Power distribution system reliability evaluation system based on cloud computing

Abstract

The invention discloses a power distribution system reliability evaluation system based on cloud computing. The system comprises a multi-source data access module, an edge computing processing module, a cloud computing evaluation module, a power distribution system reliability evaluation model construction module and an intelligent decision module. In a power distribution system reliability evaluation model construction module, a real-time failure rate of equipment in a power supply unit is generated through equipment basic data, equipment state monitoring data, a meteorological influence factor and a load fluctuation factor; the defect that a traditional fixed fault rate model only depends on a historical average value and cannot reflect the current running state of equipment is overcome, and the real-time fault recognition accuracy is improved. Due to the arrangement of the reliability influence factor, the reliable inverse value of the system is evaluated more comprehensively, and evaluation deviation caused by single data is avoided; through the arrangement of the power distribution system reliability evaluation model building module, the problem of nonlinear evaluation of multi-factor coupling in the power distribution system is solved, and the reliability inverse value evaluation accuracy of the system is improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a power distribution system reliability assessment system based on cloud computing. Background Art

[0002] The distribution system is a crucial component of the power system. It receives electricity from the transmission grid or local power plants, distributes it to various end users through distribution networks and equipment. The distribution system connects the transmission system to users and directly impacts the reliability, affordability, and quality of power supply.

[0003] Publication No. CN110162902B discloses a cloud computing-based distribution system reliability assessment method, which improves assessment efficiency through multi-source data fusion and distributed computing. This method uses a cloud computing architecture to process data and construct a basic failure rate model. However, this application suffers from the following issues: The complex coupling relationships between equipment aging, environmental stress, and load fluctuations in the distribution system are complex, but the reference document uses traditional mathematical models to handle these nonlinear relationships, resulting in poor adaptability of the assessment results to complex scenarios. Summary of the Invention

[0004] In order to solve the technical problems existing in the background technology, the present invention proposes a distribution system reliability assessment system based on cloud computing.

[0005] The present invention proposes a power distribution system reliability assessment system based on cloud computing, comprising:

[0006] Multi-source data access module, edge computing processing module, cloud computing assessment module, distribution system reliability assessment model construction module and intelligent decision-making module;

[0007] Multi-source data access module: used to collect multi-source heterogeneous data of the power distribution system in real time;

[0008] Multi-dimensional heterogeneous data includes power outage data, equipment basic data, user distribution data, geographic information data, real-time meteorological data, equipment status monitoring data, and maintenance data;

[0009] Basic equipment data includes equipment operating years;

[0010] Equipment status monitoring data includes load rate and real-time load;

[0011] Real-time meteorological data includes real-time temperature and real-time humidity;

[0012] Edge computing processing module: used to pre-process the multi-source heterogeneous data collected in the multi-source data access module;

[0013] The cloud computing evaluation module includes a rule base and a computing unit;

[0014] The rule base includes the basic equipment failure rate database and the fuzzy rule base;

[0015] The computing unit analyzes the multivariate heterogeneous data pre-processed by the edge computing processing module based on the cloud computing architecture to obtain meteorological influencing factors and load fluctuation factors;

[0016] Distribution system reliability assessment model construction module: used to build a distribution system reliability assessment model, which is used to generate the reliability inverse value K of each power supply unit pi and the system-level reliability inverse K p ;

[0017] Intelligent decision-making module: Based on the reliability evaluation model of the distribution system, the module builds the reliability inverse value K of each power supply unit output pi and the system reliability inverse K p , generate reliability assessment report and rectification suggestions, when the system reliability inverse value K p When the preset threshold is exceeded, device inspection or network reconstruction strategy is automatically triggered;

[0018] Generate reliability assessment reports and corrective action recommendations using proven, publicly available industry-standard technologies, such as expert system-based fault diagnosis methods and universal report template generation technology.

[0019] Equipment inspection or network reconstruction strategies are both well-known technical means in the field of power distribution system operation and maintenance.

[0020] Preferably, in the edge computing processing module, preprocessing includes data cleaning, format unification, and outlier detection;

[0021] The edge computing processing module is deployed at the edge nodes in the distribution area of the power distribution system. The edge nodes include intelligent distribution terminals, built-in computing modules in the equipment, or regional control cabinets. They are used to collect data physically close to the distribution equipment or user side to achieve local pre-processing and real-time communication of data.

