An abnormal alarm method, device, equipment and medium for a hydro-generator unit

Abstract

This invention discloses an abnormal alarm method, device, equipment, and medium for hydro-turbine generator sets. It involves acquiring the current vibration data of the target hydro-turbine generator set in real time, preprocessing the current vibration data to obtain standard current vibration data, extracting features from the standard current vibration data using a preset vibration data transformation method to obtain current spectral kurtosis features, inputting the current spectral kurtosis features into a pre-trained abnormal alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis features, and if the cluster type is an abnormal alarm cluster, then an abnormal alarm operation is performed on the target hydro-turbine generator set. This solves the problem of difficulty in accurately alarming hydro-turbine generator sets based on collected vibration data, improves the accuracy and rate of hydro-turbine generator set alarms, and ensures the safety of the hydro-turbine generator set.

Description

Technical Field

[0001] This invention relates to the field of hydro-generator monitoring technology, and in particular to an abnormal alarm method, device, equipment and medium for hydro-generators. Background Technology

[0002] Hydropower generator sets are key and important equipment in the power production field, and are the core of whether a power plant can complete its production tasks. Among the existing power plant equipment monitoring systems, the condition monitoring of hydropower generator sets is the most comprehensive and rigorous.

[0003] In the process of realizing this invention, the inventors discovered the following defects in the prior art: Currently, existing abnormal alarm methods for hydro-generator units are generally concentrated on threshold alarms, and joint analysis is not carried out in the monitoring of vibration data of hydro-generator units, lacking the reasonable application of power big data. Summary of the Invention

[0004] This invention provides an abnormal alarm method, device, equipment, and medium for hydro-generator sets, so as to improve the accuracy and speed of alarms for hydro-generator sets and ensure the safety of hydro-generator sets.

[0005] According to one aspect of the present invention, an abnormal alarm method for a hydro-generator set is provided, comprising:

[0006] Real-time acquisition of current vibration data of the target hydro-generator unit, and data preprocessing of the current vibration data to obtain standard current vibration data;

[0007] By using a preset vibration data transformation method, feature extraction is performed on the current standard vibration data to obtain the current spectral kurtosis features;

[0008] The current spectral kurtosis feature is input into a pre-trained anomaly alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis feature;

[0009] If the cluster type is an abnormal alarm cluster, then an abnormal alarm operation is performed on the target hydro-generator unit.

[0010] According to another aspect of the present invention, an abnormal alarm device for a hydro-generator set is provided, comprising:

[0011] The standard current vibration data determination module is used to acquire the current vibration data of the target hydro-generator unit in real time, and to preprocess the current vibration data to obtain standard current vibration data.

[0012] The current spectral kurtosis feature determination module is used to extract features from the standard current vibration data using a preset vibration data transformation method to obtain the current spectral kurtosis feature;

[0013] The cluster type determination module is used to input the current spectral kurtosis feature into a pre-trained anomaly alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis feature;

[0014] The target hydro-generator unit abnormal alarm module is used to perform abnormal alarm operations on the target hydro-generator unit if the cluster type is an abnormal alarm cluster.

[0015] According to another aspect of the present invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the abnormal alarm method for a hydro-generator set according to any embodiment of the present invention.

[0016] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the abnormal alarm method for a hydro-generator set according to any embodiment of the present invention.

[0017] The technical solution of this invention involves acquiring the current vibration data of a target hydro-generator unit in real time, preprocessing the current vibration data to obtain standard current vibration data, extracting features from the standard current vibration data using a preset vibration data transformation method to obtain current spectral kurtosis features, inputting the current spectral kurtosis features into a pre-trained abnormal alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis features, and performing an abnormal alarm operation on the target hydro-generator unit if the cluster type is an abnormal alarm cluster. This solves the problem of difficulty in accurately alarming hydro-generator units based on collected vibration data, improves the accuracy and rate of hydro-generator unit alarms, and ensures the safety of hydro-generator units.

[0018] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1This is a flowchart of an abnormal alarm method for a hydro-generator set according to Embodiment 1 of the present invention;

[0021] Figure 2 This is a schematic diagram of the structure of an abnormal alarm device for a hydro-generator set according to Embodiment 2 of the present invention;

[0022] Figure 3 This is a schematic diagram of the structure of an electronic device provided according to Embodiment 3 of the present invention. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0024] It should be noted that the terms "target," "current," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] Example 1

[0026] Figure 1 The flowchart of an abnormal alarm method for a hydro-generator set is provided in Embodiment 1 of the present invention. This embodiment is applicable to the monitoring of vibration data of a hydro-generator set. The method can be executed by an abnormal alarm device of the hydro-generator set, which can be implemented in hardware and / or software.

