Turbine control valve operation method
By using the combined valve and sequential valve switching strategy and machine learning predictive load changes in the steam turbine, it solves the problem of load and pressure fluctuations existing during the valve sequence switching process of the turbine.
Patent Information
- Application Number
- CN202411323701.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-23
- Publication Date
- 2025-01-28
AI Technical Summary
The steam turbine door control has load and pressure fluctuations during valve sequence switching, which may lead to mechanical damage and abnormal operation, and the existing control systems lack intelligent prediction and adaptation to load changes.
The switching strategy of the combination valve and the order valve is used to switch between the control point at the critical load point, and the valve switching is performed by the synergistic operation of the manual confirmation and the digital electrical control system. At the same time, use machine learning algorithm to predict the trend of load changes, and automatically adjust the door opening to maintain the stable load.
Reduced the pressure of the unit through stable transitional control to reduce the pressure of the load, improve the safety and flexibility of operation, optimize the use of resource, enhance the refinement of load control, and reduce the time of non -planning shutdown.
Smart Images

Figure BDA0005054771330000114 
Figure BDA0005054771330000121 
Figure BDA0005054771330000122
Abstract
Description
Technical Field
[0001] The invention relates to a method for operating a steam turbine regulating valve, and belongs to the field of steam turbine regulating systems for thermal power units. Background Art
[0002] The valve control of the steam turbine is a key link in the operation of the power plant, which directly affects the stability and safety of the unit. Whether the valve control is proper or not is directly related to the economic benefits and operational reliability of the entire power plant. When switching the valve sequence of the steam turbine, there are some technical challenges and potential risks, which may cause large fluctuations in parameters such as load and pressure, and even cause accidents.
[0003] First, the valve control system of the steam turbine usually needs to ensure the stability of steam parameters (such as temperature and pressure) while ensuring that the output power meets the needs of the power grid. During the valve sequence switching process, the opening and closing sequence of the valves may change sharply, causing fluctuations in load and pressure. If this fluctuation is not controlled, it may cause damage to the blades and bearings of the steam turbine, and even cause abnormal vibration or breakage of the unit.
[0004] Secondly, traditional valve control systems often rely on the experience and skills of operators, which may lead to untimely response or operational errors in emergency situations. The uncertainty of human factors increases the risk of power plant operation. In addition, existing control systems may lack sufficient intelligence to predict and adapt to load changes, resulting in the inability to make fast and accurate adjustments during valve sequence switching.
[0005] Furthermore, the dynamic response characteristics of steam turbines are complex, involving multiple fields such as thermodynamics, fluid mechanics, and mechanical vibration. When switching valve sequences, these factors need to be considered comprehensively to ensure a smooth transition of the system. However, existing control strategies may not fully consider all these dynamic factors, resulting in control blind spots during the switching process.
[0006] Finally, the valve control system of the steam turbine usually needs to work in coordination with other systems in the power plant (such as boilers, generators, and power grids). During the valve sequence switching process, if the coordination between these systems is not good, it may lead to a reduction in the operating efficiency of the entire power plant, or even trigger a chain reaction, affecting the stability of the power grid. Summary of the invention
[0007] In order to overcome the defects of the prior art, the present invention provides a method for operating a steam turbine valve adjustment. The technical solution of the present invention is:
[0008] A method for operating a steam turbine valve adjustment valve comprises the following steps:
[0009] During startup, the control mode of the regulating valve is set to the combination valve control mode, and when the unit load reaches 500MW, it is switched to the sequence valve control mode;
[0010] During shutdown, when the unit load drops to 500MW, it switches to combined valve control;
[0011] During the valve switching process, confirm that the unit is running stably, switch the unit coordinated control mode to manual control mode, manually open the steam turbine main control to 100%, and perform valve switching. After the valve switching is completed, adjust the steam turbine main control to the corresponding opening of the load.
[0012] In the valve switching process, the switching is performed manually; when the combination valve is switched to the sequence valve, on the "control panel" of the digital electro-hydraulic control system, click the valve switching button, check that the indicator light of the switching in progress is on, indicating that the valve switching is in progress, and after the switching is completed, check that the sequence valve indicator light is on and the switching valve indicator light is off, indicating that the sequence valve control has been switched.
