Fan temperature measurement data processing method and system

Abstract

The invention relates to the technical field of wind generating set temperature measurement data processing, in particular to a fan temperature measurement data processing method and system. The method comprises the following steps: acquiring real-time active power data and real-time temperature data; time alignment processing is carried out, a power temperature synchronization data sequence is constructed, a dynamic response working condition is identified, a dynamic response power input sequence and a dynamic response temperature output sequence are extracted, dynamic response characteristic parameters are calculated, and reference values of the dynamic response characteristic parameters are recorded; constructing a dynamic response characteristic parameter sequence; judging whether the dynamic response characteristic parameter sequence is in a change mode or not; after the benign response change is identified, updating the reference value of the dynamic response characteristic parameter; and judging whether the dynamic response characteristic parameter sequence is in a change mode, and identifying a real fault. The method has the advantage of solving the problems of misjudgment, unnecessary operation and maintenance cost and reduced credibility of a diagnosis system caused by difficulty in distinguishing benign response change and real equipment fault in the existing system.

Description

Technical Field

[0001] The present application relates to the technical field of wind turbine temperature measurement data processing, and in particular to a method and system for processing wind turbine temperature measurement data. Background Art

[0002] Modern wind power generation technology plays an increasingly important role in ensuring the stable operation of power grids. To meet the grid's rapid peak-shaving needs, wind turbines must frequently and drastically adjust their power output. This process causes dramatic temperature fluctuations in key components within the wind turbine, placing higher demands on the accuracy and reliability of temperature measurement systems. However, during actual operation and maintenance, even minor changes in the hardware characteristics of the temperature measurement signal transmission link, such as replacing a new signal cable, can cause subtle distortions in the dynamic characteristics of the temperature signal.

[0003] Existing fan temperature measurement data processing systems often use diagnostic logic calibrated based on legacy hardware characteristics, making it difficult to distinguish signal changes caused by benign hardware changes from actual equipment failures. For example, when a signal cable in the temperature measurement signal transmission link is replaced with a new model, even though the new cable itself performs well, its inherent electrical characteristics may cause systematic deviations in the dynamic response characteristics of the temperature signal (such as the rate of change or the acceleration of the rate of change). This deviation causes the received signal to be inconsistent with the "healthy" model established based on the old cable data, resulting in a misinterpretation of the signal as "abnormal sensor response" or "abnormal signal transmission link."

[0004] This misjudgment led the O&M team to frequently and unnecessarily conduct on-site inspections and troubleshooting, resulting in costly turbine downtime and wasted spare parts and human resources. More seriously, these recurring "false fault" alarms, which could not be resolved through conventional means, severely undermined the O&M team's trust in the diagnostic system and could even lead to system functionality being downgraded or blocked, reducing turbine O&M efficiency and defeating the original purpose of introducing the intelligent predictive system.

[0005] In view of the above problems, the existing technology is in urgent need of improvement. Summary of the Invention

[0006] The present application discloses a fan temperature measurement data processing method and system, which aims to solve the problem that the existing fan temperature measurement data processing system has difficulty in distinguishing between benign response changes caused by hardware characteristic changes and real equipment failures, thereby leading to misjudgment, unnecessary operation and maintenance costs and reduced trust in the diagnostic system.

[0007] The technical solution of this application is as follows: In a first aspect, the present application discloses a method for processing fan temperature measurement data, comprising: Obtain real-time active power data of wind turbines and real-time temperature data of generator windings; Perform time alignment processing on the collected real-time active power data and real-time temperature data to construct a power and temperature synchronization data sequence; Based on the power-temperature synchronization data sequence, the dynamic response operating conditions with significant power fluctuation characteristics are identified, and the dynamic response power input sequence and dynamic response temperature output sequence in the corresponding power-temperature synchronization data sequence are extracted; Based on the dynamic response power input sequence and the dynamic response temperature output sequence, the dynamic response characteristic parameters describing the response characteristics are calculated and their reference values ​​are recorded; Continuously record dynamic response characteristic parameters and construct a dynamic response characteristic parameter sequence; Determine whether the dynamic response characteristic parameter sequence has a one-time deviation and stable change pattern to identify benign response changes caused by hardware characteristic changes; After identifying a benign response change, updating a reference baseline value of a dynamic response characteristic parameter; Determine whether the dynamic response characteristic parameter sequence has a change pattern of continuous deviation or increased fluctuation to identify the real fault in the temperature measurement signal transmission link.

[0008] Through the above technical solution, the present application can effectively distinguish between benign response changes caused by hardware characteristic changes and real faults in the fan temperature measurement signal transmission link, avoiding misjudgment caused by the inability of traditional methods to distinguish between the two, thereby reducing unnecessary operation and maintenance costs and downtime losses, and improving the accuracy and reliability of the diagnostic system.

[0009] Furthermore, judging whether the dynamic response characteristic parameter sequence has a one-time deviation and stable change pattern includes: After identifying the dynamic response condition, the dynamic response power input sequence and the dynamic response temperature output sequence within a fixed time window are intercepted to construct the response morphology segment; Normalize the response morphology fragments and calculate the difference with the historical benchmark morphology to obtain the deviation index; Add the deviation index to the dynamic response characteristic parameter sequence; When the deviation indicator shows a sudden change in average value within a preset time period and the fluctuation amplitude is lower than a preset fluctuation threshold, it is identified as a benign response change caused by a change in hardware characteristics.

[0010] Through the above technical solution, it is possible to accurately identify benign response changes by normalizing and calculating the differences of the response morphology fragments, combining the mutation and fluctuation amplitude of the deviation index, avoiding misjudgment of small and stable changes, and improving the system's adaptability to changes in hardware characteristics.

[0011] Furthermore, after identifying a benign response change, updating the reference baseline value of the dynamic response characteristic parameter includes: After identifying a benign response change, a plurality of subsequent response morphology segments are acquired and normalized; Perform stability assessment on multiple normalized response morphology fragments and select sample fragments that meet preset stability conditions; Aggregate the sample segments to generate an updated response benchmark form; An updated reference reference value of the dynamic response characteristic parameter is calculated based on the response reference form.

[0012] Through the above technical solution, the reference baseline value can be dynamically updated based on the aggregation processing of subsequent stable sample fragments, ensuring that the system can quickly adapt and re-establish an accurate health baseline after a benign change in hardware characteristics, thereby maintaining the accuracy of the diagnosis.

[0013] Furthermore, it is determined whether the dynamic response characteristic parameter sequence has a pattern of continuous deviation or increased fluctuation to identify the real fault in the temperature measurement signal transmission link, including: Perform sliding segmentation processing on the dynamic response characteristic parameter sequence to construct multiple dynamic response characteristic parameter segments; Calculate the mean and fluctuation amplitude of each dynamic response characteristic parameter segment respectively; When the mean value deviates continuously from the reference benchmark value, or when the fluctuation amplitude is continuously higher than the preset fluctuation threshold, a real fault is identified.

[0014] Through the above technical solution, it is possible to effectively identify persistent signal anomalies by segmenting the dynamic response characteristic parameter sequence and analyzing its mean and fluctuation amplitude, thereby accurately judging the real fault in the temperature measurement signal transmission link and avoiding misjudgment of instantaneous fluctuations.

[0015] Furthermore, extracting the dynamic response power input sequence and the dynamic response temperature output sequence from the corresponding power-temperature synchronization data sequence includes: Obtain peak load control instructions for wind turbines; Calculate the rate of change of real-time active power data within a set time window; When the peak load control instruction is in an activated state and the change rate exceeds a preset change threshold, the current period is determined to be a dynamic response operating condition; The power input sequence and temperature output sequence corresponding to the power-temperature synchronization data sequence in the current period are extracted as the dynamic response power input sequence and the dynamic response temperature output sequence.