[0027] Correspondingly, such as Figure 1 As shown, the method includes:

[0028] S110. Acquire the current vibration data of the target hydro-generator unit in real time, and perform data preprocessing on the current vibration data to obtain standard current vibration data.

[0029] The current vibration data can be used to describe the vibration status of the hydro-generator unit. Standard current vibration data can be obtained by preprocessing the current vibration data.

[0030] Specifically, the current vibration data can include vibration data of the top cover of the hydro-generator unit, vibration data of the upper guide bearing, and vibration data of the water guide bearing.

[0031] In this embodiment, it is necessary to acquire the current vibration data of the target hydro-generator unit in real time. Since the current vibration data may contain blank vibration data or the length of the vibration data may not meet the spectral kurtosis feature length threshold, it is necessary to perform data preprocessing on the current vibration data. After data preprocessing, standard current vibration data can be obtained.

[0032] Optionally, before acquiring the current vibration data of the target hydro-generator unit in real time and performing data preprocessing on the current vibration data to obtain standard current vibration data, the method further includes: acquiring multiple sets of historical vibration data of the hydro-generator unit, and processing the multiple sets of historical vibration data using a pre-set data preprocessing method to obtain multiple sets of standard historical vibration data; extracting features from the multiple sets of standard historical vibration data using a preset vibration data transformation method to obtain multiple sets of historical spectral kurtosis features; using at least three sets of historical spectral kurtosis features as inputs to the initial clustering algorithm model for model training, and training an abnormal alarm cluster confirmation model when the cluster confirmation accuracy corresponding to the initial clustering algorithm model meets the accuracy threshold.

[0033] Historical vibration data can be vibration data used to describe the history of the target hydro-generator unit. Standard historical vibration data can be data obtained by preprocessing historical vibration data. Historical spectral kurtosis features can be spectral kurtosis features obtained by feature extraction from standard historical vibration data.

[0034] In this embodiment, the acquired historical vibration data needs to be preprocessed to obtain multiple sets of standard historical vibration data. Furthermore, feature extraction needs to be performed on these multiple sets of standard historical vibration data to obtain corresponding sets of historical spectral kurtosis features.

[0035] Accordingly, multiple sets of historical spectral kurtosis features are processed in batches of at least three sets of historical spectral kurtosis features. After batch processing, a batch of historical spectral kurtosis features is gradually input into the initial clustering algorithm model for cluster training. When the cluster confirmation accuracy of the initial clustering algorithm model meets the accuracy threshold, the abnormal alarm cluster confirmation model is trained.

[0036] Specifically, the anomaly alarm cluster confirmation model can determine whether the input spectral kurtosis features belong to an anomaly alarm cluster. This model can be trained based on the k-medoids algorithm; no specific limitations are specified here.

[0037] Optionally, the data preprocessing method includes: a data completion method, a data normalization method, and a data slicing method; the step of processing multiple sets of historical vibration data using a pre-set data preprocessing method to obtain multiple sets of standard historical vibration data includes: determining whether there are blank historical vibration data in the multiple sets of historical vibration data; if so, obtaining the adjacent historical vibration data corresponding to the blank historical vibration data, and processing the data using the data completion method to obtain multiple sets of historically completed vibration data; and applying the data normalization formula... Data processing is performed on multiple sets of historical completed vibration data V to obtain multiple sets of historical normalized vibration data V′; the spectral kurtosis feature length threshold is obtained, and the multiple sets of historical normalized vibration data are processed by data segmentation to obtain multiple sets of standard historical vibration data.

[0038] Among them, historical completed vibration data can be data obtained by completing blank historical vibration data. Historical normalized vibration data can be data obtained by normalizing historical completed vibration data. The spectral kurtosis feature length threshold can be a pre-set size of the spectral kurtosis feature length threshold.

[0039] In this embodiment, assuming there are 80 sets of historical vibration data, it is necessary to determine whether there are any blank historical vibration data sets among these 80 sets. If there are blank historical vibration data sets, they are set 15 and set 50. Further, it is necessary to obtain set 14 and set 16 of historical vibration data sets, and calculate the average historical vibration data set based on the two sets of historical vibration data sets, using this average as set 15 of historical vibration data sets.

