CN106095785B - Fault code diagnosis vehicle work item and spare part retrieval method based on decision tree classification - Google Patents

Abstract

Fault code diagnosis vehicle worker's item and spare part retrieval method based on decision tree classification belongs to the information retrieval field, because different spare parts are under the condition that the major group number is the same, and the group number still differs, in order to solve the problem of accurate matching spare part, the technical essential is: resolving a vehicle VIN code to obtain variables, wherein the variables comprise engine displacement, a vehicle body type and an engine gearbox type which are obtained by resolving the VIN code; performing decision tree analysis on the spare part codes corresponding to the variables, finishing classification of variable data to form spare part information, and establishing indexes to form a diagnosis knowledge base; creating a language model, establishing a cell word bank, cutting words in the cell word bank to search cell words, arranging the cell words, and forming a diagnosis database of the fault codes corresponding to the work items by utilizing decision classification of a decision tree model; associating the diagnosis database with a diagnosis knowledge base and establishing a main key; the effect is as follows: after the fault code is obtained, the solution of common faults and corresponding spare parts and work items can be quickly found.

Description

Fault code diagnosis vehicle work item and spare part retrieval method based on decision tree classification

Technical Field

The invention belongs to the field of information retrieval, and relates to a method for vehicle remote diagnosis and spare part retrieval

Background

At present, the automobile maintenance industry in China develops from a stage of diagnosis completely depending on the feeling and practical experience of inspectors to a stage of comprehensive detection and diagnosis by using special equipment, but a plurality of problems generally exist in the traditional automobile maintenance industry, such as technical aging of maintenance workers, and the problems that the faults can not be solved quickly and economically by using technical strength in all aspects are often caused; with the increasing automobile keeping quantity, various services in the automobile after-market are greatly developed like bamboo shoots in spring after rain. From the perspective of the vehicle owner, how to better and more comprehensively know the vehicle condition, how to quickly acquire the solution to be solved for the vehicle, the required working hours and the relevant information of the spare parts when the vehicle breaks down, and the accurate wearable device of the vehicle is completely necessary for meeting the real-time requirements of the vehicle owner. General OBD vehicle-mounted equipment can only read relevant vehicle fault information, and cannot make detailed solutions to faults and relevant maintenance labor cost and spare part cost, so that a vehicle owner can enter a store blindly and consume blindly.

Disclosure of Invention

In order to solve the problem that when a fault code vehicle which does not need to be identified and classified breaks down, the fault code can accurately and quickly match work items and spare parts corresponding to the fault code, the invention provides the following technical scheme: a fault code diagnosis vehicle work item and spare part retrieval method based on decision tree classification comprises

Acquiring vehicle information data;

resolving the VIN code of the vehicle to obtain variables, wherein the variables comprise the engine displacement, the vehicle body type and the engine gearbox type which are obtained by resolving the VIN code;

thirdly, performing decision tree analysis on the spare part codes corresponding to the variables, finishing classification of the variable data to form spare part information, and establishing indexes to form a diagnosis knowledge base;

creating a language model, establishing a cell word bank, cutting words in the cell word bank to search cell words, arranging the cell words, and forming a diagnosis database of the fault codes corresponding to the work items by utilizing decision classification of a decision tree model;

associating the diagnosis database with the diagnosis knowledge base and establishing a main key;

has the advantages that: according to the invention, after the fault code is obtained, the solution of common faults and corresponding spare parts and work items can be quickly found. Effectively solves the problem of experience limitation of technicians and spare parts personnel, and obtains a fault solution from big data. Step four of the invention constructs a maintenance cell word stock from fault description to word vector decomposition; by means of word segmentation retrieval of fault description in a cell word bank and according to a syntactic rule established by cell word weight, a maintenance decision tree from a decision process to a decision suggestion to a maintenance solution is realized, and finally, remote fault diagnosis is realized

Drawings

FIG. 1 is a flow chart of a method for remote vehicle diagnosis and spare part retrieval in accordance with the present invention;

fig. 2 is a schematic diagram illustrating the translation of the VIN number of the vehicle with the chassis number LFV5a14B8Y 3000001.

