CN113537770A - Decision tree configuration life prediction method and system based on cloud computing - Google Patents
Decision tree configuration life prediction method and system based on cloud computing Download PDFInfo
- Publication number
- CN113537770A CN113537770A CN202110801130.5A CN202110801130A CN113537770A CN 113537770 A CN113537770 A CN 113537770A CN 202110801130 A CN202110801130 A CN 202110801130A CN 113537770 A CN113537770 A CN 113537770A
- Authority
- CN
- China
- Prior art keywords
- decision tree
- distribution network
- network equipment
- histogram
- cloud computing
- 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
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
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The invention discloses a method and a system for predicting the configuration life of a decision tree based on cloud computing, wherein the method comprises the following steps: acquiring the attribute of the distribution network equipment and fault information of a preset time period; calculating the residual service life of the distribution network equipment according to the distribution network equipment attribute and the fault information of the preset time period based on a decision tree algorithm; and dividing different risk level strategies for the distribution network equipment based on the residual service life, and pushing the risk level strategies to corresponding maintainers. According to the method and the system for predicting the service life of the decision tree configuration based on the cloud computing, the service life is predicted through the cloud algorithm, and automation and intellectualization of the whole service life prediction are achieved.
Description
Technical Field
The invention relates to the technical field of power computers, in particular to a decision tree configuration life prediction method and system based on cloud computing.
Background
With the rapid development of economy in China, the demand of various industries on electricity is increasingly greater, and the phenomenon of unbalanced electricity consumption at different times is increasingly serious. In order to relieve the increasingly sharp power supply and demand contradiction in China, adjust a load curve, improve the phenomenon of unbalanced power consumption, comprehensively implement a peak, average and valley time-sharing power price system, perform peak clipping and valley filling, improve the national power utilization efficiency and reasonably utilize power resources. The usage amount of distribution network equipment is rapidly increased, and the problems of maintenance and arrangement of the distribution network equipment are increasingly important.
After the distribution network equipment generally breaks down, maintenance personnel patrol and repair, for example, the micro-grid measuring equipment, the maintenance personnel disassemble the electric energy meter after the electric energy meter breaks down, and then the intelligent meter needs to be subjected to fault analysis to find out the reason of the fault of the intelligent meter. The hidden defects of the intelligent meter can be found in advance only by regular inspection, but the regular inspection is low in efficiency and long in time consumption, and the intelligent meter is supported by a large amount of human resources under popularization of massive electric energy meters. Other distribution network equipment is maintained by similar routing inspection logics according to different uses and volumes and different maintenance difficulties, but the optimal utilization of resources cannot be achieved all the time, and the importance of the service life prediction of the distribution network equipment is highlighted.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides a method and a system for predicting the service life of decision tree configuration based on cloud computing.
In order to solve the above technical problem, an embodiment of the present invention provides a method for predicting a lifetime of a decision tree configuration based on cloud computing, where the method includes:
acquiring the attribute of the distribution network equipment and fault information of a preset time period;
calculating the residual service life of the distribution network equipment according to the distribution network equipment attribute and the fault information of the preset time period based on a decision tree algorithm;
and dividing different risk level strategies for the distribution network equipment based on the residual service life, and pushing the risk level strategies to corresponding maintainers.
The distribution network equipment attributes comprise: equipment code, manufacturer and service market; the fault information includes: fault phenomenon, abnormal interval, abnormal registration and abnormal code collection.
The step of calculating the remaining life of the distribution network equipment based on the distribution network equipment attribute and the fault information of the preset time period based on a decision tree algorithm comprises the following steps:
first, generation of a decision tree: a process of generating a decision tree from a training sample set;
step two, pruning the decision tree: the pruning of the decision tree is the process of checking, correcting and repairing the decision tree generated at the last stage, and the preliminary rule generated in the process of generating the decision tree is verified by using the data in the new sample data set, so that the branch which influences the accuracy of the pre-balance is pruned.
The step of calculating the remaining life of the distribution network equipment based on the distribution network equipment attribute and the fault information of the preset time period based on a decision tree algorithm comprises the following steps:
and calculating the remaining service life of the distribution network equipment by using a decision tree algorithm based on the histogram.
