SVG power grid construction and situation awareness method supporting multi-source data fusion

Abstract

The invention relates to the technical field of power supply management systems, in particular to an SVG power grid construction and situation awareness method supporting multi-source data fusion, and the method comprises the steps: S1, collecting the multi-source data of a power grid, and S2, constructing a dynamic knowledge graph, according to the method, a situation awareness SVG model is constructed through the step S3, the output process of the situation awareness SVG model is adjusted and updated through the step S4, and reason analysis is conducted on an abnormal load area through the step S5. The situation awareness SVG model is constructed according to the multi-source data of the power grid, and the situation awareness SVG model is corrected in real time according to the multi-source data of the power grid, so that the operation state of the power grid is comprehensively mastered, potential risks are predicted, and the awareness efficiency and accuracy of the situation of the power grid are improved.

Description

Technical Field

[0001] This invention relates to the field of power supply management system technology, and in particular to an SVG power grid construction and situational awareness method that supports multi-source data fusion. Background Technology

[0002] Against the backdrop of rapid development in global information and industrial technologies, power grids exist extensively in the form of internet-connected data grids. However, the large-scale integration of wind and solar power has brought significant uncertainties and volatility to power grids, increasing the risk of anomalies in grid operation and posing challenges to the safe and stable operation of the grid. Power grid situational awareness is one of the important factors in ensuring the safe operation of the power system. It uses technologies such as machine learning and data mining to predict future electricity demand and power failures. However, the single data collection mode of the power grid leads to perception results that are out of touch with reality and difficult to apply. A reasonable power grid situational awareness method is key to evaluating the effectiveness of the power grid system; conversely, it may lead to erroneous power grid status assessments. Therefore, there is an urgent need to propose a reasonable situational awareness method.

[0003] Chinese patent CN119669965A discloses a method and system for power grid anomaly situation perception. The method includes: acquiring power grid data in real time and calculating different power grid situation assessment indicators; assigning weights to each power grid situation assessment indicator and correcting the weights of anomaly indicators, then linearly weighting all power grid situation assessment indicators to obtain a situation assessment value; constructing a prediction model with a two-layer gated recurrent neural network, an attention layer, and an output layer cascaded, inputting historical data of the power grid situation assessment indicators, and outputting predicted future situation assessment values; inputting new historical data of the power grid situation assessment indicators into the trained prediction model to obtain the corresponding predicted future situation assessment values, and matching the predicted future situation assessment values ​​with pre-defined security levels to determine whether an anomaly will occur. Therefore, this solution still suffers from problems such as a single source of power grid data acquisition and a lack of real-time and accurate perception of the power grid situation, resulting in insufficient accuracy and low efficiency in power grid situation perception. Summary of the Invention

[0004] To address this, the present invention provides an SVG power grid construction and situational awareness method that supports multi-source data fusion, thereby overcoming the problems of insufficient accuracy and low efficiency in the perception of power grid situation caused by the single source of power grid data acquisition and the lack of real-time and accurate perception of power grid situation in the prior art.

[0005] To achieve the above objectives, this invention provides an SVG power grid construction and situational awareness method supporting multi-source data fusion, the method comprising: Step S1: Collect multi-source data from the power grid; Step S2: The multi-source data of the power grid is fused to obtain a standardized multi-source data package of the power grid, and a dynamic knowledge graph is constructed based on the standardized multi-source data package of the power grid to obtain the dynamic knowledge graph; Step S3: Construct the situational awareness SVG model based on the dynamic knowledge graph, output the situational awareness SVG model, and optimize the output process of the situational awareness SVG model according to the device spacing. Step S4 involves adjusting the load based on the load factor during the distance optimization process, and updating the load based on the three-phase imbalance during the load adjustment process. Step S5: Obtain the abnormal load area based on the load rate, perform cause analysis on the abnormal load area, obtain the load cause analysis results, and correct the distance optimization based on the load cause analysis results.

[0006] Furthermore, step S1 involves collecting multi-source data from the power grid, including real-time power grid data, geographic coordinate information, power grid topology model, three-phase imbalance, actual load, and rated load. When step S2 performs fusion processing on the multi-source data of the power grid according to the fusion processing method, the fusion processing method includes: Step A01: Perform data cleaning on the multi-source data of the power grid to obtain the cleaned multi-source data of the power grid; Step A02: Convert the format of the cleaned multi-source power grid data to obtain unified multi-source power grid data; Step A03: Perform entity alignment on the unified power grid multi-source data to obtain a standardized power grid multi-source data packet.

[0007] Furthermore, when step S2 constructs the dynamic knowledge graph based on standardized power grid multi-source data packets using the dynamic knowledge graph construction method, the dynamic knowledge graph construction method includes: Step B01: Define entities for the standardized power grid multi-source data packets to obtain the defined set of entity points; Step B02: Construct connection relationships based on the defined set of entity points to obtain the basic graph; Step B03: Generate a dynamic chart based on the basic knowledge graph to obtain a dynamic chart, and use the dynamic chart as a dynamic knowledge graph.

[0008] Furthermore, when step S3 constructs the situation awareness SVG model based on the dynamic knowledge graph using the situation awareness SVG model construction method, the situation awareness SVG model construction method includes: Step C01: Input the geographic coordinate information and power grid topology model into the dynamic knowledge graph to obtain the device coordinates and device connection patterns output by the dynamic knowledge graph; Step C02: Perform vector conversion on the device connection configuration to obtain the device vector symbol. Step C03: Input the device coordinates and device vector symbols into the SVG construction tool to obtain the situational awareness SVG model.

