A method and system for controlling the opening of a turbine governor

Abstract

The present invention provides a method and system for controlling the opening of a turbine governor. The method comprises the following steps: obtaining water head data, a frequency signal, and an initial PID control signal of the turbine; generating a difference correction coefficient by combining the water head data and a preset power setting; correcting the initial difference coefficient preset in the turbine using the difference correction coefficient to obtain an adjustment difference coefficient; generating an adjustment PID control signal by combining the frequency signal, the initial PID control signal, and the adjustment difference coefficient; generating an opening control signal by combining the water head data, the power setting, and a preset opening setting; and superimposing the adjustment PID control signal with the opening control signal to generate a control signal for the turbine. The present invention eliminates the need to set a dead zone for power or opening regulation and solves the problem that a single frequency regulation may not operate or operate incorrectly under small frequency differences.

Description

Technical Field

[0001] The invention belongs to the technical field of hydropower unit control, and in particular relates to a method and system for controlling the opening of a turbine governor. Background Art

[0002] In the power system, hydropower units shoulder a significant portion of the peak and frequency regulation tasks. Hydropower units must promptly and rapidly adjust their output power based on grid dispatch requirements and grid frequency fluctuations to meet the grid's active power balance and frequency regulation needs. Two control methods are commonly used for grid-connected hydropower units. One, aperture control, targets aperture control and offers excellent stability, is unaffected by external interference, and its regulation parameters are minimally affected by operating conditions. However, this control objective struggles to meet the grid's power regulation requirements. Therefore, another regulation method, power control, is widely adopted in practice.

[0003] However, the power regulation mode is a feedback control method, which uses the difference between target power and actual power to generate the turbine relay control output according to a certain power law. Its control parameters are significantly affected by operating conditions, making parameter tuning difficult. In addition, due to the influence of turbine water pressure pulsation, the feedback power will fluctuate slightly under stable conditions. To reduce frequent and unnecessary operation of the turbine governor, a certain power dead zone is often set. Due to the existence of the power regulation dead zone, when the target power caused by grid frequency fluctuations is less than the dead zone set value, the turbine governor will not produce any regulatory effect. As a result, the primary frequency regulation performance of the small frequency difference turbine governor cannot meet the grid assessment requirements. To address this situation, more reasonable and effective turbine governor control methods are needed. Summary of the Invention

[0004] The present invention provides a method and system for controlling the opening of a turbine governor, so as to solve the problem that the primary frequency regulation performance of a small frequency difference turbine governor cannot meet the requirements of power grid assessment.

[0005] In a first aspect, the present invention provides a method for controlling the opening of a turbine governor, the method comprising the following steps:

[0006] Obtain the water head data, frequency signal and initial PID control signal of the turbine;

[0007] generating a difference correction coefficient by combining the water head data and a preset power setting;

[0008] Correcting an initial differential coefficient preset in the turbine by using the differential correction coefficient to obtain an adjusted differential coefficient;

[0009] generating an adjusted PID control signal by combining the frequency signal, the initial PID control signal, and the adjustment difference coefficient;

[0010] generating an opening control signal by combining the water head data, the power setting and the preset opening setting;

[0011] The adjustment PID control signal is superimposed on the opening control signal to generate a control signal for the water turbine.

[0012] Optionally, generating a difference correction coefficient by combining the water head data and a preset power setting includes the following steps:

[0013] Perform positive bias on the preset power setting to obtain the positive bias operating condition point;

[0014] Calculating a first opening prediction value by combining the positive bias operating point and the water head data;

[0015] Negatively bias the preset power setting to obtain the negative bias operating condition point;

[0016] Calculating a second opening prediction value by combining the negative bias operating condition point and the water head data;

[0017] A difference correction coefficient is calculated by combining the first opening prediction value and the second opening prediction value.

[0018] Optionally, the frequency signal includes frequency data and a preset frequency setting, and the generating of the adjusted PID control signal by combining the frequency signal, the initial PID control signal, and the adjustment difference coefficient includes the following steps:

[0019] Calculating a frequency error according to the frequency data and the frequency setting;

[0020] generating a control error signal by combining the frequency error, the initial PID control signal, and the adjustment difference coefficient;

[0021] An adjustment PID control signal is generated based on the control error signal through a PID control algorithm.