The step of calculating the remaining life of the distribution network equipment by the histogram-based decision tree algorithm comprises the following steps:
firstly, discretizing continuous floating point characteristic values into k integers, and simultaneously constructing a histogram with the width of k; when data is traversed, statistics are accumulated in the histogram according to the discretized value serving as an index, after the data is traversed once, the histogram accumulates needed statistics, and then the optimal segmentation point is searched in a traversing mode according to the discretized value of the histogram.
The step of dividing the distribution network equipment into different risk level strategies based on the remaining service life comprises the following steps:
marking the electric energy meters with the fault type prediction results at the first level and the residual life prediction results within one week as high risks, and prompting a timely maintenance strategy aiming at the high risks;
marking the electric energy meter with the residual life prediction result less than 30 days as a medium risk, and aiming at the medium risk prompt plan maintenance strategy;
and marking the electric energy meter with the residual life prediction result less than half a year as a low risk, and carrying out a sampling maintenance strategy aiming at the low risk.
Correspondingly, the invention also provides a system for predicting the service life of the decision tree configuration based on cloud computing, which comprises:
the processing module is used for acquiring the attribute of the distribution network equipment and fault information of a preset time period;
the calculation module is used for calculating the residual service life of the distribution network equipment according to the distribution network equipment attribute and the fault information of the preset time period based on a decision tree algorithm;
and the risk module is used for dividing different risk grade strategies for the distribution network equipment based on the residual service life and pushing the risk grade strategies to corresponding maintainers.
The distribution network equipment attributes comprise: equipment code, manufacturer and service market; the fault information includes: fault phenomenon, abnormal interval, abnormal registration and abnormal code collection.
And the calculation module calculates the remaining service life of the distribution network equipment based on a decision tree algorithm of the histogram.
The step of calculating the remaining life of the distribution network equipment by the histogram-based decision tree algorithm comprises the following steps: firstly, discretizing continuous floating point characteristic values into k integers, and simultaneously constructing a histogram with the width of k; when data is traversed, statistics are accumulated in the histogram according to the discretized value serving as an index, after the data is traversed once, the histogram accumulates needed statistics, and then the optimal segmentation point is searched in a traversing mode according to the discretized value of the histogram.
In the embodiment of the invention, firstly, fault data are fully utilized, and abnormal information uploaded by the distribution network equipment before the fault is combined with the characteristics of the distribution network equipment; secondly, a decision tree model algorithm of cloud computing is used, a cloud algorithm can be called to quickly help operation and maintenance personnel to locate the reason of the failure of the distribution network equipment and predict the service life of the distribution network equipment, and the operation and maintenance personnel can conveniently arrange a maintenance plan.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of a method for cloud computing based decision tree configuration life prediction in an embodiment of the invention;
FIG. 2 is a schematic diagram of a decision tree in an embodiment of the invention;
FIG. 3 is a schematic diagram of a histogram algorithm in an embodiment of the invention;
fig. 4 is a schematic structural diagram of a system for predicting the life of a decision tree configuration based on cloud computing in an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and 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.
According to the embodiment of the invention, the abnormal information uploaded by the distribution network equipment before the fault is combined with the characteristics of the distribution network equipment, the characteristics are utilized to train a distribution network equipment fault analysis model based on a decision tree model through a machine learning method, the fault type of the distribution network equipment is judged, the model is trained by using the distribution network equipment record to predict the service life of the distribution network equipment, and operation and maintenance personnel can conveniently arrange a maintenance plan. Meanwhile, with the development of the cloud computing technology, the service life prediction can be carried out by directly calling a cloud algorithm through the API, so that the automation and the intellectualization of the whole service life prediction are realized.
Fig. 1 is a flowchart of a method for predicting a life of a cloud-computing-based decision tree configuration in an embodiment of the present invention, where the method includes the following steps:
the technology for predicting the service life of the distribution network equipment based on the decision tree aims at the problems that the distribution network equipment is long in fault maintenance time and routing inspection depends on a large amount of manpower, and fault data of the distribution network equipment and model characteristic data characteristics of the distribution network equipment are analyzed through machine learning, so that the service life prediction is realized, and operation and maintenance personnel are assisted to arrange a maintenance plan. Meanwhile, the capability of cloud computing is utilized to enable the algorithm and the operation package to provide an API (application programming interface) so that the prediction technology is more intelligent and automatic to realize.