[0009] Further, in step S3, when optimizing the output process of the situational awareness SVG model based on the device spacing, the device spacing CX is compared with the preset device spacing CX0. The safety of the device spacing is judged based on the comparison result, and the output process of the situational awareness SVG model is optimized based on the judgment result, wherein: When CX≤CX0, the safety status of the device spacing is determined to be safe, and distance optimization is not performed on the output process of the situational awareness SVG model; When CX > CX0, the safety status of the device spacing is determined to be unsafe. Distance optimization is performed on the output process of the situational awareness SVG model: the device coordinates are aligned to obtain optimized device coordinates. The optimized device coordinates are then replaced with the optimized device coordinates. The device coordinates and device vector symbols are then re-input into the SVG construction tool to obtain the situational awareness SVG model.

[0010] Further, in step S4, when adjusting the load based on the distance optimization process according to the load factor, the load factor FLP is calculated based on the actual load AL and rated load ALP from the multi-source data of the power grid. FLP is set to AL / ALP to obtain the load factor FLP. The load factor FLP is then compared with the preset load factor FLP0. Based on the comparison result, the state of the load factor is judged, and the load is adjusted according to the preset equipment spacing CX0 based on the judgment result. Wherein: When FLP≤FLP0, the load rate is determined to be normal, and no load adjustment is performed on the preset equipment spacing CX0; When FLP > FLP0, the load rate is determined to be abnormal. The load is adjusted according to the preset equipment spacing CX0 based on the adjustment coefficient tz, which is set to tz = 0.67 + 0.22 × e. -(FLP-FLP0) Where e is the base of the natural logarithm, the adjusted preset device spacing CX0` is obtained, CX0` is set to CX0×tz, the preset device spacing CX0 is replaced with the adjusted preset device spacing CX0`, and the device spacing CX is re-compared with the preset device spacing CX0.

[0011] Further, in step S4, when updating the load based on the three-phase unbalance, the three-phase unbalance SV is compared with the preset three-phase unbalance SV0. Based on the comparison result, the three-phase unbalance is judged, and based on the judgment result, the preset load rate FLP0 is updated. Wherein: When SV≤SV0, the three-phase imbalance is considered to be within the acceptable range, and no load update is performed on the preset load rate FLP0. When SV > SV0, the three-phase imbalance is deemed substandard. The preset load factor FLP0 is then updated based on the update factor gz, which is set to gz = 0.72 + 0.23 × e. -(SV-SV0) Where e is the base of the natural logarithm, the updated preset load rate FLP0` is obtained, and FLP0` is set to FLP0×gz. The preset load rate FLP0 is replaced with the updated preset load rate FLP0`, and the load rate FLP is compared with the preset load rate FLP0 again.

[0012] Further, in step S5, when acquiring abnormal load areas based on load rates and performing cause analysis on abnormal load areas, the load rate FLP is compared with a preset high load rate FLP1. Based on the comparison result, the load attributes of the load rate are determined, and the abnormal load areas are output based on the determination result. Wherein: When FLP≤FLP1, the load attribute of the load factor is determined to be overloaded, and no output is output for abnormal load areas; When FLP > FLP1, the load attribute of the load rate is determined to be overload, the abnormal load area is output, and the cause analysis of the abnormal load area is performed.

[0013] Furthermore, when performing cause analysis on the abnormal load area in step S5 according to the cause analysis method, the cause analysis method includes: Step D01: Perform vector transformation on abnormal current, abnormal voltage, and abnormal equipment temperature to obtain current vector value OP, voltage vector value KP, and equipment temperature vector value WP. Step D02: Calculate the current evaluation value OPAU based on the current vector value OP and the current mean value OPA, and set OPAU = OP / OPA; Step D03: Calculate the voltage evaluation value KPAU based on the voltage vector value KP and the voltage mean KPA, and set KPAU = KP / KPA; Step D04: Calculate the equipment temperature assessment value WPAU based on the equipment temperature vector value WP and the equipment temperature mean value WPA, and set WPAU = WP / WPA; Step D05: Calculate the abnormal descent assessment value ZSDF based on the current assessment value OPAU, voltage assessment value KPAU, equipment temperature assessment value WPAU, current assessment weight w1, voltage assessment weight w2, equipment temperature assessment weight w3, and load factor v. Set ZSDF = (OPAU × w1 + KPAU × w2 + WPAU × w3) / v to obtain the abnormal descent assessment value ZSDF. Step D06: Compare the abnormal drop assessment value ZSDF with the preset abnormal drop assessment value ZSDF0. Determine the status of the abnormal drop assessment value based on the comparison result, and output the load cause analysis result based on the determination result. Wherein: When ZSDF≤ZSDF0, the abnormal drop evaluation value is determined to be normal, and non-fault causes are output as load cause analysis results. When ZSDF > ZSDF0, the abnormal drop evaluation value is determined to be abnormal, and the cause of the fault is output as the result of the load cause analysis.

[0014] Further, in step S5, the distance optimization is corrected based on the load cause analysis results. When the fault cause is output as a load cause analysis result, the fault assessment probability model is input according to the fault triggering frequency to obtain the fault triggering probability GS output by the fault assessment probability model. The fault triggering probability GS is compared with the preset fault triggering probability GSO. The state of the fault triggering probability is judged according to the comparison result, and the preset equipment spacing CX0 is corrected according to the judgment result, wherein: When GS≤GS0, the fault trigger probability is determined to be normal, and the preset device spacing CX0 is not corrected. When GS > GS0, the fault trigger probability is determined to be abnormal. The preset equipment spacing CX0 is then corrected according to the correction coefficient qc, which is set to qc = 0.77 + 0.21 × e -(GS-GS0) Where e is the base of the natural logarithm, the corrected preset device spacing CX0` is obtained, CX0` is set to CX0×qc, the preset device spacing CX0 is replaced with the corrected preset device spacing CX0`, and the device spacing CX is re-compared with the preset device spacing CX0.