[0040] Accordingly, it is necessary to obtain the 49th and 51st sets of historical vibration data, and calculate the average historical vibration data based on the two sets of historical vibration data, which will then be used as the 50th set of historical vibration data.

[0041] Additionally, through the data normalization formula The historical completed vibration data is normalized to obtain historical normalized vibration data. After obtaining the historical normalized vibration data, the data is further segmented to obtain multiple sets of standard historical vibration data.

[0042] Specifically, the length of the historical normalized vibration data is L1, and the spectral kurtosis feature length threshold is k. mThe requirement is that the length of the historical normalized vibration data is greater than or equal to the spectral kurtosis feature length threshold. If the length of the historical normalized vibration data is less than the spectral kurtosis feature length threshold, the length of the historical normalized vibration data is discarded.

[0043] For example, assuming the spectral kurtosis feature length threshold is 100 and the length of the historical normalized vibration data is 125, the corresponding standard historical vibration data can be obtained through data segmentation. Specifically, the length of the standard historical vibration data is 100. Alternatively, if the length of the historical normalized vibration data is 25, then the historical normalized vibration data can be discarded.

[0044] Optionally, the step of extracting features from multiple sets of standard historical vibration data using a preset vibration data transformation method to obtain multiple sets of historical spectral kurtosis features includes: using formulas... Multiple sets of standard historical vibration data are subjected to Hilbert transform to obtain multiple sets of standard historical vibration data signals H(t); where V1(r) represents the standard historical vibration data; t represents the time point; τ represents the time component, and the value of τ ranges from (-∞, +∞); the multiple sets of standard historical vibration data signals are filtered by low-pass and high-pass filters respectively to obtain multiple sets of standard historical vibration data filtered signals; the multiple sets of standard historical vibration data filtered signals are then subjected to virtual-real decomposition processing to obtain multiple sets of standard historical vibration data decomposed signals; and finally, the signals are processed by formula... Feature extraction was performed on multiple sets of standard historical vibration data decomposition signals to obtain multiple sets of historical spectral kurtosis features K; where V″ represents the standard historical vibration data decomposition signal; C 4 For the calculation of the instantaneous moment of the fourth-order spectrum, C 2 Calculation of the instantaneous moment of the second-order spectrum.

[0045] The standard historical vibration data signal can be obtained by applying a Hilbert transform to the standard historical vibration data. The filtered standard historical vibration data signal can be obtained by applying low-pass and high-pass filters. The historical spectral kurtosis feature can be obtained by feature extraction from the decomposed standard historical vibration data signal.

[0046] Specifically, spectral kurtosis features can include the mean of spectral kurtosis, the standard deviation of spectral kurtosis, the skewness of spectral kurtosis, and the kurtosis of spectral kurtosis. Furthermore, spectral kurtosis is highly sensitive to transient impulse components in a signal, and can effectively identify transient impulses and their distribution in the frequency band from signals containing background noise.

[0047] S120. Using a preset vibration data transformation method, feature extraction is performed on the current standard vibration data to obtain the current spectral kurtosis features.

[0048] Among them, the current spectral kurtosis feature can be a feature used to describe the spectral kurtosis of the standard current vibration data.

[0049] Specifically, vibration data transformation methods include Hilbert transform, low-pass and high-pass filtering, virtual-real decomposition, and feature extraction. After these vibration data transformation methods, the current spectral kurtosis characteristics can be further obtained.

[0050] S130. Input the current spectral kurtosis feature into the pre-trained anomaly alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis feature.

[0051] The abnormal alarm cluster confirmation model can be a model that judges and analyzes the current spectral kurtosis characteristics to determine whether it belongs to an abnormal alarm cluster. The cluster type can include abnormal alarm clusters and normal alarm clusters.

[0052] In this embodiment, an abnormal alarm cluster confirmation model is needed to determine the cluster type corresponding to the current spectral kurtosis feature. Appropriate processing operations are then performed based on the cluster type.

[0053] S140. If the cluster type is an abnormal alarm cluster, then perform an abnormal alarm operation on the target hydro-generator unit.

[0054] Among them, the abnormal alarm cluster can be a cluster that describes the abnormality of the current spectral kurtosis feature.

