CN106168799B - A method of batteries of electric automobile predictive maintenance is carried out based on big data machine learning - Google Patents
A method of batteries of electric automobile predictive maintenance is carried out based on big data machine learning Download PDFInfo
- Publication number
- CN106168799B CN106168799B CN201610504843.4A CN201610504843A CN106168799B CN 106168799 B CN106168799 B CN 106168799B CN 201610504843 A CN201610504843 A CN 201610504843A CN 106168799 B CN106168799 B CN 106168799B
- Authority
- CN
- China
- Prior art keywords
- data
- battery
- batteries
- electric automobile
- model
- 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.)
- Active
Links
- 238000012423 maintenance Methods 0.000 title claims abstract description 46
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000010801 machine learning Methods 0.000 title abstract description 9
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 54
- 238000012549 training Methods 0.000 claims abstract description 27
- 238000003066 decision tree Methods 0.000 claims description 21
- 230000003044 adaptive effect Effects 0.000 claims description 15
- 238000011156 evaluation Methods 0.000 claims description 14
- 238000006116 polymerization reaction Methods 0.000 claims description 13
- 238000012360 testing method Methods 0.000 claims description 13
- 238000005096 rolling process Methods 0.000 claims description 12
- 238000002360 preparation method Methods 0.000 claims description 11
- 230000015556 catabolic process Effects 0.000 claims description 9
- 238000004140 cleaning Methods 0.000 claims description 9
- 238000013459 approach Methods 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 8
- 238000007477 logistic regression Methods 0.000 claims description 8
- 238000012512 characterization method Methods 0.000 claims description 7
- 238000002790 cross-validation Methods 0.000 claims description 6
- 230000000379 polymerizing effect Effects 0.000 claims description 6
- 238000012795 verification Methods 0.000 claims description 6
- 230000015572 biosynthetic process Effects 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 5
- 238000003786 synthesis reaction Methods 0.000 claims description 5
- 238000013145 classification model Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 3
- 230000005611 electricity Effects 0.000 claims description 3
- 238000002474 experimental method Methods 0.000 claims description 3
- 230000007257 malfunction Effects 0.000 claims description 3
- HUTDUHSNJYTCAR-UHFFFAOYSA-N ancymidol Chemical compound C1=CC(OC)=CC=C1C(O)(C=1C=NC=NC=1)C1CC1 HUTDUHSNJYTCAR-UHFFFAOYSA-N 0.000 claims description 2
- 230000001537 neural effect Effects 0.000 claims description 2
- 238000012937 correction Methods 0.000 claims 1
- 230000008569 process Effects 0.000 abstract description 5
- 238000011217 control strategy Methods 0.000 abstract 1
- 230000004044 response Effects 0.000 abstract 1
- 230000006870 function Effects 0.000 description 9
- 238000012545 processing Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000000047 product Substances 0.000 description 3
- 208000032953 Device battery issue Diseases 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 210000004556 brain Anatomy 0.000 description 2
- 210000004027 cell Anatomy 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 210000004218 nerve net Anatomy 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 238000005312 nonlinear dynamic Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000032258 transport Effects 0.000 description 1
- 238000005406 washing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
- G05B23/0283—Predictive maintenance, e.g. involving the monitoring of a system and, based on the monitoring results, taking decisions on the maintenance schedule of the monitored system; Estimating remaining useful life [RUL]
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Secondary Cells (AREA)
Abstract
This patent is related to a kind of method for carrying out batteries of electric automobile life prediction maintenance based on big data machine learning, this method by corresponding application architecture, process, computation model form.This method is based on the battery real time data to the acquisition in batteries of electric automobile operational process, in conjunction with electric car vehicle others operation data, pass through the model training and proof of algorithm of machine learning, and the assessment of different angle is carried out to result, to establish the control strategy to batteries of electric automobile operation predictive maintenance and response, the maintenance and replacement for optimizing battery, improve the safety indexes of car owner, reach the balance of system performance and economic benefit.
Description
Technical field
The present invention relates to a kind of application analysis method that battery predictive maintenance is carried out based on big data machine learning, applications
Field be batteries of electric automobile Maintenance forecast replacement and maintenance field.
Background technique
As electric car is in the popularization of China and the application of car networking technology, more and more electric cars, which enter, to disappear
Expense person market and travelling data is acquired in real time.For the battery management system of one of electric car core component, still stop
Stay in the stage judged by the threshold value of predefined.Maintenance management to battery is by inspecting periodically and based on event
Method does not comprehensively consider the true driving situation of vehicle, personalized analysis is carried out for different driving behaviors, especially in electricity
Before pond failure occurs, predictive behave can not be taken, affects vehicle maintenance expense in this way, and then to client's body of car owner
It tests and adversely affects, electric car manufacturer leads to high cost of serving and product recall due to that can not identify later product problem.
At present to the management of batteries of electric automobile data, substantially dependence historical empirical data show that one is substantially repaired
Time and life curve, what the battery management after factory substantially carried out on this basis.It is different due to driving situation complexity
Vehicle condition and driving behavior all have a great impact to the performance of battery, and empirical data only has referential and can not effectively instruct true
The maintenance of truth condition, at present lack a data-driven method system battery use is analyzed, thus obtain whether
The indexs such as faulty and remaining power service life.
