Remote real-time monitoring method for port shore power system

Abstract

The invention provides a remote real-time monitoring method for a port shore power system, and relates to the technical field of power system monitoring, and the method comprises the steps: extracting power generation parameters from a mutual inductance data set during the power transmission of the shore power system; performing anomaly check on the mutual inductance data set to obtain a mutual inductance error, and determining an error induction node based on the mutual inductance error and the power generation parameter; the abnormal probability of the shore power system during power transmission is determined through the topological structure of the transmission nodes and the error induction nodes, and then split point anomaly analysis is carried out on the abnormal probability to obtain an abnormal isolation path; performing fault prediction on the error sensing node according to historical power transmission data to obtain an abnormal node sequence, and determining an abnormal load factor through the abnormal node sequence and an abnormal isolation path; and performing abnormity monitoring on the port shore power system based on the abnormal load factor to obtain a monitoring result of an abnormal transmission node, thereby realizing real-time fault monitoring and isolation of the port shore power system so as to improve the credibility of the monitoring result of the shore power system.

Description

Technical Field

[0001] This application relates to the technical field of power system monitoring. More specifically, this application relates to a remote real-time monitoring method for a port shore power system. Background Art

[0002] With the continuous development of smart grids and energy informatization, the monitoring and fault detection technologies of power systems have gradually evolved from traditional manual inspections and static alarms to a real-time monitoring system relying on sensor networks, edge computing, and data intelligent analysis. By continuously collecting and analyzing system operation data, it is possible to achieve dynamic perception of the grid operation state, fault early warning, and rapid response, thereby improving the stability and security of grid operation.

[0003] The port shore power system provides onshore power for berthed ships, reducing ship fuel consumption and environmental pollution, which has significant economic and environmental protection significance. However, due to its complex power transmission structure, frequent connection / disconnection behaviors, and various external interference factors during operation, it poses great challenges to fault location and anomaly analysis of the shore power system. Currently, the existing methods of setting fixed thresholds to trigger alarms, recording the state of transmission nodes, and static trend analysis can be used for the monitoring of the port shore power system. However, the existing technology lacks a dynamic correlation analysis mechanism, unable to establish an electrical logic and error propagation path between multiple data transmission nodes, and can only respond passively after a fault occurs, unable to predict in advance, resulting in difficulty in accurately identifying the fault source transmission node, poor robustness in dealing with complex fault modes, and low credibility of monitoring results. Therefore, how to achieve real-time fault monitoring and isolation of the port shore power system to improve the credibility of shore power system monitoring results has become a difficult problem faced by the current industry. Summary of the Invention

[0004] This application provides a remote real-time monitoring method for a port shore power system, which can achieve real-time fault monitoring and isolation of the port shore power system to improve the credibility of shore power system monitoring results.

[0005] This application provides a remote real-time monitoring method for a port shore power system. The real-time monitoring method includes the following steps: Collect the mutual inductance data set during the power transmission of the port shore power system, and extract the power generation parameters from the mutual inductance data set; Conduct an anomaly test on the mutual inductance data set to obtain a mutual inductance error, and determine an error induction node based on the mutual inductance error and the power generation parameters; Determine the anomaly probability during the power transmission of the shore power system through the topological structure of the port shore power system transmission node and the error induction node, and then conduct a split point anomaly analysis on the anomaly probability to obtain an anomaly isolation path; Obtain the historical power transmission data of the port shore power system, predict the faults of the error induction nodes based on the historical power transmission data to obtain an abnormal node sequence, and then determine the abnormal load factor through the abnormal node sequence and the abnormal isolation path; Perform abnormal monitoring on the port shore power system based on the abnormal load factor to obtain the monitoring results of the abnormal transmission nodes.

[0006] In this embodiment, a mutual inductance data set of the port shore power system is collected through a preset sensor array.

[0007] In this embodiment, a power generation parameter is extracted from the mutual inductance data set through a data extraction model.

[0008] In this embodiment, the abnormal inspection of the mutual inductance data set to obtain the mutual inductance error specifically includes: Perform denoising, normalization, and synchronous correction processing on the mutual inductance data in the mutual inductance data set to obtain the processed mutual inductance data; Perform time series abnormal detection analysis on the processed mutual inductance data to obtain the mutual inductance error.

[0009] In this embodiment, determining the error induction nodes based on the mutual inductance error and the power generation parameter specifically includes: Map the mutual inductance error and the power generation parameter to the same scale through normalization processing to obtain the normalized mutual inductance error and power generation parameter; Determine the abnormal sensitivity of the transmission nodes of the port shore power system based on the normalized mutual inductance error and power generation parameter; Perform abnormal screening on the transmission nodes of the port shore power system according to the abnormal sensitivity to obtain the error induction nodes.

