Multi-source measurement data fusion and state sensing method and system for power distribution network

Abstract

The invention belongs to the technical field of power system automation, and provides a power distribution network multi-source measurement data fusion and state sensing method and system.Multi-source measurement equipment is integrated through a data acquisition module, data quality is monitored, and time alignment and evidence theory fusion are achieved through a data fusion module; the power distribution network state estimation module outputs an accurate state in combination with dynamic weighting and a power supply uncertainty model, the fault diagnosis module locates a fault and identifies the type based on a Bayesian network, the risk assessment module analyzes the risk and then performs early warning, and the visualization module graphically displays information. After the system is started, multi-source data are automatically collected, state estimation is carried out after fusion processing, states are monitored in real time, faults are rapidly diagnosed and positioned, risks are synchronously assessed and early warned, and operation and maintenance personnel grasp conditions and process the conditions through a visual interface.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power system automation, and particularly relates to a power distribution network multi-source measurement data fusion and state perception method and system. BACKGROUND

[0002] With large-scale access of distributed power sources and wide application of power electronic devices, the topology structure and operation characteristics of the power distribution network are increasingly complex, and the traditional single measurement system has been difficult to meet the high-precision state perception demand. In the prior art, multi-source measurement data often has problems such as non-uniform time scale and strong data heterogeneity due to different types of collection devices and different communication protocols, resulting in low data fusion precision and being difficult to support fine operation and management of the power distribution network.

[0003] Meanwhile, the randomness and volatility of the distributed power sources make the state variables of the power distribution network present strong uncertainty, the traditional state estimation model does not fully consider such influences, and estimation deviation is prone to occur; in addition, fault diagnosis is mostly dependent on single measurement data or simple rule reasoning, and has problems such as low positioning precision and slow type identification, and is difficult to adapt to the fault handling demand of the complex power distribution network, and a fusion and perception method that can integrate multi-source information and take into account uncertainty is urgently needed. SUMMARY

[0004] To solve the problems proposed in the background, the application provides a power distribution network multi-source measurement data fusion and state perception method and system to solve the problems of low positioning precision and slow type identification of the power distribution network.

[0005] To achieve the above-mentioned purpose, the application provides the following technical scheme: a power distribution network multi-source measurement data fusion and state perception system, comprising a data collection module, a data fusion module, a power distribution network state estimation module, a fault diagnosis module, a risk assessment module and a visualization module, wherein:

[0006] The data collection module: integrates various types of measurement devices through a standardized interface, realizes collection of various types of measurement data in the operation process of the power distribution network, and performs quality monitoring on the collected data to provide basic data support for subsequent data processing;

[0007] The data fusion module: performs time alignment processing on the collected multi-source measurement data of different time scales, and uses a multi-evidence source fusion model to fuse the aligned multi-source measurement data to improve the accuracy and reliability of the data;

[0008] The power distribution network state estimation module: based on the fused measurement data, constructs a state estimation model, estimates the node voltage, power flow and other state variables of the power distribution network, and considers the influence of the uncertainty of the distributed power source access on the state of the power distribution network to realize accurate perception of the operation state of the power distribution network;

[0009] Fault diagnosis module: based on the measurement data and state estimation results of the power distribution network, a fault diagnosis model is constructed to locate the fault components and identify the type of fault, providing basis for fault handling;

[0010] Risk assessment module: considering various risk factors in the operation process of the power distribution network, a risk assessment model is constructed to assess the operation risk level of the power distribution network, and according to the assessment results, a corresponding level of risk warning is issued to ensure the safe operation of the power distribution network;

[0011] Visualization module: the topological structure, operating state information and risk assessment results of the power distribution network are displayed in an intuitive graphical manner, which is convenient for operation and maintenance personnel to master the overall operation condition and risk situation of the power distribution network.

