Method and device for detecting an anomaly in a variable speed converter system of a wind turbine generator

Abstract

Provided are an abnormality detection method and device for a variable flow system of a wind turbine generator. The abnormality detection method comprises: determining a temperature threshold of a target component in the variable flow system based on power of the wind turbine generator; determining whether the temperature threshold of the target component and an operating temperature of the target component satisfy a preset condition, wherein the preset condition comprises a difference between the operating temperature and the temperature threshold being greater than a preset difference; in response to the temperature threshold and the operating temperature satisfying the preset condition, determining whether an operating state of a water-cooling heat dissipation system of the wind turbine generator is abnormal; and in response to the operating state of the water-cooling heat dissipation system being normal, determining that the target component has an abnormal temperature. The abnormality detection method and device for the variable flow system of the wind turbine generator improve the effect of abnormality detection of the wind turbine generator.

Description

Technical Field

[0001] This disclosure relates to the field of wind power generation technology. More specifically, this disclosure relates to an anomaly detection method and apparatus for the converter system of a wind turbine generator set. Background Technology

[0002] The converter system is one of the most important systems in a wind turbine generator set. Its main function is to convert the unstable AC power generated by the wind turbine into stable AC power that meets the requirements of the power grid. Therefore, the stable operation of the converter system plays a crucial role in the normal power generation of the wind turbine generator set. The generator-side rectifier in the converter system converts the unstable AC power generated by the wind turbine into stable DC power, which is then converted into AC power that meets the grid requirements by the grid-side inverter. Because these devices essentially change the amplitude and frequency of the power generated by the wind turbine, the temperature of these components is strongly correlated with the changes in the active power of the wind turbine generator set. Furthermore, the operating temperature of these components varies depending on the ambient temperature and the specific wind turbine generator set model.

[0003] As wind turbines age, especially those over 10 years old, their converter systems and related cooling systems inevitably experience performance degradation. This can manifest as clogged water-cooled filters, abnormal temperatures in water-cooled inlet and outlet valves, and damaged cooling fans. When components such as inverters and rectifiers in the converter system experience abnormal temperatures, or when the water-cooling system malfunctions and exceeds its fault settings, it not only disrupts grid connection and power generation but also results in power loss, impacting customer revenue. Furthermore, prolonged exposure to high temperatures can cause component performance degradation, potentially leading to damage and explosion of the rectifier and inverter, resulting in safety accidents.

[0004] Current research on anomaly detection in inverters, rectifiers and other components of converter systems mainly focuses on anomaly detection or fault diagnosis of specific components, which has the problem of high probability of missed or false alarms. Summary of the Invention

[0005] According to an exemplary embodiment of this disclosure, an anomaly detection method for a converter system of a wind turbine generator set is provided, comprising: determining a temperature threshold of a target component in the converter system based on the power of the wind turbine generator set; determining whether the temperature threshold of the target component and the operating temperature of the target component meet a preset condition, wherein the preset condition includes a difference between the operating temperature and the temperature threshold being greater than a preset difference; in response to the temperature threshold and the operating temperature meeting the preset condition, determining whether the operating state of the water-cooled heat dissipation system of the wind turbine generator set is abnormal; and in response to the normal operating state of the water-cooled heat dissipation system, determining that the temperature of the target component is abnormal.

[0006] Optionally, determining the temperature threshold of the target component in the converter system based on the power of the wind turbine generator set may include: selecting the temperature threshold of the target component corresponding to the power of the wind turbine generator set based on the relationship between the power of the wind turbine generator set and the temperature threshold of the target component.

[0007] Optionally, the relationship between the power of the wind turbine generator set and the temperature threshold of the target component can be obtained by the following operations: acquiring the operating data of at least some wind turbine generator sets in the wind farm where the wind turbine generator set is located within a preset time period, the operating data including the temperature of the target component and the power of the wind turbine generator set; and determining the relationship between the power of the wind turbine generator set and the temperature threshold of the target component based on the operating data.

[0008] Optionally, determining the relationship between the power of the wind turbine generator set and the temperature threshold of the target component based on the operating data may include: dividing the temperature of the target component included in the operating data into power windows according to the power included in the operating data; determining the temperature threshold of the target component corresponding to the power in each power window for each power window; and determining the relationship between the power of the wind turbine generator set and the temperature threshold of the target component based on the temperature threshold of the target component corresponding to the power in each power window.

[0009] Optionally, determining the temperature threshold of the target component corresponding to the power in each power window may include: calculating the average temperature of the target component in each power window and three times the standard deviation of the target component temperature; if three times the standard deviation of the target component temperature is greater than a predetermined three times the standard deviation of the target component temperature corresponding to each power window, calculating the difference between three times the standard deviation of the target component temperature and the predetermined three times the standard deviation of the target component temperature to obtain a first difference; calculating the sum of the average temperature of the target component, the predetermined value of the target component temperature corresponding to each power window, and the first difference to obtain the temperature threshold of the target component corresponding to the power in each power window.

[0010] Optionally, determining the temperature threshold of the target component corresponding to the power in each power window may further include: calculating the average value of the target component temperature and the sum of the average value of the target component temperature and the predetermined value of the target component temperature corresponding to each power window, provided that three times the standard deviation of the target component temperature is not greater than three times the standard deviation of the target component temperature corresponding to each power window, to obtain the temperature threshold of the target component corresponding to the power in each power window.

[0011] Optionally, the power of the wind turbine generator set may include the active power of the wind turbine generator set.

[0012] Optionally, the target component may include at least one of an inverter and a rectifier.

[0013] Optionally, the anomaly detection method may further include: when the water cooling system is malfunctioning, determining that the target component is normal and the water cooling system is malfunctioning.

[0014] Optionally, the anomaly detection method may further include: in response to an anomaly in the water cooling system or a target component, displaying data about the anomaly on a display screen, wherein the data about the anomaly includes at least one of the wind turbine generator number, anomaly time, anomaly component, anomaly cause, and a visualization image of the anomaly.