[0016] Through the above technical solution, the peak-shaving control instructions and power change rate can be combined to accurately identify the dynamic response conditions of the wind turbine during the peak-shaving process, ensuring that the extracted power input and temperature output sequences can truly reflect the response characteristics of the system under severe working conditions, thereby improving the accuracy of data extraction.

[0017] Furthermore, determining whether the mean value continuously deviates from the reference benchmark value, or when the fluctuation amplitude continuously exceeds a preset fluctuation threshold, includes: Perform statistical process monitoring on dynamic response characteristic parameter segments to establish control limits for reference baseline values; Determine whether the mean or fluctuation amplitude of each dynamic response characteristic parameter segment exceeds the control limit continuously, or whether it shows a non-random deviation trend; When any one of the judgment conditions is met, it is determined that there is a change pattern of persistent deviation or increased fluctuation.

[0018] Through the above technical solution, a statistical process monitoring method is introduced. By constructing control limits and judging whether the mean or fluctuation amplitude is continuously exceeded or shows a non-random deviation trend, it is possible to identify persistent anomalies more scientifically and sensitively, thereby improving the reliability of fault diagnosis.

[0019] Furthermore, statistical process monitoring is performed on the dynamic response characteristic parameter segments to establish control limits of reference benchmark values, including: Extract dynamic response characteristic parameter fragments within multiple historical time periods and calculate their long-term mean and fluctuation standard deviation; Construct the initial control limit interval based on the long-term mean and fluctuation standard deviation; Perform trend analysis on the dynamic response characteristic parameter segments of the current time period. When a stable drift mode is determined, update the long-term mean and fluctuation standard deviation. Based on the updated long-term mean and fluctuation standard deviation, the upper and lower thresholds of the initial control limit interval are dynamically corrected and used as control limits.

[0020] Through the above technical solution, the control limits can be dynamically corrected to adapt to the slow and stable drift that may occur during the operation of the wind turbine, avoiding false alarms or missed alarms caused by fixed limits and improving the adaptability of monitoring.

[0021] Furthermore, a trend analysis is performed on the dynamic response characteristic parameter segments of the current time period, including: Calculate the mean and fluctuation range of the dynamic response characteristic parameter segment in the current time period; Use a sliding window method to compare the mean and fluctuation range of multiple consecutive time periods to identify trend evolution patterns; When the mean continues to change slowly within the preset range and the fluctuation amplitude is within the preset stable range, the trend evolution mode is identified as drift mode.

[0022] Through the above technical solution, it is possible to analyze the continuous changes of the mean and fluctuation amplitude through a sliding window, accurately identify the stable drift pattern, and provide an accurate basis for the subsequent dynamic correction of the control limit.

[0023] Furthermore, when a stable drift mode is determined, the long-term mean and fluctuation standard deviation are updated, including: Based on multiple dynamic response characteristic parameter fragments within the current time period, the long-term mean is recalculated using a sliding average or exponential weighting method to generate an updated reference benchmark value; Adjust the upper and lower thresholds of the initial control limit interval based on the updated reference benchmark value and fluctuation standard deviation.

[0024] Through the above technical solution, the long-term mean can be smoothly updated by using sliding average or exponential weighting, ensuring that the update process of the reference benchmark value is stable and representative, further enhancing the system's adaptability to long-term drift.

[0025] In a second aspect, the present application further discloses a fan temperature measurement data processing system, the system comprising: A data acquisition module is used to obtain real-time active power data of the wind turbine generator set and real-time temperature data of the generator winding; A synchronization processing module is used to perform time alignment processing on the collected real-time active power data and real-time temperature data to construct a power and temperature synchronization data sequence; The operating condition identification and recording module is used to identify the dynamic response operating condition with significant power fluctuation characteristics based on the power-temperature synchronization data sequence, and extract the dynamic response power input sequence and dynamic response temperature output sequence from the corresponding power-temperature synchronization data sequence; A parameter calculation module calculates dynamic response characteristic parameters describing the response characteristics based on the dynamic response power input sequence and the dynamic response temperature output sequence, and records the reference baseline values; Parameter tracking module, used to continuously record dynamic response characteristic parameters and construct dynamic response characteristic parameter sequences; Benign change identification module, used to determine whether there is a one-time deviation and stable change pattern in the dynamic response characteristic parameter sequence, so as to identify benign response changes caused by hardware characteristic changes; A benchmark updating module, configured to update a reference benchmark value of a dynamic response characteristic parameter after identifying a benign response change; The real fault identification module is used to determine whether there is a change pattern of continuous deviation or increased fluctuation in the dynamic response characteristic parameter sequence, so as to identify the real fault in the temperature measurement signal transmission link.

[0026] Through the above technical solution, this application provides a system that can effectively distinguish between benign changes in hardware characteristics and real faults. Through modular design, it realizes the complete process of data acquisition, synchronous processing, working condition identification, parameter calculation, parameter tracking, benign change identification, benchmark update and real fault identification, providing a hardware foundation for the intelligent diagnosis of the fan temperature measurement system.

[0027] Beneficial effects

[0028] The present application effectively solves the problem that the existing fan temperature measurement data processing system has difficulty distinguishing between benign signal changes and real equipment failures when faced with slight changes in hardware characteristics. Specifically, the present application introduces dynamic response characteristic parameters and conducts a refined analysis of their change patterns, which can accurately identify one-time, stable response changes caused by hardware characteristic changes such as replacing signal cables, and promptly update the reference baseline value, thereby avoiding misjudging such benign changes as failures. At the same time, for change patterns with persistent offsets or increased fluctuations, the present application can accurately identify them as real temperature measurement signal transmission link failures. It can be seen that the method of the present application significantly improves the accuracy and reliability of the fan temperature measurement system diagnosis, reduces unnecessary on-site inspections and downtime losses, saves operation and maintenance costs, and enhances the operation and maintenance team's trust in the diagnostic system, overcoming the shortcomings of the prior art of frequent "pseudo-fault" alarms and reduced operation and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A flow chart of a method for processing fan temperature measurement data provided in this application.

[0030] Figure 2 This is a program flowchart of a fan temperature measurement data processing system provided in this application.

[0031] In the figure: 1. Data acquisition module; 2. Synchronous processing module; 3. Working condition identification and recording module; 4. Parameter calculation module; 5. Parameter tracking module; 6. Benign change identification module; 7. Benchmark update module; 8. Real fault identification module. DETAILED DESCRIPTION

[0032] The technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. The components of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for which protection is claimed, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of this application.

[0033] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0034] Traditional wind turbine temperature measurement data processing systems struggle to accurately distinguish between distortions in the dynamic characteristics of temperature signals caused by subtle hardware changes in the temperature signal transmission link (such as replacing a new signal cable) and actual equipment failures when wind turbines frequently adjust their power output. This diagnostic logic is often calibrated based on legacy hardware characteristics, causing the signals received by the system to mismatch the "healthy" model established based on the old data, resulting in misdiagnosis as "abnormal sensor response" or "abnormal signal transmission link." If these issues are not addressed, these misdiagnoses will lead to frequent and unnecessary on-site inspections and troubleshooting by the operation and maintenance team, resulting in costly wind turbine downtime, wasted spare parts and human resources, and a serious impact on the operation and maintenance team's trust in the diagnostic system, reducing wind turbine operation and maintenance efficiency.

