Intelligent control system of mixed-flow water turbine and control method of intelligent control system

Abstract

The invention discloses a mixed-flow water turbine intelligent control system and a control method thereof, and belongs to the technical field of mixed-flow water turbine intelligent control. The method comprises the steps that a mixed-flow water turbine is monitored in real time, and monitoring data are obtained; analyzing the monitoring data in real time, and determining a power generation efficiency curve; determining an optimal working condition point in real time according to the power generation efficiency curve, and generating a control instruction based on the optimal working condition point; obtaining an adjustment record of the power generation efficiency curve in real time; classifying the adjustment records according to the working condition points to obtain working condition efficiency records of the working condition points; obtaining a standard power generation curve, and generating an efficiency change curve of the working condition points according to the standard power generation curve and the working condition efficiency record; performing early warning analysis according to the efficiency change curve of each working condition point to obtain a corresponding early warning analysis result, and performing corresponding early warning processing according to the early warning analysis result; and the start-stop record of the mixed-flow water turbine is obtained in real time, and start-stop optimization management is conducted on the mixed-flow water turbine according to the start-stop record.

Description

Technical Field

[0001] The present invention belongs to the technical field of Francis turbine intelligent control, and in particular relates to a Francis turbine intelligent control system and a control method thereof. Background Art

[0002] Francis turbines, known for their high efficiency and stability, are widely used in water conservancy and hydropower projects. However, with the continued expansion of hydropower stations and the increasing complexity of operating conditions, the performance requirements for Francis turbines are becoming increasingly stringent. While significant advances have been made in the design and manufacturing of existing Francis turbines, deficiencies remain in intelligent control. Traditional control systems often rely on manual operation and empirical judgment, making it difficult to achieve precise and efficient automated control.

[0003] With the rapid development of technologies such as the Internet of Things, big data, and artificial intelligence, intelligent control systems have been widely used in various fields. However, their application in Francis turbines is still in its infancy. Most existing intelligent control systems only provide simple monitoring and alarm functions, lacking comprehensive, real-time, and precise control of Francis turbine operating conditions. Existing intelligent control systems are often only adaptable to specific operating conditions and lack sufficient adaptability to the complex and changing environment of hydropower stations.

[0004] Based on this, in order to solve the above problems, the present invention provides an intelligent control system for a Francis turbine and a control method thereof. Summary of the Invention

[0005] In order to solve the problems existing in the above solutions, the present invention provides an intelligent control system for a Francis turbine and a control method thereof, so as to solve the intelligent control problem of the existing Francis turbine.

[0006] The purpose of the present invention can be achieved through the following technical solutions: An intelligent control system for a Francis turbine, comprising a monitoring module, an efficiency analysis module, a control module and an abnormality monitoring module; The monitoring module is used to monitor the Francis turbine in real time and obtain monitoring data of the Francis turbine.

[0007] The efficiency analysis module is used to perform real-time analysis on the monitoring data of the Francis turbine to determine a power generation efficiency curve of the Francis turbine, wherein the vertical axis of the power generation efficiency curve is the power generation efficiency and the horizontal axis is the operating point; An optimal operating point is determined in real time according to the power generation efficiency curve, a corresponding control instruction is generated based on the optimal operating point, and the control instruction is sent to a control module.

[0008] Furthermore, the determination of the power generation efficiency curve includes: Obtain standard power generation curves; identify corresponding control parameter items based on the standard power generation curves; obtain material analysis data of Francis turbines in real time, including historical monitoring data and historical power generation efficiency; performing feature recognition on the material analysis data according to the control parameter items to obtain control feature data, wherein the control feature data includes a control combination feature and a power generation efficiency feature; classifying the control feature data according to the control combination feature to obtain efficiency optimization materials corresponding to the control combination feature; The efficiency optimization material is analyzed to obtain the power generation adjustment efficiency corresponding to the control combination characteristics; the control combination characteristics and the power generation adjustment efficiency are integrated into curve adjustment data; and the standard power generation curve is adjusted in real time according to the curve adjustment data to obtain the power generation efficiency curve of the Francis turbine.

