CN112836921A - Data monitoring analysis and optimization method based on power grid micro-data fusion - Google Patents
Data monitoring analysis and optimization method based on power grid micro-data fusion Download PDFInfo
- Publication number
- CN112836921A CN112836921A CN202011403624.XA CN202011403624A CN112836921A CN 112836921 A CN112836921 A CN 112836921A CN 202011403624 A CN202011403624 A CN 202011403624A CN 112836921 A CN112836921 A CN 112836921A
- Authority
- CN
- China
- Prior art keywords
- data
- power grid
- index
- project
- indexes
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 77
- 238000004458 analytical method Methods 0.000 title claims abstract description 33
- 238000005457 optimization Methods 0.000 title claims abstract description 28
- 230000004927 fusion Effects 0.000 title claims abstract description 27
- 238000012544 monitoring process Methods 0.000 title claims abstract description 20
- 238000004140 cleaning Methods 0.000 claims abstract description 12
- 238000012216 screening Methods 0.000 claims abstract description 12
- 230000002159 abnormal effect Effects 0.000 claims abstract description 5
- 238000000547 structure data Methods 0.000 claims abstract description 5
- 239000011159 matrix material Substances 0.000 claims description 29
- 238000004422 calculation algorithm Methods 0.000 claims description 22
- 230000001419 dependent effect Effects 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000011156 evaluation Methods 0.000 claims description 11
- 238000007637 random forest analysis Methods 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 6
- 238000007418 data mining Methods 0.000 claims description 6
- 238000004445 quantitative analysis Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 3
- 238000013278 delphi method Methods 0.000 claims description 3
- 238000013178 mathematical model Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000003860 storage Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 8
- 238000012098 association analyses Methods 0.000 abstract description 2
- 230000007547 defect Effects 0.000 abstract description 2
- 238000007726 management method Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 7
- 238000012163 sequencing technique Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 4
- 102100032216 Calcium and integrin-binding protein 1 Human genes 0.000 description 3
- 101000943475 Homo sapiens Calcium and integrin-binding protein 1 Proteins 0.000 description 3
- 101000726148 Homo sapiens Protein crumbs homolog 1 Proteins 0.000 description 3
- 102100027331 Protein crumbs homolog 1 Human genes 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000013486 operation strategy Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 238000005406 washing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
- G06Q10/06375—Prediction of business process outcome or impact based on a proposed change
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Health & Medical Sciences (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
Aiming at the defects of the prior art, the method utilizes the micro-data fusion technology, can carry out targeted screening on the big data before carrying out data association analysis, and effectively improves the relevant indexes of the project. In order to achieve the purpose, the invention is realized by the following technical scheme: a data monitoring analysis and optimization method based on power grid micro-data fusion is applied to electric power projects in designated areas and comprises the following steps: acquiring power grid structure data, power grid operation data and planning project data of a region where the project is located through a data platform; cleaning power grid data based on the power data dictionary, and cleaning or repairing dirty data such as repeated values, abnormal data and the like in the data to reach an available standard; the invention considers the problem that the project in the prior art can not be optimized according to the field situation, provides a method for optimizing the power project by considering the actual environment, and makes up the inadaptability of the original power grid project optimization system.
Description
Technical Field
The invention relates to the field of data analysis, in particular to a power data monitoring, analyzing and optimizing method based on micro data fusion.
Background
With the comprehensive deepening of the power grid service, the management mode and the operation strategy of each power company are deeply influenced. Especially, for actively adapting to the new trend of development, according to the construction thought of a large-scale operation system, an operation management mechanism and a research mechanism are improved, and an operation management online platform which is orderly, optimized, powerful in execution, checked, intensive and efficient is created
At present, most project optimization methods are based on a big data mining technology, the required data volume is huge, the requirement on analysis and calculation efficiency is high, meanwhile, dirty data in a large amount of data are relatively troublesome to process, and management personnel can hardly intervene in the whole analysis process, so that many analysis results cannot be well combined with local actual conditions, the project is built, but limited resources are wasted due to the fact that the method is low in effect.
In order to optimize the feasibility and efficiency of the project, for example, patent application numbers: 201811228414.4 discloses a power grid project two-stage decision optimization method and system, including the following: analyzing factors influencing the power grid project sequencing, and constructing an evaluation index system for optimizing and sequencing the power grid projects; establishing a power grid project optimization sequencing model based on physical element extension according to an evaluation index system of power grid project optimization sequencing; screening power grid items meeting the requirements of corresponding evaluation indexes according to the power grid item optimization sequencing model, and acquiring basic information of the screened power grid items; according to a power grid development target, a power grid development auxiliary decision model based on Weingartner is established in advance, and constraint conditions of power grid projects are determined; the optimal power grid project or combination of the power grid projects is determined based on the power grid investment allowance according to a pre-established power grid development aid decision model, the screened power grid projects and basic information of the power grid projects.
However, in the whole optimization process, the setting of indexes and the like are the manifestation of the intelligence activity of the decision maker. Decisions made by decision makers at different levels are different even if the same information is obtained. Therefore, although such a system can make a decision maker know data information and project dynamics conveniently, it cannot directly bring about optimization of a project.
In addition, from the technical point of view, the analysis method can still meet diversified query requirements in a temporary calculation mode in function, and only meets the historical offline data query requirements in a pre-calculation technology; the method is only suitable for data developers and modeling personnel familiar with data in the aspect of usability, and cannot be directly used by front-line operators, so that the method is not friendly enough.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a data monitoring analysis and optimization method based on power grid micro-data fusion, which utilizes a micro-data fusion technology, can carry out targeted screening on big data before carrying out data association analysis, and finds out an index system suitable for local site characteristics in project association indexes through artificial intelligence and in combination with actual conditions to form a micro-data set, so that the data volume used by subsequent analysis is reduced, the optimal result is directly matched with the local actual conditions, the understanding and execution of a front-line operator are facilitated, and the project-related indexes are effectively improved.
