Wind driven generator stator winding fault detection method and system based on electrical characteristics

Abstract

The invention relates to the related technical field of wind power equipment fault detection, and discloses a wind driven generator stator winding fault detection method and system based on electrical characteristics. External image data of motor equipment and active data which are independently output or indirectly generated are synchronously monitored; and deviation evaluation is carried out through expected value active parameters obtained through fitting and actually monitored active data parameters, deviations between different parameter objects and expected values are established, and then empirical association matching can be carried out through a plurality of deviation values, so that the fault possibility is judged and positioned, and the accuracy of fault diagnosis is improved. And the problem that the fault source of a simple structure object such as a stator winding is difficult to monitor can be effectively solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wind power equipment fault detection, in particular to a wind generator stator winding fault detection method and system based on electrical characteristics. BACKGROUND

[0002] The generator stator winding is the heart of the mechanical equipment in wind power generation and is one of the core keys of the entire system, but unlike the peripheral electrical equipment which can be easily monitored and faulted in use (for example, circuit problems can be quickly repaired or electronic component replaced by monitoring the damaged position with auxiliary equipment), the stator winding has a simple structure, and regardless of the fault types such as insulation aging fault, turn-to-turn short circuit fault, phase-to-phase short circuit fault, single ground fault, winding open circuit fault, and thermal fatigue fault, the final fault object is always the simple winding itself. Due to the structure type and use form, the stator winding cannot be accurately and effectively monitored and managed during use.

[0003] The prior art mostly adopts a periodic maintenance method, but due to the limitations of the maintenance method, the internal state of the stator winding cannot be completely disassembled and judged, and the use of the prior art for fault judgment through external data parameters is also limited, mostly limited to judging the waveform curve of the output voltage and current or the heating condition of the equipment. However, the structure of the stator winding is relatively simple, and in many cases, when a fault occurs, many external associated parameters will change to a certain extent, so it is difficult to accurately locate and judge the fault by a single parameter. SUMMARY

[0004] The present application provides a wind generator stator winding fault detection method and system based on electrical characteristics to solve the problems in the background art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0006] A wind generator stator winding fault detection method based on electrical characteristics, comprising:

[0007] Real-time synchronization of multi-source monitoring data of the motor equipment, and mapping of the multi-source monitoring data in the twin monitoring space through data association matching, wherein the multi-source monitoring data is used to represent the active data parameters of the motor equipment, and the data association is used to represent the type and data attribution of the corresponding data;

[0008] Synchronization of external associated data of the motor equipment and mapping in the twin monitoring space, wherein the external associated data is used to represent the environmental data parameters and input data parameters of the external environment affected by the non-motor equipment;

[0009] based on a conversion characteristic model of the corresponding motor equipment, performing twin simulation on the external correlation data to obtain expected active parameters of the motor equipment under the current external correlation environment, the conversion characteristic model being used to represent a mapping relationship between the external correlation data and the active data parameters of the corresponding motor equipment fitted through historical data;

[0010] evaluating a deviation value of the active data parameters and the expected active parameters, establishing a deviation information pair for a parameter object corresponding to the deviation value, and performing correlation matching on a plurality of deviation information pairs to screen matched fault information, the deviation information pair being used to represent a data record form including an object and a corresponding deviation value.

[0011] As a further scheme of the present application, the active data parameters are used to represent characteristic data of relevant physical parameter changes and output power generated by the motor equipment in operation and use.

[0012] The multi-source monitoring data specifically include power output parameters, motor thermal sensing data, motor vibration monitoring data, and motor noise data, and the power output parameters include characteristic curves of a plurality of output power parameters.

[0013] The external correlation data specifically include environmental temperature and humidity, wind intensity information, and wind direction data.

[0014] As a further scheme of the present application, the method further comprises the steps of:

[0015] obtaining historical maintenance diagnosis records of a motor equipment group and corresponding multi-source monitoring data and external correlation data;

[0016] based on the fault information of the historical maintenance diagnosis records, performing group division on the multi-source monitoring data and the external correlation data, establishing a deviation information pair for a parameter object within the same fault information group, and performing correlation in units of single faults;

[0017] statistically analyzing the deviation information pairs within the same fault information group in units to evaluate the correlation between the fault information and the deviation information pairs, and establishing a fault correlation mapping relationship.