[0009] Furthermore, the efficiency optimization materials are analyzed, including: Step SA1: Identify power generation efficiency features corresponding to the efficiency optimization material, classify identical power generation efficiency features into a category, and label them as unit efficiency features; identify the unit value and unit number corresponding to the unit efficiency feature, where the unit number is the number of power generation efficiency features included in the unit efficiency feature, and the unit value is the power generation efficiency corresponding to the unit efficiency feature; Step SA2: Merge any two unit efficiency features with the largest number of units according to a preset representative merging formula to obtain a new unit efficiency feature; the representative merging formula is: ; d=d1+d2; Where: L is the unit value of the new unit efficiency feature; L1 and L2 are the unit values ​​of the two unit efficiency features to be merged; d1 and d2 are the number of units of the two unit efficiency features to be merged; d is the number of units of the new unit efficiency feature; Step SA3: looping step SA2 until there is only one unit efficiency feature, and determining the power generation adjustment efficiency according to the unit value of the unit efficiency feature.

[0010] Furthermore, the optimal operating point is determined in real time based on the power generation efficiency curve, including: Identifying operating points corresponding to various power generation efficiencies according to the power generation efficiency curve, marking them as calibration operating points, and performing calibration analysis on the calibration operating points to obtain a number of preliminary selected operating points; Identify the highest power generation efficiency among the pre-selected operating points, and mark the pre-selected operating point corresponding to the highest power generation efficiency as the operating point to be selected; Screen the operating points to be selected and obtain the best operating point.

[0011] Furthermore, the operating point is calibrated and analyzed, including: Identify the reference operating point based on the monitoring data; establish a control calibration model, analyze the reference operating point and the calibration operating point through the control calibration model, and obtain the calibration result of the calibration operating point, the calibration result includes calibration pass and calibration fail; mark the calibration operating point with the calibration result of calibration pass as the preliminary operating point.

[0012] Furthermore, the expression of the control calibration model is: ; Where: (GK, HK) is the input data, GK is the reference operating point, HK is the calibration operating point; GK→HK means that the reference operating point can be adjusted to the corresponding calibration operating point through the control module; the output data is the control calibration value KP(GK, HK), which is 1 or 0; The reference operating point and the calibration operating point are analyzed through the control calibration model to obtain the corresponding control calibration value; When the control calibration value is 1, the calibration result is calibration passed; When the control calibration value is 0, the calibration result is calibration failure.

[0013] Furthermore, before determining the optimal operating point based on the power generation efficiency curve, the environmental adjustment item is first determined, and real-time monitoring is performed based on the environmental adjustment item to obtain environmental adjustment data; environmental adjustment interval data is determined based on the environmental adjustment data; and the optimal operating point is determined based on the environmental adjustment interval data and the power generation efficiency curve.

[0014] The control module is used to perform control adjustments according to received control instructions.

[0015] The abnormality monitoring module is used to perform abnormality early warning analysis based on the power generation efficiency curve and obtain the adjustment record of the power generation efficiency curve in real time; Classifying the adjustment records according to the operating point to obtain the operating efficiency records of the operating point; obtaining the standard power generation curve, and generating the efficiency change curve of the operating point according to the standard power generation curve and the operating efficiency records, wherein the horizontal axis of the efficiency change curve is time and the vertical axis is the efficiency change value; Perform early warning analysis based on the efficiency change curve of each operating point, obtain corresponding early warning analysis results, and perform corresponding early warning processing based on the early warning analysis results.

[0016] Furthermore, early warning analysis is performed based on the efficiency change curves of each operating point, including: Fit the efficiency change curve to obtain the efficiency change function, which is marked as HY(t), where t is time; Establish an efficiency early warning model. The expression of the efficiency early warning model is: ; Where: [HY(t)] is the input data, X1 and X2 are thresholds; kt represents the slope of the efficiency change function corresponding to time; the output data is the efficiency warning value QS[HY(t)], which is 1, 2, or 0; The efficiency change function is analyzed through the efficiency warning model to obtain the corresponding efficiency warning value; the warning analysis result is determined according to the efficiency warning value.

