Hydropower station gate detection method and system

Abstract

The invention discloses a hydropower station gate detection method and system, and relates to the technical field of hydropower station gate detection, and the method comprises the steps: collecting environment parameters, building a dynamic environment model, and determining a real-time impact factor on a gate state; synchronously collecting gate vibration and sound wave data to obtain a composite signal; constructing a short-term and long-term trend model, and identifying a potential abnormal trend; automatically adjusting the sampling frequency of the sensor according to an abnormal detection result; performing fault classification on the multi-parameter signals by using an event pattern recognition algorithm; and generating an automatic response strategy according to a fault classification result. According to the hydropower station gate detection method provided by the invention, the accuracy and efficiency of hydropower station gate detection are improved through multi-dimensional parameter fusion and adaptive sampling adjustment. Through monitoring of various parameters, the system can comprehensively analyze the state of the gate, the misjudgment risk caused by a single signal is reduced, and comprehensive recognition of faults is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of hydropower station gate detection, and in particular to a hydropower station gate detection method and system. Background Art

[0002] The reliable operation of the gates of a hydropower station is crucial to the safety and stability of the hydropower station. The gate status monitoring and fault detection directly affect the operating efficiency, equipment life and safety of the hydropower station. However, the current hydropower station gate monitoring method usually relies on sensor data collection with a single parameter or a fixed sampling frequency, which cannot timely reflect the gate operation status in a complex environment, and it is difficult to effectively detect multiple types of faults, resulting in potential faults that are difficult to detect in time. The existing detection technology has the following main defects:

[0003] In the existing technology, gate status monitoring mainly relies on single parameter monitoring of vibration signals or sound wave signals, lacking comprehensive consideration of environmental factors such as ambient temperature, air pressure and water flow rate. Since environmental parameters will have different degrees of influence on the gate operation status, a single parameter detection method is difficult to fully and accurately capture the changes in the gate status, resulting in large deviations in the detection results.

[0004] In traditional hydropower station gate detection systems, the sampling frequency of sensors is usually fixed and not automatically adjusted for abnormal situations. This fixed-frequency sampling method has major defects in data redundancy and the accuracy of capturing important fault signals. Low sampling frequency may cause early fault signals to be lost, while high sampling frequency will lead to unnecessary data storage overhead and increase the data processing burden.

[0005] Current fault detection technologies mostly use simple threshold judgments, which cannot distinguish between abnormalities caused by mechanical failures, electrical failures, or environmental factors, resulting in low accuracy of early warning signals. A single threshold setting is also difficult to effectively identify potential fault trends under different risk levels, making it difficult for traditional methods to meet the needs of accurate classification and risk-based early warning of complex faults in actual operations.

[0006] In current technology, hydropower station gate detection systems usually rely on human judgment and manual control to respond after detecting an abnormality. This approach has the risk of delayed response, especially in high-risk situations, which may lead to safety hazards. In addition, traditional systems lack automated fault classification response strategies, making it difficult to automatically generate reasonable maintenance and operation decisions based on the degree of fault risk. Summary of the invention

[0007] In view of the above-mentioned problems, the present invention is proposed.

[0008] Therefore, the technical problem solved by the present invention is: the existing hydropower station gate detection method has the problems of single signal detection leading to false alarms and missed alarms, fixed sampling frequency unable to respond to fault changes in time, and lack of fault classification and automated response, as well as how to achieve optimization of early warning and efficient response to faults in complex environments.

[0009] In order to solve the above technical problems, the present invention provides the following technical solutions: a hydropower station gate detection method, comprising:

[0010] Collect environmental parameters, establish a dynamic environmental model, and determine the real-time influencing factors on the gate status;

[0011] Synchronously collect gate vibration and sound wave data to obtain composite signals;

[0012] Build short-term and long-term trend models to identify potential abnormal trends;

[0013] Automatically adjust sensor sampling frequency based on anomaly detection results;

[0014] Use event pattern recognition algorithm to classify faults of multi-parameter signals;

[0015] Generate automated response strategies based on fault classification results.

