Fan state monitoring and fault judging method based on parameter sequence analysis
Through the fan status monitoring method based on parameter sequence analysis, a baseline parameter sequence is constructed and the real-time difference value is calculated to identify abnormal parameters, which solves the problem of difficulty in diagnosing faults in traditional fan monitoring methods and improves the reliability of fan operation and maintenance efficiency.
Patent Information
- Application Number
- CN202510880642.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-27
- Publication Date
- 2025-09-12
AI Technical Summary
Traditional fan monitoring methods have difficulty capturing subtle changes in the fan's operating status in real time, and are unable to accurately diagnose the type and extent of faults, making it difficult to ensure equipment reliability and stability.
The wind turbine condition monitoring method based on parameter sequence analysis constructs a baseline parameter sequence, calculates the real-time difference value and compares it with the threshold range, identifies abnormal parameters, and analyzes the similarity between historical and real-time abnormal parameters to achieve fault diagnosis.
It has improved the reliability and safety of wind turbine operation, reduced maintenance costs, improved the pertinence and efficiency of operation and maintenance, and optimized the management level of wind farms.
Smart Images
Figure CN120626428A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of wind turbine monitoring, and in particular to a method for wind turbine status monitoring and fault diagnosis based on parameter sequence analysis. Background Art
[0002] As the global energy structure accelerates its transformation, wind power, with its significant advantages as a clean and renewable energy source, has become a key development direction in the energy sector, with its installed capacity showing rapid growth over the past few decades. As the core equipment of wind power generation systems, wind turbines are exposed to the natural environment year-round, facing extremely complex and harsh operating conditions.
[0003] Traditional wind turbine monitoring methods, such as regular manual inspections, have significant limitations. Their detection cycles are long, making it difficult to capture subtle changes in the wind turbine's operating status in real time. Problems are often only detected after a fault has occurred, and they cannot meet the stringent requirements of modern wind power generation for equipment reliability and stability. Monitoring methods based on single parameters or simple threshold judgments, due to their lack of comprehensive analysis capabilities of the wind turbine's overall operating parameters, are prone to false alarms or missed alarms, making it impossible to accurately diagnose the type and extent of faults, making it difficult to effectively ensure efficient operation and timely maintenance of the wind turbine. Against this backdrop, the development of a more advanced, accurate, and efficient method for wind turbine status monitoring and fault diagnosis is particularly urgent. Summary of the Invention
[0004] To address the shortcomings of existing technologies, this paper proposes a method for fan status monitoring and fault diagnosis based on parameter sequence analysis. This method accurately collects fan operating parameters and, by rationally setting thresholds and analyzing parameter sequence similarities, enables timely and effective monitoring of fan status and fault diagnosis, improving the reliability and safety of fan operation and reducing maintenance costs.
[0005] The technical solutions of the present invention are as follows:
[0006] A method for monitoring and diagnosing a fan state based on parameter sequence analysis, comprising:
[0007] According to the design standards, various benchmark parameters of the wind turbine under normal operating conditions are obtained to construct a wind turbine benchmark parameter sequence; historical operating parameters of the wind turbine are retrieved, the historical abnormal operating parameter sequence of the wind turbine under abnormal conditions is excluded, and the historical normal operating parameters are organized into a historical normal operating parameter sequence;
[0008] Compare the historical normal operation parameter sequence with the wind turbine benchmark parameter sequence to obtain the historical difference values of each parameter during the historical operation of the wind turbine, select the maximum and minimum real-time difference values of each parameter as the state judgment threshold, and construct the threshold range sequence;
[0009] Acquire real-time operating parameters from various sensors of the wind turbine, organize the real-time operating parameters into a real-time operating parameter sequence, and calculate the real-time difference values of various parameters of the wind turbine under the real-time operating state based on the real-time operating parameter sequence and the wind turbine benchmark parameter sequence;
[0010] Compare the real-time difference value and the threshold range sequence, identify the real-time operating parameters corresponding to the real-time difference value that exceeds the threshold range sequence as abnormal parameters, extract the abnormal parameters from the real-time operating parameter sequence according to the time series, and form a real-time abnormal state parameter sequence;
[0011] The similarity between the historical abnormal operation parameter sequence and the real-time abnormal state parameter sequence of the wind turbine under abnormal state is analyzed to diagnose the fault state of the wind turbine.