In order to achieve the purpose, the invention is realized by the following technical scheme:
a data monitoring analysis and optimization method based on power grid micro-data fusion is applied to electric power projects in designated areas and comprises the following steps:
step 1: acquiring power grid structure data, power grid operation data and planning project data of a region where the project is located through a data platform;
step 2: cleaning power grid data based on the power data dictionary, and cleaning or repairing dirty data such as repeated values, abnormal data and the like in the data to reach an available standard;
and step 3: performing index screening on the power grid data by using correlation coefficient analysis, and selecting a data index set suitable for the local power grid condition from all data indexes;
and 4, step 4: through an influence factor quantitative analysis data mining method, indexes with large influences on the reliability, adaptability and the like of a power grid in the region are found out from the data index set to form an index micro data set influence factor coefficient;
and 5: substituting indexes of each project to be built by using a fusion weighting method to obtain a power grid project association relation;
step 6: establishing a project dynamic adjustment framework according to the incidence relation of the projects;
and 7: the adjustment of the items is done in the form of a manual or automatic trigger.
Preferably, the step 2 of cleaning the grid data based on the power data dictionary at least includes:
1. processing incomplete data;
2. detecting and processing an error value;
3. detecting and eliminating repeated records;
4. detecting and processing inconsistent data;
the steps are sequentially executed from 1 to 4, and are repeatedly executed after the first round of circulation is finished until the processed numerical value is not changed. The power data obtained is reduced in this manner to remove duplicate records and to convert the remainder to a standard acceptable format. The electric power data cleaning processes the problems of data loss value, out-of-bounds value, inconsistent code, repeated data and the like from the aspects of data accuracy, integrity, consistency, uniqueness, timeliness and effectiveness.
Preferably, the specific steps of the index screening in the step 3 are as follows:
firstly, determining a core index of a power grid of a concerned area;
other grid indexes and core indexes are put together for comparison, the change trend relationship between the two groups of grid indexes is determined in three directions of positive correlation, negative correlation and irrelevance of the index screening, and a mathematical model of correlation coefficients is established;
if the core index data set of the power grid is X and other related index data sets are Y, the covariance of the two sets of data is as follows:
Cov(X，Y)＝E(X-E(X))(Y-E(Y))
where Cov (X, Y) is covariance, i.e. the error of the ensemble of the two sets of data, e (X) and e (Y) are the expected values of the data sets, respectively;
after Cov (X, Y) is obtained, further pass through correlation coefficient rhoXYTo judge the degree of correlation between the two sets of indexes:
correlation coefficient ρXYThe calculation formula of (2) is as follows:
where D (X) is the variance of the reliability index data set, D (Y) is the variance of the hypothetical correlation index data set, ρXYIs a correlation coefficient; by correlation coefficient pXYTo determine the degree of correlation between the core index and the other indexes.
Preferably, the impact factor quantitative analysis data mining method adopted in the step 4 is an embedding method based on a random forest prediction algorithm, and the method specifically comprises the following steps:
1) firstly, determining independent variables and dependent variables of a prediction algorithm;
2) taking 75% of historical dependent variables and characteristic data as an input training set of a random forest prediction algorithm, setting model parameters, and fitting a prediction model;
3) using the residual 25% of historical dependent variables and characteristic data as a test set of a prediction algorithm, predicting the dependent variables by using a prediction model, comparing the dependent variables with an actual dependent variable value, and judging the error of the prediction model;
4) when the error is within the acceptable range, outputting the characteristic importance degree sequence at the moment;
5) and when the error exceeds an allowable value, adjusting the model setting parameters or increasing the features for modeling again until the error is within an acceptable range, and outputting the feature importance degree sequence at the moment.
Further, after the indexes obtained through the random forest prediction algorithm are initially selected, the indexes are analyzed through a grey correlation method to determine the influence of the indexes on the power grid, a preferred index system is further determined, the correlation degree between the indexes and the power grid can be obtained according to the grey correlation degree, the degree of influence of the indexes on the power grid is represented by the magnitude of the correlation degree, and the indexes to be used can be determined through setting a threshold value of the correlation degree.
Wherein grey correlation is generated, the data is firstly standardized. Let Fi＝(fi1，fi2，fi3，...，fij，...，fiN) The ith (i ═ 1, 2., M) sequence of influencing factors, which is the index of influence of the similarity data, is the ith comparison sequence. Where N represents the statistical year or region of the factor and M represents the number of factors.
Data were normalized using the following equation Xi＝(xi1，xi2，xi3，...，xij，…，xiN)：
Determining analysis factors:
by a first step on the reference sequence Ik＝(ik1，ik2，ik3，...，ikj，...，ikN) Normalizing k to 1, 2 to obtain Xk＝(xk1，xk2，xk3，...，xkj，…，xkN)。
Calculating a gray correlation coefficient
Calculate x using the formulaijAnd x0jThe proximity of (a).
Wherein: gamma (x)0j，xij) Is xijAnd x0jThe correlation coefficient and xi are resolution coefficients, and xi is in the scope of [0, 1 ]]Generally, 0.5 is taken, and:
Δij＝|xkj-xij|
calculating the degree of correlation of gray
Reference sequence X0And comparison of sequence XiCan be represented by the following formula
Wherein: r is0iIs a reference sequence X0And comparison of sequence XiThe value of the degree of association of (a).
And obtaining the association degree between the index and the power grid according to the grey association degree, wherein the size of the association degree value represents the influence degree of the index on the power grid, and the index required to be used can be determined by setting a threshold value of the association degree.