[0017] Furthermore, a start-stop management module is included, which is used to optimize the start-stop management of the Francis turbine, obtain the start-stop records of the Francis turbine in real time, and the stop records include the start-stop time, start-stop reasons, and the number of starts and stops; identify the start-stop reasons in the start-stop records, and classify the start-stop reasons into management and adjustment categories according to whether the number of starts and stops can be reduced by controlling and managing the Francis turbine; A corresponding optimization management method is set for the start and stop reasons corresponding to the management class, and the Francis turbine is managed according to the optimization management method.

[0018] A method for controlling a Francis turbine, the method comprising: Conduct real-time monitoring of Francis turbines and obtain monitoring data of Francis turbines; Conduct real-time analysis of monitoring data to determine the power generation efficiency curve of the Francis turbine; Determine the optimal operating point in real time based on the power generation efficiency curve, generate control instructions based on the optimal operating point, and adjust the Francis turbine control according to the control instructions; Acquire adjustment records of the power generation efficiency curve in real time; classify the adjustment records according to operating points to obtain operating efficiency records of the operating points; obtain a standard power generation curve, and generate an efficiency change curve of the operating point based on the standard power generation curve and the operating efficiency records; Conduct early warning analysis based on the efficiency change curve of each operating point, obtain corresponding early warning analysis results, and perform corresponding early warning processing based on the early warning analysis results; The start and stop records of the Francis turbine are obtained in real time, and the start and stop optimization management of the Francis turbine is performed based on the start and stop records.

[0019] Compared with the prior art, the present invention has the following beneficial effects: Through the mutual cooperation between the monitoring module, efficiency analysis module, control module and abnormality monitoring module, comprehensive monitoring and precise control of the operating status of the Francis turbine are achieved; the control strategy can be automatically adjusted according to the actual operating conditions and conditions of the hydropower station to ensure that the intelligent control system can maintain optimal performance in various environments; this high adaptability not only improves the reliability and stability of the system, but also reduces dependence on human intervention; by setting up the start-stop management module, the number of starts and stops of the Francis turbine can be intelligently reduced, the stability of the Francis turbine can be improved, and at the same time, the service life can be extended and the economic benefits of the hydropower station can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 This is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0022] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0023] like Figure 1 As shown, an intelligent control system for a Francis turbine includes a monitoring module, an efficiency analysis module, a control module, an abnormality monitoring module, and a start-stop management module; The monitoring module is used to monitor the Francis turbine in real time and obtain corresponding monitoring data, such as flow rate, pressure, speed, temperature and other related data; and use the configured relevant sensors to perform corresponding monitoring.

[0024] The efficiency analysis module is used to perform real-time analysis on the monitoring data of the Francis turbine to determine the power generation efficiency curve of the Francis turbine. The vertical axis of the power generation efficiency curve is the power generation efficiency, and the horizontal axis is the various control parameters that affect the power generation efficiency, such as guide vane opening, speed, flow, head, etc. Subsequently, control adjustment is performed based on the adjustable control parameters. For example, it is determined based on actual conditions whether parameters such as flow and head can be adjusted. If not, the power generation efficiency is adjusted mainly by adjusting the guide vane opening and speed of the Francis turbine. The corresponding control parameter combination is marked as an operating point, that is, the horizontal axis is composed of the corresponding operating points; the optimal operating point is determined in real time based on the power generation efficiency curve, and the corresponding control instructions are generated based on the optimal operating point, and the control instructions are sent to the control module.

[0025] Among them, the above-mentioned Francis turbine refers to a Francis turbine that needs to be intelligently controlled after installation and application.

[0026] In one embodiment, the power generation efficiency curve can be determined based on existing methods, such as applying the power generation efficiency curve determined by the previous real machine experiment, and then synchronously correcting the above power generation efficiency curve during equipment maintenance; or an intelligent model can be established based on intelligent algorithms such as deep learning algorithms, and the monitoring data can be analyzed by the successfully trained intelligent model to obtain the power generation efficiency curve.

[0027] In one embodiment, determining the power generation efficiency curve includes: Mark the power generation efficiency curve obtained by the previous real machine experiment or other simulation experiment method as the standard power generation curve; According to the standard power generation curve, the corresponding control parameter items are identified and the material analysis data of the Francis turbine is obtained in real time. The material analysis data refers to the recent historical monitoring data of the Francis turbine and the corresponding historical power generation efficiency. For example, within a day or a week, the material amount of the corresponding power generation efficiency is generally collected. For example, for power generation efficiency A, one hour of historical monitoring data at the power generation efficiency is required, and then the historical monitoring data is collected according to the corresponding time. The material analysis data is collected according to the actual situation.

