An intelligent operation and maintenance decision-making method and system for distribution cables

Abstract

An intelligent operation and maintenance decision-making method and system for distribution cables. The method includes: obtaining the operation and maintenance dataset of distribution network cables by using the statistical method and the cable monitoring dataset of the distribution network; calculating the similarity between the operation and maintenance data of distribution network cables and the operation and maintenance standard data based on the similarity algorithm, and extracting the operation and maintenance data with a similarity greater than the set threshold as the operation and maintenance early warning dataset of distribution network cables; performing cluster analysis on the operation and maintenance early warning dataset of distribution network cables to obtain the cluster clusters of potential operation and maintenance hazards of distribution network cables; extracting the top-ranked lines under each operation and maintenance index according to the annual operation and maintenance plan quantity threshold based on the operation and maintenance indexes, and formulating operation and maintenance plans for these lines; extracting the implementation data from the operation and maintenance plans of each line, and judging whether the plan meets the feasibility based on the grid carrying capacity and the fund plan. The present invention proposes a cable maintenance strategy and plan formulation based on the operating state, which is essentially different from the traditional strategy of carrying out regular maintenance on important lines.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent operation and maintenance of power equipment, and particularly relates to a method and system for intelligent operation and maintenance decision-making of distribution cables. Background Art

[0002] With the development of the national economy and the advancement of urbanization, the electricity load is continuously increasing, and the importance of power cables in the urban power grid is becoming increasingly prominent. They have become the "aorta" for urban power transmission. The length of in-service distribution cable lines has exceeded 8 million kilometers, the total number of distribution cable equipment has maintained an annual rapid growth of 15%, the cableization rate of urban power grids has continued to climb and has exceeded 50%, and the cableization rate of some large urban power grids has exceeded 95%.

[0003] With the rapid development of technologies such as "big data, cloud computing, Internet of Things, mobile communication, and artificial intelligence", the traditional distribution network is developing towards digitalization, informatization, and diversification, and is gradually becoming a new generation of distribution and power utilization network with intelligent wisdom, multi-source flexible complementarity, and flexible interaction in multiple user states. In particular, the number of distribution cable lines is increasing year by year, but the operation and maintenance of cables are relatively difficult, and the control requirements for construction, technology, etc. are relatively strict. There is an urgent need for intelligent operation and maintenance decision-making and control methods to effectively ensure their safety and reliability.

[0004] Existing Technology 1 (CN104616090B) "Cable Maintenance Strategy Method Based on Risk Assessment" proposes a cable line maintenance strategy based on risk assessment according to the maintenance characteristics of cable lines, and simultaneously considers the probability of cable line faults and the severity of fault consequences. First, evaluate the cable line unit based on real-time information, calculate the overall health index of the cable line according to the analytic hierarchy process, and then deduce the cable line fault probability; determine the comprehensive risk of the cable line through the asset loss risk and operation loss risk of the cable line itself; finally, decide the cable line maintenance strategy according to the risk level of the cable line comprehensive risk. Existing Technology 1 further conducts risk assessment on the cable line on the basis of cable line status assessment, and formulates a cable line maintenance strategy according to the ALARP criterion, providing a new idea for cable line maintenance. The basis for the maintenance strategy of Existing Technology 1 is the fault probability of the cable line. According to the fault probability calculation formula, the fault probability of important lines is often relatively large, and the risk assessment value is also relatively high. Therefore, the cable maintenance strategy given by Existing Technology 1 will still take important lines as the main objects of maintenance, which highly overlaps with the current regular maintenance strategy and cannot achieve a maintenance strategy based on the operating status. Summary of the Invention

[0005] To address the deficiencies in the existing technologies, the objective of the present invention is to provide an intelligent operation and maintenance decision-making method and system for distribution cables. By statistically analyzing the operation status of distribution cables, the generation of operation and maintenance strategy plans and the intelligent formulation of operation and maintenance repair plans are realized, which is of practical significance for improving power supply reliability. This method defines the cable repair strategy and plan formulation based on the operation status, which is essentially different from the traditional strategy of conducting regular repairs for important lines.

[0006] The present invention adopts the following technical solutions.