[0012] Optionally, the data acquisition module comprises a multi-source measurement device integration sub-module and a data quality monitoring sub-module;

[0013] The multi-source measurement device integration sub-module designs a standardized interface to realize data access of different types of measurement devices such as supervisory control and data acquisition system, phasor measurement unit and advanced measurement system, so that various types of measurement data can be real-time converged to the data processing center;

[0014] The data quality monitoring sub-module processes the collected measurement data by statistical methods, detects the outliers in the data by calculating the variance, correlation and using 3σ rule, evaluates the stability and reliability of the data, and marks the abnormal data.

[0015] Optionally, the data fusion module comprises a time scale alignment sub-module and an evidence theory data fusion model sub-module;

[0016] The time scale alignment sub-module, in view of the difference in data upload period of different measurement devices, adopts a combination of linear interpolation and spline interpolation to convert high-frequency measurement data and low-frequency measurement data to the same time scale, realizing the unification of data in time dimension, wherein linear interpolation is used to process data change period, and spline interpolation is used to process data fluctuation period;

[0017] The evidence theory data fusion model sub-module takes the measurement values of the same electrical quantity by different measurement devices as evidence sources, calculates the fused basic probability assignment function through D-S synthesis rule, and integrates the information of multiple evidence sources to improve the accuracy and reliability of data.

[0018] Optionally, the power distribution network state estimation module further comprises a dynamic weighted state estimation sub-module and a power supply uncertainty state estimation sub-module;

[0019] The dynamic weighting state estimation submodule is based on a weighted least square method, and a function with a minimum measurement data residual square sum as a target is constructed, wherein weights are dynamically adjusted according to the data quality evaluation results, not only considering measurement device accuracy, but also giving lower weights to abnormal data in combination with data quality, so as to improve state estimation accuracy.

[0020] The power uncertainty state estimation submodule regards distributed power output as a random variable, describes the uncertainty thereof through a probability distribution function, generates a large number of output scenes through Monte Carlo simulation, performs state estimation on each scene, and statistically obtains the probability distribution of system state variables.

[0021] Optionally, the fault diagnosis module further comprises an element state diagnosis submodule and a fault type identification submodule.

[0022] The element state diagnosis submodule constructs a Bayesian network with power distribution network element states as nodes and electrical connection relationships and fault propagation paths between elements as edges, determines node conditional probability distribution through learning historical data, updates node probability according to measurement data when a fault occurs, and calculates element fault probability through Bayesian inference to locate a fault element.

[0023] The fault type identification submodule analyzes amplitude and phase change characteristics of current and voltage for short-circuit faults, extracts high-frequency components of fault signals through a signal processing method, and determines short-circuit fault types according to characteristic parameters; for overload faults, device power and current parameters are monitored and analyzed in combination with a load curve to determine overload fault degree and duration.

[0024] Optionally, the risk assessment module further comprises a multi-level risk factor analysis submodule and a risk early warning submodule.

[0025] The multi-level risk factor analysis submodule divides risk factors into multiple levels based on an analytic hierarchy process and fuzzy comprehensive evaluation, determines weights of each factor through the analytic hierarchy process and performs consistency test, quantifies each risk factor into a fuzzy evaluation index through a fuzzy membership function, and further calculates a comprehensive risk level of the power distribution network.

[0026] The risk early warning submodule sets a warning threshold according to a risk assessment result, issues warning signals of different levels when the comprehensive risk level exceeds the threshold, and sends warning information to operation and maintenance personnel through short messages, emails and the like.

[0027] Optionally, the visualization module further comprises a power distribution network operation state visualization submodule and a risk information visualization submodule.

[0028] The power distribution network operation state visualization submodule utilizes geographic information system technology to display the power distribution network topological structure, equipment position, operation state and the like information in a graphical interface, uses different colors and icons to represent the equipment operation state, and displays the electrical quantity values and change trend in real time;

[0029] The risk information visualization submodule displays the risk assessment results on a geographic information system map in the form of a risk heat map to show the risk levels of different regions, and presents each risk factor and its corresponding risk grade, weight and the like information in a list form.