[0018] As a further scheme of the present application, the evaluation of the deviation value of the active data parameters and the expected active parameters, the establishment of a deviation information pair for a parameter object corresponding to the deviation value, and the correlation matching on a plurality of deviation information pairs to screen matched fault information specifically comprise:

[0019] one-to-one corresponding matching of data objects in the active data parameters and the expected active parameters;

[0020] Calculate the deviation value of the active data parameter and the expected active parameter under the corresponding data object, and correspondingly establish the deviation information pair with the data object as the identification feature;

[0021] Match and fit the plurality of deviation information pairs through the pre-set fault association mapping relationship to obtain a plurality of matched fault information, and the fault information corresponds to the matching degree of the plurality of data pairs.

[0022] As a further scheme of the application: when the deviation value is evaluated, a pre-fitting step is further included, specifically including:

[0023] A sinking floating range is established based on external associated data, the sinking floating range is used to represent the value range of the external associated data, and the value direction of the value range corresponds to the power output reduction direction of the motor equipment;

[0024] The active data parameter is fitted based on the array expected active parameter simulated by the sinking floating range, and the expected active parameter with the highest power output parameter fitting degree is screened out for deviation value evaluation;

[0025] The output offset parameter of the expected active parameter with the highest parameter fitting degree and the expected active parameter corresponding to the external associated environment is calculated, the output offset parameter is used to represent the overall performance attenuation aging degree of the motor equipment, and is used to feedback the execution of the maintenance program.

[0026] The embodiment of the application aims to provide a wind turbine stator winding fault detection system based on electrical characteristics, comprising:

[0027] The object monitoring synchronization module is used for real-time synchronization of the multi-source monitoring data of the motor equipment, and the multi-source monitoring data is mapped in the twin monitoring space through data association matching, the multi-source monitoring data is used to represent the active data parameter of the motor equipment, and the data association is used to represent the type and data ownership of the corresponding data;

[0028] The environment monitoring synchronization module is used for synchronously associating the external associated data of the motor equipment and mapping through the twin monitoring space, the external associated data is used to represent the environmental data parameter and input data parameter of the external environment affected by the non-motor equipment;

[0029] The twin expected simulation module is used for twin simulation of the external associated data based on the conversion characteristic model of the corresponding motor equipment, to obtain the expected active parameter of the motor equipment under the current external associated environment, and the conversion characteristic model is used to represent the mapping relationship between the external associated data and the active data parameter of the corresponding motor equipment fitted through historical data;

[0030] The fault factor matching module is used for evaluating deviation values of the active data parameters from the expected active parameters, establishing deviation information pairs corresponding to parameter objects with the deviation values, and performing correlation matching on the plurality of deviation information pairs to screen matched fault information, wherein the deviation information pairs are used for representing data record forms including objects and corresponding deviation values.

[0031] As a further scheme of the present application, the active data parameters are used for representing characteristic data of relevant physical parameter changes and output power generated by the motor equipment in operation and use.

[0032] The plurality of source monitoring data specifically include power output parameters, motor thermal sensing data, motor vibration monitoring data and motor noise data, and the power output parameters include characteristic curves of a plurality of output power parameters.

[0033] The external correlation data specifically include environmental temperature and humidity, wind intensity information and wind direction data.

[0034] As a further scheme of the present application, the present application further comprises:

[0035] The maintenance data synchronization unit is used for acquiring historical maintenance diagnosis records of the motor equipment group and corresponding plurality of source monitoring data and external correlation data.

[0036] The fault data grouping unit is used for grouping the plurality of source monitoring data and the external correlation data based on fault information of the historical maintenance diagnosis records, establishing deviation information pairs of parameter objects in the same fault information group, and correlating the same in units of single faults.

[0037] The fault correlation evaluation unit is used for statistically evaluating the correlation of the fault information and the deviation information pairs in the same fault information group in units of the same, and establishing a fault correlation mapping relationship.

[0038] As a further scheme of the present application, the fault factor matching module comprises:

[0039] The parameter object matching unit is used for one-to-one corresponding matching of different data objects in the active data parameters and the expected active parameters.

[0040] The data pair establishing unit is used for calculating deviation values of the active data parameters from the expected active parameters under corresponding data objects, and establishing deviation information pairs with the data objects as identification features.