[0012] In a possible improvement of the wind turbine condition monitoring and fault diagnosis method based on parameter sequence analysis, the various benchmark parameters include: rated speed, maximum speed, maximum torque, rated torque, vibration limit, electrical parameters and control parameters. The constructed wind turbine benchmark parameter sequence is: in represents the i-th benchmark parameter, and n represents the number of benchmark parameters;
[0013] The historical normal operating parameter sequence is: in Represents the i-th historical parameter.
[0014] The real-time operating parameter sequence is: in, Represents the i-th real-time operating parameter.
[0015] In a possible improvement of the wind turbine condition monitoring and fault diagnosis method based on parameter sequence analysis, the positions of the parameters in the wind turbine benchmark parameter sequence, the historical normal operating parameter sequence, and the threshold range sequence correspond one to one.
[0016] In a possible improvement of the wind turbine state monitoring and fault diagnosis method based on parameter sequence analysis, the historical difference value calculation formula of each parameter of the wind turbine during the historical operation process is:
[0017]
[0018] in, is the historical difference value during the historical operation process, is the parameter in the historical normal operation parameter sequence, is the corresponding benchmark parameter;
[0019] Based on the historical difference values of each parameter, the difference value sequence of each parameter is obtained in, Represents the i-th historical difference value.
[0020] A method for monitoring and fault diagnosis of a fan state based on parameter sequence analysis is proposed. In one possible improvement, the maximum historical difference value of each parameter is found according to the difference value sequence. and the minimum historical difference The maximum and minimum historical difference values of each parameter are combined into a threshold range sequence
[0021]
[0022] In a possible improvement of the wind turbine state monitoring and fault diagnosis method based on parameter sequence analysis, the real-time difference value under the real-time operating state is calculated using the following formula:
[0023]
[0024] in, is the real-time difference value in real-time running state, For real-time operation parameter sequence P real Each real-time operating parameter in;
[0025] By calculating the real-time difference value in real-time running state, the real-time difference value sequence is obtained. in, Represents the i-th real-time difference value.
[0026] In a possible improvement of the wind turbine state monitoring and fault diagnosis method based on parameter sequence analysis, the method for forming a real-time abnormal state parameter sequence includes:
[0027] Compare the real-time difference value with the threshold, that is, the real-time difference value of each parameter The threshold range corresponding to the threshold range sequence T Compare them one by one; if Then the real-time operating parameter is determined to be an abnormal parameter;
[0028] Extract abnormal parameters from the real-time operation parameter sequence according to the time series to form a real-time abnormal state parameter sequence Among them, t j is the timestamp of abnormal parameter collection, and j is the number of abnormal parameters.
[0029] In a possible improvement of the wind turbine state monitoring and fault diagnosis method based on parameter sequence analysis, the method for calculating the similarity between the historical abnormal parameters and the real-time abnormal parameters includes Euclidean distance or cosine similarity.
[0030] In a possible improvement of the wind turbine state monitoring and fault judgment method based on parameter sequence analysis, the fault state of the wind turbine is diagnosed by performing real-time abnormal parameter fault type judgment based on similarity measurement results and the fault type corresponding to the historical abnormal state parameter sequence.
[0031] The beneficial effects brought about by the technical solutions provided in the embodiments of the present application include at least the following:
[0032] Beneficial effects:
[0033] 1. By accurately collecting various baseline parameters of wind turbines in normal operation, constructing a comprehensive baseline parameter sequence, and meticulously screening and organizing historical operating parameters, a solid foundation is provided for determining wind turbine operating status. During wind turbine operation, the difference between real-time operating parameters and baseline parameters is continuously calculated and compared, enabling the timely detection of abnormal parameters exceeding the threshold range. This real-time monitoring mechanism enables operations and maintenance personnel to take action at the nascent stage of a fault, effectively preventing minor faults from escalating into major failures, thereby significantly reducing the occurrence of sudden wind turbine downtime.
