A real-time detection system and early warning method for lubrication system of wind power generation equipment

Through the real-time detection system of the lubrication system of wind power generation equipment, the multi-parameter monitoring and data acquisition modules are used to monitor and early warning in real time, solving the problem of traditional low maintenance efficiency and improving the operating reliability and maintenance efficiency of wind power generation equipment.

CN118653969BActive Publication Date: 2024-11-05OVERSEAS VISION (BEIJING) TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410984811.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-05
Estimated Expiration
2044-07-22

AI Technical Summary

Technical Problem

The maintenance efficiency of traditional wind power equipment lubrication systems is low, and it is impossible to monitor and warn of potential failures in real time, resulting in increased equipment failure rates, increased maintenance costs and potential downtime.

Method used

Design a real-time detection system for lubrication systems for wind power generation equipment, including multi-parameter monitoring module and data acquisition module. By installing oil pressure, oil temperature, oil level, oil quality, vibration and current sensors, data is monitored and collected in real time, and the frequency is used. The shadow value and shadow difference value are optimized for the acquisition frequency, and abnormalities are identified through prediction models and statistical methods, and an early warning is issued automatically.

Benefits of technology

Real-time monitoring and early warning of the lubrication system of wind power generation equipment is realized, monitoring efficiency and accuracy are improved, fault downtime and maintenance costs are reduced, and equipment operation reliability and maintenance efficiency are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118653969B_ABST
    Figure CN118653969B_ABST
Patent Text Reader

Abstract

The invention discloses a real-time detection system and an early warning method for a lubrication system of a wind power generation equipment, and relates to the technical field of equipment detection and early warning. The invention comprises a multi-parameter monitoring module and a data acquisition module; the multi-parameter monitoring module is used for real-time monitoring of different parameters of the lubrication system of the wind power generation equipment; according to the monitoring requirements of the lubrication system of the wind power generation equipment, the parameters to be monitored are determined, and then various sensors are installed at corresponding positions in the lubrication system according to the parameters to be monitored, including; the invention can timely discover potential abnormalities and faults by real-time monitoring of key parameters of the lubrication system of the wind power generation equipment, such as oil pressure, oil temperature, oil level, oil quality, vibration and current. Compared with the traditional periodic inspection method, this real-time monitoring method greatly improves the efficiency and accuracy of monitoring, and reduces the downtime and maintenance costs caused by equipment failure.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of equipment detection and early warning, and in particular to a real-time detection system and an early warning method for a lubrication system of a wind power generation equipment. Background Art

[0002] As a clean and renewable energy, wind power generation plays an increasingly important role in the world's energy structure. However, the operating environment of wind power generation equipment is complex and changeable, and the reliability and maintenance of equipment have become key factors restricting its development. In particular, the lubrication system of wind power generation equipment, as an important part to ensure the normal operation of the equipment, has a direct impact on the stability and safety of the entire power generation system.

[0003] Traditional wind power equipment lubrication systems mostly use regular maintenance and inspection methods, but this method has obvious time lags and cannot monitor the status of the lubrication system in real time, resulting in the inability to detect and deal with potential faults in a timely manner. As the equipment runs longer, the failure rate of the lubrication system gradually increases, which not only increases maintenance costs, but may also cause equipment damage or even downtime accidents, resulting in huge economic losses.

[0004] In order to improve the operational reliability and maintenance efficiency of wind power generation equipment, it is necessary to develop a real-time monitoring and early warning system to monitor the key parameters of the lubrication system in real time, and issue early warnings in time when abnormalities are detected to guide maintenance personnel to carry out necessary inspections and repairs. Although some existing monitoring systems can realize real-time monitoring of some parameters, they often lack systematicity and real-time performance, and have deficiencies in data processing and early warning algorithms, and cannot meet the needs of wind power generation equipment for efficient and accurate maintenance.