[0016] As a preferred solution of the hydropower station gate detection method of the present invention, wherein: the determination of the real-time influencing factors on the gate state includes real-time collection of environmental parameter data, including water temperature, air pressure and water flow velocity, through an environmental sensor network, and recording the environmental parameters according to timestamps to form a time series data stream of each environmental parameter;

[0017] Normalize the collected water temperature, air pressure and water flow velocity data to adjust the data of each parameter to a unified numerical range;

[0018] Pearson correlation coefficient was used to calculate the correlation between each environmental parameter and the change in gate status;

[0019] The absolute value of the correlation degree of each environmental parameter is used as the initial weight, and the normalization factor is obtained by summing up all the absolute values ​​of the correlations. The final weight value is generated by dividing the absolute value of each parameter by the normalization factor;

[0020] Multiply the normalized environmental parameter data with the corresponding weights, calculate the weighted value of each parameter, and obtain the relative influence of each parameter on the gate status;

[0021] The various weighted results are added together to generate a comprehensive impact factor, which represents the overall degree of influence of environmental factors on the gate status.

[0022] As a preferred solution of the hydropower station gate detection method described in the present invention, wherein: the obtaining of the composite signal includes increasing the sensitivity of the vibration sensor to a preset level and adjusting the frequency bandwidth of the acoustic wave sensor to a preset range when the impact factor value is higher than a preset threshold value;

[0023] When the environmental impact factor value is lower than a preset threshold, the sensitivity of the vibration sensor is set to be lowered to a preset level and the sound wave frequency bandwidth range is reduced according to a preset range;

[0024] Based on the adjusted acquisition parameters, start the synchronous acquisition of vibration and sound wave signals;

[0025] Align the collected vibration data and sound wave data based on the time axis to generate a composite signal;

[0026] Perform spectrum analysis on the generated composite signal, convert the composite signal in the frequency domain, extract the frequency components and amplitude values ​​of the signal, and determine whether there are abnormal signal characteristics based on the characteristic values ​​of frequency and amplitude. Mark the frequency and amplitude combination that meets the characteristic values ​​of mechanical fatigue or structural looseness as abnormal signals, record the spectrum characteristic values ​​of the composite signal, and mark the timestamp and frequency range of the abnormal characteristic values.

[0027] As a preferred solution of the hydropower station gate detection method of the present invention, wherein: the identifying potential abnormal trends includes arranging the spectral characteristic values, timestamps and frequency ranges of the recorded abnormal signals in the order of acquisition time and storing them as time series data;

[0028] On a time scale of seconds, a fixed second-level time window is set, the spectrum characteristic values ​​of the abnormal signal are extracted from each time window, the frequency and amplitude fluctuations in the time window are calculated, and a short-term trend model including the fluctuations in the second-level time window is generated;

[0029] On the hourly time scale, a fixed hourly time window is set, the spectrum feature values ​​of abnormal signals are extracted from each hourly time window, the cumulative changes of the spectrum feature values ​​of each time window are calculated, and a long-term trend model that reflects the long-term change trend is generated;

[0030] In the current hourly time window, compare the cumulative change of the abnormal signal in the window with the cumulative change in the previous hourly window, calculate the change in frequency and amplitude, and if the change exceeds the preset threshold, mark the hourly window as an abnormal window and record its characteristic value;

[0031] When rapid fluctuations in frequency or amplitude occur in the second-level time window of the short-term trend model, the fluctuation data is compared with the abnormal window data in the long-term trend model; if the short-term fluctuation occurs in the hour-level window marked as abnormal, and the fluctuation amplitude exceeds the preset threshold, the time point is marked as a potential fault signal, and the spectral characteristic value, timestamp and frequency range of the fault signal are recorded.

[0032] As a preferred solution of the hydropower station gate detection method described in the present invention, the automatic adjustment of the sensor sampling frequency includes adjusting the sensor sampling frequency of the potential fault signal according to a preset rule.

[0033] As a preferred solution of the hydropower station gate detection method described in the present invention, wherein: the fault classification includes inputting the adjusted composite signal into the event pattern recognition algorithm, comparing each parameter with the characteristic value combination in the fault mode library through the multi-layer branch conditions of the decision tree, and realizing the classification of mechanical, electrical and environmental abnormal fault types;

[0034] According to the characteristic parameters, the faults are further classified into initial faults, potential faults and persistent faults, and each fault is marked with a corresponding risk level label.

[0035] As a preferred solution of the hydropower station gate detection method described in the present invention, the generation of an automated response strategy includes dividing the health status of the gate into three risk levels: low, medium and high, according to the fault classification results, corresponding to the response strategies of data recording and trend analysis, maintenance reminder and early warning adjustment, emergency closure operation and remote control switching, respectively.

