A gate state monitoring method and system for a hydropower station

Abstract

The application provides a gate state monitoring method and system for a hydropower station. The method comprises the following steps: acquiring gate operation monitoring data of the hydropower station at a current time and each reference state threshold of the gate, and preprocessing the gate operation monitoring data to obtain preprocessed gate operation monitoring data; determining a gate reference state deviation of the hydropower station according to the preprocessed gate operation monitoring data, the each reference state threshold and a pre-established state reference model; inputting the gate reference state deviation of the hydropower station into a pre-established fault prediction model to obtain a gate fault diagnosis result of the hydropower station at the current time; and the gate fault diagnosis result comprises a fault mode of the gate and a risk value. The technical scheme provided by the application can accurately monitor the gate state, greatly improve the safety, reliability and operation efficiency of the gate of the hydropower station, reduce the maintenance cost and prolong the service life of the equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of device monitoring, in particular to a gate state monitoring method and system for a hydropower station. BACKGROUND

[0002] The gate in a hydropower station is an important water flow control facility, which plays a key role in regulating water level, protecting the dam, and controlling the safe operation of power generation equipment. With the continuous expansion of the scale of the hydropower station and the extension of the operation cycle, the health management of the gate faces greater challenges. The traditional gate monitoring method mainly relies on manual inspection and regular inspection, which can ensure the safe operation of the gate to a certain extent, but due to the low frequency of inspection and the limitations of the inspection method, it is often difficult to discover potential faults in the operation of the gate in a timely manner. In addition, the environment of the hydropower station is complex, and the gate may have problems such as jamming, wear and tear, corrosion, etc. under the influence of long-time high-pressure water flow and environment, and the traditional monitoring method cannot efficiently and accurately identify these potential problems, and when a fault occurs, it is often in a serious stage.

[0003] In order to improve the operation efficiency and safety of the gate of the hydropower station, in recent years, intelligent monitoring technology has been gradually applied to the health management of the equipment of the hydropower station. The existing intelligent monitoring system mainly evaluates the operation state of the gate through real-time data acquisition, sensor analysis and data processing, but most of the systems still have great deficiencies in the accuracy of fault prediction and diagnosis, especially when dealing with complex operating environments and fault modes, there are often false positives or false negatives. At the same time, most of the existing technologies focus on a single monitoring index, lack a comprehensive method of fusing multiple monitoring data, and do not optimize the model in real time to improve the accuracy and real-time performance of fault diagnosis. Therefore, it is urgent to propose a scheme that can accurately monitor the state of the gate. SUMMARY

[0004] The present application provides a gate state monitoring method and system for a hydropower station to at least solve the technical problem of low gate monitoring accuracy.

[0005] The first aspect embodiment of the present application provides a gate state monitoring method for a hydropower station, which comprises:

[0006] Obtaining gate operation monitoring data of the hydropower station at the current time and each reference state threshold of the gate, and preprocessing the gate operation monitoring data to obtain preprocessed gate operation monitoring data;

[0007] Determining a gate reference state deviation of the hydropower station according to the preprocessed gate operation monitoring data, the each reference state threshold and a pre-established state reference model;

[0008] The gate reference state deviation amount of the hydropower station is input into a pre-established fault prediction model to obtain a gate fault diagnosis result of the hydropower station at the current time;

[0009] The gate fault diagnosis result includes a fault mode and a risk value of the gate.

[0010] Preferably, the pre-processing of the gate operation monitoring data to obtain pre-processed gate operation monitoring data includes:

[0011] The gate operation monitoring data is sequentially subjected to cleaning, removal of abnormal values, completion of missing data and filtering processing to obtain the pre-processed gate operation monitoring data.

[0012] The gate operation monitoring data includes external driving force, friction force, water flow force, force generated when the gate is opened, force generated when the gate is closed, vibration frequency, water flow and water level.

[0013] Further, the pre-established state reference model includes:

[0014] A gate position state model, a mechanical state model, a vibration and wear model, and a water flow and environment model.

[0015] The gate reference state deviation amount of the hydropower station includes:

[0016] A gate position deviation amount, a force deviation amount, a vibration speed deviation amount and a water flow impact force deviation amount.