[0022] Optionally, the generating of the opening control signal by combining the water head data, the power setting and the preset opening setting comprises the following steps:

[0023] Determining whether the operation mode of the turbine is a grid-connected operation mode or a non-grid-connected operation mode;

[0024] If the operation mode of the turbine is the non-grid-connected operation mode, a preset opening is given as an opening control signal;

[0025] If the operation mode of the turbine is the grid-connected operation mode, the opening control signal is generated in combination with the water head data and the power setting.

[0026] Optionally, the step of combining the water head data and the power setting to generate the opening control signal comprises the following steps:

[0027] Acquiring historical operating data of the turbine;

[0028] Building and training a convolutional neural network model based on the historical operating data;

[0029] Generate an output vector according to the water head data and the power setting;

[0030] The output vector is output to the convolutional neural network model to obtain an opening prediction signal, and the opening prediction signal is used as the opening control signal.

[0031] In a second aspect, the present invention further provides a turbine governor opening control system, the system comprising:

[0032] A data acquisition module is used to obtain the water head data, frequency signal and initial PID control signal of the turbine;

[0033] A correction coefficient generating module, configured to generate a difference correction coefficient by combining the water head data and a preset power setting;

[0034] A difference coefficient adjustment module, configured to correct an initial difference coefficient preset in the turbine using the difference correction coefficient to obtain an adjusted difference coefficient;

[0035] a PID control module, configured to generate an adjusted PID control signal by combining the frequency signal, the initial PID control signal, and the adjustment difference coefficient;

[0036] An opening control switching module, configured to generate an opening control signal by combining the water head data, the power setting and a preset opening setting;

[0037] A control output module is used to superimpose the adjustment PID control signal and the opening control signal to generate a control signal for the turbine.

[0038] Optionally, the correction coefficient generation module includes:

[0039] The positive bias unit is used to positively bias the preset power setting to obtain the positive bias operating condition point;

[0040] A first opening prediction unit, configured to calculate a first opening prediction value by combining the positive bias operating point and the water head data;

[0041] A negative bias unit is used to negatively bias the preset power setting to obtain a negative bias operating condition point;

[0042] A second opening prediction unit, configured to calculate a second opening prediction value by combining the negative bias operating point and the water head data;

[0043] A correction coefficient calculation unit is used to calculate a difference correction coefficient by combining the first opening prediction value and the second opening prediction value.

[0044] Optionally, the frequency signal includes frequency data and a preset frequency setting, and the PID control module includes:

[0045] A frequency error calculation unit, configured to calculate a frequency error based on the frequency data and the given frequency;

[0046] an error signal generating unit, configured to generate a control error signal by combining the frequency error, the initial PID control signal, and the adjustment difference coefficient;

[0047] The PID control signal generating unit is used to generate an adjustment PID control signal according to the control error signal and through a PID control algorithm.

[0048] Optionally, the opening control switching module includes:

[0049] an operation mode determination unit, configured to determine whether the operation mode of the hydraulic turbine is a grid-connected operation mode or a non-grid-connected operation mode;

[0050] The first opening control unit is preset with an opening setting, and is used to output the opening setting as an opening control signal when the judgment result of the operation mode judgment unit is that the operation mode of the turbine is the non-grid-connected operation mode.

[0051] Optionally, the opening control switching module further includes:

[0052] A historical data acquisition unit, configured to acquire historical operating data of the hydraulic turbine;

[0053] A model building unit, configured to build and train a convolutional neural network model based on the historical operating data;

[0054] A vector generating unit, configured to generate an output vector according to the water head data and the power setting;

[0055] A second opening control unit is used to output the output vector to the convolutional neural network model to obtain an opening prediction signal when the judgment result of the operation mode judgment unit is that the operation mode of the turbine is the grid-connected operation mode, and output the opening prediction signal as the opening control signal.

[0056] The beneficial effects of the present invention are:

[0057] The turbine governor opening control method provided by the present invention includes the following steps: obtaining the turbine's head data, frequency signal, and initial PID control signal; combining the head data and a preset power setting to generate a difference correction coefficient; correcting the preset initial difference coefficient in the turbine using the difference correction coefficient to obtain an adjusted difference coefficient; combining the frequency signal, the initial PID control signal, and the adjusted difference coefficient to generate an adjusted PID control signal; combining the head data, the power setting, and a preset opening setting to generate an opening control signal; and superimposing the adjusted PID control signal with the opening control signal to generate a control signal for the turbine. Therefore, the present invention retains the advantages of the traditional turbine governor opening control method, which is good stability and is not subject to interference from external factors, while achieving fine control of turbine power through power opening prediction and accurate control of primary frequency modulation through adaptive difference coefficient correction and adjustment. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a flow chart of a method for controlling the opening of a turbine governor in one embodiment of the present application.

