CN111582338A - Method and system for evaluating running state of electric vehicle charging facility - Google Patents
Method and system for evaluating running state of electric vehicle charging facility Download PDFInfo
- Publication number
- CN111582338A CN111582338A CN202010335544.9A CN202010335544A CN111582338A CN 111582338 A CN111582338 A CN 111582338A CN 202010335544 A CN202010335544 A CN 202010335544A CN 111582338 A CN111582338 A CN 111582338A
- Authority
- CN
- China
- Prior art keywords
- state
- charging facility
- decision tree
- evaluation model
- electric vehicle
- 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.)
- Withdrawn
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000003066 decision tree Methods 0.000 claims abstract description 59
- 238000013210 evaluation model Methods 0.000 claims abstract description 56
- 238000011156 evaluation Methods 0.000 claims abstract description 30
- 238000012795 verification Methods 0.000 claims abstract description 21
- 238000012549 training Methods 0.000 claims abstract description 16
- 238000009966 trimming Methods 0.000 claims abstract description 14
- 238000012216 screening Methods 0.000 claims abstract description 12
- 238000012360 testing method Methods 0.000 claims abstract description 12
- 238000012423 maintenance Methods 0.000 claims description 15
- 238000004891 communication Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 10
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 238000013138 pruning Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 238000005192 partition Methods 0.000 claims description 3
- 238000009413 insulation Methods 0.000 claims description 2
- 238000010801 machine learning Methods 0.000 abstract 1
- 230000008569 process Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- 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
-
- 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
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Entrepreneurship & Innovation (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The application discloses an electric automobile charging facility running state evaluation method and system, wherein the method comprises the following steps: screening state characteristic quantities according to the running state characteristics of the electric vehicle charging facility, distinguishing discrete state characteristic quantities from continuous state characteristic quantities, and dividing the running state types of the charging facility; acquiring historical operating data and test data, evaluating the operating state category of each piece of data, dividing all the acquired data into a training data set and a verification data set, and establishing a decision tree evaluation model for the operating state evaluation of the electric vehicle charging facility; trimming partial nodes in the decision tree evaluation model; and collecting the operation data of the electric automobile charging facility, and evaluating the operation state of the electric automobile charging facility based on the trimmed decision tree evaluation model. The comprehensive, accurate and effective state evaluation model is established based on the machine learning classification method with strong visibilities and readability.
Description
Technical Field
The invention belongs to the technical field of operation maintenance of electric vehicle charging stations, relates to an electric vehicle charging facility operation state evaluation technology, and particularly relates to a method and a system for classifying electric vehicle charging facility operation states by adopting a C4.5 decision tree.
Background
The gradual popularization of electric vehicles has great significance for improving national energy safety, reducing emission, protecting environment and promoting the development of smart power grids. In recent years, a large number of electric vehicle charging stations have been built in cities such as beijing, shanghai, suzhou and the like, and the popularization and application of electric vehicles have entered a critical period.
The electric vehicle charging facility is a necessary condition for maintaining normal charging of the electric vehicle, and is also an important infrastructure in the electric vehicle industry chain. The electric vehicle charging facility mainly comprises a charging station and accessory facilities thereof, such as a rectification module, power distribution equipment, a charging station information acquisition system, safety protection facilities and the like. At present, because facility production technology and the equipment quality of charging are uneven, and the facility operating environment that charges is located the open air mostly, and ambient temperature, humidity, external force damage etc. factor are all uncontrollable, and its running state is changeable, has caused the influence of certain degree to large-scale electric automobile user's demand of charging.
With the continuous expansion of the investment scale of the electric vehicle charging station, the method has great practical significance for establishing an effective and accurate evaluation model for the running state of the charging facility. The state evaluation result not only can provide powerful information support for the operation and maintenance company and the operation and maintenance personnel, but also is convenient for making a scientific and efficient operation and maintenance plan and reasonably distributing operation and maintenance time and operation and maintenance resources.