[0034] 2. By enabling early detection of potential faults and accurate diagnosis of fault types, maintenance work becomes more targeted. Maintenance personnel can prepare necessary repair tools and parts in advance, reducing maintenance time and labor costs. Furthermore, accurate fault diagnosis avoids unnecessary component replacement, reducing maintenance costs.
[0035] 3. During the long-term monitoring process, the analysis of fan operating parameters not only helps to detect faults, but also discovers performance fluctuations and potential optimization space during the operation of the fan.
[0036] 4. This method provides rich parameter support and scientific decision-making basis for wind farm operation and maintenance management. Based on real-time wind turbine status and fault prediction information, wind farm managers can rationally arrange work tasks and inspection plans for operation and maintenance personnel, achieving optimal resource allocation. Furthermore, through comprehensive analysis of the operating parameters of multiple wind turbines, common problems or potential risks within the wind farm can be identified, allowing for the development of proactive countermeasures, improving the overall management level and risk prevention capabilities of the wind farm and ensuring its safe and stable operation. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 This is a schematic diagram of the overall process of a wind turbine condition monitoring and fault diagnosis method based on parameter sequence analysis;
[0038] Figure 2 Detailed flow chart of step S100 of a method for monitoring and diagnosing a wind turbine state based on parameter sequence analysis;
[0039] Figure 3 Detailed flow chart of step S200 of a method for monitoring and diagnosing a wind turbine state based on parameter sequence analysis;
[0040] Figure 4 Detailed flow chart of step S300 of a method for monitoring and diagnosing a wind turbine state based on parameter sequence analysis;
[0041] Figure 5 Detailed flow chart of step S400 of a method for monitoring and diagnosing a wind turbine state based on parameter sequence analysis;
[0042] Figure 6 This is a detailed flow chart of step S500 of a method for monitoring and diagnosing a fan status based on parameter sequence analysis. DETAILED DESCRIPTION
[0043] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of this application.
[0044] With the rapid development of the wind power industry, the reliable operation of wind turbines is crucial. Wind turbines, operating in complex natural environments for long periods of time, are prone to various faults. Failure to promptly monitor and diagnose these faults can lead to reduced power generation efficiency, equipment damage, or even downtime, resulting in significant economic losses. Traditional wind turbine monitoring methods have limitations, making it difficult to accurately and efficiently identify early-stage faults and complex fault types.
[0045] To resolve the above issues, please refer to Figure 1 , which shows a method for monitoring and diagnosing a wind turbine state based on parameter sequence analysis provided by an embodiment of the present invention, the method comprising:
[0046] S100: Obtain various benchmark parameters of the wind turbine under normal operating conditions according to design standards, and construct a benchmark parameter sequence of the wind turbine; retrieve historical operating parameters of the wind turbine, exclude historical abnormal operating parameter sequences of the wind turbine under abnormal conditions, and organize the historical normal operating parameters into a historical normal operating parameter sequence.
[0047] S200: Compare the historical normal operation parameter sequence and the wind turbine benchmark parameter sequence to obtain the historical difference values of each parameter during the historical operation of the wind turbine, select the maximum and minimum real-time difference values of each parameter as the state judgment threshold, and construct a threshold range sequence.
[0048] S300: Acquire real-time operating parameters from various sensors of the wind turbine, organize the real-time operating parameters into a real-time operating parameter sequence, and calculate real-time difference values of various parameters of the wind turbine in the real-time operating state based on the real-time operating parameter sequence and the wind turbine benchmark parameter sequence.
[0049] S400: Compare the real-time difference value and the threshold range sequence, identify the real-time operating parameters corresponding to the real-time difference value exceeding the threshold range sequence as abnormal parameters, extract the abnormal parameters from the real-time operating parameter sequence according to the time series, and form a real-time abnormal state parameter sequence.