[0059] Figure 2 This is a schematic diagram of a process for generating a difference correction coefficient in one embodiment of the present application.

[0060] Figure 3 This is a flow chart of generating and adjusting a PID control signal in one embodiment of the present application.

[0061] Figure 4 This is a schematic diagram of the process of generating an opening control signal in one embodiment of the present application. Figure 1 .

[0062] Figure 5 This is a schematic diagram of the process of generating an opening control signal in one embodiment of the present application. Figure 2 .

[0063] Figure 6 This is a structural diagram of a turbine governor opening control system in one embodiment of the present application.

[0064] Description of reference numerals:

[0065] 1. Data acquisition module; 2. Correction coefficient generation module; 3. Difference coefficient adjustment module; 4. PID control module; 5. Opening control switching module; 6. Control output module; 21. Positive bias unit; 22. First opening prediction unit; 23. Negative bias unit; 24. Second opening prediction unit; 25. Correction coefficient calculation unit; 41. Frequency error calculation unit; 42. Error signal generation unit; 43. PID control signal generation unit. DETAILED DESCRIPTION

[0066] The invention discloses a method for controlling the opening of a turbine governor.

[0067] In one embodiment, referring to Figure 1 The turbine governor opening control method specifically includes the following steps:

[0068] S101. Obtain the water head data, frequency signal and initial PID control signal of the turbine.

[0069] Head data typically refers to the height difference between water flow and the mechanical energy it can convert into turbine output—the difference between kinetic and potential energy carried by the water flow. Head is related to parameters such as water velocity, flow rate, and pipe cross-sectional area, and is generally expressed as head per unit area. The frequency signal includes both the real-time frequency data from the turbine governor and the preset frequency reference of the governor. The initial PID control signal comes from the turbine's PID control module, a control system primarily used to regulate and control parameters such as turbine speed, head, and load. The PID control module consists of three components: proportional, integral, and differential control. Proportional control generates a control amplitude by measuring the difference between the system output and the desired value. Integral control can identify past system errors, making the control amplitude more precise. Differential control can adjust the control amplitude based on current system feedback, improving system stability and responsiveness. The turbine's PID control module can be adjusted and optimized based on specific application scenarios and requirements, achieving more efficient and precise control.

[0070] S102. Generate a difference correction coefficient by combining the water head data and the preset power setting.

[0071] Among them, the power setting is the preset setting value of the turbine speed governor. The power setting can be positively or negatively biased to obtain the positive and negative biased operating conditions. Then, the opening value of the turbine speed governor is predicted in combination with the head data, and the difference correction coefficient is finally calculated based on the opening value.

[0072] S103. Correct the initial difference coefficient preset in the turbine using the difference correction coefficient to obtain an adjusted difference coefficient.

[0073] The calculation formula for the adjustment difference coefficient is as follows:

[0074]

[0075] Where: kb is the difference correction coefficient, Bp is the adjustment difference coefficient, and bp0 is the preset initial difference coefficient.

[0076] S104. Generate an adjusted PID control signal by combining the frequency signal, the initial PID control signal and the adjustment difference coefficient.

[0077] Among them, the frequency error is first calculated based on the frequency signal, and then the control error signal is generated by combining the frequency error, the initial PID control signal and the adjustment difference coefficient. Finally, the adjustment PID control signal is generated through the PID control algorithm.

[0078] S105. Generate an opening control signal by combining the water head data, the power setting and the preset opening setting.

[0079] Among them, it is necessary to first determine whether the operation mode of the turbine is grid-connected operation, and then select different methods to generate the opening signal according to the judgment result.

[0080] S106. Superimpose the adjusted PID control signal and the opening control signal to generate a control signal for the turbine.

[0081] The implementation principle of this embodiment is:

[0082] Based on the power given, the required opening control signal is generated through power opening prediction control, thereby improving the power control accuracy of the opening control method. At the same time, the initial difference coefficient is corrected by generating a difference correction coefficient, thereby achieving accurate control of the primary frequency regulation of the turbine speed governor.