The state evaluation model of the electric vehicle charging facility is actually a multi-classification task model taking evaluation indexes as input and the overall running state as output. Due to the fact that the structure of equipment in the charging station is complex, the factors influencing the running state of the charging station are many, the traditional evaluation method depends on expert experience too much, the nonlinear relation between the sample characteristics and the classification result is difficult to process, and a comprehensive, effective and accurate evaluation model cannot be established.
The C4.5 Decision Tree (Decision Tree) algorithm is a common supervised learning (Supervised learning) model, the node division basis is Information Gain rate (Information Gain Ratio), the method is commonly used for classification and regression, and the method has the advantages of readability, high classification speed, complete dependence on historical objective operation data and test data, and capability of overcoming judgment errors caused by traditional experience evaluation.
Disclosure of Invention
In order to overcome the defects in the prior art, the application provides the method and the system for evaluating the running state of the electric vehicle charging facility, and the real-time running state of the electric vehicle charging facility is effectively evaluated by taking mass big data of historical records as learning samples and adopting a C4.5 decision tree algorithm based on the distribution structure and the running characteristics of the charging facility.
In order to achieve the above object, the first invention of the present application adopts the following technical solutions:
an electric vehicle charging facility running state assessment method comprises the following steps:
step 1: according to the running state characteristics of the electric vehicle charging facility, screening state characteristic quantities capable of reflecting the whole running state of the charging facility, distinguishing discrete state characteristic quantities from continuous state characteristic quantities, and dividing the running state categories of the charging facility, wherein the running state categories of the charging facility comprise a normal working state, an abnormal early warning state and a serious warning state;
the charging equipment comprises a rectifier cabinet, an internal rectifier module, a charging pile body and a charging gun wire;
the state characteristic quantities reflecting the operation states of the rectifier cabinet and the rectifier module include but are not limited to: inputting alternating current three-phase average voltage, inputting alternating current three-phase average current, frequency, harmonic content grade, alternating current side breaker switching state, normal working module number, outputting direct current voltage, outputting direct current, cabinet temperature, contactor switching state and filter working state;
the state characteristic quantity reflecting the operation state of the charging pile body includes but is not limited to: the charging control system comprises an online state, a card reader communication state, an electric energy meter communication state, an auxiliary power supply voltage, a BMS communication state, a lightning arrester state, a TCU and charging controller communication state and charging pile internal temperature;
the state characteristic quantities reflecting the operating state of the charging gun line include, but are not limited to: the sheath wear state, the insulation condition of the gun head, the fastening condition of the gun head and the temperature of the gun head;
step 2: collecting historical operating data and test data of an electric vehicle charging facility, evaluating discrete state characteristic quantity or continuous state characteristic quantity of each piece of data, and dividing all the collected data into a training data set and a verification data set;
and step 3: establishing a decision tree evaluation model for evaluating the running state of the electric vehicle charging facility on the basis of calculation of the information gain rate;
and 4, step 4: trimming part of nodes in the decision tree evaluation model to prevent overfitting of the model and improve the generalization of the decision tree evaluation model;
and 5: and acquiring real-time operation data of the electric automobile charging facility, and evaluating the operation state of the electric automobile charging facility based on the trimmed decision tree evaluation model.
The invention further comprises the following preferred embodiments:
preferably, the method further comprises: in step 4, a new set of operating data is selected as a test data set for the trimmed decision tree evaluation model, and the accuracy, recall rate and F1 value of the trimmed decision tree evaluation model are verified.
Preferably, the accuracy rate refers to the proportion of the number of samples which can be correctly classified to the total number of samples, and is used for verifying the overall classification effect of the decision tree evaluation model;
the recall rate refers to the proportion of correctly classified samples in each classification label to all samples under the classification label, and is used for measuring the effectiveness of a classification mode of the decision tree evaluation model.
The F1 value is an accuracy recall harmonic mean value of the comprehensive measurement accuracy and recall ratio.