[0015] Compared with existing technologies, the beneficial effects of the present invention are as follows: The method collects multi-source data of the power grid in step S1 to comprehensively perceive the power grid situation based on data from different sources, thereby improving the accuracy of power grid situation perception. The method also constructs a dynamic knowledge graph in step S2 to present the relationship attributes between devices in the power grid in a visual manner, and updates the dynamic knowledge graph in real time by updating the multi-source data of the power grid, dynamically reflecting the power grid situation, thereby improving the efficiency of power grid situation perception. The method also constructs a situation perception SVG model in step S3 to directly perceive the power grid situation in real time based on the situation perception SVG model, further improving the efficiency of power grid situation perception. The method also adjusts and updates the output process of the situation perception SVG model in step S4 to increase the accuracy of the situation perception SVG model, thereby improving the environmental adaptability and security of the situation perception SVG model. The method also performs cause analysis on abnormal load areas in step S5 to determine whether there are faults in the abnormal load areas, so as to correct the situation perception SVG model, thereby improving the accuracy and efficiency of power grid situation perception. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the SVG power grid construction and situational awareness method that supports multi-source data fusion in this embodiment. Detailed Implementation

[0017] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0018] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0019] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0020] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0021] Please see Figure 1 As shown, this is a flowchart illustrating the SVG power grid construction and situational awareness method supporting multi-source data fusion in this embodiment. The method includes: Step S1: Collect multi-source data from the power grid; Step S2: The multi-source data of the power grid is fused to obtain a standardized multi-source data package of the power grid, and a dynamic knowledge graph is constructed based on the standardized multi-source data package of the power grid to obtain the dynamic knowledge graph; Step S3: Construct the situational awareness SVG model based on the dynamic knowledge graph, output the situational awareness SVG model, and optimize the output process of the situational awareness SVG model according to the device spacing. Step S4 involves adjusting the load based on the load factor during the distance optimization process, and updating the load based on the three-phase imbalance during the load adjustment process. Step S5: Obtain the abnormal load area based on the load rate, perform cause analysis on the abnormal load area, obtain the load cause analysis results, and correct the distance optimization based on the load cause analysis results.

[0022] Specifically, the SVG power grid construction and situational awareness method supporting multi-source data fusion is applied to power grid monitoring terminals, such as substation integrated automation systems. The method constructs a situational awareness SVG model based on multi-source power grid data and continuously modifies the model in real time based on the data. This facilitates a comprehensive understanding of the power grid's operating status and prediction of potential risks, thereby improving the efficiency and accuracy of power grid situational awareness. Specifically, step S1 involves collecting multi-source power grid data to comprehensively perceive the power grid situation based on data from different sources, thus improving the accuracy of situational awareness. Step S2 involves constructing a dynamic knowledge graph to visualize the relationships between devices in the power grid and continuously updating the multi-source power grid data in real time. The method updates the dynamic knowledge graph in real time to dynamically reflect the power grid situation, thereby improving the efficiency of power grid situation perception. Step S3 involves constructing a situational awareness SVG model to directly perceive the power grid situation in real time, further improving the efficiency of power grid situation perception. Step S4 adjusts and updates the output process of the situational awareness SVG model to increase its accuracy, thereby improving its environmental adaptability and security. Step S5 performs cause analysis on abnormal load areas to determine whether there are faults in these areas, allowing for correction of the situational awareness SVG model and improving the accuracy and efficiency of power grid situation perception.

[0023] Specifically, step S1 involves collecting multi-source data from the power grid, including real-time power grid data, geographic coordinate information, power grid topology model, three-phase imbalance, actual load, and rated load.

[0024] Specifically, the real-time power grid data includes voltage, current, and equipment temperature. Voltage refers to the operating voltage of the power grid electrical equipment, current refers to the directional flow of charge in the power grid system, and equipment temperature refers to the temperature value of the power grid electrical equipment under operating conditions. This embodiment does not limit the specific method of collecting the real-time power grid data; those skilled in the art can freely choose according to actual needs, such as collecting the real-time power grid data through a monitoring and data acquisition system. The geographic coordinate information refers to the location coordinates of the power grid electrical equipment collected through GIS, such as the first equipment coordinates Ap(x1,y1,z1) and the second equipment coordinates Bp(x2,y2,z2), where x1 is the value of the first equipment coordinate on the X-axis in a three-dimensional spatial coordinate system, and y1 is the value of the first equipment coordinate on the X-axis. The coordinates are defined as follows: z1 is the Y-axis value of the first device coordinate on the Z-axis, x2 is the X-axis value of the second device coordinate on the X-axis, y2 is the Y-axis value of the second device coordinate on the Z-axis, and z2 is the Z-axis value of the second device coordinate on the Z-axis. This embodiment does not limit the construction method of the spatial three-dimensional coordinate system; those skilled in the art can freely choose according to actual needs. For example, the power dispatch control center can be used as the origin to construct the spatial three-dimensional coordinate system. The power dispatch control center refers to the central location for overall control of the power grid. GIS stands for Geographic Information System, which is software for collecting geographic coordinate information. InformationSystem, the power grid topology model refers to a mathematical model describing the connection relationship and network structure of electrical equipment in a power system. This embodiment does not limit the specific method of obtaining the power grid topology model, such as obtaining the power grid topology model through a public information model. The three-phase imbalance refers to the degree to which the phase difference between the three-phase voltage and the three-phase current in the power grid deviates from 120°. The three-phase voltage refers to an AC voltage signal with a phase difference of 120°. The three-phase current refers to the current synchronized with the three-phase voltage. This embodiment does not limit the specific method of obtaining the three-phase imbalance. Those skilled in the art can freely choose according to actual needs, such as collecting the three-phase imbalance through a power quality analyzer. The actual load refers to the sum of the active and reactive power actually consumed by the electrical equipment in the power grid. The rated load refers to the sum of the active power of the electrical equipment in the power grid under ideal conditions. This embodiment does not limit the specific method of collecting the actual load and the rated load. Those skilled in the art can freely choose according to actual needs, such as collecting the actual load and the rated load through a monitoring and data acquisition system.