[0041] The fault feature matching unit is used for matching and fitting a plurality of deviation information pairs through a pre-set fault correlation mapping relationship to obtain a plurality of matched fault information, and the fault information corresponds to matching degrees of a plurality of data pairs.

[0042] As a further further scheme of the present application, further comprising:

[0043] The floating defining unit is used to establish a sinking floating range based on the external correlation formula data, the sinking floating range is used to represent the value range of the external correlation data, and the value direction of the value range corresponds to the power output reduction direction of the motor equipment;

[0044] The floating fitting unit is used to fit the active data parameters based on the array expected active parameters simulated based on the sinking floating range, and the expected active parameter with the highest power output parameter fitting degree is selected for bias value evaluation;

[0045] The systematic evaluation unit is used to calculate the output offset parameter of the expected active parameter with the highest parameter fitting degree and the expected active parameter corresponding to the external correlation environment, the output offset parameter is used to represent the overall performance attenuation aging degree of the motor equipment, and is used to feedback the execution maintenance program.

[0046] Compared with the prior art, the present application has the beneficial effects that: by synchronously monitoring the external image data and the active data generated by the motor equipment, and by evaluating the deviation between the expected value active parameter and the actual monitored active data parameter, the deviation between different parameter objects and the expected value is established, and then the experience correlation matching is realized through multiple deviation values, the fault possibility is judged and positioned, and the problem of difficulty in monitoring the fault source of the stator winding and other simple structure objects is effectively solved. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 The flowchart of the wind turbine stator winding fault detection method based on electrical characteristics.

[0048] Figure 2 The flowchart of the pre-fitting step in the wind turbine stator winding fault detection method based on electrical characteristics.

[0049] Figure 3 The composition block diagram of the wind turbine stator winding fault detection system based on electrical characteristics. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application is further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application, and are not used to limit the present application.

[0051] The specific implementation mode of the present application is described in detail below in combination with specific examples.

[0052] As Figure 1The wind generator stator winding fault detection method based on electrical characteristics provided by the embodiment of the application comprises the following steps:

[0053] S10, real-time synchronization is performed on multi-source monitoring data of the motor equipment, and the multi-source monitoring data is mapped in a twin monitoring space through data correlation matching, the multi-source monitoring data is used to represent active data parameters of the motor equipment, and the data correlation is used to represent the type and data attribution of corresponding data;

[0054] S20, external correlation data of the motor equipment is synchronously associated and mapped through the twin monitoring space, the external correlation data is used to represent environmental data parameters and input data parameters of an external environment that are influenced and changed by non-motor equipment;

[0055] S30, twin simulation is performed on the external correlation data based on a conversion characteristic model of the corresponding motor equipment, expected active parameters of the motor equipment under a current external correlation environment are obtained, and the conversion characteristic model is used to represent a mapping relationship between the external correlation data and the active data parameters of the corresponding motor equipment fitted through historical data;

[0056] S40, a deviation value of the active data parameters from the expected active parameters is evaluated, a deviation information pair including an object and a corresponding deviation value is established, a plurality of deviation information pairs are associated and matched to screen matched fault information, and the deviation information pair is used to represent a data record form.

[0057] In this embodiment, by synchronously monitoring the external image data of the motor equipment and the active data autonomously output or indirectly generated, and by fitting the deviation evaluation of the expected value active parameter and the actual monitoring active data parameter, the deviation of different parameter objects and the expected value is established, and then the empirical correlation matching can be performed through multiple deviation values, the fault possibility is judged and positioned, and the difficulty problem of monitoring the fault source of the stator winding of such a simple structure object can be effectively solved. In wind power generation, the generator stator winding is the heart part of the mechanical equipment and is one of the core keys of the whole system, but unlike the peripheral electrical equipment which can be easily monitored and faulted in use (for example, the circuit problem can be quickly repaired or the electronic component is replaced by monitoring the damaged position by the auxiliary equipment), the stator winding has a simple structure, and regardless of the insulation aging fault, turn-to-turn short circuit fault, phase-to-phase short circuit fault, single ground fault, winding open circuit fault, thermal fatigue fault and other fault types, the final fault object is a simple winding itself. Because of the structure type and use form, the internal state of the stator winding cannot be accurately and effectively monitored during use, and because the probability of failure of the structure in actual use is low, in most cases, the existing technology more adopts the periodic maintenance method, but due to the limitation of the maintenance method, the internal state of the stator winding cannot be completely disassembled and judged, and the fault judgment method by the external data parameters is also limited in the existing technology, and is mostly limited to the judgment of the waveform curve of the output voltage and current or the heating condition of the equipment. Because the structure of the stator winding is relatively simple, in many cases, when a fault occurs, many external associated parameters will change to a certain extent, so it is difficult to accurately locate and judge the fault by a single parameter. In this application, multiple data monitoring schemes are combined to monitor and collect multiple associated data objects, and are grouped according to the input and output influence relationship of the motor equipment, so that the expected value of the corresponding parameter can be obtained by simulation, and the deviation value calculation is performed with the actual monitoring corresponding value. In this way, multiple deviation information pairs of "object-deviation value" can be obtained, and the error influence of different data types is different under different faults. Through comprehensive evaluation and screening of the mapping correlation table of multiple deviation information pairs and fault maintenance experience, the final fault positioning can be obtained.

