Method and system for predicting main steam parameters before entering a preheater of a combined cycle unit

Abstract

The application provides a method and system for predicting main steam parameters before steam warming of a combined cycle unit, the method comprising the following steps: obtaining historical operation data of a target power generating unit, the historical operation data comprising a plurality of observable input parameters and output parameters closely related to a main steam system; constructing an NARX neural network model, training the neural network model based on the historical operation data, and obtaining a trained main steam parameter prediction model; deploying the main steam parameter prediction model on a general computing device; obtaining initial state parameters under a current working condition before starting the power generating unit, and inputting the initial state parameters into the main steam parameter prediction model; and outputting a predicted trajectory of main steam key parameters before the high-pressure bypass system is put into operation from the main steam parameter prediction model.

Description

Technical Field

[0001] This invention belongs to the field of combined cycle unit parameter prediction technology, specifically relating to a method and system for predicting the main steam parameters before the inlet steam warm-up of a combined cycle unit. Background Technology

[0002] In recent years, the installed capacity of new energy power generation has continued to grow rapidly, and the intermittency and volatility of new energy sources pose a severe challenge to grid stability: daily power fluctuations can reach more than 70% of the installed capacity. Existing energy storage technologies are constrained by cost, lifespan, and scale limitations, resulting in severely insufficient peak-shaving capabilities. New energy storage projects, represented by electrochemical energy storage, generally have an effective peak-shaving duration of less than 4 hours, and their cost per kilowatt-hour is as high as 0.6-0.8 yuan / kWh, making it difficult to support large-scale peak shaving and valley filling demands. Against this backdrop, thermal power generating units are forced to undertake deep peak-shaving tasks. Traditional coal-fired units experience sharp efficiency drops and worsened emissions under low-load conditions, urgently requiring alternative solutions with superior peak-shaving performance.

[0003] Gas-steam combined cycle units have become ideal peak-shaving carriers due to their inherent characteristics. Their core advantages are reflected in three aspects: (1) rapid start-up and shutdown capability: gas turbines only need 90-120 minutes from shutdown to full load (warm start), which is much shorter than the 5-8 hours of coal-fired units; (2) high-efficiency operation under wide load: power supply efficiency can still be maintained at more than 50% within the 40%-100% load range (compared to the efficiency decline of more than 15% for coal-fired units of the same capacity); (3) outstanding environmental performance: NOx emission concentration is less than 30mg / Nm³, and there is no dust or SO2 emission. Therefore, combined cycle units significantly alleviate the pressure of new energy consumption.

[0004] A typical combined cycle power unit currently consists of three core components: a gas turbine, a waste heat boiler, and a steam turbine. The high-pressure bypass system, a critical auxiliary device, is located between the outlet of the high-pressure superheater of the waste heat boiler and the condenser. It mainly comprises the following components: a high-pressure bypass valve for regulating the main steam pressure; a desuperheating device to reduce the steam temperature from high to a suitable level before it enters the condenser via spray water; and an isolation valve and control system for switching steam paths and ensuring safe isolation. This high-pressure bypass system plays a crucial role in unit start-up, shutdown, and load changes, performing core functions such as main steam pressure control, superheater protection, and steam turbine thermal stress control. The accuracy of its parameters directly impacts the unit's safety and economic efficiency.

[0005] Specifically, the start-up of a combined cycle unit is divided into three operating conditions: cold start (shutdown after >72h), warm start (shutdown after 8-72h), and hot start (shutdown after <8h). Taking cold start as an example, its key stages include: (1) gas turbine ignition and speed-up (unit speed 0-3000r / min); (2) waste heat boiler steam generation stage: when the high-pressure steam pressure reaches the set value, the high-pressure bypass valve is opened; (3) turbine start-up: when the steam parameters meet the standard, the main steam regulating valve is opened and steam enters the turbine. The core functions of the high-pressure bypass system in this process include: establishing boiler steam flow and preventing superheater overheating; maintaining the main steam pressure at the set value; and absorbing more than 80% of the excess steam flow. If the bypass parameters are out of control, it will cause sudden changes in the turbine inlet steam temperature or pressure oscillation, leading to serious accidents such as excessive rotor thermal stress.