[0015] According to an exemplary embodiment of this disclosure, an anomaly detection device for a converter system of a wind turbine generator set is provided, comprising: a threshold determination unit configured to determine a temperature threshold of a target component in the converter system based on the power of the wind turbine generator set; a condition determination unit configured to determine whether the temperature threshold of the target component and the operating temperature of the target component meet a preset condition, wherein the preset condition includes a difference between the operating temperature and the temperature threshold being greater than a preset difference; a heat dissipation system detection unit configured to determine whether the operating state of the water-cooled heat dissipation system of the wind turbine generator set is abnormal in response to the temperature threshold and the operating temperature meeting the preset condition; and a temperature anomaly determination unit configured to determine that the temperature of the target component is abnormal in response to the normal operating state of the water-cooled heat dissipation system.

[0016] Optionally, the threshold determination unit may be configured to: select the temperature threshold of the target component corresponding to the power of the wind turbine generator set based on the relationship between the power of the wind turbine generator set and the temperature threshold of the target component.

[0017] Optionally, the anomaly detection device may further include: a correspondence determination unit, configured to acquire operating data of at least some wind turbines in the wind farm where the wind turbine is located within a preset time period, the operating data including the temperature of the target component and the power of the wind turbine; and determine the relationship between the power of the wind turbine and the temperature threshold of the target component based on the operating data.

[0018] Optionally, the correspondence determination unit may be configured to: divide the target component temperature included in the operating data into power windows according to the power included in the operating data; for each power window after power windowing, determine the temperature threshold of the target component corresponding to the power in each power window; and determine the relationship between the power of the wind turbine generator set and the temperature threshold of the target component based on the temperature threshold of the target component corresponding to the power in each power window.

[0019] Optionally, the correspondence determination unit may be configured to: calculate the average value of the target component temperature and three times the standard deviation of the target component temperature in each power window; if three times the standard deviation of the target component temperature is greater than a predetermined three times the standard deviation of the target component temperature corresponding to each power window, calculate the difference between three times the standard deviation of the target component temperature and the predetermined three times the standard deviation of the target component temperature to obtain a first difference; calculate the sum of the average value of the target component temperature, the predetermined value of the target component temperature corresponding to each power window, and the first difference to obtain the temperature threshold of the target component corresponding to the power in each power window.

[0020] Optionally, the correspondence determination unit may be configured to: calculate the average value of the target component temperature and the sum of the average value of the target component temperature and the predetermined value of the target component temperature corresponding to each power window, provided that three times the standard deviation of the target component temperature is not greater than three times the standard deviation of the target component temperature corresponding to each power window, to obtain the temperature threshold of the target component corresponding to the power in each power window.

[0021] Optionally, the power of the wind turbine generator set may include the active power of the wind turbine generator set.

[0022] Optionally, the target component may include at least one of an inverter and a rectifier.

[0023] Optionally, the anomaly detection device may further include: a heat dissipation anomaly determination unit, configured to determine that the target component is normal and the water cooling system is abnormal when the water cooling system is operating abnormally.

[0024] Optionally, the anomaly detection device may further include: an anomaly display unit, configured to display data about the anomaly on a display screen in response to an anomaly in the water cooling system or a target component, wherein the data about the anomaly includes at least one of the wind turbine generator number, anomaly time, anomaly component, anomaly cause, and a visual image about the anomaly.

[0025] According to exemplary embodiments of the present disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements an anomaly detection method according to exemplary embodiments of the present disclosure.

[0026] According to an exemplary embodiment of the present disclosure, a computing device is provided, including: a processor; and a memory storing a computer program, wherein when the computer program is executed by the processor, it implements an anomaly detection method according to an exemplary embodiment of the present disclosure.

[0027] According to exemplary embodiments of the present disclosure, a computer program product is provided, wherein the instructions in the computer program product are executable by a processor of a computer device to perform an anomaly detection method according to exemplary embodiments of the present disclosure.

[0028] An anomaly detection method and apparatus for a converter system of a wind turbine generator set according to an exemplary embodiment of the present disclosure determines the temperature threshold of a target component in the converter system based on the power of the wind turbine generator set, and determines whether the temperature threshold of the target component and the operating temperature of the target component meet preset conditions. The preset conditions include that the difference between the operating temperature and the temperature threshold is greater than a preset difference. In response to the temperature threshold and the operating temperature meeting the preset conditions, it is determined whether the operating state of the water cooling system of the wind turbine generator set is abnormal. In response to the normal operating state of the water cooling system, it is determined that the temperature of the target component is abnormal, thereby improving the effectiveness of anomaly detection for the wind turbine generator set.

[0029] Further aspects and / or advantages of the general concept of this disclosure will be set forth in part in the description which follows, and in part will be clear from the description or may be learned by practice of the general concept of this disclosure. Attached Figure Description

[0030] The above and other objects and features of exemplary embodiments of this disclosure will become clearer from the following description taken in conjunction with the accompanying drawings, which exemplarily illustrate the embodiments, wherein:

[0031] Figure 1A A flowchart illustrating an anomaly detection method for a converter system of a wind turbine generator set according to an exemplary embodiment of the present disclosure is shown.

[0032] Figure 1B A flowchart illustrating an anomaly detection method for a converter system of a wind turbine generator set according to another exemplary embodiment of the present disclosure is shown.

[0033] Figure 1C A training logic diagram of a dynamic threshold setting model according to an exemplary embodiment of the present disclosure is shown;

[0034] Figure 1D A training logic diagram of a water-cooled heat dissipation anomaly detection model according to an exemplary embodiment of the present disclosure is shown.

[0035] Figure 2A A flowchart illustrating an anomaly detection method for a converter system of a wind turbine generator set according to an exemplary embodiment of the present disclosure is provided.

[0036] Figure 2B An interface showing the results of anomaly detection in the converter system of a wind turbine generator set according to an exemplary embodiment of the present disclosure is provided.