[0005] In addition, since most wind power generation equipment is located in remote areas with harsh environmental conditions, higher requirements are placed on the stability and reliability of the monitoring system. Moreover, with the development of technology and the improvement of environmental protection requirements, the market has put forward more stringent standards for the energy efficiency and maintenance costs of wind power generation equipment. Therefore, the development of a real-time detection system for the lubrication system of wind power generation equipment that can adapt to harsh environments, is stable and reliable, and has intelligent early warning functions has become an urgent need for the development of the industry.

[0006] In order to solve the above defects, a technical solution is now provided. Summary of the invention

[0007] The purpose of the present invention is to solve the problems of low maintenance efficiency and inability to monitor and warn of potential faults in the lubrication system of traditional wind power generation equipment, and to propose a real-time detection system and an early warning method for the lubrication system of wind power generation equipment.

[0008] The purpose of the present invention can be achieved by the following technical solutions:

[0009] A real-time detection system and early warning method for a lubrication system of a wind power generation equipment, comprising:

[0010] The multi-parameter monitoring module is used to monitor different parameters of the lubrication system of wind power generation equipment in real time. The specific process is as follows:

[0011] According to the monitoring requirements of the lubrication system of wind power generation equipment, determine the parameters that need to be monitored, and then install various sensors at the corresponding positions of the lubrication system according to the parameters that need to be monitored, including oil pressure sensors, oil temperature sensors, oil level sensors, oil quality sensors, vibration sensors and current sensors;

[0012] Then check the accuracy of the installed sensors. If the check result is abnormal, calibrate or reset the corresponding sensors.

[0013] The data acquisition module is used to collect real-time data of various parameters through different sensors. The specific steps are as follows:

[0014] Select data acquisition hardware and connect the sensor to the data acquisition hardware through the corresponding interface;

[0015] Configure data acquisition software, set data acquisition parameters, including sampling rate and channel configuration, and realize real-time reading, processing and storage of data through programming;

[0016] Start the data acquisition module, start collecting sensor data in real time, monitor the data flow, set the data acquisition frequency, and perform comprehensive analysis based on different sensor frequency judgment parameters to obtain the frequency value;

[0017] Then the obtained frequency shadow value is compared with the preset frequency shadow threshold. When the frequency shadow value is greater than the preset frequency shadow threshold, it is determined that the data acquisition frequency of the sensor needs to be optimized, and the difference between the frequency shadow value and the preset frequency shadow threshold is calculated and recorded as the shadow difference value;

[0018] The obtained shadow difference value is compared with several preset shadow difference value intervals. Several shadow difference value intervals correspond to different percentage frequency optimization levels. When the shadow difference interval to which the shadow difference value belongs is determined, the corresponding percentage frequency optimization level is obtained, and the frequency duration of the corresponding percentage frequency optimization level is increased on the basis of the original acquisition frequency of the sensor.

[0019] Furthermore, the data collected by different sensors in the multi-parameter monitoring module include:

[0020] Oil pressure sensor: monitors the oil pressure in the lubrication system;

[0021] Oil temperature sensor: real-time monitoring of lubricating oil temperature;

[0022] Oil level sensor: monitors the oil level in the tank;

[0023] Oil quality sensor: monitor the quality of lubricating oil, including viscosity and contamination level, through the oil quality sensor;

[0024] Vibration sensor: monitors the vibration of the equipment during operation as an auxiliary judgment of the equipment status;

[0025] Current sensor: monitors the current of lubrication pumps and other electrical equipment to evaluate the working status of the system.

[0026] Furthermore, the specific operation steps of the multi-parameter monitoring module to verify the accuracy of the installed sensors are as follows:

[0027] First, determine the calibration standard for sensor data accuracy, which is obtained by analyzing the equipment manufacturer's specifications and industry standards;

[0028] Verify the output of the sensor using a standard reference device or calibration equipment of known accuracy;

[0029] Then test different sensors, collect data from the sensor and the reference device at the same time, ensure that the data is collected under the same environmental conditions, compare the sensor data with the readings of the reference device, analyze the differences between the two, and evaluate the accuracy of the data by using statistical methods, including mean deviation and standard deviation. The specific process is as follows:

[0030] Collect data from sensors and reference or standard devices simultaneously, measure the same physical quantity, and ensure that the readings of the sensor and reference device are obtained at the same time or under the same conditions;