Preferably, the step 4 and the step 5 further include the formation of a judgment matrix: and calculating the weight of each evaluation index by combining the high-influence factor coefficient index and utilizing an analytic hierarchy process, namely comparing every two factors by applying a Delphi method, giving a judgment matrix, and verifying the consistency to be qualified to obtain the effective weight.
Preferably, after the judgment matrix is obtained, an analytic hierarchy process is applied to obtain a power supply construction evaluation index system under the project site environment through the following steps:
first, the geometric mean value of each row element is calculated
The formula is as follows:
wherein i is 1, 2.. n, aij is each value in the determination matrix;
will be provided with
And (3) normalization calculation:
wherein, i is 1,2.... n, to give wi＝(w1，w2，...，wn) I.e. the relative weight of each factor.
Calculating the maximum eigenvalue lambda of the judgment matrixmax：
Wherein n is the order of the judgment matrix.
Calculating a consistency index CI
Finding corresponding average random consistency index RI
Calculating CR:
when CR <0.1 is satisfied, the consistency of the judgment matrix is considered to be satisfactory, namely the available feature vector w at the momenti＝(w1，w2，...，wn) As a weight vector; if not, the judgment matrix is reconstructed,
until the consistency check passes.
And 5: substituting indexes of each project to be built by using a fusion weighting method to obtain a power grid project association relation;
step 6: establishing a project dynamic adjustment framework according to the incidence relation of the projects; as shown in fig. 6.
And 7: the adjustment of the items is done in the form of a manual or automatic trigger.
Preferably, a trigger scenario is preset in step 7, the trigger scenario parameters are not seen by an operator, the corresponding trigger scenario at least includes 4 sets of trigger parameters, and when all the trigger parameters are completely matched, the project adjustment is automatically performed, and the update and the sending records are sent to all the members of the project group. This allows the configuration of the items to be automatically adjusted when the trigger condition is reached. Here, similar item information of other areas is added to the database. If the previous project was optimized and brought a good profit, it can be automatically matched into the settings of similar projects when the local project is highly similar to it. This process will inform all project personnel and requires management layer validation before change. However, the management layer only needs to judge whether to execute or not and does not need to make a decision by itself.
In order to facilitate the implementation of the invention, the invention also comprises an operation terminal, which comprises a display part, an operation part, a processor and a storage device, wherein the memory is used for storing the executable instruction of the processor; the processor is configured to perform the steps of the grid micro data fusion based data monitoring analysis and optimization method via execution of the executable instructions.
Preferably, the system further comprises a communication module, which is used for downloading and updating the related algorithm related to the data monitoring analysis and optimization method based on the power grid micro data fusion.
According to the technical scheme, the problems of overlarge analysis data volume and disjointed results and areas where projects are located can be solved by adopting the micro data technology, and the overall reasonability and usability are improved. Through the data fusion technology, the accuracy of project optimization scoring is improved, and therefore optimization accuracy is improved.
Compared with the prior art, the invention mainly achieves the following technical effects:
1. the invention considers the problem that the project in the prior art can not be optimized according to the field situation, and the system provides a method for optimizing the power project by considering the actual environment, thereby making up the inadaptability of the original power grid project optimization system;
2. the micro data set method provided by the invention considers the problem of huge data volume in the traditional optimization method, reduces the indexes with smaller influence, makes the optimization indexes clearer, simplifies the data volume, improves the analysis efficiency and avoids the influence of dirty data on the optimization result.
3. And by means of similarity calculation, processing schemes of similar items in similar scenes are used for reference. When the scenes are very similar, parameters in other scenes can be directly called to adjust the local item.
4. The automatic adjustment of the project is enabled through the project adjustment framework. In this state, the relevant personnel can acquire the adjustment information in time. The management layer only needs to select whether to permit adjustment, and does not need to carry out analysis and decision by itself. The basic level staff can directly know the adjustment content and quickly carry out the work after project adjustment.
Description of the drawings:
FIG. 1 is a schematic diagram of the positive correlation feature of the index screening in step 3 of the present invention.
FIG. 2 is a schematic diagram of negative correlation feature screening by index in step 3 according to the present invention.
FIG. 3 is a schematic diagram of the method for screening irrelevant features by index in step 3.
FIG. 4 is a diagram illustrating the specific steps of the embedding method of the random forest prediction algorithm of the present invention.
FIG. 5 is a flow chart of the gray level correlation used in the present invention.
FIG. 6 is a diagram of a project adjustment framework according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described, and it is obvious that the described embodiments are a part of the embodiments of the present invention, but not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention is described in further detail below with reference to the figures and the detailed description.
The embodiment of the invention takes the data of a certain municipal power grid as the basis and takes the construction project of the power supply to be selected as an example, and the current construction project of the power supply to be selected is shown in the following table.
Project profile of power supply construction project to be optimized in certain city
Step 1: acquiring power grid structure data, power grid operation data and planning project data of a region where the project is located through a data platform; the collected grid structure data forms a feature library. The feature library is established based on standardized project names, content information, function targets, problem codes, user information, national strategy types, civil service types, operation types and lifting directions, and provides a data basis for multi-aspect project association. The project name, the content information and the function target are automatically extracted by a system, the problem code, the user information, the national strategy, the civil service, the operation type and the promotion direction information are input as selectable items in the project input, and if no related information can be filled, the project cannot be associated with the dimension when being associated.
Step 2: cleaning power grid data based on the power data dictionary, and cleaning or repairing dirty data such as repeated values, abnormal data and the like in the data to reach an available standard;
in the invention, firstly, power grid data is cleaned based on a power data dictionary, the obtained power data is simplified to remove repeated records, and the rest part is converted into a standard acceptable format, which specifically comprises the following steps:
1. incomplete data processing
In most cases, the missing grid data values are derived from the data source or other data sources by a data fitting technology, and the missing values are replaced by probability estimation, so that the cleaning purpose is achieved.