Preferably, in step 1, the normal operating state represents that the charging facility is in a stable operating range, and the state characteristic quantities are all maintained in a normal data range; the abnormal early warning state represents that although the charging facility can still continue to charge, part of components deviate from the normal operation state and cannot continuously operate for a long time, and operation and maintenance personnel are required to carry out remote or on-site inspection; the serious alarm state represents that the operation data of the charging facility is out of limit at the moment, the charging pile is immediately shut down, and the site power failure maintenance of operation and maintenance personnel is arranged.
Preferably, the step 3 of establishing a decision tree evaluation model for the evaluation of the operating state of the electric vehicle charging facility based on the calculation of the information gain ratio specifically includes:
after the information entropy of the training data set is calculated, the information gain rates of all state characteristic quantities in the training data set are calculated, the state characteristic with the largest information gain rate value is selected as the optimal partition attribute of the root node, and a second layer of subset nodes are established according to the value of the state characteristic;
and continuously calculating the information gain rate on the second-layer subset nodes, and selecting the optimal division attribute of the current node, so as to continuously divide the lower-layer subset nodes until the classification of all internal nodes is completed, and forming a decision tree evaluation model for the evaluation of the running state of the electric vehicle charging facility.
Preferably, in step 3, after the information entropy of the training data set is calculated, the information gain rate is directly calculated for the discrete state characteristic quantity; for the continuous state characteristic quantity, a plurality of intermediate points are firstly divided as candidate points by adopting a dichotomy, and then the information gain rate of each point is calculated in sequence.
Preferably, part of the nodes in the trimmed decision tree evaluation model in step 4 specifically are:
traversing the internal nodes of the decision tree evaluation model from bottom to top, judging the current verification precision of each node and replacing the verification precision of each node with a leaf node by adopting a verification data set, and pruning the node to transform the node into the leaf node when the verification data set is higher.
The application also discloses another invention, namely an electric vehicle charging facility running state evaluation system, which comprises a screening module, an acquisition module, a modeling module, a trimming module and an evaluation module;
the screening module is used for screening state characteristic quantities capable of reflecting the integral operation state of the charging facility according to the operation state characteristics of the charging facility of the electric automobile, distinguishing discrete state characteristic quantities and continuous state characteristic quantities and classifying the operation state types of the charging facility;
the acquisition module is used for acquiring historical operating data and test data of the electric automobile charging facility, evaluating discrete state characteristic quantity or continuous state characteristic quantity of each piece of data, and dividing all the acquired data into a training data set and a verification data set;
the modeling module is used for establishing a decision tree evaluation model for the evaluation of the running state of the electric vehicle charging facility on the basis of the calculation of the information gain rate;
the trimming module is used for trimming partial nodes in the decision tree evaluation model so as to prevent overfitting of the model and improve the generalization of the decision tree evaluation model;
the evaluation module is used for acquiring the operation data of the electric automobile charging facility and evaluating the operation state of the electric automobile charging facility based on the trimmed decision tree evaluation model.
The beneficial effect that this application reached:
1. the method and the device form a relatively perfect key characteristic data system of the charging facility, and are convenient for monitoring, analyzing and data sharing of the running state of the charging station;
2. the model has high response speed, and can quickly determine the current running state according to the real-time running data of the electric automobile charging facility;
3. the method has strong visualization, the classification process can be seen clearly and intuitively, and the prediction result is easy to research and analyze;
4. the state evaluation performed by the method is completely based on the objective data of historical operation, does not contain any subjective factor and experience judgment, and the obtained evaluation result accords with the actual operation condition;
5. the method and the device can effectively prevent the over-fitting of the tree model and have high fault tolerance rate.
Drawings
FIG. 1 is a flow chart of a method for evaluating an operating condition of an electric vehicle charging facility according to the present application;
FIG. 2 is a schematic diagram of power transmission of an electric vehicle charging facility according to an embodiment of the present application;
FIG. 3 is a diagram illustrating an evaluation index factor of the state of an electric vehicle charging facility in an embodiment of the present application;
FIG. 4 is a complete decision tree evaluation model for electric vehicle charging facility status evaluation in an embodiment of the present application;
FIG. 5 is a complete decision tree evaluation model tailored in an embodiment of the present application;
fig. 6 is a block diagram of an electric vehicle charging facility operating condition evaluation system according to the present application.
