CN111025041A - Electric vehicle charging pile monitoring method and system, computer equipment and medium - Google Patents

Electric vehicle charging pile monitoring method and system, computer equipment and medium Download PDF

Info

Publication number
CN111025041A
CN111025041A CN201911081849.5A CN201911081849A CN111025041A CN 111025041 A CN111025041 A CN 111025041A CN 201911081849 A CN201911081849 A CN 201911081849A CN 111025041 A CN111025041 A CN 111025041A
Authority
CN
China
Prior art keywords
charging pile
time series
power factor
charging
value
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
Application number
CN201911081849.5A
Other languages
Chinese (zh)
Inventor
李兰哲
崔永
张效声
孙淑霞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Power Supply Bureau Co Ltd
Original Assignee
Shenzhen Power Supply Bureau Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenzhen Power Supply Bureau Co Ltd filed Critical Shenzhen Power Supply Bureau Co Ltd
Priority to CN201911081849.5A priority Critical patent/CN111025041A/en
Publication of CN111025041A publication Critical patent/CN111025041A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R21/00—Arrangements for measuring electric power or power factor
    • G01R21/006—Measuring power factor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning

Landscapes

  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Data Mining & Analysis (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • Power Engineering (AREA)
  • Charge And Discharge Circuits For Batteries Or The Like (AREA)
  • Electric Propulsion And Braking For Vehicles (AREA)

Abstract

本发明提供电动汽车充电桩监测方法及其系统、计算机设备、介质，包括如下步骤：获取第一时间序列；所述第一时间序列包括过去时间内按充电时间次序排序的多个平均功率因数值，其中每一平均功率因数值指每一次充电过程中充电桩的平均功率因数值；将所述第一时间序列输入预先训练好的极限学习机模型进行未来时间内充电桩功率因数预测，输出第二时间序列；所述第二时间序列包括未来时间内按充电时间次序排序的多个平均功率因数值；获取充电桩退化至异常时所对应的功率因数临界值；根据所述第二时间序列和所述功率因数临界值确定是否需对充电桩进行维修。本发明能够对电动汽车充电桩的状态进行监测和根据监测数据确定是否对充电桩进行维修。

Figure 201911081849

The present invention provides a method for monitoring electric vehicle charging piles, a system, computer equipment, and a medium, including the following steps: obtaining a first time series; the first time series includes a plurality of average power factor values sorted in the order of charging time in the past time , where each average power factor value refers to the average power factor value of the charging pile during each charging process; input the first time series into the pre-trained extreme learning machine model to predict the power factor of the charging pile in the future, and output the first time series. Two time series; the second time series includes a plurality of average power factor values sorted in the order of charging time in the future; obtains the power factor threshold value corresponding to when the charging pile degrades to an abnormality; according to the second time series and The power factor threshold determines whether maintenance of the charging pile is required. The invention can monitor the state of the electric vehicle charging pile and determine whether to repair the charging pile according to the monitoring data.

