Power grid state evaluation method and system applying multi-source data fusion

Abstract

The invention discloses a power grid state evaluation method and system applying multi-source data fusion, and relates to the technical field of data analysis, and the method comprises the steps: carrying out the expansion of a to-be-evaluated shaft segment through combining the dynamic characteristics of a power grid and the correlation degree, and enabling the state of the power grid under the to-be-evaluated shaft segment to be always related to the data at the previous time, a reasonable forward time node is determined by combining the dynamic characteristics of the power grid and the correlation degree, and data expansion is performed on a to-be-evaluated shaft segment, so that the comprehensiveness and rationality of data are ensured, and a reliable basis is provided for subsequent multi-source information feature extraction. Uncertain information is fused through a combination rule, and the problem that the boundary state is misjudged by a traditional threshold method is solved. The evidence theory can process conflicts among multi-source information, and a single data source is prevented from leading an evaluation result through a discount coefficient and a dynamic evidence weight. The accuracy and adaptability of power grid state evaluation are improved, and the reasonability of power grid state monitoring and the stability and safety of the power grid operation state are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a power grid state assessment method and system using multi-source data fusion. Background Art

[0002] As new power systems evolve toward the characteristics of "double highs" (a high proportion of renewable energy and power electronic equipment) and "double randomness" (randomness of load and randomness of renewable energy output), the operating state of the power grid exhibits strong uncertainty, multi-scale coupling at multiple spatiotemporal scales, and dynamic evolution. Traditional single-data source assessment methods (such as topological analysis based on SCADA measurements and phasor measurement based on PMUs) struggle to accurately capture hidden faults, parameter drift, and cascading risks under complex operating conditions due to their single data dimension and poor anti-interference capabilities. Multi-source data fusion technology, by integrating the complementary features of heterogeneous data sources (such as wide-area measurement data, equipment sensor signals, meteorological and environmental information, historical operation and maintenance records, etc.), combined with algorithms such as evidence theory, Bayesian networks, and deep learning, can overcome the perception limitations of a single data source and achieve cross-domain correlation modeling and dynamic assessment of power grid status.

[0003] In the existing technology, the correlation between multi-source data and the temporal correlation of data are not taken into account, resulting in poor accuracy and adaptability of power grid status assessment, failure to ensure the rationality of power grid status monitoring, and poor power grid task efficiency.

[0004] Therefore, how to improve the accuracy and adaptability of power grid status assessment is a technical problem that needs to be solved. Summary of the Invention

[0005] The purpose of the present invention is to solve the problem of poor accuracy and adaptability of power grid status assessment in the prior art, and to propose a power grid status assessment method using multi-source data fusion, which includes: Collect multi-source information of the power grid, establish a time axis of the multi-source information of the power grid according to the timestamps of the multi-source information, determine the time period to be evaluated, and mark the axis segment to be evaluated on the time axis of the multi-source information of the power grid; Determine the dynamic characteristics of the power grid under the axis segment to be evaluated, analyze the correlation degree of multi-source information on the axis segment to be evaluated on the time axis of the multi-source information of the power grid, and expand the axis segment to be evaluated based on the dynamic characteristics of the power grid and the correlation degree; Define the grid status quality level based on the time axis of the grid multi-source information, and extract the multi-source information features based on the expanded axis segment to be evaluated; According to the quality level of power grid status, an identification framework of evidence theory is constructed, and the basic probability of evidence theory is constructed. Multi-source information features are input into the evidence theory to identify the quality level of power grid status in the time period to be evaluated.

[0006] In some embodiments of the present application, determining the dynamic characteristics of the power grid under the shaft segment to be evaluated includes: The dynamic characteristics of the power grid include two dynamic modes: transient mode and steady-state mode; The time domain features and frequency domain features of the shaft segment to be evaluated are extracted, the dynamic mode is judged according to the range of the time domain features and frequency domain features, and the dynamic value of the dynamic mode is determined by combining the time domain features and frequency domain features.