Detailed Description
The present application is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present application is not limited thereby.
As shown in fig. 1, the method for evaluating the operating state of an electric vehicle charging facility of the present application includes the following steps:
step 1: according to the running state characteristics of the electric automobile charging facility, the state characteristic quantity which has strong representativeness and small coupling relation among data and can effectively and quickly reflect the state change is screened. For example, the communication state of the card reader for executing the charging function of the user by the charging pile is directly influenced, the number of working modules which can influence the normal charging speed of the electric automobile is influenced, the two state characteristic quantities reflecting the key functions of the charging pile have strong representativeness to the running state of the charging pile, and the affiliated components also run independently without coupling relation. Distinguishing discrete state characteristic quantity and continuous state characteristic quantity, and dividing the operation state type of the charging facility;
in the embodiment, as shown in fig. 2, electric energy is charged to the electric vehicle from the power distribution network through the processes of voltage reduction by a transformer, rectification by a rectifier cabinet, charging by a charging pile, gun line power transmission and the like. The equipment included in each independent charging process is a rectifying cabinet, an internal rectifying module, a charging pile body and a charging gun wire, so that the running characteristic analysis is performed on the components, and 24 state characteristic quantities are selected as index factors for state evaluation, as shown in fig. 3.
The state characteristic quantity is distinguished from a continuous value, the mark is (D) after the state characteristic quantity is dispersed, and the mark is (C) after the state characteristic quantity is continuous.
The overall operating state of the charging facility is divided into three categories: normal working state, abnormal early warning state and serious warning state. Wherein, the normal working state represents that the charging facility is in a stable operation range, and the state characteristic quantities are all maintained in a normal data range; the abnormal early warning state represents that although the charging facility can still continue to charge, part of components deviate from the normal operation state and cannot continuously operate for a long time, and operation and maintenance personnel are required to carry out remote or on-site inspection; the serious alarm state represents that the operation data of the charging facility is out of limit at the moment, the charging pile is immediately shut down, and the site power failure maintenance of operation and maintenance personnel is arranged.
Step 2: acquiring massive historical operating data and test data of an electric vehicle charging facility, evaluating discrete state characteristic quantity or continuous state characteristic quantity of each piece of data, and dividing all the acquired data into a training data set and a verification data set;
in the embodiment, historical operating data and test data of the electric automobile charging facility are sorted and classified, and each piece of data is evaluated according to discrete state characteristic quantity or continuous state characteristic quantity. Characteristic quantities such as voltage, current, and temperature, whose values are continuously varied, and thus evaluated as continuous-state characteristic quantities; and characteristic quantities such as the contactor switch state, the lightning arrester state and the BMS communication state are operated in a 1 state (working) or a 0 state (failure), so that the characteristic quantities are evaluated as discrete state characteristic quantities.
Writing a training data set as a sample set D { (x)1,y1k),(x2,y2k),L,(xn,ynk) N is the total number of sample data, xnIs a state feature quantity vector of the nth sample, ynkIndicating that the nth sample belongs to the kth operating state class, k being 1,2, L, v, where v is the total number of operating state classes, three operating states are considered in the present invention, and thus v being 3.
And step 3: establishing a decision tree evaluation model for evaluating the running state of the electric vehicle charging facility on the basis of calculation of the information gain rate;
in the embodiment, the information entropy of the whole sample is firstly calculated, and the information gain rate of each feature quantity is calculated on the basis of the information entropy, so that which feature quantity has the largest contribution to the whole information acquisition is deduced and is taken as the current division node.
After the information entropy of the training data set is calculated, the information gain rate is directly calculated for discrete state characteristic quantities such as an on-off state, a communication state and the like, a plurality of middle points are firstly divided as candidate points for continuous state characteristic quantities such as voltage, current, power and the like by adopting a dichotomy, then the information gain rate of each point is sequentially calculated, and the point with the largest information gain rate value is selected as the optimal division point of the state characteristic quantities.
