Marine ranching monitoring system fault positioning method, device, equipment and medium

Abstract

The application provides a fault positioning method, device, equipment and medium of a mariculture monitoring system. The method comprises: collecting current operation data of a plurality of monitoring points in the mariculture monitoring system and historical operation data of a plurality of historical time points; obtaining predicted data of a pre-constructed digital twin at the current operation of each monitoring point according to the historical operation data; determining whether there is a monitoring point with operation anomaly according to the predicted data of the current operation and the current operation data; if there is, extracting a feature vector of the current operation data; determining whether there is a fault in each monitoring point according to the feature vector and a fault feature vector library; if there is a fault, obtaining a fault space correlation degree of each monitoring point, the fault space correlation degree is used to quantify the probability of a fault occurring in each monitoring point, so as to determine the fault position of the mariculture monitoring system. Potential faults can be perceived in advance and faults can be positioned, so that system data interruption when a sudden fault occurs can be avoided.

Description

Technical Field

[0001] This application relates to the field of marine ranching monitoring technology, and in particular to a fault location method, device, equipment and medium for a marine ranching monitoring system. Background Technology

[0002] Marine ranching monitoring systems are widely used in unmanned aquaculture areas in deep and open seas. Their core function is to monitor marine environmental parameters (such as meteorology, hydrology, and water quality) and the status of aquaculture facilities in real time by deploying a wind-solar hybrid self-powered system, data transmission and communication systems, structural accessories, and a sensor network. This system needs to operate stably for extended periods in complex and variable marine environments, facing challenges such as high salinity, highly corrosive seawater, and extreme weather events like typhoons. Therefore, it is necessary to predict potential faults in the equipment within the marine ranching monitoring system and pinpoint their locations to ensure the system's continuous and stable operation.

[0003] Currently, existing maintenance models primarily rely on regular manual inspections or reactive repairs after equipment failures occur. However, these methods fail to detect potential faults in advance or pinpoint their location, making them prone to causing system data interruptions during sudden failures. Summary of the Invention

[0004] This application provides a fault location method, apparatus, equipment, and medium for a marine ranch monitoring system, which can achieve the effect of early detection and fault location of potential faults, and avoid system data interruption in the event of a sudden fault.

[0005] In a first aspect, embodiments of this application provide a fault location method for a marine ranching monitoring system, applied to an electronic device, comprising: collecting current operating data of multiple monitoring points in the marine ranching monitoring system, and historical operating data of multiple historical moments; wherein the number of multiple historical moments is determined according to the operating conditions of the marine ranching monitoring system, and the current operating data includes the communication data packet loss rate; obtaining, based on the historical operating data, predictive data of the current operation of a pre-constructed digital twin at each monitoring point; wherein the pre-constructed digital twin is a pre-constructed digital model corresponding to the marine ranching monitoring system; determining, based on the predicted data and the current operating data, whether there are any monitoring points with abnormal operation; if an abnormal operation is determined to exist... For each monitoring point, multiple time-domain and frequency-domain features of the current operating data are extracted. Based on these features, a feature vector is obtained. Using the feature vector and a fault feature vector library, it is determined whether a fault exists at each monitoring point. If a fault is identified, the operating deviation of each monitoring point is obtained based on the predicted and current operating data. The communication link reliability of each monitoring point is obtained based on the packet loss rate. The fault spatial correlation of each monitoring point is obtained based on the operating deviation and communication link reliability. This fault spatial correlation is used to quantify the probability of a fault occurring at each monitoring point. Based on the fault spatial correlation of each monitoring point, the fault location of the marine ranch monitoring system is determined.

[0006] In one possible implementation, the formula for obtaining the predicted data of the current operation of the pre-constructed digital twin at each monitoring point based on historical operational data is:

[0007]

[0008]

[0009] in, This represents the predicted data of the pre-constructed digital twin at time t at each monitoring point, where n represents the number of historical times, and k represents the k-th historical time. This represents the weighting coefficients of historical data at k historical moments. This represents the historical running data at historical time tk. This represents the error correction factor. Represents the prediction error at historical time t-1; where, This represents the historical running data at historical moment t-1. This represents the current running prediction data at historical time t-1.

[0010] In one possible implementation, determining whether there is a monitoring point with operational anomalies based on the predicted data and the current operating data includes: obtaining the absolute difference between the predicted data and the current operating data; if the absolute difference is greater than a preset absolute difference threshold, then it is determined that there is a monitoring point with operational anomalies.

[0011] In one possible implementation, the fault feature vector library stores fault feature vectors for various faults; accordingly, based on the feature vectors and the fault feature vector library, it is determined whether each monitoring point has a fault, including: obtaining the similarity between the feature vector and each fault feature vector in the fault feature vector library; if the similarity is greater than or equal to a preset similarity threshold, it is determined that the current monitoring point has a fault.

[0012] In one possible implementation, the formula for obtaining the fault spatial correlation degree of each monitoring point based on the operational deviation degree and communication link reliability is as follows:

[0013]

[0014] in, This represents the spatial correlation of the fault at the i-th monitoring point. The weighting coefficients representing the degree of operational deviation. Indicates the degree of operational deviation; Weighting coefficients representing the reliability of communication links. Indicates the reliability of the communication link.

[0015] In one possible implementation, the fault location of the marine ranch monitoring system is determined based on the fault spatial correlation of each monitoring point, including: determining whether there is a consecutive preset number of monitoring points whose fault spatial correlation all exceed a preset fault spatial correlation threshold; if it is determined to exist, then the consecutive preset number of monitoring points are determined as the fault location of the marine ranch monitoring system; if it is determined not to exist, then the monitoring point with the highest fault spatial correlation is determined as the fault location of the marine ranch monitoring system.