[0037] Figure 2C A daily data trend graph showing the abnormality of the water cooling system and the abnormal temperature of the target components of a wind turbine generator set according to an exemplary embodiment of the present disclosure;

[0038] Figure 2D A graph showing the data variation trend of the target component temperature of an abnormal wind turbine generator set compared with the entire wind farm according to an exemplary embodiment of the present disclosure;

[0039] Figure 2E This diagram illustrates the daily data trend of a wind turbine generator set with a normal water-cooled heat dissipation system and abnormal target component temperatures, according to an exemplary embodiment of the present disclosure.

[0040] Figure 2F A graph showing the data variation trend of the target component temperature of an abnormal wind turbine generator set compared with the entire wind farm according to an exemplary embodiment of the present disclosure;

[0041] Figure 3 A block diagram of an anomaly detection device for a converter system of a wind turbine generator set according to an exemplary embodiment of the present disclosure is shown.

[0042] Figure 4 A block diagram showing an anomaly detection device for a converter system of a wind turbine generator set according to an exemplary embodiment of the present disclosure; and

[0043] Figure 5 A schematic diagram of a computing device according to an exemplary embodiment of the present disclosure is shown. Detailed Implementation

[0044] Exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings, examples of which are illustrated in the drawings, wherein the same reference numerals always refer to the same components. The embodiments will now be described with reference to the accompanying drawings in order to explain the present disclosure.

[0045] When components such as inverters and rectifiers in the converter system of wind turbines experience abnormal temperatures, prolonged exposure to high temperatures can lead to performance degradation. When component temperatures exceed fault limits, it can cause frequent turbine shutdowns, disrupting normal grid connection and power generation, resulting in power loss and impacting the economic benefits of wind farms. Severe performance degradation due to high temperatures can even cause component damage or explosion, affecting the converter cabinet and other components within the nacelle, leading to significant property damage and safety accidents. Existing methods do not eliminate the interference of water cooling on detection results, and the complex algorithm design of the model itself contributes to issues such as missed and false alarms.

[0046] To address the aforementioned issues, this disclosure proposes an anomaly detection method for the converter system of a wind turbine generator set. Based on accurate detection using appropriate detection methods tailored to the data characteristics of different components, it eliminates interference from water-cooling system anomalies on the converter system, effectively achieving accurate detection and precise location of abnormal components in the converter system. The anomaly detection method for the converter system of a wind turbine generator set proposed in this disclosure can detect abnormal temperatures in the rectifier and inverter of the wind turbine generator set's converter system in advance, preventing frequent shutdowns due to overheating, damage due to performance degradation, and safety accidents caused by component explosions. Furthermore, it can help on-site maintenance personnel more accurately locate abnormal components in the wind turbine generator set, effectively reducing the operation and maintenance costs of the wind turbine generator set.

[0047] The anomaly detection method for the converter system of a wind turbine generator set according to the exemplary embodiments of this disclosure can be executed by the field-level controller of the wind farm, or by the main controller of the wind turbine generator set, or by any processing device (such as a server) with data analysis, processing and control functions. This disclosure does not limit the scope of the method.

[0048] Taking server execution as an example, Figure 1A A flowchart illustrating an anomaly detection method for a converter system of a wind turbine generator set according to an exemplary embodiment of the present disclosure is shown.

[0049] like Figure 1A As shown, the server first acquires data, including the active power of the wind turbine generator, grid-side inverter temperature, turbine-side rectifier temperature, water-cooled inlet and outlet valve temperatures, and water-cooling system flow and pressure. Next, data cleaning and processing (including data preprocessing and feature extraction) are performed. Then, based on real-time data, dynamic thresholds for inverter and rectifier temperatures are set, and an anomaly detection model for the water-cooling system is constructed. The server then reads the wind turbine generator's detection data and uses dynamic thresholds to determine rectifier and inverter anomalies. If components exhibit abnormal temperatures, the water-cooling system anomaly detection model is used to identify the anomaly and pinpoint the faulty component of the wind turbine generator based on the final detection results. After receiving the anomaly result, the server sends it to a display device, which shows the faulty component, the specific cause of the anomaly, and a visualized image of the day's data, allowing on-site maintenance personnel to understand the wind turbine generator's anomaly from a data perspective.

[0050] Figure 1B A flowchart illustrating an anomaly detection method for a converter system of a wind turbine generator set according to another exemplary embodiment of the present disclosure is shown. Figure 1C A training logic diagram of a dynamic threshold setting model according to an exemplary embodiment of the present disclosure is shown. Figure 1D This diagram illustrates a training logic diagram of a water-cooled heat dissipation anomaly detection model according to an exemplary embodiment of the present disclosure. The anomaly detection method in this disclosure is based on at least a portion of data such as the active power of the wind turbine generator during operation, grid-side inverter temperature, turbine-side rectifier temperature, water-cooled inlet and outlet valve temperatures, and water-cooled heat dissipation system flow rate and pressure. This data can also be cleaned and processed.

[0051] As an example, this data can be obtained by reading transient data from the Supervisory Control and Data Acquisition (SCADA) system of the same type of wind turbine generator set in multiple wind farms for at least one year, which includes the temperature of the grid-side inverter and the temperature of the generator-side rectifier.

[0052] As an example, when performing data cleaning, empty data and dead value data can be deleted, and time format conversion can be performed.

[0053] As an example, during data processing, data from grid-connected power generation can be filtered to ensure that the rectifier and inverter are operating at that time. The temperatures of the grid-side inverter and generator-side rectifier in the converter system are normally maintained within a certain range; therefore, temperature data within this fixed range can be filtered out.

[0054] Each model of the converter system contains multiple components with identical functions, and each component has a corresponding temperature measurement point. Therefore, the data to be analyzed includes the temperature of the grid-side inverter components themselves and the temperature difference between any two grid-side inverters, the temperature of the generator-side rectifier components themselves and the temperature difference between any two generator-side rectifiers, etc.

[0055] Reference Figure 1B In step S101, the temperature threshold of the target components in the converter system is determined based on the power of the wind turbine generator set.

