Wind turbine health condition assessment method

Abstract

The application provides a wind turbine health state evaluation method, and belongs to the technical field of intelligent evaluation, and comprises the following steps: performing data cleaning on historical operation information of an interactive wind turbine to obtain a data cleaning result, reading state evaluation indexes of the wind turbine, constructing a state evaluation index set and performing index classification to obtain an index classification result, performing data interaction on the wind turbine, performing credible evaluation after index data corresponding to the state evaluation index set is collected and obtained, setting a credible coefficient of the index data and performing data integration calculation on the index data, performing classification degradation degree calculation according to the index classification result and the data cleaning result respectively, and then performing comprehensive degradation degree calculation, and generating a health state evaluation result of the wind turbine through the comprehensive degradation degree calculation result, and the application solves the technical problem that the state of the wind turbine is not evaluated in the prior art, so that the health state of the wind turbine cannot be ensured.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent evaluation, and particularly relates to a wind turbine health state evaluation method. BACKGROUND

[0002] With the advent of the era of electric power big data, a large number of high-speed real-time data streams are becoming more and more common, and big data analysis technology provides more stable and powerful data analysis capabilities for the development of the electric power industry. Effective dynamic data streams provide rich state information and decision support information for the health state evaluation of the monitored object, but unlike traditional analysis methods, the analysis method based on real-time data streams has higher requirements for algorithm execution efficiency. Therefore, the research on the real-time processing method of wind turbine monitoring data streams is of great significance for the stability and safety of wind power generation equipment, and for changing the maintenance strategy of the wind turbine from traditional abnormal state monitoring to health management.

[0003] Traditional equipment health evaluation methods have achieved many results, but they are all based on static data sets. Even if there is real-time evaluation, it is based on small-scale data. When the wind turbine state monitoring data stream is continuous and high-speed, these methods will be forced to discard a lot of information to achieve fast processing as much as possible, and the result is to reduce the accuracy of the algorithm. Therefore, the traditional method is not suitable for processing real-time data streams. In addition, the existing methods are generally based on fuzzy theory methods, only considering the fuzziness in the uncertainty of information, and not considering the randomness of information. The wind turbine operating conditions are complex and changeable, and are affected by the uncertainty of wind speed, the uncertainty in wind power conversion, and the internal and external uncertainty of the wind turbine system, so that the health state also has the characteristics of uncertainty.

[0004] However, the prior art lacks evaluation of the state of the wind turbine, resulting in the technical problem that the health state of the wind turbine cannot be guaranteed. SUMMARY

[0005] The application provides a wind turbine health state evaluation method, which solves the technical problem that the state of the wind turbine is not evaluated, resulting in the health state of the wind turbine cannot be guaranteed.

[0006] The application provides a wind turbine health state evaluation method, which includes: interacting with the historical operation information of the wind turbine, and performing data cleaning on the historical operation information to obtain a data cleaning result;

[0007] Read the state evaluation index of the wind turbine, and construct a state evaluation index set. Perform index classification on the state evaluation index set to obtain an index classification result;

[0008] Data interaction is performed on the wind turbine to collect index data corresponding to the set of state evaluation indexes;

[0009] The index data is subjected to a credibility evaluation, and a credibility coefficient of the index data is set;

[0010] Based on the credibility coefficient, data integration calculation is performed on the index data, and classification degradation degree calculation is performed according to the index classification result and the data cleaning result respectively to obtain a classification degradation degree calculation result;

[0011] Based on the classification degradation degree calculation result, comprehensive degradation degree calculation is performed, and a health state evaluation result of the wind turbine is generated through the comprehensive degradation degree calculation result.

[0012] By adopting the above technical solution, including:

[0013] When performing classification degradation degree calculation, it is judged whether the index classification result corresponding to the data is a small optimal index classification;

[0014] When the index classification result is a small optimal index classification, degradation degree calculation under the current classification is performed through a formula, and the calculation formula is as follows:

[0015]

[0016] wherein k is any index data, δ is the credibility coefficient of the index data k, k min is the lower limit value of the evaluation index, and k max is the upper limit value of the evaluation index.