2. Method for detecting and solving error value
The data accuracy is ensured from multiple dimensions by identifying possible error values or abnormal values by statistical analysis methods such as deviation analysis, identifying values that do not comply with distributions or regression equations, and the like, and checking the data values against the power data fields in the power data dictionary.
3. Method for detecting and eliminating duplicate records
Records in the power data set having the same attribute value are regarded as duplicate records, and whether the records are equal is detected by determining whether the attribute values between the records are equal, and the equal records are merged into one record.
4. Method for detecting and solving inconsistent data
Because the source of the power data is often the case of multiple data sources, and the power data integrated from the multiple data sources can conflict, the invention detects the inconsistency by defining the power data integrity constraint, so that the data is kept consistent.
In the invention, all the 1-4 are required to be circulated for many times, so that the final numerical value is ensured to be determined without errors. Through electric power data washing, can reach "relative clean" degree with raw data processing, provide reliable basic data for follow-up analysis.
The electrical association is performed by taking a transformer substation and a line as an association main body and extracting power grid infrastructure, technical improvement and overhaul projects related to the same transformer substation and the same line for association.
The facility association is to associate a new construction, a modification, and a major repair occurring in the same association body with an office or production facility purchase item not included in a project, with the infrastructure such as a road, a railway, a port, an office or production room, and the like as the association body. And (3) associating the related power grid infrastructure, technical improvement and overhaul on the same facility with office or production equipment purchasing items which are not contained in the project.
The functional association is that besides electrical association and facility association, according to national policy and major key work requirements of companies, projects with the same construction target are associated with each other, according to national policy and major key work requirements of companies, projects with the same investment target are associated with each other, project functional characteristic elements are extracted, and project association is carried out.
And step 3: performing index screening on the power grid data through correlation coefficient analysis, and selecting a data index set suitable for the local power grid condition from all data indexes;
in the present invention, first, the relevant index to the core index of the local power grid is determined, and the reliability is taken as an example.
All the power grid indexes and the reliability indexes are put together for comparison, the change trend relation between two groups of power grid indexes is determined from the positive correlation direction, the negative correlation direction and the irrelevant direction, and a mathematical model of a correlation coefficient is established and can be represented by the formulas (2-1) and (2-2).
If the grid reliability index data set is X and the associated index data set is Y, the covariance of the two sets of data is:
Cov(X，Y)＝E(X-E(X))(Y-E(Y)) (2-1)
cov (X, Y) is the covariance, i.e. the overall error of the two sets of data is indicated. E (X) and E (Y) are the expected values of the data set, respectively.
The covariance represents the error of the ensemble of the two sets of data, and the Cov (X, Y) is positive if the reliability index dataset is consistent with the trend of the assumed associated index dataset, i.e. if one is greater than its expected value and the other is also greater than its expected value. If the two sets of data have opposite trends, i.e., one is greater than its expected value and the other is less than its expected value, then Cov (X, Y) is negative.
After Cov (X, Y) is obtained, further pass through correlation coefficient rhoXYTo determine the degree of correlation between the two sets of indicators.
Correlation coefficient ρXYThe calculation formula of (2) is as follows:
of the formulaWhere D (X) is the variance of the reliability index data set, D (Y) is the variance of the hypothetical associated index data set, ρXYIs the correlation coefficient.
By correlation coefficient pXYJudging the degree of correlation between the reliability index and the assumed correlation index:
(1) positive correlation
FIG. 1 shows the positive correlation, the left scatter plot shows a generally upward trend, Y increases with increasing X, and the right scatter plots of X and Y show a straight line, which is a complete positive correlation and ρXY＝1。
Correlation coefficient ρXY∈(0，1]The positive correlation relationship between the two is represented, the larger the correlation coefficient is, the stronger the linear correlation between the two is represented, and the scatter diagram in the graph is more concentrated.
(2) Negative correlation
FIG. 2 shows the negative correlation feature, the left plot shows a generally downward trend, Y decreases with increasing X, and the right plot shows a straight line between X and Y, which is the complete negative correlation tangent ρXY＝-1。
Correlation coefficient ρXYE < -1, 0) representing a negative correlation relationship between the two, and the smaller the correlation coefficient, the stronger the linear correlation between the two, and the more concentrated the scatter diagram in the graph.
(3) Is not related
No matter how X changes, Y is not affected at all, and X is also not affected by Y, as shown in fig. 3.
According to the correlation coefficient analysis, the fact that the linear relation between the reliability indexes of the power grid and the indexes is strong, namely the linear correlation is high can be analyzed. The common index for judging the correlation by the correlation coefficient is shown as the following table.
Correlation coefficient and relation correspondence table
After the indexes with strong correlation are obtained, the further analysis can be carried out on the relevant indexes of the reliability of each group of power grids.
According to the method, indexes related to main indexes of the local power grid can be analyzed one by one, and the analyzed indexes are used for forming the micro data set.
Still, the example analysis is performed through the power grid data of a certain city, and the example analysis is performed by taking the selection of the relevant indexes of the power grid reliability indexes as an example.
The reliability index set of this market per year is obtained first as follows, and the data set is denoted by X.
And respectively obtaining other power grid indexes from the information system, wherein the three indexes of capacity margin, 110kV line number and annual power consumption are taken as examples for analysis.
The capacity margins are shown in the following table, using Y for the data set1And (4) showing.
The number of 110kV lines is shown in the following table, and Y is used for data set2And (4) showing.
The annual power consumption is shown in the following table, and the data set is Y3And (4) showing.