The practical calculation example is microgrid measuring equipment with extremely large popularization, the specific implementation method is that input abnormal data and characteristic data of the electric energy meter are used for uploading to the cloud, the predicted residual life of the electric energy meter is output through a decision tree algorithm, risk grades are divided, high wind is that maintenance is urgently needed, medium risk is planned maintenance, and low risk is that sampling maintenance can be carried out.
S101, acquiring the attribute of distribution network equipment and fault information of a preset time period;
here, the distribution network device attributes include: equipment code, manufacturer and service market; the fault information includes: fault phenomenon, abnormal interval, abnormal registration and abnormal code collection.
The distribution network equipment can be microgrid measuring equipment, and the specific implementation method comprises the steps of uploading input abnormal data and characteristic data of the electric energy meter to a cloud end, outputting the predicted residual life of the electric energy meter through a decision tree algorithm, and dividing risk grades, wherein high wind is urgent, maintenance is required, medium risk is planned maintenance, and low risk is sampling maintenance.
The data of the distribution network equipment comprises two types, wherein the first type of characteristics are the attributes of the distribution network equipment, and the second type of characteristics are the latest alarm information, namely fault information.
Here, the microgrid measurement device is taken as an example:
the first type of feature includes equipment code, manufacturer, length of service.
The service duration is obtained by converting the installation time and the dismantling time into timestamps and subtracting the installation time and the dismantling time from the timestamps, so that the service duration of each electric energy meter is favorably compared. The data codes of the two characteristics are counted, and the distribution conditions of the manufacturer and the equipment codes can be visually seen. The first category of characteristics represents the shipping attributes as well as the operational attributes of the power meter itself.
The second category of characteristics includes fault phenomenon, abnormal interval, abnormal grade, and collected abnormal code.
The abnormal interval is obtained by converting the abnormal time and the installation time into a time stamp minus, and represents how long the last abnormal alarm is away from the installation time. The fault phenomenon and the logic of the abnormal level installation processing manufacturer and the equipment code are also subjected to coding counting processing. The exception code itself is already encoded and processed without further processing. The second type of characteristics is obtained by information sent when the electric energy meter fails, and belongs to dynamic characteristics in the operation of the electric energy meter and fault characteristics in the failure.
Most researches on the service life prediction of the distribution network equipment are based on the information of the distribution network equipment in operation, the distribution network equipment is evaluated and the service life is predicted from the aspect of reliability, and fault data is not fully utilized for analysis. The influence of the attribute of the distribution network equipment on the service life of the electric energy meter is considered, the fault information is also fully considered, and the service life of the distribution network equipment is comprehensively predicted from two aspects.
S102, calculating the residual service life of the distribution network equipment according to the distribution network equipment attribute and the fault information of the preset time period based on a decision tree algorithm;
a decision tree algorithm is a method of approximating discrete function values. It is a typical classification method that first processes the data, generates readable rules and decision trees using a generalisation algorithm, and then uses the decisions to analyze the new data. In essence, the decision tree is a process of classifying data through a series of rules, and how to construct a decision tree with high precision and small scale is the core content of a decision tree algorithm. Decision tree construction can be performed in two steps. First, generation of a decision tree: a process of generating a decision tree from a training sample set. In general, a training sample data set is a data set which has a history according to actual needs and a certain degree of integration and is used for data analysis processing. Step two, pruning the decision tree: the pruning of the decision tree is a process of checking, correcting and repairing the decision tree generated at the previous stage, and is mainly to use data in a new sample data set (called a test data set) to check a preliminary rule generated in the process of generating the decision tree and prune branches influencing the accuracy of pre-balance.