[0022] Preferably, in the calculation unit, the meteorological impact factor and the load fluctuation factor are generated as follows:

[0023] Assume that the meteorological impact factor is α(T, H), the real-time temperature and real-time humidity in the real-time meteorological data are T and H respectively, and the calculation formula of α(T, H) is:

[0024]

[0025] Where γ is the environmental sensitivity coefficient, T0 is the operating temperature of the device under the preset ideal state, and H0 is the operating humidity of the device under the preset ideal state;

[0026] Assume that the load fluctuation factor is β(L), and the real-time load in the equipment status monitoring data is L. The calculation formula of β(L) is:

[0027]

[0028] Where, L n For rated load.

[0029] Preferably, in the distribution system reliability assessment model construction module, the distribution system reliability assessment model is constructed as follows:

[0030] Obtain the multi-dimensional heterogeneous data pre-processed in the edge computing processing module, and use the dynamic grid partitioning algorithm to divide the distribution area into multiple power supply units based on the geographic information data and user distribution data in the multi-dimensional heterogeneous data;

[0031] According to the distribution network topology in the power distribution system, a set of physical entities supplying power to the power supply unit is set as a distribution network subunit;

[0032] Each power supply unit corresponds to a power distribution network subunit;

[0033] The set of physical entities includes the distribution lines, transformers, and switchgear that supply power to the supply unit;

[0034] Generate the real-time failure rate of the equipment in the power supply unit based on the basic equipment data and equipment status monitoring data in the multi-heterogeneous data preprocessed by the edge computing processing module, as well as the meteorological influence factors and load fluctuation factors in the computing unit;

[0035] Based on the multi-dimensional heterogeneous data pre-processed by the edge computing processing module and the real-time failure rate of the equipment in the power supply unit, the reliability impact factor of the power supply unit is generated. The reliability impact factors include the average number of power outages per month, the average power outage duration, the number of users, and the maintenance response time.

[0036] The reliability influencing factors are used as input variables of the fuzzy membership function, and the input variable reliability influencing factors are mapped to the fuzzy state interval through the fuzzy membership function to generate the fuzzy level of each reliability influencing factor;

[0037] The fuzzy state interval includes three fuzzy levels: low, medium, and high;

[0038] Based on the fuzzy rule base in the rule base, according to the fuzzy level of each reliability influencing factor, the fuzzy result is obtained by matching the rules in the fuzzy rule base through fuzzy reasoning. The fuzzy result is defuzzified by the centroid method to obtain the reliability score inverse value K pi ;

[0039] Calculate the weighted sum of the reliability inverse values of all power supply units to obtain the system reliability inverse value K p ;

[0040] K p Reflects the reliability level of the entire power distribution system.

[0041] Preferably, in the power distribution system reliability assessment model building module, the real-time failure rate of the equipment in the power supply unit is generated in the following manner:

[0042] Based on the equipment basic failure rate database in the rule base;

[0043] Obtain equipment basic data and equipment status monitoring data, and generate the initial failure rate function λ0(t) through the Weibull distribution model;

[0044] Obtain the meteorological influence factor α(T, H) and load fluctuation factor β(L) in the calculation unit;

[0045] The calculation formula for the real-time failure rate λ(t) of the equipment in the power supply unit is:

[0046] λ(t)=λ0(t)×α(T,H)×β(L).

[0047] Preferably, in the distribution system reliability assessment model construction module, the reliability influencing factors of the power supply unit are generated based on the multivariate heterogeneous data preprocessed by the edge computing processing module and the real-time failure rate of the equipment in the power supply unit as follows:

[0048] Based on the multivariate heterogeneous data pre-processed by the edge computing processing module, the average monthly power outage times, average power outage duration, number of users, and maintenance response time of the power supply unit are obtained through statistical analysis.

[0049] Based on the multivariate heterogeneous data preprocessed by the edge computing processing module and the real-time failure rate of the equipment in the power supply unit, the equipment health index of the power supply unit is obtained through the neural network algorithm.

[0050] Preferably, it also includes:

[0051] Visualization interaction module: used to convert the reliability score inverse value K of each power supply unit generated by the distribution system reliability assessment model construction module pi and the system-level reliability inverse K p , and the reliability assessment report and rectification suggestions generated by the intelligent decision-making module are presented in a visual way.