Figure 201911081849

Description

Electric vehicle charging pile monitoring method and system, computer equipment and medium
Technical Field
The invention relates to the field of charging piles, in particular to a method and a system for monitoring a charging pile of an electric automobile, computer equipment and a medium.
Background
With the development of electric automobile industry in China, the sale and maintenance quantity of electric automobiles is continuously increased, and the matched charging service is also greatly improved. The charging pile is the most widely used charging equipment for the electric automobile, and plays an irreplaceable important role in supplying power energy sources for the electric automobile and promoting popularization of the electric automobile. Under the background, it is very important to evaluate and predict the state of the charging device in time and to perform maintenance according to the state of the charging device in order to ensure safe and stable operation of the charging device.
At present, some charging equipment manufacturers are matched with monitoring systems for charging piles, and have the functions of acquiring and storing state parameters of the charging piles in real time, however, the parameters are often only used for real-time observation and display and then stored in a database, and further effective and full utilization is not achieved. On the other hand, although the country combines the actual development of electric vehicle charging equipment in China, a plurality of national standards related to the charging equipment are released in 2016, and the new standards also increase the regulations on the aspect of charging safety, the monitoring standards and the monitoring methods for the universal key parameters of the charging pile are not provided. At present, the health management mode of the electric vehicle charging equipment still remains in the stage of regular maintenance or after-fault maintenance. This kind of maintenance mode efficiency is not high, and the means is laggard, and carries out periodic maintenance to healthy electric pile and also can lead to the waste of maintenance resource, is unfavorable for the economic benefits of the electric pile is filled in the maximize performance.
Disclosure of Invention
The invention aims to provide an electric vehicle charging pile monitoring method and system, computer equipment and a medium thereof, so as to monitor the state of an electric vehicle charging pile and determine whether to maintain the charging pile according to monitoring data.
In a first aspect, an embodiment of the present invention provides a method for monitoring a charging pile of an electric vehicle, including the following steps:
acquiring a first time sequence; the first time sequence comprises a plurality of average power factor values which are sequenced in the past time according to the charging time sequence, wherein each average power factor value refers to the average power factor value of the charging pile in each charging process;
inputting the first time sequence into a pre-trained extreme learning machine model to predict the power factor of the charging pile in the future time, and outputting a second time sequence; the second time series includes a plurality of average power factor values in a future time ordered in a charging time order;
acquiring a power factor critical value corresponding to the condition that the charging pile is degraded to be abnormal;
and determining whether the charging pile needs to be maintained or not according to the second time sequence and the power factor critical value.
Inputting the first time sequence into a pre-trained extreme learning machine model for prediction and outputting a second time sequence comprises the following steps:
step S101, taking the first time sequence as an input time sequence
Figure BDA0002264223080000021
Inputting the extreme learning machine model for processing to obtain an output time sequence
Figure BDA0002264223080000022
Step S102, time sequence
Figure BDA0002264223080000023
And
Figure BDA0002264223080000024
combining to form a new vector as an input time series of the new model
Figure BDA0002264223080000025
Returning to the step S1 according to the preset iteration times to carry out an iteration loop;
step S103, obtaining an input time sequence according to multiple iterationsColumn(s) of
Figure BDA0002264223080000026
A second time series is obtained.
Wherein the extreme learning machine model training process is as follows:
step 201, generating training and testing data by using historical monitoring data of a charging pile, and sampling a time sequence of the monitoring data at intervals according to a certain interval e, wherein the sampling length is t + d, (t)>d) Splitting each sampled sequence into one sample input xin＝{x1,x2,...,xtAnd the corresponding output xout＝{xt+1,xt+2,...,xt+dThen each sample is x ═ xin,xoutDividing all samples into a training set and a testing set according to a preset proportion;
step 202, setting the number of input nodes, hidden nodes and output nodes of the extreme learning machine to establish an extreme learning machine model; wherein, the number of input nodes is the dimension of training sample input, namely t; the number of output nodes is the dimension d of the training sample output; the number of nodes of the hidden layer is initially n-10 d, and a weight matrix W and a bias vector b between the input layer and the hidden layer are initialized randomly by normal distribution;
step 203, training the established extreme learning machine model by using the training sample set x according to the batch processing mode of the extreme learning machine, and calculating the output weight value of the extreme learning machine model
Figure BDA0002264223080000031
Wherein, H ═ g (W.X + b) is the hidden layer output, I is the diagonal matrix, C is the preset value;
step 204, inputting the input time sequence of the test sample into the established extreme learning machine model, and calculating the output time sequence of the extreme learning machine model; comparing the predicted time sequence with the output time sequence of the test sample to obtain the root mean square error of the predicted time sequence and the output time sequence of the test sample; if the root mean square error meets the preset precision threshold, the training of the extreme learning machine model is completed, otherwise, the number n of the nodes of the hidden layer is increased to n +100, and the step 202 is returned.
Wherein, the obtaining of the power factor critical value corresponding to the degradation of the charging pile to the abnormality includes:
step 301, give initial normal time sequence { XnormalAnd then the time series { X } which needs to be judgednew}；
Step 302, calculating the Mean and variance Var of the normal time sequence;
step 303, comparing the time series { X ] one by using LayAda criterionnewEach value x innew_iDifference from Mean;
step 304, if the difference value satisfies | xnew_i-Mean|<3Var, then the new value xnew_iAdding to the original normal time series, i.e. { Xnormal}＝{Xnormal,xnew_iI is made to be i +1 and returns to step 302; if xnew_i-Mean|>3Var, then i ═ i +1, and proceed to step 305;
step 305, compare the value x using the Lavian criterionnew_iAnd the difference value of the Mean value, if the difference value satisfies | xnew_i-Mean|<3Var, then the value xnew_iAdding the original normal sequence and returning to the step 302; if xnew_i-Mean|>3Var, then the value xnew_iIs counted as a power factor critical value.
Wherein, the determining whether to maintain the charging pile according to the second time series and the power factor critical value comprises:
determining a state change time sequence corresponding to the power factor critical value in the second time sequence;
and judging whether the charging time of the charging pile reaches the state change time sequence, if so, determining that the charging pile needs to be maintained, and if not, determining that the charging pile does not need to be maintained.
Wherein the method further comprises:
acquiring charging monitoring data of a charging pile;
and generating a first time sequence according to the charging monitoring data.
In a second aspect, an embodiment of the present invention provides an electric vehicle charging pile monitoring system, which is used for implementing the electric vehicle charging pile monitoring method according to the embodiment, and includes:
a sequence acquisition unit configured to acquire a first time sequence; the first time sequence comprises a plurality of average power factor values which are sequenced in the past time according to the charging time sequence, wherein each average power factor value refers to the average power factor value of the charging pile in each charging process;
