Running state risk assessment method and system for wind generating set

Abstract

The invention relates to the technical field of wind power generation, and discloses an operation state risk assessment method and system for a wind generating set, and the method comprises the steps: building a risk assessment model based on a neural network model, collecting historical first operation data and corresponding state data of the wind generating set in a normal operation state, and obtaining a risk assessment result; training a risk assessment model; acquiring real-time operation data of the wind generating set, acquiring actually measured state data corresponding to the real-time operation data, inputting the real-time operation data into the risk assessment model, and outputting predicted state data corresponding to the real-time operation data by the risk assessment model; the difference reference value corresponding to the current environment data is obtained, the difference between the prediction state data and the actual measurement state data is calculated, the difference is compared with the difference reference value, the operation state of the wind generating set is determined, and the risk assessment accuracy is improved.

Description

Technical Field

[0001] The present application relates to the field of wind power generation technology, and in particular to a method and system for risk assessment of the operating status of a wind turbine generator set. Background Art

[0002] Traditional wind turbine condition monitoring relies heavily on fixed thresholds to identify anomalies, which presents significant limitations. In complex and changing environments, such as sudden temperature changes and wind speed fluctuations, fixed thresholds are difficult to adapt to different operating conditions, leading to false alarms or missed reports. For example, at high temperatures and low wind speeds, the heat dissipation efficiency of the gearbox decreases, and normal temperature fluctuations can be easily misjudged as faults. Meanwhile, brief increases in vibration amplitude at low temperatures and high wind speeds may be overlooked, leading to missed reports of potential risks. Existing methods fail to effectively integrate historical fault data with real-time environmental parameters, making it impossible to dynamically adjust thresholds. Furthermore, relying on single sensor data, they struggle to capture the collaborative degradation characteristics of multiple parameters.

[0003] A similar prior art includes a Chinese patent application with publication number CN119829957A, which discloses a method and system for analyzing equipment fault detection data for digital factories, including S1: using high-precision clock synchronization technology to synchronously collect comprehensive maintenance data of equipment; S2: using the location data of the equipment as graph nodes, and the functional associations and data interaction data between equipment as graph edges to construct a graph neural network, and constructing an equipment association model based on the graph neural network, using the status data of the equipment as the input of the equipment association model, and outputting a status association factor; S3: using the comprehensive maintenance data and the status association factor as the state input of the reinforcement learning agent, and using the Q learning algorithm to train the agent, inputting the real-time status data into the agent, outputting the status score, and setting an abnormal status threshold. When the status score is lower than the abnormal status threshold, the corresponding real-time data is marked as abnormal. A similar prior art includes a Chinese patent application with publication number CN119826908A, which discloses a method and device for detecting abnormal operation status of a crane pipe. The method includes: collecting crane pipe operation parameters in real time during crane pipe operation; fusing the operation parameters to obtain fused data; inputting the fused data into a pre-trained anomaly detection model, obtaining a reconstruction error based on the output of the anomaly detection model; and determining whether the crane pipe operation status is abnormal based on the reconstruction error.

[0004] However, both of the above technical solutions do not consider the dynamic impact of environmental parameters (such as temperature and wind speed) on the equipment status, and there are still deficiencies in the accuracy of the evaluation. Summary of the Invention

[0005] In order to solve the above technical problems, the present application provides a method and system for risk assessment of the operating status of a wind turbine generator set, so as to improve the accuracy of risk assessment.

[0006] In a first aspect, the present application provides a method for risk assessment of the operating status of a wind turbine generator set, wherein sensors are installed on various components of the wind turbine generator set, and the method comprises:

[0007] Step S1: constructing a risk assessment model based on a neural network model, collecting and training the risk assessment model based on historical first operating data and corresponding status data of the wind turbine generator set under normal operating conditions;

[0008] Step S2: acquiring real-time operating data of the wind turbine generator set and also acquiring measured state data corresponding to the real-time operating data, inputting the real-time operating data into the risk assessment model, and the risk assessment model outputting predicted state data corresponding to the real-time operating data;

[0009] Step S3: Obtain a difference reference value corresponding to the current environmental data, calculate and compare the difference between the predicted state data and the measured state data with the difference reference value, and determine the operating state of the wind turbine generator set, wherein the operating state includes a normal operating state, a low-risk operating state, and a high-risk operating state.

[0010] In combination with the first aspect, in a first implementation of the first aspect of the present application, in step S3, obtaining a difference reference value corresponding to the current environmental data includes:

[0011] collecting historical second operating data and corresponding status data of the wind turbine generator set under abnormal operating conditions, grouping the status data corresponding to the historical second operating data according to environmental data in the historical second operating data to obtain a plurality of environmental groups, clustering the status data in each of the environmental groups, and obtaining and calculating, based on the clustering results, a difference reference value corresponding to the environmental group;

[0012] Based on the environmental grouping and the difference reference value corresponding to the environmental grouping, a mapping table of environmental grouping and difference reference value is established, the current environmental data is matched with the environmental grouping, and the corresponding difference reference value is obtained from the mapping table of environmental grouping and difference reference value based on the matched environmental grouping.