[0052] A method for evaluating the reliability of a power distribution system based on cloud computing includes the following steps:

[0053] S1, real-time collection of multi-dimensional heterogeneous data of the power distribution system;

[0054] Multivariate heterogeneous data;

[0055] Multi-dimensional heterogeneous data includes power outage data, equipment basic data, user distribution data, geographic information data, real-time meteorological data, equipment status monitoring data, and maintenance data;

[0056] Basic equipment data includes equipment operating years;

[0057] Equipment status monitoring data includes load rate and real-time load;

[0058] Real-time meteorological data includes real-time temperature and real-time humidity;

[0059] S2, preprocessing the multivariate heterogeneous data collected in S1;

[0060] S3, based on the cloud computing architecture, analyzes the multivariate heterogeneous data pre-processed in S2 to obtain meteorological influencing factors and load fluctuation factors;

[0061] S4. Obtain the multi-dimensional heterogeneous data pre-processed in the edge computing processing module, and use a dynamic grid partitioning algorithm to divide the distribution area into multiple power supply units based on the geographic information data and user distribution data in the multi-dimensional heterogeneous data;

[0062] According to the distribution network topology in the power distribution system, a set of physical entities supplying power to the power supply unit is set as a distribution network subunit;

[0063] Each power supply unit corresponds to a power distribution network subunit;

[0064] The set of physical entities includes the distribution lines, transformers, and switchgear that supply power to the supply unit;

[0065] Generate the real-time failure rate of the equipment in the power supply unit based on the basic equipment data and equipment status monitoring data in the multi-heterogeneous data preprocessed by the edge computing processing module, as well as the meteorological influence factors and load fluctuation factors in the computing unit;

[0066] S5. Generate reliability influencing factors of the power supply unit based on the multivariate heterogeneous data preprocessed in S2 and the real-time failure rate of the equipment in the power supply unit in S4. The reliability influencing factors include the average number of power outages per month, the average power outage duration, the number of users, and the maintenance response time.

[0067] The reliability influencing factors are used as input variables of the fuzzy membership function, and the input variable reliability influencing factors are mapped to the fuzzy state interval through the fuzzy membership function to generate the fuzzy level of each reliability influencing factor;

[0068] Based on the fuzzy rule base in the rule base, according to the fuzzy level of each reliability influencing factor, the fuzzy result is obtained by matching the rules in the fuzzy rule base through fuzzy reasoning. The fuzzy result is defuzzified by the centroid method to obtain the reliability score inverse value K pi ;

[0069] Calculate the weighted sum of the reliability inverse values of all power supply units to obtain the system reliability inverse value K p ;

[0070] S6, according to the output of S5, the reliable inverse value K of each power supply unit pi and the system reliability inverse K p , generate reliability assessment report and rectification suggestions, when the system reliability inverse value K p When the preset threshold is exceeded, device inspection or network reconstruction strategy is automatically triggered.

[0071] The proposed cloud computing-based distribution system reliability assessment system has the following beneficial technical effects:

[0072] 1. In the distribution system reliability assessment model construction module, the real-time failure rate of equipment in the power supply unit is generated by integrating basic equipment data, equipment status monitoring data, meteorological influence factors, and load fluctuation factors. This overcomes the defect of traditional fixed failure rate models that rely solely on historical averages and cannot reflect the current operating status of equipment. The initial failure rate function based on the Weibull distribution is combined with dynamically generated meteorological influence factors and load fluctuation factors to improve the accuracy of real-time fault identification.

[0073] Based on the multi-dimensional heterogeneous data pre-processed by the edge computing processing module and the real-time failure rate of the equipment in the power supply unit, the reliability influencing factors of the power supply unit are generated. The reliability influencing factors include the average monthly number of power outages, the average power outage duration, the number of users, and the maintenance response time. The system reliability inverse value is evaluated more comprehensively to avoid the evaluation bias caused by single data; based on the fuzzy rule base in the rule base, according to the fuzzy level of each reliability influencing factor, the fuzzy results are obtained by matching the rules in the fuzzy rule base through fuzzy reasoning. The fuzzy results are defuzzified by the centroid method to obtain the reliability score inverse value, which solves the nonlinear evaluation problem of multi-factor coupling in the distribution system and improves the evaluation accuracy of the system reliability inverse value; thus, the specific problem area is located through the reliability score inverse value of the power supply unit, and the system reliability inverse value reflects the reliability level of the entire distribution system.

[0074] 2. The calculation unit quantifies meteorological influencing factors and load fluctuation factors to respond to the impact of weather on equipment failure rates in real time, avoiding the evaluation distortion of traditional fixed failure rate models in complex environments. The load fluctuation factor reflects the impact of equipment load on reliability in real time, realizing dynamic load perception.

[0075] 3. When the system reliability inverse exceeds a preset threshold, it automatically triggers equipment inspections or network reconstruction strategies. This data-driven automation mechanism can quickly respond to potential risks without human intervention, significantly shortening the time interval between fault discovery and resolution. It is suitable for real-time monitoring of large-scale and complex distribution networks, effectively improving operation and maintenance efficiency and reducing the risk of human misjudgment.

[0076] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 It is a principle block diagram of the system of the present invention;

[0078] Figure 2 Flowchart of the method of the present invention. DETAILED DESCRIPTION

[0079] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar symbols throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.