Fig. 2 is a schematic structural diagram of a decision tree in the embodiment of the present invention, in which a root node is a complete set containing samples, an internal node is a corresponding feature attribute test, and a leaf node is a result of decision. Generally, when the attribute represented by a node cannot be judged, the node is divided into 2 child nodes (if the node is not a binary tree, the node is divided into n child nodes), and the attribute test is to select an appropriate threshold value so as to minimize the classification error rate. The decision tree generation algorithm includes ID3, C4.5 and CART, etc., and the most basic ID3 is determined by the Entropy increase (Encopy) principle to be the parent node, and the parent node needs to be split. For a set of data, a smaller entropy indicates a better classification result. Entropy is defined as follows:
wherein, p (x)i) Is class xiThe probability of occurrence, n, is the number of classifications. It can be seen that the magnitude of the entropy is only related to the probability distribution of the variables. For the conditional entropy of Y under the condition of X, which refers to the magnitude of the information amount (uncertainty) of the variable Y after the information of X, the calculation formula is as follows:
for example, when there are only class A and class B, and the probabilities are the same, then the magnitude of the entropy is:
Entropy＝-(0.5log2(0.5)+0.5log2(0.5))＝1
when there is only a class a or B,
Entropy＝-(1log2(1))＝0
so when control is at most 1, it is the state of the worst classification effect, and when it is at least 0, it is the state of the complete classification. Since it is an ideal situation that the entropy is equal to zero, in general practice, the entropy is between 0 and 1.
In the embodiment of the invention, a GBDT (gradient Boosting Decision Tree) model is adopted, iterative training is carried out by utilizing a Decision tree to obtain the most effective model, and the model has good training effect and is not suitable for overfitting. The framework for realizing the GBDT algorithm is a light GBM (light Gradient Boosting machine), which supports high-efficiency parallel training and has the advantages of higher training speed, lower memory consumption, better accuracy, supporting distributed type, and the like, and can quickly process mass data.
LightGBM is a histogram-based decision tree algorithm, and FIG. 3 shows a schematic diagram of a histogram algorithm in an embodiment of the present invention, where the histogram algorithm is to first discretize continuous floating-point feature values into k integers and construct a k-wide histogram. When data is traversed, statistics are accumulated in the histogram according to the discretized value serving as an index, after the data is traversed once, the histogram accumulates needed statistics, and then the optimal segmentation point is searched in a traversing mode according to the discretized value of the histogram.
The histogram has the benefits of less memory footprint and less computational cost. The histogram can be used for difference acceleration, the histogram of one leaf can be obtained by the difference between the histogram of the parent node and the histogram of the brother node, and the speed can be doubled.
And (4) bringing the cleaned and sorted data into a LightGBM framework, and predicting the residual life by utilizing the characteristic data and the fault information data of the distribution network equipment.
And S103, dividing different risk level strategies for the distribution network equipment based on the residual service life, and pushing the risk level strategies to corresponding maintainers.
Here, the classifying the distribution network device into different risk level strategies based on the remaining life includes:
marking the electric energy meters with the fault type prediction results at the first level and the residual life prediction results within one week as high risks, and prompting a timely maintenance strategy aiming at the high risks; marking the electric energy meter with the residual life prediction result less than 30 days as a medium risk, and aiming at the medium risk prompt plan maintenance strategy; and marking the electric energy meter with the residual life prediction result less than half a year as a low risk, and carrying out a sampling maintenance strategy aiming at the low risk.
The output of the algorithm based on the embodiment of the invention is graded according to the possible faults and the predicted residual life of the distribution network equipment so as to reasonably arrange maintenance personnel to overhaul.
By taking the micro-grid measuring equipment as an example, the fault phenomenon of the electric energy meter is predicted in an algorithm frame, and according to the calculation result of the accuracy of hitting the real fault in the most probable N fault types, the Top-1, the Top-3 and the Top-5 can be obtained to be 70.01%, 89.32% and 92.90%. The life prediction accuracy of the model is evaluated by using the real life of the meter in the database, the average absolute percentage error of the life prediction of the microgrid measurement equipment by the model is 8.6%, and the accuracy rate is over 90%. The electric energy meters with the failure type prediction results at Top-1 and the residual life prediction results within one week are marked as high-risk electric energy meters which are urgently required to be overhauled, and other electric energy meters with the residual life prediction results smaller than 30 are marked as medium-risk electric energy meters which are required to be overhauled. The electric energy meter with the residual life prediction result less than half a year can be sampled and overhauled. General distribution network equipment carries out the stepping with this classification logic according to specific service environment and specific distribution network equipment, and maintenance personal carries out maintenance work in more reasonable arrangement. Meanwhile, the prediction algorithm is packaged in the cloud, and the service life prediction function can be realized only by calling the API, so that the whole prediction process is more automatic and more convenient to use.