Summary of the invention
In order to solve this problem, it the present invention provides a kind of approaches of predictive maintenance of data-driven, is namely based on big
The applied analysis system of batteries of electric automobile predictive maintenance is established in data machine learning.
The present invention provides a kind of batteries of electric automobile approaches of predictive maintenance, the method packets to solve the above-mentioned problems
Include: step 001 data preparation step obtains and uses relevant data to batteries of electric automobile；The batteries of electric automobile uses
Correlation includes the use data of breakdown maintenance data and battery；Wherein, before the breakdown maintenance data include cell malfunctions
Data record and/or battery mantenance data；The use data of the battery are included in relevant to battery when normal use
Battery itself data and vehicle condition data；The breakdown maintenance data, the use data of battery are all based on time series
Stream data；Step 002 data preparation step, to the batteries of electric automobile using relevant data carry out cleaning and will be clear
The batteries of electric automobile after washing is based on time quantum using relevant data and carries out data building；It is described that data are carried out clearly
Wash bags include, and the assignment of vacant variable is carried out using the average value or median or neighbor interpolation that take a trip variable；Pass through
Setting batteries of electric automobile checks that data will exceed normal model whether meet the requirement using the threshold value of each variable of related data
The data enclosed are deleted or are corrected；The mutual constraint and dependence of related data are used by setting batteries of electric automobile,
Data unreasonable or conflicting in logic are deleted or corrected；Data building includes, suitable according to the time
Sequence integrates the other data collected；Step 003 data characterization step, the number that will be obtained by data preparation step
According to summarizing and extracting, the data after characterizing are obtained；Summary and extraction for data include rolling polymerization, the rolling
Polymerization refers to one time window of setting, calculates the polymerizing value in scheduled variable in the time window, the polymerizing value can
To be summation, the average value either standard deviation of data；The summary and extraction further include being extended characteristic variable, described
Extension includes to initial characteristic variable according to the corresponding number of mean value increase for rolling polymerization, and to initial characteristic variable
Increase corresponding number according to the standard deviation for rolling polymerization；Step 004 establishes model step, is established based on the data after characterization
Battery predictive safeguards adaptive model；The problem of for battery predictive maintenance, resolving into first subproblem is that battery is
It is no will break down and second subproblem battery there are also how long can break down；For first subproblem be battery whether
It will break down, and establish the battery predictive maintenance adaptive model in present embodiment using binary classification model；
For second subproblem battery there are also that how long can break down, the battery predictive maintenance is established using regression model certainly
Adaptive model；Step 005 trains verification step, is trained and verifies to adaptive model to optimize the adaptive model；Institute
It states trained verification step and preferably includes cross validation, the cross validation includes that original data are randomly divided into K portion first
Point, select one of part as test data in this K part, remaining K-1 part is obtained as training data
Corresponding experimental result；Then, another part is selected as test data, and remaining K-1 part is used as training data；
And so on, repeat K crosscheck, experiment all selects a different part as test from K part every time
Data guarantee that the data of K part all did test data respectively, and remaining K-1 are tested as training data；Finally
K obtained experimental result is averaged；Optimal data classification is determined based on the experimental result；Step 006 algorithm evaluation step
Suddenly, prediction result of the assessment data under algorithms of different, optimal algorithm is selected based on assessment；The assessment is commented including accuracy
Estimate, recall rate assessment or comprehensive evaluation index are assessed；The accuracy refers to the ratio that is consistent that prediction result actually really occurs
Rate, accuracy assessment take algorithm corresponding to maximum numerical value；The recall rate refers to that how many really occurred is predicted just
Really, accuracy assessment takes algorithm corresponding to maximum numerical value；Comprehensive evaluation indexWherein, α is meter
Parameter is calculated, P is accuracy, R is recall rate, and the result F obtained according to algorithms of different judges different algorithms in different rings
Superiority under border.
This method has determined the key problem of battery failures and remaining life in batteries of electric automobile management, for the core
The acquisition and calibration of heart problem progress data and progress Data Integration and Feature Engineering, explicit data are defined and are carried out preliminary
Processing carries out the definition of feature and label by predefined rule.It is finally to carry out model training and assessment, is led by data
Enter, using the different models of machine learning, algorithms of different is selected to carry out matching verifying, and issued, become the production of structuring
Product, and with accumulated time and data rich, the forecasting accuracy of model can be promoted constantly.
Detailed description of the invention
Fig. 1 is batteries of electric automobile predictive maintenance embodiment；
Fig. 2 is system structure diagram of the invention；
Fig. 3 is big data machine learning block diagram of the invention；
Fig. 4 is that polymerization schematic diagram is rolled in the present invention.
Specific embodiment
Specific implementation of the patent mode is described in detail with reference to the accompanying drawing, it should be pointed out that the specific reality
Applying mode is only the citing to optimal technical scheme of the present invention, can not be interpreted as limiting the scope of the invention.