[0010] In this embodiment, determining the abnormal probability during the power transmission of the shore power system through the topological structure of the transmission nodes of the port shore power system and the error induction nodes specifically includes: Convert the topological structure of the transmission nodes of the port shore power system into a directed weighted graph; Derive error propagation based on the error induction nodes for the directed weighted graph to obtain the abnormal transfer probabilities of different transmission nodes in the directed weighted graph; Determine the abnormal probability during the power transmission of the shore power system through all the abnormal transfer probabilities.

[0011] In this embodiment, performing split point abnormal analysis on the abnormal probability to obtain the abnormal isolation path specifically includes: Divide the abnormal split points through the abnormal probability during the power transmission of the shore power system; Perform partition processing on the abnormal split points to obtain the abnormal transmission node subgraph of the topological structure of the transmission nodes of the port shore power system; Determine the abnormal isolation path through the abnormal transmission node subgraph of the topological structure of the transmission nodes of the onshore power supply system at the port.

[0012] In this embodiment, the historical power transmission data of the onshore power supply system at the port is obtained through the data acquisition server.

[0013] In this embodiment, fault prediction is performed on the error induction nodes based on the historical power transmission data to obtain the abnormal node sequence, which specifically includes: Perform time series prediction on the historical power transmission data to obtain the operation curve of the transmission nodes; Determine the fault monitoring value based on the operation curve; Determine the abnormal node sequence based on the fault monitoring value and the operation value of the error induction nodes.

[0014] In this embodiment, abnormal monitoring is performed on the onshore power supply system at the port based on the abnormal load factor, and the monitoring result of the abnormal transmission node specifically includes: Obtain the operation parameters of the onshore power supply system at the port, and then determine the abnormal risk value of the transmission node through the operation parameters and the abnormal load factor; Perform cumulative judgment on the abnormal risk value to obtain the monitoring result of the abnormal transmission node.

[0015] The technical solution provided by the disclosed embodiment of the present application has the following beneficial effects: By collecting the mutual inductance data set during power transmission of the onshore power supply system at the port, extracting the power generation parameters from the mutual inductance data set; performing abnormal inspection on the mutual inductance data set to obtain the mutual inductance error, determining the error induction node based on the mutual inductance error and the power generation parameters; determining the abnormal probability during power transmission of the onshore power supply system through the topological structure of the transmission nodes of the onshore power supply system and the error induction node, and then performing split point abnormal analysis on the abnormal probability to obtain the abnormal isolation path; obtaining the historical power transmission data of the onshore power supply system at the port, performing fault prediction on the error induction nodes based on the historical power transmission data to obtain the abnormal node sequence, and then determining the abnormal load factor through the abnormal node sequence and the abnormal isolation path; performing abnormal monitoring on the onshore power supply system at the port based on the abnormal load factor to obtain the monitoring result of the abnormal transmission node.

[0016] It can be seen that in this application, real-time fault monitoring and isolation of the onshore power supply system of the port can be achieved. First, by collecting the mutual inductance data set during the power transmission process of the onshore power supply system of the port and extracting key power generation parameters from it, the operating status of each transmission node of the system can be comprehensively, continuously, and real-time grasped, avoiding information delay or omission, and providing a high-quality data basis for subsequent abnormal judgment. Second, by performing abnormal inspection on the mutual inductance data set to obtain the mutual inductance error, and jointly judging the error induction node based on this error and the power generation parameters, the misjudgment problem caused by single-threshold judgment in the traditional method is avoided, effectively improving the sensitivity and accuracy of fault diagnosis. Then, the transmission nodes of the onshore power supply system of the port are constructed into a topological network, and combined with the error induction node, the abnormal probability analysis during the power transmission of the onshore power supply system is carried out, and further split-point abnormal analysis is carried out. Finally, an abnormal isolation path is constructed, realizing accurate fault location and effective isolation of the affected area, and improving the response efficiency of the onshore power supply system to sudden abnormalities. Next, combined with the historical power transmission data of the onshore power supply system of the port, a fault prediction model is constructed to pre-identify the potential abnormal trend of the error induction node, thereby enhancing the active defense ability of the system. Finally, by jointly calculating the abnormal load factor with the abnormal node sequence and the abnormal isolation path, and based on this factor for the final monitoring judgment, an abnormal detection closed-loop mechanism is formed, which can improve the credibility of the monitoring results.