[0028] Feature identification is performed on the material analysis data according to the control parameter items to obtain control feature data. The control feature data includes control combination features and power generation efficiency features. The control combination features are the combination of control parameters corresponding to each control parameter item, and the power generation efficiency features refer to the historical power generation efficiency corresponding to the control combination features. The control feature data are classified according to the control combination features to obtain efficiency optimization materials corresponding to the control combination features.

[0029] The efficiency optimization material is analyzed to obtain the power generation adjustment efficiency corresponding to the corresponding control combination characteristics; the control combination characteristics and the power generation adjustment efficiency are integrated into curve adjustment data, and the standard power generation curve is adjusted in real time according to the curve adjustment data to obtain the power generation efficiency curve of the Francis turbine; that is, the curve adjustment data is used as the new curve coordinates for curve adjustment.

[0030] In one embodiment, the efficiency optimization material is analyzed based on existing mathematical statistical methods to determine the representative power generation efficiency corresponding to the control combination characteristics according to the efficiency optimization material, and mark it as the power generation adjustment efficiency; for example, the power generation adjustment efficiency is determined by mathematical statistical methods such as mode and mean.

[0031] In one embodiment, analyzing the efficiency optimization material includes: Step SA1: Identify the power generation efficiency characteristics corresponding to the efficiency optimization material, group the same power generation efficiency characteristics into one category, label them as unit efficiency characteristics, and identify the unit value and unit number corresponding to the unit efficiency characteristics. The unit number is the number of power generation efficiency characteristics corresponding to the unit efficiency characteristic, and the unit value is the historical power generation efficiency corresponding to the unit efficiency characteristic or the new unit value calculated by subsequent merging. Step SA2: Merge any two unit efficiency characteristics with the largest number of units according to the preset representative merging formula to obtain a new unit efficiency characteristic; if the unit with the largest number of units has two unit efficiency characteristics, then merge these two unit efficiency characteristics; if the unit with the largest number of units has 5, then select any two from the 5 to merge, and so on.

[0032] Identify the unit value and number of units corresponding to the new unit efficiency feature; the representative merging formula is: ; Where: L is the unit value corresponding to the combination of the two unit efficiency characteristics; L1 and L2 are the unit values ​​of the two unit efficiency characteristics; d1 and d2 are the number of units of the two unit efficiency characteristics; The number of cells for the new cell efficiency feature is: d=d1+d2; d is the number of units of the new unit efficiency feature; Merging two unit efficiency features into a new unit efficiency feature is equivalent to set union; Step SA3: loop step SA2 until there is only one unit efficiency feature, and mark the unit value of the unit efficiency feature as the power generation adjustment efficiency.

[0033] In one embodiment, an optimal operating point is determined in real time based on a power generation efficiency curve. The operating point refers to the operating state of a Francis turbine under specific flow, head, speed, and other conditions. The optimal operating point is the state in which the Francis turbine achieves the highest power generation efficiency and most stable operation under these conditions. The optimal operating point is determined based on the above definition using existing methods.

[0034] In one embodiment, determining the optimal operating point in real time based on the power generation efficiency curve includes: Identify the operating points corresponding to each power generation efficiency based on the power generation efficiency curve, mark them as calibration operating points, perform calibration analysis on the calibration operating points, and mark the achievable calibration operating points as preliminary operating points; identify the highest power generation efficiency among the preliminary operating points, and mark the preliminary operating point corresponding to the highest power generation efficiency as the candidate operating point; Screen the operating points to be selected and obtain the best operating point.

[0035] In one embodiment, the calibration analysis of the operating point can be performed based on an existing calibration method, such as performing calibration according to actual controllable conditions to determine whether the current operating point can be adjusted to a corresponding operating point.