And after calculating the information gain rate of all the state characteristic quantities, selecting the state characteristic with the maximum information gain rate value as the optimal partition attribute of the root node, and establishing a second layer of subset nodes according to the value of the state characteristic.
And continuously calculating the information gain rate on the second-layer subset nodes, and selecting the optimal division attribute of the current node, so as to continuously divide the lower-layer subset nodes until the classification of all internal nodes is completed, and forming a decision tree evaluation model for the evaluation of the running state of the electric vehicle charging facility.
By taking 567 historical data collected by a rapid charging station and a bus charging station of a certain urban area from 2018 to 2019 in Suzhou city as an example, 480 pieces of data are used as a training data set, and 87 pieces of data are used as a verification data set. By the calculation method in the step, the state characteristic quantity g is selected4(harmonic content level of input end in the rectifier cabinet) as the optimal division attribute of the root node. If the harmonic content grade is first grade, continuing to judge the branches at the left end; if the branch is in the second level or the third level, the branch at the right end is continuously judged. Thereby establishing a second level of subset nodes.
The left end node selects the state characteristic quantity g18(charging pile internal lightning arrester state), state characteristic quantity g is selected by right end node11(switch state of contact in rectifier cabinet). The method is extended downward until the classification is completed and a complete decision tree state evaluation model is formed, as shown in fig. 4. The decision tree has 75 nodes, 1 root node, 36 internal nodes and 38 leaf nodes.
And 4, step 4: trimming part of nodes in the decision tree evaluation model to prevent overfitting of the model and promote the generalization of the decision tree evaluation model, and the method specifically comprises the following steps:
traversing the internal nodes of the decision tree evaluation model from bottom to top, judging the current verification precision of each node and replacing the verification precision of each node with a leaf node by adopting a verification data set, and pruning the node to transform the node into the leaf node when the verification data set is higher.
In the embodiment, the decision tree model trimmed by the post-pruning method is shown in fig. 5, the number of nodes is reduced to 63, and the model simplified has better generalization performance and certain improvement in calculation speed. Wherein 1 root node, 30 internal nodes, 32 leaf nodes.
And 5: and acquiring real-time operation data of the electric automobile charging facility, and evaluating the operation state of the electric automobile charging facility based on the trimmed decision tree evaluation model.
In an embodiment, the method further comprises: in step 4, selecting a group of new operation data as a test data set aiming at the trimmed decision tree evaluation model, and verifying the accuracy, the recall rate and the F1 value of the trimmed decision tree evaluation model;
the accuracy rate refers to the proportion of the number of samples which can be correctly classified to the total number of samples, and is used for verifying the overall classification effect of the decision tree evaluation model;
the recall rate refers to the proportion of correctly classified samples in each classification label to all samples under the classification label, and is used for measuring the effectiveness of a classification mode of the decision tree evaluation model.
The F1 value is an accuracy recall harmonic mean value of the comprehensive measurement accuracy and recall ratio.
And selecting 2019 new data of 60 groups to verify the effectiveness of the model: the number of correctly classified samples was 55, the number of correctly classified samples in each label was [36,11,8], thus the accuracy was 91.67%, the recall rate was [ 94.73%, 84.62%, 88.89% ], and the F1 values were [ 93.17%, 88.00%, 90.26% ]. As can be seen, all verification parameters are close to 1, and the feasibility and the effectiveness of the decision tree algorithm on the problem of state evaluation of the electric vehicle charging facility are verified.
