[0084] The model parameter library can be understood as a dedicated database module deployed within the DCS platform, specifically used to store various validated and effective process control parameter templates. Newly generated parameter sets must undergo rigorous format verification, boundary checks, and consistency reviews before being officially added to the library to ensure good compatibility and security.

[0085] In the above embodiments, through dynamic monitoring and feedback adjustment of coal consumption for power generation during unit operation, automated optimization of key parameters in the sliding pressure control strategy is achieved. This technical solution improves the adaptability and energy-saving potential of thermal power units under complex variable load environments, while reducing the frequency of human intervention and lowering operation and maintenance costs.

[0086] Reference Figure 7 As a further implementation of the turbine operation optimization control method, after generating the turbine control valve command, the following steps are also included: Step S701: Collect the actual main steam pressure value; Specifically, this step relies on a high-precision piezoresistive pressure transmitter (such as the ROSEMOUNT 3051 series) installed on the main steam pipeline. It uses the piezoelectric effect of semiconductor materials to convert static pressure information in the physical field into a continuously varying 4-20mA standard electrical signal. This analog signal is then converted from analog to digital by the AI ​​module in the distributed control system (DCS) to obtain a digital representation of the actual main steam pressure value.

[0087] Step S702: Calculate the real-time deviation between the actual main steam pressure and the target main steam pressure; Specifically, an incremental deviation assessment model was adopted in this process, mathematically expressed as ΔP=|P_actual-P_target| / P_rated×100%, where P_actual represents the actual main steam pressure value, P_target represents the target main steam pressure, and P_rated represents the rated main steam pressure level parameter of the unit. This normalization method effectively eliminates comparison errors caused by dimensional differences between units of different capacity levels, improving cross-platform adaptability.

[0088] Furthermore, to enhance anti-interference performance, a sliding time window filtering strategy is introduced into the algorithm. By performing a weighted average calculation on sampling points within a range of at least 30 seconds, the impact of high-frequency noise caused by uneven combustion or wind and coal disturbance is weakened. Simultaneously, transient impact signals generated during the instantaneous action of regulating valves are eliminated to prevent erroneous triggering of the protection mechanism due to local abrupt changes. In engineering practice, the allowable deviation threshold is generally set to no more than ±1.5% of the design pressure. Once exceeded, it means that the control system has entered the nonlinear region or is even close to the edge of instability, requiring the activation of higher-level safety intervention measures.

[0089] Step S703: When the real-time deviation value continues to exceed the preset deviation threshold for a preset duration, the emergency correction mode for the sliding pressure curve parameter is activated. When the current operating condition is determined to have significantly deviated from the expected trajectory, decisive action must be taken to prevent potential risks from escalating. The trigger logic for activating the emergency correction mode of the sliding pressure curve parameters employs a dual-condition AND gate decision mechanism: on the one hand, the deviation amplitude must meet the over-limit condition; on the other hand, it must confirm that the duration of the abnormal state reaches a predetermined threshold value (usually between 120 and 300 seconds). This design aims to avoid misjudgments that may be caused by short-term disturbances such as turbine tripping. Once both prerequisites are met, the system will immediately terminate the original control strategy execution process, switch to a special response procedure guided by the expert rule base, and simultaneously push audible and visual warning information to the operator workstation for timely monitoring intervention. Considering the significant differences in thermal inertia among different types of units, a longer delay time (e.g., 180 seconds) is preferred for subcritical parameter configurations, while the delay time is correspondingly shortened to around 120 seconds for supercritical and above parameter levels, reflecting a differentiated control concept.

[0090] Step S704: In emergency correction mode, the current sliding pressure curve parameter set is overwritten based on the historical best sliding pressure curve parameter set; Specifically, an energy consumption minimization optimization analysis was conducted on all operating condition samples recorded within the same load range (±5%) over the past month. Using the K-means clustering algorithm, several Pareto optimal solution sets balancing economy and stability were extracted. These empirical parameter combinations derived through statistical learning constitute a pool of alternative resources available for replacement. During the actual switching operation, a dual-buffer management architecture was employed for protection: currently used operating parameters were placed in a read-only locked state to prevent tampering or damage, while simultaneously, selected historically optimal parameters were loaded into a backup memory area awaiting activation commands. Finally, a seamless transition between the old and new versions was rapidly completed in a single atomic-level hot-swap operation, with the entire process taking no more than the upper limit of a single control cycle (≤200 milliseconds), thus ensuring that no control interruptions occurred during the entire transition period.

[0091] Step S705: Input the covered sliding pressure curve parameter set into the three-dimensional mapping model for real-time reconstruction.

[0092] The three-dimensional mapping model is essentially a highly abstract nonlinear dynamic description framework, which can be expressed in the form of a set of state-space equations as follows: x = f(x, u, θ), y = g(x), where x represents the internal hidden state variable vector, u is the external controllable input sequence, and θ corresponds to the set of key parameter vectors that determine the system characteristics, including but not limited to the sliding pressure regulation correlation coefficient matrix.