[0017] By adopting the above technical solution, including:

[0018] When the index classification result corresponding to the data is not a small optimal index classification, degradation degree calculation is performed through a formula, and the calculation formula is as follows:

[0019]

[0020] wherein [k a , k b ] is the best operating range of the evaluation index.

[0021] By adopting the above technical solution, including:

[0022] An obvious degradation degree threshold is set;

[0023] After all classification degradation degree calculations are completed, m offset level threshold values of offset coefficients are set according to the total number n of evaluation indexes;

[0024] Based on the classification degradation degree calculation result and the obvious degradation degree threshold, an obvious degradation number is obtained;

[0025] By the obvious degradation quantity matching the offset level threshold, a matching offset coefficient corresponding to the level is obtained;

[0026] The comprehensive degradation degree is calculated according to the matching offset coefficient and the classification degradation degree calculation result.

[0027] By adopting the technical solution, including:

[0028] The comprehensive degradation degree is obtained by formula calculation, and the calculation formula is as follows:

[0029]

[0030] Wherein, n is the total number of evaluation indexes, g k is the classification degradation degree calculation result of any one index, and ζ is an offset coefficient.

[0031] By adopting the technical solution, including:

[0032] The wind turbine is subjected to unit feature extraction according to the data cleaning result, and a unit feature extraction result is obtained;

[0033] The key features are determined according to the unit feature extraction result, and a credible constraint value of the key features is set;

[0034] The classification degradation degree is calculated according to the key features and the credible constraint value.

[0035] By adopting the technical solution, including:

[0036] It is judged whether the credible coefficient of the index data corresponding to the key features meets the credible constraint value;

[0037] When it cannot be met, the index data collection is re-performed until the index coefficients of all the key features meet the credible constraint value, and the collection is completed;

[0038] The key features are subjected to feature assignment;

[0039] The classification degradation degree is calculated according to the re-collected index data and the key feature assignment result.

[0040] The embodiment of the application provides a wind turbine health state evaluation system, including: a data cleaning module, the data cleaning module is used for interacting historical running information of a wind turbine, and performing data cleaning on the historical running information to obtain a data cleaning result;

[0041] An index classification module, the index classification module is used for reading state evaluation indexes of the wind turbine, constructing a state evaluation index set, performing index classification on the state evaluation index set, and obtaining an index classification result;

[0042] a data collection module, configured to interact with the wind turbine generator to collect index data corresponding to the set of state evaluation indexes;

[0043] a trusted evaluation module, configured to evaluate the index data and set a trusted coefficient of the index data;

[0044] a classification degradation degree calculation module, configured to integrate and calculate the index data based on the trusted coefficient, and calculate a classification degradation degree based on the index classification result and the data cleaning result respectively, to obtain a classification degradation degree calculation result;

[0045] a comprehensive degradation degree calculation module, configured to calculate a comprehensive degradation degree based on the classification degradation degree calculation result, and generate a health state evaluation result of the wind turbine generator through the comprehensive degradation degree calculation result.

[0046] The present application has the following advantages:

[0047] 1. The present application performs data cleaning on historical operation information of the wind turbine generator to obtain a data cleaning result, reads state evaluation indexes of the wind turbine generator, constructs a set of state evaluation indexes, classifies the indexes to obtain an index classification result, interacts with the wind turbine generator to collect index data corresponding to the set of state evaluation indexes, evaluates the index data to set a trusted coefficient of the index data and integrate and calculate the index data, calculates a classification degradation degree based on the index classification result and the data cleaning result respectively, and then calculates a comprehensive degradation degree, to generate a health state evaluation result of the wind turbine generator through the comprehensive degradation degree calculation result, so as to realize accurate evaluation of the state of the wind turbine generator and ensure evaluation of the health state of the wind turbine generator.