[0007] In some embodiments of the present application, analyzing the correlation degree of multi-source information on the axis segment to be evaluated on the time axis of the multi-source information of the power grid includes: The multi-source information is classified into data categories. Based on the time axis of the power grid multi-source information, the autocorrelation function between the axis segment to be evaluated and the similar data of the axis segment in the previous time is calculated, and the autocorrelation curve that changes with time is plotted. The median and maximum values are determined on the autocorrelation curve to determine the strong correlation interval on the autocorrelation curve. The autocorrelation time node is determined based on the strong correlation interval; Integrate the autocorrelation time nodes of each type of data to obtain an autocorrelation time node set, and determine a first forward time point according to the autocorrelation time node set; Determining dependent variable elements on the axis segment to be evaluated, where the dependent variable elements are data or events, and determining independent variables corresponding to each dependent variable element, calculating a cross-correlation function between the independent variable and the dependent variable elements based on an axis segment range between a first forward time point and the axis segment to be evaluated, and plotting a cross-correlation curve graph that changes over time; The mode, original mean, standard deviation and maximum value are counted on the cross-correlation curve graph, the average of the mode and maximum value is calculated based on the two, the multiple is determined based on the average value, and the threshold is determined by combining the original mean, standard deviation and multiple; The strong correlation area on the cross-correlation curve is divided by the threshold, and the area of the strong correlation area is calculated to determine the correlation degree corresponding to each dependent variable element.

[0008] In some embodiments of the present application, the axis segments to be evaluated are expanded in combination with the dynamic characteristics of the power grid and the degree of correlation, including: The correlation degree corresponding to all dependent variable elements is comprehensively considered, and the second forward time point is determined according to the dynamic mode type, dynamic value and correlation degree. The axis segment to be evaluated is expanded based on the second forward time point.

[0009] In some embodiments of the present application, the grid status quality level is defined based on the time axis of the grid multi-source information, including: The multi-source information includes the power grid's generation side data, transmission side data, and distribution side data. The power grid's generation side data, transmission side data, and distribution side data are extracted on the power grid's multi-source information timeline. Extract balance indicators and grid characteristics based on the power grid's generation, transmission, and distribution data; Input the grid characteristics into the preset grid unbalanced state model, output the initial state of the grid, and define the grid state quality level based on the initial state of the grid and the balance index; The initial state of the power grid describes the state of the power grid in an unbalanced direction, and the balance index describes the state of the power grid in a balanced direction.

[0010] In some embodiments of the present application, an identification framework of evidence theory is constructed according to the quality level of the power grid status, including: All values of the grid status quality level are determined, and a set of propositions of the evidence theory is allocated one by one according to the value range of the grid status quality level, and the proposition set includes all grid status quality levels.

[0011] In some embodiments of the present application, the basic probability of the theory of evidence is constructed, including: Normalize the multi-source features, divide the multi-source features into multiple evidences, define the membership function between each evidence and the proposition, and calculate the original BPA; Analyze multiple pieces of evidence to determine conflicting evidence, calculate the KL divergence of the conflicting evidence, determine a discount factor based on the KL divergence, adjust the original BPA based on the discount factor, and thus determine the BPA of each piece of evidence.

[0012] In some embodiments of the present application, inputting multi-source information features into the evidence theory includes: Statistics the real-time noise of each piece of evidence, calculate the cross-degree and mutual support between multiple pieces of evidence, and dynamically determine the fusion weight of each piece of evidence in the evidence fusion process based on the real-time noise, cross-degree and mutual support, to achieve the fusion of multiple pieces of evidence and output the proposition result.

[0013] Correspondingly, a power grid status assessment system using multi-source data fusion includes: The first module is used to collect multi-source information of the power grid, establish a time axis of the multi-source information of the power grid according to the timestamps of the multi-source information, determine the time period to be evaluated, and mark the axis segment to be evaluated on the time axis of the multi-source information of the power grid; The second module is used to determine the dynamic characteristics of the power grid under the axis segment to be evaluated, analyze the correlation degree of multi-source information on the axis segment to be evaluated on the time axis of the multi-source information of the power grid, and expand the axis segment to be evaluated based on the dynamic characteristics of the power grid and the correlation degree; The third module is used to define the grid status quality level according to the time axis of the grid multi-source information, and extract the multi-source information features based on the expanded axis segment to be evaluated; The fourth module is used to construct an identification framework of the evidence theory based on the quality level of the power grid status, and to construct the basic probability of the evidence theory. It inputs multi-source information features into the evidence theory to identify the quality level of the power grid status in the time period to be evaluated.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. Expand the segment to be evaluated based on the dynamic characteristics and correlation level of the power grid. The power grid status in the segment to be evaluated is often related to data from previous time periods. By combining the dynamic characteristics of the power grid and the degree of correlation, a reasonable forward time node is determined and data expansion is performed on the segment to be evaluated. This ensures the comprehensiveness and rationality of the data and provides a reliable foundation for subsequent multi-source information feature extraction. Define the grid status quality level, taking into account the balanced and unbalanced directions of the power grid to comprehensively define the grid status quality level, ensuring the feasibility and pertinence of the implementation of the evidence theory and adapting it to the actual operating conditions of the power grid.

