Fan monitoring method and monitoring system

Abstract

The invention discloses a draught fan monitoring method and system, and belongs to the technical field of draught fan monitoring, and the method comprises the steps that a server receives a first data set collected by collection equipment, and sets an offset reference value of the collection equipment; screening out first correction data and first early warning data according to the offset reference value; correcting the first correction data, and generating second early warning data; and generating an execution signal according to the first early warning data and the second early warning data, and sending the execution signal to an execution device. In the implementation process of the technical scheme, the offset reference value of the acquisition equipment is set, the first data set is subjected to statistical analysis in combination with the offset reference value, the first correction data and the first early warning data are obtained, early warning is carried out according to the first early warning data, early warning is carried out again after the first correction data are corrected, and the accuracy of early warning is improved. A grading early warning mechanism is achieved, the operation stability of the fan is improved, and the influence of icing disasters on the operation stability of the fan is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of fan monitoring, and specifically to a fan monitoring method and a monitoring system. Background Art

[0002] A wind turbine is a power generation device that converts wind energy into electrical energy. It is widely used in areas with abundant wind energy, thereby achieving clean electricity. The widespread use of wind turbines has played a vital role in reducing carbon emissions.

[0003] In actual applications, wind turbines cannot always maintain stable operation, especially in winter. Most wind-rich areas will have meteorological factors such as low temperature and high humidity in winter. These meteorological factors will affect the operating efficiency of the wind turbine. In addition, low temperature and high humidity will be accompanied by blade icing disasters, resulting in ice on the blade surface, further causing problems such as increased wind turbine load and reduced power generation efficiency. In the prior art, a sensor group is usually installed on the wind turbine to monitor the blades and other components in real time, and a maintenance signal is issued based on the monitoring results, and then the wind turbine is maintained manually or with other equipment.

[0004] However, when a low temperature disaster occurs, not only will icing disasters affect the blades and other components, but it will also affect the working state of the sensor, resulting in reduced monitoring accuracy. The collected data cannot accurately reflect the actual situation of the wind turbine, which in turn leads to untimely response to maintenance signals and inability to implement maintenance operations in a timely manner, affecting the stable operation of the fan in the above-mentioned low temperature environment.

[0005] Therefore, it is necessary to provide a fan monitoring method and a monitoring system to solve the above problems.

[0006] It should be noted that the above information disclosed in this background technology section is only for understanding the background technology of the present application concept, and therefore, it may contain information that does not constitute the prior art. Summary of the invention

[0007] Based on the above-mentioned problems existing in the prior art, the problem to be solved by the present application is: to provide a wind turbine monitoring method and a monitoring system, by performing graded early warning on the collection equipment and correcting the collected data during the early warning process to make it conform to the actual situation of the wind turbine operation, thereby reducing the impact of icing disasters on the stability of wind turbine operation.

[0008] The technical solution adopted by the present application to solve the technical problem is: a fan monitoring method, comprising:

[0009] The server receives a first data set collected by a collection device and sets an offset reference value of the collection device;

[0010] Performing statistical analysis on the first data set according to the offset reference value to screen out first correction data and first warning data;

[0011] The first correction data is further processed, the ice thickness is predicted by establishing a correction model, the prediction result is verified, the first correction data is corrected according to the verification result, and second warning data is generated;

[0012] An execution signal is generated according to the first warning data and the second warning data, and the execution signal is sent to the execution device. The execution device automatically starts the warning and corrective measures according to the received signal.

[0013] During the implementation of the technical solution of the present application, by setting the offset reference value of the acquisition equipment, a statistical analysis is performed on the first data set in combination with the offset reference value to obtain first correction data and first warning data, a warning is given based on the first warning data, and a warning is given again after correcting the first correction data, thereby realizing a hierarchical warning mechanism, improving the stability of wind turbine operation, and reducing the impact of icing disasters on the stability of wind turbine operation.

[0014] Furthermore, the acquisition device is an ultrasonic sensor.

[0015] Furthermore, the relationship between the propagation speed of ultrasound in the air and the temperature is obtained: v=331.5+0.6t, where v is the propagation speed and t is the Celsius temperature. Then, the factory error range [a,b] of the ultrasonic sensor is obtained in millimeters, and the right endpoint b of the factory error range is taken for calculation. The size of the offset reference value is b / (331.5+0.6t).