[0036] In one embodiment, performing calibration analysis on the operating point includes: Identify the current operating point based on the monitoring data and mark it as the reference operating point; establish a control calibration model, which is used to be established based on the actual control situation of the Francis turbine and trained using the set training data, such as the simulated reference operating point, calibration operating point, and calibration results; analyze the reference operating point and calibration operating point through the control calibration model to obtain corresponding calibration results, which include calibration pass and calibration fail, and calibration pass indicates that control adjustment can be performed; The preliminary operating point is determined based on the obtained calibration results.

[0037] In one embodiment, the control calibration model can be established based on a neural network such as a CNN network or a DNN network.

[0038] In one embodiment, the expression of the control calibration model is: ; Where: (GK, HK) is the input data, GK is the reference operating point, HK is the calibration operating point; GK→HK means that the reference operating point can be adjusted to the corresponding calibration operating point through the control module; the output data is the control calibration value KP(GK, HK), which is 1 or 0; The reference operating point and the calibration operating point are analyzed through the control calibration model to obtain the corresponding control calibration value; When the control calibration value is 1, the calibration result is calibration passed; When the control calibration value is 0, the calibration result is calibration failure.

[0039] In one embodiment, the operating points to be selected are screened, mainly based on screening factors such as operating stability, maintenance cost, and control adjustment time, to determine the optimal operating point. For example, weight coefficients of various screening factors are preset, and features are extracted based on the screening factors. Then, priorities are determined based on the weight coefficients of the screening factors, and the optimal operating point is determined based on the priorities.

[0040] In one embodiment, the operating points to be selected may be screened and analyzed based on other existing screening methods to determine the optimal operating point.

[0041] In one embodiment, when the working environment of the Francis turbine has conditions for adjusting the water flow rate, water head, etc., the environmental adjustment items are determined according to the actual situation and set according to the adjustable water head, etc.; intelligent analysis technology can also be used to recommend to the user whether the corresponding environmental adjustment items can be applied in the actual environment and the best implementation method, thereby improving the working environment of the Francis turbine and achieving the adjustment of some or all environmental influencing factors; specifically: Determine the environmental adjustment item, perform real-time monitoring based on the environmental adjustment item, and obtain environmental adjustment data; determine the environmental adjustment interval data based on the environmental adjustment data, that is, use the control range of the environmental adjustment item to determine the adjustment range of the environmental adjustment data, and then determine the environmental adjustment interval data; determine the optimal operating point based on the environmental adjustment interval data and the power generation efficiency curve, that is, the difference from the above embodiment is that the environmental adjustment item can be changed to achieve higher power generation efficiency.

[0042] The control module is used to perform control adjustments according to received control instructions.

[0043] The abnormality monitoring module is used to perform abnormality early warning analysis based on the power generation efficiency curve, obtain the adjustment record of the power generation efficiency curve in real time, that is, the change adjustment record of the power generation efficiency corresponding to each operating point, classify the adjustment record according to the operating point, and obtain the operating efficiency record of the corresponding operating point; generate the efficiency change curve of the corresponding operating point based on the operating efficiency record, that is, generate the efficiency curve based on the change value of the power generation efficiency, and compare and calculate it based on the standard power generation curve; the horizontal axis of the efficiency change curve is time, and the vertical axis is the efficiency change value; Conduct early warning analysis based on the efficiency change curve of each operating point to obtain corresponding early warning analysis results, and perform corresponding early warning processing based on the early warning analysis results, such as issuing early warnings to management personnel, emergency processing, etc.

[0044] In one embodiment, a warning analysis is performed based on the efficiency change curves of each operating point, mainly based on the changes in the slope of each efficiency change curve at the same time, the size of the efficiency change value, etc. The efficiency change curve can be used for warning analysis in combination with existing methods; for example, an intelligent warning model is established based on a deep learning algorithm, and a corresponding training set is established manually for training. The training set includes input data and output data, the input data is the efficiency change curve of each operating point, and the output data is the corresponding warning analysis result; and analysis is performed using the intelligent warning model after successful training.