[0080] like Figure 1 A power distribution system reliability assessment system based on cloud computing is shown, comprising:

[0081] Multi-source data access module, edge computing processing module, cloud computing assessment module, distribution system reliability assessment model construction module and intelligent decision-making module;

[0082] Multi-source data access module: used to collect multi-source heterogeneous data of the power distribution system in real time;

[0083] Multi-dimensional heterogeneous data includes power outage data, equipment basic data, user distribution data, geographic information data, real-time meteorological data, equipment status monitoring data, and maintenance data;

[0084] Basic equipment data includes equipment operating years;

[0085] Equipment status monitoring data includes load rate and real-time load;

[0086] Real-time meteorological data includes real-time temperature and real-time humidity;

[0087] Edge computing processing module: used to pre-process the multi-source heterogeneous data collected in the multi-source data access module;

[0088] In an optional embodiment, in the edge computing processing module, preprocessing includes data cleaning, format unification, and outlier detection;

[0089] The edge computing processing module is deployed at the edge nodes in the distribution area of the power distribution system. The edge nodes include intelligent distribution terminals, equipment-built-in computing modules, or regional control cabinets. They are used to collect data physically close to the distribution equipment or user side, enabling local pre-processing and real-time communication of data.

[0090] The edge computing processing module uses a triple pre-processing mechanism of data cleaning, format unification, and outlier detection to effectively filter out duplicate, missing, and abnormal data, ensuring the accuracy of data input into the cloud computing evaluation module and avoiding evaluation bias caused by data errors. Data pre-processing is completed at the edge node, reducing the amount of invalid data processed in the cloud, alleviating network transmission pressure, and improving data processing efficiency.

[0091] The cloud computing evaluation module includes a rule base and a computing unit;

[0092] The rule base includes the basic equipment failure rate database and the fuzzy rule base;

[0093] The computing unit analyzes the multivariate heterogeneous data pre-processed by the edge computing processing module based on the cloud computing architecture to obtain meteorological influencing factors and load fluctuation factors;

[0094] Distribution system reliability assessment model construction module: used to build a distribution system reliability assessment model, which is used to generate the reliability inverse value K of each power supply unit pi and the system-level reliability inverse K p ;

[0095] Intelligent decision-making module: Based on the reliability evaluation model of the distribution system, the module builds the reliability inverse value K of each power supply unit output pi and the system reliability inverse K p , generate reliability assessment report and rectification suggestions, when the system reliability inverse value K p When the preset threshold is exceeded, device inspection or network reconstruction strategy is automatically triggered;

[0096] Generate reliability assessment reports and corrective action recommendations using proven, publicly available industry-standard technologies, such as expert system-based fault diagnosis methods and universal report template generation technology.

[0097] Equipment inspection or network reconstruction strategies are both well-known technical means in the field of power distribution system operation and maintenance;

[0098] When the system reliability inverse exceeds the preset threshold, it can automatically trigger equipment inspection or network reconstruction strategies. This data-driven automation mechanism can quickly respond to potential risks without human intervention, significantly shortening the time interval between fault discovery and resolution. It is suitable for real-time monitoring of large-scale and complex distribution networks, effectively improving operation and maintenance efficiency and reducing the risk of human misjudgment.

[0099] Visualization interaction module: used to convert the reliability score inverse value K of each power supply unit generated by the distribution system reliability assessment model construction module pi and the system-level reliability inverse K p , and the reliability assessment report and rectification suggestions generated by the intelligent decision-making module are presented in a visual way.

[0100] In an optional embodiment, in the calculation unit, the meteorological impact factor and the load fluctuation factor are generated as follows:

[0101] Assume that the meteorological impact factor is α(T, H), the real-time temperature and real-time humidity in the real-time meteorological data are T and H respectively, and the calculation formula of α(T, H) is:

[0102]

[0103] Where γ is the environmental sensitivity coefficient, T0 is the operating temperature of the device under the preset ideal state, and H0 is the operating humidity of the device under the preset ideal state;

[0104] Assume that the load fluctuation factor is β(L), and the real-time load in the equipment status monitoring data is L. The calculation formula of β(L) is:

[0105]

[0106] Where, L n is the rated load;

[0107] The calculation unit quantifies the meteorological impact factor and the load fluctuation factor:

[0108] Respond to the impact of weather on equipment failure rates in real time, avoiding the evaluation distortion of traditional fixed failure rate models in complex environments;

[0109] The load fluctuation factor reflects the impact of equipment load on reliability in real time, achieving dynamic load perception;