[0031] For each set of data, the difference between the sensor reading and the reference device reading is calculated, and then all deviations are summed up and divided by the number of measurements to obtain the average deviation. The square of the difference between the deviation of each measurement and the average deviation is calculated to obtain the standard deviation;

[0032] Draw deviation distribution diagrams, including histograms or box plots, to visualize the distribution of data and check whether the data distribution is symmetrical and whether there are outliers;

[0033] When the average deviation is less than the preset value, the sensor readings have no systematic deviation;

[0034] Then calculate the confidence interval of the sensor readings, which is the mean plus or minus two standard deviations;

[0035] Record the results of the deviation analysis, including the mean deviation, standard deviation, deviation distribution, and confidence interval;

[0036] When the deviation exceeds the acceptable range, the corresponding sensor is calibrated or reset.

[0037] Furthermore, the specific operation steps of the data acquisition module to obtain the frequency shadow value by comprehensive analysis according to the frequency judgment parameters of different sensors are as follows:

[0038] Frequency judgment parameters include:

[0039] Sensor response time: The speed at which a sensor responds to changes, recorded as the response rate value;

[0040] Data update demand: The application's demand for data update speed, including real-time control or historical records. The data update demand is scored from 1 to 100 and recorded as a value. The higher the value, the greater the data update demand.

[0041] Historical data analysis: By analyzing the historical usage data of the sensor, including the total usage time, number of failures, number of maintenances and average maintenance time, the total usage time, number of failures, number of maintenances and average maintenance time are calibrated as st, gc, wc and ws respectively, and then normalized and entered into the following formula: The historical evaluation value LPZ is obtained and used as a standard to measure the historical usage data of the sensor. The larger the historical evaluation value is, the lower the stability of the sensor is. On the contrary, the higher the stability of the sensor is.

[0042] Sensor life and durability: The expected life of the sensor is the remaining service time and durability score. After normalization, a circle is built with the remaining service life as the base radius, and a cone model is built with the durability score as the highest. The surface area of ​​the cone model is calculated and recorded as the transmission evaluation value, which is used as the standard for measuring the life and durability of the sensor;

[0043] Then, the obtained response speed value, change evaluation value, history evaluation value and transmission evaluation value are normalized and input into the following formula: To get the frequency shadow value YBZ, where XSZ, GPZ, LPZ and CPZ are the response speed value, the change evaluation value, the history evaluation value and the transmission evaluation value respectively. They are the preset weight coefficients for the response speed value, the change evaluation value, the history evaluation value and the transmission evaluation value, and the obtained frequency impact value is used as a standard to measure the influence of the sensor data collection frequency.

[0044] A lubrication system early warning method for wind power generation equipment comprises the following steps:

[0045] First, the different sensor parameter data collected by the data acquisition module are obtained, and the data is cleaned, outliers and noise are proposed, and the data is standardized or normalized so that the data are on the same scale. Key features are extracted from the raw data, including mean, standard deviation, trend and peak value, to determine which features are directly related to the abnormality of the lubrication system;

[0046] Analyze the trends of key parameters and use statistical methods or machine learning models to identify parameter trends;

[0047] Based on historical data and known abnormal cases, train the prediction model, select the corresponding algorithm, test the model using the validation set, evaluate the accuracy of the model prediction, adjust the model parameters, and optimize the model performance;

[0048] Deploy the trained model to the real-time monitoring system, input sensor data in real time, and use the model to predict potential anomalies; then set the warning threshold based on the model prediction results and actual application needs. The threshold can be a specific parameter value or parameter change rate;

[0049] Integrate early warning with data collection and monitoring to ensure that the system can automatically issue early warnings when potential anomalies are detected, and then preset early warning notification mechanisms, including sound and light alarms, SMS notifications or email reminders, to ensure that relevant personnel can receive early warning information in the first place;

[0050] After the warning is issued, the changes of various parameters are continuously monitored, and the authenticity of the warning is judged based on the changes of parameters related to the warning event;

[0051] When the inspection results show that the system is normal and there is a probability of false alarm, the cause of the false alarm is recorded and the warning is cancelled. When it is confirmed to be a false alarm, the warning threshold or model parameters are adjusted according to the cause of the false alarm to optimize the warning analysis process;

[0052] The entire warning processing process is recorded, including warning parameters, analysis results and final processing measures, and automatically entered into the warning processing report for future reference and system improvement.

