Plan generation method and storage medium for differentiated operation and maintenance based on risk identification

Abstract

The application discloses a kind of plan generation methods of differentiating operation and maintenance based on risk identification, comprising: the existing defects of comprehensive equipment, hidden danger, countermeasure and the procedure of pre-trial fixed inspection, through risk degree evaluation, according to e-learn algorithm, according to the importance degree of equipment proposed by power grid annual operation mode from health degree and importance, collate equipment risk matrix, determine the management and control level of each equipment, and form the plan pool of equipment according to management and control level;According to the data of plan pool, according to time period input autoplan algorithm, the priority of plan is confirmed, and plan is generated according to time point.The application provides a kind of differentiating operation and maintenance strategy, that is generated by e-autoplan quickly.

Description

Technical Field

[0001] The present invention relates to the compilation of a production plan for the operation and maintenance of power grid transmission equipment, and in particular to a plan generation method and storage medium for differentiated operation and maintenance based on risk identification. Background Art

[0002] In the power transmission operation and maintenance work of the power grid industry, a large number of on-site inspections are carried out every year according to the health status of the transmission equipment. There are many transmission line towers and cable equipment, and the equipment operates in the natural environment for a long time. In the transmission line channel, it is affected by many factors such as the natural environment and the equipment materials itself, which directly affect the normal operation of the equipment. At present, power grid companies evaluate the monitoring status of power grid transmission equipment through a variety of data such as line defects, hidden dangers, countermeasures, pre-tests and inspections, natural environment data, machine patrol data, etc., and then prepare annual and monthly production plans. A lot of manpower, material and financial resources need to be spent on this work every year. The plan based on manual inspection has low accuracy and is time-consuming and labor-intensive. Summary of the Invention

[0003] The present invention aims to provide a differentiated O&M plan generation method and storage medium based on risk identification. This method utilizes a risk-based assessment that integrates existing equipment defects, hidden dangers, countermeasures, and pre-test inspections. Based on an e-learning algorithm and the importance of equipment proposed for the annual grid operation model, the health and importance of each device are determined, and the control level is compared against the device risk matrix. A plan pool is then generated for each device based on the control level. Data from the plan pool is input into the autoplan algorithm by time period, confirming the plan priority and generating a plan based on the time point.

[0004] The method of the present invention comprises the following steps:

[0005] Step 1: Obtain the defects, hidden dangers, countermeasures, pre-test and scheduled inspection data of the equipment in the corresponding area and store them centrally in the database;

[0006] Step 2: The system performs calculations using the e-learning algorithm. The steps are:

[0007] Step 2.1: Defects are divided into the number of defects and the level of defects; hidden dangers are divided into the number of hidden dangers, the level of hidden dangers, the handling of hidden dangers, the content of countermeasures, the procedures for pre-test and regular inspection, and other factors. Calculate the weight of each dimension, evaluate the hazard level based on the weight ratio, and include it in the plan after determining the hazard level.

[0008] Step 2.2: Test the Kth dimension UK according to the calculation target Si to obtain the index value αik about Si and UK. , , the matrix B'= [aik]n*m is called the evaluation matrix of S on the data set Q. According to experience, we generate the defect weight vector U1= [u11, u12,... u1m], the hidden danger weight vector U2= [u21, u22,... u2m], the countermeasure hidden danger weight vector U3= [u31, u32,... u3m], and the pre-trial inspection U4= [u41, u42,... u4m], to obtain the weight vector matrix U= [ukj]4xm, k=1, 2, 3, 4; wherein ukj is the jth weight in the weight vector obtained under the kth weighting method, that is:

[0009] ,

[0010] And the weight vector satisfies the normalization, that is:

[0011] ,

[0012] The improved deviation function is introduced to determine the weighted least squares optimization model:

[0013] ,

[0014] wherein, is the combined weight, , K=1, 2, 3, 4 is the weight coefficient of u k， k=1, 2, 3, 4, , the e-learn function is constructed

[0015] ,

[0016] The set is formed, and enters ;