the sequence prediction unit is used for inputting the first time sequence into a pre-trained extreme learning machine model to predict the power factor of the charging pile in the future time and outputting a second time sequence; the second time series includes a plurality of average power factor values in a future time ordered in a charging time order;
the critical value obtaining unit is used for obtaining a corresponding power factor critical value when the charging pile degrades to be abnormal;
and the maintenance judging unit is used for determining whether the charging pile needs to be maintained or not according to the second time sequence and the power factor critical value.
Wherein the system further comprises:
the monitoring data acquisition unit is used for acquiring charging monitoring data of the charging pile;
and the sequence generating unit is used for generating a first time sequence according to the charging monitoring data.
In a third aspect, an embodiment of the present invention provides a computer device, including: the electric vehicle charging pile monitoring system is characterized by comprising an electric vehicle charging pile monitoring system; or, a memory and a processor, wherein the memory stores computer readable instructions, and the computer readable instructions, when executed by the processor, cause the processor to execute the steps of the electric vehicle charging pile monitoring method according to the embodiment.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the electric vehicle charging pile monitoring method according to the embodiment.
The embodiment of the invention provides a method and a system for monitoring an electric vehicle charging pile, computer equipment and a medium, which only need to provide historical data of the charging pile, learn from the historical data by using an extreme learning machine and acquire the state change trend of the charging pile. Because the monitoring parameters of the actual charging pile truly reflect various factors such as the electrical characteristics, the use characteristics, the environment and the like, a mathematical model or other information related to the equipment is not required to be established any more, and the monitoring data of the state parameters of the charging pile can be applied as long as enough monitoring data are available, so that the embodiment of the invention has good universality and can be widely applied to various charging pile equipment.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings. Of course, not all of the advantages described above need to be achieved at the same time in the practice of any one product or method of the invention.
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 schematic flow chart of a monitoring method for an electric vehicle charging pile according to an embodiment of the invention.
Fig. 2 is a schematic diagram of a charging pile power factor curve according to an embodiment of the invention.
Fig. 3 is a schematic diagram of a training process of a limit learning machine according to an embodiment of the present invention.
Fig. 4 is a schematic flowchart of step S300 according to an embodiment of the present invention.
Fig. 5 is a schematic flow chart of a monitoring system for an electric vehicle charging pile according to a second embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying 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.
It should be noted that, in order to avoid obscuring the present invention with unnecessary details, only the structures closely related to the solution according to the present invention are shown in the drawings, and other details not closely related to the present invention are omitted.
Example one
The embodiment of the invention provides a method for monitoring a charging pile of an electric vehicle, and fig. 1 is a schematic flow chart of the method in the embodiment of the invention, and with reference to fig. 1, the method comprises the following steps:
s100, acquiring a first time sequence; the first time sequence comprises a plurality of average power factor values which are sequenced in the past time according to the charging time sequence, wherein each average power factor value refers to the average power factor value of the charging pile in each charging process;
specifically, as shown in fig. 2, the first time series is known data, the average power factor value of each charging is obtained by monitoring the charging state of the charging pile, and then the charging sequence is arranged to obtain the first time series, where the abscissa of the point on the curve in fig. 2 represents the number of charging times and the ordinate represents the average power factor value.
S200, inputting the first time sequence into a pre-trained extreme learning machine model to predict a power factor value of a charging pile in the future time, and outputting a second time sequence; the second time series includes a plurality of average power factor values in a future time ordered in a charging time order;
specifically, as shown in fig. 2, the second time series is prediction data indicating how many corresponding charging pile power factor values are for each subsequent charging on the basis of the first time series.
And step S300, acquiring a power factor critical value corresponding to the condition that the charging pile is degraded to be abnormal.
Specifically, the power factor threshold refers to that when the power factor of the charging pile is reduced to the power factor threshold, the state of the charging pile is degraded to be abnormal, and at this time, the charging pile needs to be maintained.
And S400, determining whether the charging pile needs to be maintained or not according to the second time sequence and the power factor critical value.
Specifically, when the power factor of the charging pile is decreased to the power factor critical value, as shown in fig. 2, the corresponding charging times reach 556 times, and therefore, by monitoring the charging times of the charging pile, when the charging times reach 556 times or approach 556 times, the charging pile is maintained.
Wherein the method further comprises:
acquiring charging monitoring data of a charging pile;
and generating a first time sequence according to the charging monitoring data.
Specifically, the charging pile is used for acquiring t points from the latest point to the front as a first time sequence of a model according to the current parameter monitoring data
Figure BDA0002264223080000061
Each point represents a charge, the value of the point represents the average power factor value during the charge, and the last charge is the most recent point.
Wherein, the step S200 includes:
step S101, the first time sequence is carried out
Figure BDA0002264223080000071
Inputting the extreme learning machine model as an input time sequence to be processed to obtain an output time sequence
Figure BDA0002264223080000072
Step S102, time sequence
Figure BDA0002264223080000073
And
Figure BDA0002264223080000074
combining to form a new vector, and taking t points from back to front as an input time sequence of a new model
Figure BDA0002264223080000075
Input time series as a new model
Figure BDA0002264223080000076
Returning to the step S1 for iteration circulation according to the preset iteration times until the iteration times are finished;
step S103, obtaining an input time sequence according to multiple iterations
Figure BDA0002264223080000077
A second time series is obtained.
Referring to fig. 3, the extreme learning machine model training process is as follows:
step 201, generating training and testing data by using historical monitoring data of a charging pile, and sampling a time sequence of the monitoring data at intervals according to a certain interval e, wherein the sampling length is t + d, (t)>d) Splitting each sampled sequence into one sample input xin＝{x1,x2,...,xtAnd the corresponding output xout＝{xt+1,xt+2,...,xt+dThen each sample is x ═ xin,xoutDividing all samples into a training set and a testing set according to a preset proportion;
for example, since the power factor of the charging pile has a great influence on the charging efficiency and the charging time, the health state of the charging pile can be reflected, and therefore, historical monitoring data of the parameters are selected as training and test sample data. The input of the sample is 40 dimensions, namely the average power factor of 40 charging times continuously, and the output is the average power factor of 10 charging times next for the 40 charging times. 500 points of historical monitoring data of the power factor of the charging pile are selected to be used as training and testing samples. With the first 400 data as training sets and the last 100 as test sets, 199 training samples and 49 test samples can be generated.
