Hydro-generator fault monitoring method, system and equipment based on complex network and medium

Abstract

The invention relates to the technical field of hydroelectric generator fault diagnosis, in particular to a hydroelectric generator fault monitoring method, system and device based on a complex network and a medium, a complex network model is constructed based on an actual hydroelectric generator set, and all parts of the hydroelectric generator set serve as nodes of the complex network model. Performing node edge connection according to the weight coefficients of the features between the nodes to obtain an adjacent matrix of the complex network model; acquiring operation data in real time according to sensors arranged on all parts of the actual water-turbine generator set, wherein the operation data comprises at least one of temperature, pressure, current, voltage and vibration frequency; calculating a complex network index according to the operation data and the complex network model; and comparing the complex network index with the network index reference value, and identifying the abnormal condition of the network index by using a statistical analysis method based on a comparison result. According to the method, fault monitoring and analysis can be accurately carried out.

Description

Technical Field

[0001] The invention relates to the technical field of hydraulic generator fault diagnosis, and in particular to a hydraulic generator fault monitoring method, system, equipment and medium based on a complex network. Background Art

[0002] As the global demand for clean energy continues to increase, hydropower, as an important renewable energy technology, has received widespread attention and application. However, as the core equipment of a hydropower station, the operating status of a hydropower generator directly affects the efficiency and stability of the entire power generation system. Traditional methods for monitoring and fault detection of hydropower generators mainly rely on regular maintenance and manual inspections, which often have problems such as poor real-time performance, low efficiency, and insufficient detection accuracy, and cannot meet the requirements of modern power systems for efficient and reliable operation.

[0003] In this context, complex network theory, as an important tool for studying the overall characteristics and dynamic behavior of a system, has gradually been applied to the state monitoring and analysis of various industrial systems. Complex networks can reveal the mutual relationship and influence of various components in the system through the topological structure of nodes and edges, providing a new perspective and method for fault detection and propagation analysis.

[0004] In recent years, with the advancement of sensor technology and data acquisition technology, the operating data of hydropower generators can be collected and transmitted more efficiently and accurately. This provides a reliable data basis for state assessment and fault analysis based on complex networks. However, in practical applications, how to organically combine complex network theory with hydropower generator state monitoring to achieve real-time assessment of the state of hydropower generators and fault prediction still faces many challenges and technical difficulties. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a method, system, equipment and medium for monitoring faults of a hydro-turbine generator based on a complex network in view of the deficiencies in the above-mentioned prior art, which solves the problem that the current status of a hydro-turbine generator set cannot be effectively evaluated in real time and faults cannot be accurately predicted.

[0006] The objective of the present invention is achieved by the following technical solutions: In a first aspect, the present invention provides a method for monitoring faults of a hydro-generator based on a complex network, comprising: A complex network model is constructed based on an actual hydro-generator set, and each component of the hydro-generator set is used as a node of the complex network model. The nodes are connected according to the weight coefficients of the characteristics between the nodes to obtain an adjacency matrix of the complex network model, wherein the characteristics include data correlation and location proximity; Acquiring operating data in real time according to sensors installed in various components of the actual hydro-generator set, wherein the operating data includes at least one of temperature, pressure, current, voltage and vibration frequency; Calculating complex network indicators based on the operating data and the complex network model; Based on the comparison between complex network indicators and network indicator reference values, statistical analysis methods are used to identify abnormal network indicators based on the comparison results. As a further improvement of the present invention, the components of the hydro-generator set are used as nodes of the complex network model, and the nodes are connected according to the weight coefficients of the characteristics between the nodes to obtain the adjacency matrix of the complex network model, which specifically includes: The upper guide bearing, lower guide bearing, water guide bearing, thrust bearing and its bracket, stator and its base, top cover and various bearings of the hydro-generator set are used as nodes of the complex network model; The Pearson correlation coefficient method is used to calculate the data correlation of temperature, vibration, pressure and current between each node. The Euclidean distance is used to calculate the physical distance between two nodes in the actual hydro-turbine generator set to determine the location proximity between the nodes. The comprehensive weight coefficient is obtained according to the data correlation, location proximity and the weight coefficients of data correlation and location proximity. The two nodes whose comprehensive weight coefficients exceed the set threshold are connected.

