Intelligent early warning method, system, equipment and medium for speed regulator oil pressing device of hydropower station

Abstract

The invention discloses an intelligent early warning method, system, equipment and medium for a speed regulator oil pressing device of a hydropower station, and belongs to the technical field of intelligent early warning of speed regulators of hydropower stations. Calculating the total oil amount of the oil pressing system according to the liquid level data and the cross sectional area of the corresponding oil storage equipment; recording the change data of the total oil amount in the monitoring period to form an oil amount historical curve; comparing the total oil amount change of the current time period with an oil amount historical curve; when the reduction of the total oil amount in the current time period relative to the reference value of the oil amount historical curve exceeds a judgment threshold value, judging that an oil leakage condition exists and generating early warning information; wherein the judgment threshold values are respectively set according to different operation conditions of the speed regulator oil pressing device. According to the invention, the oil leakage condition of the hydropower station governor oil pressing device is accurately judged, the abnormal operation condition is timely identified, leakage and leakage are intelligently detected, and meanwhile, the accuracy and timeliness of early warning are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent early warning technology for hydropower station governors, specifically to a method, system, equipment, and medium for intelligent early warning of the hydraulic oil device of a hydropower station governor. Background Technology

[0002] A turbine governor is a device that controls the speed of a hydroelectric turbine. It automatically adjusts the turbine's speed to ensure optimal operation. In hydroelectric generator sets, the turbine governor plays a crucial role because it directly affects the turbine's output power and efficiency. The hydraulic pressure device is an important component of the turbine governor; it controls the governor's piston movement by regulating oil pressure, thereby achieving turbine speed control.

[0003] Current monitoring of hydropower station governor oil pressure devices primarily relies on manual, periodic checks of instantaneous data such as oil pressure and level, as well as on-site inspections of the equipment's appearance, to determine the presence of any abnormalities. This traditional monitoring method has the following technical problems: First, relying solely on instantaneous data cannot accurately determine the trend of oil volume changes, making it difficult to detect slow oil leaks in a timely manner; second, monitoring of a single physical quantity lacks correlation analysis with equipment operating conditions, easily leading to false alarms or missed alarms; third, manual inspection is inefficient, and untimely detection of emergencies can affect the safe operation of the hydropower station. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is: how to achieve accurate judgment of oil leakage status of the oil pressure device of the governor of a hydropower station, timely identification of abnormal operating conditions, and intelligent detection of leaks by using dynamic comparative analysis based on historical oil volume curves and multi-dimensional data fusion early warning strategy, while improving the accuracy and timeliness of early warning.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for intelligent early warning of a hydropower station governor oil pressurization device, comprising: acquiring liquid level data of the oil collection tank and the oil pressurization tank in the governor oil pressurization device; calculating the total oil volume of the oil pressurization system based on the liquid level data and the cross-sectional area of ​​the corresponding oil storage equipment; recording the change data of the total oil volume within a monitoring period to form an oil volume history curve; comparing the change of the total oil volume in the current time period with the oil volume history curve; when the total oil volume in the current time period decreases by more than a judgment threshold relative to the benchmark value of the oil volume history curve, determining that there is an oil leakage and generating an early warning message; wherein, the judgment threshold is set according to different operating conditions of the governor oil pressurization device.

[0007] As a preferred embodiment of the intelligent early warning method for the hydraulic oil pressurization device of a hydropower station speed governor according to the present invention, the calculation of the total oil volume of the hydraulic oil pressurization system includes: obtaining the liquid level value and the cross-sectional area value of each oil storage device; calculating the volume of oil in each oil storage device based on the liquid level value and the cross-sectional area value; and determining the total oil volume based on the volume of oil in each oil storage device. The beneficial effect of this preferred technical solution is that by refining the total oil volume calculation process into three specific steps—obtaining the liquid level value and cross-sectional area value, calculating the oil volume of each oil storage device, and determining the total oil volume—it achieves accurate quantitative calculation of the oil volume of the hydraulic oil pressurization system. Compared with traditional empirical estimation or single-parameter monitoring methods, it can perform scientific calculations based on accurate geometric parameters and real-time liquid level data, effectively eliminating the impact of human estimation errors and equipment differences on monitoring accuracy. Simultaneously, the step-by-step calculation method facilitates independent monitoring and fault location of each oil storage device. When an anomaly occurs, it can quickly identify the specific problematic device, improving fault diagnosis efficiency and providing a reliable data foundation for subsequent oil leak judgment and early warning.

[0008] As a preferred embodiment of the intelligent early warning method for the hydraulic oil pressure device of a hydropower station governor according to the present invention, the baseline value of the historical oil volume curve is the statistical value of the total oil volume within the historical monitoring period; the different operating conditions of the hydraulic oil pressure device of the governor include at least one of the following: oil pump operating state, air replenishment valve operating state, and valve displacement state; comparing the change in the total oil volume in the current time period with the historical oil volume curve includes: obtaining the total oil volume within the current monitoring period; obtaining the total oil volume within the historical monitoring period; calculating the decrease ratio of the total oil volume within the current monitoring period relative to the total oil volume within the historical monitoring period, and determining whether it exceeds the judgment threshold. The beneficial effect of this preferred technical solution is that by clearly defining the baseline value of the historical oil volume curve as the statistical value of the total oil volume within the historical monitoring period, and by specifying the detailed implementation of the comparative analysis, a scientific and reasonable anomaly judgment standard is established, overcoming the deficiency of traditional monitoring methods in lacking historical baseline comparison. It can establish personalized judgment standards based on the actual operating history of the equipment, effectively avoiding misjudgment problems caused by equipment differences, seasonal changes, and other factors. Meanwhile, by taking different operating conditions into account, the invention enables differentiated setting of judgment thresholds, allowing it to adapt to the characteristics of equipment in different states such as start-up, shutdown, maintenance, and normal operation. This improves the accuracy and adaptability of oil leak detection and reduces false alarms and missed alarms.