[0017] Further, the establishment process of the fault prediction model includes:

[0018] Obtaining gate reference state values at each time in a historical period and corresponding fault modes at each time;

[0019] Using machine learning technology, taking the gate reference state values at each time in the historical period as input and taking the corresponding fault modes at each time as output, an initial fault prediction model is optimized and trained to obtain a trained fault prediction model.

[0020] The fault mode includes sticking, water leakage and wear.

[0021] Further, the risk value is calculated according to the following formula:

[0022]

[0023] In the formula, f(t) is the risk value at time t, α i is the weight of the i-th gate reference state value, x i (t) is the i-th gate reference state value at time t, and n is the total number of gate reference state values.

[0024] Further, the method further comprises:

[0025] determining whether the risk value is greater than a preset risk threshold, and generating an early warning signal for failure early warning when the risk value is greater than the preset risk threshold.

[0026] Further, the method further comprises:

[0027] performing adaptive optimization training on the pre-established failure prediction model based on the gate failure diagnosis result.

[0028] Further, the method further comprises:

[0029] collecting gate operation monitoring data by using sensors arranged at preset positions;

[0030] determining a failure occurrence position based on the failure mode of the gate in the gate failure diagnosis result and the gate operation monitoring data collected by the sensors arranged at the preset positions.

[0031] The second aspect embodiment of the present application provides a gate state monitoring system for a hydropower station, comprising:

[0032] an acquisition module, configured to acquire gate operation monitoring data of the hydropower station at a current time and each reference state threshold of the gate, and pre-process the gate operation monitoring data to obtain pre-processed gate operation monitoring data;

[0033] a determination module, configured to determine a gate reference state deviation amount of the hydropower station according to the pre-processed gate operation monitoring data, the each reference state threshold and a pre-established state reference model;

[0034] a diagnosis module, configured to input the gate reference state deviation amount of the hydropower station into a pre-established failure prediction model to obtain a gate failure diagnosis result of the hydropower station at the current time;

[0035] The gate failure diagnosis result comprises a failure mode of the gate and a risk value.

[0036] The third aspect embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the method according to the first aspect embodiment.

[0037] The technical scheme provided by the embodiments of the present application at least brings the following beneficial effects:

[0038] The application provides a gate state monitoring method and system for a hydropower station. The method comprises the following steps: acquiring gate operation monitoring data of the hydropower station at a current time and each reference state threshold of the gate, and preprocessing the gate operation monitoring data to obtain preprocessed gate operation monitoring data; determining a gate reference state deviation of the hydropower station according to the preprocessed gate operation monitoring data, the each reference state threshold and a pre-established state reference model; inputting the gate reference state deviation of the hydropower station into a pre-established fault prediction model to obtain a gate fault diagnosis result of the hydropower station at the current time; and the gate fault diagnosis result comprises a fault mode and a risk value of the gate. The technical scheme provided by the application can greatly improve the safety, reliability and operation efficiency of the gate of the hydropower station, reduce the maintenance cost and prolong the service life of the equipment.

[0039] The additional aspects and advantages of the application will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS

[0040] The above and / or additional aspects and advantages of the application will become apparent and be readily appreciated from the following description, including the accompanying drawings, wherein:

[0041] Figure 1 A flow chart of a gate state monitoring method for a hydropower station according to an embodiment of the application is provided;

[0042] Figure 2 A first structural diagram of a gate state monitoring system for a hydropower station according to an embodiment of the application is provided;

[0043] Figure 3 A second structural diagram of a gate state monitoring system for a hydropower station according to an embodiment of the application is provided. DETAILED DESCRIPTION

[0044] The embodiments of the application are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference signs represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below are exemplary and are intended to explain the application, and cannot be understood as a limitation of the application.

[0045] The application provides a gate state monitoring method and system for a hydropower station.

[0046] A gate state monitoring method and system for a hydropower station are described below with reference to the accompanying drawings.

[0047] Embodiment one

[0048] Figure 1 A flowchart of a gate state monitoring method for a hydropower station according to an embodiment of the application is shown in Figure 1 The method comprises the following steps.