[0025] Specifically, step S1 involves collecting multi-source data from the power grid through multiple acquisition dimensions to enable a comprehensive perception of the power grid situation based on data from different sources, thereby improving the accuracy of power grid situation perception.

[0026] Specifically, when step S2 performs fusion processing on the multi-source data of the power grid according to the fusion processing method, the fusion processing method includes: Step A01: Perform data cleaning on the multi-source data of the power grid to obtain the cleaned multi-source data of the power grid; Step A02: Convert the format of the cleaned multi-source power grid data to obtain unified multi-source power grid data; Step A03: Perform entity alignment on the unified power grid multi-source data to obtain a standardized power grid multi-source data packet.

[0027] Specifically, data cleaning refers to the process of removing noise from multi-source power grid data. This embodiment does not limit the specific method of data cleaning; those skilled in the art can freely choose according to actual needs, such as using wavelet denoising to clean multi-source power grid data. The format conversion refers to the process of converting the CIM / XML files in the cleaned multi-source power grid data into graph structure data. The CIM / XML files are standardized data files used in the power system to describe power grid equipment, relationships, and attributes. The graph structure data is a data format that uses "points" and "lines" to represent the relationships between power grids, such as using substations and generators as "points" and line connections as "lines." This embodiment does not limit the specific implementation method of format conversion; those skilled in the art can freely choose according to actual needs, such as using CIMpy for format conversion. CIMpy stands for Common Information Model. ForPython, the Chinese name is Python Public Information Modeling Tool Library. Entity alignment refers to the process of matching the representation of location between real-time power grid data collected in real time by the monitoring and data acquisition system and geographic coordinate information collected by GIS. This embodiment does not limit the specific method of entity alignment. Those skilled in the art can freely choose according to actual needs, such as using Paython for entity alignment.

[0028] Specifically, step S1 involves fusing multi-source data from the power grid to remove noise and standardize file formats and coordinates, thereby improving the efficiency of constructing a dynamic knowledge graph based on standardized multi-source data packets from the power grid.

[0029] Specifically, when step S2 constructs a dynamic knowledge graph based on standardized power grid multi-source data packets using a dynamic knowledge graph construction method, the dynamic knowledge graph construction method includes: Step B01: Define entities for the standardized power grid multi-source data packets to obtain the defined set of entity points; Step B02: Construct connection relationships based on the defined set of entity points to obtain the basic graph; Step B03: Generate a dynamic chart based on the basic knowledge graph to obtain a dynamic chart, and use the dynamic chart as a dynamic knowledge graph.

[0030] Specifically, the entity definition refers to the process of defining devices, attributes, and events in standardized power grid multi-source data packets. For example, defining devices as transformers and wires, attributes as transformer capacity and wire length, and events as faults and maintenance records. This embodiment does not limit the specific implementation method of entity definition; those skilled in the art can freely choose according to actual needs, such as automatic identification through Python code. The connection relationship refers to the process of constructing the relationship between devices, attributes, and events in the defined entity point set, such as "transformer - connection - wire". The dynamic chart refers to the chart obtained by presenting the basic graph in chart form. This embodiment does not limit the specific method of generating the dynamic chart; those skilled in the art can freely choose according to actual needs, such as generating the dynamic chart through Gephi, where Gephi refers to a relationship graph visualization tool.

[0031] Specifically, step S2 involves constructing a dynamic knowledge graph to visualize the relationships between devices in the power grid. The dynamic knowledge graph is then updated in real time using multi-source data from the power grid, dynamically reflecting the power grid situation and thus improving the efficiency of power grid situation perception.

[0032] Specifically, when step S3 constructs the situation awareness SVG model based on the dynamic knowledge graph using the situation awareness SVG model construction method, the situation awareness SVG model construction method includes: Step C01: Input the geographic coordinate information and power grid topology model into the dynamic knowledge graph to obtain the device coordinates and device connection patterns output by the dynamic knowledge graph; Step C02: Perform vector conversion on the device connection configuration to obtain the device vector symbol. Step C03: Input the device coordinates and device vector symbols into the SVG construction tool to obtain the situational awareness SVG model.

[0033] Specifically, the device coordinates refer to the coordinates of all devices in the power grid obtained by inputting geographic coordinate information into a dynamic knowledge graph. The device connection form refers to the connection form and status of all devices in the power grid obtained by inputting the power grid topology model into the dynamic knowledge graph. The vector conversion refers to the process of converting the device connection form into vector graphics for representation. This embodiment does not limit the specific implementation method of vector conversion. Those skilled in the art can freely choose according to actual needs, such as performing vector conversion through symbol layer rules in GIS. The SVG construction tool refers to existing tools and software for constructing situational awareness SVG models based on device coordinates and device vector symbols. This embodiment does not limit the SVG construction tool. Those skilled in the art can freely choose according to actual needs, such as the SVG generation tool in Paython.

[0034] Specifically, step S3 involves constructing a situational awareness SVG model to enable real-time perception of the power grid situation directly based on the situational awareness SVG model, thereby improving the efficiency of power grid situation perception.

[0035] Specifically, in step S3, when optimizing the output process of the situational awareness SVG model based on the device spacing, the device spacing CX is compared with the preset device spacing CX0. The safety of the device spacing is judged based on the comparison result, and the output process of the situational awareness SVG model is optimized based on the judgment result. Wherein: When CX≤CX0, the safety status of the device spacing is determined to be safe, and distance optimization is not performed on the output process of the situational awareness SVG model; When CX > CX0, the safety status of the device spacing is determined to be unsafe. Distance optimization is performed on the output process of the situational awareness SVG model: the device coordinates are aligned to obtain optimized device coordinates. The optimized device coordinates are then replaced with the optimized device coordinates. The device coordinates and device vector symbols are then re-input into the SVG construction tool to obtain the situational awareness SVG model.