[0035] Reference Figure 1 In this regard, this application proposes a method for processing fan temperature measurement data, including: S1000: Acquires real-time active power data of wind turbine generators and real-time temperature data of generator windings; S2000: performing time alignment processing on the collected real-time active power data and real-time temperature data to construct a power and temperature synchronization data sequence; S3000: Based on the power-temperature synchronization data sequence, identify a dynamic response operating condition with significant power fluctuation characteristics, and extract a dynamic response power input sequence and a dynamic response temperature output sequence from the corresponding power-temperature synchronization data sequence; S4000: Calculate dynamic response characteristic parameters describing the response characteristics based on the dynamic response power input sequence and the dynamic response temperature output sequence, and record their reference baseline values; S5000: Continuously record dynamic response characteristic parameters and construct a dynamic response characteristic parameter sequence; S6000: Determine whether a dynamic response characteristic parameter sequence has a one-time deviation and then a stable change pattern, so as to identify a benign response change caused by a hardware characteristic change; S7000: After identifying a benign response change, update a reference baseline value of a dynamic response characteristic parameter; S8000: Determine whether the dynamic response characteristic parameter sequence has a pattern of continuous deviation or increased fluctuation to identify the actual fault in the temperature measurement signal transmission link.

[0036] Therefore, this method effectively solves the problem of misjudging benign hardware changes and real faults in the existing technology, and improves the accuracy of diagnosis and operation and maintenance efficiency.

[0037] In this embodiment, the real-time active power data of the wind turbine generator set generally refers to the electric power data output by the wind turbine generator set in real time during operation, which can be collected by a power sensor on the grid side or inside the generator set.

[0038] The real-time temperature data of the generator winding usually refers to the real-time temperature of the winding inside the wind turbine generator, which can be collected by an embedded temperature sensor (such as a PT100 thermal resistor).

[0039] Time alignment involves synchronizing data from different sources and sampling frequencies (e.g., power and temperature data) on the time axis through methods such as timestamps or interpolation to facilitate joint analysis. A synchronized power and temperature data sequence is a corresponding data set consisting of power and temperature data arranged in chronological order after time alignment.

[0040] Dynamic response conditions refer to the operating state of a wind turbine when its power output fluctuates significantly (e.g., during peak shaving, startup and shutdown, or sudden load changes). A dynamic response power input sequence refers to the time-varying sequence of a wind turbine's power output under dynamic response conditions. A dynamic response temperature output sequence refers to the time-varying sequence of the generator winding temperature under dynamic response conditions, with the changes occurring in response to changes in power input. Dynamic response characteristic parameters describe the dynamic response characteristics between power input and temperature output, such as response time, overshoot, damping ratio, rise time, settling time, or specific transfer function parameters. These parameters can quantify the system's temperature response to power changes.

[0041] A reference baseline value is the expected or standard value of a dynamic response characteristic parameter under normal system conditions or with a specific hardware configuration. It is used to subsequently determine whether the parameter is abnormal. A one-time deviation followed by a stable change pattern refers to a one-time, non-continuous jump in a dynamic response characteristic parameter at a certain point in time, followed by stabilization at a new level. This is usually caused by hardware replacement or system parameter adjustment.

[0042] A pattern of change with persistent offset or increased fluctuation refers to the continuous deviation of the mean value of the dynamic response characteristic parameter from the reference benchmark value, or the continuous increase in its fluctuation amplitude, which usually indicates that the system has a fault or performance degradation. Benign response change refers to a change in the system response characteristics caused by non-fault factors (such as hardware upgrades, component replacements), which does not affect the normal function of the system. A real fault in the temperature measurement signal transmission link refers to a hardware or software failure in the signal transmission path from the sensor to the data processing unit of the temperature measurement system, resulting in signal distortion or loss, affecting the accuracy of temperature measurement. This method can usually be implemented in the monitoring center of the wind farm or the supervisory control and data acquisition (SCADA) system of the wind turbine generator set, using industrial control computers or dedicated data processing servers for data processing and analysis.

[0043] The implementation principle of the present application is as follows: First, it is necessary to obtain the real-time active power data of the wind turbine generator set and the real-time temperature data of the generator winding. As an implementation method, the field instrument data can be manually read regularly and manually input into the data processing system. As another implementation method, a simple analog signal acquisition card can be used to convert the analog signal output by the sensor into a digital signal, but the acquisition card may not have high precision or multi-channel synchronous acquisition capabilities. As another implementation method, a preset fixed time interval (for example, every minute) can be set to pull data from the SCADA system of the wind turbine generator set without considering the real-time nature of the data update or event triggering.

[0044] Secondly, the collected real-time active power data and real-time temperature data are time-aligned to construct a power-temperature synchronized data sequence. In one embodiment, time alignment can be achieved through a simple timestamp matching method. If the timestamps of different data sources are not completely consistent, the mismatched data points are discarded. In another embodiment, linear interpolation can be used to fill in data with inconsistent timestamps, but this method may not be suitable for nonlinearly varying data. In yet another embodiment, a wide time window can be set, and data falling within this window can be considered synchronized. However, this method may introduce significant time deviations. Based on the power-temperature synchronized data sequence, dynamic response conditions with significant power fluctuation characteristics are identified, and the dynamic response power input sequence and dynamic response temperature output sequence from the corresponding power-temperature synchronized data sequence are extracted. In one embodiment, a fixed power change threshold can be manually set. When the absolute value change of power exceeds this threshold, it is considered that the current dynamic response condition has been entered. In another embodiment, the absolute value change of the power data can be simply monitored, without considering its rate of change or its correlation with specific control instructions. As another implementation, a sliding window with a fixed time period can be used to identify power fluctuations. However, this method may not accurately capture the start and end points of the dynamic response. Once a dynamic response condition is identified, the corresponding power input sequence and temperature output sequence within that period can be extracted.

[0045] On this basis, based on the dynamic response power input sequence and the dynamic response temperature output sequence, the dynamic response characteristic parameters describing its response characteristics are calculated, and their reference baseline values ​​are recorded. As an embodiment, the simple lag time or maximum temperature rise value of the temperature response can be calculated as the dynamic response characteristic parameter. As another embodiment, the historical data can be statistically averaged once to obtain a fixed reference baseline value, without considering its changes over time or working conditions. As another embodiment, data under several typical working conditions can be manually selected and their characteristic parameters calculated as a reference baseline. Subsequently, the dynamic response characteristic parameters are continuously recorded to construct a dynamic response characteristic parameter sequence. As an embodiment, the dynamic response characteristic parameters obtained each time can be simply appended to the database to form a time series. As another embodiment, a fixed frequency (for example, every hour) can be set to record the parameters, regardless of the frequency of the dynamic response working condition.

[0046] Next, determine whether the dynamic response characteristic parameter sequence has a one-time deviation and a stable change pattern to identify benign response changes caused by hardware characteristic changes. A fixed threshold can be set. When the difference between the current value of the dynamic response characteristic parameter and the reference baseline value exceeds the threshold, it is determined to be a deviation. In another embodiment, the chart of the dynamic response characteristic parameter sequence can be manually visually inspected to determine whether there is a one-time deviation. As another embodiment, the average values ​​of the two time periods before and after can be compared. If the difference is large, it is considered that there is a deviation, but no further determination is made as to whether it is stabilizing.

[0047] After a benign response change is identified, the reference baseline value of the dynamic response characteristic parameter is updated. In one embodiment, the first stable parameter value after the benign change is identified can be directly used as the new reference baseline value. In another embodiment, after confirming a benign change, a new reference baseline value can be manually input through human intervention.

[0048] Finally, determine whether the dynamic response characteristic parameter sequence has a pattern of change that shows persistent deviation or increased fluctuations to identify real faults in the temperature measurement signal transmission link. As an implementation method, a fixed upper and lower limit can be set. When the dynamic response characteristic parameter exceeds the limit value N times in a row, it is determined to be a persistent deviation. As another implementation method, the standard deviation of the parameter sequence can be simply calculated. When the standard deviation exceeds a preset threshold, it is determined to be an increased fluctuation. As another implementation method, the trend chart of the parameter sequence can be manually reviewed regularly to determine whether there is a persistent deviation or increased fluctuation.