[0083] In one embodiment, referring to Figure 2 Step S102, i.e., combining the water head data and the preset power setting to generate the difference correction coefficient, specifically includes the following steps:

[0084] S201. Perform a positive bias on the preset power setting to obtain a positive bias operating condition point.

[0085] Positively biasing the power reference means setting the system's power setpoint or expected value to a certain value higher than the actual output power, thereby enabling the controller to more precisely control the system's output power. Positive biasing can place the control system's operating point in a higher output power range, thereby reducing adjustment errors and improving the control system's stability and response speed. In turbine control systems, positive biasing is typically achieved by changing the opening of the hydraulic governor or adjusting the control valve. This can be adjusted and optimized based on specific turbine parameters and operating conditions to achieve superior regulation.

[0086] For example, assuming that the head data of the turbine is H, the preset power setting is Cp, and the positive bias is expected to be 5%, the power setting after positive bias is Cp1=105%Cp, so the positive bias operating point S1 is [Cp1, H].

[0087] S202. Calculate a first opening prediction value by combining the positive bias operating point and the water head data.

[0088] In one embodiment, the following steps are included before step S202: obtaining historical hydraulic head data, historical power settings, and historical governor opening values ​​of the turbine; establishing an opening prediction model based on a convolutional neural network; and training the opening prediction model using the historical hydraulic head data, historical power settings, and historical governor opening values. After the opening prediction model is trained, the positive bias operating condition point and hydraulic head data are input into the trained opening prediction model to calculate a first opening prediction value.

[0089] S203. Negatively bias the preset power setting to obtain a negative bias operating condition point.

[0090] For example, assuming that the head data of the turbine is H, the preset power setting is Cp, and the negative bias is expected to be 5%, the power setting after negative bias is Cp2=95%Cp, so the negative bias operating point S2 is [Cp2, H].

[0091] S204. Calculate the second opening prediction value by combining the negative bias operating point and the water head data.

[0092] In one embodiment, the following steps are included before step S204: obtaining historical hydraulic head data, historical power settings, and historical governor opening values ​​of the turbine; establishing an opening prediction model based on a convolutional neural network; and training the opening prediction model using the historical hydraulic head data, historical power settings, and historical governor opening values. After the opening prediction model is trained, the negative bias operating condition point and hydraulic head data are input into the trained opening prediction model to calculate a second opening prediction value.

[0093] S205. Calculate the difference correction coefficient by combining the first opening prediction value and the second opening prediction value.

[0094] The calculation formula of the difference correction coefficient is as follows:

[0095]

[0096] Where: kb is the difference correction coefficient, y1 is the first opening prediction value, and y2 is the second opening prediction value.

[0097] In one embodiment, the frequency signal includes frequency data and a preset frequency setting, referring to Figure 3 Step S104, i.e., generating an adjusted PID control signal by combining the frequency signal, the initial PID control signal, and the adjustment difference coefficient, specifically includes the following steps:

[0098] S301. Calculate the frequency error based on the frequency data and the frequency setting.

[0099] The frequency error is calculated as follows:

[0100] Xe=Cf-freq

[0101] Where: Xe is the frequency error, Cf is the frequency reference, and freq is the frequency data.

[0102] S302. Generate a control error signal by combining the frequency error, the initial PID control signal and the adjustment difference coefficient.

[0103] Among them, the initial PID control signal is the initial frequency value output by the turbine PID controller, and the control error signal can be understood as the actual control error difference. Therefore, the calculation formula of the actual control error difference is as follows:

[0104] Ce=Xe-Bp×f0

[0105] Where: Ce is the actual control error difference, Xe is the frequency error, Bp is the adjustment difference coefficient, and f0 is the initial frequency value output by the turbine PID controller.

[0106] S303 . Generate an adjustment PID control signal based on the control error signal and through a PID control algorithm.

[0107] The adjusted PID control signal is the frequency value output by the PID control module after error correction. The formula for generating the adjusted PID control signal through the PID control algorithm is as follows:

[0108]

[0109] Where: f pid To adjust the PID control signal, kp is the proportional control coefficient, ki is the integral control coefficient, and kd is the differential control coefficient.