[0007] An intelligent operation and maintenance decision-making method for distribution cables, comprising:

[0008] Step 1, collect production management data, distribution network cable operation management data, power grid asset management data, and power supply service command data to form a distribution network cable monitoring data set;

[0009] Step 2, based on statistical methods, use the distribution network cable monitoring data set to obtain a distribution network cable operation and maintenance data set;

[0010] Step 3, use the distribution network cable operation and maintenance standard data to form a distribution network cable operation standard data set. Based on the similarity algorithm, calculate the similarity between the distribution network cable operation and maintenance data and the operation and maintenance standard data, and extract the operation and maintenance data with a similarity greater than the set threshold as the distribution network cable operation and maintenance early warning data set;

[0011] Step 4, based on the k-means clustering method, perform clustering analysis on the distribution network cable operation and maintenance early warning data set to obtain clustering clusters of distribution network cable operation and maintenance hidden dangers; among them, each clustering cluster corresponds to a type of distribution network cable operation and maintenance hidden danger;

[0012] Step 5, according to the operation and maintenance indicators, compare and sort the operation and maintenance assessment data of each line in each clustering cluster; according to the annual operation and maintenance plan quantity threshold, extract the lines ranked in the front under each operation and maintenance indicator, and formulate operation and maintenance plans for these lines;

[0013] Step 6, extract the implementation data from the operation and maintenance plans of each line, and based on the power grid carrying capacity and the capital plan, judge whether the plan meets the feasibility; use the plan that meets the feasibility as the intelligent operation and maintenance decision-making result of the distribution network cable; for the plan that does not meet the feasibility, extract the distribution network cable corresponding to the plan, and repeat steps 4 and 5 for plan correction.

[0014] Preferably, in step 1, the production management data includes: data ledger, service life;

[0015] The distribution network cable operation management data includes: voltage monitoring value, current monitoring value, temperature monitoring value;

[0016] Power grid asset management data includes: manufacturer, tender batch;

[0017] Power supply service command data includes: emergency repair records, line historical power outage information.

[0018] Preferably, step 2 includes:

[0019] Step 2.1, according to the grid structure of the distribution network, based on statistical methods, obtain the operation data of the distribution network cables from the distribution network cable monitoring dataset; among them, the operation data includes: the length, operation years, overload times, and heavy load times of each distribution network cable;

[0020] Step 2.2, according to the construction plan of the distribution network, based on statistical methods, obtain the construction data of the distribution network cables from the distribution network cable monitoring dataset; among them, the construction data includes: the grounding method, type, manufacturer, construction unit, laying method, fire protection, and current status of monitoring device configuration of each distribution network cable;

[0021] Step 2.3, integrate the operation data and the construction data into a distribution network cable operation and maintenance dataset.

[0022] Preferably, in step 2.3, the distribution network cable operation and maintenance dataset includes an operation and maintenance data graph and an operation and maintenance data table.

[0023] Preferably, step 3 includes:

[0024] Step 3.1, extract standard statements and standard values from the management standards and technical standards of the distribution network cables; use the standard statements and standard values as the distribution network cable operation and maintenance standard dataset;

[0025] Step 3.2, divide the distribution network cable operation and maintenance data into operation and maintenance statements and operation and maintenance values;

[0026] Step 3.3, based on the similarity algorithm, calculate the similarity between the operation and maintenance statements and the standard statements, and the similarity between the operation and maintenance values and the standard values respectively;

[0027] Step 3.4, extract the operation data with a similarity greater than the set threshold as the distribution network cable operation and maintenance warning dataset; among them, the value of the set threshold is not less than 0.8.

[0028] Preferably, in step 3.3, calculating the similarity between the operation and maintenance statements and the standard statements includes:

[0029] Step 3.3.1, based on the word embedding technology, for any standard statement, map all the words in the statement to standard vectors, and use the average value of each standard vector as the standard feature vector;

[0030] Step 3.3.2: Based on the word embedding technology, for any operation and maintenance statement, map all the words in the statement into running vectors, and use the average value of each running vector as the running feature vector.

[0031] Step 3.3.3: Based on the cosine similarity algorithm, calculate the cosine value of the angle between the standard feature vector and the operation and maintenance feature vector, and use the cosine value as the similarity between the standard statement and the operation and maintenance statement.

[0032] Preferably, in Step 3.3, based on the Euclidean distance algorithm, calculate the distance value between the operation and maintenance value and the standard value, and use the distance value as the similarity between the operation and maintenance value and the standard value.

[0033] Preferably, in the operation and maintenance data graph and operation and maintenance data table of the distribution network cable, record the operation and maintenance values of each cable line, and at the same time record the standard values of each cable line.

[0034] Preferably, in Step 4, based on the k-means clustering method, perform clustering analysis on the operation and maintenance warning data set of the distribution network cable; among them, the number of clustering clusters is set to 6.

[0035] Each clustering cluster corresponds to the fault type, the number of faults, the power restoration time, the fault occurrence time, the load fluctuation, and the weather condition respectively.