[0093] Furthermore, the core objective of the reconstruction task is to adjust θ as quickly as possible to re-match the requirements of the new operating environment when existing parameters are found to be no longer applicable. To this end, this application proposes a fast solution method based on the idea of ​​pre-computation of the Jacobian matrix. During the online reconstruction phase, it is only necessary to call the inverse of the Jacobian partial derivative matrix prepared offline and combine it with the residual term between the current desired output Y_desired and the measured feedback Y_current to deduce the updated parameter estimate θ_new = J in one step. -1 ·(Y_desired-Y_current)+θ_old. Thanks to the support of the FPGA hardware acceleration platform, the entire re-estimation process can be completed in less than fifty milliseconds, greatly improving the system's adaptability and fault tolerance margin.

[0094] The above implementation achieves an effective leap from the conventional PID control level to the model-driven adaptive compensation level. This solution can autonomously complete a series of complex linked operations such as fault location, parameter rollback, and model reconstruction with almost no impact on normal power generation. It can bring the unit back to near its optimal efficiency operating point in a short time, significantly reducing the frequency of unplanned shutdowns while also curbing unnecessary energy waste.

[0095] This application also discloses a turbine operation optimization control system based on dynamic sliding pressure curves.

[0096] A turbine operation optimization control system based on dynamic sliding pressure curve, the operation optimization control system includes: The data acquisition module is used to collect real-time operating data of the steam turbine unit, including the current unit load, ambient temperature, and main steam flow. The initial parameter set generation module is used to establish a three-dimensional mapping model of back pressure-load-main steam flow based on real-time operating data and a pre-stored test database, and output the initial sliding pressure curve parameter set; The temperature compensation module is used to obtain the ambient temperature compensation factor based on the ambient temperature, and to correct the initial sliding pressure curve parameter set based on the ambient temperature compensation factor to generate the compensated sliding pressure curve parameter set. The optimization module is used to generate an optimized sliding pressure curve function based on the compensated sliding pressure curve parameter set, the predefined sliding pressure curve equation, and the unit design constraints. The regulating valve control command generation module is used to calculate the target main steam pressure based on the current unit load and the optimized sliding pressure curve function, and generate turbine regulating valve control commands based on the target main steam pressure; The optimization and update module is used to obtain actual power supply coal consumption data. Based on the comparison results between the actual power supply coal consumption data and the preset coal consumption threshold, the sliding pressure curve parameter set is optimized and updated, and the updated sliding pressure curve parameter set is output and applied to the reconstruction of the three-dimensional mapping model.

[0097] The turbine operation optimization control system based on dynamic sliding pressure curve of this application embodiment can realize any of the above-mentioned turbine operation optimization control methods, and the specific working process of each module in the turbine operation optimization control system can refer to the corresponding process in the above-mentioned method embodiments.

[0098] In the several embodiments provided in this application, it should be understood that the provided methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for example, the division of a certain module is merely a logical functional division, and in actual implementation there may be other division methods, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0099] This application also discloses a computer device.

[0100] The computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described method for optimizing the operation control of a steam turbine based on a dynamic sliding pressure curve.

[0101] This application also discloses a computer-readable storage medium.

[0102] A computer-readable storage medium storing a computer program that can be loaded by a processor and executed as described above in any of the methods of turbine operation optimization control based on dynamic sliding pressure curves.

[0103] The computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device; the program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0104] In this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0105] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce a good effect.

[0106] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A method for operation optimization control of a steam turbine based on a dynamic sliding pressure curve, characterized in that, The optimization control method comprises: Collecting real-time operation data of the steam turbine unit, including current unit load, ambient temperature and main steam flow; Based on the real-time operation data and the pre-stored test database, a three-dimensional mapping model of back pressure-load-main steam flow is established, and an initial sliding pressure curve parameter set is output; According to the ambient temperature, an ambient temperature compensation factor is obtained, and the initial sliding pressure curve parameter set is modified based on the ambient temperature compensation factor to generate a compensated sliding pressure curve parameter set; Based on the compensated sliding pressure curve parameter set and a predefined sliding pressure curve equation and unit design constraint conditions, an optimized sliding pressure curve function is generated; According to the current unit load and the optimized sliding pressure curve function, a target main steam pressure is calculated, and a steam turbine governing valve control instruction is generated based on the target main steam pressure; Obtaining actual power supply coal consumption data, based on the comparison result of the actual power supply coal consumption data and the preset coal consumption threshold, the sliding pressure curve parameter set is optimized and updated, and the updated sliding pressure curve parameter set is output and applied to the reconstruction of the three-dimensional mapping model.

2. The method for steam turbine operation optimization control based on dynamic sliding pressure curve according to claim 1, characterized in that, The step of establishing a three-dimensional mapping model of back pressure-load-main steam flow based on the real-time operation data and the pre-stored test database, and outputting an initial sliding pressure curve parameter set comprises: Analyzing the real-time operation data of the steam turbine unit to obtain the current unit load, ambient temperature and main steam flow; Determine the current working condition interval based on the current unit load and the ambient temperature; Extracting historical test data matching the current working condition interval from the pre-stored test database; A three-dimensional coordinate system is constructed with the current unit load as the first coordinate axis, the back pressure as the second coordinate axis, and the main steam flow as the third coordinate axis; In the three-dimensional coordinate system, a continuous surface model of back pressure-load-main steam flow is generated based on the historical test data; Parameterized calculation is performed on the continuous surface model to output an initial sliding pressure curve parameter set.