[0049] Step 1: Obtain gate operation monitoring data of the hydropower station at the current time and each reference state threshold of the gate, and preprocess the gate operation monitoring data to obtain preprocessed gate operation monitoring data.

[0050] It should be noted that the gate operation monitoring data comprises external driving force, friction, water flow force, force generated when the gate is opened, force generated when the gate is closed, vibration frequency, water flow, and water level.

[0051] It should be noted that a state reference model is established by considering various factors such as the position, movement, mechanical behavior, and vibration of the gate, in combination with the operating conditions under the specific environment of the hydropower station, so as to obtain each reference state threshold, thereby laying a foundation for subsequent fault analysis.

[0052] For example, the values or ranges of each variable under normal operating conditions are determined through historical data and experimental data. The reference values, i.e., each reference state threshold, under normal operation are set according to the equipment specifications, usage, and environmental conditions.

[0053] In the embodiments of the present disclosure, the preprocessing of the gate operation monitoring data to obtain preprocessed gate operation monitoring data comprises the following steps.

[0054] The gate operation monitoring data is sequentially subjected to cleaning, removal of abnormal values, completion of missing data and filtering processing to obtain preprocessed gate operation monitoring data.

[0055] It should be noted that the preprocessing includes:

[0056] The data is cleaned, abnormal values are removed, missing data is completed and filtering processing is performed;

[0057] The collected abnormal data is filtered.

[0058] The collected monitoring data may have problems such as noise, missing values, abnormal fluctuations, and therefore needs to be preprocessed. Data preprocessing includes data cleaning, removal of abnormal values, completion of missing data, and filtering processing. For abnormal data, a method based on statistical analysis and pattern recognition is used for filtering. For example, by setting a suitable threshold range, data that exceeds the reasonable range (such as excessively high pressure, displacement out of limits, etc.) can be detected and removed, and data that does not conform to the actual physical law can be removed. At the same time, missing values are filled in to ensure the integrity of the data. Through these preprocessing steps, subsequent analysis and decision-making are based on accurate and clean data, thereby avoiding false positives and false negatives.

[0059] Step 2: determining the gate reference state deviation of the hydropower station according to the preprocessed gate operation monitoring data, the reference state threshold and the pre-established state reference model;

[0060] In the embodiments of the present disclosure, the pre-established state reference model includes:

[0061] a gate position state model, a mechanical state model, a vibration and wear model, and a water flow and environment model;

[0062] It should be noted that the gate position state model can describe the position and motion state of the gate.

[0063] The gate position state model sets that the motion of the gate is driven by an external force and is affected by friction and water flow force, and the position of the gate is described by the following first-order dynamics equation:

[0064]

[0065] In the formula, x(t) is the gate position, F external (t) is the external driving force, F friction (t) is the friction force, F water (t) is the water flow force, γ1 is the damping coefficient reflecting the influence of friction, and k1 is the elastic coefficient reflecting the elastic characteristics of the system.

[0066] The equation describes the position change of the gate under the action of external force driving, friction and water flow force, and can simulate the motion characteristics of the gate in the normal operation process.

[0067] The mechanical state model can describe the forces acting on the gate during operation, including opening and closing forces, resistance;

[0068] The mechanical state model, the mechanical characteristics experienced by the gate during operation, is described by the following formula:

[0069] F(t)=F opening (t)+F closing (t)

[0070] In the formula, F(t) is the opening and closing force required by the gate, F opening (t) is the force generated when the gate is opened, and F closing (t) is the force generated when the gate is closed.

[0071] The model can help staff assess the opening and closing forces required by the gate under different operating modes, and compare them with the reference mechanical state to determine whether there are abnormalities, such as excessive friction or operating resistance.

[0072] The vibration and wear model can describe the vibration and wear state of the gate during operation;

[0073] The vibration and wear model, the vibration state of the gate, is as follows:

[0074] v(t)=a1×sin(ωt)+a2×cos(ωt)+η(t)

[0075] In the formula, v(t) is the vibration speed at time t, a1 is the first vibration amplitude coefficient, a2 is the second vibration amplitude coefficient, ω is the angular velocity, and η(t) is the random noise caused by wear factors.