[0056] In an exemplary embodiment of this disclosure, when determining the temperature threshold of a target component in a converter system based on the power of a wind turbine generator set, the temperature threshold of the target component corresponding to the power of the wind turbine generator set can be selected based on the relationship between the power of the wind turbine generator set and the temperature threshold of the target component.

[0057] In an exemplary embodiment of this disclosure, the relationship between the power of the wind turbine generator set and the temperature threshold of the target component can be obtained by the following operations: acquiring the operating data of at least some wind turbine generator sets in the wind farm where the wind turbine generator set is located within a preset time period, the operating data including the temperature of the target component and the power of the wind turbine generator set; and determining the relationship between the power of the wind turbine generator set and the temperature threshold of the target component based on the operating data.

[0058] In an exemplary embodiment of this disclosure, when determining the relationship between the power of the wind turbine generator set and the temperature threshold of the target component based on the operating data, the temperature of the target component included in the operating data can first be divided into power windows according to the power included in the operating data. For each power window after power windowing, the temperature threshold of the target component corresponding to the power in each power window is determined. Then, based on the temperature threshold of the target component corresponding to the power in each power window, the relationship between the power of the wind turbine generator set and the temperature threshold of the target component is determined.

[0059] In an exemplary embodiment of this disclosure, when determining the temperature threshold of the target component corresponding to the power in each power window, the average value of the target component temperature and three times the standard deviation of the target component temperature in each power window can be calculated first. If three times the standard deviation of the target component temperature is greater than a predetermined three times the standard deviation of the target component temperature corresponding to each power window, the difference between three times the standard deviation of the target component temperature and the predetermined three times the standard deviation of the target component temperature is calculated to obtain a first difference. Then, the sum of the average value of the target component temperature, the predetermined value of the target component temperature corresponding to each power window, and the first difference is calculated to obtain the temperature threshold of the target component corresponding to the power in each power window.

[0060] In an exemplary embodiment of this disclosure, when determining the temperature threshold of the target component corresponding to the power in each power window, the average value of the target component temperature and the sum of the average value of the target component temperature and the predetermined value of the target component temperature corresponding to each power window can be calculated, provided that three times the standard deviation of the target component temperature is not greater than three times the standard deviation of the target component temperature corresponding to each power window, to obtain the temperature threshold of the target component corresponding to the power in each power window.

[0061] In exemplary embodiments of this disclosure, the power of the wind turbine generator set may include the active power of the wind turbine generator set.

[0062] In exemplary embodiments of this disclosure, the target component may include at least one of an inverter (also known as a grid-side inverter) and a rectifier (also known as a machine-side rectifier).

[0063] The temperature threshold of a target component can be determined using a dynamic threshold setting model. Before determining the temperature threshold of the target component using the dynamic threshold setting model, the dynamic threshold setting model needs to be trained to obtain a trained dynamic threshold setting model.

[0064] like Figure 1CAs shown, when training the dynamic threshold setting model, the temperature data of the target components (e.g., grid-side inverters and turbine-side rectifiers) of the same type of wind turbine generator set in multiple wind farms are first read, and then the data is cleaned and processed.

[0065] As an example, temperature data can be obtained by reading transient data from the Supervisory Control and Data Acquisition (SCADA) system of the same type of wind turbine generator set in multiple wind farms for at least one year, which includes the temperature of the grid-side inverter and the temperature of the generator-side rectifier.

[0066] As an example, when performing data cleaning, empty data and dead value data can be deleted, and time format conversion can be performed.

[0067] As an example, during data processing, data from grid-connected power generation can be filtered to ensure that the rectifier and inverter are operating at that time. The temperatures of the grid-side inverter and generator-side rectifier in the converter system are normally maintained within a certain range; therefore, temperature data within this fixed range can be filtered out.

[0068] Each model of the converter system contains multiple components with the same function, and each component has a corresponding temperature measurement point. When the target components are the grid-side inverter and the machine-side rectifier, the data to be analyzed includes the temperature of the grid-side inverter components themselves and the temperature difference between any two grid-side inverters, the temperature of the machine-side rectifier components themselves and the temperature difference between any two machine-side rectifiers, etc.

[0069] When the amount of data after data processing reaches the required amount of data for training, the dynamic threshold calculation logic is set, and the parameters such as the threshold (also known as the fixed threshold) and normal three standard deviation are determined in the dynamic threshold calculation process. It is then judged whether the threshold (also known as the fixed threshold) and normal three standard deviation are reasonable. When it is judged that the threshold (also known as the fixed threshold) and normal three standard deviation are reasonable, the threshold (also known as the fixed threshold) and normal three standard deviation are saved.

[0070] The temperature of target components such as grid-side inverters and turbine-side rectifiers in the converter system is strongly correlated with active power. Therefore, a power windowing approach is chosen to determine the model threshold. When determining the temperature threshold (i.e., the dynamic threshold) of the target components, relevant data from multiple wind turbine generators of the same model are first read. The average temperature within each window is calculated for each component, using a1 as a window, and a fixed threshold is added to obtain the over-temperature threshold for that component. If the actual three-standard-deviation temperature within a window is greater than the normal three-standard-deviation obtained during model training, the threshold for that window is calculated by adding the fixed threshold to the average temperature and the difference between the actual and normal three-standard-deviation values.

[0071] In step S102, it is determined whether the temperature threshold and the operating temperature of the target component meet preset conditions. Here, the preset conditions may include the difference between the operating temperature and the temperature threshold being greater than a preset difference.

[0072] As an example, taking data from a single wind turbine generator set, specifically the grid-side inverter, the temperature data of the grid-side inverter is compared with the calculated dynamic temperature threshold of the inverter. If the amount of data showing the actual temperature exceeding the dynamic threshold is greater than p3, the grid-side inverter is determined to be overheating. The actual temperature difference between each pair of grid-side inverters is compared with the dynamic threshold for temperature difference between grid-side inverters. If the amount of data showing the actual temperature difference exceeding the dynamic threshold is greater than p4, the temperature difference between each pair of inverters is determined to be too large.