[0006] Therefore, there is an urgent need for a method and system to predict the main steam parameters before the combined cycle unit's inlet steam warm-up, in order to solve the technical problem that power plants cannot accurately predict the main steam parameters, which seriously restricts the development of advanced applications such as start-up optimization and life management. Summary of the Invention

[0007] The present invention aims to at least solve one of the technical problems existing in the prior art, and to provide a new technical solution for a method and system for predicting the main steam parameters before the inlet steam warm-up of a combined cycle unit.

[0008] According to a first aspect of the present invention, a method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit is provided, comprising the following steps: Step S1: Obtain historical operating data of the target generator set, including various observable input and output parameters closely related to the main steam system; Step S2: Construct a NARX neural network model and train the neural network model based on historical operating data to obtain a trained main steam parameter prediction model; wherein, the NARX neural network model is used to learn the nonlinear dynamic mapping relationship and time series evolution characteristics between the input parameters and the key output parameters of the main steam. Step S3: Deploy the main steam parameter prediction model on a general-purpose computing device; Step S4: Before starting the generator set, obtain the initial state parameters under the current operating conditions and input the initial state parameters into the main steam parameter prediction model; Step S5: The main steam parameter prediction model outputs the predicted trajectory of the key parameters of the main steam before the high-pressure bypass system is put into operation. The key parameters include at least the main steam pressure and the main steam temperature.

[0009] Optionally, the input parameters include gas turbine exhaust temperature, steam turbine high-pressure cylinder metal temperature, main steam pressure, main steam temperature, and high-pressure bypass valve opening.

[0010] Optionally, the output parameters include the gas turbine exhaust temperature, main steam pressure, and main steam temperature change curve before the high-pressure bypass valve is opened.

[0011] Optionally, the main steam parameter prediction model operates under the condition that the computational resource consumption is lower than a preset threshold.

[0012] Optionally, the historical operating data of the target generator set includes historical operating data during normal operation and startup.

[0013] Optionally, the NARX neural network model is configured to operate in a closed-loop feedback mode during the prediction process, wherein the model output at the current time step is fed back and used as one of the inputs for subsequent time steps to achieve continuous recursive prediction of the dynamic behavior of the system.

[0014] Optionally, after each iteration of prediction, the NARX neural network model determines whether the predicted main steam pressure has reached the preset target pressure threshold. When the determination condition is met, the subsequent prediction calculation is terminated, and the complete evolution trajectory of the main steam parameters is output.

[0015] According to a second aspect of the present invention, a system for predicting the main steam parameters before the inlet steam heater of a combined cycle unit is provided, employing the method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit as described in the first aspect, comprising: The data acquisition module is used to acquire historical operating data of the target generator set, including various observable input and output parameters closely related to the main steam system. The model building and training module is used to build a NARX neural network model, train the neural network model based on historical operating data, and obtain a trained main steam parameter prediction model; wherein, the NARX neural network model is used to learn the nonlinear dynamic mapping relationship and time series evolution characteristics between the input parameters and the key output parameters of the main steam. The model deployment module is used to deploy the main steam parameter prediction model on a general-purpose computing device. The parameter input module is used to obtain the initial state parameters under the current operating conditions before the generator set is started, and input the initial state parameters into the main steam parameter prediction model. The prediction execution module is used to call the main steam parameter prediction model and output the prediction trajectory of the key parameters of the main steam before the high-pressure bypass system is put into operation. The key parameters include at least the main steam pressure and the main steam temperature.

[0016] According to a third aspect of the present invention, an electronic device is provided, comprising: One or more processors; A storage unit for storing one or more programs that, when executed by one or more processors, enable the one or more processors to implement the method for predicting the main steam parameters before the combined cycle unit's inlet steam heater, as described in the first aspect.