[0076] The role of the model is to evaluate the vibration characteristics of the gate during use and identify possible structural problems in advance. For example, wear of gate components can cause changes in frequency or irregular vibration patterns.

[0077] The water flow and environmental model considers the influence of water flow and environmental factors such as climate on the operation of the gate.

[0078] The water flow and environmental model, the model describing the impact of water flow mechanics on the gate, is as follows:

[0079] w(t)=β1×Q(t)+β2×H(t)

[0080] Wherein, w(t) is the impact force generated by water flow, Q(t) is water flow, H(t) is water level, β1 is the influence coefficient of water flow, and β2 is the influence coefficient of water head;

[0081] The model can predict the working condition of the gate according to the water flow and the water level change, and help identify the additional pressure or resistance generated by the gate due to environmental changes.

[0082] It should be noted that by installing high-precision sensors and collection equipment, the data of the gate in different operating states are monitored and collected in real time, and the monitored data includes the data input to the state reference model.

[0083] In the embodiment, the input data includes but is not limited to the opening and closing state of the gate, position sensor data, pressure and force sensor feedback, flow and water level data, etc. The monitored data is not only limited to the basic position of the gate, but also includes mechanical stress, vibration and other details, so as to identify potential wear or abnormal phenomena.

[0084] Further, the gate reference state deviation of the hydropower station includes:

[0085] Gate position deviation, stress deviation, vibration speed deviation, and water flow impact force deviation.

[0086] Step 3: inputting the gate reference state deviation of the hydropower station into the pre-established fault prediction model to obtain a gate fault diagnosis result of the hydropower station at the current time;

[0087] The gate fault diagnosis result includes a fault mode of the gate and a risk value.

[0088] In the embodiment of the present disclosure, the establishment process of the fault prediction model includes:

[0089] Obtaining the gate reference state values at each time in a historical period and the fault modes corresponding to each time;

[0090] Using machine learning technology, taking the gate reference state values at each time in the historical period as input and taking the fault modes corresponding to each time as output, the initial fault prediction model is optimized and trained to obtain a trained fault prediction model;

[0091] The fault mode includes: sticking, water leakage, and wear.

[0092] It should be noted that the risk value is calculated as follows:

[0093]

[0094] Wherein, f(t) is the risk value at time t, α i is the weight of the i-th gate reference state value, xi (t) is the i-th reference state value of the gate at time t, and n is the total number of gate reference state values.

[0095] It should be noted that the weight of each reference state value can be achieved by the following methods:

[0096] Weight setting based on expert experience. When there is not enough historical data or complex algorithm support, the weight coefficient can be manually set by expert experience or domain knowledge. For example, based on the working characteristics of the gate, it is believed that the position change has a greater impact on the failure, and the vibration frequency may be less important, that is, the weight of the gate position state model is large, and the weight of the vibration and wear model is less important.

[0097] Statistical analysis method based on data. Use statistical analysis (such as Pearson correlation coefficient, analysis of variance, etc.) to calculate the correlation between each state variable and the occurrence of failure, and then determine the weight coefficient. For example, if a certain variable (such as the vibration and wear model) has a high correlation with the occurrence of failure, then the weight coefficient of the variable will be relatively large. That is, a large amount of historical failure data and normal operation data are collected, the correlation between each state variable and the occurrence of failure is calculated, and the state variable with strong correlation is given a higher weight coefficient.

[0098] Further, the embodiment also includes:

[0099] The real-time monitoring data of the gate state is extracted and compared with the reference model to determine whether problems such as jamming, leakage, abnormal vibration, etc. have occurred. Through comparative analysis, conditions that are significantly different from the normal state can be identified, and specific diagnostic reports can be generated. The core of fault diagnosis is to provide targeted solutions based on known fault patterns and current data.

[0100] It should be noted that the fault prediction model first trains the historical operation data of the gate through machine learning technology. The training data includes the reference state during normal operation and various known fault patterns, including sample data of jamming, leakage, and wear;

[0101] The input data of the fault prediction model directly comes from the state variables and the output of the state reference model, combined with historical operation data, to form a multi-dimensional data set with time sequence characteristics.