[0036] Specifically, the device spacing refers to the distance between operating power grid devices. This embodiment does not limit the specific method of obtaining the device spacing, such as based on the first device coordinate Ap ( , , ) and second device coordinates Bp ( , , Obtain and set the device spacing CX. The device spacing is obtained. The preset device spacing refers to a preset value for judging the safety of the device spacing. This embodiment does not limit the specific value of the preset device spacing CX0. Those skilled in the art can freely choose according to actual needs. For example, in this embodiment, CX0 is set to 7.95. The safety of the device spacing refers to the degree of safety of the device spacing judged based on the device spacing and the preset device spacing. The safety of the device spacing includes safe and unsafe. The coordinate alignment processing refers to the operation process of eliminating small deviations and irregularities in the device coordinates, such as performing coordinate alignment processing through Protodyakonov analysis.

[0037] Specifically, step S3 involves judging the safety status of the device spacing. When the device spacing is deemed unsafe, the device coordinates are aligned to increase the accuracy of the situational awareness SVG model, thereby improving the environmental adaptability and safety of the situational awareness SVG model.

[0038] Specifically, in step S4, when adjusting the load based on the distance optimization process according to the load factor, the load factor FLP is calculated based on the actual load AL and rated load ALP from the multi-source data of the power grid. FLP is set to AL / ALP to obtain the load factor FLP. The load factor FLP is then compared with the preset load factor FLP0. Based on the comparison result, the state of the load factor is judged, and the load is adjusted according to the preset equipment spacing CX0 based on the judgment result. When FLP≤FLP0, the load rate is determined to be normal, and no load adjustment is performed on the preset equipment spacing CX0; When FLP > FLP0, the load rate is determined to be abnormal. The load is adjusted according to the preset equipment spacing CX0 based on the adjustment coefficient tz, which is set to tz = 0.67 + 0.22 × e. -(FLP-FLP0) Where e is the base of the natural logarithm, the adjusted preset device spacing CX0` is obtained, CX0` is set to CX0×tz, the preset device spacing CX0 is replaced with the adjusted preset device spacing CX0`, and the device spacing CX is re-compared with the preset device spacing CX0.

[0039] Specifically, the preset load rate refers to a preset value for judging the state of the load rate. This embodiment does not limit the specific value of the preset load rate. Those skilled in the art can freely choose according to actual needs. For example, in this embodiment, the preset load rate FLP0 is set to 90%. The state of the load rate refers to the normality of the load rate judged based on the load rate and the preset load rate. The state of the load rate includes normal and abnormal.

[0040] Specifically, step S4 involves judging the state of the load factor. When the load factor is abnormal, the preset equipment spacing CX0 is reduced according to an adjustment coefficient that decreases from 0.89 to 0.67. The constant term 0.67 of the adjustment coefficient is the minimum value that the adjustment coefficient can reach, and the constant coefficient 0.22 is the amplitude of the adjustment coefficient change. This is to reasonably reduce the impact of the load factor on the safety judgment of equipment spacing, thereby improving the efficiency of power grid situation awareness.

[0041] Specifically, in step S4, when updating the load based on the three-phase imbalance, the three-phase imbalance SV is compared with the preset three-phase imbalance SV0. Based on the comparison result, the three-phase imbalance is judged, and based on the judgment result, the preset load factor FLP0 is updated. When SV≤SV0, the three-phase imbalance is considered to be within the acceptable range, and no load update is performed on the preset load rate FLP0. When SV > SV0, the three-phase imbalance is deemed substandard. The preset load factor FLP0 is then updated based on the update factor gz, which is set to gz = 0.72 + 0.23 × e. -(SV-SV0) Where e is the base of the natural logarithm, the updated preset load rate FLP0` is obtained, and FLP0` is set to FLP0×gz. The preset load rate FLP0 is replaced with the updated preset load rate FLP0`, and the load rate FLP is compared with the preset load rate FLP0 again.

[0042] Specifically, the preset three-phase imbalance degree refers to a preset value for judging the three-phase imbalance degree. This embodiment does not limit the specific value of the preset three-phase imbalance degree. Those skilled in the art can freely choose according to actual needs. For example, in this embodiment, the preset three-phase imbalance degree SV0 is set to 15%. The three-phase imbalance degree refers to the degree of compliance of the three-phase imbalance degree judged based on the three-phase imbalance degree and the preset three-phase imbalance degree. The three-phase imbalance degree includes compliance and non-compliance.

[0043] Specifically, step S4 involves judging the three-phase imbalance. When the three-phase imbalance is below standard, the preset load rate is steadily reduced based on an update coefficient that decreases from 0.95 to 0.72. The constant term 0.72 of the update coefficient is the minimum value that the update coefficient can reach, and the constant coefficient 0.23 represents the range of change of the update coefficient. This is to reasonably reduce the impact of excessively high three-phase imbalance on the accuracy of load rate status judgment, thereby improving the accuracy of power grid situation awareness.

[0044] Specifically, in step S5, when acquiring abnormal load areas based on load rates and performing cause analysis on these areas, the load rate FLP is compared with a preset high load rate FLP1. Based on the comparison result, the load attributes of the load rate are determined, and the abnormal load areas are output based on the determination result. When FLP≤FLP1, the load attribute of the load factor is determined to be overloaded, and no output is output for abnormal load areas; When FLP > FLP1, the load attribute of the load rate is determined to be overload, the abnormal load area is output, and the cause analysis of the abnormal load area is performed.

[0045] Specifically, the preset high load rate refers to a preset value for judging the load attribute of the load rate. This embodiment does not limit the specific value of the preset high load rate. Those skilled in the art can freely choose according to actual needs. For example, in this embodiment, the preset high load rate FLP1 is set to 92%. The load attribute of the load rate refers to the overload degree of the load rate judged based on the load rate and the preset high load rate. The load attribute of the load rate includes overload and no overload.