[0053] Furthermore, the specific steps for judging the authenticity of the warning according to the changes in parameters related to the warning event are as follows:

[0054] Record the warning time and the current status of related parameters, automatically correct the parameter data collection frequency of the corresponding sensor to real-time collection, analyze the change trend of parameters before and after the warning again according to the new collection frequency, and use chart tools, including trend lines, line charts, etc., to visualize parameter changes;

[0055] Then compare the current parameters with historical data, find the change pattern with a similarity of 80%, and evaluate whether the current change is consistent with the historical abnormal event cases;

[0056] Then analyze the correlation between multiple related parameters to determine whether the changes in related parameters jointly indicate the pattern of the abnormal event. When the data analysis shows that there is indeed a trend of abnormal events, the used chart tool together with the upgraded early warning alert will be sent to the relevant personnel again;

[0057] At the same time, diagnostic tests are performed to confirm whether there are other abnormalities. When other abnormal trends exist at the same time, the corresponding abnormal trends are synchronized to relevant personnel to provide technical support for the technical staff's correlation analysis.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] The present invention can timely detect potential abnormalities and faults by real-time monitoring of key parameters of the lubrication system of wind power generation equipment, such as oil pressure, oil temperature, oil level, oil quality, vibration and current. Compared with the traditional periodic inspection method, this real-time monitoring method greatly improves the efficiency and accuracy of monitoring, and reduces the downtime and maintenance costs caused by equipment failure.

[0060] The present invention dynamically adjusts the data acquisition frequency according to the response time and data update requirements of different sensors through the data acquisition module. By calculating the frequency shadow value and the shadow difference value and comparing them with the preset threshold value, the system can intelligently optimize the acquisition frequency, thereby ensuring the real-time and accuracy of the data, while avoiding data redundancy and resource waste.

[0061] The early warning method of the present invention can not only analyze sensor data in real time, extract key features, and train prediction models to predict potential anomalies, but also judge the authenticity of the early warning by continuously monitoring and analyzing the changing trends of relevant parameters after the early warning is issued. The intelligent early warning system can respond quickly and provide accurate fault diagnosis information to technical personnel, thereby effectively improving the operating reliability and maintenance efficiency of wind power generation equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0063] Figure 1 It is the overall block diagram of the system of the present invention;

[0064] Figure 2 It is a flow chart of the early warning method in the present invention. DETAILED DESCRIPTION

[0065] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] It should be understood that the terms "include" and "comprising" used in the specification and claims of the present disclosure indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0067] It should also be understood that the terms used in this disclosure are only for the purpose of describing specific embodiments and are not intended to limit the disclosure. As used in this disclosure and claims, the singular forms of "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in this disclosure and claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations.

[0068] like Figure 1 As shown, a real-time detection system for the lubrication system of a wind power generation equipment includes a multi-parameter monitoring module and a data acquisition module;

[0069] The multi-parameter monitoring module is used to monitor different parameters of the lubrication system of wind power generation equipment in real time;

[0070] According to the monitoring requirements of the lubrication system of wind power generation equipment, the parameters that need to be monitored are determined, and then various sensors are installed at the corresponding positions in the lubrication system according to the parameters that need to be monitored, including:

[0071] Oil pressure sensor: monitors the oil pressure in the lubrication system to ensure that the oil can be effectively delivered to the parts that need lubrication; Oil temperature sensor: monitors the temperature of the lubricating oil in real time to prevent the lubrication effect from being affected by excessively high or low temperatures; Oil level sensor: monitors the oil level in the oil tank to ensure that the oil level is within a safe range to prevent the oil pump from sucking in air or insufficient oil; Oil quality sensor: monitors the quality of the lubricating oil, such as viscosity and contamination level, through the oil quality sensor; Vibration sensor: monitors the vibration of the equipment during operation as an auxiliary judgment of the equipment status; Current sensor: monitors the current of electrical equipment such as lubrication pumps to evaluate their working status;