Step 202, setting the number of input nodes, hidden nodes and output nodes of the extreme learning machine to establish an extreme learning machine model; wherein, the number of input nodes is the dimension of training sample input, namely t; the number of output nodes is the dimension d of the training sample output; the number of nodes of the hidden layer is initially n-10 d, and a weight matrix W and a bias vector b between the input layer and the hidden layer are initialized randomly by normal distribution;
for example, 40 input nodes and 10 output nodes of the extreme learning machine are set. And (5) setting the number n of the initial hidden layer nodes to be 400, and then randomly initializing a weight matrix W and a bias vector b between the input layer and the hidden layer by utilizing normal distribution. And giving a preset precision threshold errth＝0.0001。
Step 203, training the established extreme learning machine model by using the training sample set X according to the batch processing mode of the extreme learning machine, and calculating the output weight value of the extreme learning machine model
Figure BDA0002264223080000081
Where H ═ g (W.X + b) is the hidden layer output, I is the diagonal matrix, and C is a constant.
Step 204, inputting the input time sequence of the test sample into the established extreme learning machine model, and calculating the output time sequence of the extreme learning machine model; comparing the predicted time sequence with the output time sequence of the test sample to obtain the root mean square error of the predicted time sequence and the output time sequence of the test sample; if the root mean square error meets the preset precision threshold, the training of the extreme learning machine model is completed, otherwise, the number n of the nodes of the hidden layer is increased to n +100, and the step 202 is returned.
For example, the predicted time series is compared with the output of the test sample, and the root mean square error of the two is calculated. In the case of n being 400, the calculated accuracy err being 0.0000554<errthThe requirement is met, so that the number of nodes of the hidden layer does not need to be increased, and the training is finished.
In one embodiment, referring to fig. 4, the step S300 includes:
step 301, give initial normal time sequence { XnormalAnd then the time series { X } which needs to be judgednew}；
Step 302, calculating the Mean and variance Var of the normal time sequence;
step 303, comparing the time series { X ] one by using LayAda criterionnewEach value x innew_iDifference from Mean;
step 304, if the difference value satisfies | xnew_i-Mean|<3Var, then the new value xnew_iAdding to the original normal time series, i.e. { Xnormal}＝{Xnormal,xnew_iI is made to be i +1 and returns to step 302; if xnew_i-Mean|>3Var, then i ═ i +1, and proceed to step 305;
step 305, compare the value x using the Lavian criterionnew_iThe difference value of the Mean value is smaller, and | x is satisfiednew_i-Mean|<3Var, then the value xnew_iAdding the original normal sequence and returning to the step 302; if | x is satisfiednew_i-Mean|>3Var, then the value xnew_iPower factor threshold value: x is the number ofcr＝xnew_i。
Specifically, with the continuous operation of the charging pile, the situations of equipment loss, aging and the like occur, the equipment state is judged in several stages from normal operation to abnormal operation to failure and the like according to the state parameters, however, the range of the parameter values is normal, the range of the parameter values is abnormal, and the parameter values are obtained by a large amount of experiments or by a specialist with strong experience, and sometimes it is difficult to provide such a critical value. The embodiment provides the Lauda criterion of mathematical statistics for judgment, is convenient to implement, and solves the problem that the analysis parameters need experience or experiments.
Wherein the step S400 includes:
step 401, determining a state change time sequence corresponding to the power factor and the power factor critical value in the second time sequence;
and step 402, judging whether the charging time of the charging pile reaches the state change time sequence, if so, determining that the charging pile needs to be maintained, and if not, determining that the charging pile does not need to be maintained.
Specifically, referring to FIG. 2, the data of 200 points in front can be taken as { XnormalFrom point 201 to 600 as { X }newAnd point 556 is a critical point, when the charging reaches the 556 th time, the state of the charging pile is degraded to be abnormal, and the charging pile is recommended to be subjected to maintenance and inspection.
According to the method, special mathematical modeling is not needed for the charging piles of various models, or the state turning points of the state parameters of the charging piles are determined through tests, only historical data of the monitoring parameters of the charging piles are needed to be provided, and a data training extreme learning machine model is utilized, so that health information contained in the data is mined, the future state of the charging piles is predicted, corresponding maintenance suggestions are given, and the automation level of health management of the charging piles is effectively improved.
Example two
As shown in fig. 5, a second embodiment of the present invention provides an electric vehicle charging pile monitoring system, which is used for implementing the electric vehicle charging pile monitoring method according to the first embodiment, and includes:
a sequence acquisition unit 1 for acquiring a first time sequence; the first time sequence comprises a plurality of average power factor values which are sequenced in the past time according to the charging time sequence, wherein each average power factor value refers to the average power factor value of the charging pile in each charging process;
the sequence prediction unit 2 is used for inputting the first time sequence into a pre-trained extreme learning machine model to predict the power factor of the charging pile in the future time and outputting a second time sequence; the second time series includes a plurality of average power factor values in a future time ordered in a charging time order;
the critical value obtaining unit 3 is used for obtaining a power factor critical value corresponding to the charging pile when the charging pile degrades to be abnormal;
and the maintenance judging unit 4 is used for determining whether the charging pile needs to be maintained or not according to the second time sequence and the power factor critical value.
In one embodiment, the method further comprises:
the monitoring data acquisition unit 5 is used for acquiring charging monitoring data of the charging pile;
and the sequence generating unit 6 is used for generating a first time sequence according to the charging monitoring data.
It should be noted that the system according to the second embodiment corresponds to the method according to the first embodiment, and therefore, a part of the system according to the second embodiment that is not described in detail can be obtained by referring to the content of the method according to the first embodiment, and is not described again here.
It is to be noted that, based on the content, those skilled in the art can clearly understand that the embodiments of the present invention can be implemented by software and necessary general hardware, and certainly can be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the embodiments of the present invention may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) to implement the methods/systems described in the foregoing embodiments.
EXAMPLE III
An embodiment of the present invention provides a computer device, including: the electric vehicle charging pile monitoring system is characterized by comprising an electric vehicle charging pile monitoring system; or, a memory and a processor, wherein the memory stores computer readable instructions, and the computer readable instructions, when executed by the processor, cause the processor to execute the steps of the electric vehicle charging pile monitoring method according to the embodiment.
Of course, the computer device may also have components such as a wired or wireless network interface, a keyboard, and an input/output interface, so as to perform input/output, and the computer device may also include other components for implementing the functions of the device, which are not described herein again.
Example four
The fourth embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the steps of the method for monitoring a charging pile of an electric vehicle according to the first embodiment.
The foregoing is directed to embodiments of the present invention, and it is understood that various modifications and improvements can be made by those skilled in the art without departing from the spirit of the invention.