[0016] Furthermore, statistical analysis is performed on the first data set, and first correction data and first warning data are screened out, including: obtaining sensor acquisition data of the blade when it is not covered with ice, and using it as baseline data; establishing a linear model of ice thickness and temperature on the blade surface, and predicting theoretical ice thickness at different temperatures; converting the theoretical ice thickness at different temperatures into a distance signal of an acquisition device, and the distance signal is the time it takes for the acquisition device to complete an acquisition process; comparing the actual acquisition data with the offset reference value based on the converted distance signal, and using the acquisition data that is less than or equal to the offset reference value as the first correction data, and using the acquisition data that is greater than the offset reference value as the first warning data.

[0017] Furthermore, the linear model is implemented using the Stefan equation.

[0018] Furthermore, converting the theoretical ice thickness at different temperatures into the distance signal of the acquisition device includes: dividing the ultrasonic propagation distance by the ultrasonic change equation with temperature, that is, the distance signal is (initial distance-theoretical ice thickness) / (331.5+0.6t).

[0019] Furthermore, the correction of the first correction data is implemented by adopting a correction model, wherein the correction model is a Huber loss function, wherein the loss function has an adjustable parameter, and the first correction data is corrected by combining the adjustable parameter with an offset reference value.

[0020] Furthermore, before sending the execution signal to the execution device, the accuracy of the execution signal is verified, and the accuracy verification process is implemented by using dynamic error analysis and feedback adjustment methods.

[0021] A fan monitoring system, comprising:

[0022] An offset reference value setting module is used for the server to receive a first data set collected by a collection device and set an offset reference value of the collection device;

[0023] A statistical analysis module, used for performing statistical analysis on the first data set according to the offset reference value, and screening out first correction data and first warning data;

[0024] A correction module, used for further processing the first correction data, predicting the ice thickness by establishing a correction model, verifying the prediction result, correcting the first correction data according to the verification result, and generating second warning data;

[0025] The early warning execution module is used to generate an execution signal according to the first early warning data and the second early warning data, and send the execution signal to the execution device. The execution device automatically starts the early warning and corrective measures according to the received signal.

[0026] The beneficial effects of the present application are as follows: a wind turbine monitoring method and monitoring system provided by the present application, by setting an offset reference value of an acquisition device, statistically analyzing a first data set in combination with the offset reference value, obtaining first correction data and first warning data, issuing a warning based on the first warning data, and issuing a warning again after correcting the first correction data, thereby realizing a graded warning mechanism, improving the stability of wind turbine operation, and reducing the impact of icing disasters on the stability of wind turbine operation.

[0027] In addition to the above-described purposes, features and advantages, the present application also has other purposes, features and advantages. The present application will be further described in detail with reference to the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The drawings in the specification, which constitute a part of the present application, are used to provide further understanding of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0029] In the attached picture:

[0030] Figure 1 A schematic diagram of a flow chart of a fan monitoring method in this application;

[0031] Figure 2 This is a schematic diagram of the module structure of a fan monitoring system in this application. DETAILED DESCRIPTION

[0032] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0033] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0034] Embodiment 1: Figure 1 As shown, the present application provides a fan monitoring method, which is applied to the monitoring process of the fan, especially for the monitoring of the fan in a low temperature environment, to realize the calibration of the monitoring data without replacing the monitoring sensor, so that it can still accurately reflect the actual operating status of the fan under low temperature conditions, thereby improving the response of the maintenance signal and maintaining the fan in time. Specifically, the fan monitoring method includes the following steps:

[0035] Step 10: The server receives the first data set collected by the collection device and sets the offset reference value of the collection device;

[0036] In order to monitor the fan in a low temperature environment and prevent ice disasters, it is necessary to use acquisition equipment to collect data from key components such as blades. The acquisition equipment is various sensors, such as optical sensors, which are used to monitor the light reflection changes on the blade surface to identify the formation of ice; ultrasonic sensors, which use the ultrasonic reflection principle to identify the thickness of the ice layer on the blade surface; vibration sensors, which are used to identify the vibration changes during the movement of the blade, so as to determine whether there are changes in vibration characteristics caused by ice, etc. These sensors constitute the acquisition equipment and can be used alone or in combination. In this embodiment, an ultrasonic sensor is taken as an example;