Accordingly, fig. 4 illustrates a system for cloud computing-based decision tree configuration lifetime prediction in an embodiment of the present invention, where the system includes:
the processing module is used for acquiring the attribute of the distribution network equipment and fault information of a preset time period;
the calculation module is used for calculating the residual service life of the distribution network equipment according to the distribution network equipment attribute and the fault information of the preset time period based on a decision tree algorithm;
and the risk module is used for dividing different risk grade strategies for the distribution network equipment based on the residual service life and pushing the risk grade strategies to corresponding maintainers.
The distribution network equipment attributes comprise: equipment code, manufacturer and service market; the fault information includes: fault phenomenon, abnormal interval, abnormal registration and abnormal code collection.
And the calculation module calculates the remaining service life of the distribution network equipment based on a decision tree algorithm of the histogram.
The step of calculating the remaining life of the distribution network equipment by the histogram-based decision tree algorithm comprises the following steps: firstly, discretizing continuous floating point characteristic values into k integers, and simultaneously constructing a histogram with the width of k; when data is traversed, statistics are accumulated in the histogram according to the discretized value serving as an index, after the data is traversed once, the histogram accumulates needed statistics, and then the optimal segmentation point is searched in a traversing mode according to the discretized value of the histogram.
In the embodiment of the invention, firstly, fault data are fully utilized, and abnormal information uploaded by the distribution network equipment before the fault is combined with the characteristics of the distribution network equipment; secondly, a decision tree model algorithm of cloud computing is used, a cloud algorithm can be called to quickly help operation and maintenance personnel to locate the reason of the failure of the distribution network equipment and predict the service life of the distribution network equipment, and the operation and maintenance personnel can conveniently arrange a maintenance plan.
The foregoing detailed description of the embodiments of the present invention has been presented for the purposes of illustrating the principles and implementations of the present invention and is provided only for the purpose of facilitating an understanding of the principles and core concepts of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A method for predicting equipment risk based on a cloud-computing decision tree, the method comprising:
acquiring the attribute of the distribution network equipment and fault information of a preset time period;
calculating the residual service life of the distribution network equipment according to the distribution network equipment attribute and the fault information of the preset time period based on a decision tree algorithm;
and dividing different risk level strategies for the distribution network equipment based on the residual service life, and pushing the risk level strategies to corresponding maintainers.
2. The method of cloud computing-based decision tree configuration lifetime prediction of claim 1, wherein said distribution network device attributes comprise: equipment code, manufacturer and service market; the fault information includes: fault phenomenon, abnormal interval, abnormal registration and abnormal code collection.
3. The method of cloud computing-based decision tree configuration life prediction as claimed in claim 2, wherein the calculating the remaining life of the distribution network device based on the distribution network device attributes and the fault information of the preset time period based on the decision tree algorithm comprises:
first, generation of a decision tree: a process of generating a decision tree from a training sample set;
step two, pruning the decision tree: the pruning of the decision tree is the process of checking, correcting and repairing the decision tree generated at the last stage, and the preliminary rule generated in the process of generating the decision tree is verified by using the data in the new sample data set, so that the branch which influences the accuracy of the pre-balance is pruned.
4. The method of cloud computing-based decision tree configuration life prediction as claimed in claim 2, wherein the calculating the remaining life of the distribution network device based on the distribution network device attributes and the fault information of the preset time period based on the decision tree algorithm comprises:
and calculating the remaining service life of the distribution network equipment by using a decision tree algorithm based on the histogram.
5. The method of cloud computing-based decision tree configuration life prediction of claim 4, wherein the histogram-based decision tree algorithm calculating remaining life of distribution network equipment comprises:
firstly, discretizing continuous floating point characteristic values into k integers, and simultaneously constructing a histogram with the width of k; when data is traversed, statistics are accumulated in the histogram according to the discretized value serving as an index, after the data is traversed once, the histogram accumulates needed statistics, and then the optimal segmentation point is searched in a traversing mode according to the discretized value of the histogram.