[0015] 2. Constructing the basic probability of evidence theory, inputting multi-source information features into the evidence theory, and integrating uncertain information through the Dempster combination rule, addressing the problem of traditional threshold methods misjudging boundary states. Evidence theory can handle conflicts between multi-source information. By using discount coefficients and dynamic evidence weighting, it prevents a single data source from dominating the assessment results, improves decision-making robustness under complex operating conditions, and provides dispatchers with a quantifiable state confidence reference. This improves the accuracy and adaptability of power grid state assessment, ensuring the rationality of power grid state monitoring and the stability and security of power grid operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flow chart of a power grid status assessment method using multi-source data fusion proposed by the present invention; Figure 2 This is a structural diagram of a power grid status assessment system using multi-source data fusion proposed by the present invention. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0018] Reference Figure 1 , a power grid state assessment method using multi-source data fusion, comprising the following steps: Step S101 : collecting multi-source information of the power grid, establishing a time axis of the multi-source information of the power grid according to the timestamps of the multi-source information, determining a time period to be evaluated, and marking the axis segment to be evaluated on the time axis of the multi-source information of the power grid.

[0019] In this embodiment, multi-source information includes three types of data: power grid data from the generation, transmission, and distribution sides. For example, this includes SCADA measurements (voltage, current), PMU phasor data (frequency, phase angle), meteorological data (temperature, humidity), and equipment sensor signals (transformer oil temperature, circuit breaker status). Multi-source data is sorted by timestamp to construct a timeline for multi-source information. The time period to be evaluated (e.g., 10:00-10:10) is determined based on scheduling requirements and marked as the segment to be evaluated on the timeline.

[0020] It should be noted that existing technologies may only capture data from the time period to be evaluated (10:00-10:10) to analyze the power grid status between 10:00 and 10:10. However, power grid data is causal and time-sensitive. That is, the power grid status between 10:00 and 10:10 may be related to or affected by the power grid status before 10:00. This needs to be taken into consideration to ensure the reliability of power grid status assessment. Expanding the axis section to be evaluated exists precisely to solve these problems.

[0021] Step S102 : determining the grid dynamic characteristics of the axis segment to be evaluated, analyzing the correlation degree of multi-source information on the axis segment to be evaluated on the grid multi-source information time axis, and expanding the axis segment to be evaluated based on the grid dynamic characteristics and correlation degree.

[0022] In this embodiment, the degree of correlation between the dynamic characteristics of the power grid and the multi-source information (correlation between the forward time data and the data of the axis segment time to be evaluated) will affect the expansion demand, and the expansion time point is determined by considering both factors.

[0023] In some embodiments of the present application, determining the dynamic characteristics of the power grid under the shaft segment to be evaluated includes: The dynamic characteristics of the power grid include two dynamic modes: transient mode and steady-state mode; The time domain features and frequency domain features of the shaft segment to be evaluated are extracted, the dynamic mode is judged according to the range of the time domain features and frequency domain features, and the dynamic value of the dynamic mode is determined by combining the time domain features and frequency domain features.

[0024] In this embodiment, time-domain features (mean, variance) and frequency-domain features (FFT harmonic components) are extracted from the data within the evaluated axis segment. The transient mode is the non-stationary phase of the grid transitioning from its initial state to a new equilibrium when subjected to external disturbances (such as short circuits, renewable energy fluctuations) or sudden changes in internal parameters. The steady-state mode occurs when the grid reaches a new equilibrium state after normal operation or after a disturbance, with electrical quantities fluctuating within allowable limits and the system structure stable. The dynamic mode is determined based on the range of the time-domain and frequency-domain features. For example, if the time-domain variance exceeds a threshold and the frequency-domain higher-order harmonic ratio exceeds 10%, the transient mode (such as a short circuit fault) is determined. If the time-domain variance is less than a threshold and the frequency-domain fundamental ratio exceeds 90%, the steady-state mode (such as normal operation) is determined. For transient mode, the dynamic value = time-domain variance × frequency-domain higher-order harmonic ratio. For steady-state mode, the dynamic value = 1 / (time-domain variance + frequency-domain fundamental ratio). The dynamic value describes the degree of dynamic quantization of the grid.