[0017] In summary, the technical solution adopted in this application can realize real-time fault monitoring and isolation of the onshore power supply system of the port to improve the credibility of the monitoring results of the onshore power supply system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 is a flowchart of a remote real-time monitoring method for an onshore power supply system of a port provided according to the present application; Figure 2 is an exemplary flowchart for determining the mutual inductance error provided according to the present application; Figure 3 is an exemplary flowchart for determining the abnormal probability during the power transmission of the onshore power supply system provided according to the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0021] The embodiment of the present application provides a remote real-time monitoring method for a port shore power system. The core is to collect visual parameters on the surface of the steel structure during the cutting process; normalize the parameter measurement values during the steel cutting process according to the visual parameters, and then linearly map the normalized parameter measurement values to a closed interval. Determine abnormal parameters through the closed interval and a preset specification threshold; construct a machine learning model for regulating cutting process parameters, and use the abnormal parameters to perform self-adjustment of training parameters based on cross-validation for the machine learning model to obtain a hyperparameter feature topology, and then determine a dynamic compensation coefficient from the hyperparameter feature topology; obtain the operating parameters of the cutting device, and adaptively dynamically adjust the operating parameters through the dynamic compensation coefficient based on a programmable logic controller to construct a logical control rule, and obtain adaptive cutting parameters; determine the intelligent regulation result of the steel cutting process through the adaptive cutting parameters and a preset cutting target value.

[0022] To better understand the above technical solutions, the above technical solutions will be described in detail below in conjunction with the accompanying drawings of the specification and specific implementation manners. Refer to Figure 1 As shown, this figure is an exemplary flowchart of a remote real-time monitoring method for a port shore power system according to an embodiment of the present application. The real-time monitoring method includes the following steps: In step S1, a mutual inductance data set during the power transmission of the port shore power system is collected, and power generation parameters are extracted from the mutual inductance data set.

[0023] Specifically, when implemented, a mutual inductance data set of the port shore power system is collected through a preset sensor array, that is: the electromagnetic signal during the transmission process is converted into a digital signal by a mutual inductance sensor deployed at the port shore power transmission node, and the collected analog signal is converted into digital data by an analog-to-digital converter with a high sampling rate, and then the data is transmitted to the monitoring center through the industrial Ethernet to obtain a mutual inductance data set.

[0024] It should be noted that the mutual inductance data set in this application refers to a set composed of the mutual inductance data of the onshore power transmission nodes in the port. Among them, the mutual inductance data refers to the electromagnetic signal data collected by mutual inductance sensors, which can reflect information such as current, voltage, and phase. The transmission nodes include: substations, busbars, and load terminals. In addition, the sampling rate in the high-sampling-rate analog-to-digital converter represents the number of times the sensor collects data per second. In this embodiment, the sampling rate can be set between 1 - 10 kHz, which is not limited here.

[0025] Specifically, when implemented, the power generation parameters can be extracted from the mutual inductance data set through a data extraction model. Among them, the data extraction model can include existing data extraction models such as Ohm's law, power calculation formula, and phase angle analysis model. It should be noted that the power generation parameters are physical quantities that can be used to characterize the working state of the power generation system. The power generation parameters include: power generation power, active power, reactive power, power factor, frequency deviation, voltage deviation, current distortion rate, etc.

[0026] In step S2, an anomaly check is performed on the mutual inductance data set to obtain a mutual inductance error, and an error induction node is determined based on the mutual inductance error and the power generation parameters.

[0027] Preferably, in this embodiment, refer to Figure 2 As shown, this figure is an exemplary flowchart for determining the mutual inductance error in the embodiment of this application. In this embodiment, performing an anomaly check on the mutual inductance data set to obtain the mutual inductance error can be specifically implemented by the following steps: First, in step S21, denoising, normalization, and synchronization correction processing are performed on the mutual inductance data in the mutual inductance data set to obtain the processed mutual inductance data; Then, in step S22, a time-series anomaly detection analysis is performed on the processed mutual inductance data to obtain the mutual inductance error.

[0028] In specific implementation, first, wavelet threshold denoising is adopted to eliminate random interference in the sampling process, that is, the mutual inductance data in the mutual inductance dataset is subjected to 3-layer wavelet denoising using the db4 wavelet basis, which can retain the main components of the signal. In other embodiments, wavelet bases of other sizes or other denoising algorithms can also be used for denoising, which is not limited here. And maximum-minimum normalization is used to perform unified scale conversion on the data features of different channels, and then the data of different measurement points are aligned based on the timestamp to obtain the processed mutual inductance data, which has the characteristics of noise suppression, scale unity, and timing consistency. Then, time series modeling analysis is performed on the processed mutual inductance data, and the median absolute deviation is used to judge anomalies in the residual sequence, that is: if the residual of the data transmission node at a certain moment exceeds the set threshold (such as 3 times the median absolute deviation), it is determined as an abnormal transmission node, and the deviation amplitude of this abnormal point is output through the median absolute deviation algorithm, and then the deviation amplitude of this abnormal transmission node is used as the mutual inductance error.