[0013] When the sequence valve is switched to a combination valve, click the valve switching button and check that the valve switching light is on during the switching, indicating that valve switching is in progress. After the switching is completed, check that the sequence valve indicator light and the switching valve light are off, indicating that the combination valve control has been switched; the valve switching time is 8-12 minutes. During the switching process, it is prohibited to operate the valve switch.
[0014] When the unit load is reduced to 600MW, the unit is in sliding pressure operation with two valves fully open, and the steam distribution method is #1 and #3 valve steam intake. When the unit load is reduced to below 700MW, the unit TSI parameters are monitored.
[0015] During normal operation of the unit, when the load reaches above 600MW, the #1 and #3 regulating valves are fully opened, and the #2 and #4 regulating valves and sliding pressure are used to control the unit load.
[0016] The step of predicting the valve opening is also included, as follows:
[0017] 6.1 Collect real-time load data and historical operation data;
[0018] 6.2 Use machine learning algorithms to analyze data and identify load change trends;
[0019] 6.3 Calculate the optimal valve opening according to trend prediction;
[0020] 6.4 Automatically adjust the valve opening to maintain load stability.
[0021] The step 6.1 is specifically as follows:
[0022] Define the scope of data collection: Real-time load data includes but is not limited to the current unit load, steam flow, turbine speed and valve opening; historical operation data includes parameters recorded in past operations and operation result data;
[0023] Implement data acquisition: Use sensors and data acquisition systems to monitor turbine parameters in real time and transmit data to a central processing unit;
[0024] Data preprocessing: Clean the collected data, including removing outliers, filling missing values, and normalizing the data to ensure data quality and consistency.
[0025] The step 6.2 is specifically as follows:
[0026] Select machine learning model: choose the appropriate machine learning algorithm based on data characteristics and prediction goals;
[0027] Training model: Use historical operation data to train the selected machine learning model and adjust model parameters until the model can accurately predict load change trends;
[0028] Feature Engineering: Identify and select the features that have the greatest impact on load variation prediction.
[0029] The step 6.3 is specifically as follows:
[0030] Define optimization goals: Determine the goals of the valve operation;
[0031] Establish mathematical model: construct a mathematical model to quantify the relationship between load change trend and valve opening;
[0032] Optimization algorithm: Apply the optimization algorithm to calculate the optimal valve opening based on load forecast and optimization objectives.
[0033] The step 6.4 is specifically as follows:
[0034] Implement control strategy: According to the calculated optimal valve opening, the valve position is automatically adjusted through the digital control system;
[0035] Real-time monitoring and feedback: After the valve is adjusted, the system response is monitored in real time and compared with the predicted value;
[0036] Closed-loop control: If the actual response deviates from the predicted or target value, the system will adjust the valve opening according to the feedback to form a closed-loop control cycle to ensure load stability;
[0037] Safety and exception handling: When an abnormality or potential risk is detected, the safety protocol is immediately initiated. The specific method for establishing the mathematical model is:
[0038] (1) Define all key variables in the model: P(t): turbine load at time t (MW); x(t): throttle opening at time t (%); u(t): steam flow rate at time t (kg / s); p(t): steam pressure at time t (bar); θ(t): turbine speed at time t (RPM);
[0039] (2) Establish the basic equation:
[0040] The mass conservation equation: (Assuming the steam flow is constant);
[0041] Energy conservation equation: P(t) = η·m(t)·(h out -h in ), where η is the efficiency, h out and h in are the outlet and inlet enthalpies, respectively;
[0042] Momentum conservation equation: Where J is the moment of inertia, τ(t) is the torque, and T b is the damping torque;
[0043] (3) Considering nonlinear characteristics: The flow characteristics of the valve are usually nonlinear. The nonlinear function f(x) is obtained by fitting the experimental data:
[0044] Nonlinear flow equation: u(t) = f(x(t)) = k·x(t) n , where k and n are parameters obtained by fitting experimental data.