[0049] The overall working principle of the method provided in the present application is to achieve accurate identification of anomalies in the temperature measurement signal transmission link by finely processing and analyzing the real-time active power data of the wind turbine generator set and the real-time temperature data of the generator winding. Specifically, first, the real-time active power data and the real-time temperature data of the generator winding are acquired and time-aligned, thereby constructing a power-temperature synchronization data sequence, which lays the foundation for subsequent joint analysis. Secondly, based on the synchronization data sequence, the system can identify dynamic response conditions with significant power fluctuation characteristics, and extract the corresponding dynamic response power input sequence and dynamic response temperature output sequence therefrom. This step ensures that the focus of the analysis is on the key conditions where the response characteristics of the wind turbine best reflect its health status.

[0050] Subsequently, based on the extracted dynamic response power input sequence and dynamic response temperature output sequence, the system calculates the dynamic response characteristic parameter that describes its response characteristics and records its reference baseline value. This parameter quantifies the temperature response behavior of the fan under dynamic operating conditions, while the reference baseline value provides a basis for determining whether it is abnormal. By continuously recording these dynamic response characteristic parameters and constructing a dynamic response characteristic parameter sequence, the system can track the long-term operating status of the fan temperature measurement link.

[0051] This application introduces two different judgment modes to distinguish the types of abnormalities. On the one hand, by judging whether there is a one-time deviation and stable change pattern in the dynamic response characteristic parameter sequence, benign response changes caused by hardware characteristic changes can be identified. For example, when a new model of signal cable is replaced, the dynamic response characteristics of the temperature measurement signal may undergo a one-time, systematic deviation, but will then stabilize in a new "healthy" state. This method can identify such benign changes and promptly update the reference baseline value of the dynamic response characteristic parameter after identification, thereby avoiding misjudging such normal system adjustments as faults. On the other hand, by judging whether there is a continuous deviation or an increased fluctuation pattern in the dynamic response characteristic parameter sequence, the real fault in the temperature measurement signal transmission link can be accurately identified. For example, when the signal transmission link has poor contact or the components are aging, the dynamic response characteristic parameters may continue to deviate from their baseline values, or their fluctuation amplitude may increase significantly.

[0052] This method distinguishes between benign response changes and true faults, enabling the diagnostic system to adapt to changes in the wind turbine's operating environment and hardware configuration, avoiding the false alarms caused by hardware updates in traditional methods. The various technical features work together to form a complete, closed-loop diagnostic process, from data acquisition, preprocessing, operating condition identification, parameter calculation and tracking, to final intelligent judgment and benchmark updates. This significantly improves the accuracy and reliability of fault diagnosis in wind turbine temperature measurement systems, thereby enhancing the operational efficiency and economic benefits of wind turbine generator systems.

[0053] Furthermore, in another embodiment of the present application, S6000 includes: S6100: After identifying the dynamic response condition, it intercepts the dynamic response power input sequence and the dynamic response temperature output sequence within a fixed time window to construct a response morphology segment; S6200: Normalize the response morphology segment and calculate the difference with the historical reference morphology to obtain a deviation index; S6300: Add the deviation index to the dynamic response characteristic parameter sequence; S6400: When the deviation indicator shows a sudden change in average value within a preset time period and the fluctuation amplitude is lower than a preset fluctuation threshold, it is identified as a benign response change caused by a change in hardware characteristics.

[0054] Specifically, after identifying the dynamic response condition of a wind turbine, the dynamic response power input sequence and dynamic response temperature output sequence within a specific time window are extracted from the power and temperature synchronization data sequence to construct a response profile segment. This response profile segment characterizes the transient response relationship between power input and temperature output under the specific dynamic condition. Subsequently, the response profile segment is normalized to eliminate dimensional differences and amplitude effects across different operating or measurement conditions. The normalized response profile segment is compared with a pre-stored historical baseline profile to quantify the degree of deviation between the current response and the historical normal response, generating a deviation index. This deviation index is then incorporated into the dynamic response characteristic parameter sequence for continuous monitoring and analysis. Finally, by analyzing this deviation index, if a sudden change in its mean value within a preset time period—that is, a sudden jump from one stable level to another—and the fluctuation amplitude is below a preset fluctuation threshold, it can be identified as a benign response change caused by hardware characteristic changes. This change is typically caused by non-fault factors such as sensor aging or minor changes in line impedance, rather than a true failure of the temperature measurement signal transmission link.

[0055] The solution of this application achieves accurate identification of benign response changes by introducing response morphology segments, normalization, a deviation index, and sudden change and fluctuation analysis based on the deviation index. Specifically, the dynamic response power input sequence and dynamic response temperature output sequence within a fixed time window are intercepted and constructed into response morphology segments, aiming to capture the transient thermal response characteristics of the wind turbine under dynamic operating conditions. Normalization of these response morphology segments eliminates the impact of changes in the external environment or operating conditions on the data amplitude, making the response morphology at different time points comparable. By calculating the difference with the historical baseline morphology, the deviation between the current response and the normal response can be quantified and a deviation index can be generated. The introduction of this deviation index simplifies complex dynamic response characteristics into a single, quantifiable value, facilitating subsequent trend analysis. When the deviation index shows a sudden change in its average value within a preset time period with low fluctuation, it indicates a one-time, stable change in the system response characteristics. This is typically due to a small but stable change in the hardware's physical properties (such as sensor drift or line impedance changes), rather than an intermittent or persistent fault in the signal transmission link. This mechanism enables the system to distinguish normal aging or minor adjustments of hardware from actual failures, avoiding false alarms.

[0056] Furthermore, in another embodiment of the present application, S7000 includes: S7100: After identifying a benign response change, obtain a plurality of subsequent response morphology segments and perform normalization processing; S7200: Performing stability assessment on multiple normalized response morphology segments to select sample segments that meet preset stability conditions; S7300: Aggregate the sample segments to generate an updated response benchmark form; S7400: Calculate updated reference reference values ​​of dynamic response characteristic parameters based on the response reference form.

[0057] Specifically, after identifying a benign response change caused by a change in hardware characteristics, the system will proactively acquire multiple response morphology fragments after the benign change occurs and the system operation tends to be stable. These fragments are instantaneous snapshots of the relationship between power input and temperature output of the wind turbine under dynamic response conditions. To eliminate dimensional differences under different operating conditions or measurement conditions, these response morphology fragments are normalized so that they can be compared and analyzed on a unified scale. Normalization can be performed using a variety of methods, such as maximum-minimum normalization and Z-score normalization, to ensure the accuracy of subsequent evaluations.

[0058] Among them, the stability assessment of multiple normalized response morphology segments is aimed at ensuring that the samples used to update the benchmark are stable and representative. The stability assessment can be based on indicators such as volatility within the segment, similarity between segments, or the degree of deviation from the short-term average morphology. For example, the variance or standard deviation of each normalized segment can be calculated, and a preset stability threshold can be set. Only when the volatility of the segment is lower than the threshold is it considered stable. In addition, the similarity between different segments can be evaluated by calculating the correlation coefficient or Euclidean distance between them, thereby screening out sample segments that are highly consistent with each other.

[0059] In practical applications, sample segments that meet pre-defined stability criteria are aggregated to extract a new, representative baseline response profile from multiple stable samples. Aggregation can employ statistical methods such as averaging, median, and weighted average to fuse the information from multiple stable segments into a single, more robust profile. For example, all sample segments can be aligned along the time axis and the normalized values ​​at each time point averaged to produce a smooth baseline curve representing the new response profile.

[0060] Based on the generated updated response benchmark form, an updated reference baseline value for the dynamic response characteristic parameter can be calculated. This calculation process is consistent with the calculation method for the initial reference baseline value, ensuring that the new baseline value accurately reflects the system dynamic response characteristics after the hardware characteristics have changed. For example, if the dynamic response characteristic parameter is a response time constant or gain, the model will be refitted based on the new response benchmark form to obtain the updated time constant or gain as the new reference baseline value.