[0017] According to a fourth aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, enables the method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit as described in the first aspect.

[0018] One technical advantage of this invention is that: In this embodiment of the application, the method for predicting the main steam parameters before the combined cycle unit's inlet steam warm-up uses a variety of key parameters observable during the operation of the combined cycle unit as model inputs, including gas turbine exhaust temperature, turbine high-pressure cylinder metal temperature, main steam pressure and temperature, high-pressure bypass valve opening, etc. By integrating these real-time data reflecting the unit's status and control behavior, the basis for constructing the prediction model is built.

[0019] Furthermore, the method for predicting the main steam parameters before the combined cycle unit's inlet steam warm-up turbine combines a time-series-based NARX neural network. This mechanism aims to simultaneously capture the complex nonlinear relationships and time dynamic characteristics inherent in the system, thereby achieving a comprehensive prediction of the gas turbine exhaust temperature, main steam pressure, and main steam temperature variation patterns.

[0020] Furthermore, this method for predicting the main steam parameters before the turbine inlet of a combined cycle unit emphasizes computational efficiency. The predictive model it constructs has low computational resource requirements and can run efficiently on a standard personal computer. This allows operators to quickly obtain detailed predicted changes in the main steam parameters before the turbine high-pressure cylinder inlet before actual unit startup, providing a basis for startup strategy formulation and risk assessment. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of a combined cycle power unit; Figure 2 This is a structural diagram of the high-pressure steam system of a combined cycle unit; Figure 3 This is a graph showing the parameter changes during the start-up process of a combined cycle unit; Figure 4 This is a parameter prediction model diagram of the main steam of a combined cycle unit; Figure 5 This is a flowchart illustrating a method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit, according to an embodiment of the present invention.

[0022] In the diagram: 1. Compressor; 2. Combustion chamber; 3. Turbine; 4. Natural gas valve; 5. First economizer; 6. First steam drum; 7. First evaporator; 8. First superheater; 9. Steam valve; 10. Steam turbine; 11. Condenser; 12. Circulating water; 13. Circulating water pump; 14. Condensate pump; 15. Condensate valve; 16. Generator; 17. Transformer; 21. High-pressure feedwater pump; 22. Second economizer; 23. Second steam drum; 24. Second evaporator; 25. Second superheater; 26. Main steam regulating valve; 27. High-pressure cylinder; 28. High-pressure bypass valve; 29. ​​Water spray desuperheater. Detailed Implementation

[0023] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the present application.

[0024] The embodiments of this application will now be described in detail. Examples of these embodiments are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0025] The terms "first" and "second" in the specification and claims of this application may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise stated, "multiple" means two or more. Furthermore, "and / or" in the specification and claims indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0026] According to a first aspect of the invention, see Figure 5 This paper provides a method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit.

[0027] For example, the method for predicting the main steam parameters before the inlet steam warm-up of a combined cycle unit adopts a data-driven approach. Using the operating data of a typical F-class combined cycle unit in a natural gas power plant as training data, a parameter prediction model for the high-pressure bypass system of the combined cycle unit is established through neural network models, steam thermophysical parameters, and other methods. This enables operators to obtain the operating and variation patterns of the main steam parameters before the inlet steam warm-up before the unit starts up.

[0028] Specifically, the method for predicting the main steam parameters before the combined cycle unit's inlet steam warm-up includes the following steps: Step S1: Obtain historical operating data of the target generator set, including various observable input and output parameters closely related to the main steam system; Step S2: Construct a NARX neural network model and train the neural network model based on historical operating data to obtain a trained main steam parameter prediction model; wherein, the NARX neural network model is used to learn the nonlinear dynamic mapping relationship and time series evolution characteristics between the input parameters and the key output parameters of the main steam. Step S3: Deploy the main steam parameter prediction model on a general-purpose computing device; Step S4: Before starting the generator set, obtain the initial state parameters under the current operating conditions and input the initial state parameters into the main steam parameter prediction model; Step S5: The main steam parameter prediction model outputs the predicted trajectory of the key parameters of the main steam before the high-pressure bypass system is put into operation. The key parameters include at least the main steam pressure and the main steam temperature.