[0017] Step 2.3, data access, the database has integrated the required data types, after integration, the plan wherein r is a positive integer, ri reads a new value according to the time attribute, i is a positive number, and , the value of wherein Vt is the value at time t when ri, t, m are positive numbers, and t>0;

[0018] Step 2.4, instance mapping, access data V t Instance mapping, get the plan class data set T={T1, T2,... T e}, wherein e belongs to a positive integer, and e∈[1, n], T⊆r;

[0019] Step 2.5, set data update, take the data V t of the same type Tk Latest time t km , where k∈[1,e]; if the set P u There are t um >t km , then remove T k Monitoring time is earlier than t km The value of V is added to the latest access data t , that is, V tkm ∈T k , V t ∈T k , and t>t km ; Otherwise, only V t Add to collection T k In the middle; aggregate T into the set value calculated by the algorithm;

[0020] Step 2.6, plan pool set is formed, and T set value is imported into the equipment risk matrix, such as Figure 2 shown.

[0021] Generate judgment level value and store it in set P O Medium P O ={P O1 ,P O2 ,…P On}, where n is a positive integer and ∈[1,n], P O ⊆T; P O Save to the database to form a plan pool.

[0022] Step 3: According to the scheduling time, the annual plan is before January of the new year. The plan pool data is input into the e-autoplan algorithm for priority sorting, mainly considering weight allocation. The main steps are:

[0023] Suppose P n Factor P n ={P1,…,p n}, which has an impact on the upper criterion Z, and each time two factors p are taken ni ; and p nj , let the relative importance ratio between the two be a ij The relative importance of all factors is compared using the matrix A = (a ij ) max Indicates that A is called the judgment matrix between ZX. If x i with x j The importance ratio of Z is a ij , then x j with x i The importance ratio of Z is 1 / a ij How to determine a ijFor the value, you can refer to the scale values ​​shown in the following table:

[0024] ;

[0025] Step 4. Generate a plan through the B / s system. When the data volume is large, apply the buffer server and adopt the MyBatis caching mechanism, including the first-level cache and the second-level cache. The first-level cache is also called the local cache. The first-level cache of MyBatis is cached at the session (SqlSession) level. The second-level cache is used to solve the problem that the first-level cache cannot be shared across sessions. The scope is at the namespace level and can be shared by multiple SqlSessions to solve the problem of large data volume. The order of MyBatis querying data is: second-level cache -> first-level cache -> database.

[0026] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program can be executed by a processor to implement the steps of the plan generation method for differentiated operation and maintenance based on risk identification described in the present invention.

[0027] Beneficial effects of the present invention:

[0028] The present invention provides a differentiated operation and maintenance strategy, which quickly generates plans through e-autoplan. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 : Flowchart of the method of the present invention.

[0030] Figure 2 : Equipment risk matrix. DETAILED DESCRIPTION

[0031] like Figure 1 As shown, the method of the present invention comprises the following steps:

[0032] Step 1: Obtain the defects, hidden dangers, countermeasures, pre-test and scheduled inspection data of the equipment in the corresponding area and store them centrally in the database.

[0033] Step 2: The system performs calculations using the e-learning algorithm. The steps are:

[0034] Step 2.1: Defects are categorized into the number of defects and their level. Hidden dangers are categorized into the number of hidden dangers, their level, their handling, countermeasures, and pre-test and inspection procedures. Calculate the weight of each dimension and assess the hazard level based on the weight. Once the hazard level is determined, incorporate it into the plan.