[0029] It should be noted that the mutual inductance error in this application refers to the outlier deviation of the mutual inductance data from the expected value. The processed mutual inductance data refers to the standardized mutual inductance data that has completed noise reduction, normalization, and multi-channel synchronization correction, and is used for subsequent anomaly analysis. In addition, in this embodiment, high-sensitivity identification of hidden anomalies and gradual errors can be achieved through time series anomaly analysis and dynamic error modeling, and it can be extended to adapt to the data formats of multiple types of mutual inductance elements (such as current, voltage, combined transformers), and is applicable to heterogeneous acquisition platforms for subsequent modules to identify error induction nodes and deduce abnormal paths.

[0030] In this embodiment, the following method can be specifically adopted to determine the error induction node based on the mutual inductance error and the power generation parameters, that is: The mutual inductance error and the power generation parameters are mapped to the same scale through normalization processing to obtain the normalized mutual inductance error and power generation parameters; Based on the normalized mutual inductance error and power generation parameters, determine the anomaly sensitivity of the transmission nodes of the onshore power supply system at the port; According to the anomaly sensitivity, perform anomaly screening on the transmission nodes of the onshore power supply system at the port to obtain the error induction nodes.

[0031] In specific implementation, first, the collected mutual inductance error and the extracted power generation parameters are respectively normalized to map each data to a unified scale (such as the 0-1 interval) to eliminate the influence of different dimensions and numerical range differences on subsequent calculations. For example, the maximum-minimum normalization method is used to process the mutual inductance error and the power generation parameters so that both are in the range of 0-1, obtaining the normalized mutual inductance error and power generation parameters. In other embodiments, other normalization techniques can also be used to map to other specified intervals, which is not limited here. Then, for each power transmission node of the onshore power supply system at the port, weights are assigned according to the normalized mutual inductance error and the power generation parameters, and the weighted integral method is used to calculate the anomaly sensitivity of each transmission node based on the normalized data, and a threshold of 0.7 is set as the anomaly screening criterion. Finally, the anomaly transmission nodes are rechecked by the local statistical clustering method, that is, the K-means clustering technology is used to monitor the fluctuation of the anomaly sensitivity of the transmission nodes, and then the transmission nodes with the anomaly sensitivity fluctuation greater than the set threshold of 0.7 are all used as error induction nodes. Among them, the anomaly screening criterion can also set the threshold between 0.6 and 0.9 according to historical experience, which is not limited here.

[0032] It should be noted that the anomaly sensitivity in this application refers to a quantitative evaluation index of the anomaly risk of each transmission node after comprehensively considering the normalized mutual inductance error and the power generation parameters. The error induction node is a key transmission node with a relatively high anomaly risk determined by screening the anomaly sensitivity of the transmission nodes. In addition, in this embodiment, multi-dimensional data fusion is realized through data normalization and multi-index weighted calculation, which improves the accuracy of anomaly sensitivity determination, and further reduces the misjudgment rate through a secondary verification mechanism, improving the response accuracy of the system to the abnormal state.

[0033] In step S3, the anomaly probability during the power transmission of the onshore power supply system is determined through the topological structure of the power transmission nodes of the onshore power supply system and the error induction nodes, and then the split-point anomaly analysis is performed on the anomaly probability to obtain the anomaly isolation path.

[0034] Preferably, in this embodiment, refer to Figure 3 As shown, this figure is an exemplary flowchart for determining the anomaly probability during the power transmission of the onshore power supply system in the embodiment of this application. The specific steps for determining the anomaly probability during the power transmission of the onshore power supply system through the topological structure of the power transmission nodes of the onshore power supply system and the error induction nodes in this embodiment can be implemented as follows: First, in step S31, the topological structure of the power transmission nodes of the onshore power supply system is converted into a directed weighted graph; Then, in step S32, error propagation derivation is performed on the directed weighted graph according to the error induction nodes to obtain the anomaly transfer probabilities of different transmission nodes in the directed weighted graph; Finally, in step S33, the abnormal probability during the power transmission of the onshore power supply system is determined based on all the abnormal transmission probabilities.