Detailed Description

Example 1: a fault code diagnosis vehicle work item and spare part retrieval method based on decision tree classification comprises

Acquiring vehicle information data;

resolving the VIN code of the vehicle to obtain variables, wherein the variables comprise the engine displacement, the vehicle body type and the engine gearbox type which are obtained by resolving the VIN code;

thirdly, performing decision tree analysis on the spare part codes corresponding to the variables, finishing classification of the variable data to form spare part information, and establishing indexes to form a diagnosis knowledge base;

creating a language model, establishing a cell word bank, cutting words in the cell word bank to search cell words, arranging the cell words, and forming a diagnosis database of the fault codes corresponding to the work items by utilizing decision classification of a decision tree model;

and step five, associating the diagnosis database with the diagnosis knowledge base and establishing a main key.

According to the scheme, when the vehicle fault code is collected, the sixth step is executed, the fault code generated by the vehicle fault is identified, and the variable obtained by analyzing the vehicle VIN code through the keywords is classified and searched to obtain the work item and spare part information.

Example 2:having the same technical solution as that of the embodiment 1, more specifically, for the step four of the embodiment 1,

in the third step, the historical records of the maintenance spare part table are used as data bases, the spare parts are classified through a decision tree model, and the maintenance spare part table sample is shown as table one:

watch 1

The basic principle of the decision tree model is as follows:

firstly: determining the entropy of the different classes of spare parts in each dimension, taking VIN4 as an example, the entropy is defined as

E＝sum(-p(I)*log(p(I)))

Where I ═ 1: N (class N results, as in example 1, i.e. the spare part belongs to the vehicle type, so the probability p (I) ═ 1)

Then E (5) ═ 1/1) Log2(1/1) - (0/1) Log2(0/1) ═ 0+0 ═ 0

E(3)＝-(1/1)Log2(1/1)-(0/1)Log2(0/1)＝0+0＝0

E(4)＝-(1/1)Log2(1/1)-(0/1)Log2(0/1)＝0+0＝0

If the entropy is 0, the higher the discrimination is; if the entropy is 1, no discrimination is indicated;

the three different spare part codes may be distinguished by the VIN 4.

After determining how each dimension is classified, the priority levels of different dimensions are distinguished by information gain

Gain(Sample,Action)＝E(sample)-sum(|Sample(v)|/Sample*E(Sample(v)))

Then Gain (VIN4) ═ E(s) - (1/3) × E (5) - (1/3) × E (3) - (1/3) × E (4) ═ 1-0 ═ 1

Gain(VIN6)＝E(S)-(1/3)*E(1)-(2/3)*E(2)＝1-0-2/3＝1/3

Gain(VIN78)＝E(S)-(1/3)*E(4B)-(1/3)*E(8K)-(1/3)*E(4F)＝1-0＝1

If the information gain is larger, the classification priority is higher; conversely, the lower the priority.

Therefore, the classification priority of the 4 th order of the chassis number (VIN4) is the same as that of the 78 th order of the chassis number (VIN78), followed by the 6 th order of the chassis number (VIN 6).

Through the above key steps, spare part codes can be distinguished according to the 4 th position (VIN4) of the chassis number, the 6 th position (VIN6) of the chassis number and the 78 th position (VIN78) of the chassis number.

In summary, the basic steps of the spare part retrieval method are as follows:

distinguishing the same dimension of the maintenance spare part table according to the information entropy;

the different dimensions of the maintenance spare part table are prioritized according to information gain;

drawing a decision tree according to the priorities and the distinguishing degrees divided in the steps 1 and 2;

and inputting a regular chassis number, and outputting spare part codes under the vehicle type by the system according to the VIN123, the VIN4, the VIN6, the VIN78 and the obtained decision tree.

The spare part code obtains the Chinese name and price of the spare part, the current use state and the applicable vehicle type information of the spare part by associating a spare part price table.

The technical scheme is obtained by classifying spare part codes corresponding to different vehicle types, different displacement volumes and different engine gearbox types one by one for analysis and comparison, finding that the corresponding spare part codes are different under the premise of the same main group number, the vehicle type displacement volumes and other information are different, and using the method for finding the rule thereof to form a complete and comprehensive theoretical knowledge information base.