Fig. 1 shows the step of one of this patent specific embodiment batteries of electric automobile approaches of predictive maintenance.Wherein:
Step S001 data preparation step obtains and uses relevant data to batteries of electric automobile.
In this step, the data of the batteries of electric automobile include the use data of breakdown maintenance data and battery.Its
In, the breakdown maintenance data include the mantenance data of the data record and/or battery before cell malfunctions.The battery
Battery itself data relevant to battery and vehicle condition data when being included in normal use using data.
The stream data that time series is all based on using data of the breakdown maintenance data, battery, including but it is unlimited
In voltage, electric current, remaining capacity (SOC) etc..A kind of citing but not all data content is as shown in the table.
S002 data preparation step, to the batteries of electric automobile using relevant data carry out cleaning and will be after cleaning
The batteries of electric automobile is based on time quantum using relevant data and carries out data building.
In the present embodiment, due to being mainly based upon data processing realization, guarantee that the data of high quality are conducive to mention
The accuracy of high result, it is therefore desirable to which data preparation is carried out to the data of acquisition.The data preparation first has to carry out data
Cleaning, the present invention have formulated corresponding cleaning rule and have converted data of low quality to the data for meeting quality of data requirement.
Cleaning rule includes:
Vacant assignment: battery data is in transmission process, it is easy to and occurring to exchange causes variable to lack, in the present invention,
The main assignment that vacant variable is carried out using the average value or median or neighbor interpolation that take a trip variable.
Mistake value removal: by setting batteries of electric automobile using the reasonable value range of each variable of related data, i.e.,
Threshold value checks data whether meet the requirement, and the data that will exceed normal range (NR) are deleted or corrected.
Crosscheck: by setting batteries of electric automobile using the mutual constraint of related data and dependence, by logic
Upper unreasonable or conflicting data are deleted or are corrected.
It cleans after data, data building is carried out based on time quantum, i.e., it is other by what is collected according to the sequence of time
Data are integrated.Time quantum can be based on millisecond, second, minute etc., and time quantum can be different with the frequency of collection
It causes.
After completing data building, need to be assessed and corrected to based on the data that time quantum is constructed.Institute
Commentary is estimated including filtering out wrong data, i.e., there are those of mistake data for data itself.E.g., including but be not limited to, it lacks
Value, exceptional value, time cycle mistake and calculating specification mistake etc..After evaluation, the wrong data is corrected.Example
Such as missing values, the value that null will be present is set as 0, supplements the data of missing；For exceptional value, 0 is set by negative value, is kept away
Exempt from occur mistake in training process；For the numerical value of time cycle mistake, the time cycle should be clearly obtained, adjusts and transports again
Row data；For calculating the numerical value of specification mistake, bore adjustment and again operation data are specified.
The data obtained by data preparation step are summarized and are extracted by S003 data characterization step, obtain special
Data after signization.
Due to needing to be handled data and calculated in subsequent processing step, for ease of calculation with identification data
Feature, it is necessary first to reduced data is characterized in order to show the various features of the data consequently facilitating meter
It calculates and identifies.
It in this step, include rolling polymerization for the summary of data and extraction.The rolling polymerization refers to setting one
Time window, calculates the polymerizing value in scheduled variable in the time window, and the polymerizing value can be the summation of data, put down
Mean value either standard deviation.As shown in figure 4, such as t1 node, setting time window are 3, its rolling polymerization is exactly to calculate t1 section
Summation, mean value or the standard deviation of point and 3 nodes between the t1 node.
In this step, more preferable in order to provide learning algorithm, even additional study and predictive ability need
More multivariate data, invention are summarized and are extracted from the battery data based on time series, thus by initial S001
Characteristic variable is extended.For example, when there is 65 characteristic variables in step S001, in this example, the number being extended
According to mainly two classes: first is to increase 65-2=63 according to the mean value for rolling polymerization to initial 65 characteristic variables greatly；Second
Class is to increase 65-2=63 according to the standard deviation for rolling polymerization to 65 initial characteristic variables；The change finally obtained in this way
Amount is 65+63+63=191.This makes it possible to provide more multivariate data, so that being conducive to learning algorithm provides more preferable and prediction
Ability.
S004 data calculate step, establish battery predictive maintenance adaptive model based on the data after characterization.
The problem of for battery predictive maintenance, two sub-problems can be resolved into, first subproblem be battery whether
It will break down；Second subproblem is that there are also how long can break down for battery.Difference can be passed through for different problems
Model and algorithm go to be predicted.
Whether will break down for battery, it is pre- to establish the battery in present embodiment using binary classification model
The property surveyed maintenance adaptive model.
Specifically, the battery data of input is set as x；It is set as target judging whether battery will break down and is
Y, then the individual of y is only there are two types of selection, y=1 as breaks down, and y=0 is to break down.
The model of so binary classification is: y=f (x), wherein f is specific algorithm, battery data x can be mapped to mesh
It marks in y.
When being trained above-mentioned model using initial training data, need to carry out label to initial training data set,
Using the data to break down as positive (label 1), using the data of normal operation as reversed (label 0), it is established that
Next cycle possible breakdown or normal mode y=f (x), wherein y is whether battery will break down, and x is battery
Data, f are specific algorithm.