[0035] In specific implementation, the transmission node mapping algorithm is used to convert each transmission node in the onshore power supply system of the port into a directed weighted graph according to the actual power transmission path. Among them, the edge weight of the directed weighted graph can be comprehensively assigned based on the transmission impedance and data transmission delay, that is, the average value of the transmission impedance and data transmission delay is used as the edge weight of the directed weighted graph. The transmission impedance and data transmission delay can be determined by testing the network communication of the onshore power supply system. Then, taking the error induction node as the initial activation point, an error influence is exerted on its adjacent transmission nodes in the graph, that is, the error propagation factors of different edge weights are determined through the edge weight assignment in the prior art, and then each error propagation factor is used as the calculation weight of the recursive algorithm. The abnormal transmission probabilities of different transmission nodes in the directed weighted graph are calculated through the recursive algorithm with the set calculation weights. Finally, the abnormal transmission probabilities of each transmission node are comprehensively calculated. By normalizing all the abnormal transmission probabilities and performing weighted summation on all the normalized abnormal probabilities, the abnormal probability during the power transmission of the onshore power supply system is obtained, where the weight of the weighted summation is determined by the ratio of the error propagation factors.

[0036] It should be noted that the abnormal probability during the power transmission of the onshore power supply system in this application refers to the normalized risk index that quantifies the occurrence of abnormalities at each transmission node after comprehensively considering all the abnormal transmission probabilities. The directed weighted graph represents a graph with the transmission nodes of the onshore power supply system of the port as vertices, the connection relationships between the transmission nodes as edges, and each edge is assigned a weight reflecting the power transmission characteristics (such as impedance, load, etc.). The abnormal transmission probability represents the probability value of the abnormal influence being transmitted from the error induction node to other transmission nodes in the directed weighted graph. In addition, by globally expressing the transmission nodes of the onshore power supply system of the port through the graph model, compared with single-point monitoring, the spatial propagation analysis of the abnormal influence can be realized, and the influence of the error induction node on other transmission nodes in the whole network can be quantified through error propagation prediction, which is beneficial to identifying the risk of fault diffusion.

[0037] In this embodiment, to perform abnormal split point analysis on the abnormal probability and obtain the abnormal isolation path, the following method can be specifically adopted, that is: Divide the abnormal split points through the abnormal probability during the power transmission of the onshore power supply system; Perform partition processing on the abnormal split points to obtain the sub-graph of abnormal transmission nodes of the topological structure of the transmission nodes of the onshore power supply system of the port; Determine the abnormal isolation path through the sub-graph of abnormal transmission nodes of the topological structure of the transmission nodes of the onshore power supply system of the port.

[0038] In specific implementation, first, using clustering or threshold segmentation methods, the delivery node regions with a relatively high abnormal probability during the power delivery of the onshore power system are used as abnormal splitting points. For example, according to the set abnormal probability threshold for the power delivery of the onshore power system (such as 0.8), cluster analysis is performed on the delivery nodes with an abnormal probability exceeding this abnormal probability threshold during the power delivery of the onshore power system in the topology graph. Then, the image region composed of the nodes exceeding this abnormal probability threshold obtained by clustering is used as the abnormal splitting point region. Then, partition processing is performed on the abnormal splitting points in the divided abnormal splitting point region, that is, the minimum segmentation algorithm is used to cut the abnormal splitting point region at the abnormal splitting points, and the obtained image regions are used as the abnormal delivery node subgraphs of the topological structure of the onshore power system delivery nodes. Finally, within the abnormal delivery node subgraphs, according to the edge weights and abnormal transmission probabilities between the delivery nodes, the Dijkstra algorithm is used to calculate the shortest path from the abnormal source to the isolation boundary, and this shortest path is used as the abnormal isolation path. Among them, the abnormal splitting point refers to the key delivery node region determined by setting a threshold or cluster analysis in the abnormal probability distribution during the global onshore power system power delivery. This region plays a bridging role in abnormal propagation. The abnormal delivery node subgraph is a local subgraph divided from the original topological graph of the onshore power system delivery nodes in the port after partition processing of the abnormal splitting points. All delivery nodes within this subgraph have a relatively high abnormal risk.

[0039] It should be noted that the abnormal isolation path in this application represents the optimal path planned according to the transmission weights and abnormal transmission probabilities between the delivery nodes in the abnormal delivery node subgraph, which is used to cut off the channel for abnormal diffusion. In addition, using the splitting point partition can effectively identify and isolate the abnormal diffusion region from the global topology perspective. The isolation path constructed based on the abnormal delivery node subgraph can cut off the abnormal conduction in the shortest time and reduce the abnormal diffusion risk.