[0073] As an example, taking data from a single wind turbine generator set, specifically the turbine-side rectifier, the temperature data of the turbine-side rectifier is compared with the calculated dynamic temperature threshold of the rectifier. If the amount of data showing the actual temperature exceeding the dynamic threshold is greater than p5, the turbine-side rectifier is determined to be overheating. The actual temperature difference between each pair of turbine-side rectifiers is compared with the dynamic threshold for temperature difference between the turbine-side rectifiers. If the amount of data showing the actual temperature difference exceeding the dynamic threshold is greater than p6, the temperature difference between each pair of rectifiers is determined to be too large.

[0074] In step S103, in response to the temperature threshold and the operating temperature satisfying the preset conditions, it is determined whether the operating status of the water-cooled heat dissipation system of the wind turbine generator set is abnormal. Here, step S103 is executed if any target component, such as the grid-side inverter or the machine-side rectifier, has an abnormal temperature. If no target component, such as the grid-side inverter or the machine-side rectifier, has an abnormal temperature, it is determined that the wind turbine generator set is operating normally, and steps S103 and S104 can be omitted.

[0075] Here, a water-cooled heat dissipation system anomaly detection model can be used to determine whether the operating status of the water-cooled heat dissipation system of the wind turbine generator is abnormal.

[0076] The anomaly detection model of the water cooling system can be trained in advance to obtain the anomaly detection threshold data of the water cooling system.

[0077] like Figure 1DAs shown, when training the anomaly detection model for the water-cooled heat dissipation system, data can be read first. For example, SCADA transient data (e.g., but not limited to, water-cooled heat dissipation system inlet and outlet valve temperatures, water-cooled heat dissipation system inlet and outlet valve pressures, water-cooled heat dissipation system flow rates) can be read from multiple wind farms over a certain period of time. After reading the SCADA transient data, the data is cleaned. For example, but not limited to, deleting empty data, deleting dead value data, and performing time format conversion. After data cleaning, the cleaned data is processed. For example, since the water-cooled system temperature, pressure, and flow rate are all within a stable range under normal conditions, data within a fixed range are filtered out to avoid abnormally high or low values ​​due to sensor malfunctions or problems in the data transmission process. For example, the processed data can be the water-cooled heat dissipation system inlet and outlet valve temperature difference, the water-cooled outlet valve temperature difference between the water-cooled system and the ambient temperature, the water-cooled inlet and outlet valve pressure difference, and the water-cooled heat dissipation system coolant flow rate. Finally, if the amount of processed data meets the training requirements of the water-cooling system anomaly detection model, the processed data is used to train the model to obtain its threshold, and this threshold is saved. The relevant data of the water-cooling system is relatively stable, with small fluctuations under normal circumstances. Therefore, a fixed threshold is selected as the threshold for the water-cooling system anomaly detection model based on the actual data during model training. In step S103, the presence of anomalies in the water-cooling system can be determined based on the trained threshold (i.e., the fixed threshold). For example, the trained fixed threshold is compared with the actual data. If the amount of any abnormal data feature in the comparison result is greater than p5, the water-cooling system is determined to be abnormal; otherwise, the water-cooling system is determined to be normal.

[0078] In step S104, in response to the normal operation of the water-cooled heat dissipation system, it is determined that the temperature of the target component is abnormal.

[0079] In an exemplary embodiment of this disclosure, when the water-cooling heat dissipation system is malfunctioning, it can also be determined that the target component is normal and the water-cooling heat dissipation system is malfunctioning.

[0080] In an exemplary embodiment of this disclosure, in response to an anomaly in the water-cooling system or a target component, data regarding the anomaly can be displayed on a screen to facilitate on-site maintenance personnel in understanding the specific situation of the malfunctioning wind turbine. Here, the data regarding the anomaly includes at least one of the wind turbine's serial number, the time of the anomaly, the malfunctioning component, the cause of the anomaly, and a visual image related to the anomaly.

[0081] Figure 2AA flowchart illustrating an anomaly detection method for a converter system of a wind turbine generator according to an exemplary embodiment of the present disclosure is shown. Figure 2A The following is an example of an anomaly detection method, using inverters and rectifiers as target components.

[0082] like Figure 2AAs shown, in step 1, all wind turbine generator data (hereinafter referred to as wind turbines) of the wind farm are read. In step 2, data cleaning, data filtering, and data processing are performed. In step 3, it is determined whether the amount of data obtained in step 2 is greater than the preset P1. If the result is greater, step 4 is executed; if the result is less, step 5 is executed. In step 4, the temperature threshold (i.e., the dynamic threshold) is calculated based on the data obtained in step 2. In step 5, if the amount of data is insufficient, the anomaly detection ends. In step 6, the data of the wind turbine generator to be detected is extracted. In step 7, it is determined whether the amount of data obtained in step 6 is greater than the preset P2. If the result is greater, step 8 is executed; if the result is less, step 9 is executed. In step 8, based on the data obtained in step 6, the temperatures of the inverter and rectifier are compared with the corresponding temperature thresholds (i.e., the dynamic thresholds). In step 9, if the amount of data is insufficient, the anomaly detection ends. In step 10, based on the comparison results in step 8, determine whether the inverter and rectifier are abnormal. If the result indicates an abnormality, proceed to step 11; otherwise, proceed to step 16. In step 11, select data such as temperature, flow rate, and pressure of the water-cooled heat dissipation system of the wind turbine generator set to be tested. In step 12, obtain (import) the threshold of the water-cooled heat dissipation system abnormality detection model (i.e., the threshold set by the water-cooled heat dissipation detection model). In step 13, based on the data obtained in step 11 and the threshold obtained in step 12, determine whether the water-cooled heat dissipation system of the wind turbine generator set to be tested is abnormal. If the result indicates an abnormality, proceed to step 14; otherwise, proceed to step 15. In step 14, determine that the water-cooled heat dissipation system of the wind turbine generator set is abnormal. In step 15, determine that the water-cooled heat dissipation system of the wind turbine generator set is normal, but the temperature of the inverter or rectifier is abnormal. In step 16, save the results of step 14 or step 15. In step 17, it is determined whether all wind turbines in the wind farm have completed the inspection (i.e., early warning calculation). If the result is that all have completed the inspection, step 18 is executed. If the result is that not all have completed the inspection, step 6 is returned to execute, and the wind turbines that have not completed the inspection are inspected sequentially until all wind turbines in the wind farm have completed the inspection. In step 18, the results from step 16 are output (e.g., displayed).