[0030] A power distribution network multi-source measurement data fusion and state perception method comprises the following steps:

[0031] Various types of measurement equipment are integrated through a standardized interface to realize the collection of various types of measurement data in the operation process of the power distribution network, and the collected data is subjected to quality monitoring to provide basic data support for subsequent data processing;

[0032] The multi-source measurement data collected at different time scales are subjected to time alignment processing, and a multi-evidence source fusion model is used to fuse the aligned multi-source measurement data to improve the accuracy and reliability of the data;

[0033] Based on the fused measurement data, a state estimation model is constructed to estimate the node voltage, power flow and the like state variables of the power distribution network, while considering the uncertainty influence of the distributed power supply access on the power distribution network state, so as to realize the accurate perception of the operation state of the power distribution network;

[0034] Based on the measurement data and state estimation results of the power distribution network, a fault diagnosis model is constructed to locate the components of the power distribution network that have occurred faults and identify the type of the faults, so as to provide a basis for fault processing;

[0035] Various risk factors in the operation process of the power distribution network are comprehensively considered to construct a risk assessment model to assess the operation risk level of the power distribution network, and a risk warning of a corresponding level is issued according to the assessment results to ensure the safe operation of the power distribution network;

[0036] The topological structure, operation state information and risk assessment results of the power distribution network are displayed in an intuitive graphical manner, so as to facilitate the operation and maintenance personnel to master the overall operation condition and risk condition of the power distribution network.

[0037] Compared with the prior art, the power distribution network multi-source measurement data fusion and state perception method has the following beneficial effects:

[0038] The dynamic weighting state estimation submodule in the application adjusts the weight dynamically by introducing the data quality coefficient, solves the defects that the traditional method only relies on the inherent accuracy of equipment and ignores the real-time data quality, and reduces the state estimation error.The power uncertainty state estimation submodule regards the output of the distributed power supply as a random variable and describes it through a probability distribution function, and generates multiple scene estimation results through Monte Carlo simulation, compared with the static model, improves the state estimation accuracy of the distribution network containing a high proportion of distributed power supply, and effectively supports the state perception under the access of new energy;

[0039] The element state diagnosis submodule in the application realizes fault element positioning based on a Bayesian network, dynamically updates the conditional probability distribution through historical data learning, inferences the fault probability by using the Bayesian formula, improves the positioning speed, and controls the positioning error.The fault type identification submodule combines wavelet transform and load rate analysis, realizes the rapid differentiation of short circuit and overload faults, improves the type identification accuracy, provides accurate guidance for fault repair, and shortens the fault recovery time;

[0040] The deep cooperation of the distribution network state estimation module and the fault diagnosis module in the application forms a closed loop of state perception and fault handling: the state estimation result provides high-precision basic data for fault diagnosis, and the topology change fed back by fault diagnosis also benefits the state estimation model optimization, and the combination of the two improves the operation reliability of the distribution network.At the same time, the dynamic data fusion mechanism cooperates with the visualization module, so that the operation and maintenance personnel can master the system state in real time, improve the comprehensive operation and maintenance efficiency, and significantly reduce the operation cost and fault loss of the distribution network. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 The whole system and method flowchart in the application;

[0042] Figure 2 The data acquisition module flowchart in the application;

[0043] Figure 3 The data fusion module flowchart in the application;

[0044] Figure 4 The distribution network state estimation module flowchart in the application;

[0045] Figure 5 The fault diagnosis module flowchart in the application;

[0046] Figure 6 The risk assessment module flowchart in the application;

[0047] Figure 7 The visualization module flowchart in the application;

[0048] In the figure:

[0049] 101, data acquisition module; 102, data fusion module; 103, power distribution network state estimation module; 104, fault diagnosis module; 105, risk assessment module; 106, visualization module. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0051] As shown in Figures 1 to 7 The present application provides a power distribution network multi-source measurement data fusion and state perception system, comprising a data acquisition module 101, a data fusion module 102, a power distribution network state estimation module 103, a fault diagnosis module 104, a risk assessment module 105, and a visualization module 106, wherein:

[0052] The data acquisition module 101 integrates various types of measurement devices through standardized interfaces, realizes the acquisition of various types of measurement data in the operation process of the power distribution network, and performs quality monitoring on the collected data to provide basic data support for subsequent data processing;

[0053] The data fusion module 102 performs time alignment processing on the multi-source measurement data collected at different time scales, and uses a multi-evidence source fusion model to fuse the aligned multi-source measurement data to improve the accuracy and reliability of the data;

[0054] The power distribution network state estimation module 103 constructs a state estimation model based on the fused measurement data, estimates the node voltage, power flow, and other state variables of the power distribution network, and considers the uncertainty impact of distributed power supply access on the power distribution network state to realize accurate perception of the operation state of the power distribution network;

[0055] The fault diagnosis module 104 constructs a fault diagnosis model based on the measurement data and state estimation results of the power distribution network, locates the components that have failed in the power distribution network, and identifies the type of the fault to provide a basis for fault handling;

[0056] The risk assessment module 105 considers various risk factors in the operation process of the power distribution network, constructs a risk assessment model, assesses the operation risk level of the power distribution network, and issues a risk warning of the corresponding level according to the assessment result to ensure the safe operation of the power distribution network;

[0057] The visualization module 106: the topology of the power distribution network, the operation state information and the risk assessment results are displayed in an intuitive graphical manner, so as to help the operation and maintenance personnel to master the overall operation condition and risk situation of the power distribution network.

[0058] The data acquisition module 101 comprises a multi-source measurement equipment integration submodule and a data quality monitoring submodule.

[0059] The multi-source measurement equipment integration submodule is designed with a standardized interface, realizes data access of different types of measurement equipment such as the supervisory control and data acquisition system, the phasor measurement unit and the advanced measurement system, and enables real-time convergence of various types of measurement data to the data processing center.

[0060] The data quality monitoring submodule processes the collected measurement data by using statistical methods, detects abnormal values in the data by calculating the variance and correlation of the data and using the 3σ criterion, evaluates the stability and reliability of the data, and marks abnormal data.

[0061] The data fusion module 102 comprises a time scale alignment submodule and an evidence theory data fusion model submodule.

[0062] The time scale alignment submodule converts high-frequency measurement data and low-frequency measurement data to the same time scale by using a combination of linear interpolation and spline interpolation in view of the difference in data upload period of different measurement equipment, realizes unification of data in the time dimension, wherein the linear interpolation is used to process data change period, and the spline interpolation is used to process data fluctuation period.

[0063] The evidence theory data fusion model submodule takes the measurement values of the same electrical quantity by different measurement equipment as evidence sources, calculates the basic probability assignment function after fusion by using the D-S synthesis rule, and comprehensively considers the information of multiple evidence sources to improve the accuracy and reliability of data.

[0064] The power distribution network state estimation module 103 further comprises a dynamic weighted state estimation submodule and a power source uncertainty state estimation submodule.

[0065] The dynamic weighted state estimation submodule is based on the weighted least squares method, and constructs a function with the minimum sum of squares of measurement data residuals as the target, wherein the weight is dynamically adjusted according to the data quality evaluation result, not only considering the measurement equipment accuracy, but also giving lower weight to abnormal data in combination with data quality, so as to improve the accuracy of state estimation.

[0066] The power source uncertainty state estimation submodule regards the distributed power output as a random variable, describes the uncertainty thereof by using a probability distribution function, generates a large number of output scenarios by using Monte Carlo simulation, performs state estimation on each scenario and statistically obtains the probability distribution of system state variables.