[0055] In this embodiment, if the cluster type is determined to be an abnormal alarm cluster, it indicates that the current spectral kurtosis feature is abnormal, that is, the current vibration data collected from the target turbine unit is abnormal, and therefore alarm processing is required.

[0056] Optionally, after inputting the current spectral kurtosis feature into the pre-trained abnormal alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis feature, the method further includes: if the cluster type is not an abnormal alarm cluster, calculating the standard current vibration data distance using a distance classifier; comparing the standard current vibration data distance with a preset standard distance threshold, and processing the target hydro-generator unit based on the distance comparison result.

[0057] The distance classifier can be any classifier capable of distance calculation. The standard current vibration data distance can be the distance to the standard current vibration data calculated by the distance classifier. The standard distance threshold can be a pre-set distance threshold for determining whether an anomaly exists. The distance comparison result can be the result obtained by performing a distance comparison.

[0058] In this embodiment, if the cluster type corresponding to the current spectral kurtosis feature is not an abnormal alarm cluster, it cannot be directly determined that the target hydro-generator unit is in a normal state. It is also necessary to obtain the standard current vibration data distance through a distance classifier, further determine the distance, and process the target hydro-generator unit based on the distance comparison result.

[0059] Optionally, the step of comparing the current standard vibration data distance with a preset standard distance threshold and processing the target hydro-generator unit based on the distance comparison result includes: determining whether the current standard vibration data distance is greater than the preset standard distance threshold; if so, performing an abnormal alarm operation on the target hydro-generator unit; if not, determining that the target hydro-generator unit is in a normal state.

[0060] In this embodiment, it is necessary to determine whether the current standard vibration data distance is greater than a preset standard distance threshold. Specifically, if the current standard vibration data distance is greater than the preset standard distance threshold, it indicates that the current standard vibration data distance does not meet the requirements of the standard distance threshold, and an abnormal alarm operation needs to be performed on the target hydro-generator unit. Conversely, it can be determined that the target hydro-generator unit is in a normal state. This allows for a more accurate judgment of the state of the target hydro-generator unit, improving the reliability of the state judgment results.

[0061] Optionally, the step of calculating the standard current vibration data distance using a distance classifier includes: obtaining the three-dimensional channel current vibration data and the three-dimensional channel target cluster center corresponding to the standard current vibration data through the abnormal alarm cluster confirmation model; wherein, the three-dimensional channel current vibration data includes: the first-dimensional channel current vibration data, the second-dimensional channel current vibration data, and the third-dimensional channel current vibration data; the three-dimensional channel target cluster center includes: the first-dimensional channel target cluster center, the second-dimensional channel target cluster center, and the third-dimensional channel target cluster center; according to the formula Calculations are performed to obtain the standard current vibration data distance d; where x represents the standard current vibration data; x1 represents the current vibration data of the first-dimensional channel; x2 represents the current vibration data of the second-dimensional channel; x3 represents the current vibration data of the third-dimensional channel; m j Indicates the center of the three-dimensional channel target cluster; m j1 Indicates the center of the first-dimensional channel target cluster; m j2 Indicates the center of the target cluster in the second-dimensional channel; m j3 This indicates the center of the target cluster in the third-dimensional channel.

[0062] The current vibration data of the three-dimensional channel can be data from a three-dimensional channel associated with the standard current vibration data. The target cluster center of the three-dimensional channel can be the cluster center of the three-dimensional channel associated with the standard current vibration data.

[0063] Specifically, the current vibration data of the three-dimensional channels can include the current vibration data of the first-dimensional channel, the current vibration data of the second-dimensional channel, and the current vibration data of the third-dimensional channel. The target cluster center of the three-dimensional channels can include the target cluster center of the first-dimensional channel, the target cluster center of the second-dimensional channel, and the target cluster center of the third-dimensional channel.

[0064] In addition, the current vibration data in the three-dimensional channel can include data on abnormal conditions, pumping conditions, power generation conditions, and pumping phase adjustment conditions.

[0065] In this embodiment, the current vibration data of the three-dimensional channel and the target cluster center of the three-dimensional channel corresponding to the standard current vibration data are obtained. Further, according to the formula... The standard current vibration data distance is then calculated. Correspondingly, the calculated standard current vibration data distance is compared with a standard distance threshold, and the target hydro-generator unit is processed based on the distance comparison result.