[0007] As a further improvement of the present invention, complex network indicators are calculated based on the operating data and the complex network model, and the complex network indicators include node degree values, network clustering coefficients, and average path lengths between nodes; The node degree is used to indicate the number of connections between a node and other nodes. The node degree is calculated as follows:

[0008] in, k i For Node v i The degree value, a ij Represents the element in the i-th row and j-th column of the adjacency matrix of the complex network model. v i With Node v j If there is an edge connection between a ij =1, otherwise, a ij =0; The network clustering coefficient is obtained by averaging the clustering coefficients of all nodes. v i The clustering coefficient calculation expression is:

[0009] in, C i For Node v i The clustering coefficient, e i Representation and Node v i The number of edges that actually exist between neighbor nodes, k i Is a node v i The degree value of The node average path length is used to calculate the average value of the shortest path length between any two nodes in a complex network; the node average path length is calculated as follows:

[0010] in, L is the average node path length, N is the total number of nodes in the network, d ij For Node v i To Node v j The shortest path length is v i and v j If the two paths are unreachable, the shortest path length will be set to infinity.

[0011] As a further improvement of the present invention, the comparison based on the relevant network indicator and the network indicator reference value specifically includes: The network index reference value is obtained according to the baseline network model, and the baseline network model construction process is: obtaining the operating data of the hydro-generator set during normal operation, and preprocessing it, and establishing the baseline network model according to the preprocessed operating data; Calculating baseline network indicators according to the baseline network model, wherein the baseline network indicators include node degree centrality, clustering coefficient, and betweenness centrality; The network index reference value for normal operation of the hydro-generator set is determined according to the baseline network index.

[0012] As a further improvement of the present invention, the comparison between the complex network indicator and the network indicator reference value and the identification of abnormal network indicator conditions using a statistical analysis method based on the comparison result specifically include: Calculate the mean and standard deviation of each complex network indicator based on the baseline network model; The Z-score corresponding to each complex network indicator is calculated using the Z-score statistical analysis model, and the set indicator threshold is used to compare the Z-score corresponding to the complex network indicator of each node. When the Z-score of a node corresponding to a complex network indicator exceeds the set indicator threshold, the abnormal node is regarded as a faulty component.

[0013] As a further improvement of the present invention, the method of identifying abnormal network indicators by using statistical analysis method, when abnormal nodes are detected, also includes using SI model to simulate the fault propagation process of hydro-turbine generator sets in different scenarios.

[0014] As a further improvement of the present invention, the SI model is used to simulate the fault propagation process of the hydro-generator set in different scenarios, specifically including: Initialize all nodes to the susceptible state, and select one or more initial fault nodes to be set to the infected state; Set the infection rate and propagation time step; at each time step, detect the susceptible neighboring nodes connected to each infected node; calculate the infection probability of the susceptible neighboring nodes based on the infection rate; Update the node status according to the infection probability of the susceptible neighboring nodes, and record the number and distribution of infected nodes at the current time step; When all nodes are infected, or the number of infected nodes does not change after multiple time steps, the iteration stops; An infection path diagram is drawn according to the current node infection situation, and the infection path diagram is used to display the propagation path of the fault in the network and identify the key propagation paths and nodes.

[0015] In a second aspect, the present invention further provides a complex network-based hydro-generator fault monitoring system, which is used to implement the above-mentioned complex network-based hydro-generator fault monitoring method, comprising: A complex network model building module builds a complex network model based on an actual hydro-generator set, takes various components of the hydro-generator set as nodes of the complex network model, connects nodes according to weight coefficients of features between nodes, and obtains an adjacency matrix of the complex network model, wherein the features include data correlation and location proximity; A data acquisition module, used to obtain monitoring data according to sensors installed in various components of the actual hydro-generator set, wherein the monitoring data includes at least one of temperature, pressure, current, voltage and vibration frequency; An indicator calculation module calculates complex network indicators based on the monitoring data and the complex network model; The anomaly identification module is used to compare complex network indicators with network indicator reference values, and based on the comparison results, use statistical analysis methods to identify abnormal situations in network indicators.