[0016] Secondly, embodiments of this application provide a fault location device for a marine ranching monitoring system, applied to an electronic device, comprising: a data acquisition unit, used to acquire current operating data of multiple monitoring points in the marine ranching monitoring system, and historical operating data of multiple historical moments; wherein the number of historical moments is determined according to the operating conditions of the marine ranching monitoring system, and the current operating data includes the communication data packet loss rate; a first acquisition unit, used to acquire, based on the historical operating data, the predicted operating data of a pre-constructed digital twin at each monitoring point; wherein the pre-constructed digital twin is a pre-constructed digital model corresponding to the marine ranching monitoring system; a first judgment unit, used to determine, based on the predicted operating data and the current operating data, whether there are any monitoring points with abnormal operation; and a second acquisition unit, used to, if it is determined that there are monitoring points with abnormal operation, ... The system extracts multiple time-domain and frequency-domain features from the current operating data; and obtains the feature vector of the current operating data based on these features. A second judgment unit determines whether a fault exists at each monitoring point based on the feature vector and a fault feature vector library. A third acquisition unit, if a fault is determined to exist, obtains the operating deviation of each monitoring point based on the current predicted data and the current operating data. A fourth acquisition unit obtains the communication link reliability of each monitoring point based on the communication data packet loss rate. A fifth acquisition unit obtains the fault spatial correlation of each monitoring point based on the operating deviation and communication link reliability; the fault spatial correlation is used to quantify the probability of a fault occurring at each monitoring point. A determination unit determines the fault location of the marine ranch monitoring system based on the fault spatial correlation of each monitoring point.

[0017] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0018] The memory stores the instructions that the computer executes;

[0019] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0020] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0021] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0022] The fault location method, apparatus, equipment, and medium for a marine ranching monitoring system provided in this application collect current operating data and historical operating data from multiple historical moments at each monitoring point. The number of historical moments is determined based on the operating conditions of the marine ranching monitoring system. Operating conditions reflect the external environment and the system's own operating status; the variation patterns of operating data differ significantly under different operating conditions. Determining the number of historical moments based on operating conditions ensures accurate predictions from the digital twin. Relying on a pre-constructed digital twin, predicted current operating data for each monitoring point is obtained based on historical operating data. By comparing the predicted data with the current operating data, monitoring points with operational anomalies can be identified in advance. Multiple time-domain and frequency-domain features of the current operating data are extracted to comprehensively characterize the data's anomaly information from two dimensions: time variation patterns and frequency distribution characteristics, thereby obtaining feature vectors. Fault judgment is completed by combining the fault feature vector library, achieving proactive identification and early perception of potential faults at monitoring points. After determining that a monitoring point has a fault, the operational deviation of each monitoring point is obtained based on the predicted and actual operational data. The communication link reliability of each monitoring point is obtained based on the packet loss rate. Finally, the fault spatial correlation of each monitoring point is obtained based on the operational deviation and communication link reliability. The fault spatial correlation quantifies the probability of a fault occurring at each monitoring point, accurately determining the fault location in the marine ranch monitoring system and quickly pinpointing the specific monitoring point where the fault occurred. By proactively detecting potential faults and locating them, system data interruption can be avoided in the event of a sudden fault. Attached Figure Description

[0023] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0024] Figure 1 A schematic diagram illustrating a scenario for the fault location method of the marine ranching monitoring system provided in this application embodiment;

[0025] Figure 2 A flowchart illustrating the fault location method for the marine ranching monitoring system provided in this application embodiment;

[0026] Figure 3 A schematic diagram of the structure of the fault location device for the marine ranch monitoring system provided in this application embodiment;

[0027] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0028] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0029] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0030] Figure 1 This is a schematic diagram of a scenario for the fault location method of the marine ranching monitoring system provided in the embodiments of this application, such as... Figure 1 As shown, it includes: a receiving device 101, a processing device 102, and a display device 103.

[0031] It is understood that the structure illustrated in the embodiments of this application does not constitute a specific limitation on the fault location method of the marine ranch monitoring system. In other feasible embodiments of this application, the above architecture may include more or fewer components than illustrated, or combine some components, or split some components, or arrange different components, which can be determined according to the actual application scenario and is not limited here. Figure 1 The components shown can be implemented in hardware, software, or a combination of both.

[0032] In the specific implementation process, the receiving device 101 can be an input / output interface or a communication interface, which can collect the current operating data of multiple monitoring points in the marine ranch monitoring system, as well as the historical operating data of multiple historical moments.

[0033] The processing device 102 can acquire predicted data of the current operation of a pre-constructed digital twin at each monitoring point based on historical operating data; determine whether there are any monitoring points with operational anomalies based on the predicted data and the current operating data; if anomalies are found, extract multiple time-domain features and multiple frequency-domain features of the current operating data; and acquire feature vectors of the current operating data based on the multiple time-domain features and multiple frequency-domain features; determine whether there are any faults at each monitoring point based on the feature vectors and a fault feature vector library; if faults are found, acquire the operating deviation of each monitoring point based on the predicted data and the current operating data; acquire the communication link reliability of each monitoring point based on the communication data packet loss rate; acquire the fault spatial correlation of each monitoring point based on the operating deviation and communication link reliability; wherein the fault spatial correlation is used to quantify the probability of faults occurring at each monitoring point; and determine the fault location of the marine ranch monitoring system based on the fault spatial correlation of each monitoring point.

[0034] The display device 103 can be used to display the location of the fault.