[0036] Preferably, Step 5 includes:

[0037] Step 5.1: Use the number of faults, the power restoration time, the operation years, the number of overloads, and the number of heavy loads as the operation and maintenance indicators of each cable line.

[0038] Step 5.2: In the i-th clustering cluster, extract the operation and maintenance assessment data of l lines in the clustering cluster; the operation and maintenance assessment data includes: the number of faults, the power restoration time, the operation years, the number of overloads, and the number of heavy loads.

[0039] Step 5.3: Under the j-th operation and maintenance indicator, compare and sort the operation and maintenance assessment data of l j cable lines in descending order, and extract the cable lines ranked in the front.

[0040] Step 5.4: Set the weight factor w i for each clustering cluster respectively, and obtain the total number L j of all objects of the j-th operation and maintenance plan according to the following relational expression

[0041]

[0042] where k is the number of clustering clusters; w i ranges from 0 to 1;

[0043] L j It is set according to the quantity threshold of the j-th operation and maintenance plan within a year, and the value is not greater than 50.

[0044] Preferably, in step 5, each operation and maintenance plan includes: power distribution cable transformation plan, priority maintenance plan during Spring Festival, priority maintenance plan in autumn, targeted maintenance plan, power outage plan.

[0045] Preferably, in step 5.4, adjust the weight factor w of each clustering cluster i The value is obtained according to the quantity L of all objects of the j-th operation and maintenance plan j Different numerical values can be obtained.

[0046] Preferably, in step 6, the implementation data extracted from the j-th operation and maintenance plan includes: "N-1" passing rate of distribution network lines, short-circuit capacity, voltage fluctuation, harmonic distortion rate, maximum load rate, investment capacity ratio;

[0047] Using the "N-1" passing rate of distribution network lines, short-circuit capacity, voltage fluctuation and harmonic distortion rate, calculate the safety of the plan; using the maximum load rate and investment capacity ratio to calculate the economy of the plan;

[0048] When the safety of the plan meets the grid carrying capacity and the economy meets the capital plan, it is determined that the plan is feasible.

[0049] A power distribution cable intelligent operation and maintenance decision-making system includes: a power distribution network cable data source module, a cable operation status analysis module, a problem and hidden danger analysis and output module, an operation and maintenance strategy plan generation module, and a plan correction module;

[0050] The power distribution network cable data source module is used to collect production management data, power distribution network cable operation management data, power grid asset management data, and power supply service command data to form a power distribution network cable monitoring data set, and output the power distribution network cable monitoring data set to the cable operation status analysis module;

[0051] The cable operation status analysis module is used to obtain a power distribution network cable operation and maintenance data set from the power distribution network cable monitoring data set based on the statistical method, compare the power distribution network cable operation and maintenance data set with the power distribution network cable operation and maintenance standard data set to obtain a power distribution network cable operation and maintenance warning data set, and output the power distribution network cable operation and maintenance warning data set to the problem and hidden danger analysis and output module;

[0052] The problem and hidden danger analysis and output module is used to perform clustering analysis on the power distribution network cable operation and maintenance warning data set based on the k-means clustering method to obtain clustering clusters of power distribution network cable operation and maintenance hidden dangers, and output the clustering clusters of power distribution network cable operation and maintenance hidden dangers to the operation and maintenance strategy plan generation module;

[0053] An operation and maintenance strategy plan generation module is used to compare and sort the operation and maintenance assessment data of each line in each clustering cluster according to operation and maintenance indicators; extract the lines ranked at the top under each operation and maintenance indicator according to the annual operation and maintenance plan quantity threshold, formulate operation and maintenance plans for these lines, and output the operation and maintenance plans to the plan correction module;

[0054] A plan correction module is used to extract implementation data from the operation and maintenance plans of each line, and judge whether the plan meets the feasibility based on the grid carrying capacity and the fund plan; use the plan that meets the feasibility as the intelligent operation and maintenance decision result of the distribution cable.

[0055] The beneficial effects of the present invention are as follows. Compared with the prior art, by statistically analyzing the operation conditions of distribution cables, the potential problems existing in the operation of the distribution network can be clearly grasped; for the potential problems, according to the operation and maintenance strategy, the intelligent formulation of operation and maintenance repair plans is realized, which has practical significance for improving the power supply reliability. Based on multi-source data, algorithms such as neural networks are used to identify the cable data, and maintenance strategies are proposed by comprehensively considering the operation conditions and the account information. Brief Description of the Drawings

[0056] Figure 1 is a step block diagram of an intelligent operation and maintenance decision method for a distribution cable of the present invention;

[0057] Figure 2 is a schematic diagram of an intelligent operation and maintenance decision system for a distribution cable of the present invention. Detailed Embodiments

[0058] The following further describes the present application with reference to the drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention, and cannot be used to limit the protection scope of the present application.