[0045] (4) Dynamic response model: Consider the dynamic response of the system, such as time delay and hysteresis effects:
[0046] Dynamic equation: τ(t) = k t ·u(t)·p(t), where kt is the torque coefficient;
[0047] (5) Construct a complete mathematical model:
[0048] Combining the above equations, we can build a complete mathematical model:
[0049] Complete dynamic system model: The advantages of the present invention are:
[0050] Sliding load transition: By switching the throttle control mode at key load points (such as 500MW), the unit can ensure a smooth transition at different load stages and reduce the mechanical and thermal stress caused by sudden load changes.
[0051] Improve operational safety: During the valve switching process, manually confirming that the unit is operating stably and switching the control mode can reduce the risks caused by automatic control errors and improve operational safety.
[0052] Reduce human errors: Although valve switching requires manual operation, the possibility of human operating errors can be reduced through clear indicator light prompts and strict operating procedures.
[0053] Optimize resource utilization: In the low load stage (such as below 600MW), full-open sliding pressure operation can optimize the use of steam and improve the thermal efficiency of the unit.
[0054] Strengthen monitoring: Strengthening the monitoring of TSI parameters at key load points (such as below 700MW) will help to detect potential problems in a timely manner and avoid accidents.
[0055] Refined load control: In normal operation, the use of partial throttle and sliding pressure to control the load can achieve more precise load regulation and improve the unit's ability to respond to load changes.
[0056] Predictive maintenance: By collecting real-time and historical data and using machine learning algorithms to predict load change trends, the valve opening can be adjusted in advance to achieve predictive maintenance and reduce unplanned downtime. DETAILED DESCRIPTION
[0057] The present invention will be further described below in conjunction with specific embodiments, and the advantages and features of the present invention will become clearer as the description proceeds. However, these embodiments are exemplary only and do not constitute any limitation to the scope of the present invention. It should be understood by those skilled in the art that the details and forms of the technical solution of the present invention may be modified or replaced without departing from the spirit and scope of the present invention, but these modifications and replacements all fall within the scope of protection of the present invention.
[0058] The present invention relates to a method for operating a steam turbine valve, comprising the following steps:
[0059] During startup, the control mode of the regulating valve is set to the combination valve control mode, and when the unit load reaches 500MW, it is switched to the sequence valve control mode;
[0060] During shutdown, when the unit load drops to 500MW, it switches to combined valve control;
[0061] During the valve switching process, confirm that the unit is running stably, switch the unit coordinated control mode to manual control mode, manually open the steam turbine main control to 100%, and perform valve switching. After the valve switching is completed, adjust the steam turbine main control to the corresponding opening of the load.
[0062] Based on the settings of the above scheme, the following are achieved:
[0063] Ensure smooth transition of startup and shutdown: By switching the throttle control mode when the unit load reaches or drops to 500MW, a smooth transition at key load points can be ensured, reducing stress and wear on the unit.
[0064] Enhanced operational flexibility: Using manual control during valve switching allows operators to make fine adjustments based on the actual operating conditions of the unit, improving operational flexibility and adaptability.
[0065] Improve the safety of unit operation: By confirming that the unit is operating stably before valve switching and adjusting the steam turbine main control to the appropriate opening after switching, the safety risks caused by control errors can be reduced.
[0066] Reduce load fluctuations: After the valve is switched, the main control opening of the steam turbine is adjusted according to the load, which helps to reduce load fluctuations and improve the stability of power supply.
[0067] In the valve switching process, the switching is performed manually; when the combination valve is switched to the sequence valve, on the "control panel" of the digital electro-hydraulic control system, click the valve switching button, check that the indicator light of the switching in progress is on, indicating that the valve switching is in progress, and after the switching is completed, check that the sequence valve indicator light is on and the switching valve indicator light is off, indicating that the sequence valve control has been switched.
[0068] When the sequence valve is switched to a combination valve, click the valve switching button and check that the valve switching light is on during the switching, indicating that valve switching is in progress. After the switching is completed, check that the sequence valve indicator light and the switching valve light are off, indicating that the combination valve control has been switched; the valve switching time is 8-12 minutes. During the switching process, it is prohibited to operate the valve switch.
[0069] When the load of the unit is reduced to 600MW, the unit is in sliding pressure operation with two valves fully open, and the steam distribution mode is #1 and #3 valve steam inlet (both arranged on the upper part of the high-pressure cylinder). The unit vibration, thrust bearing temperature, axial displacement, and #1, 2, 3 support bearing temperature may all change. When the unit load is reduced to below 700MW, the unit TSI parameters are monitored, especially the main engine vibration, thrust bearing temperature, axial displacement, and #1, 2, 3 support bearing temperature.