[0037] The server refers to a hardware module with functions such as data reception and data analysis, which is used to receive data collected by the collection device, analyze it, and output the analysis results. It is different from the concept of server in the prior art;

[0038] After receiving the first collected data, it is also necessary to perform statistical analysis on it to determine whether the data collected by the collection device truly reflects the operating status of the blade. Before performing the statistical analysis, it is necessary to set an offset reference value first. The reference value will be used as a standard for statistical analysis of the first data set and subsequent calibration. Specifically, the acquisition of the offset reference value adopts the description based on the embodiment. The acquisition of the offset reference value adopts a multi-parameter comprehensive analysis strategy, combining external factors such as ambient temperature and humidity, and historical data during the operation of the fan to comprehensively obtain an offset reference value. In addition, for different types of sensors, the offset reference value is generally different. Taking an ultrasonic sensor as an example, the setting of the offset reference value needs to consider the change in the propagation speed of ultrasonic waves in a low-temperature environment, as well as the original error range of the ultrasonic sensor, so as to determine the final offset reference value;

[0039] Specifically, according to the prior art, the propagation speed of ultrasound in the air is proportional to the temperature, that is, the lower the temperature, the slower the propagation speed of ultrasound. The specific relationship is v=331.5+0.6t, where t is the Celsius temperature and v is the propagation speed in m / s. Therefore, when formulating the offset reference value, this physical property needs to be taken into account. The specific calculation method of the offset reference value is: obtain the factory error range [a, b] of the ultrasonic sensor in millimeters, and take the right end point b of the factory error range for calculation. The offset reference value = b / (331.5+0.6t) in seconds. When substituting, b needs to be converted to meters for calculation;

[0040] The offset reference value calculated by the above formula is the data offset critical value of the ultrasonic sensor in a low temperature environment. The working principle of the ultrasonic sensor is based on the time difference between the sound wave reflecting after contacting the surface of the object and then reflecting back to measure the distance, so as to judge whether there is ice disaster on the blade surface. According to the factory error range of the ultrasonic sensor, the offset reference value can be calculated by taking the right endpoint for calculation.

[0041] Step 20: Performing statistical analysis on the first data set according to the offset reference value to screen out first correction data and first warning data;

[0042] After obtaining the offset reference value, it is necessary to perform statistical analysis on the first data set according to the offset reference value, because within the error range, even if the collected data has deviations, it can still be corrected. If the error is large, it is necessary to give an early warning in time, omit the correction link, improve the early warning efficiency, and ensure real-time monitoring and adjustment of the blade operation status; Specifically, performing statistical analysis on the first data set and screening out the first correction data and the first early warning data includes the following steps:

[0043] Step 201: obtaining sensor data collected when the blade is not covered with ice, and using the data as reference data;

[0044] Before performing statistical analysis on the first data set, it is necessary to first obtain the sensor data collected when the blade is not covered with ice as the baseline data, and predict the theoretical ice thickness at each temperature based on the baseline data. Because under normal circumstances, there is a linear relationship between ice thickness and temperature within a fixed range, so the theoretical ice thickness prediction can be performed.

[0045] Step 202: Establish a linear model of ice thickness and temperature on the blade surface to predict theoretical ice thickness at different temperatures;

[0046] Based on the benchmark data, the ice thickness on the blade surface is predicted using a linear model to obtain the theoretical ice thickness at different temperatures. The linear model can be implemented using the Stefan equation in the prior art, which is: Among them, h is the height of the phase change interface (such as the thickness of the ice layer), t is the time, k is the thermal conductivity, L is the latent heat of phase change, T is the temperature, and z is the vertical coordinate, that is, the phase change cross section. Through this equation, the theoretical ice thickness of the blade surface at different temperatures can be predicted in combination with the benchmark data;

[0047] In addition, the heat conduction method can also be used to realize the theoretical ice thickness prediction. The heat conduction equation can describe the process of heat propagation in the ice, thereby revealing the influence of temperature on the ice thickness. For details, reference can be made to the prior art and will not be described in detail in this embodiment.