[0046] Specifically, step S5 involves determining the load attributes of the load rate to obtain the load area attributes, so as to output the load cause analysis results based on the load area attributes, thereby improving the accuracy and efficiency of power grid situation awareness.

[0047] Specifically, when step S5 performs cause analysis on the abnormal load area according to the cause analysis method, the cause analysis method includes: Step D01: Perform vector transformation on abnormal current, abnormal voltage, and abnormal equipment temperature to obtain current vector value OP, voltage vector value KP, and equipment temperature vector value WP. Step D02: Calculate the current evaluation value OPAU based on the current vector value OP and the current mean value OPA, and set OPAU = OP / OPA; Step D03: Calculate the voltage evaluation value KPAU based on the voltage vector value KP and the voltage mean KPA, and set KPAU = KP / KPA; Step D04: Calculate the equipment temperature assessment value WPAU based on the equipment temperature vector value WP and the equipment temperature mean value WPA, and set WPAU = WP / WPA; Step D05: Calculate the abnormal descent assessment value ZSDF based on the current assessment value OPAU, voltage assessment value KPAU, equipment temperature assessment value WPAU, current assessment weight w1, voltage assessment weight w2, equipment temperature assessment weight w3, and load factor v. Set ZSDF = (OPAU × w1 + KPAU × w2 + WPAU × w3) / v to obtain the abnormal descent assessment value ZSDF. Step D06: Compare the abnormal drop assessment value ZSDF with the preset abnormal drop assessment value ZSDF0. Determine the status of the abnormal drop assessment value based on the comparison result, and output the load cause analysis result based on the determination result. Wherein: When ZSDF≤ZSDF0, the abnormal drop evaluation value is determined to be normal, and non-fault causes are output as load cause analysis results. When ZSDF > ZSDF0, the abnormal drop evaluation value is determined to be abnormal, and the cause of the fault is output as the result of the load cause analysis.

[0048] Specifically, the abnormal current refers to the current value in the abnormal load area, the abnormal voltage refers to the voltage value in the abnormal load area, and the abnormal equipment temperature refers to the equipment temperature value in the abnormal load area. This embodiment does not limit the specific methods for obtaining the abnormal current, abnormal voltage, and abnormal equipment temperature. Those skilled in the art can freely choose according to actual needs, such as obtaining the abnormal current through a current sensor, obtaining the abnormal voltage through a voltage sensor, or obtaining the abnormal equipment temperature through a temperature sensor. The vector conversion refers to the process of converting the abnormal current, abnormal voltage, and abnormal equipment temperature into vector values ​​reflecting the characteristics of the abnormal current, abnormal voltage, and abnormal equipment temperature, respectively. This embodiment does not limit the specific method of vector transformation. Those skilled in the art can freely choose according to actual needs, such as using Fourier transform for vector transformation. The current vector value refers to the vector value describing the characteristics of the abnormal current obtained after vector transformation; the voltage vector value refers to the vector value describing the characteristics of the abnormal voltage obtained after vector transformation; the equipment temperature vector value refers to the vector value describing the characteristics of the abnormal equipment temperature obtained after vector transformation; the average current refers to the average value of historical currents; the average voltage refers to the average value of historical voltages; and the average equipment temperature refers to the average value of historical equipment temperatures. This embodiment does not limit the average current value, The specific methods for obtaining the average voltage and average equipment temperature are not limited, and those skilled in the art can freely choose according to actual needs. For example, the average current, average voltage, and average equipment temperature can be obtained through the power grid monitoring and data acquisition system. The current assessment weight refers to a coefficient that measures the importance of the current assessment value in the abnormal descent assessment value. The voltage assessment weight refers to a coefficient that measures the importance of the voltage assessment value in the abnormal descent assessment value. The equipment temperature assessment weight refers to a coefficient that measures the importance of the equipment temperature assessment value in the abnormal descent assessment value. The load factor refers to a coefficient that corrects for the influence of power grid load during the calculation of the abnormal descent assessment value. This embodiment does not specify the specific acquisition of the load factor. The method is limited, such as obtaining the load factor through expert evaluation. Expert evaluation refers to the process by which an expert with the ability to set the load factor sets the load factor. This embodiment does not limit the specific method of obtaining the load factor set by the expert; those skilled in the art can freely choose according to actual needs. For example, the load factor input by the expert in the interactive window can be obtained through the cloud. The preset abnormal drop evaluation value refers to a preset value used to judge the state of the abnormal drop evaluation value. This embodiment does not limit the specific value setting of the preset abnormal drop evaluation value ZSDF0; those skilled in the art can freely choose according to actual needs. For example, in this embodiment, ZSDF0 is set to 0.66. The state of the abnormal drop assessment value refers to the degree of normality of the abnormal drop assessment value judged based on the abnormal drop assessment value and the preset abnormal drop assessment value. The state of the abnormal drop assessment value includes normal and abnormal.

[0049] Specifically, step S5 involves analyzing the causes of abnormal load areas to determine whether there are faults in the abnormal load areas, so as to correct the situational awareness SVG model and improve the accuracy of power grid situational awareness.

[0050] Specifically, step S5 corrects the distance optimization based on the load cause analysis results. When the fault cause is output as a load cause analysis result, the fault assessment probability model is input based on the fault triggering frequency to obtain the fault triggering probability GS output by the fault assessment probability model. The fault triggering probability GS is compared with the preset fault triggering probability GS0. The state of the fault triggering probability is judged based on the comparison result, and the preset equipment spacing CX0 is corrected based on the judgment result, wherein: When GS≤GS0, the fault trigger probability is determined to be normal, and the preset device spacing CX0 is not corrected. When GS > GS0, the fault trigger probability is determined to be abnormal. The preset equipment spacing CX0 is then corrected according to the correction coefficient qc, which is set to qc = 0.77 + 0.21 × e -(GS-GS0) Where e is the base of the natural logarithm, the corrected preset device spacing CX0` is obtained, CX0` is set to CX0×qc, the preset device spacing CX0 is replaced with the corrected preset device spacing CX0`, and the device spacing CX is re-compared with the preset device spacing CX0.