[0067] Specifically, the dynamic weighting state estimation submodule is constructed to build a function with the least measurement data residual square sum as the target, and the function formula is:

[0068]

[0069] Wherein is the target function value, representing the measurement data residual square sum, the smaller the value, the smaller the deviation between the state estimation result and the actual measurement data, the more accurate the state estimation, and m is the number of measurement data, is the weight of the i th measurement data, the size of the weight determines the influence degree of the measurement data in the state estimation, the greater the weight, the greater the influence on the estimation result, is the i th actual measurement data value, is the estimation function of the i th measurement data, x is the system state variable, and the weight The calculation formula is:

[0070]

[0071] Wherein is the inherent standard deviation of the i th measurement device, reflecting the measurement accuracy of the measurement device itself, the smaller the standard deviation, the higher the device accuracy, is the data quality coefficient of the i th measurement data, ranging from 0 to 1, the better the data quality, the closer the coefficient to 1.

[0072] Specifically, the probability distribution function in the power uncertainty state estimation submodule is:

[0073]

[0074] Wherein is the probability density function value when the distributed power output is P, indicating the possibility of the output being P, and P is the output value of the distributed power, is the mean of the distributed power output, is the standard deviation of the distributed power output, reflecting the dispersion degree of the distributed power output, the larger the standard deviation, the greater the uncertainty of the output.

[0075] Specifically, the dynamic weighting state estimation submodule in the application dynamically adjusts the weight by introducing the data quality coefficient, solves the defects that the traditional method only depends on the inherent accuracy of the device and ignores the real-time data quality, and reduces the state estimation error. The power uncertainty state estimation submodule regards the distributed power output as a random variable and describes it through a probability distribution function, generates multiple scene estimation results through Monte Carlo simulation, improves the state estimation accuracy of the distribution network containing a high proportion of distributed power compared with the static model, and effectively supports the state perception under the access of new energy.

[0076] Specifically, dynamic weighting state estimation error comparison (rated voltage 10kV, rated power flow 10MW)

[0077] State variable Traditional fixed weight Dynamic weight Error reduction rate Node voltage (kV) 0.28 0.09 67.9％ Line flow (MW) 0.65 0.18 72.3％

[0078] Specifically, power supply uncertainty state estimation voltage deviation pass rate comparison

[0079] Distributed power output scenario Traditional random-neglected qualification rate Monte Carlo simulation qualification rate Qualification rate improvement increment Low output (total output ≤ 2 MW) 92.5% 99.1% 7.1% Medium output (2 MW < total output ≤ 5 MW) 85.3% 98.7% 15.7% High output (total output > 5 MW) 78.6% 97.9% 24.6%

[0080] The fault diagnosis module 104 further comprises an element state diagnosis sub-module and a fault type identification sub-module.

[0081] The element state diagnosis sub-module constructs a Bayesian network with the state of the distribution network element as the node and the electrical connection relationship between elements and the fault propagation path as the edge, determines the node conditional probability distribution through learning historical data, updates the node probability according to the measurement data when a fault occurs, and calculates the element fault probability by using Bayesian inference to locate the faulty element.

[0082] The fault type identification sub-module analyzes the amplitude and phase change characteristics of the current and voltage for short-circuit faults, extracts the high-frequency component of the fault signal by using a signal processing method, and determines the short-circuit fault type according to the characteristic parameters; for overload faults, the power and current parameters of the equipment are monitored and analyzed in combination with the load curve to determine the overload fault degree and duration.

[0083] Specifically, the element state diagnosis sub-module in the application realizes fault element positioning based on a Bayesian network, dynamically updates the conditional probability distribution through historical data learning, and infers the fault probability by using the Bayesian formula, thereby improving the positioning speed and controlling the positioning error. The fault type identification sub-module combines wavelet transform and load rate analysis to realize rapid differentiation of short-circuit and overload faults, improve the type identification accuracy, provide accurate guidance for fault repair, and shorten the fault recovery time.