[0009] As a preferred embodiment of the intelligent early warning method for the hydraulic oil pressurization device of a hydropower station according to the present invention, the early warning information includes abnormal operating condition early warning information generated based on the operating condition data of the hydraulic oil pressurization device; the abnormal operating condition early warning information is generated by acquiring the operating condition data and analyzing the operating condition data according to a preset early warning logic.

[0010] As a preferred embodiment of the intelligent early warning method for the hydraulic oil pressurization device of a hydropower station speed governor according to the present invention, the early warning information includes leakage warning information generated based on image data of the hydraulic oil pressurization device; the generation of the leakage warning information includes: acquiring image data of the oil collection tank, hydraulic oil tank, valve group, and oil pipeline; detecting oil stain areas after preprocessing the image data; and generating the leakage warning information when the oil stain areas are detected. The beneficial effect of this preferred technical solution is that it can directly detect visually visible abnormalities such as oil stains and leaks on the equipment surface and surrounding environment, compensating for the lag and limitations that may exist in judging solely based on changes in oil volume. For minor leaks and intermittent drips that will not significantly affect the total oil volume in the short term but pose safety hazards, image monitoring can achieve early detection and early warning. Furthermore, through all-weather automatic image acquisition and intelligent recognition, it replaces the traditional manual inspection method, not only improving the detection coverage and continuity but also eliminating omissions and subjective judgment errors that may exist in manual inspection, significantly improving the reliability and intelligence level of equipment safety monitoring.

[0011] As a preferred embodiment of the intelligent early warning method for the hydraulic oil pressurization device of a hydropower station speed governor according to the present invention, the operating condition data includes at least one of the following: oil pump running time, air replenishment valve operation status, and valve displacement data; the preset early warning logic includes at least one of the following: when the oil pump running time exceeds a set multiple of the historical oil pump running time statistical value or exceeds a set duration, an abnormal pump operation early warning is generated; when the oil level in the pressure tank is higher than a set oil level value and the air replenishment valve operation status is continuously active for a set duration, an air replenishment valve leakage early warning is generated; when the air replenishment interval is less than a set interval, an abnormal air replenishment frequency early warning is generated; when the oil leak pump start interval is less than a set start interval, a frequent oil leak pump start early warning is generated.

[0012] As a preferred embodiment of the intelligent early warning method for the oil pressure device of the governor of a hydropower station according to the present invention, the preprocessing of the image data includes opening the morphological convolution kernel constructed based on the oil stain diffusion pattern, and using wavelet transform to separate the high-frequency noise components.

[0013] The oil stain detection area is obtained by thresholding the preprocessed image data in the HSV color space.

[0014] To address the aforementioned technical problems, the present invention further provides the following technical solution: A system for intelligent early warning of a hydropower station governor oil pressurization device includes: a data acquisition module for acquiring liquid level data of the oil collection tank and oil pressurization tank in the governor oil pressurization device; a calculation module for calculating the total oil volume of the oil pressurization system based on the liquid level data and the cross-sectional area of ​​the corresponding oil storage equipment; a recording module for recording the change data of the total oil volume within a monitoring period to form a historical oil volume curve; a comparison module for comparing the change in the total oil volume in the current time period with the historical oil volume curve; and a judgment module for determining that there is an oil leak and generating an early warning message when the total oil volume in the current time period decreases more than a judgment threshold relative to the baseline value of the historical oil volume curve; wherein, the judgment threshold is set according to different operating conditions of the governor oil pressurization device.

[0015] The present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, characterized in that the processor executes the computer program to implement the steps of the method for intelligent early warning of the hydraulic oil pressure device of a hydropower station governor.

[0016] The present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of the method for intelligent early warning of a hydropower station governor oil pressure device.

[0017] The beneficial effects of the present invention are as follows: (1) The present invention forms a historical oil volume curve by recording the change data of the total oil volume within the monitoring period, and dynamically compares and analyzes the change of the total oil volume in the current time period with the historical curve. When the total oil volume decreases more than the judgment threshold relative to the benchmark value, the oil leakage situation is accurately judged. This overcomes the shortcomings of the prior art which relies solely on instantaneous data monitoring, and can detect slow oil leakage in a timely manner, significantly improving the accuracy and timeliness of oil leakage detection, and effectively avoiding equipment failures and safety accidents caused by oil leakage.

[0018] (2) This invention achieves comprehensive intelligent monitoring of the governor's oil pressure device through a multi-dimensional early warning strategy that integrates operational condition data and image data. By analyzing operational condition data, it identifies operational problems such as abnormal oil pump operation, air leakage from the air supply valve, and abnormal air supply frequency. Image data processing utilizes morphological convolution kernel opening operations and wavelet transform preprocessing, followed by threshold segmentation in the HSV color space to accurately detect oil stain areas, achieving intelligent identification of leaks. Compared to single monitoring methods, this invention significantly improves the comprehensiveness and reliability of anomaly detection, reducing false alarms and missed alarms.