[0016] In a third aspect, the present invention also provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, which, when executed by a computing device, enable the computing device to execute the above-mentioned complex network-based hydro-turbine generator fault monitoring method.

[0017] In a fourth aspect, the present invention further provides a computing device, comprising: One or more processors, a memory and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include steps for executing a complex network-based hydro-generator fault monitoring method.

[0018] The beneficial effect of the present invention is that in the hydraulic generator fault monitoring method based on complex network of the present invention, by constructing a complex network model and analyzing the operation data in real time, real-time monitoring and early warning of the state of the hydraulic generator set can be achieved. Once an abnormal situation is found, measures can be taken in time to deal with it, so as to avoid the expansion of the fault or cause serious consequences. The key nodes and fault points in the hydraulic turbine system can be effectively identified through the complex network model, the accuracy and sensitivity of fault detection can be improved, and the risk of misjudgment and missed judgment can be reduced.

[0019] Furthermore, by calculating the complex network indicators of the turbine generator set, we can deeply understand the structure and dynamic characteristics of the hydropower generator system. This helps to identify key components and potential risks in the system, and can also guide condition monitoring, fault prediction and maintenance optimization, thereby improving the reliability and safety of the entire generator system.

[0020] Furthermore, by simulating the propagation of faults in complex networks through the SI model, the propagation path and scope of faults can be effectively predicted and analyzed. This model helps to identify key nodes and weak links in the system, providing strong support for the system's fault prevention and maintenance strategies. By analyzing the results of the propagation model, more accurate fault isolation and protection strategies can be formulated to improve the overall stability and reliability of the system.

[0021] In summary, the present invention can reduce the frequency and severity of faults through accurate fault detection and effective propagation analysis, thereby reducing equipment damage and downtime, and saving maintenance and operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 is a schematic flow chart of a method for monitoring a fault of a hydro-turbine generator based on a complex network in an embodiment of the present invention; Figure 2 is a schematic diagram of a method for monitoring faults of a hydro-generator based on a complex network in an embodiment of the present invention; Figure 3 It is a structural schematic diagram of a complex network-based hydraulic generator fault monitoring system in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose and technical solution of the present invention clearer and easier to understand, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0025] The concept of the present invention is to provide a method, system, device and medium for monitoring faults of a hydro-generator based on a complex network. The method mainly includes: A complex network model is constructed based on an actual hydro-generator set, and each component of the hydro-generator set is used as a node of the complex network model. The nodes are connected according to the weight coefficients of the characteristics between the nodes to obtain an adjacency matrix of the complex network model, wherein the characteristics include data correlation and location proximity; Acquiring operating data in real time according to sensors installed in various components of the actual hydro-generator set, wherein the operating data includes at least one of temperature, pressure, current, voltage and vibration frequency; Calculating complex network indicators based on the operating data and the complex network model; Based on the comparison between complex network indicators and network indicator reference values, statistical analysis methods are used to identify abnormal network indicators based on the comparison results.

[0026] By constructing a complex network model, the present invention can effectively identify key nodes and fault points in the turbine system, improve the accuracy and sensitivity of fault detection, and reduce the risks of misjudgment and missed judgment.

[0027] The technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings and specific embodiments, wherein the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments.

[0028] Example 1 like Figure 1 , Figure 2 As shown, this embodiment provides a method for monitoring faults of a hydro-turbine generator based on a complex network. By constructing a complex network model of a hydro-turbine generator set, the present invention can accurately capture the relationships between key components of the generator and use network analysis technology to conduct in-depth analysis and monitoring of these relationships.

[0029] Firstly, a complex network model is constructed based on the actual hydro-turbine generator set. The various components of the hydro-turbine generator set are used as nodes of the complex network model. The nodes are connected according to the weight coefficients of the characteristics between the nodes to obtain the adjacency matrix of the complex network model. The characteristics include data correlation and location proximity.