[0058] As another preferred embodiment of the application, the active data parameter is used to represent the related physical parameter change and output power characteristic data generated by the motor equipment in operation.

[0059] The multi-source monitoring data specifically includes power output parameters, motor thermal sensing data, motor vibration monitoring data, and motor noise data, and the power output parameters include characteristic curves of multiple output power parameters.

[0060] The external correlation data specifically includes environmental temperature and humidity, wind intensity information, and wind direction data.

[0061] In this embodiment, the two types of data are described respectively, which can be simply summarized as follows: the external correlation data is an external preposition parameter that affects the operation of the motor equipment, and when these parameters change, the operation state of the motor equipment is affected; and the multi-source monitoring data is a related parameter that changes due to the operation of the motor equipment; the state evaluation of the motor equipment can be performed from multiple angles through the monitoring of the two types of data.

[0062] As another preferred embodiment of the present application, the method further comprises the steps of:

[0063] obtaining historical maintenance diagnosis records of the motor equipment group and corresponding multi-source monitoring data and external correlation data;

[0064] grouping the multi-source monitoring data and the external correlation data based on fault information of the historical maintenance diagnosis records, establishing deviation information pairs of parameter objects in the same fault information group, and correlating the same as a unit of single fault;

[0065] statistically analyzing the deviation information pairs in the same fault information group as a unit to evaluate the correlation between the fault information and the deviation information pairs, and establishing a fault correlation mapping relationship.

[0066] In this embodiment, the purpose of this step is to establish a database for matching multiple deviation information pairs, i.e., a fault correlation mapping relationship. Here, the data referred to are mainly historical maintenance data and correlated monitoring data. In addition, when the amount of data is insufficient, a certain fault condition can be pre-set by forward simulation to generate various state parameters under different faults, so as to establish a reference correlation relationship. However, this generated database needs to be continuously optimized and updated through actual data feedback in subsequent maintenance.

[0067] As another preferred embodiment of the present application, the evaluation of the deviation value of the active data parameter from the expected active parameter, the establishment of the deviation information pair of the parameter object corresponding to the deviation value, the correlation and matching of multiple deviation information pairs, and the screening of matched fault information specifically include:

[0068] According to the one-to-one corresponding matching of the data objects in the active data parameter and the expected active parameter;

[0069] corresponding data object, and a deviation information pair corresponding to the deviation value is established;

[0070] The plurality of deviation information pairs are matched and fitted through a pre-set fault correlation mapping relationship to obtain a plurality of matched fault information, and the fault information corresponds to a matching degree of the plurality of data pairs.

[0071] In this embodiment, there are two parts. One is the calculation of the deviation value. In this process, the corresponding data parameter object needs to be matched, and after the deviation value is calculated, a one-to-one corresponding deviation information pair is established. In the process of calculating and establishing the deviation information pair, the process of data selection will also be involved. Because there is a certain error in the parameters of the associated data generated in each generation and recording, a certain error tolerance needs to be considered in the calculation. This process needs to be evaluated through the floating changes of various parameters of historical data. The other is to retrieve and match the plurality of deviation information pairs to obtain the fault condition with the highest matching degree of the plurality of deviation information pairs.