[0110] In one embodiment, referring to Figure 4 Step S105, i.e., combining the water head data, the power setting, and the preset opening setting to generate the opening control signal, specifically includes the following steps:

[0111] S401. Determine whether the operation mode of the turbine is the grid-connected operation mode or the non-grid-connected operation mode. If the operation mode of the turbine is the non-grid-connected operation mode, execute step S402; if the operation mode of the turbine is the grid-connected operation mode, execute step S403.

[0112] S402. Set the preset opening as the opening control signal.

[0113] S403. Generate an opening control signal by combining the water head data and the power setting.

[0114] In one embodiment, referring to Figure 5 Step S403, i.e., combining the water head data and the power setting to generate the opening control signal, specifically includes the following steps:

[0115] S501. Obtain historical operating data of the turbine.

[0116] Among them, the historical operation data includes historical head data, historical power setting and historical governor opening value.

[0117] S502. Build and train a convolutional neural network model based on historical operation data.

[0118] Among them, in this embodiment, the specific steps of step S502 are as follows: constructing a convolutional neural network model; training the convolutional neural network model through historical head data, historical power settings and historical opening values ​​of the speed regulator.

[0119] S503. Generate an output vector according to the water head data and power setting.

[0120] Assuming that the water head data is H and the power setting is Cp, the output vector Sp generated according to the water head data and the power setting is [Cp, H].

[0121] S504. Output the output vector to the convolutional neural network model to obtain an opening prediction signal, and use the opening prediction signal as the opening control signal.

[0122] The invention also discloses a turbine governor opening control system.

[0123] In one embodiment, referring to Figure 6 The turbine governor opening control system includes a data acquisition module, a correction coefficient generation module, a difference coefficient adjustment module, a PID control module, an opening control switching module, and a control output module. The data acquisition module is connected to the PID control module to obtain the initial PID control signal from the PID control module. The data acquisition module is also used to obtain the turbine's head data and frequency signal. The frequency signal includes frequency data and a preset frequency reference.

[0124] The correction coefficient generation module is connected to the data acquisition module to receive water head data. The correction coefficient generation module combines the received water head data with the preset power setting to generate a difference correction coefficient. The difference coefficient adjustment module is connected to the correction coefficient generation module. The difference coefficient adjustment module receives the difference correction coefficient and uses the difference correction coefficient to correct the initial difference coefficient preset in the turbine, outputting an adjusted difference coefficient. The calculation formula for the adjusted difference coefficient is as follows:

[0125]

[0126] Where: kb is the difference correction coefficient, Bp is the adjustment difference coefficient, and bp0 is the preset initial difference coefficient.

[0127] The PID control module is connected to the data acquisition module and the difference coefficient adjustment module. The PID control module receives the frequency signal from the data acquisition module, obtains the adjustment difference coefficient and the initial PID control signal from the difference coefficient adjustment module, and then combines the frequency signal, the initial PID control signal and the adjustment difference coefficient to generate the adjustment PID control signal.

[0128] The opening control switching module is connected to the data acquisition module to receive water head data and a preset opening setting, thereby generating an opening control signal by combining the water head data, power setting, and the preset opening setting. The control output module is connected to the PID control module and the opening control switching module respectively. The control output module is used to superimpose the adjusted PID control signal with the opening control signal to generate a control signal for the turbine speed governor.

[0129] Reference Figure 6 In this embodiment, the correction coefficient generation module includes:

[0130] The positive bias unit is used to positively bias the preset power setting to obtain a positive bias operating condition point.

[0131] The first opening prediction unit is used to calculate a first opening prediction value by combining the positive bias operating point and the water head data.

[0132] The negative bias unit is used to negatively bias the preset power setting to obtain a negative bias operating condition point.

[0133] The second opening prediction unit is used to calculate the second opening prediction value by combining the negative bias operating point and the water head data.

[0134] The correction coefficient calculation unit is used to calculate the difference correction coefficient by combining the first opening prediction value and the second opening prediction value.

[0135] Among them, assuming that the head data of the turbine is H, the preset power setting is Cp, the positive bias expectation is 5%, and the negative bias expectation is 5%, then the power setting after positive bias is Cp1=105%Cp, and the positive bias operating condition point S1 is [Cp1, H], and the power setting after negative bias is Cp2=95%Cp, and the negative bias operating condition point S2 is [Cp2, H].