[0036] Another object of the present invention is to provide a hydropower station gate detection system, which can solve the problems of insufficient comprehensive judgment of multiple factors, delayed dynamic adjustment and low response efficiency in the existing hydropower station gate monitoring by constructing an intelligent detection system based on multi-parameter fusion, adaptive sampling adjustment and event pattern recognition, and realize comprehensive monitoring of gate status, accurate classification of fault types and automatic graded response of risk levels.

[0037] In order to solve the above technical problems, the present invention provides the following technical solutions: a hydropower station gate detection system, comprising: a data acquisition module, a data fusion module, an abnormality identification module, a frequency adjustment module, a fault classification module and an abnormality adjustment module; the data acquisition module is used to collect environmental parameters, establish a dynamic environmental model, and determine the real-time influencing factors on the gate state; the data fusion module is used to synchronously collect gate vibration and sound wave data to obtain a composite signal; the abnormality identification module is used to construct short-term and long-term trend models to identify potential abnormal trends; the frequency adjustment module is used to automatically adjust the sensor sampling frequency according to the abnormality detection results; the fault classification module is used to use an event pattern recognition algorithm to classify multi-parameter signals for faults; the abnormality adjustment module is used to generate an automated response strategy according to the fault classification results.

[0038] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned hydropower station gate detection method when executing the computer program.

[0039] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the hydropower station gate detection method as described above.

[0040] Beneficial effects of the present invention: The hydropower station gate detection method provided by the present invention improves the accuracy and efficiency of hydropower station gate detection through multi-dimensional parameter fusion and adaptive sampling adjustment. By using multiple parameter monitoring, the system can comprehensively analyze the gate status, reduce the risk of misjudgment caused by a single signal, and enhance the comprehensive identification of faults. The adaptive sampling adjustment mechanism enables the system to dynamically adjust the acquisition frequency according to the actual detection situation, ensure sensitive detection in high-risk situations, and reduce resource consumption in low-risk situations.

[0041] In addition, the trend analysis model of the present invention effectively identifies short-term and long-term fault change trends, enabling the system to monitor fault trends and identify potential risks earlier. Through the hierarchical response strategy, the system can quickly take corresponding measures based on the fault classification results, improving the real-time response capability of gate detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0043] Figure 1 An overall flow chart of a hydropower station gate detection method provided for one embodiment of the present invention.

[0044] Figure 2 An overall structural diagram of a hydropower station gate detection system provided for one embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0046] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0047] Example 1

[0048] Reference Figure 1 , as an embodiment of the present invention, provides a hydropower station gate detection method, comprising:

[0049] Collect environmental parameters, establish a dynamic environmental model, and determine the real-time influencing factors on the gate status;

[0050] Synchronously collect gate vibration and sound wave data to obtain composite signals;

[0051] Build short-term and long-term trend models to identify potential abnormal trends;

[0052] Automatically adjust sensor sampling frequency based on anomaly detection results;

[0053] Use event pattern recognition algorithm to classify faults of multi-parameter signals;

[0054] Generate automated response strategies based on fault classification results.

[0055] Determining the real-time influencing factors on the gate state includes collecting environmental parameter data in real time through an environmental sensor network, including water temperature, air pressure and water flow velocity, and recording the environmental parameters according to timestamps to form a time series data stream for each environmental parameter;

[0056] Normalize the collected water temperature, air pressure and water flow velocity data to adjust the data of each parameter to a unified numerical range;

[0057] Pearson correlation coefficient was used to calculate the correlation between each environmental parameter and the change in gate status;

[0058] Specifically, it includes: aligning the time series data of each environmental parameter with the data of gate status changes according to timestamps to form a multidimensional data matrix, and calculating the degree of correlation between each environmental parameter and the gate status based on the Pearson correlation method.

[0059] The absolute value of the correlation degree of each environmental parameter is used as the initial weight, and the normalization factor is obtained by summing up all the absolute values ​​of the correlations. The final weight value is generated by dividing the absolute value of each parameter by the normalization factor;

[0060] Multiply the normalized environmental parameter data with the corresponding weights, calculate the weighted value of each parameter, and obtain the relative influence of each parameter on the gate status;

[0061] The various weighted results are added together to generate a comprehensive impact factor, which represents the overall degree of influence of environmental factors on the gate status.