[0072] As shown in Figure 2 as another preferred embodiment of the present application, the evaluation of the deviation value also includes a pre-fitting step, which specifically includes:

[0073] S51, a sinking floating range is established based on external associated data, the sinking floating range is used to represent the value range of the external associated data, and the value direction of the value range corresponds to the power output reduction direction of the motor equipment;

[0074] S52, the active data parameter is fitted based on the array of the simulated sinking floating range, and the expected active parameter with the highest power output parameter fitting degree is selected for deviation value evaluation;

[0075] S53, the output offset parameter of the expected active parameter with the highest parameter fitting degree and the expected active parameter corresponding to the external associated environment is calculated, the output offset parameter is used to represent the overall performance attenuation and aging degree of the motor equipment, and is used to feedback the execution of the maintenance program.

[0076] In this embodiment, in addition to the fault, there is also a performance reduction caused by aging and weakening in long-term use without damage. This type of deviation has overall and regular characteristics. Therefore, when establishing the deviation data pair, overall deviation fitting needs to be performed to exclude this part of the deviation. This can be regarded as a noise elimination step for the deviation data pair. The eliminated offset can be used to evaluate the maintenance needs of the overall equipment and determine the maintenance task.

[0077] As shown in Figure 3As shown, the application also provides a wind generator stator winding fault detection system based on electrical characteristics, which comprises:

[0078] The object monitoring synchronization module 100 is used for real-time synchronization of multi-source monitoring data of the motor equipment, and mapping of the multi-source monitoring data in a twin monitoring space through data correlation matching, the multi-source monitoring data being used for characterizing active data parameters of the motor equipment, and the data correlation being used for characterizing types and data attribution of corresponding data.

[0079] The environment monitoring synchronization module 200 is used for synchronization and correlation of external correlation data of the motor equipment and mapping through the twin monitoring space, the external correlation data being used for characterizing environmental data parameters and input data parameters of an external environment which are influenced and changed by non-motor equipment.

[0080] The twin expectation simulation module 300 is used for twin simulation of the external correlation data based on a conversion characteristic model of the corresponding motor equipment, to obtain expected active parameters of the motor equipment under a current external correlation environment, the conversion characteristic model being used for characterizing a mapping relationship between the external correlation data and the active data parameters of the corresponding motor equipment fitted through historical data.

[0081] The fault factor matching module 400 is used for evaluating a deviation value of the active data parameters from the expected active parameters, establishing a deviation information pair for a parameter object corresponding to the deviation value, and correlating and matching a plurality of deviation information pairs to screen matched fault information, the deviation information pair being used for characterizing a data record form including an object and a corresponding deviation value.

[0082] As another preferred embodiment of the application, the active data parameters are used for characterizing relevant physical parameter changes and output power characteristic data generated by the motor equipment in operation and use.

[0083] The multi-source monitoring data specifically include power output parameters, motor thermal sensing data, motor vibration monitoring data and motor noise data, and the power output parameters include characteristic curves of a plurality of output power parameters.

[0084] The external correlation data specifically include environmental temperature and humidity, wind intensity information and wind direction data.

[0085] As another preferred embodiment of the application, the application further comprises:

[0086] The maintenance data synchronization unit is used for obtaining historical maintenance diagnosis records of the motor equipment group and corresponding multi-source monitoring data and external correlation data.

[0087] A fault data grouping unit is configured to group multi-source monitoring data and external associated data based on fault information of historical maintenance diagnosis records, and to establish a deviation information pair of a parameter object in the same fault information group, and to associate the same fault with a single fault as a unit;

[0088] A fault correlation evaluation unit is configured to statistically evaluate the correlation between the fault information and the deviation information pair in the same fault information group, and to establish a fault correlation mapping relationship.

[0089] As another preferred embodiment of the present application, the fault factor matching module comprises:

[0090] A parameter object matching unit is configured to one-to-one match different data objects in the active data parameters with the expected active parameters.

[0091] A data pair establishing unit is configured to calculate the deviation value of the active data parameters and the expected active parameters corresponding to the data objects, and to establish a deviation information pair corresponding to the data objects as an identification feature.

[0092] A fault feature matching unit is configured to match and fit a plurality of deviation information pairs through a pre-set fault correlation mapping relationship, to obtain a plurality of matched fault information, and the fault information corresponding to the matching degree of a plurality of data pairs.

[0093] As another preferred embodiment of the present application, it further comprises:

[0094] A floating limiting unit is configured to establish a sinking floating range based on the external associated data, the sinking floating range is used to represent the value range of the external associated data, and the value direction of the value range corresponds to the power output reduction direction of the motor equipment.