[0072] Then check the accuracy of the installed sensor. The specific steps are as follows:

[0073] First, it is necessary to determine the calibration standard for the accuracy of sensor data. The calibration standard is obtained by analyzing the specifications and industry standards of the equipment manufacturer. The output of the sensor is verified by using standard reference equipment or calibration equipment of known accuracy. These equipment have certified accuracy and stability. Then, different sensors are tested, and data is collected from the sensor and reference equipment at the same time to ensure that the data is collected under the same environmental conditions. The sensor data is compared with the readings of the reference equipment, and the differences between the two are analyzed. The accuracy of the data is evaluated by using statistical methods, including the mean deviation and the standard deviation. The specific process is as follows: Collect data from the sensor and the reference equipment or standard equipment at the same time, measure the same physical quantity, and ensure that the readings of the sensor and the reference equipment are obtained at the same time or under the same conditions for easy comparison. Calculate the difference between the sensor reading and the reference equipment reading for each set of data, then add up all deviations and divide them by the number of measurements to obtain the average deviation, calculate the square of the difference between the deviation of each measurement and the average deviation to obtain the standard deviation. The standard deviation indicates the degree of dispersion of the deviation. The smaller the value, the more consistent the measurement results. Draw a deviation distribution diagram, including a histogram or a box plot, to visualize the distribution of the data and check whether the data distribution is symmetrical and whether there are outliers.

[0074] When the average deviation is less than the preset value, it proves that the sensor reading has no systematic deviation; then calculate the confidence interval of the sensor reading, which is the average plus or minus two times the standard deviation; record the deviation analysis results, including the average deviation, standard deviation, deviation distribution and confidence interval; when the deviation exceeds the acceptable range, calibrate or reset the corresponding sensor.

[0075] The data acquisition module is used to collect real-time data of various parameters through different sensors;

[0076] Select data acquisition hardware, including data acquisition cards, modules or systems, ensure that the hardware is compatible with the sensor and can process the required amount of data, connect the sensor to the data acquisition hardware through the corresponding interface, such as analog signals through analog input channels and data signals through digital input channels; configure data acquisition software, set data acquisition parameters, including sampling rate and channel configuration, and realize real-time reading, processing and storage of data through programming; start the data acquisition module, start collecting sensor data in real time, and monitor the data flow to ensure data quality and system stability, then set the data acquisition frequency, and conduct a comprehensive analysis based on the frequency judgment parameters of different sensors, where the frequency judgment parameters include:

[0077] Sensor response time: the speed at which the sensor responds to changes, recorded as the response rate value; Data update demand: the application's demand for data update speed, including real-time control or historical records, and assign a score of 1-100 to the data update demand, and record it as a higher score. The higher the score, the greater the data update demand; Historical data analysis: By analyzing the historical usage data of the sensor, including the total usage time, number of failures, number of maintenances, and average maintenance time, the total usage time, number of failures, number of maintenances, and average maintenance time are calibrated as st, gc, wc, and ws, respectively, and normalized and entered into the following formula: To obtain the historical evaluation value LPZ, and use the historical evaluation value as a standard for measuring the historical usage data of the sensor. The larger the historical evaluation value, the lower the stability of the sensor, and vice versa. Sensor life and durability: The expected life of the sensor is the remaining usage time and durability score. After normalization, a circle is established with the remaining service life as the bottom circle radius, and a cone model is established with the durability score as the high. The surface area of ​​the cone model is calculated and recorded as the transmission evaluation value, and this transmission evaluation value is used as a standard for measuring the life and durability of the sensor;