[0029] In this embodiment of the application, the method for predicting the main steam parameters before the combined cycle unit's inlet steam warm-up uses a variety of key parameters observable during the operation of the combined cycle unit as model inputs, including gas turbine exhaust temperature, turbine high-pressure cylinder metal temperature, main steam pressure and temperature, high-pressure bypass valve opening, etc. By integrating these real-time data reflecting the unit's status and control behavior, the basis for constructing the prediction model is built.

[0030] Furthermore, the method for predicting the main steam parameters before the combined cycle unit's inlet steam warm-up turbine combines a time-series-based NARX neural network. This mechanism aims to simultaneously capture the complex nonlinear relationships and time dynamic characteristics inherent in the system, thereby achieving a comprehensive prediction of the gas turbine exhaust temperature, main steam pressure, and main steam temperature variation patterns.

[0031] Furthermore, this method for predicting the main steam parameters before the turbine inlet of a combined cycle unit emphasizes computational efficiency. The predictive model it constructs has low computational resource requirements and can run efficiently on a standard personal computer. This allows operators to quickly obtain detailed predicted changes in the main steam parameters before the turbine high-pressure cylinder inlet before actual unit startup, providing a basis for startup strategy formulation and risk assessment.

[0032] It should be noted that shortening the start-up time of combined cycle units is an important measure to improve the economic benefits of natural gas power plants and increase the peak-shaving and frequency regulation capabilities of the power system. However, currently, the commonly used H-class and F-class combined cycle units in China are all supplied by international manufacturers, and the unit control logic is completely encapsulated. Operators can only see the input and output values ​​and cannot obtain the internal calculation process. This invention proposes a method for predicting the main steam parameters before the inlet steam warm-up turbine of a combined cycle unit. This method can effectively solve the above-mentioned technical bottlenecks, and it consumes less computational resources and can quickly predict the main steam parameters before the inlet steam warm-up turbine of the combined cycle unit before start-up.

[0033] In this embodiment, the neural network model is a NARX neural network model based on time series data. Furthermore, this invention requires minimal computational resources; the prediction of high-voltage bypass system parameters can be completed using a personal computer.

[0034] Therefore, this invention innovatively proposes a main steam parameter prediction method based on a neural network model using a data-driven approach. Its breakthrough lies in constructing an explicit mathematical model of "gas turbine exhaust parameters - boiler heat storage - bypass control commands," eliminating reliance on the OEM's core control algorithm. This method will empower the needs of subsequent power plant control systems and equipment. In other words, this application bypasses the direct analysis of encapsulated control logic, instead employing a technical route combining data-driven approaches and physical laws, representing an innovative main steam parameter prediction method.

[0035] Optionally, the input parameters include gas turbine exhaust temperature, steam turbine high-pressure cylinder metal temperature, main steam pressure, main steam temperature, and high-pressure bypass valve opening.

[0036] In the above implementation, the input parameter set comprehensively considers the energy input characteristics, equipment thermal inertia, control action response and system dynamic feedback mechanism during the start-up process of the combined cycle unit. It not only has clear physical meaning and engineering measurability, but also provides sufficient and efficient feature input for the NARX neural network model, thereby achieving high-precision and robust prediction of the main steam pressure and temperature change trajectory while ensuring low computational resource consumption.

[0037] Optionally, the output parameters include the gas turbine exhaust temperature, main steam pressure, and main steam temperature change curve before the high-pressure bypass valve opens. Using the gas turbine exhaust temperature, main steam pressure, and main steam temperature change curve before the high-pressure bypass valve opens as output parameters not only fully depicts the thermodynamic dynamic process during the initial startup phase, but also provides crucial information for safe startup, precise control, and intelligent optimization, significantly enhancing the model's engineering value and practical significance.