Claims (10)

1.一种电动汽车充电桩监测方法，其特征在于，包括如下步骤：1. an electric vehicle charging pile monitoring method, is characterized in that, comprises the steps: 获取第一时间序列；所述第一时间序列包括过去时间内按充电时间次序排序的多个平均功率因数值，其中每一平均功率因数值指每一次充电过程中充电桩的平均功率因数值；Obtaining a first time series; the first time series includes a plurality of average power factor values sorted in the order of charging time in the past, wherein each average power factor value refers to the average power factor value of the charging pile in each charging process; 将所述第一时间序列输入预先训练好的极限学习机模型进行未来时间内充电桩功率因数预测，输出第二时间序列；所述第二时间序列包括未来时间内按充电时间次序排序的多个平均功率因数值；Input the first time series into the pre-trained extreme learning machine model to predict the power factor of charging piles in the future, and output a second time series; the second time series includes a plurality of Average power factor value; 获取充电桩退化至异常时所对应的功率因数临界值；Obtain the critical value of the power factor when the charging pile degrades to an abnormal state; 根据所述第二时间序列和所述功率因数临界值确定是否需对充电桩进行维修。Whether the charging pile needs to be repaired is determined according to the second time series and the power factor threshold. 2.如权利要求1所述的电动汽车充电桩监测方法，其特征在于，将所述第一时间序列输入预先训练好的极限学习机模型进行预测输出第二时间序列包括：2. The electric vehicle charging pile monitoring method according to claim 1, wherein inputting the first time series into a pre-trained extreme learning machine model to predict and output the second time series comprises: 步骤S101、将第一时间序列作为输入时间序列
Figure FDA0002264223070000011
输入所述极限学习机模型进行处理得到输出时间序列
Figure FDA0002264223070000012
Step S101, take the first time series as the input time series
Figure FDA0002264223070000011
Input the extreme learning machine model for processing to obtain the output time series
Figure FDA0002264223070000012
 步骤S102、将时间序列
Figure FDA0002264223070000013
与
Figure FDA0002264223070000014
合并组成新的向量，并作为新的模型的输入时间序列
Figure FDA0002264223070000015
按照预定迭代次数返回步骤S1进行迭代循环；Step S102, the time series
Figure FDA0002264223070000013
and
Figure FDA0002264223070000014
Combined to form a new vector and used as the input time series of the new model
Figure FDA0002264223070000015
Return to step S1 to perform an iterative loop according to the predetermined number of iterations; 步骤S103、根据多次迭代得到的输入时间序列
Figure FDA0002264223070000016
得到第二时间序列。Step S103, according to the input time series obtained by multiple iterations
Figure FDA0002264223070000016
Get the second time series. 3.如权利要求1所述的电动汽车充电桩监测方法，其特征在于，所述极限学习机模型训练过程如下：3. The electric vehicle charging pile monitoring method according to claim 1, wherein the extreme learning machine model training process is as follows: 步骤201、利用充电桩历史监控数据生成训练和测试数据，按照一定间隔e将监控数据的时间序列进行间隔采样，采样长度为t+d,(t>d)，将每个采样的序列分割为一个样本的输入xin＝{x1,x2,...,xt}和相应的输出xout＝{xt+1,xt+2,...,xt+d}，则每个样本为x＝{xin,xout}，按照预设的比例，将所有样本分为训练集和测试集；Step 201: Use the historical monitoring data of the charging pile to generate training and test data, perform interval sampling of the time series of the monitoring data according to a certain interval e, the sampling length is t+d, (t>d), and divide each sampling sequence into A sample input x in = {x 1 , x 2 ,...,x t } and corresponding output x out ={x t+1 ,x t+2 ,...,x t+d }, then Each sample is x={x in ,x out }, according to the preset ratio, all samples are divided into training set and test set; 步骤202、设置极限学习机的输入节点、隐含节点和输出节点个数以建立极限学习机模型；其中，输入节点数为训练样本输入的维度，即t；输出节点数为训练样本输出的维度d；隐含层节点数初始取n＝10d，利用正态分布随机初始化输入层和隐含层之间的权值矩阵W和偏置向量b；Step 202, setting the number of input nodes, hidden nodes and output nodes of the extreme learning machine to establish the extreme learning machine model; wherein, the number of input nodes is the dimension of the input of the training sample, namely t; the number of output nodes is the dimension of the output of the training sample d; the number of hidden layer nodes is initially n=10d, and the weight matrix W and the bias vector b between the input layer and the hidden layer are randomly initialized using normal distribution; 步骤203、利用训练样本集合x，按照极限学习机批处理方式对建立的极限学习机模型进行训练，计算其输出权值
Figure FDA0002264223070000021
其中，H＝g(W.X+b)为隐含层输出，I为对角矩阵，C为预设值；Step 203: Use the training sample set x to train the established extreme learning machine model according to the batch processing mode of the extreme learning machine, and calculate its output weights
Figure FDA0002264223070000021
Among them, H=g(W.X+b) is the hidden layer output, I is the diagonal matrix, and C is the preset value; 步骤204、将测试样本的输入时间序列输入建立的极限学习机模型，计算其输出时间序列；并将预测的时间序列和测试样本的输出时间序列进行对比得到两者的均方根误差；若均方根误差满足预设精度阈值，则极限学习机模型训练完成，否则增加隐含层节点数目n＝n+100，并返回步骤202。Step 204: Input the input time series of the test sample into the established extreme learning machine model, and calculate its output time series; compare the predicted time series and the output time series of the test sample to obtain the root mean square error of both; If the square root error satisfies the preset accuracy threshold, the extreme learning machine model training is completed; otherwise, the number of hidden layer nodes is increased by n=n+100, and the process returns to step 202 . 4.如权利要求1所述的电动汽车充电桩监测方法，其特征在于，所述获取充电桩退化至异常时所对应的功率因数临界值包括：4 . The method for monitoring electric vehicle charging piles according to claim 1 , wherein the obtaining the power factor threshold value corresponding to when the charging pile degrades to an abnormality comprises: 5 . 