Wherein, the specific algorithm f selectively includes: logistic regression, promotes decision tree, decision forest and nerve net
Network.
The logistic regression algorithm assumes that the example of class is linear separability, passes through the gain of parameter of direct estimation discriminate
Final prediction model.Consider vector x '=(x that there is P independent variable for the data of electric car predictive maintenance1,
x2... xp), if conditional probability P (Y=1 | x)=p is the probability occurred according to observed quantity relative to certain event.Logistic regression is collinear
Property return equally required a hypothesis function, Sigmoid function is introduced in this algorithmWherein π
(x) domain is (- ∞ ,+∞), and codomain is (0,1).According to defined above, formula used by the logistic regression algorithm
Are as follows:
The promotion decision Tree algorithms are by combining the hierarchical data structure of decision tree divide-and-conquer strategy to initial classification
The data weighting of last misclassification is improved every time and is a little classified again by generated classifying rules, such loop iteration
Obtain objective result.
If D is the division that use classes carry out training tuple, then the entropy of D indicates are as follows:
Wherein, pi indicates what i-th of classification occurred in entire training tuple
Probability can be used and belong to the quantity of this class elements divided by training tuple elements total quantity as estimating.The practical significance table of entropy
Show it is average information required for the class label of tuple in D.For this prediction technique, D is battery failures situation, is had
Failure and normal two states, so m=2.
If training tuple D is divided by attribute A, wherein A be after characterization, battery data one of them
Feature, then the expectation information that A divides D are as follows:Wherein j indicates certain of attribute A
A type, V indicate the classification sum of attribute A；And the information gain of attribute A is the difference of the two: gain (A)=info (D)-
infoA(D).Need to calculate the information gain of each attribute in battery data training tuple at every secondary clearing (division), then
The selection maximum attribute of ratio of profit increase is layered, and the decision tree for being able to carry out electric car predictive maintenance thus can be formed.
The forest that decision forest is made of multiple decision trees, algorithm classification result are voted to obtain by these decision trees, decision
Tree adds random process on line direction and column direction respectively in the process of generation, uses when constructing decision tree on line direction
Sampling with replacement (bootstraping) obtains training data, puts back to random sampling using nothing on column direction and obtains character subset, and
Its optimal cut-off is obtained accordingly.Decision forest is a built-up pattern, and inside is still based on decision tree, with single decision
Unlike tree classification, decision forest is classified by multiple decision tree voting results, and algorithm is not easy overfitting occur
Problem.
Neural network is exactly that the second way of human brain thinking is simulated using its algorithm characteristic, it is a Nonlinear Dynamic
Mechanical system is able to carry out concurrent collaborative processing although the structure of single neuron and its simple.It is different in neural network
The output layer of scene corresponds to different cost functions, and in this method, output layer is K logistic regression, the cost letter of whole network
Number is exactly the adduction of this K Logic Regression Models cost function, can carry out batteries of electric automobile failure by this cost function
Prediction, the assessment of cost function carries out according to s006 algorithm evaluation.
For battery there are also that how long can break down, the battery is established in present embodiment using regression model
Predictive maintenance adaptive model.
Regression model determines the relationship between variable to the credible of these relational expressions from one group of sample data
Degree carries out various statistical checks, and the influence for finding out from all multivariables for influencing a certain particular variables which variable is significant,
Which is not significant.
Time to break down marks each battery data from the time that time upper distance breaks down as Y
Labelization；For example, when battery is used for 5 days, fault time is 300, remaining time represented by the label is
300-5=295；In another example when battery is used for 10 days, fault time is 280, remaining time represented by the label
For 280-10=270.Sample each in this way can have a remaining pot life.Specific label is as shown in the table:
The battery data of input is set as x；The model of regression algorithm is Y=f (x).Wherein, the regression model is used
Specific algorithm f include decision forest algorithm return, promoted decision tree return, Poisson regression and neural net regression.
Promoting decision tree and returning with decision forest recurrence is made of decision tree one or several decision trees, is
The combination of decision tree and the battery whether will break down it is middle using the relevant algorithm of decision tree as, battery also
In the regression model that how long can be broken down, also using judging to promote what decision tree and decision forest returned using information gain
Quality passes through difference:
Gain (A)=info (D)-infoA(D), judge.
In Poisson regression, modeled using the Poisson regression model recorded extensively in the prior art.
Neural network is exactly a kind of algorithm for the simulation human brain thinking recorded extensively in the prior art.Neural network
In, the output layer of different scenes corresponds to different cost functions.In this method, output layer can be K logistic regression, entire net
The cost function of network is exactly the adduction of this K Logic Regression Models cost function.
S005 trains verification step, is trained and verifies to adaptive model to optimize the adaptive model.
On the basis of establishing above-mentioned model, the work for needing to be trained and verify carrys out Optimized model.In order to improve
The accuracy of model.
In this embodiment, the trained verification step preferably includes cross validation and minority class sampling.