[0019] (3) This invention sets judgment thresholds according to different operating conditions of the governor's oil pressurization device (oil pump operating state, air replenishment valve operating state, valve displacement state), realizing flexible adjustment of the early warning logic. Compared with the traditional fixed threshold method, it can more accurately adapt to the characteristic changes of the equipment under different operating conditions, improve the accuracy of early warning, effectively reduce the workload of manual monitoring, and promote the transformation of hydropower stations towards intelligent operation and maintenance mode. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used 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 those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is an overall flowchart of a method for intelligent early warning of a hydropower station governor oil pressure device according to an embodiment of the present invention;

[0022] Figure 2 This is an overall schematic diagram of a method for intelligent early warning of a hydropower station governor oil pressure device according to an embodiment of the present invention;

[0023] Figure 3 A schematic diagram of a local data acquisition server for an intelligent early warning method for a hydropower station governor oil pressure device, provided in one embodiment of the present invention;

[0024] Figure 4 A schematic diagram of the remote data terminal server principle of a method for intelligent early warning of hydraulic oil device of governor in hydropower station provided in an embodiment of the present invention;

[0025] Figure 5 This is a schematic diagram of the oil pressure device principle of a method for intelligent early warning of a hydropower station governor oil pressure device according to an embodiment of the present invention. Detailed Implementation

[0026] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0027] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a method for intelligent early warning of the hydraulic oil pressure device of a hydropower station governor, comprising:

[0028] Step S100: Obtain the liquid level data of the oil collection tank and oil pressure tank in the governor oil pressure device.

[0029] In one optional embodiment, the liquid level data of the oil collection tank and the oil pressure tank in the governor oil pressure device can be collected by a liquid level sensor.

[0030] The liquid level sensor can be of various types, such as float type, capacitive type, ultrasonic type, or radar type. The appropriate sensor type is selected based on the site environment and accuracy requirements. The liquid level data acquisition frequency can be set according to actual monitoring needs, including but not limited to once per second, once per minute, or once per hour. In this embodiment, a frequency of once per minute is used. The acquired liquid level data is transmitted to the data processing module in real time through the data acquisition system to ensure the real-time performance and accuracy of the data.

[0031] Step S200: Calculate the total oil volume of the oil pressure system based on the liquid level data and the cross-sectional area of ​​the corresponding oil storage equipment.

[0032] It should be noted that the volume is calculated based on the liquid level data obtained in step S100, combined with the cross-sectional area parameters of each oil storage device.

[0033] Cross-sectional area can be obtained through on-site measurement, calculation from design drawings, or 3D modeling. For example, for oil storage devices with regular geometric shapes, standard geometric formulas are used for calculation; for oil storage devices with irregular shapes, segmented calculation or table lookup methods can be used to determine the volume. The calculation process also needs to consider the influence of factors such as the tilt angle of the oil storage device and pipeline connections on the total oil volume. In this embodiment, the total oil volume of the pressure oil system is obtained by multiplying the liquid level value of each oil storage device by its corresponding cross-sectional area and summing the results.

[0034] Step S300: Record the changes in total oil volume during the monitoring period to form a historical oil volume curve.

[0035] It should be noted that the total oil volume calculated in step S200 is recorded and stored in a time series.

[0036] The monitoring cycle can be set according to actual operational needs, including but not limited to hourly, daily, weekly, or monthly monitoring. In this embodiment, a 24-hour monitoring cycle is used. By continuously recording data from multiple monitoring cycles, a historical oil volume curve is generated using data fitting technology. The historical curve can take different forms such as linear fitting, polynomial fitting, or spline fitting, and the optimal fitting method is selected based on the data characteristics. Under normal operating conditions, the curve should show a relatively stable trend and be able to intuitively reflect the change pattern of oil volume in the hydraulic oil system.

[0037] Step S400: Compare the change in total oil volume in the current time period with the historical oil volume curve.

[0038] It should be noted that extracting the total oil volume within the current monitoring period and comparing it with the established historical oil volume curve can employ various methods, including numerical comparison, trend analysis, and statistical testing. Through comparative analysis, the degree of deviation and trend of the current total oil volume relative to the historical benchmark can be identified. During the comparison process, the impact of changes in normal operating conditions on the oil volume needs to be excluded, such as normal fluctuations caused by equipment maintenance, oil replenishment, and other operations. In this embodiment, the relative deviation between the current total oil volume and the historical benchmark value is calculated to provide a quantitative basis for subsequent anomaly judgment.

[0039] Step S500: When the total oil volume in the current time period decreases by more than the judgment threshold relative to the baseline value of the historical oil volume curve, it is determined that there is an oil leak and an early warning message is generated.

[0040] It should be noted that the benchmark value can be a statistical quantity such as the average, median, or weighted average of the total oil volume over a historical monitoring period. The judgment threshold is set according to different operating conditions of the governor's oil pressure device, including but not limited to normal operation, maintenance, and emergency conditions. Different value ranges can be set for the judgment threshold under different operating conditions; for example, 3%-10% for normal operation, 5%-15% for maintenance, and 1%-5% for emergency conditions. When the detected oil volume decrease exceeds the corresponding judgment threshold, an oil leak is automatically identified, and a warning message is immediately generated. The warning message can include various forms such as audible and visual alarms, SMS notifications, and system log records to ensure that relevant personnel can be promptly informed of the abnormality and take appropriate measures.

[0041] Through the above steps S100 to S500, the present invention can realize intelligent monitoring and early warning of the operating status of the oil pressure device of the governor of the hydropower station, timely detect potential oil leakage hazards, ensure the safe and stable operation of the equipment, effectively reduce the workload of manual inspection, and thus improve the intelligent management level of the hydropower station.

[0042] Example 2, refer to Figures 1 to 5 The second embodiment of the present invention provides a method for intelligent early warning of the oil pressure device of the governor of a hydropower station, which is achieved through the collaborative work of system components such as a local data acquisition server and a remote data terminal server.