[0059] As Figure 1 , an intelligent operation and maintenance decision method for a distribution cable includes:

[0060] Step 1, collect production management data, distribution network cable operation management data, power grid asset management data, and power supply service command data to form a distribution network cable monitoring data set.

[0061] Specifically, in Step 1, the production management data includes but is not limited to: data ledger, service life; the distribution network cable operation management data includes but is not limited to: voltage monitoring value, current monitoring value, temperature monitoring value; the power grid asset management data includes but is not limited to: manufacturer, bidding batch; the power supply service command data includes but is not limited to: repair records, line historical power outage information.

[0062] It should be noted that in the preferred embodiment of the present invention, the data extracted from each system is a non - restrictive better choice. Those skilled in the art can select different data according to the cable operation status and operation and maintenance requirements to form a distribution network cable monitoring data set.

[0063] Step 2: Based on statistical methods, use the distribution network cable monitoring data set to obtain a distribution network cable operation and maintenance data set.

[0064] Specifically, step 2 includes:

[0065] Step 2.1: According to the grid structure of the distribution network, based on statistical methods, obtain the operation data of the distribution network cables from the distribution network cable monitoring data set; among them, the operation data includes: the length, operation years, overload times, and heavy load times of each distribution network cable.

[0066] Step 2.2: According to the construction plan of the distribution network, based on statistical methods, obtain the construction data of the distribution network cables from the distribution network cable monitoring data set; among them, the construction data includes: the grounding method, type, manufacturer, construction unit, laying method, fire protection, and current status of monitoring device configuration of each distribution network cable.

[0067] Step 2.3: Integrate the operation data and the construction data into a distribution network cable operation and maintenance data set.

[0068] Furthermore, in step 2.3, the distribution network cable operation and maintenance data set includes an operation and maintenance data graph and an operation and maintenance data table.

[0069] It should be noted that in the preferred embodiment of the present invention, the operation data and construction data extracted from each system are a non - restrictive better choice. Those skilled in the art can select different data according to the cable operation status and operation and maintenance requirements to form a distribution network cable operation and maintenance data set.

[0070] Step 3: Use the distribution network cable operation and maintenance standard data to form a distribution network cable operation standard data set. Based on the similarity algorithm, calculate the similarity between the distribution network cable operation and maintenance data and the operation and maintenance standard data, and extract the operation and maintenance data with a similarity greater than the set threshold as the distribution network cable operation and maintenance warning data set.

[0071] Specifically, step 3 includes:

[0072] Step 3.1: Extract standard statements and standard values from the management standards and technical standards of the distribution network cables; use the standard statements and standard values as the distribution network cable operation and maintenance standard data set.

[0073] Step 3.2: Divide the distribution network cable operation and maintenance data into operation and maintenance statements and operation and maintenance values.

[0074] Step 3.3: Calculate the similarity between the operation and maintenance statements and the standard statements, and the similarity between the operation and maintenance values and the standard values respectively, based on the similarity algorithm.

[0075] Further, in Step 3.3, calculating the similarity between the operation and maintenance statements and the standard statements includes:

[0076] Step 3.3.1: Based on the word embedding technology, for any standard statement, map all the words in the statement to standard vectors, and use the average value of the standard vectors as the standard feature vector.

[0077] Step 3.3.2: Based on the word embedding technology, for any operation and maintenance statement, map all the words in the statement to operation vectors, and use the average value of the operation vectors as the operation feature vector.

[0078] Step 3.3.3: Based on the cosine similarity algorithm, calculate the cosine value of the angle between the standard feature vector and the operation and maintenance feature vector, and use the cosine value as the similarity between the standard statement and the operation and maintenance statement.

[0079] Further, in Step 3.3, based on the Euclidean distance algorithm, calculate the distance value between the operation and maintenance values and the standard values, and use the distance value as the similarity between the operation and maintenance values and the standard values.

[0080] Step 3.4: Extract the operation data with similarity greater than the set threshold as the warning dataset for the operation and maintenance of the distribution network cables; wherein, the value of the set threshold is not less than 0.8.

[0081] It should be noted that the value of the set threshold in the preferred embodiment of the present invention is a non-restrictive and relatively optimal choice, and those skilled in the art can select different thresholds according to the requirements of the operation and maintenance warnings to be output.