[0062] The obtaining of the composite signal includes increasing the sensitivity of the vibration sensor to a preset level and adjusting the frequency bandwidth of the acoustic sensor to a preset range when the impact factor value is higher than a preset threshold value, so that the sensor can adapt to the signal collection requirements under the current environmental conditions;

[0063] When the environmental impact factor value is lower than a preset threshold, the sensitivity of the vibration sensor is set to be lowered to a preset level and the sound wave frequency bandwidth range is reduced according to a preset range;

[0064] Based on the adjusted acquisition parameters, start the synchronous acquisition of vibration and acoustic wave signals; keep the sampling frequency of vibration and acoustic wave signal acquisition consistent, collect the vibration data and acoustic wave data of the gate at the same time interval, and synchronize the recorded signal timestamps to ensure the time consistency of the data.

[0065] The collected vibration data and acoustic wave data are aligned based on the time axis to generate a composite signal. By matching the data points with the same timestamp on the time axis, a pair of vibration and acoustic wave data with consistent time is formed. The median filter algorithm is used to process each data pair to remove random noise in the signal to enhance the stability of the data. The processed vibration data and acoustic wave data are integrated into a composite signal.

[0066] Perform spectrum analysis on the generated composite signal, convert the composite signal in the frequency domain, extract the frequency components and amplitude values ​​of the signal, and determine whether there are abnormal signal characteristics based on the characteristic values ​​of frequency and amplitude. Mark the frequency and amplitude combination that meets the characteristic values ​​of mechanical fatigue or structural looseness as abnormal signals, record the spectrum characteristic values ​​of the composite signal, and mark the timestamp and frequency range of the abnormal characteristic values.

[0067] The identifying of potential abnormal trends includes arranging the frequency spectrum characteristic values, timestamps and frequency ranges of the recorded abnormal signals in the order of acquisition time and storing them as time series data;

[0068] On a time scale of seconds, a fixed second-level time window is set, the spectrum characteristic values ​​of the abnormal signal are extracted from each time window, the frequency and amplitude fluctuations in the time window are calculated, and a short-term trend model including the fluctuations in the second-level time window is generated;

[0069] On the hourly time scale, a fixed hourly time window is set, the spectrum characteristic values ​​of abnormal signals are extracted from each hourly time window, the cumulative changes of the spectrum characteristic values ​​of each time window are calculated, and a long-term trend model reflecting the long-term change trend is generated to characterize the cumulative changes of the gate status;

[0070] In the current hourly time window, the cumulative change of the abnormal signal in the window is compared with the cumulative change in the previous hourly window, and the change in frequency and amplitude is calculated. If the change exceeds the preset threshold, the hourly window is marked as an abnormal window, and its characteristic value is recorded to characterize the cumulative change of the gate state;

[0071] When rapid fluctuations in frequency or amplitude occur in the second-level time window of the short-term trend model, the fluctuation data is compared with the abnormal window data in the long-term trend model; if the short-term fluctuation occurs in the hour-level window marked as abnormal, and the fluctuation amplitude exceeds the preset threshold, the time point is marked as a potential fault signal, and the spectral characteristic value, timestamp and frequency range of the fault signal are recorded.

[0072] The automatically adjusting the sensor sampling frequency includes adjusting the sensor sampling frequency of the potential fault signal according to a preset rule.

[0073] It should be noted that when a potential fault signal is detected and marked as an abnormal signal, the sampling frequency is adjusted to twice the original sampling frequency to improve the signal acquisition accuracy, and the changes in the spectrum characteristic values ​​are recorded in detail;

[0074] If no persistent abnormal signal or potential fault signal is detected in the short-term trend model and the long-term trend model, the sampling frequency is adjusted down to 50% of the original frequency to reduce the data storage burden.

[0075] The fault classification includes inputting the adjusted composite signal into the event pattern recognition algorithm, comparing the characteristic value combination of each parameter with the fault pattern library through the multi-layer branch conditions of the decision tree, and realizing the classification of the fault types of mechanical, electrical and environmental abnormalities;

[0076] According to the characteristic parameters, including frequency range, amplitude fluctuation and cumulative change rate, the faults are further classified into initial faults, potential faults and persistent faults, and each fault is marked with a corresponding risk level label.