[0048] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be achieved and obtained by the structure particularly pointed out in the description and the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0049] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, illustrate embodiments of the present application, and are used to explain the present application, and do not constitute a limitation of the present application. In the drawings:

[0050] Figure 1 a flowchart of a wind turbine generator health state evaluation method according to an embodiment of the present application;

[0051] Figure 2A wind turbine health state evaluation method embodiment of the present application is provided with a classification degradation degree calculation flowchart.

[0052] Figure 3 A wind turbine health state evaluation system structure diagram is provided in the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical scheme and advantages of the technical scheme of the present application more clear, the technical scheme of the present application embodiment will be described clearly and completely in the following with reference to the drawings of the present application embodiment. The same reference signs in the drawings represent the same parts. It should be noted that the described embodiment is part of the embodiment of the present application, not all the embodiments. Based on the described embodiment of the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0054] Referring to Figure 1 The wind turbine health state evaluation method embodiment of the present application is provided, which comprises: interacting historical operation information of a wind turbine, and performing data cleaning on the historical operation information to obtain a data cleaning result.

[0055] The embodiment is specifically: in use, in order to ensure the accuracy of the wind turbine health state evaluation, the data of the interactive operation process of the wind turbine in the past time needs to be extracted first. That is, the historical operation information of the wind turbine, which can include historical average wind speed data, historical effective wind time data, historical average air density data, etc. Further, the data cleaning of the data contained in the historical operation information refers to the process of re-examining and verifying the historical average wind speed data, the historical effective wind time data and the historical average air density data, which aims to delete the repeated information contained in the historical average wind speed data, the historical effective wind time data and the historical average air density data, correct the existing errors, and provide the consistency of the historical average wind speed data, the historical effective wind time data and the historical average air density data, so as to summarize the cleaned historical average wind speed data, the historical effective wind time data and the historical average air density data as the data cleaning result, which serves as an important reference for the later health state evaluation of the wind turbine.

[0056] Referring to Figure 1 The wind turbine health state evaluation method embodiment of the present application is provided, which comprises: reading state evaluation indexes of the wind turbine, and constructing a state evaluation index set, classifying the state evaluation index set to obtain an index classification result.

[0057] The embodiment is specific: in use, in order to evaluate the real-time state of the wind turbine, firstly, the state evaluation index of the current wind turbine needs to be read, the state evaluation index of the wind turbine can be constructed on the basis of the historical average wind speed data, the historical effective wind time data and the historical average air density data after data cleaning, the historical average wind speed data is the average value of the wind speed in a given time, the historical effective wind time data is the cumulative value of the wind speed between the cut-in wind speed and the cut-out wind speed at the hub height of the wind turbine, and the historical average air density data is the average value of the air density of the wind turbine in the evaluation period, so as to construct the corresponding state evaluation index of the wind turbine according to different data, and then obtain the state evaluation index set, and classify the obtained state evaluation index set according to the average wind speed, the effective wind time and the average air density, and then obtain the index classification result, so as to ensure the evaluation of the health state of the wind turbine.

[0058] With reference to Figure 1 The embodiment of the present application proposes a wind turbine health state evaluation method, which comprises: data interaction of the wind turbine, and acquisition of index data corresponding to the state evaluation index set;

[0059] The embodiment is specific: in use, the state evaluation index set constructed above is taken as the basis to perform data interaction of the real-time running state of the current wind turbine, the data interaction can be performed on the basis of the wind turbine data corresponding to each state evaluation index in the state evaluation index set, for example, the current wind speed and the wind speed evaluation index of the current state of the wind turbine are matched, the matching result is recorded as the index data corresponding to the wind speed evaluation index of the current state of the wind turbine, and further, each corresponding index data in the state evaluation index set is collected, so as to lay a foundation for subsequent evaluation of the health state of the wind turbine.