[0038] Optionally, the main steam parameter prediction model operates under the condition that the computational resource consumption is below a preset threshold. By controlling the computational resource consumption of the main steam parameter prediction model within the preset threshold, it is ensured that the prediction results can be calculated and output within the control cycle, meeting the real-time response requirements of the unit's distributed control system (DCS) or the turbine's digital electro-hydraulic control system (DEH), and avoiding control lag due to model delay.

[0039] Optionally, the historical operating data of the target generator set includes historical operating data during normal operation and startup.

[0040] In the above embodiments, this invention uses historical operating data of the target generator unit during normal operation and startup as the basis for model training. This not only achieves a comprehensive characterization of the steady-state characteristics and dynamic behavior of the unit's thermodynamic system, but also significantly improves the generalization ability, prediction accuracy, and engineering applicability of the constructed NARX neural network model. This data strategy effectively supports reliable predictions under different startup conditions, solves the prediction distortion problem caused by insufficient data coverage in traditional methods, and provides a solid data foundation for intelligent auxiliary decision-making during unit startup.

[0041] Optionally, the NARX neural network model is configured to operate in a closed-loop feedback mode during the prediction process, wherein the model output at the current time step is fed back and used as one of the inputs for subsequent time steps to achieve continuous recursive prediction of the dynamic behavior of the system.

[0042] In the above implementation, the trained NARX neural network model adopts a closed-loop prediction architecture in the practical application stage. In this mode, the model's output at time t (such as main steam pressure and temperature) is re-injected into the model's input layer to participate in the prediction process at time t+1, thus enabling multi-step forward rolling prediction without external measured feedback. This mechanism effectively simulates the inertia and cumulative effects of a thermodynamic system.

[0043] Optionally, after each iteration of prediction, the NARX neural network model determines whether the predicted main steam pressure has reached the preset target pressure threshold. When the determination condition is met, the subsequent prediction calculation is terminated, and the complete evolution trajectory of the main steam parameters is output.

[0044] In the above implementation, an automatic termination logic is set: when the main steam pressure obtained by continuous prediction reaches the target pressure for steam inlet heating specified in the operating procedure, the system automatically ends the prediction process and outputs the pressure and temperature change curves of the entire process from the initial state to the target state, thereby significantly improving the practicality of the model and the efficiency of human-computer interaction.

[0045] In the embodiments of this application, see Figure 1, Figure 1 This is a structural schematic diagram of a combined cycle power unit. Among them, Figure 1 This demonstrates the typical structure of an F-class combined cycle unit. The gas turbine consists of three main components: compressor 1, combustion chamber 2, and turbine 3. The compressor and turbine are located on the same rotor, with the turbine providing power to the compressor. Air first enters the compressor for compression, then enters the combustion chamber to mix with natural gas and burn. The natural gas flow rate is controlled by natural gas valve 4. The high-temperature flue gas after combustion enters the turbine for expansion and work, and the waste heat from the flue gas is recovered and utilized by a waste heat boiler. The waste heat boiler consists of a first economizer 5, a first steam drum 6, a first evaporator 7, and a first superheater 8. Currently, most common F-class or H-class combined cycle units use a three-pressure reheat type waste heat boiler. Condensate enters the first economizer and is heated to saturated water. Steam-water separation is achieved in the first steam drum. The saturated water vaporizes into steam in the first evaporator, and the steam then enters the first superheater for further heating to become superheated steam. The superheated steam is controlled by steam valve 9 to enter the turbine 10 for expansion and work. The steam valves in the high-pressure steam system are called main steam control valves. The steam that has completed its work enters the condenser 11 and is cooled into condensate by the circulating water 12. Then, it passes through the condensate pump 14 and the condensate valve 15 and re-enters the waste heat boiler to complete the cycle. The gas turbine and the steam turbine together drive the generator 16 to generate electricity, which is then fed into the power grid via the transformer 17.