[0035] It should be understood that the aforementioned processor can be implemented by reading instructions from memory and executing those instructions, or it can be implemented through chip circuitry.

[0036] Furthermore, the network architecture and business scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0037] To address the aforementioned technical problems, this application proposes the following technical concept: Considering that relying primarily on manual periodic inspections or reactive repairs after equipment failures occur makes it impossible to proactively detect and locate potential faults, easily leading to system data interruptions during sudden failures, the inventors devised a pre-constructed digital model corresponding to the marine ranching monitoring system. Combined with historical operational data, this model obtains predicted data on the current operation of the pre-constructed digital twin at each monitoring point. The number of historical time points is determined based on the operating conditions of the marine ranching monitoring system. These operating conditions reflect the external environment and the system's own operational status, with significant differences in the patterns of data change under different conditions. Based on the predicted data and the actual current operational data, it is determined whether any monitoring points exhibit abnormal operation. When an abnormal monitoring point is identified, multiple time-domain and frequency-domain features of the current operational data are extracted. Based on these features, a feature vector is obtained and combined with a fault feature vector library to determine whether each monitoring point has a fault. This allows for the early detection of potential faults. Subsequently, based on the predicted data and the current operating data, the operating deviation of each monitoring point is obtained; based on the packet loss rate of the communication data, the communication link reliability of each monitoring point is obtained; based on the operating deviation and the communication link reliability, the fault spatial correlation of each monitoring point is obtained; the fault spatial correlation is used to quantify the probability of each monitoring point experiencing a fault; based on the fault spatial correlation of each monitoring point, the fault location of the marine ranch monitoring system is determined.

[0038] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0039] Figure 2 This is a flowchart illustrating the fault location method for the marine ranching monitoring system provided in this application embodiment, as shown below. Figure 2 As shown, the method includes:

[0040] S201: Collect current operating data from multiple monitoring points in the marine ranch monitoring system, as well as historical operating data from multiple historical moments; the number of historical moments is determined based on the operating conditions of the marine ranch monitoring system, and the current operating data includes the communication data packet loss rate.

[0041] Among them, the operating conditions reflect the external environment and the operating status of the marine ranch monitoring system, and the variation patterns of the operating data under different operating conditions are significantly different.

[0042] In this embodiment, the monitoring point range covers all key monitoring points in the entire chain of the marine ranch monitoring system, including power supply nodes of the wind-solar hybrid self-powered system, segment monitoring points of data transmission cables, sensor interface nodes, shore base station communication nodes, etc.

[0043] In this embodiment, the current operating data are all quantitative physical parameters that can characterize the operating status of the monitoring point, including voltage, current, temperature, and impedance.

[0044] In this embodiment, the frequency can be dynamically adjusted according to the operating conditions of the marine ranch monitoring system to collect real-time data from each monitoring point. After collection, the data is transmitted through an edge gateway and a hybrid transmission link (industrial Ethernet / wireless short-range communication) to ensure the real-time performance and stability of the data transmission.

[0045] In this embodiment, the operation data of each monitoring point at multiple past moments are retrieved from the historical data archive of the marine ranch monitoring system. This historical data archive stores the operation data of the monitoring point throughout its entire life cycle.

[0046] It is important to ensure that the current data collection time of all monitoring points is synchronized to avoid errors in subsequent data analysis due to time differences.

[0047] In this embodiment, data cleaning, standardization, and fusion are performed on the current running data and historical running data; data cleaning adopts... The principles and trends are used to identify and process outlier data, and random noise data is supplemented using linear interpolation; data standardization is achieved using the Z-score standardization formula. This converts data with different dimensions into dimensionless data, where X represents the original operational data. The average of the running data, To determine the standard deviation of the running data, the data fusion uses the Kalman filter algorithm to fuse sensor data (voltage sensor, current sensor, and temperature sensor, etc.) with the same physical parameter, thereby reducing measurement errors.

[0048] in, The principle, also known as the three-standard-deviation principle, is a method for identifying and judging abnormal data based on the normal distribution characteristics of data. It is used to quickly and objectively identify abnormal data deviating from the normal range from multi-source operational data of marine ranching monitoring systems. This principle is based on the premise that normally collected operational data (such as voltage, current, and temperature) will follow a normal distribution (when a marine ranching system is operating stably, the values ​​of each monitoring parameter will fluctuate slightly around the mean, conforming to the normal distribution law). For operational data that follows a normal distribution, the probability of the data falling within the range of the mean μ ± 3 standard deviations is approximately 99.73%. Data falling outside this range are considered low-probability outliers and are thus judged as abnormal data.

[0049] S202: Based on historical operational data, obtain the predicted data of the current operation of the pre-constructed digital twin at each monitoring point; wherein the pre-constructed digital twin is a pre-constructed digital model corresponding to the marine ranch monitoring system.

[0050] The pre-constructed digital twin is modeled using 3D modeling technology, based on the actual size and structure of the marine ranch monitoring system. For the wind-solar hybrid system: the system layout, power supply module locations, and communication interface distribution are recreated; for Ethernet: the laying path (e.g., underwater direction, burial depth), cross-sectional structure (copper core, insulation layer, protective layer), etc., are recreated; for the power board: the physical layout of the internal power modules, communication modules, and sensor interfaces, etc., is recreated; for sensor interfaces: the number of pins, connection methods, and packaging structure, etc., are recreated.