[0078] Then, the obtained response speed value, change evaluation value, history evaluation value and transmission evaluation value are normalized and input into the following formula: To get the frequency shadow value YBZ, where XSZ, GPZ, LPZ and CPZ are the response speed value, the change evaluation value, the history evaluation value and the transmission evaluation value respectively. The preset weight coefficients are respectively the response speed value, the change evaluation value, the history evaluation value and the transmission evaluation value, and the values ​​are set between 1.2-1.5, 0.6-0.9, 1.1-1.4 and 0.9-1.3 respectively; and the obtained frequency shadow value is used as a standard to measure the influence of the sensor data collection frequency;

[0079] The obtained frequency shadow value is then compared with the preset frequency shadow threshold. When the frequency shadow value is greater than the preset frequency shadow threshold, it is determined that the data acquisition frequency of the sensor can be optimized, and the difference between the frequency shadow value and the preset frequency shadow threshold is calculated and recorded as the shadow difference value. The obtained shadow difference value is compared with several preset shadow difference value intervals. Several shadow difference value intervals correspond to frequency optimization levels with different percentages. When the shadow difference interval to which the shadow difference value belongs is determined, the corresponding percentage frequency optimization level is obtained, and the frequency duration of the corresponding percentage frequency optimization level is increased on the basis of the original acquisition frequency of the sensor.

[0080] like Figure 2 As shown, a lubrication system early warning method for wind power generation equipment includes the following steps:

[0081] First, the different sensor parameter data collected by the data acquisition module are obtained, and the data is cleaned, outliers and noise are proposed, and the data is standardized or normalized so that the data are on the same scale. Key features are extracted from the raw data, including mean, standard deviation, trend and peak value, to determine which features are directly related to the abnormality of the lubrication system;

[0082] Analyze the trends of key parameters, including the gradual increase in oil temperature and the continuous decrease in oil pressure, and use statistical methods or machine learning models to identify the changing trends of parameters; train prediction models based on historical data and known abnormal cases, select corresponding algorithms, including event sequence analysis, support vector machines or neural networks, test the models using validation sets, evaluate the accuracy of model predictions, adjust model parameters, and optimize model performance;

[0083] Deploy the trained model to the real-time monitoring system, input sensor data in real time, and use the model to predict potential anomalies; then set the warning threshold based on the model prediction results and actual application requirements. The threshold can be a specific parameter value or parameter change rate; integrate the warning with data collection and monitoring to ensure that the system can automatically issue a warning when a potential anomaly is detected, and then preset the warning notification mechanism, including sound and light alarms, SMS notifications or email reminders, to ensure that relevant personnel can receive the warning information in the first time; after the warning is issued, continue to monitor the changes in each parameter, and judge the authenticity of the warning based on the changes in parameters related to the warning event. The specific process is as follows:

[0084] Record the warning time and the current status of related parameters, automatically correct the parameter data collection frequency of the corresponding sensor to real-time, analyze the change trend of the parameters before and after the warning again for the new collection frequency, and use chart tools, including trend lines, line charts, etc., to visualize parameter changes; then compare the current parameters with historical data, find the change pattern with a similarity of 80%, and evaluate whether the current change is consistent with the historical abnormal event case; then analyze the correlation between multiple related parameters to determine whether the changes in related parameters jointly indicate the pattern of the abnormal event. When the data analysis shows that there is indeed a trend of abnormal events, the chart tools used will be sent to the relevant personnel together with the upgraded warning alert; at the same time, perform diagnostic tests to confirm whether there are other abnormalities. When other abnormal trends exist at the same time, synchronize the corresponding abnormal trends to the relevant personnel to provide technical support for the correlation analysis of technical personnel;

[0085] When the inspection results show that the system is normal, it may be a false alarm. The cause of the false alarm is recorded and the warning is canceled. When it is confirmed to be a false alarm, the warning threshold or model parameters are adjusted according to the cause of the false alarm to optimize the warning analysis process; the entire warning processing process is recorded, including warning parameters, analysis results and final processing measures, and automatically entered into the warning processing report for future reference and system improvement.