[0035] Step 2.2: Test the Kth dimension UK according to the calculation target Si to obtain the index value αik about Si and UK. , , the matrix B'=[αik]n*m is called the evaluation matrix of S for the data set Q. According to experience, we generate the defect weight vector U1=[u11,u12,...u1m], the hidden danger weight vector generates the defect weight vector U2=[u21,u22,...u2m], the countermeasure hidden danger weight vector generates the defect weight vector U3=[u31,u32,...u3m], and the pre-test inspection U4=[u41,u42,...u4m],

[0036] The weight vector matrix U = [ukj] 4 × m, k = 1, 2, 3, 4 is obtained; where ukj is the jth weight in the weight vector obtained under the kth weighting method, that is:

[0037] ,

[0038] And the weight vector satisfies normalization, that is:

[0039] ,

[0040] Improve the introduction of deviation function and determine the weighted least squares optimization model:

[0041] ,

[0042] in, is the combined weight, ,K=1,2,3,4 is u k， The weight coefficient of k=1, 2, 3, 4, , construct e-learn function

[0043] ,

[0044] The collection formed, enter ;

[0045] Step 2.3, data access, the database has integrated the data types required, and after integration, the plan is obtained. Where r is a positive integer, ri reads the new value according to the time attribute, i is a positive number, and ,get Where Vt is the value of ri at time t, t and m are positive numbers, and t>0;

[0046] Step 2.4, instance mapping, access data V t Instance mapping, the planning class data set T={T1,T2,…T e}, where e is a positive integer, and e∈[1,n], T⊆r;

[0047] Step 2.5, update the collection data, take the data V accessed in the second step t Same type T k Latest time t km , where k∈[1,e]; if the set P u There are t um >t km , then remove T k Monitoring time is earlier than t km The value of V is added to the latest access data t , that is, V tkm ∈T k , V t ∈T k , and t>t km ; Otherwise, only V t Add to collection T k In the algorithm, T is aggregated into a set value calculated by the algorithm.

[0048] Step 2.6, plan pool set is formed, and T set value is imported into the equipment risk matrix, such as Figure 2 shown.

[0049] Generate judgment level value and store it in set P O Medium P O ={P O1 ,P O2 ,…P On}, where n is a positive integer and ∈[1,n], P O ⊆T; P O Save to the database to form a plan pool.

[0050] Step 3: According to the scheduling time, the annual plan is before January of the new year. The plan pool data is input into the e-autoplan algorithm for priority sorting, mainly considering weight allocation. The main steps are:

[0051] Suppose P n Factor P n ={P1,…,p n}, which has an impact on the upper criterion Z, and each time two factors p are taken ni ; and p nj , let the relative importance ratio between the two be a ij The relative importance of all factors is compared using the matrix A = (a ij ) max Indicates that A is called the judgment matrix between ZX. If x i with x j The importance ratio of Z is aij , then x j with x i The importance ratio of Z is 1 / a ij How to determine a ij For the value, you can refer to the scale values ​​shown in the following table:

[0052] ;

[0053] Step 4. Generate a plan through the B / s system. When the data volume is large, apply the buffer server and adopt the MyBatis caching mechanism, including the first-level cache and the second-level cache. The first-level cache is also called the local cache. The first-level cache of MyBatis is cached at the session (SqlSession) level. The second-level cache is used to solve the problem that the first-level cache cannot be shared across sessions. The scope is at the namespace level and can be shared by multiple SqlSessions to solve the problem of large data volume. The order in which MyBatis queries data is: second-level cache -> first-level cache -> database.