The data set was taken into the formula Cov (X, Y) ═ E (X-E (X)) (Y-E (Y))), and the covariance of the three indices of reliability and capacity margin, the number of 110kV lines, and annual power consumption was obtained, and the covariance results were as follows.
COV(X,Y1)＝0.78
COV(X,Y2)＝0.06
COV(X,Y3)＝-0.23
Correlation coefficient rho is further calculatedXYThe following results were obtained.
ρXY1＝0.81
ρXY2＝0.13
ρXY3＝-0.52
According to the result, the capacity margin index and the power grid reliability index have high positive correlation, and the capacity margin index can be used as a judgment basis.
The number of 110kV lines has little relation with the reliability index of the power grid and cannot be used as a judgment index.
The annual power consumption has obvious negative correlation with the reliability index of the power grid, and can be used as an alternative for completely judging and reading basis.
And 4, step 4: through an influence factor quantitative analysis data mining technology, indexes which may influence the reliability, adaptability and the like of a power grid in the region greatly are found out, and an index micro data set influence factor coefficient is formed;
1) embedding method
The embedding method is more complex in principle, and firstly trains by using certain machine learning algorithms and models to obtain the weight coefficient of each feature, and the features are selected according to the weight coefficient from large to small. Similar to the filtering method, but the importance of the features is determined by machine learning training instead of directly determining the importance of the features from some statistical indexes of the features, so the accuracy is higher.
The invention adopts an embedding method to calculate and sort the characteristics of the equipment utilization factor influence factors, and relies on a machine learning algorithm as a random forest prediction algorithm, and the specific flow is as follows:
1. the independent variables and the dependent variables of the prediction algorithm are determined first, and if the main transformer utilization dimension is selected for the project, the dependent variables and the independent variables can be selected as shown in the following table.
2. Taking 75% of historical dependent variables and characteristic data as an input training set of a random forest prediction algorithm, setting model parameters, and fitting a prediction model;
3. using the residual 25% of historical dependent variables and characteristic data as a test set of a prediction algorithm, predicting the dependent variables by using a prediction model, comparing the dependent variables with an actual dependent variable value, and judging the error of the prediction model;
4. if the error is within the acceptable range, outputting the feature importance ranking at the moment, and calculating the importance of the features in the prediction algorithm modeling process;
5. if the error exceeds the allowable value, the model is remodeled by adjusting the setting parameters of the model or increasing the features until the error is within the acceptable range, and the feature importance degree sequence at the moment is output.
The specific flow is shown in fig. 4.
After the indexes obtained by the random forest prediction algorithm are initially selected, the indexes are analyzed by a gray correlation method to determine the influence of the indexes on the power grid, and further determine an optimal index system. As shown in fig. 5:
grey correlation is generated and the data is first normalized. Let Fi＝(fi1，fi2，fi3，...，fij，...，fiN) The ith (i ═ 1, 2., M) sequence of influencing factors, which is the index of influence of the similarity data, is the ith comparison sequence. Where N represents the statistical year or region of the factor and M represents the number of factors.
Data were normalized using the following equation Xi＝(xi1，xi2，xi3，...，xij，...，xiN)：
(1) Determining analytical factors
By a first step on the reference sequence Ik＝(ik1，ik2，ik3，...，ikj，...，ikN) Normalizing k to 1, 2 to obtain Xk＝(xk1，xk2，xk3，...，xkj，...，xkN)。
(2) Calculating a gray correlation coefficient
Calculate x using the formulaijAnd x0jThe proximity of (a).
Wherein: gamma (x)0j，xij) Is xijAnd x0jThe correlation coefficient and xi are resolution coefficients, and xi is in the scope of [0, 1 ]]Generally, 0.5 is taken, and:
Δij＝|xkj-xij|
(3) calculating the degree of correlation of gray
Reference sequence X0And comparison of sequence XiCan be represented by the following formula
Wherein: r is0iIs a reference sequence X0And comparison of sequence XiThe value of the degree of association of (a).
And obtaining the association degree between the index and the power grid according to the grey association degree, wherein the size of the association degree value represents the influence degree of the index on the power grid, and the index required to be used can be determined by setting a threshold value of the association degree.
Through the step 4, primary indexes B1 (scale index), B2 (structure index) and B3 (market index) can be determined to form the micro data set influence factor.
At the moment, the weight of each evaluation index can be calculated by utilizing an analytic hierarchy process, namely, a Delphi method is applied to compare every two factors, a judgment matrix is given, and the weight is effective after consistency verification is passed;
and constructing a judgment matrix by adopting a 1-9 ratio scaling method, and obtaining the relative weight of each factor by applying an analytic hierarchy process. When the weight judgment matrix is constructed, the elements of the upper level are compared pairwise with the elements of the same level as a reference, the relative importance degree of the elements is determined according to a pre-selected evaluation scale, and finally, a quantitative judgment matrix is established according to the relative importance degree, wherein the evaluation scale is shown in the following table.
Judging matrix evaluating scale
The method combines the project site environment, constructs a judgment matrix for the index system, applies an analytic hierarchy process, and obtains the power construction evaluation index system under the project site environment through the following steps:
(1) first, the geometric mean value of each row element is calculated
The formula is as follows:
wherein i is 1, 2.. n, aij is each value in the determination matrix;
(2) will be provided with
And (3) normalization calculation:
wherein i is 1, 2.. n, yielding wi＝(w1，w2，...，wn) I.e. the relative weight of each factor.
(3) Calculating the maximum eigenvalue lambda of the judgment matrixmax：
Wherein n is the order of the judgment matrix.