The cross validation method of parameter frame in to(for) each model optimizes.Such as disaggregated model above-mentioned
(decision forest algorithm returns, promotes decision for (logistic regression promotes decision tree, decision forest and neural network) and regression model
Tree algorithm returns, Poisson algorithm returns and neural network algorithm returns), the reliability of these algorithms relies on parameter frame, is exactly
Say which battery data for generation the result is that most effective.
In this embodiment, in order to improve the quality of parameter frame, original data are randomly divided into K first
Part.In this K part, select one of part as test data, remaining K-1 part is obtained as training data
To corresponding experimental result.Then, another part is selected as test data, and remaining K-1 part is as training number
According to, and so on, repeat K crosscheck.Every time experiment all selected from K part a different part as
Test data guarantees that the data of K part all did test data respectively, and remaining K-1 are tested as training data.
Finally K obtained experimental result is averaged, the experimental result may include accuracy, recall rate and comprehensive evaluation index
Deng.According to the purpose of each predictive maintenance, in the selection of accuracy, recall rate and three kinds of comprehensive evaluation index of mean value, thus
Determine optimal classification, the training of implementation model.
The minority class sampling is when only having small number of training sample for a kind of data, and data set is unbalanced
It is used when situation.It, can be by will be a small number of in present embodiment when a kind of data only have a small amount of training sample
The new minority class sample data of fault sample Data Synthesis carry out the training of model.Such as in the data collection of battery,
Only discovery has a small amount of fault record data, in order to generate more data for carrying out machine learning from a small amount of fault data,
It needs to carry out Data Synthesis.Specifically, a sample B is selected at random from its arest neighbors to each minority class sample A, this
In distance be to be calculated according to the distance in time and variogram, then randomly choosed a bit on the line between A and B
As newly synthesized minority class sample.Continuous synthesis in this way, a small amount of sample A can be become to have multidata
Sample A+ will not be generated in calculating because of mistake caused by data nonbalance to reach the data demand of predictive maintenance
Fitting or distortion.
S006 algorithm evaluation step is assessed prediction result of the data under algorithms of different, optimal calculation is selected based on assessment
Method.
In the predictive maintenance of battery, based on different prediction targets or it is different data source, using different
The obtained result of algorithm is also different, and thus needs to select preferable algorithm for different situations.
Usually in batteries of electric automobile predictive maintenance, accuracy (Precision), recall rate can be used
(Recall) or comprehensive evaluation index (F1-Measure) carrys out assessment prediction as a result, more in varied situations using different
Whether the obtained result of algorithm is optimal, to select optimal algorithm.
Wherein, accuracy is that how many is practical in the model prediction is broken down for prediction result sample
The sample really to break down, usually the higher the better.The recall rate is how many quilt really to break down in sample
Predict correct, usually the higher the better.
In battery predictive maintenance, the two usually conflicts.In order to improve for the reasonable of more excellent algorithms selection
Property, F1-Measure comprehensive evaluation index is preferably used in this embodiment, it combines accuracy and recall rate
Weighted average, the higher the better for value.Formula isWherein P is accuracy, and R is recall rate, when parameter alpha=1
When, it is exactly the most common F1, namelyThe result F or F1 obtained according to algorithms of different is different to judge
The superiority of algorithm in different environments.Such as a certain group of specific data and prediction target, after calculating relatively
It was found that such data and target select to promote decision Tree algorithms in disaggregated model and select neural network in regression model
Regression algorithm result is optimal.
Claims (5)
1. a kind of batteries of electric automobile approaches of predictive maintenance, which is characterized in that the described method includes:
Step 001 data preparation step obtains and uses relevant data to batteries of electric automobile；
The batteries of electric automobile uses the related use data including breakdown maintenance data and battery；Wherein, the failure dimension
Repair the mantenance data that data include the data record and/or battery before cell malfunctions；The use data of the battery include
Battery itself data relevant to battery and vehicle condition data in normal use；The breakdown maintenance data, battery
The stream data of time series is all based on using data；
Step 002 data preparation step, to the batteries of electric automobile using relevant data carry out cleaning and will be after cleaning
The batteries of electric automobile is based on time quantum using relevant data and carries out data building；It is described that cleaning packet is carried out to data
It includes, the assignment of vacant variable is carried out using the average value or median or neighbor interpolation that take a trip variable；Pass through setting electricity
Electrical automobile battery checks that data will exceed the number of normal range (NR) whether meet the requirement using the threshold value of each variable of related data