As shown in fig. 6, an evaluation system for an operation state of an electric vehicle charging facility includes a screening module, an acquisition module, a modeling module, a trimming module, and an evaluation module;
the screening module is used for screening state characteristic quantities capable of reflecting the integral operation state of the charging facility according to the operation state characteristics of the charging facility of the electric automobile, distinguishing discrete state characteristic quantities and continuous state characteristic quantities and classifying the operation state types of the charging facility;
the acquisition module is used for acquiring historical operating data and test data of the electric automobile charging facility, evaluating discrete state characteristic quantity or continuous state characteristic quantity of each piece of data, and dividing all the acquired data into a training data set and a verification data set;
the modeling module is used for establishing a decision tree evaluation model for the evaluation of the running state of the electric vehicle charging facility on the basis of the calculation of the information gain rate;
the trimming module is used for trimming partial nodes in the decision tree evaluation model so as to prevent overfitting of the model and improve the generalization of the decision tree evaluation model;
the evaluation module is used for acquiring the operation data of the electric automobile charging facility and evaluating the operation state of the electric automobile charging facility based on the trimmed decision tree evaluation model.
The present applicant has described and illustrated embodiments of the present invention in detail with reference to the accompanying drawings, but it should be understood by those skilled in the art that the above embodiments are merely preferred embodiments of the present invention, and the detailed description is only for the purpose of helping the reader to better understand the spirit of the present invention, and not for limiting the scope of the present invention, and on the contrary, any improvement or modification made based on the spirit of the present invention should fall within the scope of the present invention.
Claims (8)
1. An electric automobile charging facility running state assessment method is characterized in that:
the method comprises the following steps:
step 1: according to the running state characteristics of the electric vehicle charging facility, screening state characteristic quantities capable of reflecting the whole running state of the charging facility, distinguishing discrete state characteristic quantities from continuous state characteristic quantities, and dividing the running state categories of the charging facility, wherein the running state categories of the charging facility comprise a normal working state, an abnormal early warning state and a serious warning state;
the charging equipment comprises a rectifier cabinet, an internal rectifier module, a charging pile body and a charging gun wire;
the state characteristic quantities reflecting the operation states of the rectifier cabinet and the rectifier module include but are not limited to: inputting alternating current three-phase average voltage, inputting alternating current three-phase average current, frequency, harmonic content grade, alternating current side breaker switching state, normal working module number, outputting direct current voltage, outputting direct current, cabinet temperature, contactor switching state and filter working state;
the state characteristic quantity reflecting the operation state of the charging pile body includes but is not limited to: the charging control system comprises an online state, a card reader communication state, an electric energy meter communication state, an auxiliary power supply voltage, a BMS communication state, a lightning arrester state, a TCU and charging controller communication state and charging pile internal temperature;
the state characteristic quantities reflecting the operating state of the charging gun line include, but are not limited to: the sheath wear state, the insulation condition of the gun head, the fastening condition of the gun head and the temperature of the gun head;
step 2: collecting historical operating data and test data of an electric vehicle charging facility, evaluating discrete state characteristic quantity or continuous state characteristic quantity of each piece of data, and dividing all the collected data into a training data set and a verification data set;
and step 3: establishing a decision tree evaluation model for evaluating the running state of the electric vehicle charging facility on the basis of calculation of the information gain rate;
and 4, step 4: trimming part of nodes in the decision tree evaluation model to prevent overfitting of the model and improve the generalization of the decision tree evaluation model;
and 5: and acquiring real-time operation data of the electric automobile charging facility, and evaluating the operation state of the electric automobile charging facility based on the trimmed decision tree evaluation model.
2. The method for evaluating the operating condition of an electric vehicle charging facility according to claim 1, wherein:
the method further comprises the following steps: in step 4, a new set of operating data is selected as a test data set for the trimmed decision tree evaluation model, and the accuracy, recall rate and F1 value of the trimmed decision tree evaluation model are verified.
3. The method for evaluating the operating state of an electric vehicle charging facility according to claim 2, wherein:
the accuracy rate refers to the proportion of the number of samples which can be correctly classified to the total number of samples, and is used for verifying the overall classification effect of the decision tree evaluation model;
the recall rate refers to the proportion of correctly classified samples in each classification label to all samples under the classification label, and is used for measuring the effectiveness of a classification mode of the decision tree evaluation model.
The F1 value is an accuracy recall harmonic mean value of the comprehensive measurement accuracy and recall ratio.