3. The method for turbine operation optimization control based on dynamic sliding pressure curve according to claim 2, characterized in that, The step of obtaining an ambient temperature compensation factor according to the ambient temperature, and modifying the initial sliding pressure curve parameter set based on the ambient temperature compensation factor to generate a compensated sliding pressure curve parameter set comprises: Obtaining the real-time ambient temperature of the steam turbine unit; According to the real-time ambient temperature, a temperature-compensation factor mapping table is queried to obtain an ambient temperature compensation factor; Receive the initial sliding pressure curve parameter set; The initial sliding pressure curve parameter set is modified and calculated using the ambient temperature compensation factor to generate a compensated sliding pressure curve parameter set.

4. The method for turbine operation optimization control based on dynamic sliding pressure curve according to claim 3, characterized in that, The step of generating an optimized sliding pressure curve function based on the compensated sliding pressure curve parameter set and a predefined sliding pressure curve equation and unit design constraint conditions comprises: Receive the compensated sliding pressure curve parameter set; Call the pre-stored sliding pressure curve equation; Substitute the compensated sliding pressure curve parameter set into the sliding pressure curve equation to calculate the original sliding pressure curve function; Load the unit design constraint conditions; Apply the unit design constraint conditions to the boundary processing of the original sliding pressure curve function to generate an optimized sliding pressure curve function.

5. The method for steam turbine operation optimization control based on dynamic sliding pressure curve according to claim 4, characterized in that, The step of calculating a target main steam pressure according to the current unit load and the optimized sliding pressure curve function comprises: Obtaining the current unit load value; calling the optimized sliding pressure curve function; inputting the current unit load value into the optimized sliding pressure curve function to calculate an initial value of main steam pressure; performing boundary correction on the initial value of main steam pressure based on unit design constraint conditions in the optimized sliding pressure curve function to output a target main steam pressure.

6. The method for steam turbine operation optimization control based on dynamic sliding pressure curve according to claim 1, characterized in that, The step of obtaining actual power supply coal consumption data and optimizing and updating the sliding pressure curve parameter set based on a comparison result of the actual power supply coal consumption data and a preset coal consumption threshold comprises: real-time collection of unit power supply coal consumption data; reading of a preset coal consumption threshold; calculation of an absolute deviation value of the power supply coal consumption data from the preset coal consumption threshold; triggering of a parameter optimization instruction when the absolute deviation value exceeds a preset tolerance; in response to the parameter optimization instruction, performing reverse iterative calculation on the sliding pressure curve parameter set based on the absolute deviation value by using an optimization algorithm to generate an updated sliding pressure curve parameter set; storing the updated sliding pressure curve parameter set into a model parameter library.

7. The dynamic sliding pressure curve-based steam turbine operation optimization control method according to any one of claims 1 to 6, after the generation of the steam turbine governing valve control instruction, further comprising the following steps: collection of an actual main steam pressure value; calculation of a real-time deviation value of the actual main steam pressure value from the target main steam pressure; activation of a sliding pressure curve parameter emergency correction mode when the real-time deviation value continuously exceeds a preset deviation threshold for a preset time length; in the emergency correction mode, covering the current sliding pressure curve parameter set with a historical optimal sliding pressure curve parameter set; inputting the covered sliding pressure curve parameter set into the three-dimensional mapping model for immediate reconstruction.

8. A dynamic sliding pressure curve based turbine operation optimization control system, characterized in that, The operation optimization control system comprises: a collection module for collecting real-time operation data of a steam turbine unit, including current unit load, ambient temperature and main steam flow; an initial parameter set generation module for establishing a three-dimensional mapping model of back pressure-load-main steam flow based on the real-time operation data and a pre-stored test database, and outputting an initial sliding pressure curve parameter set; a temperature compensation module for obtaining an ambient temperature compensation factor according to the ambient temperature, and modifying the initial sliding pressure curve parameter set based on the ambient temperature compensation factor to generate a compensated sliding pressure curve parameter set; an optimization module for generating an optimized sliding pressure curve function based on the compensated sliding pressure curve parameter set, a predefined sliding pressure curve equation and unit design constraint conditions; a governing valve control instruction generation module for calculating a target main steam pressure according to the current unit load and the optimized sliding pressure curve function, and generating a steam turbine governing valve control instruction based on the target main steam pressure; an optimization update module for obtaining actual power supply coal consumption data, optimizing and updating the sliding pressure curve parameter set based on a comparison result of the actual power supply coal consumption data and a preset coal consumption threshold, outputting an updated sliding pressure curve parameter set and applying the same to reconstruction of the three-dimensional mapping model.

9. A computer device, characterized by: A computer program product comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of any one of claims 1 to 7 when executing the program. A computer program product comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of any one of claims 1 to 7 when executing the program.

10. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the method according to any one of claims 1 to 7.

Images