[0136] In one embodiment, the correction coefficient generation module also includes a historical data retrieval unit and a prediction model construction unit. The historical data retrieval unit is used to obtain the historical head data, historical power setting and historical opening value of the turbine, and the prediction model construction unit is used to establish an opening prediction model based on a convolutional neural network, and train the opening prediction model through the historical head data, historical power setting and historical opening value of the governor. The first opening prediction unit inputs the positive bias operating condition point and the head data into the opening prediction model to obtain the first opening prediction value y1; the second opening prediction unit inputs the negative bias operating condition point and the head data into the opening prediction model to obtain the second opening prediction value y2. The calculation formula of the difference correction coefficient kb is as follows:

[0137]

[0138] Reference Figure 6 In this embodiment, the PID control module includes:

[0139] The frequency error calculation unit is used to calculate the frequency error according to the frequency data and the frequency setting.

[0140] The error signal generating unit is used to generate a control error signal by combining the frequency error, the initial PID control signal and the adjustment difference coefficient.

[0141] The PID control signal generating unit is used to generate an adjustment PID control signal according to the control error signal and through a PID control algorithm.

[0142] The frequency error is calculated as follows:

[0143] Xe=Cf-freq

[0144] Where: Xe is the frequency error, Cf is the frequency reference, and freq is the frequency data.

[0145] The initial PID control signal is the initial frequency value output by the turbine PID controller. The control error signal can be understood as the actual control error difference. Therefore, the calculation formula for the actual control error difference is as follows:

[0146] Ce=Xe-Bp×f0

[0147] Where: Ce is the actual control error difference, Xe is the frequency error, Bp is the adjustment difference coefficient, and f0 is the initial frequency value output by the turbine PID controller.

[0148] The adjusted PID control signal is the frequency value output by the PID control module after error correction. The formula for generating the adjusted PID control signal through the PID control algorithm is as follows:

[0149]

[0150] Where: f pid To adjust the PID control signal, kp is the proportional control coefficient, ki is the integral control coefficient, and kd is the differential control coefficient.

[0151] In one embodiment, the opening control switching module includes:

[0152] The operation mode determination unit is used to determine whether the operation mode of the turbine is a grid-connected operation mode or a non-grid-connected operation mode.

[0153] The first opening control unit is preset with an opening setting, and is used to output the opening setting as an opening control signal when the operation mode judgment unit determines that the operation mode of the turbine is a non-grid-connected operation mode.

[0154] The historical data acquisition unit is used to acquire historical operating data of the turbine.

[0155] The model building unit is used to build and train a convolutional neural network model based on historical operation data.

[0156] The vector generating unit is used to generate an output vector according to the water head data and the power setting.

[0157] The second opening control unit is used to output the output vector to the convolutional neural network model to obtain an opening prediction signal when the judgment result of the operation mode judgment unit is that the operation mode of the turbine is the grid-connected operation mode, and output the opening prediction signal as the opening control signal.

[0158] The historical operating data includes historical water head data, historical power settings, and historical governor opening values. In one embodiment, the model building unit constructs a convolutional neural network model and trains the convolutional neural network model using the historical water head data, historical power settings, and historical governor opening values. Assuming the water head data is H and the power setting is Cp, the output vector Sp generated by the vector generation unit based on the water head data and power setting is [Cp, H].

[0159] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of protection of the present application is limited to these examples. In line with the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of different aspects of one or more embodiments of the present application as above, which are not provided in detail for the sake of simplicity.

[0160] The one or more embodiments of this application are intended to encompass all such substitutions, modifications, and variations that fall within the broad scope of this application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of this application should be included in the scope of protection of this application.

Claims

1. A method for controlling the opening of a turbine governor, characterized in that: The steps include: Obtain the water head data, frequency signal and initial PID control signal of the turbine; generating a difference correction coefficient by combining the water head data and a preset power setting; Correcting an initial differential coefficient preset in the turbine by using the differential correction coefficient to obtain an adjusted differential coefficient; generating an adjusted PID control signal by combining the frequency signal, the initial PID control signal, and the adjustment difference coefficient; generating an opening control signal by combining the water head data, the power setting and the preset opening setting; Superimposing the adjustment PID control signal and the opening control signal to generate a control signal for the hydraulic turbine; The step of combining the water head data and the preset power setting to generate the difference correction coefficient comprises the following steps: Perform positive bias on the preset power setting to obtain the positive bias operating condition point; Calculating a first opening prediction value by combining the positive bias operating point and the water head data; Negatively bias the preset power setting to obtain the negative bias operating condition point; Calculating a second opening prediction value by combining the negative bias operating condition point and the water head data; A difference correction coefficient is calculated by combining the first opening prediction value and the second opening prediction value.