[0025] In some embodiments of the present application, analyzing the correlation degree of multi-source information on the axis segment to be evaluated on the time axis of the multi-source information of the power grid includes: The multi-source information is classified into data categories. Based on the time axis of the power grid multi-source information, the autocorrelation function between the axis segment to be evaluated and the similar data of the axis segment in the previous time is calculated, and the autocorrelation curve that changes with time is plotted. The median and maximum values are determined on the autocorrelation curve to determine the strong correlation interval on the autocorrelation curve. The autocorrelation time node is determined based on the strong correlation interval; Integrate the autocorrelation time nodes of each type of data to obtain an autocorrelation time node set, and determine a first forward time point according to the autocorrelation time node set; Determining dependent variable elements on the axis segment to be evaluated, where the dependent variable elements are data or events, and determining independent variables corresponding to each dependent variable element, calculating a cross-correlation function between the independent variable and the dependent variable elements based on an axis segment range between a first forward time point and the axis segment to be evaluated, and plotting a cross-correlation curve graph that changes over time; The mode, original mean, standard deviation and maximum value are counted on the cross-correlation curve graph, the average of the mode and maximum value is calculated based on the two, the multiple is determined based on the average value, and the threshold is determined by combining the original mean, standard deviation and multiple; The strong correlation area on the cross-correlation curve is divided by the threshold, and the area of the strong correlation area is calculated to determine the correlation degree corresponding to each dependent variable element.

[0026] In this embodiment, the autocorrelation function (ACF) of the axis segment to be evaluated and similar data in the previous period (for example, ten minutes) is calculated, and an autocorrelation curve is drawn. The portion between the median and the maximum value is taken as the strong correlation interval, and the strong correlation interval is determined (such as ACF>0.7), and a set of autocorrelation time nodes (such as 10:00, 10:15) is extracted. A first forward time point is determined based on the set of autocorrelation time nodes, and the first forward time point is taken as the first forward time point in the set. This ensures maximum coverage, avoids omissions or missing of relevant data, and screens out strong correlation intervals, which can effectively reduce the amount of data processing and ensure analysis results.

[0027] In this embodiment, after analyzing the autocorrelation of the same type of data at different times, it is necessary to consider the cross-correlation between the segment to be evaluated and the previous time data (the first forward time point). The relationship between the independent and dependent variables can be determined based on power grid model theory or historical data. The dependent variable is the target data / event directly observed or calculated within the segment to be evaluated and is the core object of analysis. Examples include voltage sag amplitude (dependent variable: voltage amplitude), wind farm power fluctuation (dependent variable: output power), and fault trigger signals (dependent variable: protective device activation events). Independent variables are potential driving factors that influence changes in the dependent variable and may come from historical data or external inputs. Examples include renewable energy output (independent variable: photovoltaic / wind power), load switching (independent variable: user-side electricity consumption), and environmental disturbances (independent variable: wind speed / sun intensity). The cross-correlation function (CCF) calculates the average of the mode and maximum values on the graph. Different average values correspond to different multipliers. The threshold is determined by combining the original mean, standard deviation, and multiplier: threshold = mean ± 3 times the standard deviation. Each dependent variable element corresponds to an association relationship, that is, the degree of association. The area is calculated by numerical integration of the trapezoidal method, parameterized curve fitting, etc., so as to map the degree of association.

[0028] In some embodiments of the present application, the axis segments to be evaluated are expanded in combination with the dynamic characteristics of the power grid and the degree of correlation, including: The correlation degree corresponding to all dependent variable elements is comprehensively considered, and the second forward time point is determined according to the dynamic mode type, dynamic value and correlation degree. The axis segment to be evaluated is expanded based on the second forward time point.

[0029] In this embodiment, the second forward time point is determined according to the dynamic mode type, dynamic value and correlation degree. An initial time point is determined according to the correlation degree, and then a correction factor is determined according to the dynamic value. The mapping relationship of different dynamic mode types is different. The initial time point is corrected by the correction factor (correction factor * initial time point) to obtain the second forward time point, thereby expanding the axis segment to be evaluated.