[0030] In this embodiment, the complex network model G = (V, E) is composed of a node set V and an edge set E. Assume that the number of nodes in the network model G is n, and the node set is represented by V ={ v 1, v 2, v 3, ..., v n}, the number of edges is m, and the set of edges is represented by E ={ e 1, e 2, e 3, ..., e m} to indicate.

[0031] The upper guide bearing, lower guide bearing, water guide bearing, thrust bearing and its bracket, stator and its base, top cover and various bearings of the hydro-generator set are used as nodes of the complex network model. The number of nodes is N.

[0032] The edges between nodes are jointly determined by two features: data correlation and location proximity.

[0033] Specifically, the Pearson correlation coefficient method is used to calculate the data correlation or similarity of temperature, vibration, pressure and current between each node. ij .

[0034] Position proximity refers to the degree of proximity between two nodes in physical space. A high proximity means that they are physically close. Therefore, the Euclidean distance is used to calculate the physical distance between two nodes in the actual hydropower generator set to determine the position proximity between the nodes.

[0035]

[0036] In the formula, is the Euclidean distance between node i and node j, and the spatial position of node i is ( , , ), the spatial position of node j is ( ).

[0037] After comprehensively considering the above two factors, the comprehensive weight coefficient is obtained according to the data correlation, location proximity and the weight coefficient of data correlation and location proximity. The edge between nodes is determined by calculating the comprehensive weight method, and the calculation method is as follows:

[0038] In the formula, is the comprehensive weight coefficient, is the weight coefficient of data relevance, is the weight coefficient of location proximity.

[0039] Establish an edge between two nodes whose comprehensive weight coefficient exceeds the set threshold. Specifically, the simplified threshold is set to 0.5. After calculating the comprehensive weight, if the comprehensive weight w ij If the threshold exceeds 0.5, i and nodes j Establish a connection between them.

[0040] The key parts of the turbine are abstracted as nodes, and the nodes that meet the comprehensive weight requirements after comprehensive consideration of data relevance and location proximity are connected. Finally, the adjacency matrix A of the turbine complex network is obtained:

[0041] Adjacency Matrix A The dimension is N×N , the adjacency matrix A Elements in a ij Representation Node i and nodes j The connection relationship between: a ij =1 indicates a node i and nodes j Direct connection; a ij =0 means node i and nodes j Not connected.

[0042] The operating data is acquired in real time according to sensors installed in various components of the actual hydro-generator set, and the operating data includes at least one of temperature, pressure, current, voltage and vibration frequency.

[0043] Specifically, data is collected in real time through sensors installed on key components of the hydroelectric generator, including temperature sensors, vibration sensors, pressure sensors, current sensors, and voltage sensors. The collected data is transmitted to the control unit or data processing center through wired or wireless transmission technology. The collected data undergoes preprocessing steps such as denoising, normalization, and missing value processing to ensure the accuracy and consistency of the data, and is ultimately used to build a complex network model of the hydroelectric generator set for condition monitoring and fault analysis.

[0044] Calculate complex network indicators based on operating data and complex network models, including node degree distribution, network clustering coefficient, and average path length between nodes.

[0045] Specifically, based on the preprocessed operating data, the network topology of the generator is constructed, and the relationship matrix of nodes and edges is generated. The network's degree distribution, clustering coefficient, average path length and other indicators are calculated to analyze the overall structural characteristics of the network.

[0046] The degree of a node indicates the number of connections between a node and other nodes. v i The value of k i The calculation method is:

[0047] in, a ij Represents an element in the adjacency matrix. If a node v i With Node v j If there is an edge connection between a ij =1; otherwise, a ij = 0. By calculating the node degree value, the key parts in the turbine can be identified. This key part is the central hub of the turbine information or energy. By identifying the height value node, the key monitoring components can be guided to ensure their normal operation.

[0048] The network clustering coefficient reflects the degree of connection between the neighboring nodes of a node. v The clustering coefficient C i The calculation method is:

[0049] in, e i Representation and Nodev i The number of edges that actually exist between neighbor nodes, k i Is a node v i The degree value.

[0050] The global clustering coefficient of the network C It can be obtained by taking the average of the clustering coefficients of all nodes:

[0051] in, N is the total number of nodes in the network.