[0086] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A real-time detection system for the lubrication system of a wind power generation equipment, characterized in that: include: The multi-parameter monitoring module is used to monitor different parameters of the lubrication system of wind power generation equipment in real time. The specific process is as follows: According to the monitoring requirements of the lubrication system of wind power generation equipment, determine the parameters that need to be monitored, and then install various sensors at the corresponding positions of the lubrication system according to the parameters that need to be monitored, including oil pressure sensors, oil temperature sensors, oil level sensors, oil quality sensors, vibration sensors and current sensors; Then check the accuracy of the installed sensors. If the check result is abnormal, calibrate or reset the corresponding sensors. The data acquisition module is used to collect real-time data of various parameters through different sensors. The specific steps are as follows: Select data acquisition hardware and connect the sensor to the data acquisition hardware through the corresponding interface; Configure data acquisition software, set data acquisition parameters, including sampling rate and channel configuration, and realize real-time reading, processing and storage of data through programming; Start the data acquisition module, start collecting sensor data in real time, monitor the data flow, set the data acquisition frequency, and perform comprehensive analysis based on different sensor frequency judgment parameters to obtain the frequency value; The frequency judgment parameters include: Sensor response time: The speed at which a sensor responds to changes, recorded as the response rate value; Data update demand: The application's demand for data update speed, including real-time control or historical records. The data update demand is scored from 1 to 100 and recorded as a value. The higher the value, the greater the data update demand. Historical data analysis: By analyzing the historical usage data of the sensor, including the total usage time, number of failures, number of maintenances and average maintenance time, the total usage time, number of failures, number of maintenances and average maintenance time are calibrated as st, gc, wc and ws respectively, and then normalized and entered into the following formula: The historical evaluation value LPZ is obtained and used as a standard to measure the historical usage data of the sensor. The larger the historical evaluation value is, the lower the stability of the sensor is. On the contrary, the higher the stability of the sensor is. Sensor life and durability: The expected life of the sensor is the remaining service time and durability score. After normalization, a circle is built with the remaining service life as the base radius, and a cone model is built with the durability score as the highest. The surface area of ​​the cone model is calculated and recorded as the transmission evaluation value, which is used as the standard for measuring the life and durability of the sensor; Then, the obtained response speed value, change evaluation value, history evaluation value and transmission evaluation value are normalized and input into the following formula: To get the frequency shadow value YBZ, where XSZ, GPZ, LPZ and CPZ are the response speed value, the change evaluation value, the history evaluation value and the transmission evaluation value respectively. are preset weight coefficients for the response speed value, the change evaluation value, the history evaluation value and the transmission evaluation value, and the obtained frequency impact value is used as a standard to measure the impact of the sensor data collection frequency; Then the obtained frequency shadow value is compared with the preset frequency shadow threshold. When the frequency shadow value is greater than the preset frequency shadow threshold, it is determined that the data acquisition frequency of the sensor needs to be optimized, and the difference between the frequency shadow value and the preset frequency shadow threshold is calculated and recorded as the shadow difference value; The obtained shadow difference value is compared with several preset shadow difference value intervals. Several shadow difference value intervals correspond to different percentage frequency optimization levels. When the shadow difference interval to which the shadow difference value belongs is determined, the corresponding percentage frequency optimization level is obtained, and the frequency duration of the corresponding percentage frequency optimization level is increased on the basis of the original acquisition frequency of the sensor.

2. A real-time detection system for the lubrication system of a wind power generation equipment according to claim 1, characterized in that: The data collected by different sensors in the multi-parameter monitoring module include: Oil pressure sensor: monitors the oil pressure in the lubrication system; Oil temperature sensor: real-time monitoring of lubricating oil temperature; Oil level sensor: monitors the oil level in the tank; Oil quality sensor: monitor the quality of lubricating oil, including viscosity and contamination level, through the oil quality sensor; Vibration sensor: monitors the vibration of the equipment during operation as an auxiliary judgment of the equipment status; Current sensor: monitors the current of lubrication pumps and other electrical equipment to evaluate the working status of the system.