[0013] In combination with the first aspect, in a second implementation of the first aspect of the present application, collecting historical second operating data and corresponding status data of the wind turbine generator set in an abnormal operating state includes:

[0014] Based on the fault records in the fault record table of the wind turbine generator set, operating data of the entire wind turbine generator set is divided into operating data of multiple stages, and from the operating data of each stage, operating data and corresponding status data within a preset time period before a second time point are obtained as target operating data and target status data, and a first time point is determined within the preset time period based on the target status data, and the first time point is used as a time demarcation point between a normal operating state and an abnormal operating state of the wind turbine generator set;

[0015] The operating data from each fault recovery to the first time point is used as the first-stage operating data, and the operating data from the first time point to the second time point is used as the second-stage operating data. Multiple first-stage operating data constitute historical first operating data, and status data corresponding to the historical first operating data is also obtained. Multiple second-stage operating data constitute historical second operating data, and the historical second operating data and corresponding status data are also obtained, wherein the second time point is the start time of any fault in the wind turbine fault record table.

[0016] In combination with the first aspect, in a third implementation of the first aspect of the present application, calculating the difference reference value corresponding to the environmental grouping includes:

[0017] Obtain any parameter from the status data in any one of the environmental groups as the third parameter, cluster the values of the third parameter to obtain multiple clusters and the standard deviation corresponding to each cluster, calculate the standard deviation corresponding to each cluster and multiply it by a preset safety factor to obtain a difference reference value corresponding to each cluster, calculate the average difference reference value of the multiple clusters and use it as the difference reference value of the third parameter corresponding to the environmental group, repeat this step to calculate the difference reference values of other parameters corresponding to the environmental group, and the difference reference value of each parameter in the status data in the environmental group constitutes the difference reference value corresponding to the environmental group.

[0018] In combination with the first aspect, in a fourth implementation of the first aspect of the present application, determining the first time point within the preset time period based on the target state data includes:

[0019] Calculate a third time point corresponding to the first parameter in the target state data and a fourth time point corresponding to the second parameter in the target state data; if the third time point is less than or equal to the fourth time point, use the third time point as the first time point; if the third time point is greater than the fourth time point, use the fourth time point as the first time point.

[0020] In combination with the first aspect, in a fifth implementation of the first aspect of the present application, calculating the third time point corresponding to the first parameter in the target state data and the fourth time point corresponding to the second parameter in the target state data includes:

[0021] Taking any parameter in the target state data as a first parameter, calculating a first average value of the first parameter within the preset time period, calculating the difference between the values of the first parameter corresponding to all time points within the preset time period and the first average value as a first difference, obtaining and selecting a time point corresponding to the first parameter having the largest difference from the plurality of first differences as a third time point;

[0022] Any parameter other than the first parameter in the target state data is used as the second parameter, and the above steps are repeated to calculate the fourth time point corresponding to the second parameter in the target state data.

[0023] In combination with the first aspect, in a sixth implementation of the first aspect of the present application, the setting of the preset safety factor includes:

[0024] Collect the number of misjudgments of low-risk or high-risk operating states in the past N months. If there are M consecutive misjudgments, the safety factor is increased; otherwise, it remains unchanged.

[0025] If a high-risk operating state is missed once in the past N months, the safety factor will be lowered; otherwise, it will remain unchanged, where N represents a positive integer between 1 and 12, and M represents a positive integer greater than or equal to 2 and less than or equal to 5.

[0026] In combination with the first aspect, in a seventh implementation of the first aspect of the present application, determining the operating status of the wind turbine generator set includes:

[0027] If the difference between the measured value and the predicted value of each parameter in the status data is less than or equal to the difference reference value corresponding to the parameter, it is determined that the wind turbine generator set is in a normal operating state and the current monitoring cycle of the wind turbine generator set is continued;

[0028] If the difference between the measured value and the predicted value of any parameter in the status data is greater than the difference reference value corresponding to the parameter and is less than or equal to twice the difference reference value corresponding to the parameter, it is determined that the wind turbine generator set is in a low-risk operating state and the maintenance process is initiated;

[0029] If the difference between the measured values and the predicted values of all parameters in the status data is greater than the difference reference value corresponding to the parameter or the difference value of any one parameter is greater than twice the difference reference value corresponding to the parameter, the wind turbine generator set is determined to be in a high-risk operating state, the wind turbine generator set is controlled to shut down immediately, and an emergency maintenance instruction is sent to the person in charge.

[0030] In conjunction with the first aspect, in an eighth implementation of the first aspect of the present application, calculating the difference between the predicted state data and the measured state data includes:

[0031] Any parameter in the state data is used as the fourth parameter, and the formula Calculate the difference value of the fourth parameter, where λ represents the difference value of the fourth parameter, A′ represents the predicted value of the fourth parameter, and A represents the measured value of the fourth parameter; repeat this step to calculate the difference between the measured value and the predicted value of all parameters in the state data.