[0051] In this embodiment, a physical parameter calibration algorithm is used to map the inherent properties of each component of the marine ranch monitoring system to a digital twin. Electrical properties include: the power generation coefficient of the wind-solar hybrid system, the conversion efficiency of the power module of the wind-solar hybrid system, and the resistance / capacitance parameters of the cables; thermal properties include: the thermal conductivity of the solar photovoltaic panel casing and the heat dissipation coefficient of the cable insulation layer; mechanical properties include: the tensile strength of the cables, the protection level parameters of the data acquisition unit, and the wind resistance level parameters of the wind power generation unit.

[0052] In this embodiment, a real-time push and periodic calibration mechanism is adopted to ensure the consistency between the digital twin and the marine ranch monitoring system. Real-time data from the marine ranch monitoring system is collected and pushed to the digital twin to update the twin's status. Every hour, the digital twin is precisely corrected using manual calibration data from key nodes of the marine ranch monitoring system (such as voltage calibration of the wind-solar hybrid system, cable impedance calibration, and battery power supply / charging voltage calibration) to eliminate accumulated errors.

[0053] Specifically, based on historical operational data, the formula for obtaining the predicted operational data of the pre-constructed digital twin at each monitoring point is as follows:

[0054]

[0055]

[0056] in, This represents the predicted data of the pre-constructed digital twin at time t at each monitoring point, where n represents the number of historical times, and k represents the k-th historical time. This represents the weighting coefficients of historical data at k historical moments. This represents the historical running data at historical time tk. This represents the error correction factor. This represents the prediction error at historical time t-1.

[0057] in, This represents the historical running data at historical moment t-1. This represents the current running prediction data at historical time t-1.

[0058] In this embodiment, the current prediction time of the currently running prediction data is t, and historical running data of n historical times (t-1, t-2...tn) are retrieved. The number of n is determined by the operating complexity of the marine ranch monitoring system.

[0059] The weight coefficients of historical data at different historical moments represent the degree of influence of different historical moments on historical data, and are obtained through training using the gradient descent algorithm.

[0060] The error correction coefficient ranges from [0,1] and is used to correct the impact of the previous prediction error on the current prediction, reducing the cumulative error. For monitoring points with more complex operating conditions, the error correction coefficient is closer to 1. Examples include offshore / deep-sea submarine cable monitoring points and power modules of wind-solar hybrid self-powered systems, which are greatly affected by marine environment and weather changes, resulting in significant fluctuations in operating parameters and a stronger cumulative effect of historical prediction errors, thus requiring a higher error correction coefficient. Conversely, for monitoring points with more stable operating conditions, the error correction coefficient is closer to 0. Examples include communication nodes at shore base stations and fixed sensor interfaces, where the operating environment is stable, parameter fluctuations are small, and the cumulative effect of historical prediction errors is weak, allowing for a lower error correction coefficient.

[0061] S203: Based on the current predicted data and the current running data, determine whether there are any monitoring points with abnormal operation.

[0062] Specifically, the absolute difference between the predicted data and the current running data is obtained; if the absolute difference is greater than the preset absolute difference threshold, it is determined that there is a monitoring point with abnormal operation.

[0063] In this embodiment, the current running data and the predicted data for the current running are matched one-to-one at each monitoring point to ensure consistency of the comparison object; for each monitoring point, the absolute difference between the current running data and the predicted data for the current running is calculated: .in, t represents the absolute difference of the data. This represents the current operating data of the i-th monitoring point. This represents the current operational prediction data for the i-th monitoring point.

[0064] In this embodiment, a threshold for the absolute difference of data is preset (dynamically set according to equipment type and operating conditions); if any monitoring point exists... If t > preset absolute difference threshold, the equipment is deemed to be malfunctioning; if If t ≤ preset absolute difference threshold, the monitoring point is considered to be operating normally.

[0065] S204: If it is determined that there is a monitoring point with abnormal operation, extract multiple time-domain features and multiple frequency-domain features of the current operating data; and obtain the feature vector of the current operating data based on the multiple time-domain features and multiple frequency-domain features.

[0066] Specifically, multiple time-domain features and multiple frequency-domain features are extracted from the current running data using wavelet transform algorithms.

[0067] Among them, wavelet transform algorithm is a mathematical tool used to extract the time domain features and frequency domain features of a signal.

[0068] In this embodiment, time-domain features reflect the overall trend and fluctuation of operational data, including mean, variance, peak value, and peak-to-peak value. Among them, peak-to-peak value (PP) is the core indicator in time-domain features used to quantify the fluctuation range and dispersion of operational data. It intuitively reflects the difference between the maximum and minimum values ​​of operational data within a certain period and is an important basis for judging whether the operational status of the monitoring point is stable.

[0069] In this embodiment, the frequency domain characteristics reflect the frequency properties of the operating data, including the dominant frequency, harmonic content, and spectral energy distribution.

[0070] In this embodiment, the extracted time-domain features and multiple frequency-domain features are combined in a fixed order to form a feature vector of dimension m, where m is the total number of extracted features, such as 4 time-domain features + 3 frequency-domain features, m=7.

[0071] S205: Determine whether there is a fault at each monitoring point based on the feature vector and the fault feature vector library.

[0072] The fault feature vector library stores fault feature vectors for various faults, covering all fault types across the entire chain. Each fault type corresponds to a fault feature vector, and the dimension is consistent with the feature vector of the current running data.

[0073] In this embodiment, the failure of the wind-solar hybrid power generation system and the performance degradation of the power module are characterized by increased output voltage fluctuation and decreased wind-solar hybrid power generation conversion efficiency.

[0074] In this embodiment, the communication module lag fault is characterized by increased data transmission latency and increased communication data packet loss rate.

[0075] In this embodiment, the sensor malfunction is characterized by decreased insulation performance, such as reduced insulation impedance and increased leakage current; and data drift, characterized by parameter output values ​​deviating from the mean range.