[0043] Step S100: Obtain the liquid level data of the oil collection tank and oil pressure tank in the governor oil pressure device.

[0044] It should be noted that the oil level data of the oil collection tank and pressure tank are collected through the liquid level sensor in the local data acquisition server. The local data acquisition server includes a local data acquisition unit and a local power management unit. The local data acquisition unit includes a pressure sensor, a liquid level sensor, a temperature sensor, and a displacement sensor.

[0045] In one optional embodiment, liquid level data can be acquired through various methods such as direct measurement, indirect measurement, and combined measurement.

[0046] (1) Direct measurement. Float-type liquid level sensors detect liquid level by changing the position of the float; capacitive liquid level sensors measure liquid level by changing the dielectric constant; ultrasonic liquid level sensors calculate liquid level by the round-trip time of ultrasonic waves; and radar liquid level sensors measure liquid level based on the principle of microwave reflection.

[0047] (2) Indirect measurement. The hydrostatic pressure at the bottom of the oil storage device is measured by a pressure sensor, and the liquid level is indirectly calculated based on the principle of liquid gravity.

[0048] (3) Composite measurement. Two or more sensors based on different principles are used simultaneously, and the measurement reliability is improved through data comparison and verification. In this embodiment, a liquid level sensor is used for direct measurement, which can reflect the changes in oil level in the oil storage equipment in real time.

[0049] It should be noted that the local power management unit provides power to the local data acquisition unit, and the local power management unit operates in a dual-power mode to prevent data acquisition failure due to a single power source failure. The dual-power mode typically includes AC mains power as the primary power source and a UPS power source as a backup power source. When the primary power supply is interrupted, an automatic switching device switches to the backup power source within milliseconds to ensure the continuity of data acquisition. The level sensor converts the detected physical quantity into a digital signal via an analog-to-digital converter. After filtering and calibration, the digital signal is sent to the local data acquisition unit. The data acquisition frequency can be set to continuous or timed acquisition mode as required.

[0050] Step S200: Calculate the total oil volume of the oil pressure system based on the liquid level data and the cross-sectional area of ​​the corresponding oil storage equipment.

[0051] It should be noted that the on-site data acquisition unit transmits liquid level data to the remote data terminal server via an isolation device. This isolation device is a forward isolation device, employing physical isolation technology to allow only unidirectional data transmission from the on-site side to the remote side, blocking any signal transmission from the remote side to the on-site side. This prevents network security risks from the remote data terminal server from affecting the on-site production system and ensures the security of hydropower station production information. Data transmission uses standard industrial communication protocols to ensure data integrity and accuracy.

[0052] Furthermore, the remote data terminal server includes a data analysis module, a data query module, an image recognition module, and an intelligent early warning module. The data analysis module is responsible for receiving and processing various monitoring data from the local area. The governor's hydraulic system includes an oil collection tank, a pressure oil tank, an oil pump, a valve group, and oil pipelines. The oil pumps include an oil collection pump for the oil collection tank and a pressure oil pump for the pressure oil tank. The pressure oil tank delivers pressurized oil to the turbine governor's hydraulic mechanism via the pressure oil pump and oil pipelines, providing the necessary hydraulic power support for the various actions of the governor's hydraulic mechanism.

[0053] In this embodiment, calculating the total oil volume of the hydraulic system includes:

[0054] Step A1: Obtain the liquid level value and cross-sectional area value of each oil storage device;

[0055] Step A2: Calculate the volume of oil in each oil storage device based on the liquid level and cross-sectional area.

[0056] Step A3: Determine the total amount of oil based on the volume of oil in each oil storage device.

[0057] Furthermore, the calculation of the total oil volume needs to consider the actual geometry of the oil storage equipment. For oil storage equipment with regular geometric shapes, such as cylindrical pressure tanks, the volume calculation uses standard geometric formulas; for oil storage equipment with irregular shapes, such as oil collection tanks with conical bottoms, it is necessary to calculate the volume of each section and then sum them. Cross-sectional area data usually comes from the equipment design drawings or actual on-site measurements. During the calculation process, the influence of factors such as the installation tilt of the oil storage equipment and the space occupied by internal components on the effective volume also needs to be considered. The data analysis module uses pre-stored equipment parameters and real-time liquid level data to obtain the actual oil volume of each oil storage device through a volume calculation algorithm, and finally summarizes the total oil volume of the pressure oil system.

[0058] For example, in the oil pressure device of a hydropower station's governor, the oil collection tank is a cuboid structure with an effective length of 3.2 meters and a width of 1.8 meters. When the level sensor detects a liquid level of 1.6 meters, the volume of oil in the collection tank is 3.2 × 1.8 × 1.6 = 9.216 cubic meters. The pressure tank is a cylindrical structure with an inner diameter of 2.4 meters. When the level sensor detects a liquid level of 2.2 meters, the volume of oil in the pressure tank is π × (1.2). 2 ×2.2=9.952 cubic meters; therefore, the total amount of oil calculated is 9.216+9.952=19.168 cubic meters.

[0059] Furthermore, the hydraulic pressurization system in a hydropower station performs three functions: lubrication, heat dissipation, and hydraulic operation. Under normal operating conditions, the hydraulic pressurization system primarily functions as a lubricant and heat dissipator. Its working principle is as follows: oil from the pressurization tank enters each bearing system through oil pumps and pipelines, lubricating the bearings while carrying away the heat generated during equipment operation to prevent overheating. The heated oil then flows back to the oil collection tank for cooling. When the oil level in the pressurization tank drops to a set low threshold, the cooling oil in the oil collection tank is pumped back into the pressurization tank by the oil pump, forming a continuous lubrication and heat dissipation circulation system.