[0110] In an optional embodiment, in the power distribution system reliability assessment model construction module, the power distribution system reliability assessment model is constructed as follows:

[0111] Obtain the multi-dimensional heterogeneous data pre-processed in the edge computing processing module, and use the dynamic grid partitioning algorithm to divide the distribution area into multiple power supply units based on the geographic information data and user distribution data in the multi-dimensional heterogeneous data;

[0112] According to the distribution network topology in the power distribution system, a set of physical entities supplying power to the power supply unit is set as a distribution network subunit;

[0113] Each power supply unit corresponds to a power distribution network subunit;

[0114] The set of physical entities includes the distribution lines, transformers, and switchgear that supply power to the supply unit;

[0115] Through a dynamic grid division algorithm, the power distribution area is divided into multiple power supply units based on geographic information data and user distribution data, more accurately locating areas with weak reliability. Different power supply units have different power supply characteristics due to their different geographical environments, user distribution and power demand, thus distinguishing different power supply characteristics;

[0116] Dividing the distribution area into power supply units and corresponding distribution network subunits can more accurately reflect the actual structure and operation of the system. This helps improve the accuracy and reliability of the model, making the evaluation results more realistically reflect the actual status of the distribution system.

[0117] By dividing the power supply unit and distribution network subunits, the impact of load changes on system reliability can be better tracked and analyzed;

[0118] Generate the real-time failure rate of the equipment in the power supply unit based on the basic equipment data and equipment status monitoring data in the multi-heterogeneous data preprocessed by the edge computing processing module, as well as the meteorological influence factors and load fluctuation factors in the computing unit;

[0119] Based on the multi-dimensional heterogeneous data pre-processed by the edge computing processing module and the real-time failure rate of the equipment in the power supply unit, the reliability impact factor of the power supply unit is generated. The reliability impact factors include the average number of power outages per month, the average power outage duration, the number of users, and the maintenance response time.

[0120] The reliability influencing factors are used as input variables of the fuzzy membership function, and the input variable reliability influencing factors are mapped to the fuzzy state interval through the fuzzy membership function to generate the fuzzy level of each reliability influencing factor;

[0121] The fuzzy state interval includes three fuzzy levels: low, medium, and high;

[0122] Based on the fuzzy rule base in the rule base, according to the fuzzy level of each reliability influencing factor, the fuzzy result is obtained by matching the rules in the fuzzy rule base through fuzzy reasoning. The fuzzy result is defuzzified by the centroid method to obtain the reliability score inverse value K pi ;

[0123] Calculate the weighted sum of the reliability inverse values of all power supply units to obtain the system reliability inverse value Kp ;

[0124] K p Reflects the reliability level of the entire power distribution system;

[0125] Based on the multi-dimensional heterogeneous data pre-processed by the edge computing processing module and the real-time failure rate of the equipment in the power supply unit, the reliability influencing factors of the power supply unit are generated. The reliability influencing factors include the average monthly number of power outages, the average power outage duration, the number of users, and the maintenance response time. The system reliability inverse value is evaluated more comprehensively to avoid the evaluation bias caused by single data; based on the fuzzy rule base in the rule base, according to the fuzzy level of each reliability influencing factor, the fuzzy results are obtained by matching the rules in the fuzzy rule base through fuzzy reasoning. The fuzzy results are defuzzified by the centroid method to obtain the reliability score inverse value, which solves the nonlinear evaluation problem of multi-factor coupling in the distribution system and improves the evaluation accuracy of the system reliability inverse value; thus, the specific problem area is located through the reliability score inverse value of the power supply unit, and the system reliability inverse value reflects the reliability level of the entire distribution system.

[0126] In an optional embodiment, in the power distribution system reliability assessment model building module, the real-time failure rate of the equipment in the power supply unit is generated in the following manner:

[0127] Based on the equipment basic failure rate database in the rule base;

[0128] Obtain the equipment operating years from the equipment basic data and the load rate from the equipment status monitoring data, and generate the initial failure rate function λ0(t) through the Weibull distribution model;

[0129] Obtain the meteorological influence factor α(T, H) and load fluctuation factor β(L) in the calculation unit;

[0130] The calculation formula for the real-time failure rate λ(t) of the equipment in the power supply unit is:

[0131] λ(t)=λ0(t)×α(T,H)×β(L);

[0132] In the distribution system reliability assessment model construction module, the real-time failure rate of equipment in the power supply unit is generated by integrating equipment basic data, equipment status monitoring data, meteorological influencing factors and load fluctuation factors. This solves the defect that the traditional fixed failure rate model only relies on historical averages and cannot reflect the current operating status of the equipment. The initial failure rate function based on Weibull distribution is combined with dynamically generated meteorological influencing factors and load fluctuation factors to improve the accuracy of real-time fault identification.