[0076] In this embodiment, the aging of the copper core is characterized by increased transmission resistance and increased current loss.

[0077] In this embodiment, the absence of response is characterized by continuously outputting a fixed value or no data.

[0078] Specifically, the similarity between the feature vector and each fault feature vector in the fault feature vector library is obtained; if the similarity is greater than or equal to the preset similarity threshold, it is determined that there is a fault at the current monitoring point.

[0079] In this embodiment, feature vectors of various faults are retrieved from the fault feature vector library. Based on the feature vectors of the current operating data and each fault feature vector, cosine similarity is used to calculate the similarity. The similarity value ranges from [0,1], with a higher degree of matching between the current operating data and the fault features indicating a closer similarity to 1.

[0080] In this embodiment, a similarity threshold, such as 0.8, is preset in advance and can be dynamically adjusted according to the actual situation.

[0081] In this embodiment, the formula for obtaining the similarity is:

[0082]

[0083] in, The similarity between the feature vector and each fault feature vector in the fault feature vector library is represented by A, B, and m. This represents the feature component of the i-th dimension of the feature vector. This represents the fault feature component of the i-th dimension of each fault feature vector.

[0084] In this embodiment, if If the similarity is greater than or equal to a preset similarity threshold, then a fault of the corresponding fault type is determined to exist. Optionally, a fault prediction result can be output, which includes the fault type and the predicted occurrence time window; if Sim(A,B) < threshold, then no potential fault is determined.

[0085] S206: If a fault is determined to exist, the operating deviation of each monitoring point is obtained based on the predicted data and the current operating data.

[0086] In this embodiment, the operational deviation of each monitoring point characterizes the degree of abnormality in the operational data of each monitoring point.

[0087]

[0088] in: Let be the operational deviation of the i-th monitoring point. The larger the value, the more severely the operational data of that monitoring point deviates from the normal state. This represents the current operating data for the i-th monitoring point. This represents the current operational prediction data for the i-th monitoring point.

[0089] S207: Obtain the reliability of the communication link at each monitoring point based on the packet loss rate of the communication data.

[0090] In this embodiment, by monitoring the communication link status of each monitoring point, calculating the communication data packet loss rate by counting the number of data packets lost per unit time, and then obtaining the communication link reliability, a communication link reliability of approximately 1 indicates that the more reliable the communication link is, the higher the credibility of the current running data.

[0091] Specifically, the formula for communication link reliability is:

[0092]

[0093] in, Indicates the reliability of the communication link. This indicates the packet loss rate of communication data.

[0094] S208: Obtain the spatial correlation of faults at each monitoring point based on the operational deviation and communication link reliability.

[0095] In this embodiment, the formula for obtaining the fault spatial correlation degree of each monitoring point based on the operational deviation degree and communication link reliability is as follows:

[0096]

[0097] in, This represents the spatial correlation of the fault at the i-th monitoring point. The weighting coefficients representing the degree of operational deviation. Indicates the degree of operational deviation; Weighting coefficients representing the reliability of communication links. Indicates the reliability of the communication link.

[0098] In this embodiment, the range of the fault spatial correlation degree is [0,1]. The larger the value, the higher the probability of a fault occurring at the monitoring point.

[0099] The sum of the weighting coefficients for operational deviation and communication link reliability is 1. This can be adjusted based on the monitoring point type; for example, the weighting coefficient for operational deviation at shore base stations is 0.6, and the weighting coefficient for operational deviation at submarine cable stations is 0.5.

[0100] S209: Determine the fault location of the marine ranch monitoring system based on the spatial correlation of faults at each monitoring point.

[0101] Specifically, it is determined whether there exists a continuous preset number of monitoring points whose fault spatial correlation exceeds a preset fault spatial correlation threshold. If it is determined to exist, the continuous preset number of monitoring points are identified as the fault location of the marine ranch monitoring system. If it is determined not to exist, the monitoring point with the highest fault spatial correlation is identified as the fault location of the marine ranch monitoring system.

[0102] The preset quantity is 3, which can be adjusted according to the actual situation.

[0103] In this embodiment, if the spatial correlation of multiple consecutive monitoring points exceeds the preset spatial correlation threshold, such as three consecutive monitoring points of a power supply cable, it is determined to be a regional fault, such as the area 2.5km away from the power supply cable.

[0104] In this embodiment, if there are no multiple consecutive monitoring points whose fault spatial correlation exceeds the preset fault spatial correlation threshold, the fault spatial correlation of each monitoring point is sorted from high to low; if the fault spatial correlation of a single monitoring point is the maximum value, the monitoring point is directly determined as the fault location.

[0105] In this embodiment, the specific information of the final output fault location includes the equipment or component to which the monitoring point belongs, such as power module A and sensor interface 3, physical coordinates or location, such as the submarine cable 1km east of the offshore base station and the 5th group of photovoltaic panel array.