[0082] Preferably, in the operation and maintenance data graph and operation and maintenance data table of the distribution network cables, record the operation and maintenance values of each cable line, and at the same time record the standard values of each cable line. Through the operation and maintenance data graph and operation and maintenance data table, the differences between the operation and maintenance data and the operation and maintenance standards can be compared intuitively.

[0083] Step 4: Based on the k-means clustering method, perform clustering analysis on the warning dataset for the operation and maintenance of the distribution network cables to obtain the clustering clusters of the potential hazards in the operation and maintenance of the distribution network cables; wherein, each clustering cluster corresponds to a type of potential hazard in the operation and maintenance of the distribution network cables.

[0084] Specifically, in Step 4, based on the k-means clustering method, perform clustering analysis on the warning dataset for the operation and maintenance of the distribution network cables; in the preferred embodiment of the present invention, the number of clustering clusters is set to 6, and those skilled in the art can set different numbers of clustering clusters according to the number of fault types in engineering operation and maintenance. The setting in the preferred embodiment of the present invention is a non-restrictive and relatively optimal choice.

[0085] Each clustering cluster corresponds to a fault type, the number of faults, the power restoration time, the fault occurrence time, the load fluctuation, and the weather condition.

[0086] Step 5: According to the operation and maintenance indicators, compare and sort the operation and maintenance assessment data of each line in each clustering cluster; extract the lines ranked at the top under each operation and maintenance indicator according to the annual operation and maintenance plan quantity threshold, and formulate operation and maintenance plans for these lines.

[0087] Specifically, Step 5 includes:

[0088] Step 5.1: Use the number of faults, the power restoration time, the operation years, the number of overloads, and the number of heavy loads as the operation and maintenance indicators of each cable line.

[0089] Step 5.2: In the i-th clustering cluster, extract the operation and maintenance assessment data of l lines in this clustering cluster; the operation and maintenance assessment data includes: the number of faults, the power restoration time, the operation years, the number of overloads, and the number of heavy loads.

[0090] Step 5.3: Under the j-th operation and maintenance indicator, compare and sort the operation and maintenance assessment data of l j cable lines in descending order, and extract the cable lines ranked at the top.

[0091] Step 5.4: Set a weight factor w i for each clustering cluster, and obtain the number L j of all objects of the j-th operation and maintenance plan according to the following relational expression

[0092]

[0093] where k is the number of clustering clusters; w i ranges from 0 to 1;

[0094] L j is set according to the quantity threshold of the j-th operation and maintenance plan within a year, and its value is not greater than 50.

[0095] Furthermore, in Step 5.4, adjust the value of the weight factor w i of each clustering cluster. According to the number L j of all objects of the j-th operation and maintenance plan, different values can be obtained.

[0096] Furthermore, in Step 5, each operation and maintenance plan includes: a distribution cable renovation plan, a priority maintenance plan during the Spring Festival, a priority maintenance plan in autumn, a targeted maintenance plan, and a power outage plan.

[0097] Step 6: Extract implementation data from the operation and maintenance plans of each line. Based on the grid carrying capacity and the fund plan, determine whether the plan meets the feasibility; use the plan that meets the feasibility as the decision result of the intelligent operation and maintenance of the distribution network cables; for the plan that does not meet the feasibility, extract the corresponding distribution network cables of the plan, and repeat Steps 4 and 5 for plan correction.

[0098] Preferably, in Step 6, the implementation data extracted from the j-th operation and maintenance plan includes: the passing rate of "N-1" of the distribution network line, the short-circuit capacity, the voltage fluctuation, the harmonic distortion rate, the maximum load rate, and the investment capacity ratio.

[0099] Use the passing rate of "N-1" of the distribution network line, the short-circuit capacity, the voltage fluctuation, and the harmonic distortion rate to calculate the safety of the plan; use the maximum load rate and the investment capacity ratio to calculate the economy of the plan.

[0100] When the safety of the plan meets the grid carrying capacity and the economy meets the fund plan, it is determined that the plan is feasible.

[0101] For the implemented plan, timely implement the operation and maintenance strategy and conduct post-evaluation and optimization, mainly including the implementation and optimization part and the post-evaluation and optimization part.

[0102] The implementation and optimization part is mainly the strategy generated after the above-mentioned correction. Due to unexpected situations encountered during implementation, such as objective impacts such as natural disasters and large-scale epidemics that cannot be implemented as planned, it is reflected in the system and the original operation and maintenance strategy is changed.

[0103] The post-evaluation and optimization part is mainly the strategy change carried out according to the latest data when extremely serious situations occur.