4. The method for evaluating the operating condition of an electric vehicle charging facility according to claim 1, wherein:
in step 1, the normal working state represents that the charging facility is in a stable operation range, and the state characteristic quantities are all maintained in a normal data range; the abnormal early warning state represents that although the charging facility can still continue to charge, part of components deviate from the normal operation state and cannot continuously operate for a long time, and operation and maintenance personnel are required to carry out remote or on-site inspection; the serious alarm state represents that the operation data of the charging facility is out of limit at the moment, the charging pile is immediately shut down, and the site power failure maintenance of operation and maintenance personnel is arranged.
5. The method for evaluating the operating condition of an electric vehicle charging facility according to claim 1, wherein:
and 3, establishing a decision tree evaluation model for evaluating the running state of the electric vehicle charging facility on the basis of the calculation of the information gain rate, specifically:
after the information entropy of the training data set is calculated, the information gain rates of all state characteristic quantities in the training data set are calculated, the state characteristic with the largest information gain rate value is selected as the optimal partition attribute of the root node, and a second layer of subset nodes are established according to the value of the state characteristic;
and continuously calculating the information gain rate on the second-layer subset nodes, and selecting the optimal division attribute of the current node, so as to continuously divide the lower-layer subset nodes until the classification of all internal nodes is completed, and forming a decision tree evaluation model for the evaluation of the running state of the electric vehicle charging facility.
6. The method for evaluating the operating condition of an electric vehicle charging facility according to claim 5, wherein:
in step 3, after the information entropy of the training data set is calculated, the information gain rate is directly calculated for the discrete state characteristic quantity; for the continuous state characteristic quantity, a plurality of intermediate points are firstly divided as candidate points by adopting a dichotomy, and then the information gain rate of each point is calculated in sequence.
7. The method for evaluating the operating condition of an electric vehicle charging facility according to claim 1, wherein:
and 4, trimming partial nodes in the decision tree evaluation model specifically as follows:
traversing the internal nodes of the decision tree evaluation model from bottom to top, judging the current verification precision of each node and replacing the verification precision of each node with a leaf node by adopting a verification data set, and pruning the node to transform the node into the leaf node when the verification data set is higher.
8. An electric automobile facility operating condition evaluation system that charges which characterized in that:
the system comprises a screening module, an acquisition module, a modeling module, a trimming module and an evaluation module;
the screening module is used for screening state characteristic quantities capable of reflecting the integral operation state of the charging facility according to the operation state characteristics of the charging facility of the electric automobile, distinguishing discrete state characteristic quantities and continuous state characteristic quantities and classifying the operation state types of the charging facility;
the acquisition module is used for acquiring historical operating data and test data of the electric automobile charging facility, evaluating discrete state characteristic quantity or continuous state characteristic quantity of each piece of data, and dividing all the acquired data into a training data set and a verification data set;
the modeling module is used for establishing a decision tree evaluation model for the evaluation of the running state of the electric vehicle charging facility on the basis of the calculation of the information gain rate;
the trimming module is used for trimming partial nodes in the decision tree evaluation model so as to prevent overfitting of the model and improve the generalization of the decision tree evaluation model;
the evaluation module is used for acquiring the operation data of the electric automobile charging facility and evaluating the operation state of the electric automobile charging facility based on the trimmed decision tree evaluation model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010335544.9A CN111582338A (en) | 2020-04-22 | 2020-04-22 | Method and system for evaluating running state of electric vehicle charging facility |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010335544.9A CN111582338A (en) | 2020-04-22 | 2020-04-22 | Method and system for evaluating running state of electric vehicle charging facility |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111582338A true CN111582338A (en) | 2020-08-25 |