[0060] With reference to Figure 1 The embodiment of the present application proposes a wind turbine health state evaluation method, which comprises: data interaction of the wind turbine, and acquisition of index data corresponding to the state evaluation index set;

[0061] The embodiment is specific: when used, in order to ensure the accuracy of the index data corresponding to the state evaluation index set collected above, it is necessary to perform credibility evaluation on each index data corresponding to the state evaluation index set, wherein the credibility of the index data can be decomposed into availability, security, real-time performance, maintainability, survivability and other credibility attributes, and the credibility of each index data is evaluated based on the credibility attributes, so as to set the credibility coefficient of the index data. For example, each attribute in the credibility attribute accounts for 20%, if the credibility is greater than or equal to 60%, that is, the index data meets 3 and more attributes in the credibility attribute, it is determined that the corresponding index data is credible, if the credibility is less than 60%, that is, the index data only meets 2 and less attributes in the credibility attribute, it is determined that the corresponding index data is not credible, and the credibility coefficient is the credibility of the index data. The credibility coefficient has a limited effect on the health state evaluation of the wind turbine generator.

[0062] Referring to Figure 1 The embodiment of the application proposes a wind turbine generator health state evaluation method, comprising: performing data integration calculation on the index data based on the credibility coefficient, and performing classification degradation degree calculation based on the index classification result and the data cleaning result respectively to obtain a classification degradation degree calculation result.

[0063] The embodiment is specific: when used, the index data corresponding to the state evaluation index set is calculated based on the credibility coefficient set above, that is, the index data with different credibility coefficients is collected, arranged, cleaned, converted and integrated into a new data set. Further, the classification degradation degree calculation is performed based on the index classification result and the data cleaning result, the classification degradation degree calculation process can first extract the unit characteristics of the wind turbine generator through the data cleaning result, then determine the key characteristics of the wind turbine generator, calculate the classification degradation degree of the determined key characteristics and the credibility value of the key characteristics, and record the calculation result as the classification degradation degree calculation result, so as to serve as reference data for the health state evaluation of the wind turbine generator in the later stage.

[0064] Referring to Figure 1 The embodiment of the application proposes a wind turbine generator health state evaluation method, comprising: performing comprehensive degradation degree calculation based on the classification degradation degree calculation result, and generating the health state evaluation result of the wind turbine generator through the comprehensive degradation degree calculation result.

[0065] The embodiment is specifically: in use, taking the classification degradation degree calculation result as a reference, the comprehensive degradation degree calculation is carried out, first, the threshold value of the obvious degradation of the state of the wind turbine is set, when all the classification degradation degree calculations are completed, the offset level threshold value corresponding to the offset coefficient less than or equal to the total number of evaluation indexes is set according to the total number of evaluation indexes, further, the number of the classification degradation degree calculation result coinciding with the obvious degradation threshold value is taken out, and after matching with the offset level threshold value, the classification degradation degree calculation result is calculated, so that the comprehensive degradation degree is obtained, and the health state of the current wind turbine is evaluated according to the comprehensive degradation degree calculation result of the wind turbine, and the technical effect of accurately evaluating the state of the wind turbine is realized.

[0066] The embodiment of the application proposes a wind turbine health state evaluation method, comprising:

[0067] When the classification degradation degree calculation is performed, whether the index classification result corresponding to the data is a small optimal index classification is judged;

[0068] When the index classification result is a small optimal index classification, the degradation degree calculation under the current classification is performed through a formula, and the calculation formula is as follows:

[0069]

[0070] Wherein, k is any index data, δ is the confidence coefficient of the index data k, k min is the lower limit value of the evaluation index, and k max is the upper limit value of the evaluation index.

[0071] When the index classification result corresponding to the data is not a small optimal index classification, the degradation degree calculation is performed through a formula, and the calculation formula is as follows:

[0072]

[0073] Wherein, [k a ,k b ] is the best operation range of the evaluation index.