[0052] By calculating the network clustering coefficient, we can describe the degree of connection between a node and its neighboring nodes. A high clustering coefficient means that there are more local groups or modules in the network. In the complex network of generators, the modular structure may correspond to functional modules or component groups, which is convenient for identifying and analyzing modular fault propagation.

[0053] The average path length of a node is the average of the shortest path lengths between any two nodes in the network. v i To Node v j The shortest path length is d ij , the average path length of nodes L The calculation method is:

[0054] in, N is the total number of nodes in the network. v i and v j If the two are not reachable, usually d ij Set to infinity.

[0055] By calculating the average path length, the average shortest path between any two nodes in the network can be identified, reflecting the speed of information or fault propagation in the network. A shorter average path length means that information or faults can propagate quickly throughout the system, requiring a faster detection and response mechanism.

[0056] By calculating the complex network indicators of the turbine, we can deeply understand the structure and dynamic characteristics of the hydropower generator system. It helps to identify the key components and potential risks in the system, and can also guide condition monitoring, fault prediction and maintenance optimization, thereby improving the reliability and safety of the entire generator system.

[0057] Based on the comparison between complex network indicators and network indicator reference values, statistical analysis methods are used to identify abnormal network indicators based on the comparison results.

[0058] The network index reference values ​​are obtained according to the baseline network model. The baseline network model construction process is as follows: the operating data of the hydro-turbine generator set during normal operation is obtained and preprocessed, and a baseline network model is established according to the preprocessed operating data; the baseline network indexes are calculated according to the baseline network model, and the baseline network indexes include node degree values, clustering coefficients, and betweenness centrality; and the network index reference values ​​for the normal operation of the hydro-turbine generator set are determined according to the baseline network indexes.

[0059] Specifically, when the system is operating normally, a large amount of operating data such as vibration, temperature, pressure, current, etc. is collected, and the collected data is cleaned, denoised, and normalized to eliminate abnormal points and noise interference, and a baseline network model is established based on the preprocessed normal data.

[0060] According to the baseline network model, key baseline turbine complex network indicators including node degree, clustering coefficient, betweenness centrality, etc. are calculated, the statistical distribution of network indicators under normal conditions is analyzed, the characteristic values ​​and thresholds of normal operation are determined, and the indicator reference values ​​under normal conditions are formed.

[0061] Based on the comparison between complex network indicators and network indicator reference values, statistical analysis is used to identify abnormal network indicators, including: Collect turbine operation data in real time, dynamically update complex network models based on real-time data, calculate new network indicators in real time and compare them with the baseline model, detect changes in network structure and indicators in real time, compare real-time network indicators with baseline values, and analyze changing trends of indicators.

[0062] This embodiment adopts the Z-score model when analyzing abnormal network indicators. The statistical analysis method Z-score is used to identify abnormal patterns and determine whether they deviate from the normal state. Specifically: The mean and standard deviation of each complex network indicator are calculated based on the baseline network model. Specifically, for each network indicator (such as node degree, clustering coefficient, etc.), a large amount of data is collected under normal conditions to calculate its mean and standard deviation. For each indicator value collected in real time, its Z-score can be calculated to obtain its standardized deviation degree.

[0063] The Z-score corresponding to each complex network indicator is calculated using the Z-score statistical analysis model, and the set indicator threshold is used to compare the Z-score corresponding to each complex network indicator. When the Z-score of a node corresponding to a complex network indicator exceeds the set indicator threshold, the abnormal node is regarded as a faulty component.

[0064] By setting appropriate indicator thresholds (for example, indicator threshold |Z|>2), outliers that deviate significantly from the normal state can be identified. When the Z-score of a network indicator exceeds the set threshold, it can be considered that the indicator is abnormal, and the abnormal node can be located as a possible fault component.

[0065] After monitoring the abnormal nodes, complex network theory can also be used to analyze fault propagation. Since some mechanical systems or electrical equipment of the turbine require human intervention to repair once a fault occurs, there is no automatic fault recovery mechanism. Therefore, this embodiment uses the SI model (susceptible-infectious model) to simulate the fault propagation process of the turbine generator set under different scenarios.