[0032] In a second aspect, the present application provides a system for assessing the operating status risk of a wind turbine generator set, wherein sensors are installed on various components of the wind turbine generator set, and the system comprises:

[0033] a model building unit, configured to build a risk assessment model based on a neural network model, and to collect and train the risk assessment model based on historical first operating data and corresponding status data of the wind turbine generator set under normal operating conditions;

[0034] a data acquisition unit, configured to acquire real-time operating data of the wind turbine generator set and also acquire measured state data corresponding to the real-time operating data, input the real-time operating data into the risk assessment model, and the risk assessment model output predicted state data corresponding to the real-time operating data;

[0035] A risk assessment unit is used to obtain a difference reference value corresponding to the current environmental data, calculate and compare the difference between the predicted state data and the measured state data with the difference reference value, and determine the operating state of the wind turbine generator set, where the operating state includes a normal operating state, a low-risk operating state, and a high-risk operating state.

[0036] Compared with the prior art, the beneficial effects of the present invention are at least as follows:

[0037] In the technical solution provided by this application, the risk assessment model constructed based on the neural network model can efficiently learn the multi-parameter correlation characteristics under normal operating conditions, such as the nonlinear relationship between gearbox temperature and spindle speed and ambient wind speed, and reduce prediction errors. By collecting operating data in real time and comparing it with the model prediction value, it can quickly identify abnormal fluctuations and achieve the goal of unnecessary shutdown. The calculation of dynamic difference reference values is achieved through environmental grouping and cluster analysis, and the standard deviation and safety factor are combined to generate difference reference values. Compared with fixed thresholds, this method can reduce the false alarm rate. At the same time, the adaptive adjustment mechanism of the safety factor further optimizes the sensitivity of this method, ensuring that omissions and misjudgments are reduced with the dynamic changes of operating data.

[0038] This application also captures equipment degradation characteristics by integrating multi-dimensional data such as temperature, vibration, and power generation. When gusts of wind cause instantaneous load fluctuations, it can quickly identify and filter short-term vibration anomalies to avoid false triggering of shutdowns. In extreme working conditions, such as low temperatures of -20°C or high humidity environments at sea, through environmental grouping and transfer learning frameworks, the model can automatically match the difference reference values of similar historical scenarios. Improve the accuracy of bearing low-temperature embrittlement risk warnings. In summary, this application can effectively improve the accuracy of risk assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 This is a schematic diagram of an embodiment of a method for risk assessment of the operating status of a wind turbine generator set in an embodiment of the present application;

[0041] Figure 2 Schematic diagram of the clustering results of the gearbox temperature in the embodiment of the present application;

[0042] Figure 3 This is a schematic diagram of defining the historical first operating data and the historical second operating data in an embodiment of the present application;

[0043] Figure 4 Schematic diagram of an embodiment of an operating status risk assessment system for a wind turbine generator set in an embodiment of the present application. DETAILED DESCRIPTION

[0044] The embodiments of the present application provide a method and system for assessing the operating status risks of wind turbines. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices.

[0045] Example 1:

[0046] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In one embodiment of the present application, a method for risk assessment of the operating status of a wind turbine generator set includes: installing sensors on various components of the wind turbine generator set, and performing the following steps:

[0047] Step S1: Constructing a risk assessment model based on a neural network model, and training the risk assessment model based on historical first operating data and corresponding status data collected when the wind turbine is in normal operation. The operating data refers to data collected by sensors when the wind turbine is in normal operation, including at least gearbox bearing pressure, main shaft rotation speed, ambient wind speed, and ambient temperature. The status data includes at least gearbox temperature and main bearing vibration amplitude, and may also include power generation, etc.

[0048] Specifically, by collecting historical operating data and corresponding status data of wind turbines under normal operating conditions, a neural network model is used for training to construct a risk assessment model. This risk assessment model can learn the data distribution characteristics and patterns under normal operating conditions, capture the nonlinear relationship between key parameters, and provide a basis for subsequent status prediction. Through training with a large amount of historical data, the model can effectively identify the status data characteristics under normal operating conditions, accurately characterize the status of wind turbines, and improve the prediction accuracy of the risk assessment model. The historical data used here refers to the first historical operating data. How to obtain the first historical operating data will be described in detail below.

[0049] Step S2: Acquire real-time operating data of the wind turbine generator set, and also acquire measured state data corresponding to the real-time operating data, input the real-time operating data into the risk assessment model, and the risk assessment model outputs predicted state data corresponding to the real-time operating data.

[0050] Specifically, real-time operating data of the wind turbine is collected and fed into a trained risk assessment model for prediction. Simultaneously, measured status data corresponding to the real-time operating data is obtained and compared with the predicted status data to assess the wind turbine's operating status. The risk assessment model outputs predicted status data based on the real-time operating data, combined with trained features and patterns. By comparing the predicted data with the measured data, the difference is calculated and combined with a dynamic difference reference value to determine the status.