[0077] Combine the adjusted acquisition data with multiple detection parameters (including vibration, sound waves, position and pressure) to generate a multi-parameter signal data set, and normalize each parameter data to eliminate the dimensional differences between different parameters;

[0078] Establish a fault mode library based on historical data and experience. The fault mode library contains characteristic value combinations of typical mechanical faults, electrical faults, and abnormal environmental factors. Each fault mode includes characteristic parameters such as frequency range, amplitude threshold, response time, and cumulative fluctuation changes.

[0079] The multi-parameter signal data set is input to the decision tree algorithm classifier, where:

[0080] The first layer of branches classifies signals into mechanical anomalies, electrical anomalies, and environmental anomalies based on abnormal changes in the frequency range;

[0081] The second layer branch further classifies the signal into fault state or normal state according to the amplitude fluctuation range. The specific judgment conditions include whether the amplitude exceeds the normal fluctuation range in the fault mode library.

[0082] The third branch further determines the severity of the fault state based on the cumulative fluctuation change rate, and determines whether the signal is an initial fault, a potential fault, or a persistent fault;

[0083] Each classification result is marked and assigned a fault level label to provide a reference for subsequent response strategies.

[0084] The generation of the automated response strategy includes dividing the gate's health status into three risk levels: low, medium and high, according to the fault classification results, corresponding to the response strategies of data recording and trend analysis, maintenance reminder and early warning adjustment, emergency closure operation and remote control switching, respectively.

[0085] According to the fault classification results and fault level labels, the health status of the gate is divided into three levels of risk: low, medium, and high, including:

[0086] Low risk: When the initial fault signal is detected and the fluctuation range does not exceed 10%, the data is recorded and stored in the historical database for subsequent monitoring and analysis;

[0087] Medium risk: When a potential fault signal is detected and the frequency or amplitude fluctuation is between 10% and 30%, a maintenance reminder is generated and the sampling frequency is dynamically adjusted to increase the capture accuracy of the fault signal so as to monitor the fault status more carefully.

[0088] High risk: When a persistent fault signal is detected and the fluctuation exceeds 30%, the system alarm is triggered and a control command is automatically issued to perform an emergency closure operation;

[0089] Synchronize low, medium and high risk graded response strategies to the monitoring center and generate subsequent operation instructions at each risk level, including:

[0090] Trend analysis: In low and medium risk states, trend analysis is performed on the collected data at preset time intervals to confirm whether the risk state has changed;

[0091] Real-time warning adjustment: In a medium-risk state, the warning mechanism is triggered and the detection sensitivity of vibration and acoustic sensors is adjusted to further optimize the real-time monitoring of the gate status;

[0092] Remote control switching: In high-risk situations, on-site operators are notified immediately and the system automatically switches to remote control mode to facilitate timely remote monitoring and manual intervention.

[0093] Example 2

[0094] Reference Figure 2 , as an embodiment of the present invention, provides a hydropower station gate detection system, comprising:

[0095] Data acquisition module 100, data fusion module 200, abnormality identification module 300, frequency adjustment module 400, fault classification module 500 and abnormality adjustment module 600;

[0096] The data acquisition module 100 is used to collect environmental parameters, establish a dynamic environmental model, and determine the real-time influencing factors on the gate status;

[0097] The data fusion module 200 is used to synchronously collect gate vibration and sound wave data to obtain a composite signal;

[0098] The anomaly identification module 300 is used to construct short-term and long-term trend models to identify potential abnormal trends;

[0099] The frequency adjustment module 400 is used to automatically adjust the sensor sampling frequency according to the abnormality detection result;

[0100] The fault classification module 500 is used to classify the multi-parameter signal using an event pattern recognition algorithm;

[0101] The abnormal adjustment module 600 is used to generate an automatic response strategy according to the fault classification result.

[0102] Example 3

[0103] An embodiment of the present invention is different from the first two embodiments in that:

[0104] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0105] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0106] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0107] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0108] Example 4

[0109] As an embodiment of the present invention, a hydropower station gate detection method is provided. In order to verify the beneficial effects of the present invention, a scientific demonstration is carried out through simulation experiments.

[0110] Verify the advantages of the hydropower station gate detection method of the present invention in terms of detection accuracy, response time, fault identification accuracy, and dynamic adjustment of sampling frequency. By comparing strictly set parameters and processes, the improvement effect of the present invention compared with the traditional method is clearly demonstrated.