[0133] In an optional embodiment, based on the multivariate heterogeneous data preprocessed by the edge computing processing module and the real-time failure rate of the equipment in the power supply unit, the reliability influencing factor of the power supply unit is generated as follows:

[0134] Based on the multivariate heterogeneous data pre-processed by the edge computing processing module, the average monthly power outage times, average power outage duration, number of users, and maintenance response time of the power supply unit are obtained through statistical analysis.

[0135] Based on the multivariate heterogeneous data preprocessed by the edge computing processing module and the real-time failure rate of the equipment in the power supply unit, the equipment health index of the power supply unit is obtained through the neural network algorithm.

[0136] In an optional embodiment, the multi-source data access module also includes an Internet of Things data interface for accessing real-time data collected by smart meters, sensors, and drone inspection equipment to achieve full status monitoring of the power distribution equipment.

[0137] In an optional embodiment, the distribution system reliability assessment model construction module further includes a model updating unit for dynamically adjusting the distribution system reliability assessment model using a particle swarm optimization algorithm based on historical data and real-time feedback to achieve adaptive optimization of the distribution system reliability assessment model.

[0138] In an optional embodiment, the present application transmits data through a 5G communication network or industrial Ethernet to ensure the real-time and reliability of data transmission.

[0139] In an optional embodiment, the intelligent decision-making module integrates an expert knowledge base, which contains distribution system operation and maintenance rules and historical failure cases, to assist in generating targeted rectification suggestions and optimization strategies. The rectification suggestions include an equipment replacement priority sorting algorithm, which determines the equipment maintenance sequence through weighted calculation of the reliability score inverse and the equipment health index.

[0140] like Figure 2 A distribution system reliability assessment method based on cloud computing is shown, comprising the following steps:

[0141] S1, real-time collection of multi-dimensional heterogeneous data of the power distribution system;

[0142] Multivariate heterogeneous data;

[0143] Multi-dimensional heterogeneous data includes power outage data, equipment basic data, user distribution data, geographic information data, real-time meteorological data, equipment status monitoring data, and maintenance data;

[0144] Basic equipment data includes equipment operating years;

[0145] Equipment status monitoring data includes load rate and real-time load;

[0146] Real-time meteorological data includes real-time temperature and real-time humidity;

[0147] S2, preprocessing the multivariate heterogeneous data collected in S1;

[0148] S3, based on the cloud computing architecture, analyzes the multivariate heterogeneous data pre-processed in S2 to obtain meteorological influencing factors and load fluctuation factors;

[0149] S4. Obtain the multi-dimensional heterogeneous data pre-processed in the edge computing processing module, and use a dynamic grid partitioning algorithm to divide the distribution area into multiple power supply units based on the geographic information data and user distribution data in the multi-dimensional heterogeneous data;

[0150] According to the distribution network topology in the power distribution system, a set of physical entities supplying power to the power supply unit is set as a distribution network subunit;

[0151] Each power supply unit corresponds to a power distribution network subunit;

[0152] The set of physical entities includes the distribution lines, transformers, and switchgear that supply power to the supply unit;

[0153] Generate the real-time failure rate of the equipment in the power supply unit based on the basic equipment data and equipment status monitoring data in the multi-heterogeneous data preprocessed by the edge computing processing module, as well as the meteorological influence factors and load fluctuation factors in the computing unit;

[0154] S5. Generate reliability influencing factors of the power supply unit based on the multivariate heterogeneous data preprocessed in S2 and the real-time failure rate of the equipment in the power supply unit in S4. The reliability influencing factors include the average number of power outages per month, the average power outage duration, the number of users, and the maintenance response time.

[0155] The reliability influencing factors are used as input variables of the fuzzy membership function, and the input variable reliability influencing factors are mapped to the fuzzy state interval through the fuzzy membership function to generate the fuzzy level of each reliability influencing factor;

[0156] Based on the fuzzy rule base in the rule base, according to the fuzzy level of each reliability influencing factor, the fuzzy result is obtained by matching the rules in the fuzzy rule base through fuzzy reasoning, and the fuzzy result is defuzzified by the centroid method to obtain the reliability score inverse value;

[0157] Calculate the weighted sum of the reliability inverse values of all power supply units to obtain the system reliability inverse value;

[0158] S6. Generate a reliability assessment report and rectification suggestions based on the reliability inverse value of each power supply unit and the system reliability inverse value output by S5. When the system reliability inverse value exceeds the preset threshold, the equipment inspection or network reconstruction strategy is automatically triggered.