[0050] S500: Analyze the similarity between the historical abnormal operation parameter sequence and the real-time abnormal state parameter sequence of the wind turbine in an abnormal state, and diagnose the fault state of the wind turbine.
[0051] Based on design standards, the S100 selects multiple benchmark parameters, such as the wind turbine's rated speed and maximum speed, to construct a benchmark parameter sequence. It also retrieves parameters containing timestamps, parameter values, and operating status tags from a historical parameter library. It then pre-processes and selects historical normal operating parameters based on operating status, organizing them into a historical normal operating parameter sequence that corresponds one-to-one with the benchmark parameter sequence, laying the foundation for subsequent analysis.
[0052] Please refer to Figure 2 , which shows a flowchart of an exemplary method S100 of wind turbine status monitoring and fault diagnosis based on parameter sequence analysis of the present application, including:
[0053] S110: Obtaining benchmark parameters according to design standards and constructing a wind turbine benchmark parameter sequence.
[0054] The benchmark parameters of the wind turbine determined according to the design standards include but are not limited to: rated speed, maximum speed, maximum torque, rated torque, vibration limits, electrical parameters and control parameters.
[0055] Constructing a wind turbine benchmark parameter sequence in represents the i-th benchmark parameter.
[0056] S120: Retrieve historical operating parameters of the wind turbine generator, filter and sort the historical operating parameters according to the historical operating status of the wind turbine generator, and form a historical normal operating parameter sequence.
[0057] Parameters are retrieved from a historical parameter library accumulated over the long-term operation of the wind turbine. This library should contain detailed timestamps, parameter values, and operating status tags. The operating status is typically marked as normal or abnormal and can be obtained from the wind turbine's own fault diagnosis system or records kept by operation and maintenance personnel. The operating status tags are used to filter out abnormal operating parameter sequences in historical conditions and exclude them. The remaining historical normal operating parameters are also preprocessed, such as parameter cleaning and outlier processing.
[0058] Then organize it into a historical normal operating parameter sequence in Represents the i-th historical parameter.
[0059] Ensure that the parameter positions in the historical normal operation parameter sequence correspond one-to-one with the parameter positions in the benchmark parameter sequence.
[0060] S200 first calculates the historical difference between normal operating parameters and baseline parameters using a specific formula. It then traverses all parameter points to obtain a sequence of difference values, reflecting the historical fluctuations of the parameters. It then uses a sorting and comparison algorithm to calculate the maximum and minimum historical difference values of each parameter difference sequence. Finally, it combines these values into a sequence of threshold ranges, which serves as the key basis for determining whether the real-time operating parameters are abnormal.
[0061] Please refer to Figure 3 , which shows a flowchart of an exemplary method S200 of wind turbine status monitoring and fault diagnosis based on parameter sequence analysis of the present application, including:
[0062] S210: Calculate the difference value.
[0063] For each parameter in the historical normal operating parameter sequence The corresponding benchmark parameters According to the formula: Calculate the historical difference value during the historical operation
[0064] The process of calculating the real-time difference value of each parameter during the historical operation process needs to traverse all parameter points in the historical normal operation parameter sequence.
[0065] By calculating a large number of historical parameter points one by one, the difference value sequence of each parameter can be obtained in, Represents the i-th historical difference value.
[0066] The difference value sequence can reflect the fluctuation of each parameter relative to the benchmark value during the historical operation process, and provide basic parameters for the determination of subsequent thresholds.
[0067] S220: Maximum and minimum difference value statistics.
[0068] After obtaining the difference value sequence of each parameter, it is necessary to traverse these sequences to find its maximum historical difference value and the minimum historical difference
[0069] In a possible implementation, parameter statistics are efficiently implemented with the help of a sorting algorithm and a comparison algorithm in a computer program.
[0070] S230: constructing a threshold range sequence.