[0060] It should be noted that the hydraulic system also has an important hydraulic operation function, used to deliver high-pressure oil to the turbine governor via a dedicated oil pump device to control the opening and closing of the guide vanes. During the guide vane control process, a large amount of oil is required for hydraulic drive, resulting in significant fluctuations in the liquid level. However, the operation time is relatively short, and frequent start-stop operations are not required. After the operation is completed, the liquid level returns to the normal state within a certain period of time.

[0061] For example, when the unit is operating normally and no start-up or shutdown operations are performed, the oil in the hydraulic pressurization system is mainly distributed in the hydraulic oil tank, the oil collection tank, and the connected oil pipelines. Due to the lubrication and heat dissipation circulation, the liquid levels in the hydraulic oil tank and the oil collection tank exhibit dynamic changes, with the change curves resembling waves. However, there is an oil volume balance relationship between the two; that is, when the liquid level in one container rises, the liquid level in the other container correspondingly decreases. Ignoring the oil volume in the oil pipelines, the total oil volume of the hydraulic pressurization system, V = L1 × S1 + L2 × S2, remains basically constant under normal conditions, and its historical oil volume curve presents an approximately horizontal straight line. When this straight line shows a continuous downward trend, it indicates that oil loss has occurred, i.e., an oil leak has occurred. At this time, by monitoring the change trend of the total oil volume, potential oil leak faults can be detected and warned in a timely manner.

[0062] Step S300: Record the changes in total oil volume during the monitoring period to form a historical oil volume curve.

[0063] It should be noted that the data query module in the remote data terminal server is responsible for data storage, management, and display. The data query module includes two sub-modules: intelligent report display and curve data display. Intelligent report display arranges and statistically analyzes monitoring data in tabular form according to time series, making it easy for maintenance personnel to view specific values ​​and trends. Curve data display plots total oil volume data as a continuous curve along a time axis, intuitively reflecting the changing patterns and abnormal fluctuations in oil volume. Data storage uses a time-series database, supporting high-frequency data writing and fast querying, providing a technical foundation for historical data analysis.

[0064] In one alternative embodiment, the generation of the historical oil volume curve can employ different data processing methods:

[0065] (1) Original data connection method: directly connect the total oil volume values ​​at each time point with a straight line to maintain the original characteristics of the data;

[0066] (2) Moving average method: The average value of data within a certain time window is used to eliminate the impact of short-term fluctuations on the overall trend;

[0067] (3) Spline interpolation: A smooth curve is used to connect the data points, making the historical curve more continuous and aesthetically pleasing;

[0068] (4) Least square fitting method: Using mathematical fitting methods to find the curve equation that best matches the trend of data change.

[0069] This embodiment selects an appropriate processing method based on the data characteristics and analysis requirements. It typically uses a combination of moving average and spline interpolation, which can reflect the true trend of change and has a good visual effect.

[0070] It should be noted that the monitoring cycle needs to be set reasonably based on the equipment's operating characteristics and management needs. During normal equipment operation, oil levels are relatively stable, allowing for longer monitoring cycles such as 24 hours or a week. During equipment start-up, shutdown, maintenance, or abnormal operating conditions, oil levels fluctuate more frequently, requiring shorter monitoring cycles such as 1 hour or 30 minutes. For critical equipment or high-risk periods, the monitoring cycle can be further shortened to the minute level. Data storage strategies typically employ a tiered storage approach, maintaining the original accuracy of recent data while compressing older data to save storage space.

[0071] Step S400: Compare the change in total oil volume in the current time period with the historical oil volume curve.

[0072] It should be noted that the data analysis module is the core component of the remote data terminal server responsible for data processing and analysis. Hydropower station operation and maintenance personnel configure corresponding data early warning logic and alarm triggering conditions in the data analysis module based on equipment characteristics, operational experience, and historical fault cases. The comparative analysis process needs to comprehensively consider various influencing factors such as current operating conditions, seasonal factors, and equipment maintenance history to ensure the accuracy and reliability of the comparison results.

[0073] In this embodiment, the change in total oil volume during the current time period is compared with the historical oil volume curve, including:

[0074] Step B1: Obtain the total oil volume within the current monitoring period;

[0075] Step B2: Obtain the total oil volume within the historical monitoring period;

[0076] Step B3: Calculate the percentage decrease in total oil volume in the current monitoring period relative to the total oil volume in historical monitoring periods, and determine whether it exceeds the judgment threshold.

[0077] Furthermore, the implementation of comparative analysis requires the establishment of a robust data processing workflow. First, the total oil volume data for the current monitoring period and the corresponding historical monitoring periods are extracted from the database. During data extraction, validity checks are necessary to remove abnormal data caused by sensor malfunctions, communication interruptions, or other reasons. Next, the difference between the current value and the historical baseline value is calculated, either as an absolute difference or a relative percentage. Finally, the calculation result is compared with a preset threshold. When the difference exceeds the threshold, the corresponding early warning process is triggered. The entire comparison process needs to consider the reasonable fluctuation range under normal operating conditions to avoid misjudging oil volume changes caused by normal operation as abnormalities.

[0078] For example, in a certain monitoring period, the average total oil volume in the current 24-hour cycle is 18.95 cubic meters, while the corresponding historical average total oil volume for the same period is 19.40 cubic meters, resulting in a decrease of (19.40-18.95) / 19.40×100% = 2.32%. If the set judgment threshold is 3%, this decrease is considered normal fluctuation as it does not exceed the threshold. If the judgment threshold is set to 2%, this decrease exceeds the threshold, triggering further analysis and early warning processes.