[0030] Step S103 : defining the grid status quality level according to the grid multi-source information time axis, and extracting multi-source information features based on the expanded axis segment to be evaluated.

[0031] In this embodiment, the grid's status is comprehensively evaluated from two perspectives: one is the balance within the grid, and the other is the imbalanced state (situations or situations where the grid is balanced but not in good condition, such as the latent period of hidden faults, strong uncertainty impacts from renewable energy sources, and critical equipment overload conditions). The latent period of hidden faults refers to aging of internal insulation in equipment (such as partial discharge in transformers) and poor line contact, but does not cause a trip, while the power / voltage / frequency deviations all meet the standards. Strong uncertainty impacts from renewable energy sources refer to sudden drops in wind power / photovoltaic output (such as a 30% drop in 10 seconds), but rapid AGC adjustment maintains power balance, while maintaining real-time balance indicators within limits.

[0032] In some embodiments of the present application, the grid status quality level is defined based on the time axis of the grid multi-source information, including: The multi-source information includes the power grid's generation side data, transmission side data, and distribution side data. The power grid's generation side data, transmission side data, and distribution side data are extracted on the power grid's multi-source information timeline. Extract balance indicators and grid characteristics based on the power grid's generation, transmission, and distribution data; Input the grid characteristics into the preset grid unbalanced state model, output the initial state of the grid, and define the grid state quality level based on the initial state of the grid and the balance index; The initial state of the power grid describes the state of the power grid in an unbalanced direction, and the balance index describes the state of the power grid in a balanced direction.

[0033] In this embodiment, the unbalanced state model of the power grid may be a dynamic system model, a machine learning model, or the like, and the balance indicators include a voltage balance indicator, a power balance indicator, and a frequency balance indicator, specifically: Voltage balance indicator Voltage amplitude deviation: the deviation range of the three-phase voltage amplitude from the rated value.

[0034] Voltage unbalance: The ratio of the negative or zero sequence voltage component to the positive sequence voltage component.

[0035] Voltage fluctuation rate: the maximum rate of change of voltage amplitude per unit time.

[0036] Power balance indicators Active / reactive power balance rate: the deviation rate between actual output power and planned power.

[0037] Power factor: The ratio of active power to apparent power.

[0038] Reasonableness of power flow distribution: comparison between line load factor and thermal limit.

[0039] Frequency balance index Frequency deviation: The difference between the actual frequency and the rated frequency (50Hz or 60Hz).

[0040] Frequency fluctuation rate: the maximum rate of change of frequency per unit time.

[0041] Grid characteristics Topological characteristics Grid structure: radial, ring or mesh structure.

[0042] Node type: power generation node (PV / PQ node), load node, and balance node.

[0043] Line parameters: resistance, reactance, and admittance.

[0044] Operating status characteristics Load factor: The ratio of line or transformer load to rated capacity.

[0045] Flow direction: the direction and magnitude of power flow.

[0046] Switch status: the on / off status of the circuit breaker and disconnector.

[0047] Device status characteristics Equipment health: Assessment of equipment degradation based on online monitoring data.

[0048] Failure probability: prediction of equipment failure based on historical data.

[0049] Maintenance record: time and content of the equipment’s most recent maintenance.

[0050] Dynamic response characteristics Frequency response capability: the ability to regulate active power when the system frequency changes.

[0051] Voltage response capability: the ability to regulate reactive power when system voltage changes.

[0052] Transient stability: dynamic recovery capability after a fault.

[0053] In this embodiment, the formula for defining the grid status quality level based on the initial grid status and balance index is as follows: ; in, is the quality level of the power grid status, is the initial state of the power grid, To balance the number of indicators, For the The combined weight of the balance indicators, For the The size of the balance indicator, for The maximum value in is the first constant, is the second constant, It represents the correction of the maximum balance situation to the balance average value, and its value range is between 0.879-1.211. The first constant is used to balance the size of the correction function, and the second constant is used to balance the size of the grid status quality level. [] is the rounding symbol.

[0054] Step S104 , constructing an identification framework of the evidence theory based on the grid status quality level, constructing the basic probability of the evidence theory, inputting multi-source information features into the evidence theory, and identifying the grid status quality level in the time period to be evaluated.