[0066] The technical solution of this invention acquires the current vibration data of the target hydro-generator unit in real time, preprocesses the current vibration data to obtain standard current vibration data, extracts features from the standard current vibration data using a preset vibration data transformation method to obtain current spectral kurtosis features, inputs the current spectral kurtosis features into a pre-trained abnormal alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis features, and if the cluster type is an abnormal alarm cluster, performs an abnormal alarm operation on the target hydro-generator unit. This solves the problem of difficulty in accurately alarming hydro-generator units based on collected vibration data. By processing spectral kurtosis features, combining the vibration characteristics of the hydro-generator unit during operation, and jointly clustering multiple vibration data points, the accuracy and rate of hydro-generator unit alarms are improved, ensuring the safety of the hydro-generator unit.

[0067] Example 2

[0068] Figure 2 This is a schematic diagram of the structure of an abnormal alarm device for a hydro-generator set provided in Embodiment 2 of the present invention. The abnormal alarm device for a hydro-generator set provided in this embodiment can be implemented by software and / or hardware, and can be configured in a terminal device or server to implement an abnormal alarm method for a hydro-generator set according to an embodiment of the present invention. Figure 2As shown, the device includes: a standard current vibration data determination module 210, a current spectral kurtosis feature determination module 220, a cluster type determination module 230, and a target hydro-generator unit abnormal alarm module 240.

[0069] The standard current vibration data determination module 210 is used to acquire the current vibration data of the target hydro-generator unit in real time, and to perform data preprocessing on the current vibration data to obtain standard current vibration data.

[0070] The current spectral kurtosis feature determination module 220 is used to extract features from the standard current vibration data using a preset vibration data transformation method to obtain the current spectral kurtosis feature;

[0071] Cluster type determination module 230 is used to input the current spectral kurtosis feature into a pre-trained anomaly alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis feature;

[0072] The target hydro-turbine generator set abnormal alarm module 240 is used to perform an abnormal alarm operation on the target hydro-turbine generator set if the cluster type is an abnormal alarm cluster.

[0073] The technical solution of this invention involves acquiring the current vibration data of a target hydro-generator unit in real time, preprocessing the current vibration data to obtain standard current vibration data, extracting features from the standard current vibration data using a preset vibration data transformation method to obtain current spectral kurtosis features, inputting the current spectral kurtosis features into a pre-trained abnormal alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis features, and performing an abnormal alarm operation on the target hydro-generator unit if the cluster type is an abnormal alarm cluster. This solves the problem of difficulty in accurately alarming hydro-generator units based on collected vibration data, improves the accuracy and rate of hydro-generator unit alarms, and ensures the safety of hydro-generator units.

[0074] Optionally, it also includes an abnormal alarm cluster confirmation model training module, which can be specifically used to: before acquiring the current vibration data of the target hydro-generator unit in real time and performing data preprocessing on the current vibration data to obtain standard current vibration data, acquire multiple sets of historical vibration data of the hydro-generator unit, and process the multiple sets of historical vibration data through a pre-set data preprocessing method to obtain multiple sets of standard historical vibration data; extract features from the multiple sets of standard historical vibration data through a preset vibration data transformation method to obtain multiple sets of historical spectral kurtosis features; and train the model using at least three sets of historical spectral kurtosis features as inputs to the initial clustering algorithm model. When the cluster confirmation accuracy corresponding to the initial clustering algorithm model meets the accuracy threshold, the abnormal alarm cluster confirmation model is trained.

[0075] Optionally, it also includes an abnormal alarm cluster confirmation model training module, which can also be specifically used for: the data preprocessing method includes: data completion method, data normalization method, and data slicing method; the step of processing multiple sets of historical vibration data through a pre-set data preprocessing method to obtain multiple sets of standard historical vibration data includes: determining whether there is blank historical vibration data in the multiple sets of historical vibration data; if so, obtaining the adjacent historical vibration data corresponding to the blank historical vibration data, and processing the data through the data completion method to obtain multiple sets of historically completed vibration data; and using the data normalization formula... Data processing is performed on multiple sets of historical completed vibration data V to obtain multiple sets of historical normalized vibration data V′; the spectral kurtosis feature length threshold is obtained, and the multiple sets of historical normalized vibration data are processed by data segmentation to obtain multiple sets of standard historical vibration data.