[0051] Step S3: Obtain the difference reference value corresponding to the current environmental data, calculate and compare the difference between the predicted state data and the measured state data with the difference reference value, and determine the operating state of the wind turbine generator set. The operating state includes normal operating state, low-risk operating state and high-risk operating state.

[0052] Specifically, the calculation of the dynamic difference reference value is the core of this application. The difference between the predicted state data and the measured state data is calculated and compared with the difference reference value. Specifically, the difference value of each parameter is first calculated, and the operating state of the wind turbine is judged according to the setting of the difference reference value. The normal operating state means that the difference values of all parameters are within the allowable range; the low-risk operating state means that the difference values of certain parameters exceed the allowable range but do not reach the dangerous level; the high-risk operating state means that the difference values significantly exceed the allowable range, which will cause serious operating risks. Through difference analysis, the operating state of the wind turbine can be accurately judged, and a clear risk level assessment result can be provided. At the same time, the accuracy and adaptability of risk assessment can be improved.

[0053] In the embodiment of the present application, the accuracy of risk assessment can be improved through the coordination between the above steps.

[0054] In a specific embodiment, executing step S3 to obtain the difference reference value corresponding to the current environmental data specifically includes the following steps:

[0055] The historical second operating data and corresponding status data of the wind turbine generator set under abnormal operating conditions are collected, and the status data corresponding to the historical second operating data are grouped according to the environmental data in the historical second operating data to obtain multiple environmental groups. The status data in each environmental group is clustered, and the difference reference value corresponding to the environmental group is obtained and calculated based on the clustering result. Based on the environmental group and the difference reference value corresponding to the environmental group, a mapping table between the environmental group and the difference reference value is established, and the current environmental data is matched with the environmental group. Based on the matched environmental group, the corresponding difference reference value is obtained from the mapping table between the environmental group and the difference reference value. The current environmental data includes the current ambient temperature and the current ambient wind speed.

[0056] Specifically, in traditional operating status assessment methods, reference values are typically fixed, which can lead to many false positives when external environmental data changes. Therefore, it is necessary to analyze operating and status data in conjunction with environmental data to avoid misjudgments based on a single reference value under different operating conditions. First, historical secondary operating data is collected and grouped by environmental data. The specifics of this historical secondary operating data are described in detail below. For example, a wind farm collected 100 instances of operating data before a gearbox overheating failure. This historical secondary operating data includes the following parameters: environmental data: temperature (°C), wind speed (m / s); and status data: gearbox temperature (°C), main bearing vibration amplitude (mm / s), and power generation (kW). Based on the environmental data in the historical secondary operating data, the data is divided into multiple groups: Group A: High temperature, low wind speed (temperature ≥ 20°C, wind speed ≤ 5m / s); Group B: Normal temperature, medium wind speed (temperature 5°C-15°C, wind speed 6-10m / s); and Group C: Low temperature, high wind speed (temperature ≤ 0°C, wind speed ≥ 12m / s). According to the actual operation of the wind turbine, more groups will be divided.

[0057] Next, cluster one parameter in the state data within the environmental group. Take group A with high temperature and low wind speed as an example. The data sample includes 20 operating data before the fault. The gearbox temperature range is 70℃~95℃. Using the K-means algorithm in the clustering method, set K=3, the gearbox temperature is clustered. The clustering results are shown in Figure 2 Based on the clustering results, the difference reference value corresponding to each cluster is calculated. The specific method for calculating the difference reference value corresponding to each cluster will be described below. The average of the difference reference values corresponding to multiple clusters is used as the difference reference value corresponding to group A. Similarly, the difference reference values corresponding to groups B and C are calculated respectively. The specific method is described below. Through the joint analysis of environmental data and status data, it is possible to effectively avoid the misjudgment of fixed reference values under different working conditions.

[0058] In a specific embodiment, collecting historical second operating data and corresponding status data of a wind turbine generator set in an abnormal operating state specifically includes the following steps:

[0059] Based on the fault records in the wind turbine generator set fault record table, the operating data of the entire wind turbine generator set is divided into multiple stages of operating data. From the operating data of each stage, the operating data and corresponding status data within a preset time period before the second time point are obtained as target operating data and target status data. Based on the target status data, a first time point is determined from the preset time period, and the first time point is used as the time dividing point between the normal operating state and the abnormal operating state of the wind turbine generator set; the operating data from each fault recovery to the first time point is used as the first stage operating data, and the operating data from the first time point to the second time point is used as the second stage operating data. Multiple first stage operating data constitute historical first operating data, and the status data corresponding to the historical first operating data is also obtained. Multiple second stage operating data constitute historical second operating data, and the historical second operating data and corresponding status data are also obtained, wherein the second time point is the start time of any fault in the wind turbine generator set fault record table.