[0055] In this example, the Dempster-Shafer (DS) evidence theory is an effective tool for handling uncertainty and incomplete information. The basic probability assignment (BPA) is the core of the DS evidence theory and represents the degree of confidence in each hypothesis (or proposition). When constructing a BPA, grid characteristics, balance indicators, and expert knowledge must be integrated. The process is: 1. Constructing an identification framework, 2. Constructing a basic probability assignment (BPA), 3. Evidence synthesis, and 4. State identification.

[0056] In some embodiments of the present application, an identification framework of evidence theory is constructed according to the quality level of the power grid status, including: All values of the grid status quality level are determined, and a set of propositions of the evidence theory is allocated one by one according to the value range of the grid status quality level, and the proposition set includes all grid status quality levels.

[0057] In this embodiment, the grid status quality level is a proposition, and the proposition is the output result of the evidence theory.

[0058] In some embodiments of the present application, the basic probability of the theory of evidence is constructed, including: Normalize the multi-source features, divide the multi-source features into multiple evidences, define the membership function between each evidence and the proposition, and calculate the original BPA; Analyze multiple pieces of evidence to determine conflicting evidence, calculate the KL divergence of the conflicting evidence, determine a discount factor based on the KL divergence, adjust the original BPA based on the discount factor, and thus determine the BPA of each piece of evidence.

[0059] In this embodiment, the normalized features are divided into multiple pieces of evidence, and a membership function with the proposition is defined for each piece of evidence. For example, evidence is divided according to feature type, load rate, voltage imbalance, power factor, etc. The membership function is defined using fuzzy logic or expert knowledge. Conflicting evidence refers to evidence with large differences in support for the same proposition. KL divergence is used to measure the difference between two probability distributions. The KL divergence between all pieces of evidence is calculated, and conflicting evidence is identified. The discount factor is inversely proportional to the KL divergence, specifically: ; in, is the discount factor, is the adjustment coefficient, is the KL divergence, is the adjusted BPA, For the original BPA, The full set BPA (i.e., “uncertain” or “unknown” assignment) for the identification framework reflects the overall uncertainty of the evidence.

[0060] In some embodiments of the present application, inputting multi-source information features into the evidence theory includes: Statistics the real-time noise of each piece of evidence, calculate the cross-degree and mutual support between multiple pieces of evidence, and dynamically determine the fusion weight of each piece of evidence in the evidence fusion process based on the real-time noise, cross-degree and mutual support, to achieve the fusion of multiple pieces of evidence and output the proposition result.

[0061] In this embodiment, the evidence, i.e., the noise in the data, the cross-correlation between evidence, and the mutual support between evidence, affect the weights in the evidence fusion process and are therefore dynamically set. Evidence is divided according to the fusion method (different evidence combinations). For each evidence combination, the Euclidean distance (similarities) between one piece of evidence and the other pieces of evidence is calculated to determine the cross-correlation. Mutual support is the support of evidence source 1 for another evidence source 2, which is the degree of overlap of its BPA on the main proposition of the other evidence source 2. A weighted sum of the cross-correlation and mutual support is performed. Different real-time noise levels are corresponding to different adjustment factors. The weights are adjusted by adjusting the weighted summation result (multiplying it by the adjustment factor).

[0062] Understandably, in Dempster-Shafer Theory, the discount coefficient and combination weight are two distinct mechanisms, operating at the evidence preprocessing and fusion stages, respectively. While their objectives are related, they are not inherently in conflict. The discount coefficient, used in the evidence preprocessing stage, adjusts the BPA of individual evidence sources before fusion, downweighting highly conflicting or unreliable evidence. The combination weight, used in the evidence fusion stage, assigns the relative importance of each evidence source when combining multiple sources, adjusting its contribution to the fusion formula and directly reflecting its influence on decision-making.

[0063] Correspondingly, a power grid status assessment system using multi-source data fusion, such as Figure 2 Shown, including, The first module is used to collect multi-source information of the power grid, establish a time axis of the multi-source information of the power grid according to the timestamps of the multi-source information, determine the time period to be evaluated, and mark the axis segment to be evaluated on the time axis of the multi-source information of the power grid; The second module is used to determine the dynamic characteristics of the power grid under the axis segment to be evaluated, analyze the correlation degree of multi-source information on the axis segment to be evaluated on the time axis of the multi-source information of the power grid, and expand the axis segment to be evaluated based on the dynamic characteristics of the power grid and the correlation degree; The third module is used to define the grid status quality level according to the time axis of the grid multi-source information, and extract the multi-source information features based on the expanded axis segment to be evaluated; The fourth module is used to construct an identification framework of the evidence theory based on the quality level of the power grid status, and to construct the basic probability of the evidence theory. It inputs multi-source information features into the evidence theory to identify the quality level of the power grid status in the time period to be evaluated.