[0111] The traditional hydropower station gate detection system adopts a single parameter monitoring solution with a fixed sampling frequency. The system is mainly based on the vibration frequency and amplitude monitoring of the vibration sensor, and collects signals at a fixed sampling frequency of 100Hz. During detection, a single threshold is used for judgment, and the response mode is fixed. The sampling frequency cannot be adjusted dynamically, which makes it difficult to accurately identify various faults in complex environments and the response is delayed.

[0112] The detection method of the present invention adopts multi-parameter fusion monitoring and adaptive sampling adjustment mechanism to collect and fuse multiple data such as vibration, sound waves, position and pressure in real time, and combines the dynamic adjustment of environmental parameters. The event pattern recognition algorithm is used in the detection, and multi-level fault classification is performed through the decision tree model. When the system detects an abnormal signal, the sampling frequency is automatically adjusted. The fault level is divided into initial fault, potential fault and continuous fault, and the graded response mechanism is triggered according to the severity of the fault.

[0113] In a simulated hydropower station environment, experiments were conducted in three states: normal, slightly abnormal, and severely faulty. The conventional detection method was compared with the detection method of the present invention to test the false alarm rate, response time, fault identification accuracy, and sampling frequency adjustment effect.

[0114] Set the water temperature to 20℃, the air pressure to 1 atm, and the water flow rate to 1.5m / s, and create three simulation states (normal, slight abnormality, and serious failure) to test the detection system. Use the same simulation environment and ensure that the environmental parameters are stable to ensure fairness and comparability of the two methods.

[0115] The experimental steps include:

[0116] Step 1: Under normal conditions, use two detection methods to measure the false alarm rate and sampling frequency stability of the system.

[0117] Step 2: Under a slightly abnormal condition, test the response time and fault detection accuracy of the two detection methods.

[0118] Step 3: Under severe fault conditions, measure the fault identification accuracy and system response speed.

[0119] Step 4: Perform 5 tests on each set of data, take the average value, and record the final result. The experimental results are shown in Table 1.

[0120] Table 1 Comparison of experimental results

[0121]

[0122]

[0123] The present invention has achieved significant technical improvement in the detection of hydropower station gates. Through multi-parameter fusion monitoring technology, the system can effectively identify and filter false alarms caused by environmental fluctuations, greatly reduce misjudgments, and improve the stability of detection. Adaptive sampling and adjustment technology enables the system to automatically increase the sampling frequency when abnormal signals are detected, ensuring rapid response, especially in the case of severe fault conditions, it can capture subtle abnormal changes in time, ensuring the immediacy of detection.

[0124] In addition, the event pattern recognition algorithm and multi-level classification method enable the system to accurately classify fault types based on multiple parameters, overcoming the limitation of traditional methods that cannot accurately distinguish different faults and improving the accuracy of fault identification. Dynamic adjustment of sampling frequency optimizes system resources, improves sampling accuracy in high-risk conditions, and reduces frequency in low-risk conditions, balancing resource consumption and detection accuracy.

[0125] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. 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 solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for detecting a gate of a hydropower station, characterized in that: include: Collect environmental parameters, establish a dynamic environmental model, and determine the real-time influencing factors on the gate status; Synchronously collect gate vibration and sound wave data to obtain composite signals; Build short-term and long-term trend models to identify potential abnormal trends; Automatically adjust sensor sampling frequency based on anomaly detection results; Use event pattern recognition algorithm to classify faults of multi-parameter signals; Generate automated response strategies based on fault classification results.

2. The hydropower station gate detection method according to claim 1, characterized in that: Determining the real-time influencing factors on the gate state includes collecting environmental parameter data in real time through an environmental sensor network, including water temperature, air pressure and water flow velocity, and recording the environmental parameters according to timestamps to form a time series data stream for each environmental parameter; Normalize the collected water temperature, air pressure and water flow velocity data to adjust the data of each parameter to a unified numerical range; Pearson correlation coefficient was used to calculate the correlation between each environmental parameter and the change in gate status; The absolute value of the correlation degree of each environmental parameter is used as the initial weight, and the normalization factor is obtained by summing up all the absolute values ​​of the correlations. The final weight value is generated by dividing the absolute value of each parameter by the normalization factor; Multiply the normalized environmental parameter data with the corresponding weights, calculate the weighted value of each parameter, and obtain the relative influence of each parameter on the gate status; The various weighted results are added together to generate a comprehensive impact factor, which represents the overall degree of influence of environmental factors on the gate status.