[0061] The solution of the present application does not simply use the old reference benchmark value after identifying a benign response change, but actively collects and processes multiple subsequent response morphology fragments. Specifically, by obtaining multiple subsequent response morphology fragments and performing normalization processing, the consistency and comparability of the data are ensured. Furthermore, by performing stability evaluation and screening on these normalized fragments, fragments affected by transient disturbances or abnormal data are excluded, thereby ensuring the reliability of the samples used to update the benchmark. Ultimately, the screened stable sample fragments are aggregated, which can effectively smooth the data noise and extract the true and stable dynamic response characteristics of the system after the hardware characteristics change, thereby generating a response benchmark morphology that accurately reflects the current system state. The updated reference benchmark value is calculated based on this new response benchmark morphology, so that subsequent fault judgment can be based on the latest and most realistic system characteristics, thereby avoiding misjudgment caused by changes in hardware characteristics.

[0062] Through the above-mentioned technical solution, this application can effectively address the changes in the dynamic response characteristics of the temperature measurement signal transmission link that occur when benign changes occur in the hardware characteristics of the wind turbine (such as sensor drift, component aging, or replacement). By dynamically updating the reference baseline values ​​of the dynamic response characteristic parameters, the system can continuously adapt to the long-term evolution of the wind turbine's operating status, ensuring the accuracy and robustness of temperature measurement signal transmission link fault diagnosis throughout the wind turbine's lifecycle. This significantly improves the accuracy of fault identification, reduces unnecessary on-site inspections and maintenance, thereby reducing operating costs and improving the operational reliability of the wind turbine.

[0063] In some preferred embodiments, a specific example is used below to illustrate: Assume that after a period of operation, the generator winding temperature sensor of a wind turbine generator set has a slight aging effect, causing its response characteristics to undergo a one-time, stable, small change. This is recognized by the system as a benign response change.

[0064] Upon identifying this positive change, the system immediately initiates a baseline update process. Specifically, over the following hours or days, whenever the wind turbine enters a dynamic response condition such as peak load regulation, the system continuously captures and acquires multiple (e.g., 10) subsequent sequences of synchronized power and temperature data. These segments are then normalized to eliminate the effects of varying power levels and temperature ranges.

[0065] Next, the stability of these 10 normalized response morphology segments is assessed. For example, the root mean square error (RMS) or standard deviation (SD) of each segment is calculated, and a threshold is set. If a segment's RMS error is too large, it is considered unstable and removed. Assume that after screening, 8 segments are considered stable.

[0066] These eight stable sample segments are then aggregated, for example, using a point-by-point averaging method to align the eight segments on the time axis and average the normalized temperature values ​​at each time point to generate a new, smoothed response baseline morphology curve.

[0067] Finally, based on this new response benchmark curve, the dynamic response characteristic parameters are recalculated (for example, if the characteristic parameter is a response time constant, a first-order or second-order system model is refitted) to obtain an updated reference baseline value. The system then uses this updated reference baseline value to determine whether subsequent dynamic response characteristic parameter sequences exhibit persistent drift or increased fluctuations. This allows for more accurate identification of true faults in the temperature measurement signal transmission link, avoiding misinterpretation of normal response changes due to sensor aging as faults.

[0068] Furthermore, in another embodiment of the present application, step S8000 includes: S8100: Perform sliding segmentation processing on the dynamic response characteristic parameter sequence to construct multiple dynamic response characteristic parameter segments; S8200: Calculate the mean and fluctuation amplitude of each dynamic response characteristic parameter segment respectively; S8300: If the mean value continuously deviates from the reference baseline value, or if the fluctuation amplitude continuously exceeds the preset fluctuation threshold, a real fault is identified.

[0069] Sliding segmentation of the dynamic response characteristic parameter sequence refers to dividing the continuously recorded dynamic response characteristic parameter sequence into a preset time window or number of data points, thereby constructing a series of overlapping or non-overlapping dynamic response characteristic parameter segments. For example, a fixed-length sliding window can be set, each moving a step size (such as a data point or a fixed time interval) to generate a series of segments containing recent dynamic response characteristic parameters.

[0070] Calculating the mean and fluctuation amplitude for each dynamic response characteristic parameter segment involves calculating the arithmetic mean of all parameter values ​​within each constructed dynamic response characteristic parameter segment as the mean, and calculating the dispersion of the parameter values ​​within the segment as the fluctuation amplitude. For example, the fluctuation amplitude can be expressed as standard deviation, variance, or range (the difference between the maximum and minimum values) to quantify the stability or volatility of the parameters within the segment. Identifying a true fault when the mean continuously deviates from the reference baseline value, or when the fluctuation amplitude continuously exceeds a preset fluctuation threshold, involves comparing the mean of each dynamic response characteristic parameter segment with a preset reference baseline value, and comparing its fluctuation amplitude with a preset fluctuation threshold, to determine whether an anomaly exists. "Continuous deviation" or "continuously exceeding" means that for a certain number of consecutive segments, the mean value continuously exceeds the normal range of the reference baseline value (for example, beyond the statistical control limit), or the fluctuation amplitude continuously exceeds the allowable threshold. This continuity assessment helps eliminate occasional interference or momentary fluctuations, thereby more accurately identifying persistent problems caused by hardware failures, line aging, poor contact, and other factors in the temperature measurement signal transmission link.

[0071] The proposed solution, through refined sliding segmentation of the dynamic response characteristic parameter sequence and quantitative analysis of the mean and fluctuation amplitude of each segment, can effectively capture the persistent and systematic changes in the temperature measurement signal transmission link caused by real faults, avoiding misjudging benign response changes as faults. This method can significantly improve the sensitivity and specificity of fault identification, ensuring that fault alarms are only issued when there is a persistent problem in the temperature measurement signal transmission link, thereby reducing false alarm rates and improving the reliability and maintenance efficiency of wind turbine operations.

[0072] Furthermore, in another embodiment of the present application, the sub-step of S3000 is proposed: extracting the dynamic response power input sequence and the dynamic response temperature output sequence from the corresponding power-temperature synchronization data sequence, including: S3210: Obtain peak load regulation control instructions for wind turbine generator sets; S3220: Calculate the rate of change of real-time active power data within a set time window; S3230: When the peak load control instruction is in an activated state and the change rate exceeds a preset change threshold, determining that the current period is a dynamic response operating condition; S3240: Extract the power input sequence and temperature output sequence corresponding to the power-temperature synchronization data sequence in the current period as the dynamic response power input sequence and the dynamic response temperature output sequence.

[0073] Specifically, a peak-shaving control instruction refers to a power regulation instruction executed by a wind turbine in response to changes in grid frequency or power demand under grid dispatch or its own control strategy. This instruction can be obtained from the wind turbine's control system, SCADA system, or grid dispatch center. The rate of change of real-time active power data within a set time window can be obtained by performing differential calculation or regression analysis on continuously collected real-time active power data. For example, the average rate of change of active power can be calculated within a preset sliding time window, or determined by comparing the power values ​​at the start and end of the window. The length of the set time window can be adjusted according to the response characteristics of the wind turbine and the data sampling frequency to ensure that effective power fluctuations can be captured. The peak-shaving control instruction is in an active state, which means that the wind turbine is actively participating in the peak-shaving or frequency-regulating services of the grid. At this time, the rapid changes in its active power are usually controlled and purposeful responses.

[0074] The preset change threshold is a critical value determined empirically or through historical data analysis, used to distinguish normal power fluctuations from significant power changes caused by peak-shaving instructions. When the peak-shaving control instruction is active and the rate of change of the real-time active power data exceeds the preset change threshold, it can be accurately determined that the wind turbine is currently in a dynamic response condition. Once the current period of dynamic response conditions is determined, the corresponding power data within that period can be accurately extracted from the pre-established power-temperature synchronization data sequence as the dynamic response power input sequence, and the corresponding temperature data as the dynamic response temperature output sequence.