According to being deleted or corrected；By setting batteries of electric automobile using the mutual constraint of related data and dependence, by logic
Upper unreasonable or conflicting data are deleted or are corrected；The data building includes that will search according to the sequence of time
The other data collected are integrated；
The data obtained by data preparation step are summarized and are extracted by step 003 data characterization step, obtain feature
Data after change；Summary and extraction for data include rolling polymerization, and the rolling polymerization, which refers to, sets a time window,
The polymerizing value in scheduled variable in the time window is calculated, the polymerizing value is the summation of data, average value either mark
It is quasi- poor；It is described summary and extraction further include being extended characteristic variable, it is described extension include to initial characteristic variable according to
It rolls the mean value polymerizeing and increases corresponding number, and initial characteristic variable is increased accordingly according to the standard deviation for rolling polymerization
Number；
Step 004 establishes model step, establishes battery predictive maintenance adaptive model based on the data after characterization；For electricity
The problem of pond predictive maintenance, resolving into first subproblem is whether battery will break down and second subproblem battery
There are also how long can break down；Be whether battery will break down for first subproblem, using binary classification model come
Establish the battery predictive maintenance adaptive model；For second subproblem battery there are also that how long can break down, use
Regression model safeguards adaptive model to establish the battery predictive；
Step 005 trains verification step, is trained and verifies to adaptive model to optimize the adaptive model；The training
Verification step includes cross validation, and the cross validation includes that original data are randomly divided into K part first, at this K
Select one of part as test data in part, remaining K-1 part is tested accordingly as training data
As a result；Then, another part is selected as test data, and remaining K-1 part is used as training data；And so on,
Repeat K crosscheck, experiment all selects a different part as test data from K part every time, guarantees K
The data of a part all did test data respectively, and remaining K-1 are tested as training data；Finally obtained K
A experimental result is average；Optimal data classification is determined based on the experimental result；
Step 006 algorithm evaluation step is assessed prediction result of the data under algorithms of different, optimal calculation is selected based on assessment
Method；The assessment includes accuracy assessment, recall rate assessment or comprehensive evaluation index assessment；The accuracy refers to prediction
As a result the practical ratio that is consistent really occurred, accuracy assessment take algorithm corresponding to maximum numerical value；The recall rate refers to
Really occur how many be predicted correctly, recall rate assessment take algorithm corresponding to maximum numerical value；Comprehensive evaluation indexWherein, α is calculating parameter, and P is accuracy, and R is recall rate, according to the result F that algorithms of different obtains come
Judge the superiority of different algorithms in different environments.
2. a kind of batteries of electric automobile approaches of predictive maintenance according to claim 1, which is characterized in that complete data structure
After building, is assessed and corrected to based on the data that time quantum is constructed；The assessment includes garbled data itself
There are those of mistake data；After evaluation, the wrong data is corrected；The correction includes: for missing
Value, sets 0 for missing values；For exceptional value, 0 is set by negative value；For the numerical value of time cycle mistake, should clearly take
Obtain time cycle, adjustment and again operation data；For calculating the numerical value of specification mistake, specifies bore and adjust and rerun number
According to.
3. a kind of batteries of electric automobile approaches of predictive maintenance described in any one of -2 according to claim 1, which is characterized in that
The binary classification model includes: that the battery data of input is set as x；It is set as mesh judging whether battery will break down
It is designated as y, y=1, is as broken down, y=0 is to break down；The model of binary classification is: y=f (x), and wherein f is tool
Body algorithm；The specific algorithm includes: logistic regression, promotes decision tree, decision forest and neural network.
4. according to a kind of batteries of electric automobile approaches of predictive maintenance described in claim 3, which is characterized in that in the recurrence
In model, the time to break down carries out label from the time that time upper distance breaks down to each battery data as Y
Change；The battery data of input is set as x；The model of regression algorithm is Y=f (x), and wherein f is specific algorithm, comprising: decision is gloomy
Woods returns, promotes decision tree recurrence, Poisson regression and neural net regression.
5. a kind of batteries of electric automobile approaches of predictive maintenance according to claim 4, which is characterized in that the step
It further include that minority class sampling is trained the model in 005, when data a kind of in sample only have a small amount of training sample
When, by the training that the new minority class sample data of a small number of fault sample Data Synthesis is carried out to model；To each minority