[0040] In step S4, the historical power transmission data of the port onshore power system is obtained, and the error induction nodes are fault predicted based on the historical power transmission data to obtain an abnormal node sequence. Then, the abnormal load factor is determined through the abnormal node sequence and the abnormal isolation path.

[0041] In specific implementation, the historical power transmission data of the port onshore power system is obtained through a data acquisition server. Among them, the data acquisition server is a power data acquisition server equipped with a SCADA platform, which can extract historical power transmission data from the database of the SCADA platform. The historical power transmission data includes historical power parameters such as voltage, current, power, active power, and reactive power. After preprocessing the collected data using a Python script, the data is converted into a standardized format and batch written into the time series database.

[0042] In this embodiment, the following method can be specifically adopted to perform fault prediction on the error sensing nodes based on the historical power transmission data to obtain an abnormal node sequence, that is: Perform time series prediction on the historical power transmission data to obtain the operation curve of the transmission nodes; Determine the fault monitoring value according to the operation curve; Determine the abnormal node sequence based on the fault monitoring value and the operation value of the error sensing nodes.

[0043] Specifically, when implementing, preprocess the collected historical power transmission data (including missing value filling, noise filtering, timestamp alignment, etc.), and use a time series prediction model to perform trend prediction on the operation data of each transmission node, so as to obtain the operation curve of each transmission node. For example, the LSTM model can be used to train and predict the historical power transmission data to generate the operation curve of the transmission node; then, the residual analysis method is adopted to quantify the deviation between the predicted curve and the actual curve. When the deviation exceeds the preset threshold, determine the fault monitoring value of the transmission node. The fault monitoring value can be obtained by calculating key indicators such as the residual, deviation degree, and fluctuation amplitude between the actual operation value and the predicted value, which is an important basis for judging whether there is a fault trend in the transmission node. Among them, the preset threshold can be obtained by performing autoregressive analysis on the historical power transmission data through the autoregressive analysis algorithm in the prior art; finally, compare the extracted fault monitoring value with the operation value of the error sensing node, and use the transmission node where the operation value is greater than the fault monitoring value as the abnormal transmission node, and then use the set of all abnormal transmission nodes as the abnormal node sequence, where the operation value refers to the value of the power parameter when the transmission node of the onshore power system is operating.

[0044] It should be noted that the fault monitoring value in this application represents the residual or deviation degree extracted after comparing the operation curve with the actual operation data, which is a quantitative index used to reflect the potential fault risk of the transmission node; the abnormal node sequence is a set of error sensing nodes determined to have a potential fault trend after being screened by the fault monitoring value; in addition, the trend curve of the future operation state of each transmission node obtained through the time series prediction model can capture the operation trend of the transmission node, effectively improving the sensitivity and accuracy of abnormal detection.

[0045] In this embodiment, the following method can be specifically adopted to determine the abnormal load factor through the abnormal node sequence and the abnormal isolation path, that is: Determine the risk score of each abnormal transmission node in the abnormal node sequence; Perform contribution screening on the abnormal isolation path according to all the risk scores to obtain the risk contribution degrees of different transmission nodes; Determine the abnormal load factor based on all the risk scores and all the risk contribution degrees.

[0046] In specific implementation, first, by normalizing the residuals, deviation degrees, and abnormal durations of each abnormal transmission node and assigning corresponding weights, a risk score matrix of the transmission nodes is obtained, and then the risk scores of each abnormal matrix are extracted from the risk score matrix; then, the transmission nodes in the abnormal isolation path are arranged in the path order, the risk transmission and cumulative effect between the transmission nodes are analyzed, and a weighted integral or accumulation model is used to integrate the risk scores of the transmission nodes on the path to obtain the risk contribution degree of the overall path. For example: applying a weighted accumulation algorithm on the abnormal isolation path to sum the risk values from the starting transmission node of the path to the isolation boundary, and the risk contribution degrees of different transmission nodes can be obtained through this step; finally, the risk contribution degree and the risk score are weighted and fused, and combined with the global normalization method, the abnormal load factor of each transmission node or area is finally determined. For example: the abnormal load factor is the weighted average of the risk scores of each transmission node and the risk contribution of the isolation path, and the results of all transmission nodes are normalized to ensure that the load factor value is between 0 and 1.