(4) The consistency index CI is calculated according to the following formula
(5) Finding corresponding average random consistency index RI
Consistency index RI
(6) CR was calculated as follows:
when CR <0.1 is satisfied, the consistency of the matrix is judged to be satisfactory, i.e. the feature vector w obtained in step 2 can be usedi＝(w1，w2，...，wn) As a weight vector; if not, reconstructing the judgment matrix until the consistency check passes, wherein CR is less than 0.1.
Through calculation, the check results of the weight and consistency of the 4 judgment matrixes in the item 1 are as follows:
(1) building coordination overall index weight:
WA～(B1-B3)＝[0.4135 0.2156 0.1590]
λA～(B1-B3)＝3.1251
CIA～(B1-B3)＝0.0137
CRA — (B1-B3) ═ 0.0326<0.1 meets consistency check;
(2) scale index weight:
WB1～(C1-C3)＝[0.5378 0.3799 0.0534]
λB1～(C1-C3)＝3.0278
CIB1～(C1-C3)＝0.0316
CRB1 — (C1-C3) ═ 0.063<0.1, meets consistency check;
(3) structural index weight:
WB2 (C4-C5) ═ 0.650.25 ], and since there are only 2 indices, no consistency check is required;
(4) market index weight:
WB3 (C6-C7) ═ 0.50.5 ], and since there are only 2 indices, no consistency check is required; the 4 judgment matrix weights and consistency check results in item 2 are as follows:
(1) building coordination overall index weight:
WA～(B1-B3)＝[0.4135 0.2156 0.1590]
λA～(B1-B3)＝3.1251
CIA～(B1-B3)＝0.0137
CRA — (B1-B3) ═ 0.0326<0.1 meets consistency check;
(2) scale index weight:
WB1～(C1-C3)＝[0.5032 0.3599 0.0663]
λB1～(C1-C3)＝3.1698
CIB1～(C1-C3)＝0.0339
CRB1 — (C1-C3) ═ 0.0361<0.1, meets the consistency check;
(3) structural index weight:
WB2 (C4-C5) ═ 0.550.35 ], and since there are only 2 indices, no consistency check is required;
(4) market index weight:
WB3 (C6-C7) ═ 0.50.5 ], and since there are only 2 indices, no consistency check is required;
the 4 judgment matrix weights and consistency check results in item 3 are as follows:
(1) building coordination overall index weight:
WA～(B1-B3)＝[0.4313 0.3237 0.1326]
λA～(B1-B3)＝3.236
CIA～(B1-B3)＝0.0126
CRA — (B1-B3) ═ 0.0346<0.1, meets the consistency check;
(2) scale index weight:
WB1～(C1-C3)＝[0.5336 0.3587 0.0524]
λB1～(C1-C3)＝3.2398
CIB1～(C1-C3)＝0.0297
CRB1 — (C1-C3) ═ 0.075<0.1, meets the consistency check;
(3) structural index weight:
WB2 (C4-C5) ═ 0.550.45 ], and since there are only 2 indices, no consistency check is required;
(4) market index weight:
WB3 (C6-C7) ═ 0.50.5 ], and since there are only 2 indices, no consistency check is required;
and 5: substituting indexes of each project to be built by using a fusion weighting method to obtain a power grid project association relation;
step 6: establishing a project dynamic adjustment framework according to the incidence relation of the projects; as shown in fig. 6.
And 7: the adjustment of the items is done in the form of a manual or automatic trigger. And step 7, a trigger scene is preset, the parameters of the trigger scene are not seen by an operator, the corresponding trigger scene at least comprises 4 groups of trigger parameters, and when all the trigger parameters are matched, the project is automatically adjusted, and the update and the sending records are updated and sent to all the members of the project group. This allows the configuration of the items to be automatically adjusted when the trigger condition is reached. Here, similar item information of other areas is added to the database. If the previous project was optimized and brought a good profit, it can be automatically matched into the settings of similar projects when the local project is highly similar to it. This process will inform all project personnel and requires management layer validation before change. However, the management layer only needs to judge whether to execute or not and does not need to make a decision by itself. After the management layer confirms, the actual operator performs the execution according to the changed item content. And the process of the change is recorded in the history.
In order to facilitate the implementation of the invention, the invention also comprises an operation terminal, which comprises a display part, an operation part, a processor and a storage device, wherein the memory is used for storing the executable instruction of the processor; the processor is configured to perform the steps of the grid micro data fusion based data monitoring analysis and optimization method via execution of the executable instructions. The system also comprises a communication module which is used for downloading and updating the related algorithm related to the data monitoring analysis and optimization method based on the power grid micro data fusion. In the downloading process, the user authority of the downloaded device is checked, and whether the device has the authority to download the corresponding content is judged.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A data monitoring analysis and optimization method based on power grid micro-data fusion is applied to electric power projects in designated areas, and is characterized in that: the method comprises the following steps:
step 1: acquiring power grid structure data, power grid operation data and planning project data of a region where the project is located through a data platform;
step 2: cleaning power grid data based on the power data dictionary, and cleaning or repairing dirty data such as repeated values, abnormal data and the like in the data to reach an available standard;
and step 3: performing index screening on the power grid data through correlation coefficient analysis, and selecting a data index set suitable for the local power grid condition from all data indexes;
and 4, step 4: through an influence factor quantitative analysis data mining method, indexes with large influences on the reliability, adaptability and the like of a power grid in the region are found out from the data index set to form an index micro data set influence factor coefficient;
and 5: substituting indexes of each project to be built by using a fusion weighting method to obtain a power grid project association relation;
step 6: establishing a project dynamic adjustment framework according to the incidence relation of the projects;
and 7: the adjustment of the items is done in the form of a manual or automatic trigger.