[0102] The output data of the fault prediction model provides risk prediction factors and potential failure information, providing a scientific basis for fault prediction and management.

[0103] The risk value of f(t) is obtained through real-time data input into the fault prediction model, a warning signal is provided based on the risk value, which is used to determine whether the gate is in a failure critical state, and potential faults existing in the gate are identified according to abnormal values of state variables output by a certain model in the state benchmark model.

[0104] Further, the method further comprises:

[0105] determining whether the risk value is greater than a preset risk threshold, and generating a warning signal for fault warning when the risk value is greater than the preset risk threshold.

[0106] Further, the method further comprises:

[0107] based on the gate fault diagnosis result, the pre-established fault prediction model is adaptively optimized and trained.

[0108] It should be noted that, according to the results of fault diagnosis, as new training data of the fault prediction model, the diagnosis model is continuously adjusted and optimized, so as to improve the accuracy and response speed of the future fault prediction model. Through this adaptive learning method, the system can continuously improve the accuracy of fault diagnosis in long-term use.

[0109] Further, the method further comprises:

[0110] The sensor arranged at the preset position is used to collect the gate operation monitoring data;

[0111] based on the fault mode of the gate in the gate fault diagnosis result and the gate operation monitoring data collected by the sensor arranged at the preset position, the fault occurrence position is determined.

[0112] It should be noted that, according to the fault prediction model, the fault mode is identified based on the known fault mode;

[0113] According to the identified fault mode, the system accurately determines the fault occurrence position through comparison of data sources of the sensor and historical data.

[0114] Through comparison of data sources of the sensor (such as temperature sensor, vibration sensor, displacement sensor, etc.) and historical data, the system can accurately determine the fault occurrence position. For example, if the vibration sensor data is abnormal and the temperature sensor data is normal, the system may speculate that a certain mechanical component of the gate has failed; if the displacement sensor shows abnormality and other sensors are normal, it may be that the moving part of the gate is stuck.

[0115] Meanwhile, after fault location, the system generates real-time feedback based on the diagnosis results and historical maintenance records, and suggests repair measures. For example, if it is found that the gate position is abnormal due to excessive water flow pressure, the system may suggest increasing the adjustment of water flow pressure or adjusting the sealing of the gate. If it is due to mechanical failure caused by wear and tear, it may suggest replacing parts or lubricating.

[0116] In summary, the gate state monitoring method for hydropower stations proposed in this embodiment has the following advantages: 1. By establishing a state reference model of the gate, implementing multi-modal data fusion, real-time monitoring, and combining a fault prediction model, the shortcomings of the prior art are overcome. 2. This method not only enables comprehensive monitoring of the gate operating state, but also identifies potential fault risks in advance through the fault prediction model, avoiding the false positives and false negatives of traditional monitoring methods. 3. Through accurate fault diagnosis and location, the fault location and type can be quickly determined, and efficient maintenance and repair can be achieved. 4. The prediction model is dynamically adjusted according to real-time monitoring data and fault diagnosis results, thereby continuously improving the accuracy of fault diagnosis and the timeliness of system response. Ultimately, this method can significantly improve the safety, reliability, and operating efficiency of the gate of the hydropower station, reduce maintenance costs, and extend the service life of the equipment.

[0117] Embodiment Two

[0118] Figure 2 The structure diagram of a gate state monitoring system for a hydropower station according to an embodiment of the present application is shown in Figure 2 As shown in the figure, the system comprises:

[0119] The acquisition module 100 is configured to acquire gate operating monitoring data of the hydropower station at the current time and each reference state threshold of the gate, and pre-process the gate operating monitoring data to obtain pre-processed gate operating monitoring data.

[0120] The gate operating monitoring data includes external driving force, friction force, water flow force, force generated when the gate is opened, force generated when the gate is closed, vibration frequency, water flow, and water level.

[0121] The determination module 200 is configured to determine the gate reference state deviation of the hydropower station according to the pre-processed gate operating monitoring data, the each reference state threshold, and the pre-established state reference model.