[0047] It should be noted that the abnormal load factor in this application is a quantitative index that comprehensively reflects the cumulative effect of local abnormal risks and the abnormal propagation isolation effect, usually represented by a value between 0 and 1. The larger the value, the higher the abnormal load. The risk score of an abnormal transmission node is a risk index calculated based on the fault monitoring value, abnormal duration, and abnormal transmission probability of each transmission node. By integrating the abnormal node sequence and isolation path information, it not only considers the risk status of a single transmission node but also reflects the diffusion effect of the abnormality in the network, providing a more global perspective compared to traditional evaluation methods that only rely on single-point indicators.

[0048] In step S5, based on the abnormal load factor, abnormal monitoring of the onshore power supply system of the port is performed to obtain the monitoring results of the abnormal transmission nodes.

[0049] In this embodiment, the abnormal monitoring of the onshore power supply system of the port based on the abnormal load factor to obtain the monitoring results of the abnormal transmission nodes can be specifically implemented in the following manner, that is: Obtain the operating parameters of the onshore power supply system of the port, and then determine the abnormal risk value of the transmission node through the operating parameters and the abnormal load factor; Perform cumulative judgment on the abnormal risk value to obtain the monitoring results of the abnormal transmission nodes.

[0050] In specific implementation, first, the operation parameters of each transmission node are read in real time through the SCADA platform interface. The operation parameters refer to the electrical parameters collected in real time during the operation of each transmission node of the onshore power supply system in the port, including: voltage, current, power, frequency, and load status. The maximum-minimum normalization method is used to ensure that the data are all in the same dimension for subsequent calculations. Then, the abnormal risk value is calculated by calculating the parameter deviation degree between the calculated abnormal load factor and the operation parameters. Among them, the reciprocal of the Pearson correlation coefficient of the operation parameters can be used as the parameter deviation degree of the operation parameters. The abnormal risk value can be obtained through the following formula, that is: abnormal risk value = α × parameter deviation degree + β × abnormal load factor, where α and β are weight parameters, which can be preset according to historical experience. After normalizing the calculation results, the abnormal risk values of each transmission node can be obtained. Finally, through the time series integration technology, the abnormal risk values of each transmission node within a continuous monitoring period are cumulatively judged, and the persistence and cumulative effect of the risk values are evaluated through a sliding window or time series integration method. When the cumulative risk exceeds the preset threshold, it is determined that the transmission node is an abnormal transmission node, and its abnormal monitoring results (including information such as the risk level and the abnormal duration) are output.

[0051] It should be noted that the abnormal risk value in this application is a risk quantification index calculated based on operation parameters and abnormal load factors, and is used to judge the abnormal state of the transmission node. The higher its value, the greater the abnormal risk; the monitoring results of the abnormal transmission node include the determination of the abnormal state of each transmission node and the operation parameters determined as risks.

[0052] It can be seen that in this application, real-time fault monitoring and isolation of the onshore power supply system of ports can be achieved. First, by collecting the mutual inductance data set during the power transmission process of the onshore power supply system of ports and extracting key power generation parameters from it, the operating status of each transmission node of the system can be comprehensively, continuously, and real-time grasped, avoiding information delay or omission, and providing a high-quality data basis for subsequent anomaly judgment. Second, by performing anomaly detection on the mutual inductance data set to obtain the mutual inductance error, and jointly judging the error induction node based on this error and the power generation parameters, the misjudgment problem caused by single-threshold judgment in traditional methods is avoided, effectively improving the sensitivity and accuracy of fault diagnosis. Then, the transmission nodes of the onshore power supply system of ports are constructed into a topological network, and anomaly probability analysis is carried out during the power transmission of the onshore power supply system in combination with the error induction node, and further split-point anomaly analysis is carried out. Finally, an anomaly isolation path is constructed, realizing accurate fault location and effective isolation of the affected area, and improving the response efficiency of the onshore power supply system to sudden anomalies. Next, in combination with the historical power transmission data of the onshore power supply system of ports, a fault prediction model is constructed to pre-identify the potential anomaly trend of the error induction node, thereby enhancing the active defense ability of the system. Finally, by jointly calculating the anomaly load factor with the anomaly node sequence and the anomaly isolation path, and performing the final monitoring judgment based on this factor, an anomaly detection closed-loop mechanism is formed, which can improve the credibility of the monitoring results.

[0053] In summary, the technical solution adopted in this application can achieve real-time fault monitoring and isolation of the onshore power supply system of ports to improve the credibility of the monitoring results of the onshore power supply system.

[0054] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0055] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. The storage medium includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc memories, magnetic disc memories, tape memories, or any other medium that can be used to carry or store data and is computer-readable.