[0074] The embodiment is specifically: in use, according to the index classification result obtained by classifying the state evaluation index set and the data cleaning result obtained by executing data cleaning on the historical operation information, the classification degradation degree calculation is performed, and before the classification degradation degree calculation is performed, whether the index classification result corresponding to the data of the current wind turbine is a small optimal index classification is judged, the small optimal index is an index that is better the smaller, if the index classification result is a small optimal index classification, the degradation degree under the current classification is calculated through the following formula, and the calculation formula is as follows:

[0075]

[0076] wherein k is any index data, δ is a confidence coefficient of the index data k, k min is a lower limit value of the evaluation index, k max is an upper limit value of the evaluation index.

[0077] The current index classification data and the confidence coefficient corresponding to the index data are brought into the above formula, and the degradation degree under the current classification is calculated within the evaluation upper limit and the evaluation lower limit according to the evaluation index upper limit and the evaluation index lower limit in the above formula, so as to obtain the degradation degree when the index classification result is the smaller-better index classification.

[0078] Further, if the index classification result corresponding to the data of the current wind turbine is not the smaller-better index classification, the degradation degree under the current classification is calculated by the following formula, and the calculation formula is as follows:

[0079]

[0080] wherein [k a ,k b ] is the best operation range of the evaluation index.

[0081] The current index classification data and the confidence coefficient corresponding to the index data are brought into the above formula, and the degradation degree under the current classification is calculated within the best operation range of the evaluation index according to the evaluation index upper limit and the evaluation index lower limit in the above formula, so as to obtain the degradation degree when the index classification result is the larger-better index classification.

[0082] The wind turbine health state evaluation method provided in the embodiment of the application comprises the following steps.

[0083] An obvious degradation degree threshold is set.

[0084] After the degradation degrees of all classifications are calculated, m offset level threshold values of offset coefficients are set according to the total number n of evaluation indexes.

[0085] An obvious degradation number is obtained based on the classification degradation calculation result and the obvious degradation degree threshold.

[0086] The offset level threshold value is matched by the obvious degradation number, and a matching offset coefficient corresponding to the level is obtained.

[0087] The comprehensive degradation degree is calculated and obtained according to the matching offset coefficient and the classification degradation calculation result.

[0088] The comprehensive degradation degree is obtained by formula calculation, and the calculation formula is as follows:

[0089]

[0090] Wherein, n is the total number of evaluation indexes, g k is the classification degradation degree calculation result of any one index, and ζ is the offset coefficient.

[0091] The embodiment is specific: in use, the obvious degradation degree threshold is set according to the degree of obvious degradation of the wind turbine, and the degree of obvious degradation of the wind turbine can include the obvious degradation degree of the wind turbine, such as the large amplitude of the wind turbine, the serious corrosion of the blade of the wind turbine, and the decrease of the oil film quality of the wind turbine. Further, after the classification degradation degree calculation according to the index classification result and the data cleaning result is completed, the offset level threshold of m offset coefficients is set according to the total number n of evaluation indexes in the state evaluation index set, and n is a positive integer greater than 1, m is a positive integer less than or equal to n, the offset level threshold is set according to the offset coefficient, and the larger the offset coefficient is, the higher the obvious degradation degree of the wind turbine is. Further, the data matched in the obvious degradation degree threshold is extracted, and the number of the extracted data is analyzed to obtain the number of obvious degradation, and then the number of obvious degradation is matched with the offset level threshold of the m offset coefficients, the number of obvious degradation that meets the offset level in the number of obvious degradation is extracted, the matching offset coefficient corresponding to the offset level equal to the number of obvious degradation is obtained, and finally the matching offset coefficient and the classification degradation degree calculation result are comprehensively degraded by the following formula, and the calculation formula is:

[0092]

[0093] Wherein, n is the total number of evaluation indexes, g k is the classification degradation degree calculation result of any one index, and ζ is the offset coefficient.