[0071] The maximum and minimum historical difference values of each parameter are combined into a threshold range sequence
[0072]
[0073] The threshold range sequence is the core criterion for subsequently determining whether the real-time operating parameters are abnormal.
[0074] When the S300 wind turbine is operating, sensors collect parameters and transmit them to the monitoring center's processing unit, which compiles them into a real-time operating parameter sequence. A formula is then used to calculate the real-time difference between each real-time operating parameter and the corresponding baseline parameter in the sequence. This difference value sequence is then generated for subsequent comparison with the threshold.
[0075] Please refer to Figure 4 , which shows a flowchart of an exemplary method S300 of wind turbine status monitoring and fault diagnosis based on parameter sequence analysis of the present application, including:
[0076] S310: Collect wind turbine parameters in real time and build a real-time operating parameter sequence.
[0077] During the operation of the fan, the sensor continuously collects parameters and transmits the parameters in real time to the real-time parameter processing unit of the monitoring center through the established parameter transmission channel.
[0078] After receiving the parameters, the real-time parameter processing unit organizes the collected real-time operation parameters to form a real-time operation parameter sequence. in, Represents the i-th real-time operating parameter.
[0079] S320: Calculate the real-time difference value.
[0080] For real-time operation parameter sequence P real Each real-time operating parameter in The corresponding benchmark parameters Calculate the real-time difference value in real-time running state The calculation formula is:
[0081] Repeat this calculation process for all real-time operating parameters to obtain a real-time difference value sequence in, Represents the i-th real-time difference value.
[0082] The S400 compares each parameter's real-time difference value with the corresponding threshold range in the threshold range sequence. Parameters below the minimum or above the maximum are identified as abnormal. A binary search algorithm improves efficiency, and multi-parameter comparisons employ multi-threading or parallel computing. After identification, the system extracts abnormal parameters and related information based on the time series, constructing a real-time abnormal state parameter sequence to assist in fault analysis.
[0083] Please refer to Figure 5 , which shows a flowchart of an exemplary method S400 of wind turbine status monitoring and fault diagnosis based on parameter sequence analysis of the present application, including:
[0084] S410: Compare the real-time difference value with the threshold.
[0085] The real-time difference value of each parameter The threshold range corresponding to the threshold range sequence T Compare them one by one. or The real-time operating parameter is determined to be an abnormal parameter.
[0086] In a possible implementation, a binary search algorithm is used to search for the corresponding threshold range in the threshold range sequence, and then a comparison is performed.
[0087] For example, taking the bearing temperature parameter as an example, if its real-time difference value is The threshold range is because Therefore, the bearing temperature parameter at this moment is identified as an abnormal parameter.
[0088] In a possible implementation, when comparing multiple parameters simultaneously, multi-threading or parallel computing technology is used to fully utilize the computer's multi-core processor resources, speed up the comparison, and ensure that abnormal parameters can be discovered in a timely manner.
[0089] S420: Abnormal parameter extraction and real-time abnormal state parameter sequence construction.
[0090] When the abnormal parameters are identified, the abnormal parameters are extracted from the real-time operation parameter sequence according to the time series to form a real-time abnormal state parameter sequence. Among them, t j is the timestamp of abnormal parameter collection, and j is the number of abnormal parameters.
[0091] For example, if the wind speed and bearing temperature parameters are found to be abnormal at a certain moment, the real-time abnormal state parameter sequence is:
[0092] and
[0093]
[0094] When extracting abnormal parameters, it is necessary to accurately record the parameter name, parameter value and corresponding timestamp for subsequent fault analysis and diagnosis.
[0095] S500 determines the similarity between historical abnormal parameters and real-time abnormal parameters, and combines the similarity results, historical fault types and wind turbine operating conditions to determine the fault type, severity and cause of real-time abnormal parameters, and generates a report to provide decision support for operation and maintenance.
[0096] Please refer to Figure 6 , which shows a flowchart of an exemplary method S500 of wind turbine status monitoring and fault diagnosis based on parameter sequence analysis of the present application, including:
[0097] S510: Determine the similarity between the historical abnormal parameters and the real-time abnormal parameters.