步骤301、给定初始正常时间序列{Xnormal}和其后需要判断的时间序列{Xnew}；Step 301, given an initial normal time series {X normal } and a subsequent time series {X new } to be judged; 步骤302、计算正常时间序列的均值Mean和方差Var；Step 302, calculate the mean value Mean and variance Var of the normal time series; 步骤303、利用拉依达准则逐一比较时间序列{Xnew}中的每一个值xnew_i和均值Mean的差值；Step 303, using the Laida criterion to compare the difference between each value x new_i and the mean value Mean in the time series {X new } one by one; 步骤304、若差值满足|xnew_i-Mean|<3Var，则将新的值xnew_i加入原来的正常时间序列，即{Xnormal}＝{Xnormal,xnew_i}，令i＝i+1并返回步骤302；若|xnew_i-Mean|>3Var，则i＝i+1，并进入步骤305；Step 304: If the difference satisfies |x new_i -Mean|<3Var, add the new value x new_i to the original normal time series, that is, {X normal }={X normal ,x new_i }, let i=i+1 And return to step 302; if |x new_i -Mean|>3Var, then i=i+1, and go to step 305; 步骤305、利用拉依达准则比较值xnew_i和均值Mean的差值，若差值满足|xnew_i-Mean|<3Var，则将值xnew_i加入原来的正常序列，并返回步骤302；若|xnew_i-Mean|>3Var，则将值xnew_i计为功率因数临界值。Step 305: Compare the difference between the value x new_i and the mean value Mean using the Laida criterion, if the difference value satisfies |x new_i -Mean|<3Var, add the value x new_i to the original normal sequence, and return to step 302; if | x new_i -Mean|>3Var, the value x new_i is counted as the power factor threshold. 5.如权利要求1所述的电动汽车充电桩监测方法，其特征在于，所述根据所述第二时间序列和所述功率因数临界值确定是否需对充电桩进行维修包括：5 . The method for monitoring electric vehicle charging piles according to claim 1 , wherein determining whether the charging pile needs to be repaired according to the second time series and the power factor threshold value comprises: 6 . 确定所述第二时间序列中功率因数与所述功率因数临界值对应状态变化时间次序；determining the time sequence of state changes corresponding to the power factor and the power factor critical value in the second time series; 判断充电桩充电时间是否到达所述状态变化时间次序，若是，则确定需对充电桩进行维修，若否，则确定不需对充电桩进行维修。It is judged whether the charging time of the charging pile has reached the state change time sequence, and if so, it is determined that the charging pile needs to be repaired, and if not, it is determined that the charging pile does not need to be repaired. 6.如权利要求1所述的电动汽车充电桩监测方法，其特征在于，还包括：6. The electric vehicle charging pile monitoring method according to claim 1, further comprising: 获取充电桩的充电监控数据；Obtain the charging monitoring data of the charging pile; 根据所述充电监控数据生成第一时间序列。A first time series is generated based on the charge monitoring data. 7.一种电动汽车充电桩监测系统，用于实现权利要求1-5任一项所述的电动汽车充电桩监测方法，其特征在于，包括：7. An electric vehicle charging pile monitoring system for realizing the electric vehicle charging pile monitoring method according to any one of claims 1-5, characterized in that, comprising: 序列获取单元，用于获取第一时间序列；所述第一时间序列包括过去时间内按充电时间次序排序的多个平均功率因数值，其中每一平均功率因数值指每一次充电过程中充电桩的平均功率因数值；A sequence acquisition unit, configured to acquire a first time sequence; the first time sequence includes a plurality of average power factor values sorted in the order of charging time in the past, wherein each average power factor value refers to the charging pile during each charging process The average power factor value of ; 序列预测单元，用于将所述第一时间序列输入预先训练好的极限学习机模型进行未来时间内充电桩功率因数预测，输出第二时间序列；所述第二时间序列包括未来时间内按充电时间次序排序的多个平均功率因数值；The sequence prediction unit is used to input the first time sequence into the pre-trained extreme learning machine model to predict the power factor of the charging pile in the future time, and output a second time sequence; the second time sequence includes charging a plurality of average power factor values in chronological order; 临界值获取单元，用于获取充电桩退化至异常时所对应的功率因数临界值；The critical value acquiring unit is used to acquire the critical value of the power factor corresponding to when the charging pile degrades to an abnormal state; 维修判断单元，用于根据所述第二时间序列和所述功率因数临界值确定是否需对充电桩进行维修。A maintenance judging unit, configured to determine whether the charging pile needs to be maintained according to the second time series and the power factor threshold value. 8.如权利要求7所述的电动汽车充电桩监测系统，其特征在于还包括：8. The electric vehicle charging pile monitoring system according to claim 7, further comprising: 监控数据获取单元，用于获取充电桩的充电监控数据；The monitoring data acquisition unit is used to acquire the charging monitoring data of the charging pile; 序列生成单元，用于根据所述充电监控数据生成第一时间序列。A sequence generating unit, configured to generate a first time sequence according to the charging monitoring data. 9.一种计算机设备，包括：根据权利要求7-8中任一项所述的电动汽车充电桩监测系统；或者，存储器和处理器，所述存储器中存储有计算机可读指令，所述计算机可读指令被所述处理器执行时，使得所述处理器执行根据权利要求1-6中任一项所述电动汽车充电桩监测方法的步骤。9. A computer device, comprising: the electric vehicle charging pile monitoring system according to any one of claims 7-8; or, a memory and a processor, wherein computer-readable instructions are stored in the memory, and the computer The readable instructions, when executed by the processor, cause the processor to perform the steps of the method for monitoring an electric vehicle charging pile according to any one of claims 1-6. 10.一种计算机可读存储介质，其上存储有计算机程序，其特征在于：所述计算机程序被处理器执行时实现权利要求1-6中任一项所述电动汽车充电桩监测方法的步骤。10. A computer-readable storage medium on which a computer program is stored, characterized in that: when the computer program is executed by a processor, the steps of implementing the method for monitoring electric vehicle charging piles according to any one of claims 1-6 .
CN201911081849.5A 2019-11-07 2019-11-07 Electric vehicle charging pile monitoring method and system, computer equipment and medium Pending CN111025041A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911081849.5A CN111025041A (en) 2019-11-07 2019-11-07 Electric vehicle charging pile monitoring method and system, computer equipment and medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911081849.5A CN111025041A (en) 2019-11-07 2019-11-07 Electric vehicle charging pile monitoring method and system, computer equipment and medium