[0064] Compared with the prior art, the present invention has the following beneficial effects: 1. Expand the segment to be evaluated based on the dynamic characteristics and correlation level of the power grid. The power grid status in the segment to be evaluated is often related to data from previous time periods. By combining the dynamic characteristics of the power grid and the degree of correlation, a reasonable forward time node is determined and data expansion is performed on the segment to be evaluated. This ensures the comprehensiveness and rationality of the data and provides a reliable foundation for subsequent multi-source information feature extraction. Define the grid status quality level, taking into account the balanced and unbalanced directions of the power grid to comprehensively define the grid status quality level, ensuring the feasibility and pertinence of the implementation of the evidence theory and adapting it to the actual operating conditions of the power grid.

[0065] 2. Constructing the basic probability of evidence theory, inputting multi-source information features into the evidence theory, and integrating uncertain information through the Dempster combination rule, addressing the problem of traditional threshold methods misjudging boundary states. Evidence theory can handle conflicts between multi-source information. By using discount coefficients and dynamic evidence weighting, it prevents a single data source from dominating the assessment results, improves decision-making robustness under complex operating conditions, and provides dispatchers with a quantifiable state confidence reference. This improves the accuracy and adaptability of power grid state assessment, ensuring the rationality of power grid state monitoring and the stability and security of power grid operation.

[0066] Through the above description of the embodiments, those skilled in the art will clearly understand that the present invention can be implemented via hardware or via software combined with a necessary general-purpose hardware platform. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product. This software product can be stored on a non-volatile storage medium (such as a CD-ROM, USB flash drive, or external hard drive) and includes instructions for enabling a computer device (such as a personal computer, server, or network device) to execute the methods described in various implementation scenarios of the present invention.

[0067] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0068] Those skilled in the art will appreciate that the modules in the system of the implementation scenario can be distributed in the system of the implementation scenario according to the implementation scenario description, or can be modified accordingly and located in one or more systems different from the implementation scenario. The modules of the above implementation scenario can be combined into one module or further divided into multiple submodules.

[0069] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A power grid status assessment method using multi-source data fusion, characterized in that: include, Collect multi-source information of the power grid, establish a time axis of the multi-source information of the power grid according to the timestamps of the multi-source information, determine the time period to be evaluated, and mark the axis segment to be evaluated on the time axis of the multi-source information of the power grid; Determine the dynamic characteristics of the power grid under the axis segment to be evaluated, analyze the correlation degree of multi-source information on the axis segment to be evaluated on the time axis of the multi-source information of the power grid, and expand the axis segment to be evaluated based on the dynamic characteristics of the power grid and the correlation degree; Define the grid status quality level based on the time axis of the grid multi-source information, and extract the multi-source information features based on the expanded axis segment to be evaluated; According to the quality level of power grid status, an identification framework of evidence theory is constructed, and the basic probability of evidence theory is constructed. Multi-source information features are input into the evidence theory to identify the quality level of power grid status in the time period to be evaluated.

2. The power grid status assessment method using multi-source data fusion according to claim 1, characterized in that: Determine the dynamic characteristics of the power grid under the shaft segment to be evaluated, include, The dynamic characteristics of the power grid include two dynamic modes: transient mode and steady-state mode; The time domain features and frequency domain features of the shaft segment to be evaluated are extracted, the dynamic mode is judged according to the range of the time domain features and frequency domain features, and the dynamic value of the dynamic mode is determined by combining the time domain features and frequency domain features.