[0076] Optionally, it also includes an anomaly alarm cluster confirmation model training module, which can also be specifically used for: through formulas Multiple sets of standard historical vibration data are subjected to Hilbert transform to obtain multiple sets of standard historical vibration data signals H(t); where V1(τ) represents the standard historical vibration data; t represents the time point; τ represents the time component, and the value range of τ is (-∞, +∞); the multiple sets of standard historical vibration data signals are filtered by low-pass and high-pass filters respectively to obtain multiple sets of standard historical vibration data filtered signals; the multiple sets of standard historical vibration data filtered signals are subjected to virtual-real decomposition processing to obtain multiple sets of standard historical vibration data decomposed signals; and the decomposed signals are obtained by applying the formula... Feature extraction was performed on multiple sets of standard historical vibration data decomposition signals to obtain multiple sets of historical spectral kurtosis features K; where V″ represents the standard historical vibration data decomposition signal; C 4 For the calculation of the instantaneous moment of the fourth-order spectrum, C 2 Calculation of the instantaneous moment of the second-order spectrum.

[0077] Optionally, it also includes a standard current vibration data distance determination module, which can be specifically used to: after inputting the current spectral kurtosis feature into the pre-trained abnormal alarm cluster confirmation model and determining the cluster type corresponding to the current spectral kurtosis feature, if the cluster type is not an abnormal alarm cluster, calculate the standard current vibration data distance through a distance classifier; compare the standard current vibration data distance with a preset standard distance threshold, and process the target hydro-generator unit according to the distance comparison result.

[0078] Optionally, it also includes a standard current vibration data distance determination module, which can be specifically used to: determine whether the standard current vibration data distance is greater than a preset standard distance threshold; if so, perform an abnormal alarm operation on the target hydro-generator unit; if not, determine that the target hydro-generator unit is in a normal state.

[0079] Optionally, it also includes a standard current vibration data distance determination module, which can be specifically used to: obtain the three-dimensional channel current vibration data and the three-dimensional channel target cluster center corresponding to the standard current vibration data through the abnormal alarm cluster confirmation model; wherein, the three-dimensional channel current vibration data includes: the first-dimensional channel current vibration data, the second-dimensional channel current vibration data, and the third-dimensional channel current vibration data; the three-dimensional channel target cluster center includes: the first-dimensional channel target cluster center, the second-dimensional channel target cluster center, and the third-dimensional channel target cluster center; according to the formula Calculations are performed to obtain the standard current vibration data distance d; where x represents the standard current vibration data; x1 represents the current vibration data of the first-dimensional channel; x2 represents the current vibration data of the second-dimensional channel; x3 represents the current vibration data of the third-dimensional channel; m j Indicates the center of the three-dimensional channel target cluster; m j1 Indicates the center of the first-dimensional channel target cluster; m j2 Indicates the center of the target cluster in the second-dimensional channel; m j3 This indicates the center of the target cluster in the third-dimensional channel.

[0080] The abnormal alarm device for hydro-generator sets provided in this embodiment of the invention can execute the abnormal alarm method for hydro-generator sets provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0081] Example 3

[0082] Figure 3A schematic diagram of an electronic device 10, which can be used to implement Embodiment 3 of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0083] like Figure 3 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0084] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0085] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the abnormal alarm method for a hydroelectric generator set.

[0086] In some embodiments, the abnormal alarm method for the hydro-generator set can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the abnormal alarm method for the hydro-generator set described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the abnormal alarm method for the hydro-generator set by any other suitable means (e.g., by means of firmware).

[0087] The method includes: acquiring the current vibration data of the target hydro-generator unit in real time, and preprocessing the current vibration data to obtain standard current vibration data; extracting features from the standard current vibration data using a preset vibration data transformation method to obtain current spectral kurtosis features; inputting the current spectral kurtosis features into a pre-trained abnormal alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis features; if the cluster type is an abnormal alarm cluster, then performing an abnormal alarm operation on the target hydro-generator unit.

[0088] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0089] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0090] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0091] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0092] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0093] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0094] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0095] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

[0096] Example 4

[0097] Embodiment 4 of the present invention also provides a computer-readable storage medium, wherein the computer-readable instructions, when executed by a computer processor, are used to perform an abnormal alarm method for a hydro-generator unit. The method includes: acquiring the current vibration data of the target hydro-generator unit in real time, and performing data preprocessing on the current vibration data to obtain standard current vibration data; extracting features from the standard current vibration data using a preset vibration data transformation method to obtain current spectral kurtosis features; inputting the current spectral kurtosis features into a pre-trained abnormal alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis features; if the cluster type is an abnormal alarm cluster, then performing an abnormal alarm operation on the target hydro-generator unit.