6. The method for cloud computing-based decision tree configuration lifetime prediction according to any one of claims 1 to 5, wherein said partitioning the distribution network devices into different risk level policies based on remaining lifetime comprises:
marking the electric energy meters with the fault type prediction results at the first level and the residual life prediction results within one week as high risks, and prompting a timely maintenance strategy aiming at the high risks;
marking the electric energy meter with the residual life prediction result less than 30 days as a medium risk, and aiming at the medium risk prompt plan maintenance strategy;
and marking the electric energy meter with the residual life prediction result less than half a year as a low risk, and carrying out a sampling maintenance strategy aiming at the low risk.
7. A system for cloud computing based decision tree configuration life prediction, the system comprising:
the processing module is used for acquiring the attribute of the distribution network equipment and fault information of a preset time period;
the calculation module is used for calculating the residual service life of the distribution network equipment according to the distribution network equipment attribute and the fault information of the preset time period based on a decision tree algorithm;
and the risk module is used for dividing different risk grade strategies for the distribution network equipment based on the residual service life and pushing the risk grade strategies to corresponding maintainers.
8. The cloud computing-based decision tree configuration lifetime prediction system of claim 7, wherein said distribution network device attributes comprise: equipment code, manufacturer and service market; the fault information includes: fault phenomenon, abnormal interval, abnormal registration and abnormal code collection.
9. The cloud computing-based decision tree prediction device risk system of claim 8, wherein the computing module calculates a remaining life of the distribution network device based on a histogram based decision tree algorithm.
10. The system for cloud computing-based decision tree configuration life prediction of claim 9, wherein the histogram-based decision tree algorithm calculating remaining life of distribution network devices comprises: firstly, discretizing continuous floating point characteristic values into k integers, and simultaneously constructing a histogram with the width of k; when data is traversed, statistics are accumulated in the histogram according to the discretized value serving as an index, after the data is traversed once, the histogram accumulates needed statistics, and then the optimal segmentation point is searched in a traversing mode according to the discretized value of the histogram.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110801130.5A CN113537770A (en) | 2021-07-15 | 2021-07-15 | Decision tree configuration life prediction method and system based on cloud computing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110801130.5A CN113537770A (en) | 2021-07-15 | 2021-07-15 | Decision tree configuration life prediction method and system based on cloud computing |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113537770A true CN113537770A (en) | 2021-10-22 |
Family
ID=78099479
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110801130.5A Pending CN113537770A (en) | 2021-07-15 | 2021-07-15 | Decision tree configuration life prediction method and system based on cloud computing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113537770A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115225463A (en) * | 2022-09-21 | 2022-10-21 | 江苏牛掌柜科技有限公司 | Hardware fault monitoring method and system based on IT operation and maintenance |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104657130A (en) * | 2015-01-14 | 2015-05-27 | 中国移动通信集团广东有限公司 | Method for hierarchically layering business support system |
| CN105610768A (en) * | 2014-11-25 | 2016-05-25 | 阿里巴巴集团控股有限公司 | Method and device for processing network operation |
| CN107437026A (en) * | 2017-07-13 | 2017-12-05 | 西北大学 | A kind of malicious web pages commercial detection method based on advertising network topology |
| CN109460588A (en) * | 2018-10-22 | 2019-03-12 | 武汉大学 | A kind of equipment fault prediction technique promoting decision tree based on gradient |
| CN109740792A (en) * | 2018-11-29 | 2019-05-10 | 深圳市元征科技股份有限公司 | Data predication method, system, terminal and computer storage medium |
| CN111062517A (en) * | 2019-11-21 | 2020-04-24 | 上海航天智慧能源技术有限公司 | GBDT-based LightGBM model cold and heat load prediction method |
| CN111860624A (en) * | 2020-07-03 | 2020-10-30 | 国网河南省电力公司焦作供电公司 | Power grid fault information classification method based on decision tree |