[0095] A floating fitting unit is configured to fit the active data parameters based on the array of expected active parameters simulated by the sinking floating range, and to select the expected active parameter with the highest power output parameter fitting degree for deviation value evaluation.

[0096] A systematic evaluation unit is configured to calculate the output offset parameter of the expected active parameter corresponding to the external associated environment of the expected active parameter with the highest parameter fitting degree, the output offset parameter is used to represent the overall performance attenuation aging degree of the motor equipment, and is used to feedback the execution of the maintenance program.

[0097] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer readable storage medium, and when the program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, storage, databases, or other media in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0098] Other embodiments of the present disclosure will be apparent to those skilled in the art with the accomplishment of the present disclosure as reflected in the specification and embodiments. The present application is intended to cover any variations, uses, or adaptive changes of the present disclosure following the general principles of the present disclosure and including common knowledge or conventional technical means in the art not disclosed by the present disclosure. The specification and embodiments are only regarded as exemplary, and the true scope and spirit of the present disclosure are indicated by the claims.

[0099] It should be understood that the present disclosure is not limited to the precise structures described above and shown in the drawings and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A method for detecting faults in the stator windings of a wind generator based on electrical characteristics, characterized in that, Comprise: Real-time synchronization of multi-source monitoring data of motor equipment, and mapping of the multi-source monitoring data in a twin monitoring space through data correlation matching, the multi-source monitoring data being used to represent active data parameters of the motor equipment, and the data correlation being used to represent types and data ownership of corresponding data; Synchronization of external correlation data of the motor equipment and mapping through the twin monitoring space, the external correlation data being used to represent environmental data parameters and input data parameters of an external environment that are affected and changed by non-motor equipment; Based on a conversion characteristic model of a corresponding motor equipment, twin simulation of the external correlation data is performed to obtain expected active parameters of the motor equipment under a current external correlation environment, the conversion characteristic model being used to represent a mapping relationship between the external correlation data and the active data parameters of the corresponding motor equipment through historical data fitting; Evaluation of a deviation value of the active data parameters from the expected active parameters, establishment of a deviation information pair of a parameter object corresponding to the deviation value, and correlation matching of a plurality of deviation information pairs to screen matched fault information, the deviation information pair being used to represent a data record form including an object and a corresponding deviation value.

2. The electrical signature based wind generator stator winding fault detection method of claim 1, wherein, The active data parameters are used to represent characteristic data of relevant physical parameter changes and output power generated by the motor equipment in operation and use; The multi-source monitoring data specifically include power output parameters, motor thermal sensing data, motor vibration monitoring data, and motor noise data, and the power output parameters include characteristic curves of a plurality of output power parameters; The external correlation data specifically include environmental temperature and humidity, wind intensity information, and wind direction data.

3. The electrical signature based wind generator stator winding fault detection method of claim 2, wherein, Further comprising steps: Obtaining historical maintenance diagnosis records of a motor equipment group and corresponding multi-source monitoring data and external correlation data; Based on fault information of the historical maintenance diagnosis records, grouping of the multi-source monitoring data and the external correlation data, establishment of deviation information pairs of parameter objects in the same fault information group, and correlation in units of single faults; Statistical analysis of the deviation information pairs in the same fault information group in units to evaluate the correlation of the fault information and the deviation information pairs, and establishment of a fault correlation mapping relationship.

4. The electrical signature based wind generator stator winding fault detection method of claim 3, wherein, The evaluation of the deviation value of the active data parameters from the expected active parameters, the establishment of the deviation information pair of the parameter object corresponding to the deviation value, and the correlation matching of the plurality of deviation information pairs to screen the matched fault information specifically include: One-to-one matching of data objects in the active data parameters with the expected active parameters; Calculation of the deviation value of the active data parameters from the expected active parameters under the corresponding data objects, and corresponding establishment of the deviation information pair with the data object as an identification feature; Matching fitting of the plurality of deviation information pairs through a pre-set fault correlation mapping relationship to obtain a plurality of matched fault information, and the fault information corresponding to the matching degree of a plurality of data pairs.