[0079] It is important to note that when performing total oil volume comparison analysis, it is necessary to distinguish between normal hydraulic operations and abnormal oil leaks. When a significant change in oil level is detected within a short period of time, the system will automatically check for hydraulic operation records such as guide vane control. If such records exist, the data for that period will be marked as normal operation data and will not be included in the analysis for oil leak detection. If no operation records exist, further analysis will be conducted to determine whether there is a gradual decrease in oil volume, thereby accurately identifying whether a genuine oil leak is present.

[0080] Step S500: When the total oil volume in the current time period decreases by more than the judgment threshold relative to the baseline value of the historical oil volume curve, it is determined that there is an oil leak and an early warning message is generated.

[0081] It should be noted that when the data analysis module detects a decrease in total oil volume exceeding a preset threshold, it determines that an oil leak has occurred and immediately initiates the early warning process. The data analysis module transmits the early warning data containing abnormal information to the intelligent early warning module, which then executes corresponding alarm actions based on the warning level and type. The intelligent early warning module comprises three functional components: an audible and visual alarm module, an SMS sending module, and an early warning query module. The audible and visual alarm module provides real-time alarms via on-site sound and light signal devices, immediately attracting the attention of on-duty personnel. The SMS sending module sends early warning information to preset management personnel's mobile phones via mobile communication networks, ensuring that even if management personnel are not on-site, they can be promptly informed of abnormal situations. The early warning query module provides a network-based early warning information management interface, supporting the querying, confirmation, processing, and historical record management of early warning information.

[0082] The judgment threshold is set separately according to different operating conditions of the governor's oil pressure device. These different operating conditions include at least one of the following: oil pump operation status, air supply valve operation status, and valve displacement status. The characteristics of oil quantity changes in the equipment differ significantly under different operating conditions; therefore, an adaptive threshold setting strategy is needed to improve the accuracy of early warnings and reduce false alarms.

[0083] It should be noted that the baseline value of the historical oil volume curve is the statistical value of the total oil volume within the historical monitoring period. The baseline value is typically calculated using statistical methods, such as the arithmetic mean, weighted average, or median. The specific method chosen depends on the distribution characteristics and requirements of the historical data. To improve the representativeness of the baseline value, historical data from the same operating conditions and season are usually selected for statistical analysis to exclude the influence of data from abnormal operating conditions and periods of failure.

[0084] In one optional embodiment, the warning information includes abnormal operating condition warning information generated based on the operating condition data of the governor oil pressurization device.

[0085] The abnormal operating condition early warning information is generated by acquiring operating condition data and analyzing it according to preset early warning logic. The acquisition of operating condition data relies on various sensors in the local data acquisition unit: pressure sensors use piezoresistive or capacitive principles to measure the oil pressure inside the oil collection tank and pressure tank; level sensors measure the oil level in the oil storage equipment; temperature sensors use resistance temperature detectors or thermocouples to monitor changes in oil temperature; and displacement sensors use inductive or photoelectric principles to detect the opening and displacement status of various valves in the valve group. The output signals of all sensors are processed by signal conditioning circuits and converted into a standard electrical signal format before being sent to the local data acquisition unit through the data acquisition system.

[0086] It should be noted that the operating condition data includes at least one of the following: oil pump running time, air replenishment valve operation status, and valve displacement data. This operating condition data comprehensively reflects the operating status and performance of the governor's hydraulic pressurization device. Oil pump running time is obtained through motor operating status signals, reflecting the system load; air replenishment valve operation status is obtained through valve position feedback signals, reflecting the system's airtightness; and valve displacement data is obtained through displacement sensors, reflecting the working status of each actuator.

[0087] The valve assembly, as a crucial component of the governor's hydraulic pressurization system, includes various types of valves such as safety valves, load relief valves, and check valves, each with specific functions. The safety valve automatically opens to release pressure when the system pressure exceeds a set value, preventing damage to the hydraulic oil tank and related pipelines due to overpressure and protecting the entire hydraulic pressurization system. The load relief valve reduces the system load when the oil pump starts, allowing the motor to start smoothly under lower load and reducing the impact of starting current on the power grid. The check valve prevents high-pressure oil in the hydraulic oil tank from flowing back into the oil collection tank through pipelines when the oil pump stops running, maintaining stable system pressure.

[0088] The preset warning logic includes at least one of the following:

[0089] When the oil pump's running time exceeds a set multiple of the historical oil pump running time statistics or exceeds a set duration, a pump operation abnormality warning is generated.

[0090] When the oil level in the pressure tank is higher than the set oil level value and the air supply valve is continuously activated for more than the set duration, an air supply valve leakage warning is generated.

[0091] When the gas replenishment interval is less than the set interval, an abnormal gas replenishment frequency warning is generated.

[0092] When the oil leak pump start interval is less than the set start interval, a frequent oil leak pump start warning is generated.

[0093] These early warning logics are set based on actual operating experience and equipment characteristics of hydropower stations. Abnormal oil pump operating time usually indicates a leak or other malfunction requiring frequent oil replenishment; continuous operation of the air replenishment valve indicates an airtightness problem; excessively high air replenishment frequency indicates a leak in the air circuit; and excessively frequent start-up of the oil pump indicates a significant oil leak.

[0094] For example, in the actual operation of a hydropower station, historical data shows that the normal operating time of the oil pump is 25-35 hours per month. When the cumulative operating time of the oil pump in a given month reaches 52 hours, exceeding the upper limit of the historical statistical value by 1.5 times, an abnormal pump operation warning is automatically generated, reminding maintenance personnel to check for leaks or other abnormalities. Similarly, when the oil level in the pressure tank reaches the high-level limit and the air supply valve operates continuously for more than 10 minutes, it is determined that the air supply valve may have a poor seal leading to internal leakage, generating an air supply valve leakage warning.