[0046] See Figure 2 , Figure 2 The high-pressure steam system distribution structure of the combined cycle unit is shown in detail. The waste heat boiler high-pressure system draws saturated water from the low-pressure steam drum, which is then pressurized again by the high-pressure feedwater pump 21. The pressurized water then sequentially enters the second economizer 22, the second steam drum 23, the second evaporator 24, and the second superheater 25 to complete heating, evaporation, and superheating, becoming high-pressure superheated steam. The high-pressure steam has two sets of pipelines. One set, controlled by the main steam regulating valve 26, enters the high-pressure cylinder 27 of the turbine for expansion and work before returning to the reheater of the waste heat boiler. The other set of pipelines, controlled by the high-pressure bypass valve 28, passes through the spray desuperheating 29 and directly enters the condenser; the desuperheating water comes from the high-pressure feedwater. The high-pressure bypass system plays a crucial role in the unit startup process.

[0047] See Figure 3 , Figure 3 This demonstrates the changes in high-pressure steam system parameters during the startup process of a Class F combined cycle generator unit. The startup process of the combined cycle generator unit gradually activates each major piece of equipment in a predetermined sequence.

[0048] First, start the gas turbine section and monitor the unit's speed curve: Before starting, a turning gear is needed to rotate the rotor at a low speed, and all auxiliary equipment should be started. The SFC unit drives the gas turbine to complete purging at 600 r / min and then begins to reduce speed. When the speed reaches the ignition speed, the fuel control valve opens, and the fuel is ignited in the combustion chamber. The high-temperature, high-pressure gas produced by combustion drives the turbine to rotate and do work. The gas turbine speed continues to increase, and when the gas turbine speed reaches its self-sustaining speed, the SFC unit disengages. Subsequently, the gas turbine continues to increase speed to the no-load speed (3000 r / min) and runs stably for a period of time to warm up, allowing the metal components to be slowly and evenly heated. After warm-up, the fuel quantity is gradually increased, and the gas turbine begins to carry the initial load, with its exhaust temperature and flow rate increasing accordingly.

[0049] The high-temperature flue gas discharged from the gas turbine is the driving force for starting the waste heat boiler. The high-temperature flue gas enters the waste heat boiler and flows sequentially through the first superheater, the first evaporator, and the first economizer. Before boiler startup, the first steam drum has been filled with water to the specified level according to regulations. The high-temperature flue gas exchanges heat with the water or steam in the boiler's heating surfaces. Pay attention to the main steam pressure curve: the heat from the flue gas first heats the feedwater in the economizer, and then heats the water in the evaporator to produce saturated steam. As the gas turbine reaches its initial load, the flue gas volume and temperature increase, the boiler steam production increases, and the steam pressure gradually rises. The saturated steam flows through the superheater and is further heated to become superheated steam. The key to boiler startup is controlling the rate of pressure and temperature increase to prevent excessive thermal stress caused by uneven or excessively rapid heating of metal components. Strict adherence to the preset startup curve is essential.

[0050] When the main steam pressure generated by the waste heat boiler reaches the turbine start-up requirement, the steam temperature has not yet reached the required level, and the main steam regulating valve of the turbine's high-pressure cylinder cannot be opened. At this time, the high-pressure bypass system needs to be used to discharge steam to the condenser. Attention must be paid to the high-pressure bypass valve curve and the main steam regulating valve curve: During this stage, the opening of the high-pressure bypass valve must be strictly controlled to ensure that the main steam pressure remains at the set value. Simultaneously, before the steam enters the condenser through the bypass pipeline, it needs to pass through a desuperheating and pressure-reducing device, with condensate injected to lower the steam temperature and pressure to accommodate the condenser's capacity. Once the main steam temperature reaches the required level, the main steam regulating valve of the high-pressure cylinder is slowly opened, allowing a small amount of steam to enter the turbine's high-pressure cylinder for warm-up. The main purpose is to use the heat of the steam to slowly heat the cylinder and rotor, causing them to expand evenly. As the main steam regulating valve gradually opens to the fully open state, the high-pressure bypass valve gradually closes. The operation of the bypass system allows the gas turbine and waste heat boiler to start and increase load according to plan without being restricted by the turbine's status, ensuring continuous boiler operation and improving steam parameters. During turbine startup and load ramp-up, the bypass system coordinates with the turbine inlet valves to gradually transfer steam from the bypass to the turbine, achieving a smooth transition. When the turbine enters sliding parameter operation, the main steam pressure increases rapidly, and the main steam regulating valve exhibits a "jump" curve, while the high-pressure bypass valve opens and closes accordingly. Therefore, the high-pressure bypass valve opens twice during unit startup. This invention only studies and predicts the main steam parameters before the first opening of the high-pressure bypass valve.