[0083] Figure 2B This illustrates a result display interface for anomaly detection of the converter system of a wind turbine generator set according to an exemplary embodiment of the present disclosure. Figure 2CThis diagram illustrates a daily data trend of a wind turbine generator set exhibiting abnormal water cooling system and abnormal temperature of target components according to an exemplary embodiment of the present disclosure. Figure 2D This diagram illustrates the data variation trend of the target component temperature of an abnormal wind turbine generator set compared to the entire wind farm, according to an exemplary embodiment of the present disclosure. Figure 2E The diagram shows a daily data trend of a wind turbine generator set with a normal water-cooled heat dissipation system and abnormal target component temperatures, according to an exemplary embodiment of the present disclosure. Figure 2F This diagram illustrates the data variation trend of the target component temperature of an abnormal wind turbine generator set compared to the entire wind farm, according to an exemplary embodiment of the present disclosure.

[0084] like Figure 2B As shown, the results display interface may include the initial main page, the test results of a single wind turbine generator (unit), and the data visualization of a single wind turbine generator (unit).

[0085] Figure 2C This illustrates a negative example of a malfunctioning water-cooling system. For example... Figure 2C As shown in the second sub-graph from the top, the data indicates that the inlet and outlet valve temperatures are high, with a large difference compared to the ambient temperature. This suggests that the wind turbine generator is experiencing abnormal water cooling, resulting in the turbine-side rectifier temperature being higher than the grid-side inverter temperature (normally, the turbine-side rectifier temperature is slightly lower than the grid-side inverter temperature). Furthermore, there is no strong correlation between the turbine-side rectifier temperature and the active power. Figure 2C The first sub-image from top to bottom is shown. Figure 2D As shown, when comparing the temperature of the turbine-side rectifier with the average temperature of the entire wind farm in a power window of 20kW, the turbine-side rectifier temperature of the abnormal wind turbine is also higher than the average temperature of the turbine-side rectifier across the entire wind farm. Figure 2D In the diagram, the horizontal axis represents active power.

[0086] Figure 2E This shows a negative example of a normal water-cooling system. Figure 2E In this case, the rectifier on the machine side of the wind turbine generator set exhibited abnormal temperature conditions. For example... Figure 2E As shown in the first sub-figure from top to bottom, the abnormal temperature trend of the turbine-side rectifier differs from that of the grid-side inverter, and it does not show a strong correlation with active power; and as... Figure 2F The abnormal wind turbine generator set's machine-side rectifier temperature is much higher than the machine-side rectifier temperature of the entire wind farm, and the abnormal wind turbine generator set's grid-side inverter temperature is much higher than the grid-side inverter temperature of the entire wind farm.

[0087] Furthermore, according to exemplary embodiments of the present disclosure, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed, implements an anomaly detection method for the converter system of a wind turbine generator set according to exemplary embodiments of the present disclosure.

[0088] In an exemplary embodiment of this disclosure, the computer-readable storage medium may carry one or more programs that, when executed, can perform the following steps: determining a temperature threshold for a target component in a converter system based on the power of the wind turbine generator set; determining whether the temperature threshold and the operating temperature of the target component meet a preset condition, wherein the preset condition includes a difference between the operating temperature and the temperature threshold being greater than a preset difference; in response to the temperature threshold and the operating temperature meeting the preset condition, determining whether the operating state of the water-cooled heat dissipation system of the wind turbine generator set is abnormal; and in response to the normal operating state of the water-cooled heat dissipation system, determining that the temperature of the target component is abnormal.

[0089] Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In embodiments of this disclosure, a computer-readable storage medium can be any tangible medium that contains or stores a computer program that can be used by or in conjunction with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable storage medium can be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof. A computer-readable storage medium can be included in any apparatus; it can also exist independently without being assembled into that apparatus.

[0090] Furthermore, according to exemplary embodiments of the present disclosure, a computer program product is also provided, wherein the instructions in the computer program product are executable by a processor of a computer device to perform a method for anomaly detection of a converter system of a wind turbine generator set according to exemplary embodiments of the present disclosure.

[0091] The above has been combined Figures 1A to 2F An anomaly detection method for the converter system of a wind turbine generator set according to exemplary embodiments of the present disclosure has been described. Hereinafter, reference will be made to... Figure 3 and Figure 4An anomaly detection device and its unit for a converter system of a wind turbine generator set according to exemplary embodiments of the present disclosure will be described.

[0092] Figure 3 A block diagram of an anomaly detection device for a converter system of a wind turbine generator set according to an exemplary embodiment of the present disclosure is shown. Figure 4 A block diagram of an anomaly detection device for a converter system of a wind turbine generator set according to an exemplary embodiment of the present disclosure is shown.

[0093] Reference Figure 3 The anomaly detection device for the converter system of the wind turbine generator set includes a threshold determination unit 31, a condition determination unit 32, a heat dissipation system detection unit 33, and a temperature anomaly determination unit 34.

[0094] The threshold determination unit 31 is configured to determine the temperature threshold of the target component in the converter system based on the power of the wind turbine generator.

[0095] In an exemplary embodiment of this disclosure, the threshold determination unit 31 may be configured to: select a temperature threshold of the target component corresponding to the power of the wind turbine generator set based on the relationship between the power of the wind turbine generator set and the temperature threshold of the target component.

[0096] In an exemplary embodiment of this disclosure, the power of the wind turbine generator set may include the active power of the wind turbine generator set.

[0097] In exemplary embodiments of this disclosure, the target component may include at least one of an inverter and a rectifier.

[0098] The condition determination unit 32 is configured to determine whether the temperature threshold of the target component and the operating temperature of the target component meet preset conditions, wherein the preset conditions include that the difference between the operating temperature and the temperature threshold is greater than a preset difference.