[0106] In summary, the current operational data and historical operational data from multiple historical moments are collected from each monitoring point. The number of historical moments is determined based on the operating conditions of the marine ranching monitoring system. These operating conditions reflect the external environment and the system's own operational status, and the patterns of change in operational data differ significantly under different operating conditions. Determining the number of historical moments based on these operating conditions ensures the accuracy of the digital twin predictions. Using a pre-constructed digital twin, predicted current operational data for each monitoring point is obtained from historical operational data. By comparing the predicted data with the current operational data, monitoring points with operational anomalies can be identified in advance. Multiple time-domain and frequency-domain features of the current operational data are extracted to comprehensively characterize the data's anomaly information from two dimensions: temporal variation patterns and frequency distribution characteristics, thus obtaining feature vectors. Combining these with a fault feature vector library completes fault judgment, enabling proactive identification and early detection of potential faults at monitoring points. After determining that a monitoring point has a fault, the operational deviation of each monitoring point is obtained based on the predicted and actual operational data. The communication link reliability of each monitoring point is obtained based on the packet loss rate. Finally, the fault spatial correlation of each monitoring point is obtained based on the operational deviation and communication link reliability. The fault spatial correlation quantifies the probability of a fault occurring at each monitoring point, accurately determining the fault location in the marine ranch monitoring system and quickly pinpointing the specific monitoring point where the fault occurred. By proactively detecting potential faults and locating them, system data interruption can be avoided in the event of a sudden fault.

[0107] Optionally, the fault level can be determined by combining the fault type, scope of impact, and urgency, including minor faults, moderate faults, and severe faults. Minor faults include data drift of a single sensor and failure of an indicator light on the power board, which do not affect the overall operation of the marine ranching monitoring system; moderate faults include performance degradation of a module on the power board, unstable system power supply, and loss of video monitoring transmission data packets, which may affect local functions, such as unstable power supply to a few sensors; severe faults include power supply and communication link failures and power supply failures of the wind-solar hybrid self-powered system, which will lead to the complete shutdown or data interruption of the marine ranching monitoring system.

[0108] Optionally, for minor faults, a regular maintenance plan can be generated to handle them during the next system inspection, avoiding frequent downtime; for moderate faults, a priority maintenance plan can be generated, such as scheduling maintenance within 24 hours, to prevent the fault from worsening; for severe faults, an emergency maintenance plan can be generated, such as immediate shutdown and scheduling of emergency maintenance resources to reduce losses. Maintenance procedure suggestions are also output, such as system modules that need to be shut down before maintenance and security protection measures.

[0109] Optionally, two types of control commands are generated based on maintenance decisions: marine ranching monitoring system operation control commands and maintenance assistance commands. The marine ranching monitoring system operation control commands reduce the load on non-critical equipment in the event of a minor fault and trigger an emergency system shutdown in the event of a serious fault. The maintenance assistance commands control the wind-solar hybrid power generation system to switch to backup power and adjust sensors to detection mode before maintenance.

[0110] In this embodiment, the execution results of control commands from the physical system (such as successful backup power switching and system shutdown) are received, and the feedback results are transmitted to the digital twin to update the execution status of the twin. Simultaneously, feedback data is recorded for subsequent maintenance decision optimization, such as adjusting fault level judgment criteria. All data from this fault prediction, location, and maintenance are archived for optimization and fault feature library vector updates.

[0111] Optionally, a maintenance resource library is pre-stored, and resources are matched according to the fault level and location. Maintenance personnel are divided into basic, intermediate, and advanced skill levels. Basic personnel handle minor faults, intermediate personnel handle medium faults, and advanced personnel handle severe faults. Maintenance tools include underwater robots (for cable / underwater equipment maintenance) and specialized testing instruments (such as impedance testers and voltage calibrators). Spare parts include solar photovoltaic panels, wind power generation units, wind-solar hybrid controllers, power modules, cable connectors, and sensor interface components.

[0112] In this embodiment, a fault location system for a marine ranching monitoring system is provided. The fault location system for the marine ranching monitoring system comprises: a multi-source data acquisition module, a system-level digital twin construction module, a multi-source data fusion and preprocessing module, an artificial intelligence fault prediction module, a precise fault location module, a predictive maintenance decision support module, and a twin and physical system collaborative control module.

[0113] In this embodiment, the multi-source data acquisition module is used to collect various types of working status data of the marine ranch monitoring system in real time and transmit the collected data to subsequent modules. The marine ranch monitoring system includes a wind-solar hybrid self-powered system, a data transmission and communication system, structural accessories, and various sensor interfaces.

[0114] The multi-source data acquisition module specifically includes a parameter sensing unit, a data transmission unit, and an acquisition control unit. The parameter sensing unit consists of voltage sensors, current sensors, temperature sensors, and an impedance analyzer, which respectively collect the voltage of the wind-solar hybrid power generation system, cable / sensor current, cable temperature, and cable / sensor interface impedance. The data transmission unit adopts an edge gateway and hybrid transmission link architecture. Data is transmitted internally at the shore base station via industrial Ethernet, and data is transmitted via Ethernet communication links. Data is transmitted between the sensors and the acquisition unit via wireless short-range communication or Ethernet. The acquisition control unit has a built-in synchronization control chip to uniformly schedule the acquisition frequency of each sensing device and supports dynamic adjustment of the acquisition frequency according to the operating conditions of the marine ranch monitoring system.

[0115] In this embodiment, the system-level digital twin construction module is used to construct a high-fidelity digital twin that corresponds to the marine ranch monitoring system at a 1:1 scale, thereby realizing physical attribute mapping and temporal state prediction.

[0116] In this embodiment, during implementation, the marine ranching monitoring system is activated, and the multi-source data acquisition module initializes the sensors and detection equipment. The system-level digital twin construction module loads the marine ranching monitoring system drawings, constructs a 1:1 geometric model, imports physical parameters to complete attribute mapping, trains a time-series prediction model using historical data, determines the formula parameters, completes the digital twin initialization, and the marine ranching monitoring system enters a standby state. The time-series prediction model is used to predict the current operational data of the pre-constructed digital twin at each monitoring point.

[0117] In this embodiment, the multi-source data fusion and preprocessing module is used to clean, standardize and fuse the collected multi-source data to eliminate data noise and redundancy and improve data quality.