[0070] During the normal operation of the unit, when the load exceeds 600MW, the #1 and #3 regulating valves are fully opened, and the #2 and #4 regulating valves and sliding pressure are used to control the unit load. Since the adjustment curves of the #2 and #4 regulating valves are steeper, when the main control of the steam turbine changes, the #2 and #4 regulating valves fluctuate greatly.
[0071] The step of predicting the valve opening is also included, as follows:
[0072] 6.1 Collect real-time load data and historical operation data;
[0073] 6.2 Use machine learning algorithms to analyze data and identify load change trends;
[0074] 6.3 Calculate the optimal valve opening according to trend prediction;
[0075] 6.4 Automatically adjust the valve opening to maintain load stability.
[0076] The step 6.1 is specifically as follows:
[0077] Define the scope of data collection: Real-time load data includes but is not limited to the current unit load, steam flow, turbine speed and valve opening; historical operation data includes parameters recorded in past operations, as well as operation result data; such as load changes, valve response time, and system stability indicators.
[0078] Implement data acquisition: Use sensors and data acquisition systems to monitor turbine parameters in real time and transmit data to a central processing unit;
[0079] Data preprocessing: Clean the collected data, including removing outliers, filling missing values, and normalizing the data to ensure data quality and consistency.
[0080] The step 6.2 is specifically as follows:
[0081] Choose a machine learning model: Based on the data characteristics and prediction goals, choose a suitable machine learning algorithm; such as time series analysis, support vector machine, neural network or deep learning model.
[0082] Training model: Use historical operation data to train the selected machine learning model and adjust model parameters until the model can accurately predict load change trends;
[0083] Feature engineering: Identify and select the features that have the greatest impact on load change prediction, such as seasonal factors, operating modes, external environmental conditions, etc., to improve the prediction accuracy of the model. The specific steps of step 6.3 are:
[0084] Define optimization goals: Determine the goal of the throttle operation; such as minimizing load fluctuations, maximizing efficiency, or balancing multiple operating parameters.
[0085] Establish a mathematical model: Construct a mathematical model to quantify the relationship between load change trend and regulating valve opening, taking into account the dynamic response and nonlinear characteristics of the system.
[0086] Optimization algorithm: Apply optimization algorithms, such as genetic algorithm, particle swarm optimization or gradient descent method, to calculate the optimal valve opening according to load forecast and optimization objectives.
[0087] The step 6.4 is specifically as follows:
[0088] Implement control strategy: According to the calculated optimal valve opening, the valve position is automatically adjusted through the digital control system;
[0089] Real-time monitoring and feedback: After the valve is adjusted, the system response is monitored in real time, such as load changes, steam pressure, etc., and compared with the predicted values;
[0090] Closed-loop control: If the actual response deviates from the predicted or target value, the system will adjust the valve opening according to the feedback to form a closed-loop control cycle to ensure load stability;
[0091] Safety and exception handling: When an abnormality or potential risk is detected, safety protocols are immediately initiated, such as emergency shutdown or adjustment to a safe operating mode.
[0092] The specific method for establishing the mathematical model is:
[0093] (1) Define all key variables in the model: P(t): turbine load at time t (MW); x(t): throttle opening at time t (%); u(t): steam flow rate at time t (kg / s); p(t): steam pressure at time t (bar); θ(t): turbine speed at time t (RPM);
[0094] (2) Establish the basic equation:
[0095] The mass conservation equation: (Assuming the steam flow is constant);
[0096] Energy conservation equation: P(t) = η·m(t)·(h out -h in ), where η is the efficiency, h out and h in are the outlet and inlet enthalpies, respectively;
[0097] Momentum conservation equation: Where J is the moment of inertia, τ(t) is the torque, and T b is the damping torque;
[0098] (3) Considering nonlinear characteristics: The flow characteristics of the valve are usually nonlinear. The nonlinear function f(x) is obtained by fitting the experimental data:
[0099] Nonlinear flow equation: u(t) = f(x(t)) = k·x(t) n , where k and n are parameters obtained by fitting experimental data.