[0084] Specifically, the deep cooperation between the distribution network state estimation module and the fault diagnosis module in the application forms a closed loop of state perception and fault processing: the state estimation result provides high-precision basic data for fault diagnosis, and the topology change fed back by fault diagnosis in turn optimizes the state estimation model, thereby improving the operation reliability of the distribution network. At the same time, the dynamic data fusion mechanism cooperates with the visualization module to enable the operation and maintenance personnel to master the system state in real time, improve the comprehensive operation and maintenance efficiency, and significantly reduce the operation cost and fault loss of the distribution network.

[0085] The risk assessment module 105 further comprises a multi-level risk factor analysis sub-module and a risk early warning sub-module.

[0086] The multi-level risk factor analysis submodule divides the risk factors into multiple levels based on the analytic hierarchy process and fuzzy comprehensive evaluation, determines the weight of each factor by the analytic hierarchy process and performs consistency test, quantifies each risk factor into fuzzy evaluation index by using fuzzy membership function, and then calculates the comprehensive risk level of the distribution network;

[0087] The risk early warning submodule sets a warning threshold according to the risk assessment result, issues a warning signal of different levels when the comprehensive risk level exceeds the threshold, and sends the warning information to the operation and maintenance personnel through short message, email and the like;

[0088] The visualization module 106 further includes a distribution network operation state visualization submodule and a risk information visualization submodule;

[0089] The distribution network operation state visualization submodule uses geographic information system technology to display the topological structure, device location, operation state and the like of the distribution network in a graphical interface, uses different colors and icons to represent the device operation state, and displays the electrical quantity values and change trend in real time;

[0090] The risk information visualization submodule displays the risk assessment result on a geographic information system map in the form of a risk heat map to show the risk levels of different regions, and presents each risk factor, the corresponding risk grade, weight and the like in the form of a list.

[0091] Specifically, the above modules are cooperatively operated, and the key operation indicators of the distribution network are significantly optimized, as follows:

[0092] Comprehensive index Before optimization After optimization Optimization amplitude State estimation error (node voltage) ≤3.5％ ≤0.8％ 77.1％ Fault location time ≥10s ≥5s 50％ Distribution network average outage time 1.2 h / year per household 0.3 h / year per household 75％ Operation and maintenance personnel work efficiency 10 tasks / day 25 tasks / day 150％

[0093] A distribution network multi-source measurement data fusion and state perception method, comprising the following steps:

[0094] Various types of measurement devices are integrated through a standardized interface to realize the collection of various types of measurement data in the operation process of the distribution network, and the collected data is subjected to quality monitoring to provide basic data support for subsequent data processing;

[0095] The multi-source measurement data collected at different time scales are subjected to time alignment processing, and a multi-evidence source fusion model is used to fuse the aligned multi-source measurement data to improve the accuracy and reliability of the data;

[0096] Based on the fused measurement data, a state estimation model is constructed to estimate the node voltage, power flow and other state variables of the distribution network, while considering the uncertainty influence of the distributed power supply access on the state of the distribution network, so as to realize accurate perception of the operation state of the distribution network;

[0097] Based on the measurement data and state estimation results of the distribution network, a fault diagnosis model is constructed to locate the faulty components in the distribution network and identify the type of fault, providing a basis for fault handling.

[0098] Taking into account various risk factors in the operation of the distribution network, a risk assessment model is constructed to assess the operational risk level of the distribution network and issue corresponding risk warnings based on the assessment results, so as to ensure the safe operation of the distribution network.

[0099] The topology, operating status information, and risk assessment results of the distribution network are displayed in an intuitive graphical format, making it easier for operation and maintenance personnel to understand the overall operating status and risk situation of the distribution network.