[0104] Such as Figure 2 , an intelligent operation and maintenance decision-making system for distribution cables, including: a distribution network cable data source module, a cable operation status analysis module, a problem and hidden danger analysis and output module, an operation and maintenance strategy plan generation module, and a plan correction module;

[0105] The distribution network cable data source module is used to collect production management data, distribution network cable operation management data, grid asset management data, and power supply service command data to form a distribution network cable monitoring data set, and output the distribution network cable monitoring data set to the cable operation status analysis module;

[0106] The cable operation status analysis module is used to obtain the distribution network cable operation and maintenance data set from the distribution network cable monitoring data set based on the statistical method, compare the distribution network cable operation and maintenance data set with the distribution network cable operation and maintenance standard data set to obtain the distribution network cable operation and maintenance warning data set, and output the distribution network cable operation and maintenance warning data set to the problem and hidden danger analysis and output module;

[0107] A problem and potential risk analysis output module, which is used to perform clustering analysis on the distribution network cable operation and maintenance warning data set based on the k-means clustering method, obtain the clustering clusters of the distribution network cable operation and maintenance hidden dangers, and output the clustering clusters of the distribution network cable operation and maintenance hidden dangers to the operation and maintenance strategy plan generation module;

[0108] An operation and maintenance strategy plan generation module, which is used to compare and sort the operation and maintenance assessment data of each line in each clustering cluster according to the operation and maintenance indicators; extract the lines ranked in the front under each operation and maintenance indicator according to the annual operation and maintenance plan quantity threshold, formulate operation and maintenance plans for these lines, and output the operation and maintenance plans to the plan correction module;

[0109] A plan correction module, which is used to extract the implementation data from the operation and maintenance plans of each line, and judge whether the plan meets the feasibility based on the grid carrying capacity and the fund plan; use the plan that meets the feasibility as the intelligent operation and maintenance decision result of the distribution cable.

[0110] To optimize the intelligent operation and maintenance decision system for distribution cables, it is also necessary to add an operation and maintenance strategy implementation and post-evaluation optimization module, which mainly consists of an implementation optimization part and a post-evaluation optimization part.

[0111] The implementation optimization part mainly refers to the strategy generated after the above-mentioned correction. Due to unexpected situations encountered during implementation, such as objective impacts such as natural disasters and large-scale epidemic diseases that cannot be implemented as planned, it is reflected in the system and the original operation and maintenance strategy is changed.

[0112] The post-evaluation optimization part mainly refers to the strategy change made according to the update of the latest data when extremely serious situations occur.

[0113] The beneficial effects of the present invention are as follows. Compared with the prior art, by statistically analyzing the operation conditions of distribution cables, the hidden danger problems existing in the distribution network operation can be clearly grasped; for the hidden danger problems, according to the operation and maintenance strategies, intelligent formulation of operation and maintenance repair plans is realized, which has practical significance for improving the power supply reliability. Based on multi-source data, algorithms such as neural networks are used to identify the cable data, and maintenance strategies are proposed by comprehensively considering the operation conditions and the account information.

[0114] The applicant of the present invention has made a detailed description and illustration of the embodiments of the present invention in combination with the accompanying drawings of the specification. However, those skilled in the art should understand that the above embodiments are only the preferred implementation schemes of the present invention, and the detailed description is only to help readers better understand the spirit of the present invention, rather than a limitation on the protection scope of the present invention. On the contrary, any improvement or modification made based on the spirit of the present invention should fall within the protection scope of the present invention.