[0100] (4) Dynamic response model: Consider the dynamic response of the system, such as time delay and hysteresis effects:
[0101] Dynamic equation: τ(t) = k t ·u(t)·p(t), where kt is the torque coefficient;
[0102] (5) Construct a complete mathematical model:
[0103] Combining the above equations, we can build a complete mathematical model:
[0104] Complete dynamic system model: Among them, the corresponding relationship between the steam turbine master control and the valve command is as shown in Table 1:
[0105]
[0106]
[0107] Table 1 Main steam pressure corresponding to different loads controlled by steam turbine sequence valve (as shown in Table 2):
[0108]
[0109] Table 2
[0110] In the present invention, the throttle operation of the steam turbine is precisely controlled to adapt to different operating conditions and load requirements, thereby ensuring the stability and safety of the unit. The following is a detailed explanation of the working principle and key steps of the method:
[0111] Control mode switching during startup and shutdown: During startup, the combination valve control mode is initially used, which generally provides more refined control and helps to increase the load smoothly. When the unit load reaches 500MW, it switches to sequence valve control, which may provide a control strategy more suitable for high-load operation. During shutdown, when the load drops to 500MW, it switches back to combination valve control to ensure smooth load reduction.
[0112] Manual control and valve switching: During the valve switching process, first confirm that the unit is running stably, and then switch the coordinated control mode to manual control mode. The operator manually opens the turbine master control to 100%, and then switches the valve to ensure accurate control of the switching process. After the switch is completed, adjust the turbine master control to the corresponding opening according to the current load to maintain load stability.
[0113] Manual valve switching operation: On the control panel of the digital electro-hydraulic control system, the operator performs the switching by clicking the valve switching button. The switching status is monitored by observing the status of the indicator lights (such as the "in progress" and "sequential valve" indicators).
[0114] Low-load operation monitoring: When the unit load drops to 600MW, the two valves are fully opened to operate in sliding pressure mode, with special attention paid to key parameters such as unit vibration and thrust bearing temperature. When the load drops below 700MW, the monitoring of TSI parameters is strengthened to ensure the safety of the unit in critical conditions.
[0115] Load control in normal operation: In load operation above 600MW, use partial throttle and sliding pressure control to maintain the load, paying special attention to the throttle adjustment curve and possible fluctuations.
[0116] Predictive throttle opening adjustment: By collecting real-time and historical load data, machine learning algorithms are used to analyze and predict load change trends. Based on the prediction results, the optimal throttle opening is calculated and automatically adjusted through the digital control system to maintain load stability.
[0117] Establishment and application of mathematical models: define and collect key operating parameters such as load, valve opening, steam flow, pressure and speed. Establish basic equations including conservation of mass, energy and momentum, taking into account nonlinear characteristics and system dynamic response; build a complete mathematical model for predicting and optimizing valve operation.
[0118] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.
Claims
1. A method for operating a steam turbine valve, characterized in that: The following steps are involved: During startup, the control mode of the regulating valve is set to the combination valve control mode, and when the unit load reaches 500MW, it is switched to the sequence valve control mode; During shutdown, when the unit load drops to 500MW, it switches to combined valve control; During the valve switching process, confirm that the unit is running stably, switch the unit coordinated control mode to manual control mode, manually open the steam turbine main control to 100%, and perform valve switching. After the valve switching is completed, adjust the steam turbine main control to the corresponding opening of the load.
2. The method for operating a steam turbine valve regulating valve according to claim 1, characterized in that: In the valve switching process, the switching is performed manually; when the combination valve is switched to the sequence valve, on the "control panel" of the digital electro-hydraulic control system, the valve switching button is clicked to check that the indicator light during the switching is on, indicating that the valve switching is in progress. After the switching is completed, check that the indicator light of the sequence valve is on and the indicator light of the switching valve is off, indicating that the control has been switched to the sequence valve.
3. The method for operating a steam turbine valve regulating valve according to claim 1 or 2, characterized in that: When the sequence valve is switched to a combination valve, click the valve switching button and check that the valve switching light is on during the switching, indicating that valve switching is in progress. After the switching is completed, check that the sequence valve indicator light and the switching valve light are off, indicating that the combination valve control has been switched; the valve switching time is 8-12 minutes. During the switching process, it is prohibited to operate the valve switch.