[0045] In one embodiment, early warning analysis is performed based on the efficiency change curves of each operating point, including: Fit the efficiency change curve to obtain the efficiency change function, which is marked as HY(t), where t is time; Establish an efficiency early warning model. The expression of the efficiency early warning model is: ; Where: [HY(t)] is the input data, X1 and X2 are thresholds, which are set according to the user's requirements for slope change and cumulative difference, respectively. Generally, a threshold X2 and X2 that meet the user's requirements can be selected through multiple simulations; kt represents the slope of the efficiency change function over time; the output data is the efficiency warning value QS[HY(t)], which is 1, 2, or 0; The efficiency change function is analyzed through the efficiency warning model to obtain the corresponding efficiency warning value; the warning analysis result is determined according to the efficiency warning value. The efficiency warning value corresponds to different results, and the corresponding treatment measures can be preset for processing.

[0046] The start-stop management module is used to optimize the start-stop management of the Francis turbine and obtain the start-stop records of the Francis turbine in real time. The start-stop records include relevant data such as start-stop time, start-stop reasons, and number of times. For example, there are many reasons such as changes in grid demand, equipment failure, protection mechanism, and operating environment. For example, with the introduction of intermittent energy such as wind energy and solar energy, the power supply in the grid has become more unstable. When the grid load decreases, the turbine may need to reduce its output or even shut down; and when the load increases, it is necessary to start or increase the output of the turbine.

[0047] Identify the start and stop reasons in the start and stop records, and divide the start and stop reasons into two categories according to whether the start and stop times can be reduced by controlling and managing the Francis turbine, marked as management category and adjustment category; the management category refers to reducing the start and stop times by controlling the operation of the Francis turbine, such as power grid demand, equipment temperature and other reasons. The start and stop times can be avoided or delayed by controlling and adjusting the Francis turbine in advance, thereby reducing the start and stop times; the adjustment category refers to the reasons that cannot be reduced by managing the Francis turbine, and additional adjustments to equipment components are required before the start and stop times can be reduced through management and control, such as interruption of flow, deterioration of water quality, excessively high ambient temperature, equipment failure and other reasons; that is, according to the adjustments of the adjustment category, the start and stop reasons within the management category will be increased, so the management category and adjustment category formed according to the start and stop records are in dynamic update; A corresponding optimization management method is set for the start and stop reasons corresponding to the management class, and the Francis turbine is managed according to the optimization management method.

[0048] In one embodiment, the start and stop reasons are divided into management and adjustment categories, and can be classified according to the above definition using existing methods, such as using intelligent algorithms such as deep learning algorithms to perform intelligent classification using intelligent models, and then manually confirming to ensure classification accuracy; and because the number of start and stop reasons is small and well known to professionals, they can also be directly classified manually.

[0049] In one embodiment, for the start and stop reasons corresponding to the adjustment category, the available improvement plans can be intelligently analyzed and evaluated and screened based on the user's improvement requirements, such as improvement cost requirements, and improvements can be made according to the screened improvement plans.

[0050] In one embodiment, a corresponding optimization management method is set for the start and stop reasons corresponding to the management class, and an experimental simulation model that can simulate the operation of a mixed flow turbine can be established based on technologies such as digital twins; various ways to reduce the number of start and stop times corresponding to the corresponding start and stop reasons are analyzed. For example, for the main part of the grid demand reasons, the operation of the mixed flow turbine can be pre-adjusted by predicting the grid demand, and it can also be optimized and adjusted in combination with the energy storage system. There are multiple feasible ways to overcome a specific start and stop reason; the historical operating condition data of the mixed flow turbine is obtained, mainly including relevant data on the historical start and stop of the start and stop reason; the determined method is simulated through the experimental simulation model and the historical operation adjustment data to determine the simulated start and stop times, and the method with the highest priority that can be applied is determined by performing cyclic simulation and comparing the start and stop times, and marked as the optimization management method.

[0051] In one embodiment, a corresponding optimized management method is set for the start and stop reasons corresponding to the management class. With the development of technology, multiple methods can be applied to implement the optimized management method for determining the corresponding start and stop reasons.

[0052] By setting up a start-stop management module, the number of starts and stops of the Francis turbine can be intelligently reduced, the stability of the Francis turbine can be improved, and the service life can be extended while improving the economic benefits of the hydropower station.