Publications (1)

Publication Number Publication Date
CN111025041A true CN111025041A (en) 2020-04-17

Family

ID=70201146

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911081849.5A Pending CN111025041A (en) 2019-11-07 2019-11-07 Electric vehicle charging pile monitoring method and system, computer equipment and medium

Country Status (1)

Country Link
CN (1) CN111025041A (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112100920A (en) * 2020-09-15 2020-12-18 东南大学 Power distribution network three-phase voltage calculation method, device, equipment and storage medium
CN112433122A (en) * 2020-11-23 2021-03-02 广州橙行智动汽车科技有限公司 Charging pile available state detection method, device, equipment and storage medium
CN115542235A (en) * 2022-11-07 2022-12-30 北京志翔科技股份有限公司 Method, device, equipment and storage medium for determining metering error of charging gun
CN116992378A (en) * 2023-05-16 2023-11-03 江苏智臻能源科技有限公司 Charging abnormality analysis method for residential electric transportation equipment based on loop impedance
CN117074840A (en) * 2023-10-16 2023-11-17 南京思宇电气技术有限公司 DC charging pile on-line testing system

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102944769A (en) * 2012-10-18 2013-02-27 上海市电力公司 Fault diagnosis method of power transformer based on extreme learning machine
CN105046374A (en) * 2015-08-25 2015-11-11 华北电力大学 Power interval predication method based on nucleus limit learning machine model
CN105354646A (en) * 2015-12-04 2016-02-24 福州大学 Power load forecasting method for hybrid particle swarm optimization and extreme learning machine
DE102016106776A1 (en) * 2015-04-16 2016-10-20 Ford Global Technologies, Llc Vehicle high voltage system insulation test
CN106446440A (en) * 2016-10-11 2017-02-22 天津大学 Short-term photovoltaic generation power prediction method based on online sequential extreme learning machine
CN107589323A (en) * 2017-08-30 2018-01-16 安徽千里眼信息科技有限公司 A kind of charging pile electric energy quality monitoring and charging managing and control system
CN108287317A (en) * 2018-01-24 2018-07-17 上海电气分布式能源科技有限公司 Power of battery prediction model generation method and system, power forecasting method and system
CN108761346A (en) * 2018-06-20 2018-11-06 首都师范大学 A kind of vehicle lithium battery method for predicting residual useful life
WO2019035770A1 (en) * 2017-08-18 2019-02-21 Nanyang Technological University Adaptive computer system, and methods and apparatus for training the adaptive computer system
CN110133395A (en) * 2019-04-04 2019-08-16 国网电动汽车服务有限公司 Method and device for diagnosing charging pile health status
CN110333462A (en) * 2019-08-08 2019-10-15 首都师范大学 A Li-ion battery life prediction method based on DGWO-ELM under random discharge environment