[0095] In this embodiment of the invention, multiple on-site monitoring cameras are installed above the oil collection tank, pressure tank, valve group, and oil pipeline to monitor oil spills, leaks, drips, and other oil-related events. This part utilizes existing image recognition technologies to pre-train models of oil spills, leaks, drips, and other oil-related events in the oil pipeline. Then, on-site photos are collected, and image recognition is performed. When an oil spill or other event is detected, the result can be communicated to the user via image and alarm. The transmission method uses a lightweight MQTT protocol to establish an image data transmission channel, dividing the image data into 512×512 pixel segments. UDP protocol is used to accelerate transmission, and each data packet is appended with a CRC-32 checksum. Retransmission is triggered when the packet loss rate exceeds 5%.

[0096] In another optional embodiment, the warning information includes leakage warning information generated based on image data of the governor oil pressure device. Specifically, image data of the oil collection tank, oil pressure tank, valve group, and oil pipeline are acquired, and oil stain areas are detected after preprocessing the image data. When an oil stain area is detected, leakage warning information is generated.

[0097] Image data acquisition is achieved through a video surveillance acquisition server, which includes multiple local monitoring cameras and corresponding image processing equipment. The local monitoring cameras utilize industrial-grade CCD or CMOS image sensors, possessing waterproof, dustproof, and vibration-resistant characteristics, enabling them to withstand the harsh environmental conditions of hydropower stations. The camera installation locations are carefully designed to ensure coverage of all potential oil leakage areas of critical equipment such as oil collection tanks, pressure tanks, valve groups, and oil pipelines, while avoiding the impact of lighting conditions and viewing angle obstructions on image quality. Image data is transmitted via network to a remote data terminal server. Compression algorithms are used during transmission to save bandwidth while ensuring image quality meets analysis requirements.

[0098] The image recognition module in the remote data terminal server is responsible for automatically analyzing and processing image data. Employing computer vision technology, the module can automatically identify oil stains, contamination, and abnormal areas in images. When suspected abnormalities such as oil spills, leaks, or drips are detected, the relevant information is immediately transmitted to the intelligent early warning module, triggering the corresponding alarm process to ensure that maintenance personnel can promptly detect and address the problem.

[0099] It should be noted that the image data preprocessing includes opening operations on morphological convolution kernels constructed based on the oil stain diffusion pattern, and using wavelet transform to separate high-frequency noise components.

[0100] Morphological opening operations, through a process of erosion followed by dilation, effectively remove small-area noise points from images while preserving the main shape features of the oil stain area. The design of the convolution kernel needs to be optimized based on the actual diffusion morphology of the oil stain; typically, a circular or elliptical kernel structure is used, with the kernel size determined according to the expected oil stain size range. Wavelet transform techniques are used for frequency domain analysis of the image. By separating the high-frequency and low-frequency components of the image, they can effectively suppress image noise, improve the signal-to-noise ratio, and provide a better image quality foundation for subsequent feature extraction and recognition.

[0101] Detecting oil stains involves thresholding the pre-processed image data in the HSV color space. The HSV color space is more suitable for color feature analysis than the RGB color space, where H represents hue, S represents saturation, and V represents lightness. Oil stains typically possess specific color characteristics, manifesting as a specific hue range in the HSV space. Thresholding segmentation extracts pixel regions in the image that match the color characteristics of oil stains by setting threshold ranges for the H, S, and V channels, forming a binarized target region. The threshold parameters need to be adjusted and optimized based on factors such as ambient lighting conditions and the type of oil.

[0102] For example, in an image monitoring application at a hydropower station, images are captured by a camera installed 2 meters above the valve assembly. The image resolution is 1920×1080 pixels, and the acquisition frequency is one image per minute. When an irregular dark area is detected on the ground near the valve assembly, and the hue value of this area in the HSV color space is within a specific range, but the area exceeds a preset minimum threshold, it is judged as a possible oil leak. An early warning message for leaks is automatically generated, and an image containing the abnormal area is pushed to the operation and maintenance personnel for confirmation and handling.

[0103] For example, the image preprocessing process is as follows:

[0104] Morphological convolution kernels were constructed based on the oil stain diffusion pattern, and isolated noise points were eliminated through opening operations. Multi-scale edge detection was performed on the preprocessed image, and dynamic oil stain regions were segmented using HSV color space thresholding. Wavelet transform was used to decompose the high-frequency components of the image to suppress reflective noise on the metal surface.

[0105] The wavelet transform decomposition of the high-frequency components of the image includes performing a 3-level Haar wavelet transform on the image to separate the high-frequency components (noise) from the low-frequency components (oil stain areas).

[0106] In summary, through the coordinated operation of the aforementioned components, this invention achieves comprehensive, continuous, and intelligent monitoring and early warning of the operating status of the hydraulic pressure device of the hydropower station governor. It promptly detects potential equipment failures and safety hazards, reduces the monitoring workload of maintenance personnel, decreases the intensity and frequency of manual inspections, improves the timeliness and accuracy of fault detection, and promotes the transformation of hydropower stations towards a modern operation mode with fewer or even no staff. This invention can significantly improve the efficiency, safety, and intelligence level of hydropower station operation and management, thereby ensuring the safe and stable operation of power production.