3. The hydropower station gate detection method according to claim 2, characterized in that: The obtaining of the composite signal includes increasing the sensitivity of the vibration sensor to a preset level and adjusting the frequency bandwidth of the acoustic wave sensor to a preset range when the impact factor value is higher than a preset threshold value; When the environmental impact factor value is lower than a preset threshold, the sensitivity of the vibration sensor is set to be lowered to a preset level and the sound wave frequency bandwidth range is reduced according to a preset range; Based on the adjusted acquisition parameters, start the synchronous acquisition of vibration and sound wave signals; Align the collected vibration data and sound wave data based on the time axis to generate a composite signal; Perform spectrum analysis on the generated composite signal, convert the composite signal in the frequency domain, extract the frequency components and amplitude values ​​of the signal, and determine whether there are abnormal signal characteristics based on the characteristic values ​​of frequency and amplitude. Mark the frequency and amplitude combination that meets the characteristic values ​​of mechanical fatigue or structural looseness as abnormal signals, record the spectrum characteristic values ​​of the composite signal, and mark the timestamp and frequency range of the abnormal characteristic values.

4. The hydropower station gate detection method according to claim 3, characterized in that: The identifying of potential abnormal trends includes arranging the frequency spectrum characteristic values, timestamps and frequency ranges of the recorded abnormal signals in the order of acquisition time and storing them as time series data; On a time scale of seconds, a fixed second-level time window is set, the spectrum characteristic values ​​of the abnormal signal are extracted from each time window, the frequency and amplitude fluctuations in the time window are calculated, and a short-term trend model including the fluctuations in the second-level time window is generated; On the hourly time scale, a fixed hourly time window is set, the spectrum feature values ​​of abnormal signals are extracted from each hourly time window, the cumulative changes of the spectrum feature values ​​of each time window are calculated, and a long-term trend model that reflects the long-term change trend is generated; In the current hourly time window, compare the cumulative change of the abnormal signal in the window with the cumulative change in the previous hourly window, calculate the change in frequency and amplitude, and if the change exceeds the preset threshold, mark the hourly window as an abnormal window and record its characteristic value; When a rapid fluctuation in frequency or amplitude occurs in the second-level time window of the short-term trend model, the fluctuation data is compared with the abnormal window data in the long-term trend model; If a short-term fluctuation occurs within the hourly window marked as abnormal, and the fluctuation amplitude exceeds the preset threshold, the time point is marked as a potential fault signal, and the spectral characteristic value, timestamp and frequency range of the fault signal are recorded.

5. The hydropower station gate detection method according to claim 4, characterized in that: The automatically adjusting the sensor sampling frequency includes adjusting the sensor sampling frequency of the potential fault signal according to a preset rule.

6. The hydropower station gate detection method according to claim 5, characterized in that: The fault classification includes inputting the adjusted composite signal into the event pattern recognition algorithm, comparing the characteristic value combination of each parameter with the fault pattern library through the multi-layer branch conditions of the decision tree, and realizing the classification of the fault types of mechanical, electrical and environmental abnormalities; According to the characteristic parameters, the faults are further classified into initial faults, potential faults and persistent faults, and each fault is marked with a corresponding risk level label.

7. The hydropower station gate detection method according to claim 6, characterized in that: The generation of the automated response strategy includes dividing the gate's health status into three risk levels: low, medium and high, according to the fault classification results, corresponding to the response strategies of data recording and trend analysis, maintenance reminder and early warning adjustment, emergency closure operation and remote control switching, respectively.

8. A system using the hydropower station gate detection method according to any one of claims 1 to 7, characterized in that: include: A data acquisition module (100), a data fusion module (200), an abnormality identification module (300), a frequency adjustment module (400), a fault classification module (500), and an abnormality adjustment module (600); The data acquisition module (100) is used to collect environmental parameters, establish a dynamic environmental model, and determine real-time influencing factors on the gate state; The data fusion module (200) is used to synchronously collect gate vibration and sound wave data to obtain a composite signal; The anomaly identification module (300) is used to construct short-term and long-term trend models to identify potential abnormal trends; The frequency adjustment module (400) is used to automatically adjust the sensor sampling frequency according to the abnormality detection result; The fault classification module (500) is used to perform fault classification on the multi-parameter signal using an event pattern recognition algorithm; The abnormal adjustment module (600) is used to generate an automatic response strategy according to the fault classification result.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the hydropower station gate detection method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the hydropower station gate detection method described in any one of claims 1 to 7 are implemented.

Images