[0060] Specifically, see Figure 3 As shown in the figure, taking a gearbox overheating fault as an example, the second time point of the fault onset is 2024-05-10 14:00, and the preset analysis window is 72 hours before the fault. Because the values of various parameters in the status data fluctuate irregularly before the fault occurs, by analyzing these fluctuations, the first time demarcation point can be determined as 14:00 on May 7th. At this time, the gearbox temperature suddenly rose from the normal 75°C to 85°C, then dropped to 80°C, possibly fluctuating repeatedly. Even with stable wind speeds, cooling efficiency decreased significantly, which should be of concern and analyzed. The first phase data of multiple fault events constitute the historical first operating dataset, representing normal operating data. The second phase data of multiple fault events constitute the historical second operating dataset, representing abnormal operating data. By clearly separating normal and abnormal data, it is possible to distinguish between equipment degradation characteristics and true fault precursors during model training. Through dynamic time division, we can avoid misjudging short-term fluctuations as faults and capture irregular anomalies, realizing the transformation from "post-maintenance" to "predictive maintenance".

[0061] In a specific embodiment, calculating the difference reference value corresponding to the environmental grouping includes: obtaining any parameter from the status data in any environmental grouping as the third parameter, clustering the values of the third parameter, obtaining multiple clusters and the standard deviation corresponding to each cluster, calculating the standard deviation corresponding to each cluster and multiplying it by a preset safety factor to obtain the difference reference value corresponding to each cluster, calculating the average difference reference value of the multiple clusters and using it as the difference reference value of the third parameter corresponding to the environmental grouping, repeating this step, calculating the difference reference values of other parameters corresponding to the environmental grouping, and the difference reference value of each parameter in the status data in the environmental grouping constitutes the difference reference value corresponding to the environmental grouping.

[0062] Specifically, the calculation process of the difference reference value is explained by taking the high temperature and low wind speed environment group (temperature ≥ 20℃, wind speed ≤ 5m / s) as an example: the gearbox temperature is selected as the third parameter. This group contains 100 historical data with a temperature range of 70℃ to 95℃. The temperature data are clustered using the K-means algorithm (K=3), see Figure 2 As shown in the figure, when calculating the cluster-level reference difference value, assuming a safety factor of 1.5, the reference difference value for cluster 1 is 3°C × 1.5 = 4.5°C, the reference difference value for cluster 2 is 4°C × 1.5 = 6.0°C, and the reference difference value for cluster 3 is 2°C × 1.0 = 2.0°C. Taking the average of the clusters—(4.5 + 6.0 + 2.0) / 3—≈ 4.2°C, the reference difference value for the gearbox temperature in this environmental grouping is ±4.2°C. In other words, any difference between the predicted and measured gearbox temperatures within ±4.2°C indicates that the wind turbine's gearbox temperature is normal. The same process is also used to calculate the reference difference value for the main bearing vibration amplitude in this environmental grouping. When the measured gearbox temperature is 5°C higher than the predicted value (exceeding 4.2°C), the wind turbine is deemed to be in a low-risk operating state rather than a high-risk state, reducing false shutdowns by 50%. In high-temperature environments, the tolerance for temperature fluctuations is improved compared to the traditional fixed threshold of ±3°C, avoiding frequent false alarms caused by fluctuations in heat dissipation efficiency. The above method not only retains the data distribution characteristics through the weighted average of cluster-level difference reference values, but also dynamically adjusts the sensitivity of the evaluation through the safety factor, thereby achieving refined status monitoring in complex environments.

[0063] In a specific embodiment, determining a first time point within a preset time period based on target state data includes: calculating a third time point corresponding to a first parameter in the target state data and a fourth time point corresponding to a second parameter in the target state data; if the third time point is less than or equal to the fourth time point, taking the third time point as the first time point; and if the third time point is greater than the fourth time point, taking the fourth time point as the first time point.

[0064] Specifically, the first parameter can be the gearbox temperature, and the second parameter can be the main bearing vibration amplitude. The status data can also include other parameters such as power generation, which are not limited in this application. By cross-validating the multi-parameter abnormal time series, the earliest possible abnormal signal can be captured, avoiding the lagging fluctuations of relying on a single parameter. The definition of the third and fourth time points will be described below.

[0065] In a specific embodiment, calculating a third time point corresponding to a first parameter in the target state data and a fourth time point corresponding to a second parameter in the target state data includes:

[0066] Take any parameter in the target state data as the first parameter, calculate the first average value of the first parameter within a preset time period, calculate the difference between the values of the first parameter corresponding to all time points within the preset time period and the first average value as the first difference, obtain and select the time point corresponding to the first parameter with the largest difference from the multiple first differences as the third time point; take any parameter other than the first parameter in the target state data as the second parameter, repeat the above steps, and calculate the fourth time point corresponding to the second parameter in the target state data.