[0107] Example 3, the third embodiment of the present invention, provides a system for intelligent early warning of the hydraulic oil pressure device of a hydropower station governor, comprising:

[0108] The data acquisition module is used to acquire the liquid level data of the oil collection tank and oil pressure tank in the governor oil pressure device;

[0109] The calculation module is used to calculate the total amount of oil in the oil pressure system based on the liquid level data and the cross-sectional area of ​​the corresponding oil storage equipment.

[0110] The recording module is used to record the changes in total oil volume during the monitoring period and generate a historical oil volume curve.

[0111] The comparison module is used to compare the change in total oil volume in the current time period with the historical oil volume curve;

[0112] The judgment module is used to determine that there is an oil leak and generate a warning message when the total oil volume in the current time period decreases more than the benchmark value of the historical oil volume curve.

[0113] The judgment threshold is set according to the different operating conditions of the governor oil pressure device.

[0114] Example 4, the fourth embodiment of the present invention, differs from the previous three embodiments in that: if the function is implemented as a software functional unit and sold or used as an independent product, it 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 a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0115] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0116] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0117] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0118] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for intelligent early warning of hydraulic oil pressure device in a hydropower station governor, characterized in that: include: Obtain the liquid level data of the oil collection tank and oil pressure tank in the governor oil pressure device; Calculate the total oil volume of the oil pressure system based on the liquid level data and the cross-sectional area of ​​the corresponding oil storage device. Record the changes in the total oil volume during the monitoring period to form a historical oil volume curve; Compare the change in total oil volume in the current time period with the historical oil volume curve; When the total oil volume in the current time period decreases by more than the threshold value relative to the historical oil volume curve, an oil leak is detected and an early warning message is generated. The determination threshold is set according to different operating conditions of the governor oil pressurization device.

2. The method for intelligent early warning of the hydraulic oil device of a hydropower station governor as described in claim 1, characterized in that: The calculation of the total oil volume of the hydraulic system includes: Obtain the liquid level value and cross-sectional area value of each oil storage device; The volume of oil in each oil storage device is calculated based on the liquid level value and the cross-sectional area value; The total amount of oil is determined based on the volume of oil in each oil storage device.

3. The method for intelligent early warning of the hydraulic oil pressure device of a hydropower station governor as described in claim 2, characterized in that: The baseline value of the historical oil volume curve is the statistical value of the total oil volume within the historical monitoring period; The different operating conditions of the governor oil pressure device include at least one of the following: oil pump operating state, air replenishment valve operating state, and valve displacement state. The step of comparing the change in total oil volume in the current time period with the historical oil volume curve includes: Obtain the total oil volume within the current monitoring period; Obtain the total oil volume within the historical monitoring period; Calculate the percentage decrease in total oil volume during the current monitoring period relative to the total oil volume during the historical monitoring periods, and determine whether it exceeds the judgment threshold.

4. The method for intelligent early warning of the hydraulic oil pressure device of a hydropower station governor as described in claim 3, characterized in that: The early warning information includes abnormal operating condition early warning information generated based on the operating condition data of the governor oil pressure device; The abnormal operating condition warning information is generated by acquiring the operating condition data and analyzing the operating condition data according to a preset warning logic.

5. The method for intelligent early warning of the hydraulic oil pressure device of a hydropower station governor as described in claim 4, characterized in that: The early warning information includes leakage warning information generated based on image data of the governor's oil pressure device; The generation of the leakage warning information includes: Acquire image data of the oil collection tank, pressure tank, valve group, and oil pipeline; After preprocessing the image data, oil stain areas are detected. When the oil stain areas are detected, the leak warning information is generated.

6. The method for intelligent early warning of the hydraulic oil pressure device of a hydropower station governor as described in claim 4, characterized in that: The operating condition data includes at least one of the following: oil pump running time, air replenishment valve operation status, and valve displacement data. The preset warning logic includes at least one of the following: When the running time of the oil pump exceeds a set multiple of the historical oil pump running time statistics or exceeds a set duration, a pump operation abnormality warning is generated. When the oil level in the pressure tank is higher than the set oil level value and the air supply valve is continuously activated for a period of time exceeding the set duration, an air supply valve leakage warning is generated. When the gas replenishment interval is less than the set interval, an abnormal gas replenishment frequency warning is generated. When the oil leak pump start interval is less than the set start interval, a frequent oil leak pump start warning is generated.

7. A method for intelligent early warning of hydraulic oil pressure device of hydropower station governor as described in claim 5, characterized in that: The preprocessing of the image data includes opening operations on morphological convolution kernels constructed based on the oil stain diffusion pattern, and separating high-frequency noise components using wavelet transform. The oil stain detection area is obtained by thresholding the preprocessed image data in the HSV color space.

8. A system application of intelligent early warning for the hydraulic oil pressure device of a hydropower station governor, as described in any one of claims 1 to 7, characterized in that... include: The data acquisition module is used to acquire the liquid level data of the oil collection tank and oil pressure tank in the governor oil pressure device; The calculation module is used to calculate the total amount of oil in the oil pressure system based on the liquid level data and the cross-sectional area of ​​the corresponding oil storage device. The recording module is used to record the changes in the total oil volume during the monitoring period and form a historical oil volume curve. The comparison module is used to compare the change in total oil volume in the current time period with the historical oil volume curve. The judgment module is used to determine that there is an oil leak and generate a warning message when the total oil volume in the current time period decreases by more than a judgment threshold relative to the baseline value of the historical oil volume curve. The determination threshold is set according to different operating conditions of the governor oil pressurization device.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the intelligent early warning method for the hydraulic oil pressure device of the hydropower station governor as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the intelligent early warning method for the hydraulic oil device of the governor of a hydropower station as described in any one of claims 1 to 7.

Images