Class sample A selects a sample B from it at random in arest neighbors, and the distance is according to the distance in time and variogram
It is calculated, then random selection is a little used as newly synthesized minority class sample on the line between A and B；By continuous
Synthesis, by a small amount of sample A, becomes have multidata sample A+.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610504843.4A CN106168799B (en) | 2016-06-30 | 2016-06-30 | A method of batteries of electric automobile predictive maintenance is carried out based on big data machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610504843.4A CN106168799B (en) | 2016-06-30 | 2016-06-30 | A method of batteries of electric automobile predictive maintenance is carried out based on big data machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106168799A CN106168799A (en) | 2016-11-30 |
| CN106168799B true CN106168799B (en) | 2019-05-03 |
Family
ID=58065779
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610504843.4A Active CN106168799B (en) | 2016-06-30 | 2016-06-30 | A method of batteries of electric automobile predictive maintenance is carried out based on big data machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106168799B (en) |
Families Citing this family (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107122594B (en) * | 2017-04-10 | 2020-06-26 | 湖南中车时代电动汽车股份有限公司 | New energy vehicle battery health prediction method and system |
| CN110019654A (en) * | 2017-07-20 | 2019-07-16 | 南方电网传媒有限公司 | A kind of unbalance network text classification optimization system |
| CN107560849B (en) * | 2017-08-04 | 2020-02-18 | 华北电力大学 | Wind turbine generator bearing fault diagnosis method of multichannel deep convolutional neural network |
| CN108693478A (en) * | 2018-04-17 | 2018-10-23 | 北京理工大学 | A kind of method for detecting leakage of lithium-ion-power cell |
| CN108764494A (en) * | 2018-05-15 | 2018-11-06 | 中山职业技术学院 | Vehicle periodic maintenance manages system and its maintenance measure decision-making technique, computer readable storage medium |
| CN108958215A (en) * | 2018-06-01 | 2018-12-07 | 天泽信息产业股份有限公司 | A kind of engineering truck failure prediction system and its prediction technique based on data mining |
| US20200020186A1 (en) * | 2018-07-11 | 2020-01-16 | Acsys Holdings Limited | Systems and methods for providing an access management platform |
| CN109190670B (en) * | 2018-08-02 | 2020-04-07 | 大连理工大学 | Charging pile fault prediction method based on extensible hoisting tree |
| EP3620983B1 (en) | 2018-09-05 | 2023-10-25 | Sartorius Stedim Data Analytics AB | Computer-implemented method, computer program product and system for data analysis |
| CN111025153A (en) * | 2018-10-09 | 2020-04-17 | 上海汽车集团股份有限公司 | Electric vehicle battery fault diagnosis method and device |
| CN109375109A (en) * | 2018-10-16 | 2019-02-22 | 宁波工程学院 | A kind of accumulator failure diagnostic method based on data mining |
| CN109583632B (en) * | 2018-11-16 | 2021-02-26 | 北京航空航天大学 | Electric vehicle charging behavior prediction method based on internet data |
| CN109738782B (en) * | 2018-12-07 | 2020-11-06 | 上海交通大学 | Circuit board fault detection method and system based on measured data machine learning |
| CN111381166A (en) * | 2018-12-29 | 2020-07-07 | 北京骑胜科技有限公司 | Method and system for monitoring battery state |
| CN109978229A (en) * | 2019-02-12 | 2019-07-05 | 常伟 | The method that the full battery core multi-point temperature of a kind of pair of power battery pack and tie point temperature carry out thermal runaway prediction |
| CN109902875B (en) * | 2019-03-01 | 2021-06-15 | 联想(北京)有限公司 | Information processing method and information processing device |
| CN110084404B (en) * | 2019-03-31 | 2020-07-24 | 唐山百川智能机器股份有限公司 | Big data-based railway vehicle economic operation and maintenance planning method |
| CN110010987B (en) * | 2019-04-12 | 2022-08-16 | 苏州正力新能源科技有限公司 | Big data-based electric vehicle remaining charging time prediction method |
| CN110154822A (en) * | 2019-05-14 | 2019-08-23 | 中科院合肥技术创新工程院 | A kind of charge/discharge control method applied to electric car Intelligent battery management system |
| CN110610016A (en) * | 2019-07-15 | 2019-12-24 | 广东毓秀科技有限公司 | Method for predicting rail transit stopping problem based on big data machine learning |
| CN110286668A (en) * | 2019-07-15 | 2019-09-27 | 广东毓秀科技有限公司 | A kind of rail friendship signal system VIM board faults prediction technique based on big data |
| CN110596594A (en) * | 2019-09-23 | 2019-12-20 | 广东毓秀科技有限公司 | Method for predicting SOE of rail-traffic lithium battery through big data |
| KR20210077065A (en) | 2019-12-16 | 2021-06-25 | 현대자동차주식회사 | System and method for controlling degradation of power relay assembly for vehicle |
| CN111191839B (en) * | 2019-12-26 | 2023-07-07 | 上海钧正网络科技有限公司 | Electricity replacement prediction method and system and storage medium |
| CN111242172A (en) * | 2019-12-31 | 2020-06-05 | 浙江吉利新能源商用车集团有限公司 | Battery portrait label generation method and device |
| CN111222584A (en) * | 2020-01-15 | 2020-06-02 | 北京辉腾格勒石墨烯科技有限公司 | Lithium battery real-time evaluation method based on big data and deep neural network |
| CN111563612B (en) * | 2020-04-13 | 2024-03-22 | 深圳达实智能股份有限公司 | Method and system for managing predictive operation and maintenance of air conditioner of subway station |
| CN113888775A (en) * | 2020-06-19 | 2022-01-04 | 比亚迪股份有限公司 | Vehicle early warning method, server, storage medium, vehicle early warning system and vehicle |
| CN112287178B (en) * | 2020-10-30 | 2023-11-17 | 腾讯科技（深圳）有限公司 | Data cleaning method and device, electronic equipment and storage medium |
| CN112319462B (en) * | 2020-11-17 | 2022-05-13 | 河南科技大学 | Energy management method for plug-in hybrid electric vehicle |
| CN112816898B (en) * | 2021-01-26 | 2022-03-01 | 三一重工股份有限公司 | Battery failure prediction method and device, electronic equipment and storage medium |