[0100] The working principle and usage process of this invention are as follows: The data acquisition module 101 integrates multi-source measurement equipment and monitors data quality. The data fusion module 102 achieves time alignment and evidence theory fusion. The distribution network state estimation module 103 combines dynamic weighting and a power uncertainty model to output accurate state data. The fault diagnosis module 104 locates faults and identifies their types based on a Bayesian network. The risk assessment module 105 analyzes risks and issues warnings. The visualization module 106 graphically displays the information. The usage process is as follows: After startup, multi-source data is automatically collected, fused, and then state estimation is performed. The state is monitored in real time, faults are quickly diagnosed and located, risks are simultaneously assessed and warnings are issued, and maintenance personnel can understand and handle the situation through a visual interface.

[0101] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0102] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A system for fusing multi-source measurement data and assessing the condition of a power distribution network, characterized in that, It includes a data acquisition module (101), a data fusion module (102), a distribution network state estimation module (103), a fault diagnosis module (104), a risk assessment module (105), and a visualization module (106), wherein: Data acquisition module (101): It integrates various types of measurement devices through a standardized interface to collect various measurement data during the operation of the power distribution network, and monitors the quality of the collected data to provide basic data support for subsequent data processing; Data fusion module (102): performs time alignment processing on the collected multi-source measurement data at different time scales, and uses a multi-evidence source fusion model to fuse the aligned multi-source measurement data to improve the accuracy and reliability of the data; Distribution network state estimation module (103): Based on the fused measurement data, a state estimation model is constructed to estimate the state variables such as node voltage and power flow of the distribution network. At the same time, the uncertainty of the distributed power source access on the distribution network state is considered so as to achieve accurate perception of the distribution network operation status. Fault diagnosis module (104): Based on the measurement data and state estimation results of the distribution network, a fault diagnosis model is constructed to locate the faulty components in the distribution network and identify the type of fault, providing a basis for fault handling; Risk assessment module (105): Taking into account various risk factors in the operation of the distribution network, a risk assessment model is constructed to assess the operational risk level of the distribution network and issue corresponding risk warnings based on the assessment results to ensure the safe operation of the distribution network. Visualization module (106): Displays the topology, operating status information and risk assessment results of the distribution network in an intuitive graphical way, making it easier for operation and maintenance personnel to grasp the overall operating status and risk situation of the distribution network.

2. The distribution network multi-source measurement data fusion and status awareness system according to claim 1, characterized in that, The data acquisition module (101) includes: a multi-source measurement equipment integration submodule and a data quality monitoring submodule; The multi-source measurement equipment integration sub-module is designed with standardized interfaces to enable data access from different types of measurement equipment, such as monitoring and data acquisition systems, synchronous phasor measurement units, and advanced measurement systems, so that various types of measurement data can be aggregated to the data processing center in real time. The data quality monitoring submodule uses statistical methods to process the collected measurement data. By calculating the variance and correlation of the data and applying the 3σ criterion, it detects outliers in the data, evaluates the stability and reliability of the data, and marks the outliers.

3. The distribution network multi-source measurement data fusion and status awareness system according to claim 1, characterized in that, The data fusion module (102) includes: a time scale alignment submodule and an evidence theory data fusion model submodule; The time scale alignment submodule, taking into account the differences in data upload cycles of different measurement devices, uses a combination of linear interpolation and spline interpolation to convert high-frequency and low-frequency measurement data to the same time scale, thereby achieving data unification in the time dimension. Linear interpolation is used to handle periods of slow data change, while spline interpolation is used to handle periods of large data fluctuation. The evidence theory data fusion model submodule uses the measurement values ​​of the same electrical quantity from different measuring devices as evidence sources. It calculates the basic probability allocation function after fusion through the DS synthesis rule, and integrates information from multiple evidence sources to improve data accuracy and reliability.

4. The distribution network multi-source measurement data fusion and status awareness system according to claim 1, characterized in that, The distribution network state estimation module (103) further includes: a dynamic weighted state estimation submodule and a power source uncertainty state estimation submodule; The dynamic weighted state estimation submodule is based on the weighted least squares method and constructs a function with the objective of minimizing the sum of squared residuals of the measurement data. The weights are dynamically adjusted according to the data quality assessment results, taking into account not only the accuracy of the measurement equipment, but also the data quality, and assigning lower weights to abnormal data to improve the accuracy of state estimation. The power supply uncertainty state estimation submodule treats the output of distributed power sources as random variables, describes its uncertainty through a probability distribution function, generates a large number of output scenarios using Monte Carlo simulation, performs state estimation for each scenario, and statistically analyzes the probability distribution of system state variables.