[0067] Specifically, for example, analyzing data from the 72 hours before a failure, the temperature range during the preset time period was 70°C to 92°C. The vibration amplitude during the same time period ranged from 1.5 mm / s to 5.0 mm / s. The average temperature over the 72 hours was calculated to be 80°C. The absolute difference between the temperature and the average was then calculated at each time point. It was found that at 12:00 PM on the second day, the temperature suddenly rose to 90°C. The difference was 10°C, the maximum gearbox temperature difference during the preset time period. Therefore, 12:00 PM on the second day was marked as the time point of the temperature anomaly, the third time point. Using the same data from the 72 hours before the failure, the average vibration value was calculated to be 2.5 mm / s. At 1:00 PM on the third day, the vibration suddenly increased to 4.5 mm / s. The difference was 2.0 mm / s, the maximum vibration amplitude difference during the preset time period. Therefore, 1:00 PM on the third day was marked as the time point of the vibration amplitude anomaly, the fourth time point. This method allows the earliest abnormal time point to be used as the benchmark when defining normal and abnormal operating conditions, avoiding missing potential risks. At the same time, the earliest abnormal signal is located through the maximum deviation between the parameter and the mean, avoiding reliance on a single threshold or fixed rule, and providing a basis for calculating the dynamic difference reference value.

[0068] In a specific embodiment, the setting of the preset safety factor includes: collecting the number of misjudgments of low-risk operating status or high-risk operating status in the past N months, and if there are M consecutive misjudgments, the safety factor is increased, otherwise it remains unchanged; if a high-risk operating status is missed once in the past N months, the safety factor is decreased, otherwise it remains unchanged, where N represents a positive integer between 1 and 12, and M represents a positive integer greater than or equal to 2 and less than or equal to 5.

[0069] Specifically, the initial parameters were set with a safety factor of 1.5, an evaluation period of N = 3 months, and M = 3 consecutive misjudgments. During the high temperatures of summer, dust accumulation on the heat sink caused occasional temperature fluctuations. For example, from July to September, 75°C (the upper limit of normal) was mistakenly judged as high risk for three consecutive times. The threshold was ±4.5°C, and the actual measurement was 79.5°C). Then the safety factor was raised from 1.5 to 1.6, and the reference value of the difference was expanded to ±4.8°C = 3°C × 1.6. Through subsequent operation observations, the misjudgment rate dropped from 15% to 5%, reducing unnecessary shutdown inspections. At low temperatures in winter, lubricating oil viscosity can occur, masking early wear. For example, in December, a bearing wear warning was not issued in time. The measured vibration difference value was 1.8mm / s, the reference value of the difference was 1.5mm / s, and the actual required value was 1.3mm / s. The safety factor was then lowered from 1.5 to 1.4, and the vibration difference reference value was tightened to ±1.4mm / s = 1.0mm / s × 1.4. Subsequent operational observations revealed two similar anomalies the following month, shortening maintenance response time by 40%. Continuous false positives indicate that the difference reference value is too strict. Raising the factor to relax the limit (e.g., ±4.5°C → ±4.8°C) avoids the "crying wolf" effect. A single missed alarm indicates that the threshold is too loose. Lowering the factor to increase sensitivity (e.g., 1.5mm / s → ±1.4mm / s) blocks hidden risks. By dynamically adjusting the safety factor, we can precisely match environmental changes and avoid false positives or missed alarms.

[0070] In a specific embodiment, the step S3 of determining the operating status of the wind turbine generator set includes:

[0071] If the difference between the measured value and the predicted value of each parameter in the status data is less than or equal to the difference reference value corresponding to the parameter, it is determined that the wind turbine generator set is in normal operation and the current monitoring cycle of the wind turbine generator set is continued;

[0072] If the difference between the measured value and the predicted value of any parameter in the status data is greater than the difference reference value corresponding to the parameter, and less than or equal to twice the difference reference value corresponding to the parameter, the wind turbine generator set is determined to be in a low-risk operating state and the maintenance process is initiated;

[0073] If the difference between the measured values and the predicted values of all parameters in the status data is greater than the difference reference value corresponding to the parameter, or the difference value of any parameter is greater than twice the difference reference value corresponding to the parameter, the wind turbine is judged to be in a high-risk operating state, the wind turbine is controlled to shut down immediately, and an emergency maintenance instruction is sent to the person in charge.