[0118] The multi-source data fusion and preprocessing module includes a data cleaning unit, a data standardization unit, and a data fusion unit; the data cleaning unit adopts... The principle and trend judgment identifies and processes abnormal data, and completes accidental noise data through linear interpolation; the data standardization unit uses the Z-score standardization formula to convert data with different dimensions into dimensionless data; the data fusion unit uses the Kalman filter algorithm to fuse data collected by multiple sensors for the same physical parameter, thereby reducing measurement errors.

[0119] In this embodiment, the artificial intelligence fault prediction module is used to extract fault features from the processed data and compare them with the fault feature vector library to achieve early prediction of potential faults.

[0120] In this embodiment, the fault precise location module is used to calculate the fault spatial correlation degree of each monitoring point through data deviation analysis and communication link reliability assessment when a fault is detected, so as to achieve precise fault location.

[0121] In this embodiment, the predictive maintenance decision support module is used to assess the fault level and match maintenance resources based on the fault prediction results and location information, and generate predictive maintenance decisions.

[0122] In this embodiment, the digital twin and physical system collaborative control module is used to synchronize the status of the digital twin and the marine ranch monitoring system, generate control commands based on maintenance decisions, and provide feedback on the execution results.

[0123] Figure 3 This is a schematic diagram of the fault location device for the marine ranch monitoring system provided in an embodiment of this application. Figure 3 As shown, the fault location device for the marine ranch monitoring system provided in this embodiment includes: a data acquisition unit 301, a first acquisition unit 302, a first judgment unit 303, a second acquisition unit 304, a second judgment unit 305, a third acquisition unit 306, a fourth acquisition unit 307, a fifth acquisition unit 308, and a determination unit 309.

[0124] The acquisition unit 301 is used to acquire current operating data from multiple monitoring points in the marine ranch monitoring system, as well as historical operating data from multiple historical moments. The number of historical moments is determined based on the operating conditions of the marine ranch monitoring system, and the current operating data includes the communication data packet loss rate.

[0125] The first acquisition unit 302 is used to acquire the predicted data of the current operation of the pre-constructed digital twin at each monitoring point based on historical operation data; wherein the pre-constructed digital twin is a pre-constructed digital model corresponding to the marine ranch monitoring system.

[0126] The first judgment unit 303 is used to determine whether there are any monitoring points with abnormal operation based on the predicted data and the current running data.

[0127] The second acquisition unit 304 is used to extract multiple time-domain features and multiple frequency-domain features of the current operating data if it is determined that there is a monitoring point with abnormal operation; and to obtain the feature vector of the current operating data based on the multiple time-domain features and multiple frequency-domain features.

[0128] The second judgment unit 305 is used to determine whether there is a fault at each monitoring point based on the feature vector and the fault feature vector library.

[0129] The third acquisition unit 306 is used to acquire the operating deviation of each monitoring point based on the predicted data and the current operating data if a fault is determined to exist.

[0130] The fourth acquisition unit 307 is used to acquire the communication link reliability of each monitoring point based on the communication data packet loss rate.

[0131] The fifth acquisition unit 308 is used to acquire the fault spatial correlation degree of each monitoring point based on the operating deviation degree and the communication link reliability; wherein the fault spatial correlation degree is used to quantify the probability of each monitoring point experiencing a fault.

[0132] The determination unit 309 is used to determine the fault location of the marine ranch monitoring system based on the spatial correlation of faults at each monitoring point.

[0133] In one possible implementation, the formula for the first acquisition unit 302 is:

[0134]

[0135]

[0136] in, This represents the predicted data of the pre-constructed digital twin at time t at each monitoring point, where n represents the number of historical times, and k represents the k-th historical time. This represents the weighting coefficients of historical data at k historical moments. This represents the historical running data at historical time tk. This represents the error correction factor. Represents the prediction error at historical time t-1; where, This represents the historical running data at historical moment t-1. This represents the current running prediction data at historical time t-1.

[0137] In one possible implementation, the first judgment unit 303 is specifically used to: obtain the absolute difference between the currently running prediction data and the currently running data; if the absolute difference is greater than a preset absolute difference threshold, then it is determined that there is a monitoring point with abnormal operation.

[0138] In one possible implementation, the fault feature vector library stores fault feature vectors of various faults; correspondingly, the second judgment unit 305 is specifically used to: obtain the similarity between the feature vector and each fault feature vector in the fault feature vector library; if the similarity is greater than or equal to a preset similarity threshold, then it is determined that there is a fault at the current monitoring point.

[0139] In one possible implementation, the formula for the fifth acquisition unit 308 is:

[0140]

[0141] in, This represents the spatial correlation of the fault at the i-th monitoring point. The weighting coefficients representing the degree of operational deviation. Indicates the degree of operational deviation; Weighting coefficients representing the reliability of communication links. Indicates the reliability of the communication link.

[0142] In one possible implementation, the determining unit 309 is specifically used to: determine whether there is a continuous preset number of monitoring points whose fault spatial correlation exceeds a preset fault spatial correlation threshold; if it is determined that there is, then the continuous preset number of monitoring points are determined as the fault location of the marine ranch monitoring system; if it is determined that there is no, then the monitoring point with the largest fault spatial correlation is determined as the fault location of the marine ranch monitoring system.

[0143] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 4 As shown, the electronic device provided in this embodiment includes at least one processor 401 and a memory 402. Optionally, the electronic device further includes a communication component 403. The processor 401, memory 402, and communication component 403 are connected via a bus.