[0066] The SI model assumes that there are only two states for nodes in the network: Susceptible (S) and Infected (I). Once a node is infected (i.e., a fault occurs), it will always remain in the infected state and may infect other nodes directly connected to it. Then, the SI model simulates the fault propagation process of the hydropower generator set under different scenarios, including: Initialize all nodes to the susceptible state, and select one or more initial faulty nodes to be set to the infected state.

[0067] Set infection rate and the propagation time step At each time step, the following steps are performed: Detect the susceptible neighboring nodes j connected to each infected node i; calculate the infection probability of the susceptible neighboring nodes according to the infection rate; where the infection probability calculation expression of the susceptible neighboring node j is:

[0068] In the formula, is the infection probability, is the infection rate, For Node i The connection strength between the node j and the susceptible neighbor node j (i.e., the weight of the edge). When the infection probability is greater than the set threshold, it means that the susceptible neighbor node j is infected. The set threshold in this embodiment is 0.5.

[0069] Update the node status according to the infection probability of the susceptible neighboring nodes, mark the newly infected nodes as infected, and record the number and distribution of infected nodes at the current time step; When all nodes are infected, or the number of infected nodes does not change after multiple time steps, the iteration stops; Draw an infection path diagram based on the current node infection status. The infection path diagram is used to display the propagation path of the fault in the network and identify key propagation paths and nodes. Specifically, draw a curve of the number of infected nodes changing over time based on the model results to analyze the speed and scale of fault propagation. Draw an infection path diagram to display the propagation path of the fault in the network and identify key propagation paths and nodes. And use indicators such as degree centrality and betweenness centrality to identify nodes that play a key role in propagation. Evaluate the scope of influence of nodes in fault propagation and identify important propagation nodes.

[0070] By simulating the propagation of faults in complex networks through the SI model, the propagation path and scope of faults can be effectively predicted and analyzed. This model helps to identify key nodes and weak links in the system, providing strong support for the system's fault prevention and maintenance strategies. By analyzing the results of the propagation model, more accurate fault isolation and protection strategies can be formulated to improve the overall stability and reliability of the system.

[0071] In summary, this embodiment effectively improves the accuracy and efficiency of fault detection by introducing a fault propagation model and a dynamic monitoring algorithm, and can quickly identify and warn potential fault nodes and fault propagation paths. This method for hydroelectric generator status assessment and fault propagation analysis has more accurate status assessment and higher fault detection accuracy, and provides an innovative and reliable solution for the operation monitoring and fault management of hydroelectric generator sets. By real-time monitoring of the operating status of the generator set and analyzing the fault propagation characteristics, the present invention can optimize maintenance strategies, reduce operating costs, improve system stability and reliability, and meet the needs of modern power systems for efficient and safe operation.

[0072] Example 2 like Figure 3 As shown, this embodiment provides a complex network-based hydro-generator fault monitoring system, which is used to implement the above-mentioned complex network-based hydro-generator fault monitoring method, including: The complex network model building module builds a complex network model based on the actual hydro-generator set. The various components of the hydro-generator set are used as nodes of the complex network model. The nodes are connected according to the weight coefficients of the characteristics between the nodes to obtain the adjacency matrix of the complex network model. The characteristics include data correlation and location proximity. A data acquisition module, used to obtain monitoring data according to sensors installed in various components of the actual hydro-generator set, the monitoring data including at least one of temperature, pressure, current, voltage and vibration frequency; The indicator calculation module calculates the complex network indicators based on the monitoring data and the complex network model; The anomaly identification module is used to compare complex network indicators with network indicator reference values, and based on the comparison results, use statistical analysis methods to identify abnormal situations in network indicators.