[0048] Step 203: converting the theoretical ice thickness at different temperatures into a distance signal of a collection device, where the distance signal is the time it takes for the collection device to complete one collection process;

[0049] Taking the ultrasonic sensor as an example, the greater the ice thickness on the blade surface, the shorter the distance between the sensor and the reflective surface, and the shorter the time to receive the reflected signal. Therefore, it is necessary to convert the theoretical ice thickness at different temperatures into corresponding distance signals for subsequent analysis. Specifically, the distance signal is the time for the acquisition device to complete a collection process. There is an initial distance between the ultrasonic sensor and the corresponding reflective surface of the blade. As the ice is applied, the distance between the reflective surface and the ultrasonic sensor becomes smaller and smaller, which results in a shorter time for the ultrasonic sensor to complete a collection process. Therefore, the theoretical ice thickness at different temperatures is converted into a distance signal. The specific conversion method is to divide the ultrasonic propagation distance by the ultrasonic velocity variation with temperature in the aforementioned process, that is, (initial distance-theoretical ice thickness) / (331.5+0.6t), to obtain the distance signal of the acquisition device, wherein the units of the initial distance and the theoretical ice thickness are both meters;

[0050] Step 204: According to the converted distance signal, the actual collected data is compared with the offset reference value, and the collected data less than or equal to the offset reference value is used as the first correction data, and the collected data greater than the offset reference value is used as the first warning data;

[0051] After acquiring the distance signal, the actual collected data can be compared with the offset reference value, and the collected data that is less than or equal to the offset reference value can be used as the first correction data. When it does not exceed the offset reference value, it can be corrected through a subsequent correction process without affecting the normal operation of the blade. After analysis, the collected data that is greater than the offset reference value can be used as the first warning data. When the actual data is greater than the offset reference value, no correction is required and a warning is issued directly, because this indicates that the ice thickness on the blade surface may have reached or exceeded the safety range, or the collection equipment is too affected by the temperature and causes erroneous collection, so a warning is issued directly.

[0052] Step 30: further processing the first corrected data, predicting the ice thickness by establishing a correction model, verifying the prediction result, correcting the first corrected data according to the verification result, and generating second warning data;

[0053] The first correction data selected in the above process needs to be further corrected so that it can accurately reflect the ice coverage on the blade surface, and continue to issue warnings after correction to generate second warning data, thereby achieving graded warnings. By improving data accuracy, the reliability of the warning system can be ensured, and the accuracy of prediction can be improved, thereby reducing errors and ensuring the safety of blade operation. Under the graded warning mechanism, the ice coverage on the blade surface can be more accurately grasped, and corresponding measures can be taken for warnings of different levels;

[0054] Further processing of the first corrected data is implemented by using a correction model, the correction model is a Huber loss function, the loss function can correct the predicted value in the presence of an abnormal value, and the offset reference value in this embodiment is combined with an adjustable parameter in the Huber loss function, so as to correct the first corrected data;

[0055] After the correction is completed, since there is an adjustable parameter in the Huber loss function, the adjustable parameter is used as the threshold of the loss function, and the correction value of the correction process is compared with the threshold, and the data exceeding the threshold is used as the second warning data. When using the Huber loss function, the data whose comparison result is less than or equal to the threshold is predicted using the mean square error, and the data greater than the threshold is predicted using the absolute error, so that the correction process is smoother and the influence of outliers on the prediction results is reduced. The specific formula of the Huber loss function can refer to the prior art and will not be elaborated in this embodiment.

[0056] Step 40: Generate an execution signal according to the first warning data and the second warning data, and send the execution signal to the execution device. The execution device automatically starts the warning and corrective measures according to the received signal.

[0057] After generating the warning data, an execution signal can be generated according to different warning data. When the execution signal is sent to the execution device, the corresponding warning and correction measures can be implemented, such as sound and light alarms, displays, etc. Before the execution signal is sent to the execution device, accuracy verification is required. The verification process is implemented using dynamic error analysis and feedback adjustment methods, combined with real-time meteorological data and historical data to achieve reconfirmation of the warning data and execution signals. For details, please refer to the existing technology.