[0053] A method for controlling a Francis turbine, the method comprising: Conduct real-time monitoring of Francis turbines and obtain monitoring data of Francis turbines; Conduct real-time analysis of monitoring data to determine the power generation efficiency curve of the Francis turbine; Determine the optimal operating point in real time based on the power generation efficiency curve, generate control instructions based on the optimal operating point, and adjust the Francis turbine control according to the control instructions; Acquire adjustment records of the power generation efficiency curve in real time; classify the adjustment records according to operating points to obtain operating efficiency records of the operating points; obtain a standard power generation curve, and generate an efficiency change curve of the operating point based on the standard power generation curve and the operating efficiency records; Conduct early warning analysis based on the efficiency change curve of each operating point, obtain corresponding early warning analysis results, and perform corresponding early warning processing based on the early warning analysis results; The start and stop records of the Francis turbine are obtained in real time, and the start and stop optimization management of the Francis turbine is performed based on the start and stop records.

[0054] The above formulas are all calculated by removing dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technicians in this field according to actual conditions or obtained by simulating a large amount of data.

[0055] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent control system for a Francis turbine, characterized in that: Includes monitoring module, efficiency analysis module, control module and abnormality monitoring module; The monitoring module is used to monitor the Francis turbine in real time and obtain monitoring data of the Francis turbine; The efficiency analysis module is used to perform real-time analysis on the monitoring data of the Francis turbine to determine a power generation efficiency curve of the Francis turbine, wherein the vertical axis of the power generation efficiency curve is the power generation efficiency and the horizontal axis is the operating point; determining an optimal operating point in real time according to the power generation efficiency curve, generating corresponding control instructions based on the optimal operating point, and sending the control instructions to a control module; The control module is used to perform control adjustments according to the received control instructions; The abnormality monitoring module is used to perform abnormality early warning analysis based on the power generation efficiency curve and obtain the adjustment record of the power generation efficiency curve in real time; Classifying the adjustment records according to the operating point to obtain the operating efficiency records of the operating point; obtaining the standard power generation curve, and generating the efficiency change curve of the operating point according to the standard power generation curve and the operating efficiency records, wherein the horizontal axis of the efficiency change curve is time and the vertical axis is the efficiency change value; Perform early warning analysis based on the efficiency change curve of each operating point, obtain corresponding early warning analysis results, and perform corresponding early warning processing based on the early warning analysis results.

2. The intelligent control system for a Francis turbine according to claim 1, characterized in that: Determination of power generation efficiency curve, including: Obtain standard power generation curves; identify corresponding control parameter items based on the standard power generation curves; obtain material analysis data of Francis turbines in real time, including historical monitoring data and historical power generation efficiency; performing feature recognition on the material analysis data according to the control parameter items to obtain control feature data, wherein the control feature data includes a control combination feature and a power generation efficiency feature; classifying the control feature data according to the control combination feature to obtain efficiency optimization materials corresponding to the control combination feature; The efficiency optimization material is analyzed to obtain the power generation adjustment efficiency corresponding to the control combination characteristics; the control combination characteristics and the power generation adjustment efficiency are integrated into curve adjustment data; and the standard power generation curve is adjusted in real time according to the curve adjustment data to obtain the power generation efficiency curve of the Francis turbine.

3. The intelligent control system for a Francis turbine according to claim 2, characterized in that: Analyze efficiency optimization materials, including: Step SA1: Identify power generation efficiency features corresponding to the efficiency optimization material, classify identical power generation efficiency features into a category, and label them as unit efficiency features; identify the unit value and unit number corresponding to the unit efficiency feature, where the unit number is the number of power generation efficiency features included in the unit efficiency feature, and the unit value is the power generation efficiency corresponding to the unit efficiency feature; Step SA2: Merge any two unit efficiency features with the largest number of units according to a preset representative merging formula to obtain a new unit efficiency feature; the representative merging formula is: ； d=d1+d2; Where: L is the unit value of the new unit efficiency feature; L1 and L2 are the unit values ​​of the two unit efficiency features to be merged; d1 and d2 are the number of units of the two unit efficiency features to be merged; d is the number of units of the new unit efficiency feature; Step SA3: looping step SA2 until there is only one unit efficiency feature, and determining the power generation adjustment efficiency according to the unit value of the unit efficiency feature.