[0098] Of course, the computer-executable instructions provided in the embodiments of the present invention, which include a computer-readable storage medium, are not limited to the method operations described above, but can also perform related operations in the abnormal alarm method for hydro-generator sets provided in any embodiment of the present invention.

[0099] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0100] It is worth noting that in the embodiments of the above-mentioned abnormal alarm device for hydro-generator sets, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0101] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An abnormal alarm method for a hydro-generator unit, characterized in that, include: Real-time acquisition of current vibration data of the target hydro-generator unit, and data preprocessing of the current vibration data to obtain standard current vibration data; By using a preset vibration data transformation method, feature extraction is performed on the current standard vibration data to obtain the current spectral kurtosis features; The current spectral kurtosis feature is input into a pre-trained anomaly alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis feature; If the cluster type is an abnormal alarm cluster, then perform an abnormal alarm operation on the target hydro-generator unit; The process, prior to acquiring the current vibration data of the target hydro-generator unit in real time and performing data preprocessing on the current vibration data to obtain standard current vibration data, further includes: Multiple sets of historical vibration data of the hydro-generator unit are acquired, and the multiple sets of historical vibration data are processed by a pre-set data preprocessing method to obtain multiple sets of standard historical vibration data. By using a preset vibration data transformation method, features are extracted from multiple sets of standard historical vibration data to obtain multiple sets of historical spectral kurtosis features. In multiple sets of historical spectral kurtosis features, at least three sets of historical spectral kurtosis features are used as inputs to the initial clustering algorithm model for model training. When the cluster confirmation accuracy of the initial clustering algorithm model meets the accuracy threshold, the abnormal alarm cluster confirmation model is trained. The step of inputting the current spectral kurtosis feature into a pre-trained anomaly alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis feature further includes: If the cluster type is not an abnormal alarm cluster, the standard current vibration data is calculated by the distance classifier to obtain the standard current vibration data distance; The target hydro-generator unit is processed based on the distance comparison result by comparing the current standard vibration data distance with a preset standard distance threshold. The step of calculating the standard current vibration data distance using a distance classifier includes: The abnormal alarm cluster confirmation model is used to obtain the current vibration data of the three-dimensional channel and the target cluster center of the three-dimensional channel corresponding to the standard current vibration data. The current vibration data of the three-dimensional channel includes: the current vibration data of the first-dimensional channel, the current vibration data of the second-dimensional channel, and the current vibration data of the third-dimensional channel; the target cluster center of the three-dimensional channel includes: the target cluster center of the first-dimensional channel, the target cluster center of the second-dimensional channel, and the target cluster center of the third-dimensional channel. According to the formula Calculations are performed to obtain the standard current vibration data distance d; Where x represents the current standard vibration data; x1 represents the current vibration data of the first-dimensional channel; x2 represents the current vibration data of the second-dimensional channel; x3 represents the current vibration data of the third-dimensional channel; m j Indicates the center of the three-dimensional channel target cluster; m j1 Indicates the center of the first-dimensional channel target cluster; m j2 Indicates the center of the target cluster in the second-dimensional channel; m j3 This indicates the center of the target cluster in the third-dimensional channel.

2. The method according to claim 1, characterized in that, The data preprocessing methods include: data completion method, data normalization method, and data slicing method; The process of preprocessing multiple sets of historical vibration data using a pre-set data preprocessing method yields multiple sets of standard historical vibration data, including: Determine whether there are blank historical vibration data in multiple sets of historical vibration data. If there are, obtain the adjacent historical vibration data corresponding to the blank historical vibration data, and process the data through the data completion method to obtain multiple sets of historical completed vibration data. Data normalization formula Data processing was performed on multiple sets of historical completed vibration data V to obtain multiple sets of historical normalized vibration data V′; The spectral kurtosis feature length threshold is obtained, and the multiple sets of historical normalized vibration data are processed by data segmentation to obtain multiple sets of standard historical vibration data.