| CN111999649A (en) * | 2020-08-20 | 2020-11-27 | 浙江工业大学 | XGboost algorithm-based lithium battery residual life prediction method |
| CN112713649A (en) * | 2020-10-22 | 2021-04-27 | 国网浙江省电力有限公司嘉兴供电公司 | Power equipment residual life prediction method based on extreme learning machine |
-
2021
- 2021-07-15 CN CN202110801130.5A patent/CN113537770A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105610768A (en) * | 2014-11-25 | 2016-05-25 | 阿里巴巴集团控股有限公司 | Method and device for processing network operation |
| CN104657130A (en) * | 2015-01-14 | 2015-05-27 | 中国移动通信集团广东有限公司 | Method for hierarchically layering business support system |
| CN107437026A (en) * | 2017-07-13 | 2017-12-05 | 西北大学 | A kind of malicious web pages commercial detection method based on advertising network topology |
| CN109460588A (en) * | 2018-10-22 | 2019-03-12 | 武汉大学 | A kind of equipment fault prediction technique promoting decision tree based on gradient |
| CN109740792A (en) * | 2018-11-29 | 2019-05-10 | 深圳市元征科技股份有限公司 | Data predication method, system, terminal and computer storage medium |
| CN111062517A (en) * | 2019-11-21 | 2020-04-24 | 上海航天智慧能源技术有限公司 | GBDT-based LightGBM model cold and heat load prediction method |
| CN111860624A (en) * | 2020-07-03 | 2020-10-30 | 国网河南省电力公司焦作供电公司 | Power grid fault information classification method based on decision tree |
| CN111999649A (en) * | 2020-08-20 | 2020-11-27 | 浙江工业大学 | XGboost algorithm-based lithium battery residual life prediction method |
| CN112713649A (en) * | 2020-10-22 | 2021-04-27 | 国网浙江省电力有限公司嘉兴供电公司 | Power equipment residual life prediction method based on extreme learning machine |
Non-Patent Citations (1)
| Title |
|---|
| 中国科学技术信息研究所: "《科学计量与评价研究前沿聚焦》", 30 June 2018, 北京：科学技术文献出版社 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115225463A (en) * | 2022-09-21 | 2022-10-21 | 江苏牛掌柜科技有限公司 | Hardware fault monitoring method and system based on IT operation and maintenance |
| CN115225463B (en) * | 2022-09-21 | 2022-12-20 | 江苏牛掌柜科技有限公司 | Hardware fault monitoring method and system based on IT operation and maintenance |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108376982B (en) | Load phase sequence identification method and device | |
| CN112199367B (en) | Long-time scale power balance check and big data analysis method and system | |
| CN110837866A (en) | XGboost-based electric power secondary equipment defect degree evaluation method | |
| CN109308571B (en) | Distribution line variable relation detection method | |
| CN106154209A (en) | Electrical energy meter fault Forecasting Methodology based on decision Tree algorithms | |
| CN111178587B (en) | Spark framework-based short-term power load rapid prediction method | |
| CN106570778A (en) | Big data-based data integration and line loss analysis and calculation method | |
| CN106446016A (en) | Parallel association rule mining-based method for predicting running reliability of power distribution network | |
| CN113496311A (en) | Photovoltaic power station generated power prediction method and system | |
| CN110968703B (en) | Method and system for constructing abnormal metering point knowledge base based on LSTM end-to-end extraction algorithm | |
| CN112285807A (en) | Meteorological information prediction method and device | |
| CN109766416A (en) | A kind of new energy policy information abstracting method and system | |
| CN115034485A (en) | Wind power interval prediction method and device based on data space | |
| CN115859099A (en) | Sample generation method and device, electronic equipment and storage medium | |
| CN113537770A (en) | Decision tree configuration life prediction method and system based on cloud computing | |
| CN114548493A (en) | Method and system for predicting current overload of electric energy meter | |
| CN114154716A (en) | Enterprise energy consumption prediction method and device based on graph neural network | |
| CN113590607A (en) | Electric power marketing report realization method and system based on report factor | |
| CN113891342B (en) | Base station inspection method and device, electronic equipment and storage medium | |
| CN112418662A (en) | Power distribution network operation reliability analysis method using artificial neural network | |
| CN115730749B (en) | Power dispatching risk early warning method and device based on fusion power data | |
| CN112000708A (en) | Abnormal data processing method and system based on regulation and control adapted data fusion | |
| CN115905319B (en) | Automatic identification method and system for abnormal electricity fees of massive users | |
| CN115908082A (en) | Enterprise pollution discharge monitoring method and device based on electricity utilization characteristic indexes | |
| CN113657646A (en) | Method and system for acquiring equipment life prediction in real time based on cloud computing |
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: 20211022 |