2. The data monitoring, analyzing and optimizing method based on power grid micro-data fusion as claimed in claim 1, characterized in that: the step 2 of cleaning the power grid data based on the power data dictionary at least comprises the following steps:
1. processing incomplete data;
2. detecting and processing an error value;
3. detecting and eliminating repeated records;
4. detecting and processing inconsistent data;
the steps are sequentially executed from 1 to 4, and are repeatedly executed after the first round of circulation is finished until the processed numerical value is not changed.
3. The data monitoring, analyzing and optimizing method based on power grid micro-data fusion as claimed in claim 1, characterized in that: the specific steps of the index screening in the step 3 are as follows:
firstly, determining a core index of a power grid of a concerned area;
other grid indexes and core indexes are put together for comparison, the change trend relationship between the two groups of grid indexes is determined from the positive correlation direction, the negative correlation direction and the irrelevant direction, and a mathematical model of the correlation coefficient is established;
if the core index data set of the power grid is X and other related index data sets are Y, the covariance of the two sets of data is as follows:
Cov(X，Y)＝E(X-E(X))(Y-E(Y))
where Cov (X, Y) is covariance, i.e. the error of the ensemble of the two sets of data, e (X) and e (Y) are the expected values of the data sets, respectively;
after obtaining Cov (X, Y), the reaction proceedsOne-step pass correlation coefficient rhoXYTo judge the degree of correlation between the two sets of indexes:
correlation coefficient ρXYThe calculation formula of (2) is as follows:
where D (X) is the variance of the reliability index data set, D (Y) is the variance of the hypothetical correlation index data set, ρXYIs a correlation coefficient; by correlation coefficient pXYTo determine the degree of correlation between the core index and the other indexes.
4. The data monitoring, analyzing and optimizing method based on power grid micro-data fusion as claimed in claim 1, characterized in that: the influence factor quantitative analysis data mining method adopted in the step 4 is an embedding method based on a random forest prediction algorithm, and the method specifically comprises the following steps:
1) firstly, determining independent variables and dependent variables of a prediction algorithm;
2) taking 75% of historical dependent variables and characteristic data as an input training set of a random forest prediction algorithm, setting model parameters, and fitting a prediction model;
3) using the residual 25% of historical dependent variables and characteristic data as a test set of a prediction algorithm, predicting the dependent variables by using a prediction model, comparing the dependent variables with an actual dependent variable value, and judging the error of the prediction model;
4) when the error is within the acceptable range, outputting the characteristic importance degree sequence at the moment;
5) and when the error exceeds an allowable value, adjusting the model setting parameters or increasing the features for modeling again until the error is within an acceptable range, and outputting the feature importance degree sequence at the moment.
5. The data monitoring, analyzing and optimizing method based on power grid micro-data fusion as claimed in claim 4, characterized in that: after the indexes obtained through the random forest prediction algorithm are initially selected, the indexes are analyzed through a grey correlation method to determine the influence of the indexes on the power grid, a preferred index system is further determined, the correlation degree between the indexes and the power grid can be obtained according to the grey correlation degree, the degree of the correlation degree represents the influence degree of the indexes on the power grid, and the indexes to be used can be determined through setting a threshold value of the correlation degree.
6. The use method of the data monitoring analysis and optimization method based on the power grid micro-data fusion as claimed in claim 1, wherein: and step 4 and step 5 further comprise the formation of a judgment matrix: and calculating the weight of each evaluation index by combining the high-influence factor coefficient index and utilizing an analytic hierarchy process, namely comparing every two factors by applying a Delphi method, giving a judgment matrix, and verifying the consistency to be qualified to obtain the effective weight.
7. The use method of the data monitoring analysis and optimization method based on the power grid micro-data fusion as claimed in claim 6, characterized in that: after the judgment matrix is obtained, an analytic hierarchy process is applied, and a power supply construction evaluation index system under the project site environment is obtained through the following steps:
first, the geometric mean value of each row element is calculated
The formula is as follows:
wherein i is 1, 2.. n, aij is each value in the determination matrix;
will be provided with
And (3) normalization calculation:
wherein i is 1, 2.. n, yielding wi＝(w1，w2，...，wn) Namely, the relative weight of each factor;
calculating the maximum eigenvalue lambda of the judgment matrixmax：
Wherein n is the order of the judgment matrix;
calculating a consistency index CI
Finding corresponding average random consistency index RI
Calculating CR:
when CR <0.1 is satisfied, the consistency of the judgment matrix is considered to be satisfactory, namely the available feature vector w at the momenti＝(w1，w2，...，wn) As a weight vector; if not, reconstructing the judgment matrix until the consistency check passes, wherein CR is less than 0.1.
8. The data monitoring, analyzing and optimizing method based on power grid micro-data fusion as claimed in claim 1, characterized in that: and step 7, a trigger scene is preset, the parameters of the trigger scene are not seen by an operator, the corresponding trigger scene at least comprises 4 groups of trigger parameters, and when all the trigger parameters are matched, the project is automatically adjusted, and the update and the sending records are updated and sent to all the members of the project group.
9. An operation terminal comprising a display unit, an operation unit, a processor, and a storage device, characterized in that: a memory for storing executable instructions of the processor; the processor is configured to execute the steps of the grid micro data fusion based data monitoring analysis and optimization method of any one of claims 1 to 8 via execution of the executable instructions.