[0122] The pre-established state reference model includes:

[0123] a gate position state model, a mechanical state model, a vibration and wear model, and a water flow and environment model.

[0124] The gate reference state deviation of the hydropower station includes:

[0125] The gate position deviation, the force deviation, the vibration speed deviation, and the water flow impact force deviation.

[0126] The diagnosis module 300 is configured to input the gate reference state deviation of the hydropower station into a pre-established fault prediction model to obtain a gate fault diagnosis result of the hydropower station at the current moment.

[0127] The gate fault diagnosis result includes a fault mode of the gate and a risk value.

[0128] The establishment process of the fault prediction model includes:

[0129] The gate reference state values at each moment in a historical period and the fault modes corresponding to each moment are obtained.

[0130] An initial fault prediction model is optimized and trained by using a machine learning technology, taking the gate reference state values at each moment in the historical period as input and taking the fault modes corresponding to each moment as output, to obtain a trained fault prediction model.

[0131] The fault mode includes sticking, water leakage, and wear.

[0132] The risk value is calculated according to the following formula:

[0133]

[0134] In the formula, f(t) is the risk value at the moment t, α i is the weight of the i-th gate reference state value, x i (t) is the i-th gate reference state value at the moment t, and n is the total number of gate reference state values.

[0135] In the embodiments of the present disclosure, the acquisition module 100 is further configured to:

[0136] The gate operation monitoring data is sequentially subjected to cleaning, removal of abnormal values, completion of missing data, and filtering processing to obtain preprocessed gate operation monitoring data.

[0137] In the embodiments of the present disclosure, as shown in Figure 3 The system further includes a warning module 400.

[0138] The warning module 400 is configured to determine whether the risk value is greater than a preset risk threshold value, and generate a warning signal for fault warning when the risk value is greater than the preset risk threshold value.

[0139] In the embodiments of the present disclosure, as shown in Figure 3 The system further includes an optimization module 500.

[0140] The optimization module 500 is configured to perform adaptive optimization training on the pre-established fault prediction model based on the gate fault diagnosis result.

[0141] In the embodiments of the present disclosure, as shown in Figure 3 The system further comprises a fault location identification module 600.

[0142] The fault location identification module 600 is configured to collect gate operation monitoring data by using sensors arranged at preset positions.

[0143] The fault location identification module 600 is further configured to determine the fault occurrence position based on the fault mode of the gate in the gate fault diagnosis result and the gate operation monitoring data collected by the sensors arranged at the preset positions.

[0144] In summary, the gate state monitoring system for hydropower stations proposed in the embodiments can greatly improve the safety, reliability and operation efficiency of the gates of the hydropower stations, reduce the maintenance cost and prolong the service life of the equipment.

[0145] Embodiment Three

[0146] To achieve the above-mentioned embodiments, the present disclosure further proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method according to the first embodiment.

[0147] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the skilled in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples without contradiction.

[0148] Any process or method descriptions in flow charts or described elsewhere herein can be understood as representing one or more steps in a set of steps performed in one or more processes or methods. The scope of preferred embodiments of the present application encompasses numerous additional implementation sequences, examples of which are described in the description herein. The processes or methods described in flow charts or otherwise described herein can be understood as representing executable instructions stored in a computer-readable medium, which can be executed by a processor, and the scope of preferred embodiments of the present application encompasses numerous additional implementation sequences, examples of which are described in the description herein.

[0149] Although the embodiments of the present application have been shown and described above, it is understood that the above-described embodiments are exemplary and are not to be construed as limiting the present application, and that changes, modifications, substitutions and variations can be made by those skilled in the art without departing from the scope of the present application.