Claims

1. An intelligent operation and maintenance decision-making method for distribution cables, characterized in that, The method includes: Step 1: Collect production management data, distribution network cable operation management data, power grid asset management data, and power supply service command data to form a distribution network cable monitoring dataset. The production management data includes: data ledger, service life. The distribution network cable operation management data includes: voltage monitoring value, current monitoring value, temperature monitoring value. The power grid asset management data includes: manufacturer, tender batch. The power supply service command data includes: repair records, line historical power outage information. Step 2: According to the grid structure of the distribution network, based on statistical methods, obtain the operation data of the distribution network cables from the distribution network cable monitoring dataset. The operation data includes: the length, service life, overload times, and heavy load times of each distribution network cable. According to the construction plan of the distribution network, based on statistical methods, obtain the construction data of the distribution network cables from the distribution network cable monitoring dataset. Among them, the construction data includes: the grounding method, type, manufacturer, construction unit, laying method, fire protection, and current status of monitoring device configuration of each distribution network cable. Integrate the operation data and the construction data into a distribution network cable operation and maintenance dataset. The distribution network cable operation and maintenance dataset includes an operation and maintenance data graph and an operation and maintenance data table. Step 3: Use the distribution network cable operation and maintenance standard data to form a distribution network cable operation standard dataset. Based on the similarity algorithm, calculate the similarity between the distribution network cable operation and maintenance data and the operation and maintenance standard data, and extract the operation and maintenance data with a similarity greater than the set threshold as the distribution network cable operation and maintenance warning dataset. It includes: Step 3.1: Extract standard statements and standard values from the management standards and technical standards of the distribution network cables. Use the standard statements and standard values as the distribution network cable operation and maintenance standard dataset. Step 3.2: Divide the distribution network cable operation and maintenance data into operation and maintenance statements and operation and maintenance values. Step 3.3: Calculate the similarity between the operation and maintenance statements and the standard statements based on the similarity algorithm, and calculate the distance value between the operation and maintenance values and the standard values based on the Euclidean distance algorithm, and use the distance value as the similarity between the operation and maintenance values and the standard values. Calculating the similarity between the operation and maintenance statements and the standard statements includes: Based on the word embedding technology, for any standard statement, map all the words in the statement to standard vectors, and use the average value of each standard vector as the standard feature vector. Based on the word embedding technology, for any operation and maintenance statement, map all the words in the statement to operation vectors, and use the average value of each operation vector as the operation feature vector. Based on the cosine similarity algorithm, calculate the cosine value of the angle between the standard feature vector and the operation and maintenance feature vector, and use the cosine value as the similarity between the standard statement and the operation and maintenance statement. Step 3.4: Extract the operation and maintenance data with a similarity greater than the set threshold as the distribution network cable operation and maintenance warning dataset. Among them, the value of the set threshold is not less than 0.

8. Step 4: Based on the k-means clustering method, perform clustering analysis on the distribution network cable operation and maintenance warning dataset, and set the number of clustering clusters to 6; obtain the clustering clusters of distribution network cable operation and maintenance hidden dangers. Among them, each clustering cluster corresponds to the fault type, fault times, power restoration time, fault occurrence time, load fluctuation, and weather conditions respectively. Step 5: Compare and sort the operation and maintenance assessment data of each line in each cluster according to the operation and maintenance indicators; extract the lines ranked among the top according to each operation and maintenance indicator according to the annual operation and maintenance plan quantity threshold, and formulate operation and maintenance plans for these lines; Step 6: Extract the implementation data from the operation and maintenance plans of each line, and judge whether the plan meets the feasibility based on the grid carrying capacity and the fund plan; use the plan that meets the feasibility as the intelligent operation and maintenance decision result of the distribution network cable; for the plan that does not meet the feasibility, extract the distribution network cable corresponding to the plan, and repeat Steps 4 and 5 for plan correction.

2. The intelligent operation and maintenance decision-making method for distribution cables according to claim 1, characterized in that, In the distribution network cable operation and maintenance data graph and the operation and maintenance data table, the operation and maintenance values of each cable line are recorded, and the standard values of each cable line are also recorded.

3. The intelligent operation and maintenance decision-making method for distribution cables according to claim 1, characterized in that, Step 5 includes: Step 5.1: Use the number of faults, power restoration time, operation years, number of overloads, and number of heavy loads as the operation and maintenance indicators of each cable line; Step 5.2: In the i-th cluster, extract the operation and maintenance assessment data of l lines in the cluster; the operation and maintenance assessment data includes: the number of faults, power restoration time, operation years, number of overloads, and number of heavy loads; Step 5.3, under the j-th operation and maintenance index, compare and sort the operation and maintenance assessment data of l j cable lines in descending order, and extract the cable lines ranked ahead; Step 5.4, set weight factor w for each clustering cluster i , and obtain the number L of all objects of the j-th operation and maintenance plan according to the following relational expression j :[[]]END]] where k is the number of clustering clusters; w i takes values from 0 to 1; L j Set according to the quantity threshold of the j-th operation and maintenance plan within the year, with a value not greater than 50.

4. The intelligent operation and maintenance decision-making method for distribution cables according to claim 3, characterized in that, In Step 5, each operation and maintenance plan includes: distribution cable transformation plan, priority maintenance plan during Spring Festival, priority maintenance plan during autumn, targeted maintenance plan, power outage plan.

5. The intelligent operation and maintenance decision-making method for distribution cables according to claim 3, characterized in that, In step 5.4, adjust the weight factor w of each clustering cluster i value. According to the number L of all objects of the j-th operation and maintenance plan j different numerical values can be obtained.