5. The distribution network multi-source measurement data fusion and status awareness system according to claim 1, characterized in that, The fault diagnosis module (104) further includes: a component status diagnosis submodule and a fault type identification submodule; The component status diagnosis submodule constructs a Bayesian network with the status of distribution network components as nodes and the electrical connection relationship and fault propagation path between components as edges. It determines the conditional probability distribution of nodes by learning historical data, updates the node probability based on measurement data when a fault occurs, and uses Bayesian inference to calculate the component fault probability in order to locate the faulty component. The fault type identification submodule analyzes the amplitude and phase change characteristics of current and voltage for short-circuit faults, extracts high-frequency components of the fault signal using signal processing methods, and determines the type of short-circuit fault based on characteristic parameters. For overload faults, it monitors equipment power and current parameters and analyzes load curves to determine the overload fault, its degree, and duration.

6. The distribution network multi-source measurement data fusion and status awareness system according to claim 1, characterized in that, The risk assessment module (105) also includes: a multi-level risk factor analysis submodule and a risk early warning submodule; The multi-level risk factor analysis submodule, based on the analytic hierarchy process (AHP) and fuzzy comprehensive evaluation, divides risk factors into multiple levels, determines the weight of each factor through the AHP and performs consistency checks, and uses fuzzy membership functions to quantify each risk factor into fuzzy evaluation indicators, thereby calculating the comprehensive risk level of the distribution network. The risk warning submodule sets warning thresholds based on risk assessment results. When the overall risk level exceeds the threshold, it issues warning signals of different levels and sends the warning information to maintenance personnel via SMS and email.

7. The distribution network multi-source measurement data fusion and status awareness system according to claim 1, characterized in that, The visualization module (106) further includes: a power distribution network operation status visualization submodule and a risk information visualization submodule; The power distribution network operation status visualization submodule uses geographic information system technology to display the power distribution network topology, equipment location, and operation status information in a graphical interface, using different colors and icons to represent the equipment operation status, and displaying electrical quantity values ​​and changing trends in real time. The risk information visualization submodule displays the risk assessment results as a risk heat map on a geographic information system map, showing the risk levels of different regions. It also presents each risk factor and its corresponding risk level, weight, and other information in a list format.

8. A method for fusing and sensing multi-source measurement data in a distribution network, characterized in that, Includes the following steps: By integrating various types of measurement devices through standardized interfaces, it is possible to collect various measurement data during the operation of the power distribution network, monitor the quality of the collected data, and provide basic data support for subsequent data processing. Time alignment was performed on the collected multi-source measurement data at different time scales, and a multi-evidence source fusion model was used to fuse the aligned multi-source measurement data to improve the accuracy and reliability of the data. Based on the fused measurement data, a state estimation model is constructed to estimate the node voltage and power flow state variables of the distribution network. At the same time, the uncertainty of the distribution network state brought about by the access of distributed power sources is considered, so as to achieve accurate perception of the operating status of the distribution network. Based on the measurement data and state estimation results of the distribution network, a fault diagnosis model is constructed to locate the faulty components in the distribution network and identify the type of fault, providing a basis for fault handling. Taking into account various risk factors in the operation of the distribution network, a risk assessment model is constructed to assess the operational risk level of the distribution network and issue corresponding risk warnings based on the assessment results, so as to ensure the safe operation of the distribution network. The topology, operating status information, and risk assessment results of the distribution network are displayed in an intuitive graphical format, making it easier for operation and maintenance personnel to understand the overall operating status and risk situation of the distribution network.

Images