[0075] The solution of this application introduces peak-shaving control instructions as a prerequisite for determining dynamic response conditions, and combines this with the rate of change of real-time active power data for dual confirmation, enabling more accurate identification of dynamic response conditions generated by wind turbines actively responding to grid demand. This approach avoids misidentifying power fluctuations caused by non-peak-shaving factors such as wind speed changes and fault tripping as dynamic response conditions, thereby ensuring that the extracted dynamic response power input sequence and dynamic response temperature output sequence truly reflect the thermal response characteristics of the generator windings under controlled power changes. This provides a high-quality, highly relevant data foundation for the subsequent calculation of dynamic response characteristic parameters.

[0076] Furthermore, in another embodiment of the present application, it is proposed that S8300 includes: S8310: Perform statistical process monitoring on dynamic response characteristic parameter segments to establish control limits for reference baseline values; S8320: Determine whether the mean or fluctuation amplitude of each dynamic response characteristic parameter segment continuously exceeds the control limit or whether it shows a non-random deviation trend; S8330: When any judgment condition is met, it is determined that there is a change pattern of persistent deviation or increased fluctuation.

[0077] Specifically, statistical process monitoring of dynamic response characteristic parameter segments involves using statistical methods to monitor data sequences in real time to determine whether the process is in a state of statistical control. The goal is to use quantitative statistical indicators to determine whether data changes exceed normal fluctuations, thereby more accurately identifying anomalies. Establishing control limits for reference baseline values ​​can be understood as defining upper and lower limits based on historical data or a preset statistical distribution to define the normal fluctuation range of the dynamic response characteristic parameters. For example, the three-sigma rule can be used to set control limits, where a range of ±3 standard deviations from the mean is considered normal. Determining whether the mean or fluctuation amplitude of each dynamic response characteristic parameter segment continuously exceeds the control limits refers to determining an anomaly when the mean or fluctuation amplitude falls outside the control limits for multiple consecutive sampling points or time periods. Alternatively, determining whether a non-random excursion trend is present refers to determining whether the data points exhibit significant non-random variation, such as continuous increases or decreases or cyclical fluctuations, even if they do not continuously exceed the control limits. These trends may also indicate potential failures. For example, RunChart Rules or Westgard Rules can be used to identify non-random deviation trends, such as multiple consecutive points falling on one side of the mean line, or multiple consecutive points showing a monotonic upward / downward trend.

[0078] The solution of the present application introduces a statistical process monitoring method to conduct a more refined analysis of the dynamic response characteristic parameter segments. Traditional simple threshold judgments may not be able to effectively distinguish between random noise and persistent changes caused by real faults. By constructing control limits for reference baseline values, and combining the judgment of whether the mean or fluctuation amplitude continuously exceeds the control limits and whether it presents a non-random offset trend, it is possible to more accurately identify the change pattern of persistent offset or increased fluctuation caused by real faults in the temperature measurement signal transmission link based on statistical principles. This method can effectively filter out accidental fluctuations and focus on statistically significant anomalies, thereby improving the accuracy and robustness of fault identification.

[0079] Through the above technical solution, this application can significantly improve the accuracy and reliability of fault identification in the wind turbine temperature measurement signal transmission link. Compared with simple threshold judgment, the use of statistical process monitoring can more effectively identify persistent offsets or increased fluctuations caused by real faults, while reducing the false alarm rate caused by random fluctuations or changes in normal operating conditions. This ensures a more accurate assessment of the operating status of the wind turbine generator set, and timely detection and treatment of potential temperature measurement system faults, thereby ensuring the safe and stable operation of the wind turbine generator set and reducing unplanned downtime.

[0080] In some preferred embodiments, a specific example is used below to illustrate: Assume that during routine operation of a wind turbine, a series of dynamic response characteristic parameters is continuously monitored. To determine whether there is persistent drift or increased fluctuation, the system first calculates the long-term mean and standard deviation of the characteristic parameter based on historical data and sets upper and lower control limits accordingly. For example, the mean ±3 times the standard deviation is used as the control limit. During real-time monitoring, each time a new dynamic response characteristic parameter segment is acquired, its mean and fluctuation amplitude are calculated. If the mean of five consecutive segments falls outside the upper control limit, or the fluctuation amplitude of five consecutive segments exceeds the preset fluctuation threshold and exceeds the upper control limit, the system determines that a pattern of persistent drift or increased fluctuation exists and triggers a fault alarm. Furthermore, even if the mean and fluctuation amplitude do not continuously exceed the control limits, if the mean of seven consecutive segments shows a monotonically increasing trend or the mean of seven consecutive segments falls on the same side of the center line (long-term mean), the system will identify a non-random drift trend and also determine the presence of a potential fault. This statistical process monitoring-based method can effectively avoid misjudgments caused by instantaneous fluctuations, ensuring that fault identification is only performed when persistent abnormalities do occur in the temperature measurement signal transmission link.

[0081] Furthermore, in another embodiment of the present application, it is proposed that S8310 includes: S8311: Extract dynamic response characteristic parameter segments within multiple historical time periods and calculate their long-term mean and fluctuation standard deviation; S8312: Construct the initial control limit interval based on the long-term mean and fluctuation standard deviation; S8313: Perform trend analysis on the dynamic response characteristic parameter segments of the current time period. If a stable drift pattern is determined, update the long-term mean and fluctuation standard deviation. S8314: Based on the updated long-term mean and fluctuation standard deviation, the upper and lower thresholds of the initial control limit interval are dynamically revised and used as the control limits.

[0082] Specifically, extracting dynamic response characteristic parameter segments from multiple historical time periods involves selecting sufficiently long and representative time periods, such as several months or years, from the long-term operating data of the wind turbine. The goal is to obtain sufficient sample data to accurately reflect the long-term statistical characteristics of the parameter. Based on this, the long-term mean and standard deviation of these historical segments are calculated. The long-term mean can be understood as the average level of the parameter under normal operating conditions, while the standard deviation reflects the range of fluctuation around the mean. These statistics form the basis for constructing control limits. Initial control limits are constructed based on the long-term mean and standard deviation. This involves using statistical process control (SPC) principles, such as the 3σ principle, with the long-term mean as the center line and the long-term mean plus or minus a multiple of the standard deviation as the upper and lower control limits to form an initial, static control limit interval. This initial control limit interval is used to preliminarily determine whether the dynamic response characteristic parameter is within the normal range.

[0083] In practical applications, trend analysis of dynamic response characteristic parameter segments within the current time period involves monitoring and analyzing the mean and fluctuation range of these segments over a recent period (e.g., several days or weeks). A stable drift pattern can be identified when the mean exhibits a slow, sustained upward or downward trend, while the fluctuation range remains within a preset stability range. This drift is typically caused by benign factors such as sensor aging, minor changes in line impedance, or slow changes in the generator's internal physical characteristics. When a stable drift pattern is identified, the long-term mean and standard deviation of the fluctuations need to be updated to adapt the control limits to these benign changes. Based on these updated long-term mean and standard deviation, the upper and lower thresholds of the initial control limit interval are dynamically adjusted to serve as the control limits. This means that once a stable drift is identified and the baseline statistics are updated, the control limits will adjust accordingly, forming dynamically adaptable control limits. This dynamic adjustment ensures that the control limits always accurately reflect the normal operating range of the current system, avoiding misinterpretation of normal system drift as a fault.