5. The electrical signature based wind generator stator winding fault detection method of claim 4, wherein, In the evaluation of the deviation value, a pre-fitting step is further included, specifically comprising: Establishment of a sinking floating range based on the external correlation data, the sinking floating range being used to represent a value range of the external correlation data, and a value direction of the value range corresponding to a power output reduction direction of the motor equipment; The array expectation active parameter based on the sinking floating range simulation is used to fit the active data parameter, and the expectation active parameter with the highest fitting degree of the power output parameter is screened out for bias value evaluation; The output offset parameter of the expectation active parameter corresponding to the external associated environment with the highest parameter fitting degree is calculated, and the output offset parameter is used to represent the overall performance attenuation aging degree of the motor equipment, and is used for feedback to execute the maintenance program.

6. A wind generator stator winding fault detection system based on electrical characteristics, characterized in that, Comprise: The object monitoring synchronization module is used for real-time synchronization of multi-source monitoring data of the motor equipment, and mapping of the multi-source monitoring data in the twin monitoring space through data association matching, the multi-source monitoring data is used to represent the active data parameter of the motor equipment, and the data association is used to represent the type and data attribution of the corresponding data; The environment monitoring synchronization module is used for synchronizing and associating the external associated data of the motor equipment and mapping through the twin monitoring space, and the external associated data is used to represent the environmental data parameter and input data parameter of the external environment which is affected and changed by the non-motor equipment; The twin expectation simulation module is used for twin simulation of the external associated data based on the conversion characteristic model of the corresponding motor equipment, and the expectation active parameter of the motor equipment under the current external associated environment is obtained, and the conversion characteristic model is used to represent the mapping relationship between the external associated data and the active data parameter of the corresponding motor equipment fitted through historical data; The fault factor matching module is used for evaluating the bias value of the active data parameter and the expectation active parameter, establishing a bias information pair of the parameter object corresponding to the bias value, and associating and matching a plurality of bias information pairs to screen the matched fault information, and the bias information pair is used to represent the data record form including the object and the corresponding bias value.

7. The electrical signature based wind generator stator winding fault detection system of claim 6, wherein, The active data parameter is used to represent the related physical parameter change and output power characteristic data generated by the motor equipment in the running use: The multi-source monitoring data specifically includes power output parameters, motor thermal sensing data, motor vibration monitoring data and motor noise data, and the power output parameters include characteristic curves of a plurality of output power parameters; The external associated data specifically includes environmental temperature and humidity, wind intensity information and wind direction data.

8. The electrical signature based wind generator stator winding fault detection system of claim 7, wherein, Further comprise: The maintenance data synchronization unit is used for obtaining the historical maintenance diagnosis record of the motor equipment group and the corresponding multi-source monitoring data and external associated data; The fault data grouping unit is used for grouping and dividing the multi-source monitoring data and the external associated data based on the fault information of the historical maintenance diagnosis record, establishing the bias information pair of the parameter object in the same fault information group, and associating with a single fault as a unit; The fault association evaluation unit is used for statistically evaluating the bias information pair in the same fault information group in units to evaluate the association of the fault information and the bias information pair, and establishing a fault association mapping relationship.

9. The electrical signature based wind generator stator winding fault detection system of claim 8, wherein, The fault factor matching module comprises: The parameter object matching unit is used for one-to-one corresponding matching of different data objects in the active data parameter and the expectation active parameter; The data pair establishing unit is configured to calculate deviation values of active data parameters and expected active parameters corresponding to data objects, and establish deviation information pairs corresponding to the data objects as identification features; The fault feature matching unit is configured to match and fit the plurality of deviation information pairs through a pre-set fault correlation mapping relationship to obtain a plurality of matched fault information, and the fault information corresponds to matching degrees of the plurality of data pairs.

10. The electrical signature based wind generator stator winding fault detection system of claim 9, wherein, Further comprising: The floating limiting unit is configured to establish a sinking floating range based on external correlation data, the sinking floating range is used to represent a value range of the external correlation data, and a value direction of the value range corresponds to a power output reduction direction of the motor equipment; The floating fitting unit is configured to fit the active data parameters based on an array of expected active parameters simulated based on the sinking floating range, and select the expected active parameter with the highest power output parameter fitting degree for deviation value evaluation; The systematic evaluation unit is configured to calculate an output offset parameter of the expected active parameter with the highest parameter fitting degree and an expected active parameter corresponding to an external correlation environment, the output offset parameter is used to represent a whole performance attenuation aging degree of the motor equipment, and is used to feed back an execution maintenance program.

Images