[0056] It should also be noted that the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, commodity or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, commodity or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the existence of another identical element in the process, method, commodity or device comprising the element.

Claims

1. A remote real-time monitoring method for a port shore power system, characterized in that, The real-time monitoring method includes the following steps: Collect the mutual inductance data set during the power transmission of the onshore power supply system at the port, and extract the power generation parameters from the mutual inductance data set; Perform an anomaly test on the mutual inductance data set to obtain a mutual inductance error, and determine the error induction node based on the mutual inductance error and the power generation parameters; Determine the anomaly probability during the power transmission of the onshore power supply system through the topological structure of the transmission nodes of the onshore power supply system and the error induction node, and then perform split-point anomaly analysis on the anomaly probability to obtain the anomaly isolation path; Obtain the historical power transmission data of the onshore power supply system at the port, perform fault prediction on the error induction node based on the historical power transmission data to obtain an abnormal node sequence, and then determine the abnormal load factor through the abnormal node sequence and the abnormal isolation path; Perform anomaly monitoring on the onshore power supply system based on the abnormal load factor to obtain the monitoring results of the abnormal transmission nodes.

2. The remote real-time monitoring method for a port shore power system according to claim 1, characterized in that, Collect the mutual inductance data set of the onshore power supply system at the port through a preset sensor array.

3. The remote real-time monitoring method for a port shore power system according to claim 1, characterized in that, Extract the power generation parameters from the mutual inductance data set through a data extraction model.

4. The remote real-time monitoring method for a port shore power system according to claim 1, characterized in that, Performing an anomaly test on the mutual inductance data set to obtain a mutual inductance error specifically includes: Perform denoising, normalization, and synchronization correction processing on the mutual inductance data in the mutual inductance data set to obtain the processed mutual inductance data; Perform time-series anomaly detection analysis on the processed mutual inductance data to obtain a mutual inductance error.

5. The remote real-time monitoring method for a port shore power system according to claim 1, characterized in that, Determining the error induction node based on the mutual inductance error and the power generation parameters specifically includes: Map the mutual inductance error and the power generation parameters to the same scale through normalization processing to obtain the normalized mutual inductance error and power generation parameters; Determine the anomaly sensitivity of the transmission nodes of the onshore power supply system based on the normalized mutual inductance error and power generation parameters; Perform anomaly screening on the transmission nodes of the onshore power supply system according to the anomaly sensitivity to obtain the error induction node.

6. The remote real-time monitoring method for a port shore power system according to claim 1, characterized in that Determining the anomaly probability during the power transmission of the onshore power supply system through the topological structure of the transmission nodes of the onshore power supply system and the error induction node specifically includes: Convert the topological structure of the transmission nodes of the onshore power supply system into a directed weighted graph; Derive the error propagation of the directed weighted graph based on the error induction node to obtain the anomaly transfer probabilities of different transmission nodes in the directed weighted graph; Determine the anomaly probability during the power transmission of the onshore power supply system through all the anomaly transfer probabilities.

7. A remote real-time monitoring method for a port shore power system according to claim 1, characterized in that, Performing split-point anomaly analysis on the anomaly probability to obtain the anomaly isolation path specifically includes: Divide the anomaly split points through the anomaly probability during the power transmission of the onshore power supply system; Perform partition processing on the anomaly split points to obtain the abnormal transmission node subgraph of the topological structure of the transmission nodes of the onshore power supply system; Determine the anomaly isolation path through the abnormal transmission node subgraph of the topological structure of the transmission nodes of the onshore power supply system.

8. A remote real-time monitoring method for a port shore power system according to claim 1, characterized in that, Obtain the historical power transmission data of the onshore power supply system at the port through a data acquisition server.

9. The remote real-time monitoring method for a port shore power system according to claim 1, characterized in that, Performing fault prediction on the error induction node based on the historical power transmission data to obtain an abnormal node sequence specifically includes: Perform time-series prediction on the historical power transmission data to obtain the operation curve of the transmission node; Determine the fault monitoring value based on the operation curve; Determine the abnormal node sequence based on the fault monitoring value and the operation value of the error induction node.

10. A remote real-time monitoring method for a port shore power system according to claim 1, characterized in that, Based on the abnormal load factor, abnormal monitoring is carried out on the onshore power supply system of the port, and the monitoring results of abnormal transmission nodes specifically include: Obtain the operating parameters of the onshore power supply system of the port, and then determine the abnormal risk value of the transmission node through the operating parameters and the abnormal load factor; Perform cumulative judgment on the abnormal risk value to obtain the monitoring results of abnormal transmission nodes.

Images