[0084] The solution of the present application achieves adaptive adjustment of control limits by introducing trend analysis of dynamic response characteristic parameter segments and dynamically updating the long-term mean and fluctuation standard deviation based on the analysis results. Specifically, when a stable drift pattern occurs in the system, traditional fixed control limits may cause parameter values ​​to continuously exceed the limit, thereby frequently triggering false alarms. The present application identifies this benign drift and updates the reference baseline value and control limits accordingly, so that the control limits can "follow" the normal drift of the system, thereby avoiding misjudging normal changes in system characteristics as faults. At the same time, the new dynamic control limits can still effectively identify non-drifting, abnormal offsets or increased fluctuations, ensuring the sensitivity of fault diagnosis.

[0085] In some preferred embodiments, a specific example is used below to illustrate: Assume that in the early stages of a wind turbine's operation, the long-term mean of a dynamic response characteristic parameter calculated from historical data is X, and the standard deviation of its fluctuation is Y. The initial control limits constructed from these values ​​are [X-3Y, X+3Y]. As the turbine grows older, due to slight sensor aging, the mean of this characteristic parameter may slowly increase, for example, by 0.01 units per month, while its fluctuation remains stable. If fixed control limits are used, after several months, even in the absence of a true fault, the parameter's mean may consistently exceed the upper limit of X+3Y, causing the system to frequently issue fault alarms. When the system identifies this trend of slowly increasing mean values ​​and stable fluctuations, it identifies a stable drift pattern. At this point, the system recalculates the long-term mean and standard deviation based on the new data. For example, the updated long-term mean becomes X', while the standard deviation remains Y. Subsequently, the control limits are dynamically adjusted to [X'-3Y, X'+3Y]. This way, even if the parameter's mean undergoes a benign drift, as long as it remains within the new, dynamically adjusted control limits, it will not be misidentified as a fault. Only when the mean or fluctuation range of the parameter exceeds the dynamically adjusted control limit, or shows non-random, rapid abnormal changes, will it be identified as a real temperature measurement signal transmission link failure, thus avoiding false alarms and improving the accuracy of fault diagnosis.

[0086] Furthermore, in another embodiment of the present application, the sub-step of S8313 is proposed: performing trend analysis on the dynamic response characteristic parameter segments of the current time period, including: A1: Calculates the mean and fluctuation range of the dynamic response characteristic parameter segment within the current time period; A2: Use a sliding window approach to compare the mean and fluctuation range of multiple consecutive time periods to identify trend evolution patterns; A3: When the mean continues to change slowly within the preset range and the fluctuation amplitude is within the preset stable range, the trend evolution mode is identified as drift mode.

[0087] Specifically, when performing trend analysis on a segment of the dynamic response characteristic parameters within the current time period, the mean and fluctuation range of the dynamic response characteristic parameter segment within the current time period must first be calculated. The mean can be understood as the arithmetic average of all dynamic response characteristic parameters within the segment, representing the central tendency of the parameters within that time period. The fluctuation range can be measured using statistics such as standard deviation, variance, or range, representing the degree of dispersion or stability of the parameters within the segment. To identify the trend evolution pattern of the parameters, a sliding window approach is used to compare the means and fluctuation ranges of multiple consecutive time periods. Specifically, a fixed-size sliding window can be set and moved forward across the dynamic response characteristic parameter sequence. With each sliding window, the mean and fluctuation range of the parameter segments within the window are calculated. By comparing the means and fluctuation ranges of adjacent or consecutive windows, the temporal trend of the parameters can be observed. For example, if the mean value shows a continuous small increase or decrease while the fluctuation range remains relatively stable, this may indicate a trend.

[0088] Therefore, when the mean value continues to maintain a preset slow-changing state and the fluctuation amplitude is within a preset stable range, the trend evolution mode is identified as a drift mode. "Preset slow-changing state" means that the rate of change of the mean value is lower than a preset threshold, indicating that the baseline value of the parameter is shifting slowly and gradually, rather than suddenly; "within the preset stable range" means that the fluctuation amplitude remains within normal operation or expected noise levels, indicating that the intrinsic variability of the parameter has not increased significantly. When both conditions are met, it can be determined that the system response characteristics have undergone a benign, gradual drift.

[0089] The solution of this application can effectively capture subtle parameter change trends by performing a refined analysis of the mean and fluctuation amplitude of the dynamic response characteristic parameter segments and introducing a sliding window comparison mechanism. In particular, by clearly defining the identification conditions of the "drift mode", that is, the mean value changes slowly and the fluctuation amplitude is stable, the system can distinguish between benign and gradual response changes caused by hardware aging, environmental changes, etc., and abnormal fluctuations or offsets caused by temperature measurement signal transmission link failures. This differentiation capability is crucial for subsequent reference baseline value updates, ensuring the adaptability and accuracy of the monitoring system.

[0090] Furthermore, in another embodiment of the present application, sub-step S8313 is proposed: when it is determined to be a stable drift mode, updating the long-term mean and the fluctuation standard deviation includes: B1: Based on multiple dynamic response characteristic parameter segments within the current time period, recalculate their long-term average using a sliding average or exponential weighting method to generate an updated reference benchmark value; B2: Adjust the upper and lower thresholds of the initial control limit interval based on the updated reference benchmark value and fluctuation standard deviation.

[0091] Among them, "sliding average" is a commonly used data smoothing technique that eliminates short-term fluctuations by calculating the average value of data within a fixed-length window, thereby revealing the long-term trend of the data. For example, a sliding window containing N latest dynamic response characteristic parameter fragments can be set. Each time a new fragment is added, the oldest fragment is removed and the mean of all fragments in the window is recalculated. "Exponential weighting" is an averaging method that gives higher weight to recent data. Its characteristic is that it can respond to the latest changes in data more quickly while retaining the influence of historical data. For example, historical data can be weighted by an attenuation factor so that the newer data has a greater impact on the mean. Both methods are designed to make the calculated long-term mean more accurately reflect the true baseline state of the current system in stable drift mode.

[0092] Therefore, by recalculating the long-term mean using a sliding average or exponentially weighted approach, the reference baseline can dynamically adapt to slow, steady drifts in the wind turbine's temperature response characteristics during long-term operation due to factors such as component aging and environmental changes. This adaptive baseline update mechanism ensures that control limits always remain close to the system's current normal operating state, thus avoiding misinterpreting normal system characteristic drifts as faults.

[0093] The specific embodiment of the present application further discloses a fan temperature measurement data processing system, which includes: Data acquisition module 1, used to obtain real-time active power data of the wind turbine generator set and real-time temperature data of the generator winding; Synchronous processing module 2, used for time alignment processing of the collected real-time active power data and real-time temperature data, and constructing a power-temperature synchronization data sequence; The operating condition identification and recording module 3 is used to identify the dynamic response operating condition with significant power fluctuation characteristics based on the power-temperature synchronization data sequence, and extract the dynamic response power input sequence and dynamic response temperature output sequence from the corresponding power-temperature synchronization data sequence; The parameter calculation module 4 calculates the dynamic response characteristic parameters describing the response characteristics based on the dynamic response power input sequence and the dynamic response temperature output sequence, and records the reference baseline values ​​thereof; Parameter tracking module 5, used to continuously record dynamic response characteristic parameters and construct a dynamic response characteristic parameter sequence; Benign change identification module 6, used to determine whether there is a one-time deviation and stable change pattern in the dynamic response characteristic parameter sequence, so as to identify the benign response change caused by the hardware characteristic change; A reference updating module 7 is configured to update a reference reference value of a dynamic response characteristic parameter after identifying a benign response change; The real fault identification module 8 is used to determine whether there is a change pattern of continuous deviation or increased fluctuation in the dynamic response characteristic parameter sequence, so as to identify the real fault in the temperature measurement signal transmission link.