6. The intelligent operation and maintenance decision-making method for distribution cables according to claim 4, characterized in that, In Step 6, the implementation data extracted from the j-th operation and maintenance plan includes: "N-1" passing rate of the distribution network line, short-circuit capacity, voltage fluctuation, harmonic distortion rate, maximum load rate, investment capacity ratio; Use the "N-1" passing rate, short-circuit capacity, voltage fluctuation, and harmonic distortion rate of the distribution network line to calculate the safety of the plan; use the maximum load rate and investment capacity ratio to calculate the economy of the plan; When the safety of the plan meets the grid carrying capacity and the economy meets the fund plan, it is determined that the plan is feasible.

7. An intelligent operation and maintenance decision-making system for distribution cables implemented by using the intelligent operation and maintenance decision-making method according to any one of claims 1 to 6, characterized in that, The system includes: a distribution network cable data source module, a cable operation status analysis module, a problem and hidden danger analysis and output module, an operation and maintenance strategy plan generation module, and a plan correction module; The distribution network cable data source module is used to collect production management data, distribution network cable operation management data, power grid asset management data, and power supply service command data to form a distribution network cable monitoring data set, and output the distribution network cable monitoring data set to the cable operation status analysis module; the production management data includes: data ledger, years; the distribution network cable operation management data includes: voltage monitoring value, current monitoring value, temperature monitoring value; the power grid asset management data includes: manufacturer, bidding batch; the power supply service command data includes: repair records, line historical power outage information; The cable operation status analysis module is used to obtain the operation data of the distribution network cables from the distribution network cable monitoring dataset based on the grid structure of the distribution network and statistical methods; the operation data includes: the length, operation years, overload times, and heavy load times of each distribution network cable; according to the construction plan of the distribution network, obtain the construction data of the distribution network cables from the distribution network cable monitoring dataset based on statistical methods; among them, the construction data includes: the grounding method, type, manufacturer, construction unit, laying method, fire prevention, and current status of monitoring device configuration of each distribution network cable; integrate the operation data and construction data into the distribution network cable operation and maintenance dataset, and the distribution network cable operation and maintenance dataset includes an operation and maintenance data graph and an operation and maintenance data table; extract standard statements and standard values from the management standards and technical standards of the distribution network cables; use the standard statements and standard values as the distribution network cable operation and maintenance standard dataset; divide the distribution network cable operation and maintenance data into operation and maintenance statements and operation and maintenance values; calculate the similarity between the operation and maintenance statements and the standard statements respectively based on the similarity algorithm, and calculate the distance value between the operation and maintenance values and the standard values based on the Euclidean distance algorithm, and use the distance value as the similarity between the operation and maintenance values and the standard values; calculating the similarity between the operation and maintenance statements and the standard statements includes: based on the word embedding technology, for any standard statement, map all the words in the statement to standard vectors, and use the average value of each standard vector as the standard feature vector; based on the word embedding technology, for any operation and maintenance statement, map all the words in the statement to operation vectors, and use the average value of each operation vector as the operation feature vector; based on the cosine similarity algorithm, calculate the cosine value of the angle between the standard feature vector and the operation and maintenance feature vector, and use the cosine value as the similarity between the standard statement and the operation and maintenance statement; extract the operation data with a similarity greater than the set threshold as the distribution network cable operation and maintenance warning dataset; among them, the value of the set threshold is not less than 0.8; output the distribution network cable operation and maintenance warning dataset to the problem and hidden danger analysis output module; The problem and hidden danger analysis output module is used to perform clustering analysis on the distribution network cable operation and maintenance warning dataset based on the k-means clustering method, and set the number of clustering clusters to 6; obtain the clustering clusters of the distribution network cable operation and maintenance hidden dangers, and each clustering cluster corresponds to the fault type, fault times, power restoration time, fault occurrence time, load fluctuation, and weather conditions respectively; and output the clustering clusters of the distribution network cable operation and maintenance hidden dangers to the operation and maintenance strategy and plan generation module; The operation and maintenance strategy and plan generation module is used to compare and sort the operation and maintenance assessment data of each line in each clustering cluster according to the operation and maintenance indicators; extract the lines ranked in the front under each operation and maintenance indicator according to the annual operation and maintenance plan quantity threshold, formulate operation and maintenance plans for these lines, and output the operation and maintenance plans to the plan correction module; The plan correction module is used to extract the implementation data from the operation and maintenance plans of each line, and judge whether the plan meets the feasibility based on the grid carrying capacity and the capital plan; use the plan that meets the feasibility as the intelligent operation and maintenance decision result of the distribution cable.

Images