[0074] Specifically, for each parameter in the status data, such as gearbox temperature, main bearing vibration amplitude, and power generation, the difference between the measured and predicted values is calculated and compared with the corresponding reference difference value for that parameter under the current environmental data. The reference difference value is determined based on the wind turbine's historical secondary operating data and is used to measure the allowable deviation range of the parameter. The method for setting the reference difference value has been described in detail above. Secondly, the wind turbine's operating status is determined based on the comparison of the difference between the predicted and measured values and the reference difference value. Normal operation refers to the situation where the difference value for each parameter is within the preset reference difference value range (i.e., difference value ≤ reference difference value), and the wind turbine is deemed to be operating normally. The current monitoring cycle continues to monitor the equipment status. Low-risk operation refers to the situation where the difference value for any parameter exceeds the reference difference value but does not exceed twice its value (i.e., difference value > reference difference value and ≤ 2 × reference difference value), and the wind turbine is deemed to be in a low-risk state. At this point, the preventive maintenance process is initiated to prevent the potential fault from further deteriorating. In a high-risk operating state, if the difference values of all parameters exceed the difference reference value, or the difference value of any parameter exceeds twice the difference reference value, that is, the difference value is >2×difference reference value, the wind turbine is determined to be in a high-risk state and emergency measures are taken immediately, including controlling the shutdown of the wind turbine and sending emergency maintenance instructions to relevant persons in charge so that the problem can be investigated and repaired as soon as possible.

[0075] By analyzing the difference in each parameter individually and comparing it with a reference value, the operating status of the wind turbine can be accurately determined, avoiding misjudgments caused by a single parameter anomaly. Wind turbine operating risks are also classified into different levels and corresponding countermeasures are implemented, ensuring timely resolution of potential issues and avoiding unnecessary over-maintenance.

[0076] In a specific embodiment, calculating the difference between the predicted state data and the measured state data includes: taking any parameter in the state data as the fourth parameter, and using the formula Calculate the difference value of the fourth parameter, where λ represents the difference value of the fourth parameter, A′ represents the predicted value of the fourth parameter, and A represents the measured value of the fourth parameter; repeat this step to calculate the difference between the measured value and the predicted value of all parameters in the state data.

[0077] Specifically, by calculating the difference values of each parameter in the status data respectively, it is possible to identify which parameters have large deviations, and then by comparing them with the corresponding difference reference values, provide a basis for evaluating the operating status of the wind turbine generator set.

[0078] Example 2:

[0079] The above describes the operating status risk assessment method for a wind turbine generator set in the embodiment of the present application. The following describes the operating status risk assessment system for a wind turbine generator set in the embodiment of the present application. Figure 4 In one embodiment of the present application, a system for risk assessment of the operating status of a wind turbine generator set includes installing sensors on various components of the wind turbine generator set and implementing the following modules:

[0080] A model building unit is used to build a risk assessment model based on a neural network model, and to collect and train the risk assessment model based on historical first operating data and corresponding status data of the wind turbine generator set under normal operating conditions;

[0081] A data acquisition unit is used to obtain real-time operating data of the wind turbine generator set and also obtain measured state data corresponding to the real-time operating data, input the real-time operating data into the risk assessment model, and the risk assessment model outputs predicted state data corresponding to the real-time operating data;

[0082] The risk assessment unit is used to obtain the difference reference value corresponding to the current environmental data, calculate and compare the difference between the predicted state data and the measured state data with the difference reference value, and determine the operating state of the wind turbine generator set. The operating state includes normal operating state, low-risk operating state and high-risk operating state.

[0083] The accuracy of risk assessment can be improved through the coordinated cooperation of the above components.

[0084] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0085] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0086] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for risk assessment of the operating status of a wind turbine generator set, wherein sensors are installed on various components of the wind turbine generator set, characterized in that: The method comprises: Step S1: constructing a risk assessment model based on a neural network model, collecting and training the risk assessment model based on historical first operating data and corresponding status data of the wind turbine generator set under normal operating conditions; Step S2: acquiring real-time operating data of the wind turbine generator set and also acquiring measured state data corresponding to the real-time operating data, inputting the real-time operating data into the risk assessment model, and the risk assessment model outputting predicted state data corresponding to the real-time operating data; Step S3: Obtain a difference reference value corresponding to the current environmental data, calculate and compare the difference between the predicted state data and the measured state data with the difference reference value, and determine the operating state of the wind turbine generator set, wherein the operating state includes a normal operating state, a low-risk operating state, and a high-risk operating state.

2. The method according to claim 1, characterized in that In step S3, obtaining the difference reference value corresponding to the current environmental data includes: collecting historical second operating data and corresponding status data of the wind turbine generator set under abnormal operating conditions, grouping the status data corresponding to the historical second operating data according to environmental data in the historical second operating data to obtain a plurality of environmental groups, clustering the status data in each of the environmental groups, and obtaining and calculating, based on the clustering results, a difference reference value corresponding to the environmental group; Based on the environmental grouping and the difference reference value corresponding to the environmental grouping, a mapping table of environmental grouping and difference reference value is established, the current environmental data is matched with the environmental grouping, and the corresponding difference reference value is obtained from the mapping table of environmental grouping and difference reference value based on the matched environmental grouping.