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102944769A (en) * 2012-10-18 2013-02-27 上海市电力公司 Fault diagnosis method of power transformer based on extreme learning machine
DE102016106776A1 (en) * 2015-04-16 2016-10-20 Ford Global Technologies, Llc Vehicle high voltage system insulation test
CN105046374A (en) * 2015-08-25 2015-11-11 华北电力大学 Power interval predication method based on nucleus limit learning machine model
CN105354646A (en) * 2015-12-04 2016-02-24 福州大学 Power load forecasting method for hybrid particle swarm optimization and extreme learning machine
CN106446440A (en) * 2016-10-11 2017-02-22 天津大学 Short-term photovoltaic generation power prediction method based on online sequential extreme learning machine
WO2019035770A1 (en) * 2017-08-18 2019-02-21 Nanyang Technological University Adaptive computer system, and methods and apparatus for training the adaptive computer system
CN107589323A (en) * 2017-08-30 2018-01-16 安徽千里眼信息科技有限公司 A kind of charging pile electric energy quality monitoring and charging managing and control system
CN108287317A (en) * 2018-01-24 2018-07-17 上海电气分布式能源科技有限公司 Power of battery prediction model generation method and system, power forecasting method and system
CN108761346A (en) * 2018-06-20 2018-11-06 首都师范大学 A kind of vehicle lithium battery method for predicting residual useful life
CN110133395A (en) * 2019-04-04 2019-08-16 国网电动汽车服务有限公司 Method and device for diagnosing charging pile health status
CN110333462A (en) * 2019-08-08 2019-10-15 首都师范大学 A Li-ion battery life prediction method based on DGWO-ELM under random discharge environment

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
徐永海等: "基于核极限学习机的风光容量配置研究", 《电测与仪表》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112100920A (en) * 2020-09-15 2020-12-18 东南大学 Power distribution network three-phase voltage calculation method, device, equipment and storage medium
CN112433122A (en) * 2020-11-23 2021-03-02 广州橙行智动汽车科技有限公司 Charging pile available state detection method, device, equipment and storage medium
CN115542235A (en) * 2022-11-07 2022-12-30 北京志翔科技股份有限公司 Method, device, equipment and storage medium for determining metering error of charging gun
CN116992378A (en) * 2023-05-16 2023-11-03 江苏智臻能源科技有限公司 Charging abnormality analysis method for residential electric transportation equipment based on loop impedance
CN117074840A (en) * 2023-10-16 2023-11-17 南京思宇电气技术有限公司 DC charging pile on-line testing system
CN117074840B (en) * 2023-10-16 2023-12-19 南京思宇电气技术有限公司 DC charging pile on-line testing system

Similar Documents

Publication Publication Date Title
CN111025041A (en) Electric vehicle charging pile monitoring method and system, computer equipment and medium
CN113076648B (en) Fatigue acceleration test and residual life prediction method for speed reducer and storage processing system
CN113933718B (en) Retired battery capacity sorting method, retired battery capacity sorting device, retired battery capacity sorting equipment and storage medium
CN111537884B (en) Method, device, computer equipment and medium for obtaining power battery life data
CN113158589B (en) Simulation model calibration method and device of battery management system
CN116087787A (en) Battery fault judging method and system based on principal component analysis method
CN112800683B (en) System short-circuit current level evaluation method and system based on convolutional neural network
CN115392592A (en) Storage product parameter configuration recommendation method, device, equipment and medium
CN115856695A (en) Battery health state prediction method and device and electronic equipment
CN117420468B (en) Battery status evaluation method, device, equipment and storage medium
CN113300373A (en) Stability margin value prediction method and device based on PRMSE evaluation index
CN112163633A (en) Test evaluation method and device, electronic equipment and storage medium
Celen et al. Calendar ageing modelling using machine learning: an experimental investigation on lithium ion battery chemistries
CN114997271A (en) Intelligent diagnosis method and system for hydroelectric generating set faults
CN116484675A (en) Crack propagation life prediction method and system for ship engine blade
CN116384680A (en) Visual monitoring method, device and equipment for research and development quality and readable storage medium
CN115438477A (en) SOFC system degradation prediction method and system, electronic device and storage medium
CN114091243A (en) Product concept design reliability prediction method and system
CN111967774A (en) Software quality risk prediction method and device
CN118275902B (en) Battery remaining life prediction method and electronic equipment
CN118378460B (en) A planning method and device for reliability test of reusable rocket engine
CN109033667A (en) A kind of bounded-but-unknown uncertainty structural energy response predicting method based on affine perturbation
CN113128069B (en) Method, device, storage medium and electronic device for determining reliability of power battery system
BAE COMPARATIVE EVALUATION ON THE ATTRIBUTES OF SOFTWARE DEVELOPMENT COST MODEL WITH EXPONENTIAL AND INVERSE-EXPONENTIAL DISTRIBUTION PROPERTY
CN112434839B (en) Distribution transformer heavy overload risk prediction method and electronic equipment

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

Application publication date: 20200417

RJ01 Rejection of invention patent application after publication