[0159] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0160] In the embodiments provided by the present invention, it should be understood that the disclosed systems or methods can be implemented in other ways. For example, the embodiments of the invention described above are merely illustrative. For example, the division of modules is only a logical function division, and other division methods may be used in actual implementation.

[0161] Modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical modules, and may be located in one place or distributed across multiple network modules. Some or all of these modules may be selected to achieve the objectives of this embodiment based on actual needs.

[0162] In addition, the functional modules in various embodiments of the present invention may be integrated into a single processing module, each module may exist physically separately, or two or more modules may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or hardware plus software functional modules.

[0163] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the basic characteristics of the present invention.

[0164] The above are only preferred specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solutions and inventive concepts of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A power distribution system reliability assessment system based on cloud computing, characterized in that: include: Multi-source data access module, edge computing processing module, cloud computing assessment module, distribution system reliability assessment model construction module and intelligent decision-making module; Multi-source data access module: used to collect multi-source heterogeneous data of the power distribution system in real time; Multi-dimensional heterogeneous data includes power outage data, equipment basic data, user distribution data, geographic information data, real-time meteorological data, equipment status monitoring data, and maintenance data; Basic equipment data includes equipment operating years; Equipment status monitoring data includes load rate and real-time load; Real-time meteorological data includes real-time temperature and real-time humidity; Edge computing processing module: used to pre-process the multi-source heterogeneous data collected in the multi-source data access module; The cloud computing evaluation module includes a rule base and a computing unit; Rule base: including equipment basic failure rate database and fuzzy rule base; Computing unit: Based on the cloud computing architecture, it analyzes the multivariate heterogeneous data pre-processed by the edge computing processing module to obtain meteorological influencing factors and load fluctuation factors; Distribution system reliability assessment model construction module: used to build a distribution system reliability assessment model, which is used to generate the reliability inverse value K of each power supply unit pi and the system-level reliability inverse K p ; Intelligent decision-making module: Based on the reliability evaluation model of the distribution system, the module builds the reliability inverse value K of each power supply unit output pi and the system reliability inverse K p , generate reliability assessment report and rectification suggestions, when the system reliability inverse value K p When the preset threshold is exceeded, device inspection or network reconstruction strategy is automatically triggered.

2. The power distribution system reliability assessment system based on cloud computing according to claim 1, characterized in that: In the edge computing processing module, preprocessing includes data cleaning, format unification, and outlier detection.

3. The power distribution system reliability assessment system based on cloud computing according to claim 1, characterized in that: In the calculation unit, the meteorological influence factor and load fluctuation factor are generated as follows: Assume that the meteorological impact factor is α(T, H), the real-time temperature and real-time humidity in the real-time meteorological data are T and H respectively, and the calculation formula of α(T, H) is: Where γ is the environmental sensitivity coefficient, T0 is the operating temperature of the device under the preset ideal state, and H0 is the operating humidity of the device under the preset ideal state; Assume that the load fluctuation factor is β(L), and the real-time load in the equipment status monitoring data is L. The calculation formula of β(L) is: Where, L n For rated load.

4. The power distribution system reliability assessment system based on cloud computing according to claim 3, characterized in that: In the distribution system reliability assessment model construction module, the distribution system reliability assessment model is constructed as follows: Obtain the multi-dimensional heterogeneous data pre-processed in the edge computing processing module, and use the dynamic grid partitioning algorithm to divide the distribution area into multiple power supply units based on the geographic information data and user distribution data in the multi-dimensional heterogeneous data; According to the distribution network topology in the power distribution system, a set of physical entities supplying power to the power supply unit is set as a distribution network subunit; Each power supply unit corresponds to a power distribution network subunit; The set of physical entities includes the distribution lines, transformers, and switchgear that supply power to the supply unit; Generate the real-time failure rate of the equipment in the power supply unit based on the basic equipment data and equipment status monitoring data in the multi-heterogeneous data preprocessed by the edge computing processing module, as well as the meteorological influence factors and load fluctuation factors in the computing unit; Based on the multi-dimensional heterogeneous data pre-processed by the edge computing processing module and the real-time failure rate of the equipment in the power supply unit, the reliability impact factor of the power supply unit is generated. The reliability impact factors include the average number of power outages per month, the average power outage duration, the number of users, and the maintenance response time. The reliability influencing factors are used as input variables of the fuzzy membership function, and the input variable reliability influencing factors are mapped to the fuzzy state interval through the fuzzy membership function to generate the fuzzy level of each reliability influencing factor; Based on the fuzzy rule base in the rule base, according to the fuzzy level of each reliability influencing factor, the fuzzy result is obtained by matching the rules in the fuzzy rule base through fuzzy reasoning. The fuzzy result is defuzzified by the centroid method to obtain the reliability score inverse value K pi ; Calculate the weighted sum of the reliability inverse values of all power supply units to obtain the system reliability inverse value K p .