3. The method according to claim 2, characterized in that, The method involves extracting features from multiple sets of standard historical vibration data using a preset vibration data transformation method, resulting in multiple sets of historical spectral kurtosis features, including: Through formula Hilbert transform is performed on multiple sets of standard historical vibration data to obtain multiple sets of standard historical vibration data signals H(t); Where V1(τ) represents standard historical vibration data; t represents a time point; τ represents the time component, and the value range of τ is (-∞, +∞); Multiple sets of standard historical vibration data signals are filtered by low-pass and high-pass filters respectively to obtain multiple sets of filtered standard historical vibration data signals. The filtered signals of the multiple sets of standard historical vibration data are subjected to virtual-real decomposition processing to obtain the decomposed signals of the multiple sets of standard historical vibration data. Through formula Feature extraction was performed on the decomposed signals of multiple sets of standard historical vibration data to obtain multiple sets of historical spectral kurtosis features K; Where V″ represents the decomposed signal of standard historical vibration data; C 4 For the calculation of the instantaneous moment of the fourth-order spectrum, C 2 Calculation of the instantaneous moment of the second-order spectrum.

4. The method according to claim 1, characterized in that, The step of comparing the current standard vibration data distance with a preset standard distance threshold, and processing the target hydro-generator unit based on the distance comparison result, includes: Determine whether the current vibration data distance of the standard is greater than the preset standard distance threshold. If so, perform an abnormal alarm operation on the target hydro-generator unit. If not, then the target hydro-generator unit is determined to be in normal condition.

5. An abnormal alarm device for a hydro-generator set, characterized in that, include: The standard current vibration data determination module is used to acquire the current vibration data of the target hydro-generator unit in real time, and to preprocess the current vibration data to obtain standard current vibration data. The current spectral kurtosis feature determination module is used to extract features from the standard current vibration data using a preset vibration data transformation method to obtain the current spectral kurtosis feature; The cluster type determination module is used to input the current spectral kurtosis feature into a pre-trained anomaly alarm cluster confirmation model to determine the cluster type corresponding to the current spectral kurtosis feature; The target hydro-generator unit abnormal alarm module is used to perform an abnormal alarm operation on the target hydro-generator unit if the cluster type is an abnormal alarm cluster. This includes an abnormal alarm cluster confirmation model training module, used for: acquiring multiple sets of historical vibration data of the hydro-generator unit before acquiring the current vibration data of the target hydro-generator unit in real time and performing data preprocessing on the current vibration data to obtain standard current vibration data; processing the multiple sets of historical vibration data through a pre-set data preprocessing method to obtain multiple sets of standard historical vibration data; extracting features from the multiple sets of standard historical vibration data through a preset vibration data transformation method to obtain multiple sets of historical spectral kurtosis features; using at least three sets of historical spectral kurtosis features as input to the initial clustering algorithm model for model training; and training an abnormal alarm cluster confirmation model when the cluster confirmation accuracy corresponding to the initial clustering algorithm model meets the accuracy threshold. The system also includes a standard current vibration data distance determination module, used for: after inputting the current spectral kurtosis feature into a pre-trained abnormal alarm cluster confirmation model and determining the cluster type corresponding to the current spectral kurtosis feature, if the cluster type is not an abnormal alarm cluster, calculating the standard current vibration data distance using a distance classifier; comparing the standard current vibration data distance with a preset standard distance threshold, and processing the target hydro-generator unit based on the distance comparison result; This also includes a standard current vibration data distance determination module, which is further used to: obtain the three-dimensional channel current vibration data and the three-dimensional channel target cluster center corresponding to the standard current vibration data through the abnormal alarm cluster confirmation model; wherein, the three-dimensional channel current vibration data includes: the first-dimensional channel current vibration data, the second-dimensional channel current vibration data, and the third-dimensional channel current vibration data; the three-dimensional channel target cluster center includes: the first-dimensional channel target cluster center, the second-dimensional channel target cluster center, and the third-dimensional channel target cluster center; according to the formula Calculations are performed to obtain the standard current vibration data distance d; where x represents the standard current vibration data; x1 represents the current vibration data of the first-dimensional channel; x2 represents the current vibration data of the second-dimensional channel; x3 represents the current vibration data of the third-dimensional channel; m j Indicates the center of the three-dimensional channel target cluster; m j1 Indicates the center of the first-dimensional channel target cluster; m j2 Indicates the center of the target cluster in the second-dimensional channel; m j3 This indicates the center of the target cluster in the third-dimensional channel.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the abnormal alarm method for the hydro-generator set as described in any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are used to cause a processor to execute the abnormal alarm method for a hydro-generator unit as described in any one of claims 1-4.

Images