[0058] Embodiment 2: Figure 2 As shown, this embodiment proposes a fan monitoring system, which runs the fan monitoring method in embodiment 1, and the system includes:

[0059] An offset reference value setting module is used for the server to receive a first data set collected by a collection device and set an offset reference value of the collection device;

[0060] A statistical analysis module, used for performing statistical analysis on the first data set according to the offset reference value, and screening out first correction data and first warning data;

[0061] A correction module, used for further processing the first correction data, predicting the ice thickness by establishing a correction model, verifying the prediction result, correcting the first correction data according to the verification result, and generating second warning data;

[0062] The early warning execution module is used to generate an execution signal according to the first early warning data and the second early warning data, and send the execution signal to the execution device. The execution device automatically starts the early warning and corrective measures according to the received signal.

[0063] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A fan monitoring method, characterized in that: include: The server receives a first data set collected by a collection device and sets an offset reference value of the collection device; Performing statistical analysis on the first data set according to the offset reference value to screen out first correction data and first warning data; The first correction data is further processed, the ice thickness is predicted by establishing a correction model, the prediction result is verified, the first correction data is corrected according to the verification result, and second warning data is generated; An execution signal is generated according to the first warning data and the second warning data, and the execution signal is sent to the execution device. The execution device automatically starts the warning and corrective measures according to the received signal.

2. A fan monitoring method according to claim 1, characterized in that: The acquisition device is an ultrasonic sensor.

3. A fan monitoring method according to claim 2, characterized in that: The calculation method of the offset reference value is: obtain the relationship between the propagation speed of ultrasound in the air and the temperature v=331.5+0.6t, where v is the propagation speed and t is the Celsius temperature, then obtain the factory error range [a,b] of the ultrasonic sensor in millimeters, and take the right endpoint b of the factory error range for calculation. The size of the offset reference value is b / (331.5+0.6t).

4. A fan monitoring method according to claim 1, characterized in that: Performing statistical analysis on the first data set and screening out first correction data and first warning data include: obtaining sensor data collected when the blade is not covered with ice and using it as reference data; establishing a linear model of ice thickness and temperature on the blade surface to predict theoretical ice thickness at different temperatures; converting the theoretical ice thickness at different temperatures into a distance signal of a collection device, where the distance signal is the time it takes for the collection device to complete one collection process; comparing the actual collection data with an offset reference value based on the converted distance signal, using the collection data that is less than or equal to the offset reference value as the first correction data, and using the collection data that is greater than the offset reference value as the first warning data.

5. A fan monitoring method according to claim 4, characterized in that: The linear model is implemented using the Stefan equation.

6. A fan monitoring method according to claim 4, characterized in that: The conversion of the theoretical ice thickness at different temperatures into the distance signal of the acquisition device includes: dividing the ultrasonic propagation distance by the ultrasonic change with temperature, that is, the distance signal is (initial distance - theoretical ice thickness) / (331.5+0.6t)。 7. A fan monitoring method according to claim 1, characterized in that: The correction of the first correction data is implemented by using a correction model, the correction model is a Huber loss function, the loss function has an adjustable parameter, and the first correction data is corrected by combining the adjustable parameter with the offset reference value.

8. A fan monitoring method according to claim 1, characterized in that: Before sending the execution signal to the execution device, the accuracy of the execution signal is also verified, and the accuracy verification process is implemented by using dynamic error analysis and feedback adjustment methods.

9. A fan monitoring system, characterized in that: include: An offset reference value setting module is used for the server to receive a first data set collected by a collection device and set an offset reference value of the collection device; A statistical analysis module, used for performing statistical analysis on the first data set according to the offset reference value, and screening out first correction data and first warning data; A correction module, used for further processing the first correction data, predicting the ice thickness by establishing a correction model, verifying the prediction result, correcting the first correction data according to the verification result, and generating second warning data; The early warning execution module is used to generate an execution signal according to the first early warning data and the second early warning data, and send the execution signal to the execution device. The execution device automatically starts the early warning and corrective measures according to the received signal.

10. A fan monitoring system according to claim 9, characterized in that: Used to implement a wind turbine monitoring method as described in any one of claims 1 to 8.

Images