| CN112884210B (en) * | 2021-01-31 | 2022-03-15 | 中国人民解放军63963部队 | Vehicle health management system based on fuzzy clustering |
| CN112785074B (en) * | 2021-01-31 | 2022-02-01 | 中国人民解放军 63963部队 | Equipment maintenance and repair opportunity determination system and method based on deep learning |
| US11573271B1 (en) | 2021-08-25 | 2023-02-07 | Ford Global Technologies, Llc | Battery fault detection |
| CN113866659A (en) * | 2021-09-26 | 2021-12-31 | 中国华能集团清洁能源技术研究院有限公司 | Lithium battery SOH prediction method based on machine learning and ampere-hour integration method |
| CN114662710A (en) * | 2022-01-27 | 2022-06-24 | 武汉船用机械有限责任公司 | Method, device and equipment for determining maintenance requirement of marine engine room pump |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104459552A (en) * | 2014-11-28 | 2015-03-25 | 上海交通大学 | Method for evaluating influence of charging behavior on health condition of electric vehicle battery |
| CN105247379A (en) * | 2013-05-03 | 2016-01-13 | 力博特公司 | System and method for ups battery monitoring and data analysis |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8154297B2 (en) * | 2007-07-05 | 2012-04-10 | Abb Technology Ag | System and method for predictive maintenance of a battery assembly using temporal signal processing |
| US9316699B2 (en) * | 2012-04-05 | 2016-04-19 | Samsung Sdi Co., Ltd. | System for predicting lifetime of battery |
-
2016
- 2016-06-30 CN CN201610504843.4A patent/CN106168799B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105247379A (en) * | 2013-05-03 | 2016-01-13 | 力博特公司 | System and method for ups battery monitoring and data analysis |
| CN104459552A (en) * | 2014-11-28 | 2015-03-25 | 上海交通大学 | Method for evaluating influence of charging behavior on health condition of electric vehicle battery |
Non-Patent Citations (1)
| Title |
|---|
| 可重构卫星锂离子电池剩余寿命预测系统研究;周建宝 等;《仪器仪表学报》;20130930;第34卷(第9期);第2034-2044页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106168799A (en) | 2016-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106168799B (en) | A method of batteries of electric automobile predictive maintenance is carried out based on big data machine learning | |
| CN110929918B (en) | 10kV feeder fault prediction method based on CNN and LightGBM | |
| CN109978229A (en) | The method that the full battery core multi-point temperature of a kind of pair of power battery pack and tie point temperature carry out thermal runaway prediction | |
| CN110286668A (en) | A kind of rail friendship signal system VIM board faults prediction technique based on big data | |
| CN107274011A (en) | The equipment state recognition methods of comprehensive Markov model and probability net | |
| CN106912067B (en) | WSN wireless communication module fault diagnosis method based on fuzzy neural network | |
| CN106950945A (en) | A kind of fault detection method based on dimension changeable type independent component analysis model | |
| CN111275136B (en) | Fault prediction system based on small sample and early warning method thereof | |
| CN105844334A (en) | Radial basis function neural network-based temperature interpolation algorithm | |
| CN112163474B (en) | Intelligent gearbox diagnosis method based on model fusion | |
| CN104537383A (en) | Massive organizational structure data classification method and system based on particle swarm | |
| Pradeep et al. | Optimal Predictive Maintenance Technique for Manufacturing Semiconductors using Machine Learning | |
| CN109784629B (en) | Transformer substation industrial control network fault positioning method based on neural network | |
| CN112613227B (en) | Model for predicting remaining service life of aero-engine based on hybrid machine learning | |
| CN109547248A (en) | Based on artificial intelligence in orbit aerocraft ad hoc network method for diagnosing faults and device | |
| CN115392710A (en) | Wind turbine generator operation decision method and system based on data filtering | |
| CN113419496A (en) | Waterproof material production intelligent management and control method and system based on big data statistics | |
| CN116796617A (en) | Rolling bearing equipment residual life prediction method and system based on data identification | |
| Lin et al. | A method of satellite network fault synthetic diagnosis based on C4. 5 algorithm and expert knowledge database | |
| CN111143774B (en) | Power load prediction method and device based on influence factor multi-state model | |
| Zheng et al. | Machine learning-based quality evaluation of 5G wireless network in Smart Grid | |
| CN115174421B (en) | Network fault prediction method and device based on self-supervision unwrapping hypergraph attention | |
| CN117578481B (en) | Voltage sag estimation model building method, medium and system | |
| Zhao et al. | Research on machine learning-based correlation analysis method for power equipment alarms | |
| TW201347505A (en) | Telecommunication network warning correlation analysis method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221128 Address after: 200000 room 3272, building 4, No. 495, Yuejin South Road, Xinhai Town, Chongming District, Shanghai (Guangming rice industry economic Park) Patentee after: Shanghai Lejia Smart Energy Technology Co.,Ltd. Address before: 134399 Group 12, Fifth Construction Committee, Xinjian Street, Badaojiang District, Baishan City, Jilin Province Patentee before: Chang Wei |
|
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: Room 307, Building 10A, Zilang Science and Technology City, No. 60 Chongzhou Avenue, Development Zone, Nantong City, Jiangsu Province, 226010 Patentee after: Nantong Le Chuangxin Energy Co.,Ltd. Country or region after: China Address before: 200000 room 3272, building 4, No. 495, Yuejin South Road, Xinhai Town, Chongming District, Shanghai (Guangming rice industry economic Park) Patentee before: Shanghai Lejia Smart Energy Technology Co.,Ltd. Country or region before: China |