[0051] Specifically, the fault trigger frequency refers to the number of times cause analysis is performed within a preset time period. This embodiment does not limit the preset time; for example, this embodiment sets the preset time t1 to 1 hour. This embodiment does not limit the specific method for obtaining the fault trigger frequency; those skilled in the art can freely choose according to actual needs. For example, the fault trigger frequency GM = zn / t1 can be obtained by using system logs to obtain the number of cause analyses zn within the preset time period. The fault assessment probability model refers to a recurrent neural network model that uses the fault trigger frequency as input data and the fault trigger probability as output data. This embodiment does not limit the specific construction method of the fault assessment probability model; those skilled in the art can freely choose according to actual needs. For example, the fault trigger frequency can be obtained by using historical fault analysis logs to obtain the number of cause analyses zn within the preset time period. The frequency of fault triggering and its corresponding probability of fault triggering are used as training datasets to train a recurrent neural network model to obtain a fault assessment probability model. The fault triggering probability refers to the probability of a single triggering cause analysis obtained from the fault assessment probability model. The preset fault triggering probability refers to a preset value for judging the state of the fault triggering probability. This embodiment does not limit the specific value of the preset fault triggering probability. Those skilled in the art can freely choose according to actual needs. For example, in this embodiment, the preset fault triggering probability GS0 is set to 30%. The state of the fault triggering probability refers to the degree of normality of the fault triggering probability judged based on the fault triggering probability and the preset fault triggering probability. The state of the fault triggering probability includes normal and abnormal.

[0052] Specifically, step S5 involves judging the state of the fault trigger probability. When the state of the fault trigger probability is abnormal, the preset equipment spacing is smoothly reduced according to a correction coefficient that decreases from 0.98 to 0.77. The constant term 0.77 of the correction coefficient is the minimum value that the correction coefficient can reach, and the constant coefficient 0.21 represents the change range of the correction coefficient. This is to reasonably reduce the impact of the fault trigger probability on the safety judgment of equipment spacing, thereby improving the efficiency and accuracy of power grid situation awareness.

[0053] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. An SVG power grid construction and situational awareness method supporting multi-source data fusion, characterized in that, The method includes: Step S1: Collect multi-source data from the power grid; Step S2: The multi-source data of the power grid is fused to obtain a standardized multi-source data package of the power grid, and a dynamic knowledge graph is constructed based on the standardized multi-source data package of the power grid to obtain the dynamic knowledge graph; Step S3: Construct the situational awareness SVG model based on the dynamic knowledge graph, output the situational awareness SVG model, and optimize the output process of the situational awareness SVG model according to the device spacing. Step S4 involves adjusting the load based on the load factor during the distance optimization process, and updating the load based on the three-phase imbalance during the load adjustment process. Step S5: Obtain the abnormal load area based on the load rate, perform cause analysis on the abnormal load area, obtain the load cause analysis results, and correct the distance optimization based on the load cause analysis results.

2. The SVG power grid construction and situational awareness method supporting multi-source data fusion according to claim 1, characterized in that, Step S1 involves collecting multi-source data from the power grid, including real-time power grid data, geographic coordinate information, power grid topology model, three-phase imbalance, actual load, and rated load. When step S2 performs fusion processing on the multi-source data of the power grid according to the fusion processing method, the fusion processing method includes: Step A01: Perform data cleaning on the multi-source data of the power grid to obtain the cleaned multi-source data of the power grid; Step A02: Convert the format of the cleaned multi-source power grid data to obtain unified multi-source power grid data; Step A03: Perform entity alignment on the unified power grid multi-source data to obtain a standardized power grid multi-source data packet.

3. The SVG power grid construction and situational awareness method supporting multi-source data fusion according to claim 2, characterized in that, When step S2 constructs a dynamic knowledge graph based on standardized power grid multi-source data packets using a dynamic knowledge graph construction method, the dynamic knowledge graph construction method includes: Step B01: Define entities for the standardized power grid multi-source data packets to obtain the defined set of entity points; Step B02: Construct connection relationships based on the defined set of entity points to obtain the basic graph; Step B03: Generate a dynamic chart based on the basic knowledge graph to obtain a dynamic chart, and use the dynamic chart as a dynamic knowledge graph.

4. The SVG power grid construction and situational awareness method supporting multi-source data fusion according to claim 3, characterized in that, When the situational awareness SVG model is constructed based on the dynamic knowledge graph using the situational awareness SVG model construction method in step S3, the situational awareness SVG model construction method includes: Step C01: Input the geographic coordinate information and power grid topology model into the dynamic knowledge graph to obtain the device coordinates and device connection patterns output by the dynamic knowledge graph; Step C02: Perform vector conversion on the device connection configuration to obtain the device vector symbol. Step C03: Input the device coordinates and device vector symbols into the SVG construction tool to obtain the situational awareness SVG model.

5. The SVG power grid construction and situational awareness method supporting multi-source data fusion according to claim 4, characterized in that, In step S3, when optimizing the output process of the situational awareness SVG model based on the device spacing, the device spacing CX is compared with the preset device spacing CX0. The safety of the device spacing is judged based on the comparison result, and the output process of the situational awareness SVG model is optimized based on the judgment result. Specifically: When CX≤CX0, the safety status of the device spacing is determined to be safe, and distance optimization is not performed on the output process of the situational awareness SVG model; When CX > CX0, the safety status of the device spacing is determined to be unsafe. Distance optimization is performed on the output process of the situational awareness SVG model: the device coordinates are aligned to obtain optimized device coordinates. The optimized device coordinates are then replaced with the optimized device coordinates. The device coordinates and device vector symbols are then re-input into the SVG construction tool to obtain the situational awareness SVG model.