[0073] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0074] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0075] Example 3 In another embodiment of the present invention, a computing device (i.e., a terminal device) is provided, the terminal device including a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the hydro-generator fault monitoring method based on a complex network, including: A complex network model is constructed based on an actual hydro-turbine generator set, and each component of the hydro-turbine generator set is used as a node of the complex network model. The nodes are connected according to the weight coefficients of the characteristics between the nodes to obtain an adjacency matrix of the complex network model, wherein the characteristics include data correlation and location proximity; Acquiring operating data in real time according to sensors installed in various components of the actual hydro-generator set, the operating data including at least one of temperature, pressure, current, voltage and vibration frequency; Calculate complex network indicators based on operational data and complex network models; Based on the comparison between complex network indicators and network indicator reference values, statistical analysis methods are used to identify abnormal network indicators based on the comparison results. Example 4 In another embodiment of the present invention, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a terminal device for storing programs and data. It is understandable that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and the extended storage medium supported by the terminal device, and can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that more specific examples (non-exhaustive list) of the computer-readable storage medium here include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0076] Computer readable storage media also include data signals propagated in baseband or as part of a carrier wave, which carry readable program codes. Such propagated data signals can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The readable storage medium can also be any readable medium other than a readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program code contained on the readable storage medium can be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0077] Program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0078] The processor may load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the complex network-based hydro-generator fault monitoring method in the above embodiment; the processor may load and execute the following steps: A complex network model is constructed based on an actual hydro-turbine generator set, and each component of the hydro-turbine generator set is used as a node of the complex network model. The nodes are connected according to the weight coefficients of the characteristics between the nodes to obtain an adjacency matrix of the complex network model, wherein the characteristics include data correlation and location proximity; Acquiring operating data in real time according to sensors installed in various components of the actual hydro-generator set, the operating data including at least one of temperature, pressure, current, voltage and vibration frequency; Calculate complex network indicators based on operational data and complex network models; Based on the comparison between complex network indicators and network indicator reference values, statistical analysis methods are used to identify abnormal network indicators based on the comparison results. Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0079] In the embodiments provided by the present invention, it should be understood that the disclosed systems / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0080] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0081] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0082] The prior art part in the protection scope of the present invention is not limited to the embodiments given in the present application documents. All prior art that does not contradict the scheme of the present invention, including but not limited to prior patent documents, prior public publications, prior public uses, etc., can be included in the protection scope of the present invention.

Claims

1. A method for monitoring faults of a hydro-generator based on a complex network, characterized in that: include: A complex network model is constructed based on an actual hydro-generator set, and each component of the hydro-generator set is used as a node of the complex network model. The nodes are connected according to the weight coefficients of the characteristics between the nodes to obtain an adjacency matrix of the complex network model, wherein the characteristics include data correlation and location proximity; Acquiring operating data in real time according to sensors installed in various components of the actual hydro-generator set, wherein the operating data includes at least one of temperature, pressure, current, voltage and vibration frequency; Calculating complex network indicators based on the operating data and the complex network model; Based on the comparison between complex network indicators and network indicator reference values, statistical analysis methods are used to identify abnormal network indicators based on the comparison results.

2. The complex network-based hydro-generator fault monitoring method according to claim 1 is characterized in that: The method of using the components of the hydro-generator set as nodes of the complex network model and connecting the nodes according to the weight coefficients of the features between the nodes to obtain the adjacency matrix of the complex network model specifically includes: The upper guide bearing, lower guide bearing, water guide bearing, thrust bearing and its bracket, stator and its base, top cover and various bearings of the hydro-generator set are used as nodes of the complex network model; The Pearson correlation coefficient method is used to calculate the data correlation of temperature, vibration, pressure and current between each node. The Euclidean distance is used to calculate the physical distance between two nodes in the actual hydro-turbine generator set to determine the location proximity between the nodes. The comprehensive weight coefficient is obtained according to the data correlation, location proximity and the weight coefficients of data correlation and location proximity. The two nodes whose comprehensive weight coefficients exceed the set threshold are connected.