[0098] Real-time abnormal status parameter sequence of a certain parameter Determine the similarity between historical abnormal parameters and real-time abnormal parameters.
[0099] In one possible implementation, the Euclidean distance is used to determine the similarity between the historical anomaly parameters and the real-time anomaly parameters:
[0100] Assume that the historical operating parameter sequence is Real-time abnormal status parameter sequence Then the Euclidean distance d is:
[0101]
[0102] The smaller the distance, the higher the similarity between the two.
[0103] In one possible implementation, cosine similarity is used to determine the similarity between historical anomaly parameters and real-time anomaly parameters:
[0104] Assume that the historical operating parameter sequence is Real-time abnormal status parameter sequence Then the cosine similarity is:
[0105]
[0106] Its value is between -1 and 1, and the closer the value is to 1, the higher the similarity.
[0107] S520: Fault type determination.
[0108] The real-time abnormal parameter fault type is judged based on the similarity measurement results and the fault type corresponding to the historical abnormal state parameter sequence.
[0109] For example, if the Euclidean distance between a real-time abnormal state parameter sequence and a historical blade imbalance fault parameter sequence is less than a set threshold, and its spectral characteristics also conform to the characteristics of a blade imbalance fault (such as a large amplitude at a specific frequency), it can be determined that the wind turbine may currently have a blade imbalance fault. At the same time, combined with the operating conditions of the wind turbine (such as wind speed, load, etc.) and factors such as the frequency and duration of the fault, the severity and possible causes of the fault are further determined. If a wind turbine frequently experiences a certain fault at high wind speeds, and the duration of the fault gradually increases, it may mean that the fault is more serious and requires timely shutdown and maintenance.
[0110] After determining the fault type and severity, a detailed fault report can be generated, including fault time, fault type, possible causes and recommended repair measures, etc., to provide decision support for operation and maintenance personnel so that fan fault problems can be solved quickly and effectively to ensure the safe and stable operation of the fan.
[0111] The basic principles of the present application have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, and effects mentioned in this application are merely illustrative and not restrictive, and it should not be assumed that these advantages, strengths, and effects are required of each embodiment of this application. In addition, the specific details disclosed above are merely illustrative and facilitating understanding, and are not restrictive. The above details do not limit this application to necessarily being implemented using the above specific details.
[0112] The block diagrams of the devices, devices, equipment, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, devices, equipment, and systems can be connected, arranged, or configured in any manner. Words such as "include," "comprise," "have," and the like are open-ended words, meaning "including but not limited to," and can be used interchangeably therewith. The words "or" and "and" used herein refer to the words "and / or" and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to," and can be used interchangeably therewith.
[0113] It should also be noted that in the apparatus, device, and method of the present application, each component or each step can be decomposed and / or recombined, and such decomposition and / or recombination should be regarded as equivalent solutions of the present application.
[0114] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to be applied in the widest sense consistent with the principles and novel features of the present invention.
[0115] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.
Claims
1. A method for monitoring and fault diagnosis of a fan state based on parameter sequence analysis, characterized in that: include: Obtain various benchmark parameters of the wind turbine under normal operating conditions according to the design standards and construct a wind turbine benchmark parameter sequence; Retrieving historical operating parameters of the wind turbine, excluding the historical abnormal operating parameter sequence of the wind turbine in an abnormal state, and arranging the historical normal operating parameters into a historical normal operating parameter sequence; Compare the historical normal operation parameter sequence with the wind turbine benchmark parameter sequence to obtain the historical difference values of each parameter during the historical operation of the wind turbine, select the maximum and minimum real-time difference values of each parameter as the state judgment threshold, and construct the threshold range sequence; Acquire real-time operating parameters from various sensors of the wind turbine, organize the real-time operating parameters into a real-time operating parameter sequence, and calculate the real-time difference values of various parameters of the wind turbine under the real-time operating state based on the real-time operating parameter sequence and the wind turbine benchmark parameter sequence; Compare the real-time difference value and the threshold range sequence, identify the real-time operating parameters corresponding to the real-time difference value that exceeds the threshold range sequence as abnormal parameters, extract the abnormal parameters from the real-time operating parameter sequence according to the time series, and form a real-time abnormal state parameter sequence; The similarity between the historical abnormal operation parameter sequence and the real-time abnormal state parameter sequence of the wind turbine under abnormal state is analyzed to diagnose the fault state of the wind turbine.