3. The method according to claim 2, characterized in that Collecting historical second operating data and corresponding status data of the wind turbine generator set in an abnormal operating state, including: Based on the fault records in the fault record table of the wind turbine generator set, operating data of the entire wind turbine generator set is divided into operating data of multiple stages, and from the operating data of each stage, operating data and corresponding status data within a preset time period before a second time point are obtained as target operating data and target status data, and a first time point is determined within the preset time period based on the target status data, and the first time point is used as a time demarcation point between a normal operating state and an abnormal operating state of the wind turbine generator set; The operating data from each fault recovery to the first time point is used as the first-stage operating data, and the operating data from the first time point to the second time point is used as the second-stage operating data. Multiple first-stage operating data constitute historical first operating data, and status data corresponding to the historical first operating data is also obtained. Multiple second-stage operating data constitute historical second operating data, and the historical second operating data and corresponding status data are also obtained, wherein the second time point is the start time of any fault in the wind turbine fault record table.

4. The method according to claim 2, characterized in that Calculating the difference reference values corresponding to the environmental groups includes: Obtain any parameter from the status data in any one of the environmental groups as the third parameter, cluster the values of the third parameter to obtain multiple clusters and the standard deviation corresponding to each cluster, calculate the standard deviation corresponding to each cluster and multiply it by a preset safety factor to obtain a difference reference value corresponding to each cluster, calculate the average difference reference value of the multiple clusters and use it as the difference reference value of the third parameter corresponding to the environmental group, repeat this step to calculate the difference reference values of other parameters corresponding to the environmental group, and the difference reference value of each parameter in the status data in the environmental group constitutes the difference reference value corresponding to the environmental group.

5. The method according to claim 3, characterized in that Determining a first time point within the preset time period based on the target state data includes: Calculate a third time point corresponding to the first parameter in the target state data and a fourth time point corresponding to the second parameter in the target state data; if the third time point is less than or equal to the fourth time point, use the third time point as the first time point; if the third time point is greater than the fourth time point, use the fourth time point as the first time point.

6. The method according to claim 5, characterized in that Calculating a third time point corresponding to the first parameter in the target state data and a fourth time point corresponding to the second parameter in the target state data includes: Taking any parameter in the target state data as a first parameter, calculating a first average value of the first parameter within the preset time period, calculating the difference between the values of the first parameter corresponding to all time points within the preset time period and the first average value as a first difference, obtaining and selecting a time point corresponding to the first parameter having the largest difference from the plurality of first differences as a third time point; Any parameter other than the first parameter in the target state data is used as the second parameter, and the above steps are repeated to calculate the fourth time point corresponding to the second parameter in the target state data.

7. The method according to claim 4, characterized in that Preset safety factor settings, including: The number of misjudgments of low-risk or high-risk operating states in the past N months is collected. If there are M consecutive misjudgments, the safety factor is adjusted upward, otherwise it remains unchanged; if a high-risk operating state is missed once in the past N months, the safety factor is adjusted downward, otherwise it remains unchanged.

8. The method according to claim 1, characterized in that Determining the operating status of the wind turbine generator set includes: If the difference between the measured value and the predicted value of each parameter in the status data is less than or equal to the difference reference value corresponding to the parameter, it is determined that the wind turbine generator set is in a normal operating state and the current monitoring cycle of the wind turbine generator set is continued; If the difference between the measured value and the predicted value of any parameter in the status data is greater than the difference reference value corresponding to the parameter and is less than or equal to twice the difference reference value corresponding to the parameter, it is determined that the wind turbine generator set is in a low-risk operating state and the maintenance process is initiated; If the difference between the measured values and the predicted values of all parameters in the status data is greater than the difference reference value corresponding to the parameter or the difference value of any one parameter is greater than twice the difference reference value corresponding to the parameter, the wind turbine generator set is determined to be in a high-risk operating state, the wind turbine generator set is controlled to shut down immediately, and an emergency maintenance instruction is sent to the person in charge.

9. The method according to claim 1, characterized in that Calculating the difference between the predicted state data and the measured state data includes: Any parameter in the state data is used as the fourth parameter, and the formula Calculate the difference value of the fourth parameter, where λ represents the difference value of the fourth parameter, A′ represents the predicted value of the fourth parameter, and A represents the measured value of the fourth parameter; repeat this step to calculate the difference between the measured value and the predicted value of all parameters in the state data.

10. A system for assessing the operating status risk of a wind turbine generator set, for implementing the method for assessing the operating status risk of a wind turbine generator set according to any one of claims 1 to 9, wherein sensors are installed on various components of the wind turbine generator set, and wherein: The system comprises: a model building unit, configured to build a risk assessment model based on a neural network model, and to collect and train the risk assessment model based on historical first operating data and corresponding status data of the wind turbine generator set under normal operating conditions; a data acquisition unit, configured to acquire real-time operating data of the wind turbine generator set and also acquire measured state data corresponding to the real-time operating data, input the real-time operating data into the risk assessment model, and the risk assessment model output predicted state data corresponding to the real-time operating data; A risk assessment unit is used to obtain a difference reference value corresponding to the current environmental data, calculate and compare the difference between the predicted state data and the measured state data with the difference reference value, and determine the operating state of the wind turbine generator set, where the operating state includes a normal operating state, a low-risk operating state, and a high-risk operating state.

Images