2. The method for controlling the opening of a turbine governor according to claim 1, wherein: The frequency signal includes frequency data and a preset frequency setting, and the generating of the adjusted PID control signal by combining the frequency signal, the initial PID control signal, and the adjustment difference coefficient includes the following steps: Calculating a frequency error according to the frequency data and the frequency setting; generating a control error signal by combining the frequency error, the initial PID control signal, and the adjustment difference coefficient; An adjustment PID control signal is generated based on the control error signal through a PID control algorithm.

3. The method for controlling the opening of a turbine governor according to claim 1, wherein: The step of generating an opening control signal by combining the water head data, the power setting and the preset opening setting comprises the following steps: Determining whether the operation mode of the turbine is a grid-connected operation mode or a non-grid-connected operation mode; If the operation mode of the turbine is the non-grid-connected operation mode, a preset opening is given as an opening control signal; If the operation mode of the turbine is the grid-connected operation mode, the opening control signal is generated in combination with the water head data and the power setting.

4. The method for controlling the opening of a turbine governor according to claim 3, wherein: The step of combining the water head data and the power setting to generate the opening control signal comprises the following steps: Acquiring historical operating data of the turbine; Building and training a convolutional neural network model based on the historical operating data; Generate an output vector according to the water head data and the power setting; The output vector is output to the convolutional neural network model to obtain an opening prediction signal, and the opening prediction signal is used as the opening control signal.

5. A turbine governor opening control system, characterized in that: The system comprises: A data acquisition module is used to obtain the water head data, frequency signal and initial PID control signal of the turbine; A correction coefficient generating module, configured to generate a difference correction coefficient by combining the water head data and a preset power setting; A difference coefficient adjustment module, configured to correct an initial difference coefficient preset in the turbine using the difference correction coefficient to obtain an adjusted difference coefficient; a PID control module, configured to generate an adjusted PID control signal by combining the frequency signal, the initial PID control signal, and the adjustment difference coefficient; An opening control switching module, configured to generate an opening control signal by combining the water head data, the power setting and a preset opening setting; a control output module, configured to superimpose the adjustment PID control signal and the opening control signal to generate a control signal for the hydraulic turbine; The correction coefficient generation module includes: The positive bias unit is used to positively bias the preset power setting to obtain the positive bias operating condition point; A first opening prediction unit, configured to calculate a first opening prediction value by combining the positive bias operating point and the water head data; A negative bias unit is used to negatively bias the preset power setting to obtain a negative bias operating condition point; A second opening prediction unit, configured to calculate a second opening prediction value by combining the negative bias operating point and the water head data; A correction coefficient calculation unit is used to calculate a difference correction coefficient by combining the first opening prediction value and the second opening prediction value.

6. The turbine governor opening control system according to claim 5, characterized in that: The frequency signal includes frequency data and a preset frequency setting, and the PID control module includes: A frequency error calculation unit, configured to calculate a frequency error based on the frequency data and the given frequency; an error signal generating unit, configured to generate a control error signal by combining the frequency error, the initial PID control signal, and the adjustment difference coefficient; The PID control signal generating unit is used to generate an adjustment PID control signal according to the control error signal and through a PID control algorithm.

7. The turbine governor opening control system according to claim 5, characterized in that: The opening control switching module includes: an operation mode determination unit, configured to determine whether the operation mode of the hydraulic turbine is a grid-connected operation mode or a non-grid-connected operation mode; The first opening control unit is preset with an opening setting, and is used to output the opening setting as an opening control signal when the judgment result of the operation mode judgment unit is that the operation mode of the turbine is the non-grid-connected operation mode.

8. The turbine governor opening control system according to claim 7, characterized in that: The opening control switching module also includes: A historical data acquisition unit, configured to acquire historical operating data of the hydraulic turbine; A model building unit, configured to build and train a convolutional neural network model based on the historical operating data; A vector generating unit, configured to generate an output vector according to the water head data and the power setting; A second opening control unit is used to output the output vector to the convolutional neural network model to obtain an opening prediction signal when the judgment result of the operation mode judgment unit is that the operation mode of the turbine is the grid-connected operation mode, and output the opening prediction signal as the opening control signal.

Images