[0144] In a specific implementation, at least one processor 401 executes computer execution instructions stored in memory 402, causing at least one processor 401 to perform the above-described method.

[0145] The specific implementation process of processor 401 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0146] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0147] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0148] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0149] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0150] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0151] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0152] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0153] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0154] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0155] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0156] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0157] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0158] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A fault location method for a marine ranching monitoring system, characterized in that, Applied to electronic devices, including: The system collects current operating data from multiple monitoring points in the marine ranching monitoring system, as well as historical operating data from multiple historical moments. The number of historical moments is determined based on the operating conditions of the marine ranching monitoring system, and the current operating data includes the communication data packet loss rate. Based on the historical operational data, predictive data of the current operation of the pre-constructed digital twin at each monitoring point is obtained; wherein the pre-constructed digital twin is a pre-constructed digital model corresponding to the marine ranch monitoring system; Based on the predicted data and the current operating data, determine whether there are any monitoring points indicating abnormal operation. If a monitoring point is determined to be operating abnormally, then multiple time-domain features and multiple frequency-domain features of the current operating data are extracted; and based on the multiple time-domain features and the multiple frequency-domain features, the feature vector of the current operating data is obtained. Based on the feature vector and the fault feature vector library, it is determined whether each monitoring point has a fault; If a fault is determined, the operating deviation of each monitoring point is obtained based on the predicted data and the current operating data. Based on the packet loss rate of the communication data, the reliability of the communication link at each monitoring point is obtained; Based on the operational deviation and the communication link reliability, the fault spatial correlation degree of each monitoring point is obtained; wherein the fault spatial correlation degree is used to quantify the probability of each monitoring point experiencing a fault. The location of the fault in the marine ranch monitoring system is determined based on the spatial correlation of the faults at each monitoring point.

2. The method according to claim 1, characterized in that, The formula for obtaining the predicted data of the current operation of the pre-constructed digital twin at each monitoring point based on the historical operation data is as follows: in, This represents the predicted data of the pre-constructed digital twin at time t at each monitoring point, where n represents the number of the multiple historical times, and k represents the k-th historical time. This represents the weighting coefficients of historical data at k historical moments. This represents the historical running data at historical time tk. This represents the error correction factor. This represents the prediction error at historical time t-1. in, This represents the historical running data at historical moment t-1. This represents the current running prediction data at historical time t-1.

3. The method according to claim 1, characterized in that, The step of determining whether there are monitoring points indicating operational anomalies based on the current operational prediction data and the current operational data includes: Obtain the absolute difference between the currently running predicted data and the currently running data; If the absolute difference of the data is greater than the preset absolute difference threshold, then it is determined that there is a monitoring point with abnormal operation.

4. The method according to claim 1, characterized in that, The fault feature vector library stores fault feature vectors for various faults; Accordingly, determining whether a fault exists at each monitoring point based on the feature vector and the fault feature vector library includes: Obtain the similarity between the feature vector and each fault feature vector in the fault feature vector library; If the similarity is greater than or equal to the preset similarity threshold, then the current monitoring point is determined to be faulty.

5. The method according to claim 1, characterized in that, The formula for obtaining the fault spatial correlation degree of each monitoring point based on the operational deviation degree and the communication link reliability is as follows: in, This represents the spatial correlation of the fault at the i-th monitoring point. The weighting coefficient represents the operational deviation. This indicates the degree of operational deviation; The weighting coefficients representing the reliability of the communication link. This indicates the reliability of the communication link.

6. The method according to claim 1, characterized in that, Determining the fault location of the marine ranch monitoring system based on the fault spatial correlation of each monitoring point includes: Determine whether there exists a continuous preset number of monitoring points whose fault spatial correlation exceeds a preset fault spatial correlation threshold; If the fault is detected, a preset number of monitoring points will be identified as the fault location of the marine ranch monitoring system. If it is determined that there is no fault, the monitoring point with the highest spatial correlation of the fault is identified as the fault location of the marine ranch monitoring system.

7. A fault location device for a marine ranching monitoring system, characterized in that, Applied to electronic devices, including: The data acquisition unit is used to acquire current operating data from multiple monitoring points in the marine ranching monitoring system, as well as historical operating data from multiple historical moments. The number of historical moments is determined based on the operating conditions of the marine ranching monitoring system, and the current operating data includes the communication data packet loss rate. The first acquisition unit is used to acquire the predicted data of the current operation of the pre-constructed digital twin at each monitoring point based on the historical operation data; wherein the pre-constructed digital twin is a pre-constructed digital model corresponding to the marine ranch monitoring system. The first judgment unit is used to determine whether there are any monitoring points with abnormal operation based on the predicted data and the current operation data. The second acquisition unit is used to extract multiple time-domain features and multiple frequency-domain features of the current operating data if it is determined that there is a monitoring point with abnormal operation; and to obtain the feature vector of the current operating data based on the multiple time-domain features and the multiple frequency-domain features. The second judgment unit is used to determine whether there is a fault at each monitoring point based on the feature vector and the fault feature vector library; The third acquisition unit is used to acquire the operating deviation of each monitoring point based on the predicted data and the current operating data if a fault is determined to exist. The fourth acquisition unit is used to acquire the communication link reliability of each monitoring point based on the communication data packet loss rate; The fifth acquisition unit is used to acquire the fault spatial correlation degree of each monitoring point based on the operational deviation degree and the communication link reliability; wherein the fault spatial correlation degree is used to quantify the probability of each monitoring point failing. The determining unit is used to determine the fault location of the marine ranch monitoring system based on the fault spatial correlation degree of each monitoring point.

8. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-6.

Images