Family
ID=72111652
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010335544.9A Withdrawn CN111582338A (en) | 2020-04-22 | 2020-04-22 | Method and system for evaluating running state of electric vehicle charging facility |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111582338A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112836174A (en) * | 2020-12-31 | 2021-05-25 | 深圳市加码能源科技有限公司 | AHP-based real-time charging safety evaluation method and storage medium |
CN114633655A (en) * | 2020-12-15 | 2022-06-17 | 北京骑胜科技有限公司 | Charging method of shared vehicle, battery management server and system |
CN114689980A (en) * | 2022-06-01 | 2022-07-01 | 深圳市明珞锋科技有限责任公司 | Abnormal accident alarm device for charger |
WO2022241705A1 (en) * | 2021-05-20 | 2022-11-24 | 四川金瑞麒智能科学技术有限公司 | Vehicle monitoring method, apparatus, and device, and computer-readable storage medium |
-
2020
- 2020-04-22 CN CN202010335544.9A patent/CN111582338A/en not_active Withdrawn
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114633655A (en) * | 2020-12-15 | 2022-06-17 | 北京骑胜科技有限公司 | Charging method of shared vehicle, battery management server and system |
CN114633655B (en) * | 2020-12-15 | 2024-03-29 | 北京骑胜科技有限公司 | Charging method, battery management server and system for shared vehicle |
CN112836174A (en) * | 2020-12-31 | 2021-05-25 | 深圳市加码能源科技有限公司 | AHP-based real-time charging safety evaluation method and storage medium |
WO2022241705A1 (en) * | 2021-05-20 | 2022-11-24 | 四川金瑞麒智能科学技术有限公司 | Vehicle monitoring method, apparatus, and device, and computer-readable storage medium |
CN114689980A (en) * | 2022-06-01 | 2022-07-01 | 深圳市明珞锋科技有限责任公司 | Abnormal accident alarm device for charger |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111582338A (en) | Method and system for evaluating running state of electric vehicle charging facility | |
CN105512474B (en) | A kind of method for detecting abnormality of Transformer's Condition Monitoring data | |
CN108320043B (en) | Power distribution network equipment state diagnosis and prediction method based on electric power big data | |
CN111639921B (en) | Intelligent equipment fault judgment and auxiliary disposal method based on expert system | |
CN108304634A (en) | A kind of 10kV pvs (pole-mounted vacuum switch) method for evaluating state based on multi-source data | |
CN113988273A (en) | Ice disaster environment active power distribution network situation early warning and evaluation method based on deep learning | |
CN108830406A (en) | A kind of main distribution based on data mining has a power failure configuration method in advance | |
CN116614177B (en) | Optical fiber state multidimensional parameter monitoring system | |
CN113420162A (en) | Equipment operation chain state monitoring method based on knowledge graph | |
CN111738573A (en) | Health evaluation method based on electric energy meter full life cycle data | |
CN103617447A (en) | Evaluation system and method for intelligent substation | |
CN110837532A (en) | Method for detecting electricity stealing behavior of charging pile based on big data platform | |
CN110968703B (en) | Method and system for constructing abnormal metering point knowledge base based on LSTM end-to-end extraction algorithm | |
CN116432123A (en) | Electric energy meter fault early warning method based on CART decision tree algorithm | |
CN112684295A (en) | Power distribution network fault line selection method and system under high permeability based on similarity separation degree | |
CN110443481B (en) | Power distribution automation terminal state evaluation system and method based on hybrid K-nearest neighbor algorithm | |
CN113313403B (en) | Power distribution network comprehensive evaluation method, device and system based on large-scale high-power electric vehicle charging and discharging and storage medium | |
CN113256272A (en) | Fine evaluation system for power equipment | |
CN107368979B (en) | Method and device for predicting disturbance event of weather-sensitive regional power distribution network | |
CN111391694B (en) | Multi-level rapid data monitoring method and system for operation and maintenance of charging station | |
CN117614137A (en) | Power distribution network optimization system based on multi-source data fusion | |
CN112418662A (en) | Power distribution network operation reliability analysis method using artificial neural network | |
CN116204839A (en) | Transformer oil temperature anomaly analysis method | |
CN116049098A (en) | Operational data characteristic analysis method suitable for judging operational state of power distribution network | |
CN114545276A (en) | Power battery service life prediction method based on capacity test and Internet of vehicles big data |
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 | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20200825 |