4. The method for operating a steam turbine valve regulating valve according to claim 3, characterized in that: When the unit load is reduced to 600MW, the unit is in sliding pressure operation with two valves fully open, and the steam distribution method is #1 and #3 valve steam intake. When the unit load is reduced to below 700MW, the unit TSI parameters are monitored.
5. The method for operating a steam turbine valve regulating valve according to claim 4, characterized in that: During normal operation of the unit, when the load reaches above 600MW, the #1 and #3 regulating valves are fully opened, and the #2 and #4 regulating valves and sliding pressure are used to control the unit load.
6. The method for operating a steam turbine valve regulating valve according to claim 5, characterized in that: The step of predicting the valve opening is also included, as follows: 6.1 Collect real-time load data and historical operation data; 6.2 Use machine learning algorithms to analyze data and identify load change trends; 6.3 Calculate the optimal valve opening according to trend prediction; 6.4 Automatically adjust the valve opening to maintain load stability.
7. The method for operating a steam turbine valve regulating valve according to claim 6, characterized in that: The step 6.1 is specifically as follows: Define the scope of data collection: Real-time load data includes but is not limited to the current unit load, steam flow, turbine speed and valve opening; historical operation data includes parameters recorded in past operations and operation result data; Implement data acquisition: Use sensors and data acquisition systems to monitor turbine parameters in real time and transmit data to a central processing unit; Data preprocessing: Clean the collected data, including removing outliers, filling missing values, and normalizing the data to ensure data quality and consistency.
8. The method for operating a steam turbine valve regulating valve according to claim 6, characterized in that: The step 6.2 is specifically as follows: Select machine learning model: choose the appropriate machine learning algorithm based on data characteristics and prediction goals; Training model: Use historical operation data to train the selected machine learning model and adjust model parameters until the model can accurately predict load change trends; Feature Engineering: Identify and select the features that have the greatest impact on load variation prediction.
9. The method for operating a steam turbine valve regulating valve according to claim 6, characterized in that: The step 6.3 is specifically as follows: Define optimization goals: Determine the goals of the valve operation; Establish mathematical model: construct a mathematical model to quantify the relationship between load change trend and valve opening; Optimization algorithm: Apply the optimization algorithm to calculate the optimal valve opening based on load forecast and optimization objectives. The step 6.4 is specifically as follows: Implement control strategy: According to the calculated optimal valve opening, the valve position is automatically adjusted through the digital control system; Real-time monitoring and feedback: After the valve is adjusted, the system response is monitored in real time and compared with the predicted value; Closed-loop control: If the actual response deviates from the predicted or target value, the system will adjust the valve opening according to the feedback to form a closed-loop control cycle to ensure load stability; Security and exception handling: When anomalies or potential risks are detected, security protocols are immediately initiated.
10. The method for operating a steam turbine valve regulating valve according to claim 9, characterized in that: The specific method for establishing the mathematical model is: (1) Define all key variables in the model: P(t): turbine load at time t (MW); x(t): throttle opening at time t (%); u(t): steam flow rate at time t (kg / s); p(t): steam pressure at time t (bar); θ(t): turbine speed at time t (RPM); (2) Establish the basic equation: The mass conservation equation: (Assuming the steam flow is constant); Energy conservation equation: P(t) = η·m(t)·(h out -h in ), where η is the efficiency, h out and h in are the outlet and inlet enthalpies, respectively; Momentum conservation equation: Where J is the moment of inertia, τ(t) is the torque, and T b is the damping torque; (3) Considering nonlinear characteristics: The flow characteristics of the valve are usually nonlinear. The nonlinear function f(x) is obtained by fitting the experimental data: Nonlinear flow equation: u(t)=f(x(t))=k·x(t)n, where k and n are parameters obtained by fitting experimental data. (4) Dynamic response model: Consider the dynamic response of the system, such as time delay and hysteresis effects: Dynamic equation: τ(t) = k t ·u(t)·p(t), where kt is the torque coefficient; (5) Construct a complete mathematical model: Combining the above equations, we can build a complete mathematical model: Complete dynamic system model: