A Fault Diagnosis Method and System for the Excitation System of a Generator Set

Abstract

The present invention discloses a fault diagnosis method and system for an excitation system of a generator set, belonging to the technical field of fault diagnosis of the excitation system of the generator set. The fault diagnosis system includes an information acquisition module, a fault library establishment module, a fault prediction module, a fault diagnosis module, and a fault library update module. The real-time acquisition of data is realized through the information acquisition module, a multi-dimensional space fault library is established through the fault library establishment module, the fault prediction of the excitation system is carried out according to the existing node data through the fault library prediction module, the fault diagnosis of the excitation system is carried out through the fault diagnosis module, and the fault library is updated through the fault library update module. The fault prediction module uses the method of angle comparison to predict the fault of the excitation system, and the fault diagnosis module uses the methods of distance comparison, angle comparison, and coordinate comparison to realize the timely diagnosis of the fault, so that the results of fault diagnosis and prediction can be more accurate and timely.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis of generator excitation systems, and specifically to a fault diagnosis method and system for generator excitation systems. Background Art

[0002] The generator excitation system mainly includes an excitation power unit and an excitation regulator, and its main functions are to maintain the generator terminal voltage, control the reactive power distribution among parallel operating generators, and improve the transient stability of the power system, etc., which is equivalent to the heart of the generator. If a fault occurs in the generator excitation system, it will cause economic losses and there are also some potential safety hazards.

[0003] Currently, most of the fault diagnosis of the excitation system is carried out by professional technicians. Due to the multiple causes of faults, on-site personnel usually cannot quickly diagnose faults, and at the same time, they cannot predict faults based on individual abnormal data.

[0004] Therefore, a fault diagnosis method and system for generator excitation systems are needed to solve the above problems and achieve accurate and rapid diagnosis of system faults and system fault prediction. Summary of the Invention

[0005] The purpose of the present invention is to provide a fault diagnosis method and system for generator excitation systems to solve the problems raised in the above background art.

[0006] To solve the above technical problems, the present invention provides the following technical solution: A fault diagnosis method for a generator excitation system, the fault diagnosis method includes the following steps:

[0007] S1. Use the information collection module to collect and store the real-time data generated by the generator excitation system;

[0008] S2. Use the data classification unit to screen out the node abnormal correlation data of the excitation system;

[0009] S3. Place the abnormal correlation data in a multi-dimensional space to establish a fault database;

[0010] S4. Use the fault prediction module to match the real-time collected data with the data in the fault database to achieve fault prediction and use the fault diagnosis module to achieve fault diagnosis;

[0011] S5. Use the fault database update module to achieve real-time update of the fault database.

[0012] According to the above technical solution, in S2, the screening of abnormal associated data refers to determining the characteristic data of the excitation system fault and the node abnormal associated data generated during the fault occurrence process; putting the screened characteristic data of the excitation system fault into the set A = {a 1 , a 2 , a 3 , …, a n}, and putting the set of node abnormal associated data generated during the fault occurrence into the set a n = {b n1 , b n2 , b n3 , …, b ni}; by classifying the stored historical data, the calculation time in the subsequent process is reduced, and while improving the efficiency, the accuracy of fault diagnosis is also improved;

[0013] where a n represents the characteristic data of the excitation system fault, b ni represents the set of node abnormal associated data, n represents that there are n characteristic data of the excitation system fault, and i represents that one characteristic data of the excitation system fault corresponds to i sets of node abnormal associated data.

[0014] According to the above technical solution, in S3, the specific steps for establishing a fault library include:

[0015] Z1. Establish a multi-dimensional space with m pairwise intersecting and perpendicular lines, where m refers to the m-dimensional space and also refers to m nodes;

[0016] Z2. Convert each node data into a coordinate value, where the maximum and minimum values of the normal range of each node are P m and Q m , and the obtained node data is K m ;

[0017] When Q m < K m < P m , the coordinate value α m of this node is 0;

[0018] When Q m > K m , the coordinate value α m of this node is

[0019] When K m > P m , the coordinate value α m of this node is

[0020] Enable the abnormal change sequence of each node to be reflected in the spatial coordinates, making the time sequence concrete;

[0021] where t τ represents the order in which the node generates abnormal data;

[0022] Z4. Convert the node association data set into coordinates γ ni (α n1 ,α n2 ,…,α nm ,t τ ), where α m is the node coordinate value, and t τ represents the number of abnormal data generated during the failure process of the excitation system; the converted coordinates contain t τ , making the fault diagnosis more accurate;

[0023] Z5. Convert all node association data sets into coordinates, place them in an m-dimensional space, and calculate the distance D between all coordinates and the origin ni

[0024]

[0025] and place them in a set Thus, a fault library is established.

[0026] According to the above technical solution, in S4, the fault diagnosis refers to judging the cause of the excitation system fault based on the set of associated data of node anomalies. The specific steps of the fault diagnosis include:

[0027] L1. Convert the set of associated data of node anomalies into coordinates. The coordinate value conversion formula for each node data is:

[0028] When Q m <K m <P m , the coordinate value β m of this node is 0;

[0029] When Q m >K m , the coordinate value β m of this node is

[0030] When K m >P m , the coordinate value β m of this node is

[0031] Convert the node association data set into coordinates ρ(β 1 ,β 2 ,…,β m ,tκ );

[0032] where α m is the node coordinate value, and t κ represents the number of abnormal data generated during the failure of the excitation system. The maximum and minimum values of the normal range data of each node are P m and Q m respectively, and K m represents the obtained node data;

[0033] L2. Place the coordinates in an m-dimensional space coordinate system and calculate the distance d from this point to the origin

[0034]

[0035] L3. Compare the distance d with the elements in the set . When |d - D ni | = ε, then it is stipulated that D i = U l Calculate the angles between the coordinates ρ, γ i and the origin. The calculation formula is

[0036]

[0037] where l represents the number of D ni when |d - D ni | = ε;

[0038] L4. When and it meets and there is only one cosθ ι , then it proves a successful diagnosis, and the diagnosis result is the fault corresponding to the set of node associated data corresponding to the coordinate U l ;

[0039] In other cases, place the coordinates that meet into the set Ψ = {φ 1 , φ 2 , …, φ σ}, and interleave each coordinate value in the coordinate ρ with each coordinate value of φ σ . If each coordinate difference is less than , then it proves a successful diagnosis, and the diagnosis result is the fault corresponding to the set of node associated data corresponding to the coordinate φ σ , where σ represents the number of the included angles between two vectors that meet ;

[0040] By comparing distances, angles, and coordinates, the fault diagnosis result of the excitation system is made more accurate. Through step-by-step comparison, the comparison range is gradually narrowed, making the fault diagnosis efficiency higher, and reducing the amount of calculation to a certain extent;

[0041] In S4, the fault prediction module refers to establishing a prediction model to predict faults based on existing abnormal data. The specific steps of its fault prediction include:

[0042] I1. First, convert the existing node association data set into corresponding coordinates

[0043] I2. Calculate the coordinate Coordinate γ i The angle with the origin

[0044] I3. When it indicates that the fault corresponding to the node association data set corresponding to the coordinate meets the requirements The fault corresponding to the node association data set corresponding to the coordinate is the predicted fault, where represents the maximum deviation.

[0045] According to the above technical solution, in S5, the fault library update module refers to comparing the actual diagnosis result with the system's diagnosis result. If there is a deviation, the actual diagnosis result is transmitted to the fault library and matched with the original data in the fault library, and the data in the fault library is corrected in a timely manner to achieve real-time update of the fault library, making the data in the fault library more accurate and the error of the matching comparison smaller.

[0046] A fault diagnosis system for the excitation system of a generator set, which includes an information collection module, a fault library establishment module, a fault prediction module, a fault diagnosis module, and a fault library update module;

[0047] The information collection module is used to collect data of each node in real time during the operation of the excitation system;

[0048] Enabling real-time observation of node information to promptly detect node abnormalities;

[0049] The fault library establishment module is used to establish a multi-dimensional space fault library;

[0050] The fault library prediction module is used to predict faults in the excitation system based on existing node data;

[0051] Through fault prediction, corresponding measures can be taken in a timely manner to avoid, reduce or even avoid losses;

[0052] The fault diagnosis module is used to diagnose faults in the excitation system;

[0053] Enable timely and accurate diagnosis of faults and reduce losses;

[0054] The fault library update module is used to update the fault library;

[0055] Enable the data in the fault library to be updated and corrected in a timely manner, increasing the accuracy of fault diagnosis and fault prediction;

[0056] The output end of the information acquisition module is connected to the input ends of the fault library establishment module, the fault prediction module, and the fault diagnosis module, and the output end of the fault diagnosis module is connected to the input end of the fault library update module.

[0057] According to the above technical solution, the information acquisition module includes a data acquisition unit, a data storage unit, and a data classification unit;

[0058] The data acquisition unit is used to collect data of each node in real time;

[0059] The data storage unit adopts distributed storage of the collected data;

[0060] Enable efficient management of read cache and write cache and automatic hierarchical storage;

[0061] The data classification unit is used to screen out the fault feature and node anomaly correlation data set;

[0062] Reduce the computational amount in the subsequent process and reduce the error rate of fault diagnosis and prediction;

[0063] The output end of the data acquisition unit is connected to the input ends of the data storage unit and the data classification unit.

[0064] According to the above technical solution, the fault library establishment module includes a multi-dimensional space unit and a fault library establishment unit;

[0065] The multi-dimensional space unit is used to realize the establishment of a multi-dimensional space;

[0066] The fault library establishment unit is used to establish a multi-dimensional space database;

[0067] The output end of the multi-dimensional space unit is connected to the input end of the fault establishment unit.

[0068] According to the above technical solution, the fault prediction end element includes a fault prediction unit and a prediction notification unit;

[0069] The fault prediction unit is used to establish an angle prediction model for fault prediction;

[0070] Make fault prediction more accurate;

[0071] The prediction notification unit is used to notify the result of fault prediction;

[0072] It enables timely measures to be taken to reduce losses;

[0073] The output end of the fault prediction unit is connected to the input end of the prediction notification unit.

[0074] According to the above technical solution, the fault diagnosis module includes a fault diagnosis unit and a diagnosis reporting unit;

[0075] The fault diagnosis unit diagnoses faults in the excitation system through distance comparison, angle comparison, and coordinate comparison;

[0076] Through the three comparisons, the fault diagnosis result is made more accurate;

[0077] The diagnosis reporting unit is used to report the fault diagnosis result of the excitation system;

[0078] The fault library update module includes a data comparison element and a fault library update unit;

[0079] The data comparison unit extracts the latest data by comparing the data in the original fault library with the latest fault data information;

[0080] The fault library update unit is used to update the fault library;

[0081] It enables the data in the fault library to be continuously updated and corrected, continuously improving the accuracy of fault diagnosis and prediction;

[0082] The output end of the fault diagnosis unit is connected to the input ends of the diagnosis reporting unit and the data comparison unit, and the output end of the data comparison unit is connected to the input end of the fault library update unit.

[0083] Through the above technical solution, the fault of the excitation system can be accurately and quickly diagnosed, and the fault prediction of the excitation system can be realized, enabling timely solution of system faults, reducing certain economic losses, and avoiding the occurrence of safety accidents to a certain extent.

[0084] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0085] 1. By setting the fault diagnosis unit, the present invention enables rapid and accurate fault diagnosis of the excitation system. By setting the diagnosis reporting unit, it enables the staff to obtain the diagnosis result in a timely manner and make reasonable remedial measures, which can reduce economic losses and avoid the occurrence of safety accidents to a certain extent.

[0086] 2. By setting up a fault prediction unit, the present invention predicts faults in the excitation system based on existing data, enabling staff to take preventive measures in advance and perform maintenance on the equipment in advance, reducing equipment wear and tear to a certain extent and extending the equipment life.

[0087] 3. By setting up a fault library update unit, the present invention enables the fault library to be updated in a timely manner. Through continuous updates, the data in the fault library becomes more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] The drawings are used to provide a further understanding of the present invention and form a part of the specification. They are used together with the embodiments of the present invention to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0089] Figure 1 is a schematic diagram of the module composition of a fault diagnosis system for an excitation system of a generator set according to the present invention;

[0090] Figure 2 is a schematic diagram of the structural connection of a fault diagnosis system for an excitation system of a generator set according to the present invention;

[0091] Figure 3 is a schematic diagram of the step flow of a fault diagnosis method for an excitation system of a generator set according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0092] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0093] Please refer to Figures 1-3 , the present invention provides a technical solution: a fault diagnosis method for an excitation system of a generator set, and the fault diagnosis method includes the following steps:

[0094] S1. Use the information collection module to collect and store the real-time data generated by the excitation system of the generator set;

[0095] S2. Use the data classification unit to screen out the node abnormal correlation data of the excitation system;

[0096] S3. Place the abnormal correlation data in a multi-dimensional space to establish a fault library;

[0097] S4. Use the fault prediction module to match the real-time collected data with the data in the fault library to achieve fault prediction and use the fault diagnosis module to achieve fault diagnosis;

[0098] S5. Implement real-time update of the fault database using the fault database update module.

[0099] In S2, the abnormal correlation data screening refers to determining the characteristic data of the excitation system fault and the node abnormal correlation data generated during the fault occurrence process; putting the screened characteristic data of the excitation system fault into the set A = {a 1 , a 2 , a 3 , …, a n}, and putting the set of node abnormal correlation data generated during the fault occurrence into the set a n = {b n1 , b n2 , b n3 , …, b ni}; by classifying the stored historical data, the calculation time of the subsequent process is reduced, and while improving the efficiency, the accuracy of fault diagnosis is also improved. For example, in the historical stored data, 56.2 and 78.6 are the characteristic data of the excitation system fault, so they are put into the set A, A = {56.2, 78.6};

[0100] where a n represents the characteristic data of the excitation system fault, b ni represents the set of node abnormal correlation data, n represents there are n characteristic data of the excitation system fault, and i represents that one characteristic data of the excitation system fault corresponds to i sets of node abnormal correlation data.

[0101] In S3, the specific steps for establishing the fault database include:

[0102] Z1. Establish a multi-dimensional space with m pairwise intersecting and perpendicular lines, where m refers to the m-dimensional space and also refers to m nodes;

[0103] Z2. Convert each node data into a coordinate value, where the maximum and minimum values of the normal range of each node are P m and Q m , and the obtained node data is K m ;

[0104] When Q m < K m < P m , the coordinate value α m of this node is 0;

[0105] When Q m > K m , the coordinate value α m of this node is

[0106] When K m > Pm At this time, the coordinate value α of this node m is

[0107] enable the abnormal change sequence of each node to be reflected in the spatial coordinates, making the time sequence concrete;

[0108] where t τ represents the order in which this node generates abnormal data;

[0109] Z4. The node - associated data set is transformed into coordinates γ ni (α n1 , α n2 , …, α nm , t τ ), where α m is the node coordinate value, and t τ represents the number of abnormal data generated during the process of the excitation system failure; the transformed coordinates contain t τ , making the fault diagnosis more accurate;

[0110] Z5. All the node - associated data sets are transformed into coordinates and placed in an m - dimensional space, and the distance D between all the coordinates and the origin is calculated ni

[0111]

[0112] and placed in the set Thereby, a fault library is established.

[0113] In S4, the fault diagnosis refers to judging the cause of the excitation system failure according to the set of associated data of node anomalies. The specific steps of the fault diagnosis include:

[0114] L1. The set of associated data of node anomalies is transformed into coordinates. The coordinate value transformation formula for each node data is:

[0115] When Q m <K m <P m at this time, the coordinate value β of this node m is 0;

[0116] When Q m >K m at this time, the coordinate value β of this node m is

[0117] When K m >P m at this time, the coordinate value β of this node m is

[0118] The node - associated data set is transformed into coordinates ρ(β 1 , β 2 , …, β m , t κ );

[0119] For example, the maximum value P 1 of the normal range of node 1 is 90, and the minimum value Q 1 is 30. The data collected at node 1 is 60, which is within the normal range. Then the coordinate value β 1 of this node is 0;

[0120] The maximum value P 2 of the normal range of node 2 is 80, and the minimum value Q 2 is 40. The data collected at node 2 is 90, which exceeds the maximum value of the normal range, and node 2 first discovers the abnormal data during the observation process, that is, t 1 = 1. Then the coordinate value corresponding to this node 2

[0121] where α m is the node coordinate value, t κ represents the number of abnormal data generated during the failure of the excitation system. The maximum and minimum values of the normal - range data of each node are P m and Q m respectively, and K m represents the obtained node data;

[0122] L2. Place the coordinates in an m - dimensional space coordinate system and calculate the distance d of this point from the origin

[0123]

[0124] For example, for the coordinate ρ(2, 0, 1, 0, 2), the distance of the point from the origin And compare the elements in the set to screen out the coordinates that meet the conditions;

[0125] L3. Compare the distance d with the elements in the set When |d - D ni | = ε, then it is stipulated that D i = U l Calculate the angles between the coordinates ρ, coordinate γ i and the origin. The calculation formula is

[0126]

[0127] where l represents the number of D ni when |d - D ni | = ε;

[0128] L4. When and meets for cosθ ι If there is only one, it proves a successful diagnosis, and the diagnosis result is the fault corresponding to the node association data set corresponding to coordinate U l ;

[0129] In other cases, put the coordinates that meet into the set Ψ = {φ 1 , φ 2 , …, φ σ}, and interleave each coordinate value in coordinate ρ with each coordinate value in φ σ . If each coordinate difference is less than , it proves a successful diagnosis, and the diagnosis result is the fault corresponding to the node association data set corresponding to coordinate φ σ , where σ represents the number of the included angles between two vectors that meet ;

[0130] For example, only the included angle of coordinate γ 2 in coordinate ρ meets the condition, and the corresponding characteristic data is further obtained to complete the fault diagnosis;

[0131] Through distance comparison, angle comparison, and coordinate comparison, the fault diagnosis result of the excitation system is made more accurate. By comparing step by step, the comparison range is gradually narrowed, making the fault diagnosis efficiency higher, and reducing the calculation amount to a certain extent;

[0132] In S4, the fault prediction module refers to establishing a prediction model to predict faults based on the existing abnormal data. The specific steps of the fault prediction include:

[0133] I1. First, convert the existing node association data set into corresponding coordinates

[0134] I2. Calculate the included angle between coordinate i coordinate γ

[0135] I3. When , it indicates that the fault corresponding to the node association data set corresponding to the coordinate that meets is the predicted fault, where represents the maximum deviation.

[0136] In S5, the fault database update module refers to comparing the actual diagnostic results with the system's diagnostic results. If there is a deviation, the actual diagnostic results are transmitted to the fault database and matched with the original data in the fault database, and the data in the fault database is corrected in a timely manner to achieve real-time update of the fault database, making the data in the fault database more accurate and the error of matching comparison smaller.

[0137] A fault diagnosis system for the excitation system of a generator set, the fault diagnosis system includes an information acquisition module, a fault database establishment module, a fault prediction module, a fault diagnosis module and a fault database update module;

[0138] The information acquisition module is used to collect the data of the excitation system in real time during operation, and the collected data is the data of each node;

[0139] So that the node information can be observed in real time and the node abnormal conditions can be found in time;

[0140] The fault database establishment module is used to establish a multi-dimensional space fault database;

[0141] The fault database prediction module is used to predict the faults of the excitation system according to the existing node data;

[0142] Through fault prediction, corresponding measures can be taken in time to avoid, reduce or even avoid losses;

[0143] The fault diagnosis module is used to diagnose the faults of the excitation system;

[0144] So that the faults can be diagnosed accurately and in time, reducing losses;

[0145] The fault database update module is used to update the fault database;

[0146] So that the data in the fault database can be updated and corrected in time, increasing the accuracy of fault diagnosis and fault prediction;

[0147] The output end of the information acquisition module is connected to the input ends of the fault database establishment module, the fault prediction module and the fault diagnosis module, and the output end of the fault diagnosis module is connected to the input end of the fault database update module.

[0148] The information acquisition module includes a data acquisition unit, a data storage unit and a data classification unit;

[0149] The data acquisition unit is used to collect the data of each node in real time;

[0150] The data storage unit adopts distributed storage of the collected data;

[0151] So that it can efficiently manage the read cache and write cache and can automatically perform hierarchical storage;

[0152] The data classification unit is used to screen out the fault characteristics and the node anomaly correlation data set;

[0153] So as to reduce the computational amount in the subsequent process and be able to reduce the error rate of fault diagnosis and prediction;

[0154] The output end of the data acquisition unit is connected to the input ends of the data storage unit and the data classification unit.

[0155] The fault library establishment module includes a multi-dimensional space unit and a fault library establishment unit;

[0156] The multi-dimensional space unit is used to implement the establishment of a multi-dimensional space;

[0157] The fault library establishment unit is used to establish a multi-dimensional space database;

[0158] The output end of the multi-dimensional space unit is connected to the input end of the fault establishment unit.

[0159] The fault prediction end element includes a fault prediction unit and a prediction notification unit;

[0160] The fault prediction unit is used to establish an angle prediction model to conduct fault prediction;

[0161] So that the fault prediction is more accurate;

[0162] The prediction notification unit is used to notify the result of the fault prediction;

[0163] So as to be able to take corresponding measures in time to reduce losses;

[0164] The output end of the fault prediction unit is connected to the input end of the prediction notification unit.

[0165] The fault diagnosis module includes a fault diagnosis unit and a diagnosis reporting unit;

[0166] The fault diagnosis unit conducts fault diagnosis on the excitation system through distance comparison, angle comparison and coordinate comparison;

[0167] Through the three comparisons, the fault diagnosis result is more accurate;

[0168] The diagnosis reporting unit is used to report the fault diagnosis result of the excitation system; for example, predicting the possible faults of the excitation system through the abnormal data of several nodes and giving an early warning in time through the diagnosis reporting unit, and methods such as sending text messages and making phone calls can be adopted;

[0169] The fault library update module includes a data comparison end element and a fault library update unit;

[0170] The data comparison unit extracts the latest data by comparing the data in the original fault library with the latest fault data information;

[0171] The fault library update unit is used to update the fault library;

[0172] So that the data in the fault library can be continuously updated and corrected, continuously improving the accuracy of fault diagnosis and prediction;

[0173] The output end of the fault diagnosis unit is connected to the input end of the diagnosis reporting unit and the data comparison unit, and the output end of the data comparison unit is connected to the input end of the fault library update unit.

[0174] It can accurately and quickly diagnose faults in the excitation system and can realize fault prediction of the excitation system, enabling timely resolution of system faults, reducing certain economic losses and avoiding the occurrence of safety accidents to a certain extent.

[0175] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0176] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for fault diagnosis of the excitation system of a generator set, characterized in that: The fault diagnosis method includes the following steps: S1. Use the information acquisition module to collect and store the real-time data generated by the excitation system of the generator set; S2. Use the data classification unit to screen out the node abnormal correlation data of the excitation system; S3. Place the abnormal correlation data in a multi-dimensional space to establish a fault library; S4. Use the fault prediction module to match the real-time collected data with the data in the fault library to achieve fault prediction and use the fault diagnosis module to achieve fault diagnosis; S5. Use the fault library update module to achieve real-time update of the fault library; In S3, the specific steps for establishing the fault library include: Z1. Establish a multi-dimensional space with m pairwise intersecting and perpendicular lines, where m refers to m dimensions and also refers to m nodes; Z2. Convert each node data into coordinate values, where the maximum and minimum values of the normal range of each node are P m and Q m respectively, and the obtained node data is K m ; When Q m <K m <P m the coordinate value α of this node m is 0; When Q m > K m the coordinate value α of this node m is When K m > P m , the coordinate value α of this node m is where t τ represents the order in which the node generates abnormal data; Convert the Z4 and node association data sets into coordinates γ ni (α n1 , α n2 , …, α nm , t τ ), where α m is the node coordinate value, and t τ represents the number of abnormal data generated during the process of the excitation system failure; Z5. Convert the set of all node association data into coordinates and place them in an m-dimensional space, and calculate the distance D between all the coordinates and the origin. ni and put it into the set thus establishing a fault library; In S4, the fault diagnosis refers to judging the cause of the excitation system fault according to the set of correlation data of node abnormalities. The specific steps for fault diagnosis include: L1. Convert the set of correlation data of node abnormalities into coordinates, and the coordinate value conversion formula for each node data is: When Q m <K m <P m the coordinate value β of this node m is 0; When Q m >K m the coordinate value β of this node m is When K m > P m the coordinate value β of this node m is The node association data set is converted into coordinates ρ(β 1 , β 2 , …, β m , t κ ); where α m is the node coordinate value, and t κ represents the number of abnormal data generated during the fault of the excitation system. The maximum and minimum values of the normal range data of each node are P m and Q m respectively, and K m represents the node data obtained; L2. Place the coordinates in the m-dimensional space coordinate system and calculate the distance d from the coordinate ρ to the origin. L3. Compare the distance d with the elements in the set When |d - D ni | = ε, it is stipulated that D ni = U l , calculate the angles between the coordinates ρ, coordinate γ ni and the origin, and its calculation formula is where l represents the number of D when |d - D ni | = ε; D ni represents the distance between all coordinates and the origin, ε represents the error threshold, and U ni represents the absolute value corresponding to the vector γ transformed from the distance D l ; ρ is the coordinate of the node association data set established in step L1, and γ ni is the coordinate of the node association data set in the fault library; ni ni ​​ L4. When and it meets for cosθ ι if there is only one, it proves a successful diagnosis, and its diagnosis result is the fault corresponding to the node association data set corresponding to coordinate γ ni ; In other cases, coordinates that meet are put into the set Ψ = {φ 1 , φ 2 , …, φ σ}, and each coordinate value in the coordinate ρ is interleaved with each coordinate value of φ σ . If each coordinate difference is less than , it proves a successful diagnosis, and the diagnosis result is the fault corresponding to the node association data set corresponding to the coordinate φ σ , where σ represents the number of the included angles between two vectors that meet ; among them, is the angle threshold, and is the diagnosis threshold; In S4, the fault prediction module refers to establishing a prediction model to perform fault prediction based on the existing abnormal data. The specific steps for its fault prediction include: I1. First, convert the existing node association data set into corresponding coordinates I2. Calculate coordinates Coordinate γ i Angle with the origin I3. When it is the case, it indicates that the fault corresponding to the node association data set corresponding to the coordinates is the predicted fault, where represents the maximum deviation.