[0094] The current matching offset coefficient and the classification degradation degree calculation result are brought into the above formula, and the comprehensive degradation degree is calculated according to the total number of evaluation indexes, the classification degradation degree calculation result of a randomly selected index in the state evaluation index set, and the offset coefficient in the above formula, and the comprehensive degradation degree is obtained.

[0095] Referring to Figure 2 The embodiment of the application proposes a wind turbine health state evaluation method, which comprises:

[0096] The wind turbine is subjected to unit feature extraction through the data cleaning result, and a unit feature extraction result is obtained.

[0097] The key features are determined through the unit feature extraction result, and the credible constraint value of the key features is set.

[0098] The classification degradation degree is calculated through the key features and the credible constraint value.

[0099] The embodiment is specific to: in use, extracting unit characteristics of the current wind turbine through the data cleaning result, generating unit characteristics through the data cleaning result, re-examining and verifying the historical average wind speed data, the historical effective wind time data, and the historical average air density data, the purpose being to delete repeated information contained in the historical average wind speed data, the historical effective wind time data, and the historical average air density data, correct errors existing therein, thereby extracting unit characteristics of the wind speed characteristics of the wind turbine, the effective wind time characteristics of the wind turbine, and the average air density characteristics of the wind turbine, and determining key characteristics in the unit characteristic extraction result according to the influence on the wind turbine, that is, the greater the influence on the wind turbine, the more critical the characteristics, and exemplarily, the wind speed characteristics of the wind turbine and the effective wind time characteristics of the wind turbine can be set as the key characteristics, and the confidence coefficient corresponding to the key characteristics is constrained, that is, the confidence coefficient is in a confidence state within the constraint range, thereby more accurately calculating the classification degradation degree according to the key characteristics and the confidence constraint value.

[0100] The embodiment of the application proposes a wind turbine health state evaluation method, comprising:

[0101] determining whether the confidence coefficient of the index data corresponding to the key characteristics meets the confidence constraint value;

[0102] When it cannot be met, then re-perform index data collection until the index coefficients of all key characteristics meet the confidence constraint value, and complete the collection;

[0103] characteristic assignment is performed on the key characteristics;

[0104] classification degradation degree calculation is performed according to the re-collected index data and the key characteristic assignment result.

[0105] The embodiment is specific: when in use, in order to ensure the accuracy of the classification degradation degree calculation, it is necessary to judge whether the confidence coefficient of the index data corresponding to the key characteristics determined by the wind turbine characteristic result meets the confidence constraint value, if the confidence coefficient is within the confidence constraint value, the index data corresponding to the current key characteristic is considered to be in a trusted state, if the confidence coefficient of the index data corresponding to the key characteristic cannot meet the confidence constraint value, the index data corresponding to the current key characteristic is considered to be in an untrusted state, therefore, the index data of the key characteristic needs to be collected, and this iteration is performed until the confidence coefficients of the index data corresponding to the wind speed characteristic of the wind turbine and the effective wind time characteristic of the wind turbine in the key characteristic all meet the confidence constraint value, the index data re-collection operation of the current wind turbine is completed, further, the wind speed characteristic and the effective wind time characteristic in the key characteristic of the wind turbine are characterized according to the confidence coefficient, that is, the higher the confidence coefficient, the greater the characteristic value, and finally, the accuracy of the classification degradation degree calculation result is improved according to the re-collected index data and the key characteristic value.

[0106] Based on the same inventive concept as the wind turbine health state evaluation method in the foregoing embodiments, as shown in Figure 3 The application provides a wind turbine health state evaluation system, which comprises:

[0107] A data cleaning module 1 is used to interact with the historical operation information of the wind turbine, and perform data cleaning on the historical operation information to obtain a data cleaning result.