3. A real-time detection system for a lubrication system of a wind power generation equipment according to claim 1, characterized in that: The specific operation steps of the multi-parameter monitoring module to verify the accuracy of the installed sensors are as follows: First, determine the calibration standard for sensor data accuracy, which is obtained by analyzing the equipment manufacturer's specifications and industry standards; Verify the output of the sensor using a standard reference device or calibration equipment of known accuracy; Then test different sensors, collect data from the sensor and the reference device at the same time, ensure that the data is collected under the same environmental conditions, compare the sensor data with the readings of the reference device, analyze the differences between the two, and evaluate the accuracy of the data by using statistical methods, including mean deviation and standard deviation. The specific process is as follows: Collect data from sensors and reference or standard devices simultaneously, measure the same physical quantity, and ensure that the readings of the sensor and reference device are obtained at the same time or under the same conditions; For each set of data, the difference between the sensor reading and the reference device reading is calculated, and then all deviations are summed up and divided by the number of measurements to obtain the average deviation. The square of the difference between the deviation of each measurement and the average deviation is calculated to obtain the standard deviation; Draw deviation distribution diagrams, including histograms or box plots, to visualize the distribution of data and check whether the data distribution is symmetrical and whether there are outliers; When the average deviation is less than the preset value, the sensor readings have no systematic deviation; Then calculate the confidence interval of the sensor readings, which is the mean plus or minus two standard deviations; Record the results of the deviation analysis, including the mean deviation, standard deviation, deviation distribution, and confidence interval; When the deviation exceeds the preset range, the corresponding sensor is calibrated or reset.

4. A real-time detection and early warning method for the lubrication system of a wind power generation equipment according to any one of claims 1 to 3, characterized in that: The following steps are involved: First, the different sensor parameter data collected by the data acquisition module are obtained, and the data is cleaned, outliers and noise are proposed, and the data is standardized or normalized so that the data are on the same scale. Key features are extracted from the raw data, including mean, standard deviation, trend and peak value, to determine which features are directly related to the abnormality of the lubrication system; Analyze the trends of key parameters and use statistical methods or machine learning models to identify parameter trends; Based on historical data and known abnormal cases, train the prediction model, select the corresponding algorithm, test the model using the validation set, evaluate the accuracy of the model prediction, adjust the model parameters, and optimize the model performance; Deploy the trained model to a real-time monitoring system, input sensor data in real time, and use the model to predict potential anomalies; Then, set the warning threshold based on the model prediction results and actual application requirements; Integrate early warning with data collection and monitoring to ensure that the system can automatically issue early warnings when potential anomalies are detected, and then preset early warning notification mechanisms, including sound and light alarms, SMS notifications or email reminders, to ensure that relevant personnel can receive early warning information in the first place; After the warning is issued, the changes of various parameters are continuously monitored, and the authenticity of the warning is judged based on the changes of parameters related to the warning event; When the inspection results show that the system is normal and there is a probability of false alarm, the cause of the false alarm is recorded and the warning is cancelled. When it is confirmed to be a false alarm, the warning threshold or model parameters are adjusted according to the cause of the false alarm to optimize the warning analysis process; The entire warning processing process is recorded, including warning parameters, analysis results and final processing measures, and automatically entered into the warning processing report for future reference and system improvement.

5. The real-time detection and early warning method according to claim 4 is characterized in that: The specific steps to determine the authenticity of the warning based on the changes in parameters related to the warning event are as follows: Record the warning time and the current status of related parameters, automatically correct the parameter data collection frequency of the corresponding sensor to real-time collection, analyze the change trend of parameters before and after the warning again according to the new collection frequency, and use chart tools to visualize parameter changes; Then compare the current parameters with historical data, find the change pattern with a similarity of 80%, and evaluate whether the current change is consistent with the historical abnormal event cases; Then analyze the correlation between multiple related parameters to determine whether the changes in related parameters jointly indicate the pattern of the abnormal event. When the data analysis shows that there is indeed a trend of abnormal events, the used chart tool together with the upgraded early warning alert will be sent to the relevant personnel again; At the same time, diagnostic tests are performed to confirm whether there are other abnormalities. When other abnormal trends exist at the same time, the corresponding abnormal trends are synchronized to relevant personnel to provide technical support for the technical staff's correlation analysis.