Claims

1. A method for monitoring the state of a gate for a hydroelectric power plant, characterized by, The method comprises: acquiring gate operation monitoring data of a hydropower station at a current time and each reference state threshold of the gate, and preprocessing the gate operation monitoring data to obtain preprocessed gate operation monitoring data; determining a gate reference state deviation of the hydropower station according to the preprocessed gate operation monitoring data, the each reference state threshold, and a pre-established state reference model; inputting the gate reference state deviation of the hydropower station into a pre-established fault prediction model to obtain a gate fault diagnosis result of the hydropower station at the current time; wherein the gate fault diagnosis result comprises a fault mode of the gate and a risk value; the pre-established state reference model comprises: a gate position state model, a mechanical state model, a vibration and wear model, and a water flow and environment model; a calculation formula of the gate position state model is as follows: where x(t) is the gate position, F external (t) is the external driving force, F friction (t) is the friction force, F water (t) is the water flow force, γ1 is the damping coefficient reflecting the influence of the friction, and k1 is the elastic coefficient reflecting the elastic characteristics of the system. a calculation formula of the mechanical state model is as follows: F(t) = F opening (t) + F closing (t) where F(t) is the required opening and closing force of the gate, F opening (t) is the force generated when opening the gate, F closing (t) is the force generated when closing the gate; a calculation formula of the vibration and wear model is as follows: v(t)=a1×sin(ωt)+a2×cos(ωt)+η(t) wherein v(t) is a vibration speed at time t, a1 is a first vibration amplitude coefficient, a2 is a second vibration amplitude coefficient, ω is an angular velocity, and η(t) is random noise caused by wear factors; a calculation formula of the water flow and environment model is as follows: w(t)=β1×Q(t)+β2×H(t) wherein w(t) is an impact force generated by water flow, Q(t) is water flow, H(t) is water level, β1 is an influence coefficient of water flow force, and β2 is an influence coefficient of water head; the gate reference state deviation of the hydropower station comprises: a gate position deviation, a force deviation, a vibration speed deviation, and a water flow impact force deviation.

2. The method of claim 1, wherein, the preprocessing of the gate operation monitoring data to obtain the preprocessed gate operation monitoring data comprises: cleaning, removing abnormal values, completing missing data, and filtering the gate operation monitoring data in sequence to obtain the preprocessed gate operation monitoring data; the gate operation monitoring data comprises an external driving force, a friction force, a water flow force, a force generated when the gate is opened, a force generated when the gate is closed, a vibration frequency, a water flow, and a water level.

3. The method of claim 2, wherein, a process of establishing the fault prediction model comprises: acquiring each gate reference state value at each time in a historical period and a corresponding fault mode at each time; using a machine learning technology, taking the each gate reference state value at each time in the historical period as input and taking the corresponding fault mode at each time as output, optimizing and training an initial fault prediction model to obtain a trained fault prediction model; wherein the fault mode comprises sticking, water leakage, and wear.

4. The method of claim 3, wherein, a calculation formula of the risk value is as follows: In the formula, f(t) is the risk value at time t, α i is the weight of the i-th reference state value of the gate, x i (t) is the i-th reference state value of the gate at time t, and n is the total number of reference state values of the gate.

5. The method of claim 4, wherein, the method further comprises: judging whether the risk value is greater than a preset risk threshold, and generating an early warning signal for fault early warning when the risk value is greater than the preset risk threshold.

6. The method of claim 5, wherein, the method further comprises: performing adaptive optimization training on the pre-established fault prediction model based on the gate fault diagnosis result.

7. The method of claim 6, wherein, the method further comprises: collecting gate operation monitoring data by using sensors arranged at preset positions; The failure occurrence position is determined based on the failure mode of the gate in the gate failure diagnosis result and the gate operation monitoring data collected by the sensors arranged at the preset positions.

8. A gate condition monitoring system for a hydroelectric power plant based on the method of any one of the preceding claims 1 to 7, characterized in that, The system comprises: An acquisition module is configured to acquire gate operation monitoring data of a hydropower station at a current time and each reference state threshold of a gate, and to pre-process the gate operation monitoring data to obtain pre-processed gate operation monitoring data; A determination module is configured to determine a gate reference state deviation of the hydropower station according to the pre-processed gate operation monitoring data, the each reference state threshold and a pre-established state reference model; A diagnosis module is configured to input the gate reference state deviation of the hydropower station into a pre-established failure prediction model to obtain a gate failure diagnosis result of the hydropower station at the current time; The gate failure diagnosis result comprises a failure mode of the gate and a risk value.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the method of any one of claims 1-7.

Images