Claims

1. A method for generating a differentiated operation and maintenance plan based on risk identification, characterized by: Based on the equipment's existing defects, hidden dangers, countermeasures, and pre-test and inspection procedures, a risk assessment is conducted. Using an e-learning algorithm, the equipment's importance, as determined by the grid's annual operating model, is determined based on health and importance. By comparing the equipment risk matrix, the control level for each device is determined. Based on the control level, a plan pool is created for each device. Data from the plan pool is input into the autoplan algorithm by time period to confirm the plan priority and generate a plan based on the time point. The specific steps include: Step 1: Obtain the defects, hidden dangers, countermeasures, pre-test and scheduled inspection data of the equipment in the corresponding area and store them centrally in the database; Step 2: The system performs calculations using the e-learning algorithm. The steps include: Step 2.1: Defects are categorized into the number of defects and their level. Hidden dangers are categorized into the number of hidden dangers, their level, their handling, countermeasures, and procedures for pre-test inspections. Calculate the weight of each dimension and assess the hazard level based on the weight. Once the hazard level is determined, incorporate it into the plan. Step 2.2: Test the Kth dimension UK according to the calculation target Si to obtain the index value αik about Si and UK. , , the matrix B'=[αik]n*m is called the evaluation matrix of S for the data set Q; Step 2.3, data access, the database has integrated the data types required, and after integration, the plan is obtained. , where r is a positive integer, ri reads the new value according to the time attribute, i is a positive number, and ,get Where Vt is the value of ri at time t, t and m are positive numbers, and t>0; Step 2.4, instance mapping, access data V t Instance mapping, the planning class data set T={T1,T2,…T e }, where e is a positive integer, and e∈[1,n], T⊆r; Step 2.5, update the collection data, take the data V accessed in the second step t Same type T k Latest time t km , where k∈[1,e]; if the set P u There are t um >t km , then remove T k Monitoring time is earlier than t km The value of V is added to the latest access data t , that is, V tkm ∈T k , V t ∈T k , and t> t km ; Otherwise, only V t Add to collection T k In the middle; aggregate T into the set value calculated by the algorithm; Step 2.6: Plan pool set is formed, import the T set value into the equipment risk matrix, generate the judgment level value, and store it in the set P O Medium P O ={P O1 ,P O2 ,…P On }, where n is a positive integer and ∈[1,n], P O ⊆T; P O Save to the database to form a plan pool; Step 3: According to the time of scheduling, the annual plan is before January of the new year, and the plan pool data is input into the e-autoplan algorithm for priority sorting; Step 4: Generate a plan through the B / s system; The step 2.2 also includes: According to the calculation target Si, test according to the K-th dimension UK, and obtain the index value αik about Si and UK. , , the matrix B'=[αik]n*m is called the evaluation matrix of S for the data set Q, generating the defect weight vector U1=[u11,u12,...u1m], the hidden danger weight vector generates the defect weight vector U2=[u21,u22,...u2m], the countermeasure hidden danger weight vector generates the defect weight vector U3=[u31,u32,...u3m], and the pre-test inspection U4=[u41,u42,...u4m], and obtains the weight vector matrix U=[ukj]4×m, k=1,2,3,4; among them, ukj is the jth weight in the weight vector obtained under the kth weighting method, that is: ， And the weight vector satisfies normalization, that is: ， Improve the introduction of deviation function and determine the weighted least squares optimization model: ， in, is the combined weight, ,K=1,2,3,4 is u k， The weight coefficient of k=1, 2, 3, 4, , construct e-learn function ， The collection formed, enter .

2. The plan generation method according to claim 1, characterized in that: In step 3, weight distribution is considered, including: Suppose P n Factor P n ={P1,…,p n }, which has an impact on the upper criterion Z, and each time two factors p are taken ni ; and p nj , let the relative importance ratio between the two be a ij The relative importance of all factors is compared using the matrix A = (a ij ) max Indicates that A is called the judgment matrix between ZX; if x i with x j The importance ratio of Z is a ij , then x j with x i The importance ratio of Z is 1 / a ij .

3. The plan generation method according to claim 2, characterized in that: In step 3, a ij The values ​​refer to the scale values ​​shown in the following table: 。 4. The plan generation method according to claim 1, characterized in that: Step 4 also includes: When the amount of data is large, the application buffer server adopts the MyBatis caching mechanism, including the first-level cache and the second-level cache. The first-level cache is also called the local cache. The first-level cache of MyBatis is cached at the session (SqlSession) level. The second-level cache is used to solve the problem that the first-level cache cannot be shared across sessions. The scope is at the namespace level and can be shared by multiple SqlSessions to solve the problem of large data volume. The order in which MyBatis queries data is: second-level cache -> first-level cache -> database.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program can be executed by a processor to implement the steps of the plan generation method for differentiated operation and maintenance based on risk identification as described in any one of claims 1 to 4.

Images