3. The complex network-based hydro-generator fault monitoring method according to claim 2 is characterized in that: Calculating complex network indicators according to the operating data and the complex network model, wherein the complex network indicators include node degree value, network clustering coefficient, and average path length between nodes; The node degree is used to indicate the number of connections between a node and other nodes. The node degree is calculated as follows: in, k i For Node v i The degree value, a ij Represents the element in the i-th row and j-th column of the adjacency matrix of the complex network model. v i With Node v j If there is an edge connection between a ij =1, otherwise, a ij =0; The network clustering coefficient is obtained by averaging the clustering coefficients of all nodes. v i The clustering coefficient calculation expression is: in, C i For Node v i The clustering coefficient, e i Representation and Node v i The number of edges that actually exist between neighbor nodes, k i Is a node v i The degree value of The node average path length is used to calculate the average value of the shortest path length between any two nodes in a complex network; the node average path length is calculated as follows: in, L is the average node path length, N is the total number of nodes in the network, d ij For Node v i To Node v j The shortest path length is v i and v j If the two paths are unreachable, the shortest path length will be set to infinity.

4. The complex network-based hydro-generator fault monitoring method according to claim 3 is characterized in that: The comparison based on the relevant network indicators and the network indicator reference values ​​specifically includes: The network index reference value is obtained according to the baseline network model, and the baseline network model construction process is: obtaining the operating data of the hydro-generator set during normal operation, and preprocessing it, and establishing the baseline network model according to the preprocessed operating data; Calculating baseline network indicators according to the baseline network model, wherein the baseline network indicators include node degree centrality, clustering coefficient, and betweenness centrality; The network index reference value for normal operation of the hydro-generator set is determined according to the baseline network index.

5. The complex network-based hydro-generator fault monitoring method according to claim 4 is characterized in that: The method of comparing the complex network indicator with the network indicator reference value and identifying abnormal network indicator conditions by using a statistical analysis method based on the comparison result specifically includes: Calculate the mean and standard deviation of each complex network indicator based on the baseline network model; The Z-score corresponding to each complex network indicator is calculated using the Z-score statistical analysis model, and the set indicator threshold is used to compare the Z-score corresponding to the complex network indicator of each node. When the Z-score of a node corresponding to a complex network indicator exceeds the set indicator threshold, the abnormal node is regarded as a faulty component.

6. The complex network-based hydro-generator fault monitoring method according to claim 1 or 5, characterized in that: The method of using statistical analysis to identify abnormal network indicators also includes using the SI model to simulate the fault propagation process of the hydro-generator set under different scenarios after detecting abnormal nodes.

7. The complex network-based hydro-generator fault monitoring method according to claim 6 is characterized in that: The SI model is used to simulate the fault propagation process of the hydro-generator set under different scenarios, specifically including: Initialize all nodes to the susceptible state, and select one or more initial fault nodes to be set to the infected state; Set the infection rate and propagation time step; at each time step, detect the susceptible neighboring nodes connected to each infected node; calculate the infection probability of the susceptible neighboring nodes based on the infection rate; Update the node status according to the infection probability of the susceptible neighboring nodes, and record the number and distribution of infected nodes at the current time step; When all nodes are infected, or the number of infected nodes does not change after multiple time steps, the iteration stops; An infection path diagram is drawn according to the current node infection situation, and the infection path diagram is used to display the propagation path of the fault in the network and identify the key propagation paths and nodes.

8. A complex network-based hydro-generator fault monitoring system, used to implement the complex network-based hydro-generator fault monitoring method according to any one of claims 1 to 7, characterized in that: include: A complex network model building module builds a complex network model based on an actual hydro-generator set, takes various components of the hydro-generator set as nodes of the complex network model, connects nodes according to weight coefficients of features between nodes, and obtains an adjacency matrix of the complex network model, wherein the features include data correlation and location proximity; A data acquisition module, used to obtain monitoring data according to sensors installed in various components of the actual hydro-generator set, wherein the monitoring data includes at least one of temperature, pressure, current, voltage and vibration frequency; An indicator calculation module calculates complex network indicators based on the monitoring data and the complex network model; The anomaly identification module is used to compare complex network indicators with network indicator reference values, and based on the comparison results, use statistical analysis methods to identify abnormal situations in network indicators.

9. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include instructions, which, when executed by a computing device, enable the computing device to execute the complex network-based hydro-generator fault monitoring method according to any one of claims 1 to 7.

10. A computing device, characterized in that include: One or more processors, a memory and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include steps for executing the complex network-based hydro-turbine generator fault monitoring method according to any one of claims 1 to 7.

Images