[0051] From the above description, it can be seen that the main purpose of the high-pressure bypass valve is to ensure that the main steam pressure before the main steam regulating valve remains at the set value during the turbine inlet warm-up phase. Analysis of the combined cycle unit's operating data reveals that the main steam pressure set value during this period in all start-up processes of the combined cycle unit is... The main steam pressure rises from 0 to the set value before the high-pressure bypass valve opens, and the pressures are all equal. The reason why the main steam pressure curves differ during each start-up of the combined cycle unit is that the exhaust temperature of the gas turbine varies. The exhaust temperature can be divided into three stages: the first stage is from gas turbine ignition to the gas turbine reaching its initial load, and the exhaust temperature is exactly the same during each start-up; the third stage is when the gas turbine exhaust temperature is a fixed set value. This value is related to the rotor metal temperature before the high-pressure cylinder warms up. Relatedly, at this point, the high-pressure bypass valve has been opened; in the intermediate second stage, the exhaust gas temperature changes according to a fixed rate of change. Upon reaching the set value, the high-pressure bypass valve opens during this stage. Therefore, predicting the main steam pressure and temperature change curves only requires studying the intermediate second stage. Figure 4 As shown, the main steam parameter prediction model includes two NARX neural network models during the training phase, with the gas turbine exhaust temperature as the input parameter. Main steam pressure and main steam temperature The output parameters are main steam pressure and temperature, respectively. Both models are open-loop configured, with a backward lookup of 5 time steps. When using the models, the gas turbine exhaust temperature setpoint is first calculated using the following formula: ; Then, the time variation value of the gas turbine exhaust temperature can be calculated from this: ; When using the NARX neural network model, it needs to be set to closed-loop mode, that is, the output parameters at the current moment are used as the input parameters at the next moment, and the calculation terminates when the main steam pressure reaches the set value.

[0052] According to a second aspect of the present invention, a system for predicting the main steam parameters before the inlet steam heater of a combined cycle unit is provided, employing the method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit as described in the first aspect, comprising: The data acquisition module is used to acquire historical operating data of the target generator set, including various observable input and output parameters closely related to the main steam system. The model building and training module is used to build a NARX neural network model, train the neural network model based on historical operating data, and obtain a trained main steam parameter prediction model; wherein, the NARX neural network model is used to learn the nonlinear dynamic mapping relationship and time series evolution characteristics between the input parameters and the key output parameters of the main steam. The model deployment module is used to deploy the main steam parameter prediction model on a general-purpose computing device. The parameter input module is used to obtain the initial state parameters under the current operating conditions before the generator set is started, and input the initial state parameters into the main steam parameter prediction model. The prediction execution module is used to call the main steam parameter prediction model and output the prediction trajectory of the key parameters of the main steam before the high-pressure bypass system is put into operation. The key parameters include at least the main steam pressure and the main steam temperature.

[0053] In the above implementation, operators can quickly obtain detailed predicted changes in the main steam parameters before the turbine high-pressure cylinder inlet steam warm-up operation begins, providing a basis for startup strategy formulation and risk prediction.

[0054] According to a third aspect of the present invention, an electronic device is provided, comprising: One or more processors; A storage unit for storing one or more programs that, when executed by one or more processors, enable the one or more processors to implement the method for predicting the main steam parameters before the combined cycle unit's inlet steam heater, as described in the first aspect.