[0094] Through a modular design, the system achieves efficient processing and intelligent diagnosis of temperature measurement data from wind turbines, aiming to address the difficulty in distinguishing between benign hardware changes and real equipment failures in existing technologies. Specifically, the system can acquire and synchronously process the real-time active power data of the wind turbine and the real-time temperature data of the generator windings, thereby identifying the dynamic response conditions and extracting the corresponding power input sequence and temperature output sequence. On this basis, the system calculates and tracks the dynamic response characteristic parameters, and by judging their change patterns, it intelligently distinguishes between benign response changes caused by changes in hardware characteristics and real failures in the temperature measurement signal transmission link. When a benign response change is identified, the system can promptly update the reference baseline value to avoid false alarms and ensure the accuracy of the diagnosis and the adaptability of the system.

[0095] The above are merely examples of the present application and are not intended to limit the scope of protection of the present application. Those skilled in the art will appreciate that various modifications and variations are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present application shall be included within the scope of protection of the present application.

Claims

1. A method for processing fan temperature measurement data, characterized in that: include: Obtain real-time active power data of wind turbines and real-time temperature data of generator windings; Performing time alignment processing on the collected real-time active power data and the real-time temperature data to construct a power-temperature synchronization data sequence; Based on the power-temperature synchronization data sequence, identifying a dynamic response operating condition with significant power fluctuation characteristics, and extracting a dynamic response power input sequence and a dynamic response temperature output sequence from the corresponding power-temperature synchronization data sequence; Based on the dynamic response power input sequence and the dynamic response temperature output sequence, calculating a dynamic response characteristic parameter describing its response characteristics, and recording its reference baseline value; Continuously recording the dynamic response characteristic parameters to construct a dynamic response characteristic parameter sequence; Determining whether the dynamic response characteristic parameter sequence has a one-time deviation and stable change pattern, so as to identify a benign response change caused by a hardware characteristic change; After identifying the benign response change, updating the reference baseline value of the dynamic response characteristic parameter; It is determined whether the dynamic response characteristic parameter sequence has a change pattern of continuous deviation or increased fluctuation to identify the real fault in the temperature measurement signal transmission link.

2. A method for processing fan temperature measurement data according to claim 1, characterized in that: The determining whether the dynamic response characteristic parameter sequence has a one-time deviation and a stable change pattern includes: After identifying the dynamic response working condition, intercepting the dynamic response power input sequence and the dynamic response temperature output sequence within a fixed time window to construct a response morphology segment; Normalizing the response morphology segment and calculating the difference with the historical benchmark morphology to obtain a deviation index; Adding the deviation index to the dynamic response characteristic parameter sequence; When it is determined that the average value of the deviation indicator suddenly changes within a preset time period and the fluctuation amplitude is lower than a preset fluctuation threshold, it is identified as the benign response change caused by the change of hardware characteristics.

3. A method for processing fan temperature measurement data according to claim 2, characterized in that: After identifying the benign response change, updating the reference baseline value of the dynamic response characteristic parameter includes: After identifying the benign response change, obtaining a plurality of subsequent response morphology segments and performing normalization processing; Performing stability evaluation on the plurality of normalized response morphology fragments to screen out sample fragments that meet preset stability conditions; Aggregating the sample segments to generate an updated response reference form; An updated reference reference value of the dynamic response characteristic parameter is calculated based on the response reference form.

4. A method for processing fan temperature measurement data according to claim 1, characterized in that: The determining whether the dynamic response characteristic parameter sequence has a change pattern of persistent deviation or increased fluctuation to identify a real fault in the temperature measurement signal transmission link includes: Performing sliding segmentation processing on the dynamic response characteristic parameter sequence to construct a plurality of dynamic response characteristic parameter segments; respectively calculating the mean and fluctuation amplitude of each of the dynamic response characteristic parameter segments; When it is determined that the mean value continuously deviates from the reference baseline value, or when the fluctuation amplitude continuously exceeds a preset fluctuation threshold, the presence of the real fault is identified.

5. The method for processing fan temperature measurement data according to claim 1, characterized in that: The extracting the dynamic response power input sequence and the dynamic response temperature output sequence from the corresponding power-temperature synchronization data sequence includes: Obtain peak load control instructions for wind turbine generator sets; Calculating the rate of change of the real-time active power data within a set time window; When the peak shaving control instruction is in an activated state and the change rate exceeds a preset change threshold, determining that the current period is the dynamic response operating condition; A power input sequence and a temperature output sequence corresponding to the power-temperature synchronization data sequence in the current time period are extracted as the dynamic response power input sequence and the dynamic response temperature output sequence.

6. A method for processing fan temperature measurement data according to claim 4, characterized in that: The determining whether the mean value continuously deviates from the reference benchmark value, or when the fluctuation amplitude continuously exceeds a preset fluctuation threshold, includes: performing statistical process monitoring on the dynamic response characteristic parameter segments to establish control limits for the reference baseline values; Determining whether the mean or fluctuation amplitude of each of the dynamic response characteristic parameter segments continuously exceeds the control limit, or whether it presents a non-random deviation trend; When any one of the judgment conditions is met, it is determined that there is a change pattern of persistent deviation or increased fluctuation.

7. A method for processing fan temperature measurement data according to claim 6, characterized in that: The performing statistical process monitoring on the dynamic response characteristic parameter segment to establish the control limit of the reference benchmark value includes: Extracting the dynamic response characteristic parameter segments within multiple historical time periods and calculating their long-term mean and fluctuation standard deviation; Constructing an initial control limit interval based on the long-term mean and fluctuation standard deviation; Performing trend analysis on the dynamic response characteristic parameter segment in the current time period, and when determining that a stable drift mode exists, updating the long-term mean and fluctuation standard deviation; The upper and lower thresholds of the initial control limit interval are dynamically corrected based on the updated long-term mean and fluctuation standard deviation to serve as the control limits.

8. A method for processing fan temperature measurement data according to claim 7, characterized in that: The performing trend analysis on the dynamic response characteristic parameter segment in the current time period includes: Calculating the mean and fluctuation range of the dynamic response characteristic parameter segment in the current time period; Comparing the mean and the fluctuation range of multiple consecutive time periods using a sliding window method to identify trend evolution patterns; When the mean value continues to be in a preset slow-changing state and the fluctuation amplitude is within a preset stable range, the trend evolution mode is identified as a drift mode.

9. A method for processing fan temperature measurement data according to claim 7, characterized in that: When a stable drift mode is determined, the long-term mean and fluctuation standard deviation are updated, including: Based on the dynamic response characteristic parameter fragments in multiple current time periods, recalculate their long-term averages using a sliding average or exponential weighting method to generate an updated reference benchmark value; The upper and lower thresholds of the initial control limit interval are adjusted according to the updated reference benchmark value and the fluctuation standard deviation.

10. A fan temperature measurement data processing system, characterized in that: The system comprises: A data acquisition module is used to obtain real-time active power data of the wind turbine generator set and real-time temperature data of the generator winding; A synchronization processing module, configured to perform time alignment processing on the collected real-time active power data and the real-time temperature data to construct a power-temperature synchronization data sequence; an operating condition identification and recording module for identifying a dynamic response operating condition having significant power fluctuation characteristics based on the power-temperature synchronization data sequence, and extracting a dynamic response power input sequence and a dynamic response temperature output sequence from the corresponding power-temperature synchronization data sequence; a parameter calculation module, which calculates dynamic response characteristic parameters describing the response characteristics based on the dynamic response power input sequence and the dynamic response temperature output sequence, and records the reference baseline values ​​thereof; A parameter tracking module, configured to continuously record the dynamic response characteristic parameters and construct a dynamic response characteristic parameter sequence; a benign change identification module, configured to determine whether the dynamic response characteristic parameter sequence has a one-time deviation and a stable change pattern, so as to identify benign response changes caused by hardware characteristic changes; a benchmark updating module, configured to update a reference benchmark value of the dynamic response characteristic parameter after identifying the benign response change; The real fault identification module is used to determine whether the dynamic response characteristic parameter sequence has a change pattern of continuous deviation or increased fluctuation, so as to identify the real fault in the temperature measurement signal transmission link.

Images