2. A method for monitoring and diagnosing a wind turbine state based on parameter sequence analysis according to claim 1, characterized in that: The various benchmark parameters include: rated speed, maximum speed, maximum torque, rated torque, vibration limit, electrical parameters and control parameters. The constructed wind turbine benchmark parameter sequence is: Among them, P i base represents the i-th benchmark parameter, and n represents the number of benchmark parameters; The historical normal operating parameter sequence is: Among them, P i hist represents the i-th historical parameter; The real-time operating parameter sequence is: Among them, P i real Represents the i-th real-time operating parameter.
3. The method for monitoring and diagnosing a wind turbine state based on parameter sequence analysis according to claim 2, characterized in that: The positions of the parameters in the wind turbine benchmark parameter sequence, the historical normal operation parameter sequence, and the threshold range sequence correspond one to one.
4. The method for monitoring and diagnosing a wind turbine state based on parameter sequence analysis according to claim 1, characterized in that: The calculation formula for the historical difference values of various parameters of the wind turbine during its historical operation is: ΔP i hist =P i hist -P i base ; Where ΔP i hist is the historical difference value during the historical operation process, P i hist is the parameter in the historical normal operation parameter sequence, P i base is the corresponding benchmark parameter; Based on the historical difference values of each parameter, the difference value sequence of each parameter is obtained Where ΔP i hist Represents the i-th historical difference value.
5. A method for monitoring and diagnosing a wind turbine state based on parameter sequence analysis according to claim 4, characterized in that: Find the maximum historical difference value ΔP of each parameter according to the difference value sequence i max and the minimum historical difference value ΔP i min , the maximum and minimum historical difference values of each parameter are combined into a threshold range sequence 6. The method for monitoring and diagnosing a wind turbine state based on parameter sequence analysis according to claim 2, characterized in that: The real-time difference value under the real-time running state is calculated as follows: ΔP i real =P i real -P i base ; Where ΔP i real is the real-time difference value under real-time running state, P i real For real-time operation parameter sequence P real Each real-time operating parameter in; By calculating the real-time difference value in real-time running state, the real-time difference value sequence is obtained. Where ΔP i real Represents the i-th real-time difference value.
7. A method for monitoring and diagnosing a wind turbine state based on parameter sequence analysis according to claim 6, characterized in that: The method for forming a real-time abnormal state parameter sequence includes: The real-time difference value is compared with the threshold, that is, the real-time difference value ΔP of each parameter i real The threshold range corresponding to the threshold range sequence T [ΔP i min ,ΔP i max ] compare them one by one; if ΔP i real <ΔP i min or ΔP i real >ΔP i max , then the real-time operating parameter is determined to be an abnormal parameter; Extract abnormal parameters from the real-time operation parameter sequence according to the time series to form a real-time abnormal state parameter sequence Among them, t j is the timestamp of abnormal parameter collection, and j is the number of abnormal parameters.
8. The method for monitoring and diagnosing a wind turbine state based on parameter sequence analysis according to claim 7, characterized in that: The method for calculating the similarity between the historical abnormal parameters and the real-time abnormal parameters includes Euclidean distance or cosine similarity.
9. The method for monitoring and diagnosing a wind turbine state based on parameter sequence analysis according to claim 8, characterized in that: The fault state of the wind turbine is diagnosed in a manner as follows: performing real-time abnormal parameter fault type judgment based on similarity measurement results and fault types corresponding to historical abnormal state parameter sequences.