6. The SVG power grid construction and situational awareness method supporting multi-source data fusion according to claim 5, characterized in that, In step S4, when adjusting the load based on the distance optimization process according to the load factor, the load factor FLP is calculated based on the actual load AL and rated load ALP from the multi-source data of the power grid. FLP is set to AL / ALP to obtain the load factor FLP. The load factor FLP is then compared with the preset load factor FLP0. Based on the comparison result, the state of the load factor is judged, and the load is adjusted according to the preset equipment spacing CX0 based on the judgment result. Wherein: When FLP≤FLP0, the load rate is determined to be normal, and no load adjustment is performed on the preset equipment spacing CX0; When FLP > FLP0, the load rate is determined to be abnormal. The load is adjusted according to the preset equipment spacing CX0 based on the adjustment coefficient tz, which is set to tz = 0.67 + 0.22 × e. -(FLP-FLP0) Where e is the base of the natural logarithm, the adjusted preset device spacing CX0` is obtained, CX0` is set to CX0×tz, the preset device spacing CX0 is replaced with the adjusted preset device spacing CX0`, and the device spacing CX is re-compared with the preset device spacing CX0.

7. The SVG power grid construction and situational awareness method supporting multi-source data fusion according to claim 6, characterized in that, In step S4, during the load update process based on the three-phase unbalance, the three-phase unbalance SV is compared with the preset three-phase unbalance SV0. The three-phase unbalance is assessed based on the comparison result, and the preset load factor FLP0 is updated based on the assessment result. When SV≤SV0, the three-phase imbalance is considered to be within the acceptable range, and no load update is performed on the preset load rate FLP0. When SV > SV0, the three-phase imbalance is deemed substandard. The preset load factor FLP0 is then updated based on the update factor gz, which is set to gz = 0.72 + 0.23 × e. -(SV-SV0) Where e is the base of the natural logarithm, the updated preset load rate FLP0` is obtained, and FLP0` is set to FLP0×gz. The preset load rate FLP0 is replaced with the updated preset load rate FLP0`, and the load rate FLP is compared with the preset load rate FLP0 again.

8. The SVG power grid construction and situational awareness method supporting multi-source data fusion according to claim 7, characterized in that, In step S5, when acquiring abnormal load areas based on load rates and performing cause analysis on these areas, the load rate FLP is compared with a preset high load rate FLP1. Based on the comparison result, the load attributes of the load rate are determined, and the abnormal load areas are output based on the determination result. Specifically: When FLP≤FLP1, the load attribute of the load factor is determined to be overloaded, and no output is output for abnormal load areas; When FLP > FLP1, the load attribute of the load rate is determined to be overload, the abnormal load area is output, and the cause analysis of the abnormal load area is performed.

9. The SVG power grid construction and situational awareness method supporting multi-source data fusion according to claim 8, characterized in that, When performing cause analysis on the abnormal load area in step S5 according to the cause analysis method, the cause analysis method includes: Step D01: Perform vector transformation on abnormal current, abnormal voltage, and abnormal equipment temperature to obtain current vector value OP, voltage vector value KP, and equipment temperature vector value WP. Step D02: Calculate the current evaluation value OPAU based on the current vector value OP and the current mean value OPA, and set OPAU = OP / OPA; Step D03: Calculate the voltage evaluation value KPAU based on the voltage vector value KP and the voltage mean KPA, and set KPAU = KP / KPA; Step D04: Calculate the equipment temperature assessment value WPAU based on the equipment temperature vector value WP and the equipment temperature mean value WPA, and set WPAU = WP / WPA; Step D05: Calculate the abnormal descent assessment value ZSDF based on the current assessment value OPAU, voltage assessment value KPAU, equipment temperature assessment value WPAU, current assessment weight w1, voltage assessment weight w2, equipment temperature assessment weight w3, and load factor v. Set ZSDF = (OPAU × w1 + KPAU × w2 + WPAU × w3) / v to obtain the abnormal descent assessment value ZSDF. Step D06: Compare the abnormal drop assessment value ZSDF with the preset abnormal drop assessment value ZSDF0. Determine the status of the abnormal drop assessment value based on the comparison result, and output the load cause analysis result based on the determination result. Wherein: When ZSDF≤ZSDF0, the abnormal drop evaluation value is determined to be normal, and non-fault causes are output as load cause analysis results. When ZSDF > ZSDF0, the abnormal drop evaluation value is determined to be abnormal, and the cause of the fault is output as the result of the load cause analysis.

10. The SVG power grid construction and situational awareness method supporting multi-source data fusion according to claim 9, characterized in that, Step S5 corrects the distance optimization based on the load cause analysis results. When the fault cause is output as a load cause analysis result, the fault assessment probability model is input according to the fault trigger frequency to obtain the fault trigger probability GS output by the fault assessment probability model. The fault trigger probability GS is compared with the preset fault trigger probability GS0. The state of the fault trigger probability is judged according to the comparison result, and the preset equipment spacing CX0 is corrected according to the judgment result, wherein: When GS≤GS0, the fault trigger probability is determined to be normal, and the preset device spacing CX0 is not corrected. When GS > GS0, the fault trigger probability is determined to be abnormal. The preset equipment spacing CX0 is then corrected according to the correction coefficient qc, which is set to qc = 0.77 + 0.21 × e -(GS-GS0) Where e is the base of the natural logarithm, the corrected preset device spacing CX0` is obtained, CX0` is set to CX0×qc, the preset device spacing CX0 is replaced with the corrected preset device spacing CX0`, and the device spacing CX is re-compared with the preset device spacing CX0.

Images