[0108] An index classification module 2 is used to read the state evaluation index of the wind turbine, construct a state evaluation index set, perform index classification on the state evaluation index set, and obtain an index classification result.

[0109] A data collection module 3 is used to interact with the wind turbine, and collect index data corresponding to the state evaluation index set.

[0110] A confidence evaluation module 4 is used to perform confidence evaluation on the index data, and set a confidence coefficient of the index data.

[0111] A classification degradation degree calculation module 5 is used to perform data integration calculation on the index data based on the confidence coefficient, and perform classification degradation degree calculation according to the index classification result and the data cleaning result, respectively, to obtain a classification degradation degree calculation result.

[0112] The comprehensive deterioration degree calculation module 6 is configured to calculate a comprehensive deterioration degree based on the classification deterioration degree calculation results, and generate a health state evaluation result of the wind turbine through the comprehensive deterioration degree calculation result.

[0113] Further, the system further comprises:

[0114] The first judgment module is configured to judge whether the index classification result corresponding to the data is a small optimal index classification when the classification deterioration degree calculation is performed.

[0115] The second judgment module is configured to perform the deterioration degree calculation under the current classification through a formula when the index classification result is the small optimal index classification, and the calculation formula is as follows:

[0116]

[0117] wherein k is an arbitrary index data, δ is a confidence coefficient of the index data k, k min is a lower limit value of the evaluation index, and k max is an upper limit value of the evaluation index.

[0118] The third judgment module is configured to perform the deterioration degree calculation through a formula when the index classification result corresponding to the data is not the small optimal index classification, and the calculation formula is as follows:

[0119]

[0120] wherein [k a ,k b ] is an optimal operation range of the evaluation index.

[0121] Further, the system further comprises:

[0122] The first threshold setting module is configured to set an obvious deterioration degree threshold.

[0123] The second threshold setting module is configured to set an offset level threshold of m offset coefficients according to a total number n of evaluation indexes when the classification deterioration degree calculation is completed.

[0124] The number acquisition module is configured to acquire an obvious deterioration number based on the classification deterioration degree calculation result and the obvious deterioration degree threshold.

[0125] The threshold matching module is configured to match the offset level threshold through the obvious deterioration number, and acquire a matching offset coefficient corresponding to the level.

[0126] The first calculation module is configured to calculate the comprehensive deterioration degree according to the matching offset coefficient and the classification deterioration degree calculation result.

[0127] The second calculation module is configured to calculate the comprehensive deterioration degree by a formula as follows:

[0128]

[0129] wherein n is the total number of evaluation indexes, g k is the classification deterioration degree calculation result of any one index, and ζ is the offset coefficient.

[0130] Further, the system further comprises:

[0131] The feature extraction module is configured to perform unit feature extraction on the wind turbine through the data cleaning result to obtain a unit feature extraction result.

[0132] The constraint value setting module is configured to determine a key feature through the unit feature extraction result and set a credible constraint value of the key feature.

[0133] The third calculation module is configured to perform classification deterioration degree calculation through the key feature and the credible constraint value.

[0134] Further, the system further comprises:

[0135] The fourth judgment module is configured to judge whether the credible constraint value of the key feature corresponding to the index data is satisfied.

[0136] The fifth judgment module is configured to, when the constraint value is not satisfied, re-perform index data collection until the index coefficients of all key features satisfy the credible constraint value, and complete the collection.

[0137] The feature assignment module is configured to perform feature assignment on the key feature.

[0138] The fourth calculation module is configured to perform classification deterioration degree calculation according to the re-collected index data and the key feature assignment result.