[0099] The heat dissipation system detection unit 33 is configured to determine whether the operating status of the water-cooled heat dissipation system of the wind turbine generator is abnormal in response to the temperature threshold and the operating temperature meeting the preset conditions.

[0100] The temperature anomaly determination unit 34 is configured to determine that the target component has a temperature anomaly in response to the normal operation of the water cooling system.

[0101] Reference Figure 4 In addition to the threshold determination unit 31, condition determination unit 32, heat dissipation system detection unit 33 and temperature anomaly determination unit 34, the anomaly detection device of the converter system of the wind turbine generator may also include a correspondence determination unit 35, heat dissipation anomaly determination unit 36 ​​and anomaly display unit 37.

[0102] The correspondence determination unit 35 is configured to acquire operating data of at least some wind turbines in the wind farm where the wind turbine is located within a preset time period, the operating data including the temperature of the target component and the power of the wind turbine; and determine the relationship between the power of the wind turbine and the temperature threshold of the target component based on the operating data.

[0103] In an exemplary embodiment of this disclosure, the correspondence determination unit 35 may be configured to: divide the target component temperature included in the operating data into power windows according to the power included in the operating data; for each power window after power windowing, determine the temperature threshold of the target component corresponding to the power in each power window; and determine the relationship between the power of the wind turbine generator set and the temperature threshold of the target component based on the temperature threshold of the target component corresponding to the power in each power window.

[0104] In an exemplary embodiment of this disclosure, the correspondence determination unit 35 may be configured to: calculate the average value of the target component temperature and three times the standard deviation of the target component temperature in each power window; if three times the standard deviation of the target component temperature is greater than a predetermined three times the standard deviation of the target component temperature corresponding to each power window, calculate the difference between three times the standard deviation of the target component temperature and the predetermined three times the standard deviation of the target component temperature to obtain a first difference; calculate the sum of the average value of the target component temperature, the predetermined value of the target component temperature corresponding to each power window, and the first difference to obtain the temperature threshold of the target component corresponding to the power in each power window.

[0105] In an exemplary embodiment of this disclosure, the correspondence determination unit 35 may be configured to: calculate the average value of the target component temperature and the sum of the average value of the target component temperature and the predetermined value of the target component temperature corresponding to each power window, provided that the three-times standard deviation of the target component temperature is not greater than the predetermined three-times standard deviation of the target component temperature corresponding to each power window, to obtain the temperature threshold of the target component corresponding to the power in each power window.

[0106] The heat dissipation anomaly determination unit 36 ​​is configured to determine whether the target component is normal and the water cooling system is abnormal when the water cooling system is operating abnormally.

[0107] The anomaly display unit 37 is configured to display anomaly data on a display screen in response to an anomaly in the water-cooling system or a target component. Here, the anomaly data includes at least one of the following: the wind turbine generator's serial number, the time of the anomaly, the malfunctioning component, the cause of the anomaly, and a visual image of the anomaly.

[0108] The above has been combined Figure 3 and Figure 4 An anomaly detection device for the converter system of a wind turbine generator set according to an exemplary embodiment of the present disclosure has been described. Next, in conjunction with... Figure 5 A computing device according to exemplary embodiments of the present disclosure will be described.

[0109] Figure 5 A schematic diagram of a computing device according to an exemplary embodiment of the present disclosure is shown.

[0110] Reference Figure 5 The computing device 5 according to an exemplary embodiment of the present disclosure includes a memory 51 and a processor 52. The memory 51 stores a computer program, which, when executed by the processor 52, implements an anomaly detection method for the converter system of a wind turbine generator set according to an exemplary embodiment of the present disclosure.

[0111] In an exemplary embodiment of this disclosure, when the computer program is executed by the processor 52, the following steps can be implemented: determining a temperature threshold of a target component in the converter system based on the power of the wind turbine generator set; determining whether the temperature threshold of the target component and the operating temperature of the target component meet a preset condition, wherein the preset condition includes a difference between the operating temperature and the temperature threshold being greater than a preset difference; in response to the temperature threshold and the operating temperature meeting the preset condition, determining whether the operating state of the water-cooled heat dissipation system of the wind turbine generator set is abnormal; in response to the normal operating state of the water-cooled heat dissipation system, determining that the temperature of the target component is abnormal.

[0112] Figure 5 The computing device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments disclosed herein.

[0113] The above has been referred to Figures 1A to 5 An anomaly detection method and apparatus according to exemplary embodiments of the present disclosure are described. However, it should be understood that: Figure 3 and Figure 4 The anomaly detection device and its units shown can be configured as software, hardware, firmware, or any combination thereof to perform specific functions. Figure 5 The computing device shown is not limited to the components shown above, but some components may be added or removed as needed, and the above components may also be combined.

[0114] An anomaly detection method and apparatus for a converter system of a wind turbine generator set, according to an exemplary embodiment of this disclosure, determines the temperature threshold of a target component in the converter system based on the power of the wind turbine generator set, and determines whether the temperature threshold and the operating temperature of the target component meet preset conditions. The preset conditions include a difference between the operating temperature and the temperature threshold being greater than a preset difference. In response to the temperature threshold and the operating temperature meeting the preset conditions, it is determined whether the operating state of the water-cooled heat dissipation system of the wind turbine generator set is abnormal. In response to the normal operating state of the water-cooled heat dissipation system, it is determined that the temperature of the target component is abnormal, thereby improving the effectiveness of anomaly detection for the wind turbine generator set. The anomaly detection method and apparatus according to an exemplary embodiment of this disclosure can not only detect and eliminate abnormal components in advance, preventing economic losses caused by component temperature abnormalities triggering fault shutdowns or safety accidents caused by component explosions, but also eliminates the interference of water-cooled heat dissipation system abnormalities on inverter and rectifier temperature abnormalities, more clearly locating abnormal components and improving operation and maintenance efficiency.