10. The work terminal of claim 9, wherein: the system also comprises a communication module which is used for downloading and updating the related algorithm related to the data monitoring analysis and optimization method based on the power grid micro data fusion.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011403624.XA CN112836921A (en) | 2020-12-02 | 2020-12-02 | Data monitoring analysis and optimization method based on power grid micro-data fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011403624.XA CN112836921A (en) | 2020-12-02 | 2020-12-02 | Data monitoring analysis and optimization method based on power grid micro-data fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112836921A true CN112836921A (en) | 2021-05-25 |
Family
ID=75923472
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011403624.XA Pending CN112836921A (en) | 2020-12-02 | 2020-12-02 | Data monitoring analysis and optimization method based on power grid micro-data fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112836921A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113657648A (en) * | 2021-07-22 | 2021-11-16 | 广州中国科学院沈阳自动化研究所分所 | Multi-dimensional data fusion equipment health assessment method and device and operation and maintenance system |
| CN113703396A (en) * | 2021-07-26 | 2021-11-26 | 北京市机械施工集团有限公司 | Intelligent upgrading method of numerical control cutting equipment based on intelligent terminal |
| CN115713270A (en) * | 2022-11-28 | 2023-02-24 | 之江实验室 | Method and device for detecting and correcting evaluation abnormality of same-bank mutual evaluation |
| CN118157325A (en) * | 2024-05-09 | 2024-06-07 | 北京宏远创信能源科技有限公司 | Real-time monitoring method and system for new energy power |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104331844A (en) * | 2014-11-03 | 2015-02-04 | 江苏省电力公司 | Power network infrastructure project investment decision-making method |
| CN110503399A (en) * | 2019-08-23 | 2019-11-26 | 网易（杭州）网络有限公司 | Method, apparatus, storage medium and the electronic device of project management |
| CN111091229A (en) * | 2019-11-19 | 2020-05-01 | 国网安徽省电力有限公司经济技术研究院 | Accurate investment decision method for power grid infrastructure project |
| CN111259472A (en) * | 2020-01-13 | 2020-06-09 | 上海思优建筑科技有限公司 | Optimal design method and system for building structure |
-
2020
- 2020-12-02 CN CN202011403624.XA patent/CN112836921A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104331844A (en) * | 2014-11-03 | 2015-02-04 | 江苏省电力公司 | Power network infrastructure project investment decision-making method |
| CN110503399A (en) * | 2019-08-23 | 2019-11-26 | 网易（杭州）网络有限公司 | Method, apparatus, storage medium and the electronic device of project management |
| CN111091229A (en) * | 2019-11-19 | 2020-05-01 | 国网安徽省电力有限公司经济技术研究院 | Accurate investment decision method for power grid infrastructure project |
| CN111259472A (en) * | 2020-01-13 | 2020-06-09 | 上海思优建筑科技有限公司 | Optimal design method and system for building structure |
Non-Patent Citations (1)
| Title |
|---|
| 周斌，汪湧: "基于大数据的配变电压异常影响因素分析方法及应用", 《电工技术》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113657648A (en) * | 2021-07-22 | 2021-11-16 | 广州中国科学院沈阳自动化研究所分所 | Multi-dimensional data fusion equipment health assessment method and device and operation and maintenance system |
| CN113657648B (en) * | 2021-07-22 | 2024-03-19 | 广州中国科学院沈阳自动化研究所分所 | Multi-dimensional data fusion equipment health assessment method, device and operation and maintenance system |
| CN113703396A (en) * | 2021-07-26 | 2021-11-26 | 北京市机械施工集团有限公司 | Intelligent upgrading method of numerical control cutting equipment based on intelligent terminal |
| CN115713270A (en) * | 2022-11-28 | 2023-02-24 | 之江实验室 | Method and device for detecting and correcting evaluation abnormality of same-bank mutual evaluation |
| US11989167B1 (en) | 2022-11-28 | 2024-05-21 | Zhejiang Lab | Method and device for detecting and correcting abnormal scoring of peer reviews |
| CN118157325A (en) * | 2024-05-09 | 2024-06-07 | 北京宏远创信能源科技有限公司 | Real-time monitoring method and system for new energy power |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112836921A (en) | Data monitoring analysis and optimization method based on power grid micro-data fusion | |
| CN110245802B (en) | Cigarette empty-head rate prediction method and system based on improved gradient lifting decision tree | |
| US10606862B2 (en) | Method and apparatus for data processing in data modeling | |
| CN112766550B (en) | Random forest-based power failure sensitive user prediction method, system, storage medium and computer equipment | |
| Pati et al. | A comparison among ARIMA, BP-NN, and MOGA-NN for software clone evolution prediction | |
| CN105302848A (en) | Evaluation value calibration method of equipment intelligent early warning system | |
| CN111639783A (en) | Line loss prediction method and system based on LSTM neural network | |
| CN116150897A (en) | Machine tool spindle performance evaluation method and system based on digital twin | |
| CN109344907A (en) | Based on the method for discrimination for improving judgment criteria sorting algorithm | |
| CN111476274B (en) | Big data predictive analysis method, system, device and storage medium | |
| CN112446637A (en) | Building construction quality safety online risk detection method and system | |
| CN113627735A (en) | Early warning method and system for safety risk of engineering construction project | |
| CN115357764A (en) | Abnormal data detection method and device | |
| CN113642922A (en) | Small and medium-sized micro enterprise credit evaluation method and device | |
| CN114638696A (en) | Credit risk prediction model training method and system | |
| CN111612149A (en) | Main network line state detection method, system and medium based on decision tree | |
| CN117132383A (en) | Credit data processing method, device, equipment and readable storage medium | |
| CN116644965A (en) | Visual analysis method and system for project budget | |
| CN115689331A (en) | Power transmission and transformation project quantity rationality analysis method based on MLP | |
| CN112215514A (en) | Operation analysis report generation method and system | |
| Gawne et al. | A computer-based system for modelling the stage-discharge relationships in steady state conditions | |
| CN111612166A (en) | Reimbursement time prediction method based on machine learning | |
| CN117709555B (en) | Carbon emission prediction and evaluation method and system based on transformer carbon accounting model | |
| CN114331137B (en) | Data processing method and device for equipment efficiency evaluation | |
| CN114548494B (en) | Visual cost data prediction intelligent analysis system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210525 |