2. A method for fault diagnosis of the excitation system of a generator set according to claim 1, characterized in that: In S2, the abnormal correlation data screening refers to determining the characteristic data of the excitation system fault and the node abnormal correlation data generated during the fault occurrence process; putting the screened characteristic data of the excitation system fault into the set A = {a 1 , a 2 , a 3 , …, a n}, and putting the set of node abnormal correlation data generated during the fault occurrence into the set a n = {b n1 , b n2 , b n3 , …, b ni}; where a n represents the characteristic data of the excitation system fault, b ni represents the set of node abnormal correlation data, n represents that there are n pieces of characteristic data of the excitation system fault, and i represents that one piece of characteristic data of the excitation system fault corresponds to i sets of node abnormal correlation data.

3. A method for fault diagnosis of the excitation system of a generator set according to claim 1, characterized in that: In S5, the fault library update module refers to comparing the actual diagnosis result with the system's diagnosis result. If there is a deviation, the actual diagnosis result is transmitted to the fault library and matched and compared with the original data in the fault library, and the data in the fault library is corrected in time to achieve real-time update of the fault library.

4. A fault diagnosis system for the excitation system of a generator set, applicable to a method for fault diagnosis of the excitation system of a generator set according to any one of claims 1-3, characterized in that: The fault diagnosis system includes an information acquisition module, a fault library establishment module, a fault prediction module, a fault diagnosis module, and a fault library update module; The information acquisition module is used to collect the data of the excitation system during operation in real time; The fault library establishment module is used to establish a multi-dimensional space fault library; The fault library prediction module is used to perform fault prediction on the excitation system according to the existing node data; The fault diagnosis module is used to perform fault diagnosis on the excitation system; The fault library update module is used to update the fault library; The output end of the information acquisition module is connected to the input ends of the fault library establishment module, the fault prediction module, and the fault diagnosis module, and the output end of the fault diagnosis module is connected to the input end of the fault library update module.

5. A fault diagnosis system for the excitation system of a generator set according to claim 4, characterized in that: The information acquisition module includes a data acquisition unit, a data storage unit, and a data classification unit; The data acquisition unit is used to collect the data of each node in real time; The data storage unit adopts distributed storage of the collected data; The data classification unit is used to screen out the fault features and the node anomaly correlation data set; The output end of the data acquisition unit is connected to the input ends of the data storage unit and the data classification unit.

6. A fault diagnosis system for an excitation system of a generator set according to claim 4, wherein: The fault library establishment module includes a multi-dimensional space unit and a fault library establishment unit; The multi-dimensional space unit is used to establish a multi-dimensional space; The fault library establishment unit is used to establish a multi-dimensional space database; The output end of the multi-dimensional space unit is connected to the input end of the fault establishment unit.

7. A fault diagnosis system for an excitation system of a generator set according to claim 4, wherein: The fault prediction unit includes a fault prediction unit and a prediction notification unit; The fault prediction unit is used to establish an angle prediction model for fault prediction; The prediction notification unit is used to notify the result of the fault prediction; The output end of the fault prediction unit is connected to the input end of the prediction notification unit.

8. A fault diagnosis system for an excitation system of a generator set according to claim 4, wherein: The fault diagnosis module includes a fault diagnosis unit and a diagnosis reporting unit; The fault diagnosis unit diagnoses the faults of the excitation system through distance comparison, angle comparison and coordinate comparison; The diagnosis reporting unit is used to report the fault diagnosis result of the excitation system; The fault library update module includes a data comparison unit and a fault library update unit; The data comparison unit extracts the latest data by comparing the data in the original fault library with the latest fault data information; The fault library update unit is used to update the fault library; The output end of the fault diagnosis unit is connected to the input ends of the diagnosis reporting unit and the data comparison unit, and the output end of the data comparison unit is connected to the input end of the fault library update unit.

Images