3. The power grid status assessment method using multi-source data fusion according to claim 2, characterized in that: Analyze the correlation degree of multi-source information on the axis segment to be evaluated on the multi-source information time axis of the power grid, including: The multi-source information is classified into data categories. Based on the time axis of the power grid multi-source information, the autocorrelation function between the axis segment to be evaluated and the similar data of the axis segment in the previous time is calculated, and the autocorrelation curve that changes with time is plotted. The median and maximum values are determined on the autocorrelation curve to determine the strong correlation interval on the autocorrelation curve. The autocorrelation time node is determined based on the strong correlation interval; Integrate the autocorrelation time nodes of each type of data to obtain an autocorrelation time node set, and determine a first forward time point according to the autocorrelation time node set; Determining dependent variable elements on the axis segment to be evaluated, where the dependent variable elements are data or events, and determining independent variables corresponding to each dependent variable element, calculating a cross-correlation function between the independent variable and the dependent variable elements based on an axis segment range between a first forward time point and the axis segment to be evaluated, and plotting a cross-correlation curve graph that changes over time; The mode, original mean, standard deviation and maximum value are counted on the cross-correlation curve graph, the average of the mode and maximum value is calculated based on the two, the multiple is determined based on the average value, and the threshold is determined by combining the original mean, standard deviation and multiple; The strong correlation area on the cross-correlation curve is divided by the threshold, and the area of the strong correlation area is calculated to determine the correlation degree corresponding to each dependent variable element.

4. The method for power grid status assessment using multi-source data fusion according to claim 3, characterized in that: The axis segments to be evaluated are expanded based on the dynamic characteristics and correlation of the power grid, including: The correlation degree corresponding to all dependent variable elements is comprehensively considered, and the second forward time point is determined according to the dynamic mode type, dynamic value and correlation degree. The axis segment to be evaluated is expanded based on the second forward time point.

5. The power grid status assessment method using multi-source data fusion according to claim 1, characterized in that: Define the grid status quality level based on the time axis of multi-source information of the grid, including: The multi-source information includes the power grid's generation side data, transmission side data, and distribution side data. The power grid's generation side data, transmission side data, and distribution side data are extracted on the power grid's multi-source information timeline. Extract balance indicators and grid characteristics based on the power grid's generation, transmission, and distribution data; Input the grid characteristics into the preset grid unbalanced state model, output the initial state of the grid, and define the grid state quality level based on the initial state of the grid and the balance index; The initial state of the power grid describes the state of the power grid in an unbalanced direction, and the balance index describes the state of the power grid in a balanced direction.

6. The method for power grid status assessment using multi-source data fusion according to claim 1, characterized in that: According to the grid status quality level, the identification framework of evidence theory is constructed, including: All values of the grid status quality level are determined, and a set of propositions of the evidence theory is allocated one by one according to the value range of the grid status quality level, and the proposition set includes all grid status quality levels.

7. The method for power grid status assessment using multi-source data fusion according to claim 6, characterized in that: And construct the basic probability of evidence theory, including, Normalize the multi-source features, divide the multi-source features into multiple evidences, define the membership function between each evidence and the proposition, and calculate the original BPA; Analyze multiple pieces of evidence to determine conflicting evidence, calculate the KL divergence of the conflicting evidence, determine a discount factor based on the KL divergence, adjust the original BPA based on the discount factor, and thus determine the BPA of each piece of evidence.

8. The method for power grid status assessment using multi-source data fusion according to claim 7, characterized in that: Input multi-source information features into the evidence theory, include, Statistics the real-time noise of each piece of evidence, calculate the cross-degree and mutual support between multiple pieces of evidence, and dynamically determine the fusion weight of each piece of evidence in the evidence fusion process based on the real-time noise, cross-degree and mutual support, to achieve the fusion of multiple pieces of evidence and output the proposition result.

9. A power grid status assessment system using multi-source data fusion, characterized in that: include, The first module is used to collect multi-source information of the power grid, establish a time axis of the multi-source information of the power grid according to the timestamps of the multi-source information, determine the time period to be evaluated, and mark the axis segment to be evaluated on the time axis of the multi-source information of the power grid; The second module is used to determine the dynamic characteristics of the power grid under the axis segment to be evaluated, analyze the correlation degree of multi-source information on the axis segment to be evaluated on the time axis of the multi-source information of the power grid, and expand the axis segment to be evaluated based on the dynamic characteristics of the power grid and the correlation degree; The third module is used to define the grid status quality level according to the time axis of the grid multi-source information, and extract the multi-source information features based on the expanded axis segment to be evaluated; The fourth module is used to construct an identification framework of the evidence theory based on the quality level of the power grid status, and to construct the basic probability of the evidence theory. It inputs multi-source information features into the evidence theory to identify the quality level of the power grid status in the time period to be evaluated.

Images