[0055] According to a fourth aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, enables the method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit as described in the first aspect.

[0056] This invention can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of the invention.

[0057] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0058] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0059] The computer program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of the invention.

[0060] Various aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0061] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0062] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0063] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions. It will be known to those skilled in the art that implementation in hardware, implementation in software, and implementation using a combination of software and hardware are equivalent.

[0064] It is understood that the above embodiments are merely exemplary implementations used to illustrate the principles of the present invention, and the present invention is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit, characterized in that, Includes the following steps: Step S1: Obtain historical operating data of the target generator set, including various observable input and output parameters closely related to the main steam system; Step S2: Construct a NARX neural network model and train the neural network model based on historical operating data to obtain a trained main steam parameter prediction model; wherein, the NARX neural network model is used to learn the nonlinear dynamic mapping relationship and time series evolution characteristics between the input parameters and the key output parameters of the main steam. Step S3: Deploy the main steam parameter prediction model on a general-purpose computing device; Step S4: Before starting the generator set, obtain the initial state parameters under the current operating conditions and input the initial state parameters into the main steam parameter prediction model; Step S5: The main steam parameter prediction model outputs the predicted trajectory of the key parameters of the main steam before the high-pressure bypass system is put into operation. The key parameters include at least the main steam pressure and the main steam temperature.

2. The method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit according to claim 1, characterized in that, The input parameters include gas turbine exhaust temperature, steam turbine high-pressure cylinder metal temperature, main steam pressure, main steam temperature, and high-pressure bypass valve opening.

3. The method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit according to claim 2, characterized in that, The output parameters include the gas turbine exhaust temperature, main steam pressure, and main steam temperature change curve before the high-pressure bypass valve is opened.

4. The method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit according to claim 3, characterized in that, The main steam parameter prediction model operates under the condition that the computational resource consumption is lower than a preset threshold.

5. The method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit according to claim 4, characterized in that, The historical operating data of the target generator set includes historical operating data during normal operation and startup.

6. The method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit according to claim 5, characterized in that, The NARX neural network model is configured to operate in a closed-loop feedback mode during the prediction process, wherein the model output at the current time step is fed back and used as one of the inputs for subsequent time steps to achieve continuous recursive prediction of the dynamic behavior of the system.

7. The method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit according to claim 6, characterized in that, After each iteration of prediction, the NARX neural network model determines whether the predicted main steam pressure has reached the preset target pressure threshold. When the determination condition is met, the subsequent prediction calculation is terminated, and the complete evolution trajectory of the main steam parameters is output.

8. A system for predicting main steam parameters before the inlet steam heater of a combined cycle unit, characterized in that, The method for predicting the main steam parameters before the inlet steam heater of a combined cycle unit as described in any one of claims 1 to 7 includes: The data acquisition module is used to acquire historical operating data of the target generator set, including various observable input and output parameters closely related to the main steam system. The model building and training module is used to build a NARX neural network model, train the neural network model based on historical operating data, and obtain a trained main steam parameter prediction model; wherein, the NARX neural network model is used to learn the nonlinear dynamic mapping relationship and time series evolution characteristics between the input parameters and the key output parameters of the main steam. The model deployment module is used to deploy the main steam parameter prediction model on a general-purpose computing device. The parameter input module is used to obtain the initial state parameters under the current operating conditions before the generator set is started, and input the initial state parameters into the main steam parameter prediction model. The prediction execution module is used to call the main steam parameter prediction model and output the prediction trajectory of the key parameters of the main steam before the high-pressure bypass system is put into operation. The key parameters include at least the main steam pressure and the main steam temperature.

9. An electronic device, characterized in that, include: One or more processors; A storage unit for storing one or more programs that, when executed by one or more processors, enable the one or more processors to implement the method for predicting the main steam parameters before the combined cycle unit's inlet steam heater according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it can implement the method for predicting the main steam parameters before the combined cycle unit's inlet steam heater according to any one of claims 1 to 7.

Images