[0139] The above shows and describes the basic principles and main features of the present application and the advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above embodiments, and the above embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A method of wind turbine health assessment, characterized in that, The method comprises: interacting historical operation information of the wind turbine generator, and performing data cleaning on the historical operation information to obtain a data cleaning result; reading state evaluation indexes of the wind turbine generator, and constructing a state evaluation index set, performing index classification on the state evaluation index set, and obtaining an index classification result; interacting data of the wind turbine generator, and collecting index data corresponding to the state evaluation index set; performing credibility evaluation on the index data, and setting a credibility coefficient of the index data; performing data integration calculation on the index data based on the credibility coefficient, and performing classification degradation degree calculation according to the index classification result and the data cleaning result respectively to obtain a classification degradation degree calculation result; performing comprehensive degradation degree calculation based on the classification degradation degree calculation result, and generating a health state evaluation result of the wind turbine generator through the comprehensive degradation degree calculation result.

2. The wind turbine health assessment method of claim 1, wherein, The method further comprises: when performing the classification degradation degree calculation, judging whether the index classification result corresponding to the data is a small-optimization index classification; when the index classification result is the small-optimization index classification, performing degradation degree calculation under the current classification through a formula, and the calculation formula is as follows: Wherein, k is any index data, δ is the confidence coefficient of index data k, k min is the lower limit value of the evaluation index, k max is the upper limit value of the evaluation index.

3. The wind turbine health assessment method of claim 2, wherein, The method further comprises: when the index classification result corresponding to the data is not the small-optimization index classification, performing degradation degree calculation through a formula, and the calculation formula is as follows: where [k a ,k b ] is the optimal operating range of the evaluation index.

4. The wind turbine health assessment method of claim 3, wherein, The method further comprises: setting an obvious degradation degree threshold value; after completing all classification degradation degree calculations, setting offset level threshold values of m offset coefficients according to a total number n of evaluation indexes; obtaining an obvious degradation number based on the classification degradation degree calculation result and the obvious degradation degree threshold value; matching the offset level threshold values with the obvious degradation number to obtain a matching offset coefficient corresponding to a level; calculating the comprehensive degradation degree according to the matching offset coefficient and the classification degradation degree calculation result.

5. The wind turbine health assessment method of claim 4, wherein, The method further comprises: obtaining the comprehensive degradation degree through a formula, and the calculation formula is as follows: where n is the total number of evaluation indexes, g k is the classification degradation degree calculation result of any one index, and ζ is a bias coefficient.

6. The wind turbine unit health assessment method of claim 1, wherein, The method further comprises: extracting unit characteristics of the wind turbine generator through the data cleaning result to obtain a unit characteristic extraction result; determining key characteristics through the unit characteristic extraction result, and setting a credibility constraint value of the key characteristics; performing classification degradation degree calculation through the key characteristics and the credibility constraint value.

7. The wind turbine health assessment method of claim 6, wherein, The method further comprises: judging whether the credibility coefficient of the index data corresponding to the key characteristics meets the credibility constraint value; when it does not meet, re-performing index data collection until the index coefficients of all key characteristics meet the credibility constraint value, and completing the collection; performing characteristic assignment on the key characteristics; performing classification degradation degree calculation according to the re-collected index data and the key characteristic assignment result.

8. A wind turbine health assessment system, characterized in that, The system comprises: a data cleaning module, which is used for interacting historical operation information of a wind turbine generator, and performing data cleaning on the historical operation information to obtain a data cleaning result; an index classification module, which is used for reading state evaluation indexes of the wind turbine generator, and constructing a state evaluation index set, performing index classification on the state evaluation index set, and obtaining an index classification result; A data collection module is configured to interact with the wind turbine to collect index data corresponding to the set of state evaluation indexes; A trusted evaluation module is configured to perform trusted evaluation on the index data and set a trusted coefficient of the index data; A classified degradation degree calculation module is configured to perform data integration calculation on the index data based on the trusted coefficient, and perform classified degradation degree calculation according to the index classification result and the data cleaning result respectively to obtain a classified degradation degree calculation result; An integrated degradation degree calculation module is configured to perform integrated degradation degree calculation based on the classified degradation degree calculation result, and generate a health state evaluation result of the wind turbine through the integrated degradation degree calculation result.

Images