4. The intelligent control system for a Francis turbine according to claim 3, characterized in that: Determine the optimal operating point in real time based on the power generation efficiency curve, including: Identifying operating points corresponding to various power generation efficiencies according to the power generation efficiency curve, marking them as calibration operating points, and performing calibration analysis on the calibration operating points to obtain a number of preliminary selected operating points; Identify the highest power generation efficiency among the pre-selected operating points, and mark the pre-selected operating point corresponding to the highest power generation efficiency as the operating point to be selected; Screen the operating points to be selected and obtain the best operating point.

5. The intelligent control system for a Francis turbine according to claim 4, characterized in that: Perform calibration analysis on the operating points, including: Identify the reference operating point based on the monitoring data; establish a control calibration model, analyze the reference operating point and the calibration operating point through the control calibration model, and obtain the calibration result of the calibration operating point, the calibration result includes calibration pass and calibration fail; mark the calibration operating point with the calibration result of calibration pass as the preliminary operating point.

6. The intelligent control system for a Francis turbine according to claim 5, characterized in that: The expression for the control calibration model is: ； Where: (GK, HK) is the input data, GK is the reference operating point, HK is the calibration operating point; GK→HK means that the reference operating point can be adjusted to the corresponding calibration operating point through the control module; the output data is the control calibration value KP(GK, HK), which is 1 or 0; The reference operating point and the calibration operating point are analyzed through the control calibration model to obtain the corresponding control calibration value; When the control calibration value is 1, the calibration result is calibration passed; When the control calibration value is 0, the calibration result is calibration failure.

7. The intelligent control system for a Francis turbine according to claim 2, characterized in that: Before determining the optimal operating point based on the power generation efficiency curve, first determine the environmental adjustment item, perform real-time monitoring based on the environmental adjustment item, and obtain environmental adjustment data; determine environmental adjustment interval data based on the environmental adjustment data; and determine the optimal operating point based on the environmental adjustment interval data and the power generation efficiency curve.

8. The intelligent control system for a Francis turbine according to claim 1, characterized in that: Perform early warning analysis based on the efficiency change curve of each operating point, including: Fit the efficiency change curve to obtain the efficiency change function, which is marked as HY(t), where t is time; Establish an efficiency early warning model. The expression of the efficiency early warning model is: ； Where: [HY(t)] is the input data, X1 and X2 are thresholds; kt represents the slope of the efficiency change function corresponding to time; the output data is the efficiency warning value QS[HY(t)], which is 1, 2, or 0; The efficiency change function is analyzed through the efficiency warning model to obtain the corresponding efficiency warning value; the warning analysis result is determined according to the efficiency warning value.

9. The intelligent control system for a Francis turbine according to claim 1, characterized in that: The system also includes a start-stop management module, which is used to optimize the start-stop management of the Francis turbine and obtain the start-stop records of the Francis turbine in real time. The stop records include the start-stop time, start-stop reasons, and the number of starts and stops; identify the start-stop reasons in the start-stop records, and classify the start-stop reasons into management and adjustment categories based on whether the number of starts and stops can be reduced through control and management of the Francis turbine; A corresponding optimization management method is set for the start and stop reasons corresponding to the management class, and the Francis turbine is managed according to the optimization management method.

10. A method for controlling a Francis turbine, characterized in that: The method applied to an intelligent control system for a Francis turbine according to any one of claims 1 to 9 comprises: Conduct real-time monitoring of Francis turbines and obtain monitoring data of Francis turbines; Conduct real-time analysis of monitoring data to determine the power generation efficiency curve of the Francis turbine; Determine the optimal operating point in real time based on the power generation efficiency curve, generate control instructions based on the optimal operating point, and adjust the Francis turbine control according to the control instructions; Acquire adjustment records of the power generation efficiency curve in real time; classify the adjustment records according to operating points to obtain operating efficiency records of the operating points; obtain a standard power generation curve, and generate an efficiency change curve of the operating point based on the standard power generation curve and the operating efficiency records; Conduct early warning analysis based on the efficiency change curve of each operating point, obtain corresponding early warning analysis results, and perform corresponding early warning processing based on the early warning analysis results; The start and stop records of the Francis turbine are obtained in real time, and the start and stop optimization management of the Francis turbine is performed based on the start and stop records.

Images