5. The power distribution system reliability assessment system based on cloud computing according to claim 4, characterized in that: In the distribution system reliability assessment model construction module, the real-time failure rate of the equipment in the power supply unit is generated in the following way: Based on the equipment basic failure rate database in the rule base; Obtain equipment basic data and equipment status monitoring data, and generate the initial failure rate function λ0(t) through the Weibull distribution model; Obtain the meteorological influence factor α(T, H) and load fluctuation factor β(L) in the calculation unit; The calculation formula for the real-time failure rate λ(t) of the equipment in the power supply unit is: λ(t)=λ0(t)×α(T,H)×β(L).

6. The power distribution system reliability assessment system based on cloud computing according to claim 5, characterized in that: In the distribution system reliability assessment model construction module, based on the multi-dimensional heterogeneous data preprocessed by the edge computing processing module and the real-time failure rate of the equipment in the power supply unit, the reliability influencing factors of the power supply unit are generated as follows: Based on the multivariate heterogeneous data pre-processed by the edge computing processing module, the average monthly power outage times, average power outage duration, number of users, and maintenance response time of the power supply unit are obtained through statistical analysis. Based on the multivariate heterogeneous data preprocessed by the edge computing processing module and the real-time failure rate of the equipment in the power supply unit, the equipment health index of the power supply unit is obtained through the neural network algorithm.

7. The power distribution system reliability assessment system based on cloud computing according to claim 1 is characterized in that: Also includes: Visualization interaction module: used to convert the reliability score inverse value K of each power supply unit generated by the distribution system reliability assessment model construction module pi and the system-level reliability inverse K p , and the reliability assessment report and rectification suggestions generated by the intelligent decision-making module are presented in a visual way.

8. A method for evaluating the reliability of a power distribution system based on cloud computing according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1, real-time collection of multi-dimensional heterogeneous data of the power distribution system; Multivariate heterogeneous data; Multi-dimensional heterogeneous data includes power outage data, equipment basic data, user distribution data, geographic information data, real-time meteorological data, equipment status monitoring data, and maintenance data; Basic equipment data includes equipment operating years; Equipment status monitoring data includes load rate and real-time load; Real-time meteorological data includes real-time temperature and real-time humidity; S2, preprocessing the multivariate heterogeneous data collected in S1; S3, based on the cloud computing architecture, analyzes the multivariate heterogeneous data pre-processed in S2 to obtain meteorological influencing factors and load fluctuation factors; S4. Obtain the multi-dimensional heterogeneous data pre-processed in the edge computing processing module, and use a dynamic grid partitioning algorithm to divide the distribution area into multiple power supply units based on the geographic information data and user distribution data in the multi-dimensional heterogeneous data; According to the distribution network topology in the power distribution system, a set of physical entities supplying power to the power supply unit is set as a distribution network subunit; Each power supply unit corresponds to a power distribution network subunit; The set of physical entities includes the distribution lines, transformers, and switchgear that supply power to the supply unit; Generate the real-time failure rate of the equipment in the power supply unit based on the basic equipment data and equipment status monitoring data in the multi-heterogeneous data preprocessed by the edge computing processing module, as well as the meteorological influence factors and load fluctuation factors in the computing unit; S5. Generate reliability influencing factors of the power supply unit based on the multivariate heterogeneous data preprocessed in S2 and the real-time failure rate of the equipment in the power supply unit in S4. The reliability influencing factors include the average number of power outages per month, the average power outage duration, the number of users, and the maintenance response time. The reliability influencing factors are used as input variables of the fuzzy membership function, and the input variable reliability influencing factors are mapped to the fuzzy state interval through the fuzzy membership function to generate the fuzzy level of each reliability influencing factor; Based on the fuzzy rule base in the rule base, according to the fuzzy level of each reliability influencing factor, the fuzzy result is obtained by matching the rules in the fuzzy rule base through fuzzy reasoning. The fuzzy result is defuzzified by the centroid method to obtain the reliability score inverse value K pi ; Calculate the weighted sum of the reliability inverse values of all power supply units to obtain the system reliability inverse value K p ; S6, according to the output of S5, the reliable inverse value K of each power supply unit pi and the system reliability inverse K p , generate reliability assessment report and rectification suggestions, when the system reliability inverse value K p When the preset threshold is exceeded, device inspection or network reconstruction strategy is automatically triggered.

Images