[0115] Furthermore, the anomaly detection method and apparatus for the converter system of a wind turbine generator set according to the exemplary embodiments of this disclosure use a power windowing method to detect anomalies in the inverter and rectifier temperatures. Based on data from the entire wind farm, a dynamic threshold is determined, reducing false alarms and missed alarms caused by the single threshold method, thereby improving the accuracy of anomaly detection and saving maintenance time.

[0116] Furthermore, the anomaly detection method and apparatus for the converter system of a wind turbine generator set according to the exemplary embodiments of this disclosure can be deployed not only at the field-level controller of a wind farm, but also at the main controller of the wind turbine generator set, and can also be deployed on any processing device (such as a server) with data analysis, processing, and control functions. This disclosure does not limit the scope of the invention in this regard. Although this disclosure has been specifically shown and described with reference to its exemplary embodiments, those skilled in the art should understand that various changes in form and detail can be made thereto without departing from the spirit and scope of this disclosure as defined by the claims.

Claims

1. A method for detecting anomalies in the converter system of a wind turbine generator set, characterized in that, The anomaly detection method includes: Determine the temperature threshold of target components in the converter system based on the power of the wind turbine generator set; Determine whether the temperature threshold and the operating temperature of the target component meet preset conditions, wherein the preset conditions include the difference between the operating temperature and the temperature threshold being greater than a preset difference; In response to the temperature threshold and the operating temperature meeting the preset conditions, it is determined whether the operating status of the water-cooled heat dissipation system of the wind turbine generator is abnormal. In response to the normal operation of the water-cooled heat dissipation system, it is determined that the temperature of the target component is abnormal. The method of determining the temperature threshold of target components in the converter system based on the power of the wind turbine generator set includes: Based on the relationship between the power of the wind turbine generator set and the temperature threshold of the target component, the temperature threshold of the target component corresponding to the power of the wind turbine generator set is selected. The relationship between the power of the wind turbine generator set and the temperature threshold of the target component is obtained through the following operation: The operation data of at least some of the wind turbines in the wind farm where the wind turbines are located is obtained within a preset time period, and the operation data includes the temperature of the target components and the power of the wind turbines. Based on the operational data, the relationship between the power of the wind turbine generator set and the temperature threshold of the target components is determined.

2. The anomaly detection method according to claim 1, characterized in that, Determining the relationship between the power of the wind turbine generator and the temperature threshold of the target components based on the operating data includes: The target component temperatures included in the operating data are divided into power windows according to the power included in the operating data; For each power window after power windowing, determine the temperature threshold of the target component corresponding to the power in each power window; Based on the temperature threshold of the target component corresponding to the power in each power window, the relationship between the power of the wind turbine generator set and the temperature threshold of the target component is determined.

3. The anomaly detection method according to claim 2, characterized in that, Determining the temperature threshold of the target component corresponding to the power in each power window includes: Calculate the average value of the target component temperature and three times the standard deviation of the target component temperature for each power window; If the three-times standard deviation of the target component temperature is greater than the predetermined three-times standard deviation of the target component temperature corresponding to each power window, the difference between the three-times standard deviation of the target component temperature and the predetermined three-times standard deviation of the target component temperature is calculated to obtain the first difference. The average temperature of the target component, the predetermined value of the target component temperature corresponding to each power window, and the first difference are calculated to obtain the temperature threshold of the target component corresponding to the power in each power window.

4. The anomaly detection method according to claim 3, characterized in that, The step of determining the temperature threshold of the target component corresponding to the power in each power window further includes: If the three-standard-deviation of the target component temperature is not greater than the predetermined three-standard-deviation of the target component temperature corresponding to each power window, the average value of the target component temperature and the sum of the predetermined value of the target component temperature corresponding to each power window are calculated to obtain the temperature threshold of the target component corresponding to the power in each power window.

5. The anomaly detection method according to any one of claims 1-4, characterized in that, The power of a wind turbine generator set includes its active power.

6. The anomaly detection method according to any one of claims 1-4, wherein, The target components include at least one of an inverter and a rectifier.

7. The anomaly detection method according to claim 6, characterized in that, Also includes: When the water cooling system malfunctions, it is determined that the target component is normal and the water cooling system is abnormal.

8. The anomaly detection method according to claim 1, characterized in that, Also includes: In response to a malfunction in the water-cooling system or a target component, data regarding the malfunction will be displayed on the screen. The data concerning the anomaly includes at least one of the following: the wind turbine generator's serial number, the time of the anomaly, the abnormal component, the cause of the anomaly, and a visual image of the anomaly.

9. An anomaly detection device, comprising: The threshold determination unit is configured to determine the temperature threshold of target components in the converter system based on the power of the wind turbine generator set. The condition determination unit is configured to determine whether the temperature threshold of the target component and the operating temperature of the target component meet preset conditions, wherein the preset conditions include the difference between the operating temperature and the temperature threshold being greater than a preset difference. A heat dissipation system detection unit is configured to determine whether the operating status of the water-cooled heat dissipation system of the wind turbine generator is abnormal in response to the temperature threshold and the operating temperature meeting the preset condition; and The temperature anomaly determination unit is configured to determine that the target component has a temperature anomaly in response to the normal operation of the water cooling system. The threshold determination unit is further configured to select a temperature threshold of a target component corresponding to the power of the wind turbine generator set based on the relationship between the power of the wind turbine generator set and the temperature threshold of the target component. The relationship between the power of the wind turbine generator set and the temperature threshold of the target component is obtained through the following operations: acquiring the operating data of at least some wind turbine generator sets in the wind farm where the wind turbine generator set is located within a preset time period, the operating data including the temperature of the target component and the power of the wind turbine generator set; and determining the relationship between the power of the wind turbine generator set and the temperature threshold of the target component based on the operating data.

10. A computer-readable storage medium storing a computer program, wherein, When the computer program is executed by a processor, it implements the anomaly detection method according to any one of claims 1 to 8.

11. A computing device, comprising: processor; A memory storing a computer program that, when executed by the processor, implements the anomaly detection method according to any one of claims 1 to 8.

Images