Example 3:having the same technical solution as that of the embodiment 1 or 2, more specifically, for the step four of the embodiment 1,

the creating of the language model in the fourth step and the building of the cell word stock comprise the following steps:

s1.1, collecting a professional fault description language;

s1.2, performing word vector decomposition on the professional fault description language.

The language model is created based on the assumption that the nth cell word occurs in relation to only the first n-1 cell words; the calculation formula of the occurrence weight of the fault description statement T is as follows:

P(T)＝P(w₁,w₂,w₃,…,w_n)

＝P(w₁)×P(w₂|w₁)×P(w₃|w₁,w₂)×…×P(w_n|w₁,w₂,…,w_n-1)

≈P(w₁)×P(w₂|w₁)×P(w₃|w₂)…P(w_n|w_n-1)；

wherein P (T) is the weight of the fault description statement T, P (w)_n|w₁,w₂,…,w_n-1) Is the weight of the nth cell word.

The step of cutting words in the cell word bank in the step four to search cell words and arranging the cell words comprises the following steps:

s2.1, performing word segmentation retrieval in the cell word bank aiming at the fault description statement T;

s2.2, if the cell words are searched, calculating the weight of the cell words;

s2.3, arranging the cell words according to the weight of the cell words.

S2.4, if the cell words are not searched, storing the cell words which are not searched in the time into a newly-added cell word bank.

The method for calculating the cell word weight in the step S2.2 comprises the following steps:

s2.2.1 calculating chi-square statistic for each of said cell words;

s2.2.2, taking the ith cell word with the highest chi-square statistic score, and calculating the occurrence frequency of the ith cell word in the jth fault description; the calculation method of the chi-square statistic score in the S2.2.1 step is as follows:

weight＝round((10×(1+lg(tf_ij))/(1+lg(l_j))),n)；

wherein tf is_ijThe number of times the ith cellular word appeared in the jth failure description, l_jN is an integer from 3 to 6 for the length of the jth fault description.

And forming a diagnosis database of the fault codes corresponding to the work items by utilizing decision classification of the decision tree model, wherein the step also comprises the step of providing a maintenance solution (work item) by combining component measurement values.

More specifically, the present invention is directed to a method for producing,

the present embodiment achieves the conversion of the problem of natural language understanding into the problem of machine self-learning by digitizing the natural language.

(1) Constructing a base corpus

The method of creating the base corpus is to give a string whose natural language probability is P (w1, w2, w3, …, wn), w1 to wn being the respective cell words of this sentence in turn.

The model is based on the assumption that the occurrence of the nth word is only related to the first n-1 words and not to any other words, and that the probability of a complete sentence is the product of the probabilities of occurrence of the respective participles.

For a fault description T, its probability calculation formula is:

P(T)＝P(w₁,w₂,w₃,…,w_n)

＝P(w₁)×P(w₂|w₁)×P(w₃|w₁,w₂)×…×P(w_n|w₁,w₂,…,w_n-1)

≈P(w₁)×P(w₂|w₁)×P(w₃|w₂)…P(w_n|w_n-1)

(2) and (3) decomposing the word vector: the word vector is a sentence that is segmented into a number of cell words. Each fault description T may be represented as an n-dimensional word vector, T ═ w₁,w₂,...,w_n) Wherein w is₁To w_nFor each cell word contained in each word vector.

Taking the automobile fault as an example:

the fault is described as:

P0021A camshaft position (Cylinder Row 2) -timing over-advanced

The fault description statement is expressed by a word vector, and the fault description statement can be expressed by a word vector (w)₁,w₂,w₃,…,w_n). Wherein, w₁Being a camshaft, w₂Is a position, w₃Is a cylinder, w₄Is a positive timing, w₅Is excessive. Selecting each cell word as a characteristic item, wherein w_iThe ith feature item is represented.

Calculating chi-square statistic of each cell word, and taking the ith cell word with the highest score in the training set as theRepresents a word to find a word vector (w)₁,w₂,w₃,…,w_n) The least relevant cell word.

w_iThe formula for calculating the weight of (a) is:

weight＝round((10×(1+lg(tf_ij))/(1+lg(l_j))),3)；

wherein tf is_ijFor the number of times the ith word vector appears in the jth fault description, l_jThe length of the description for the jth fault. The Round function is rounded to the designated decimal place, which is the third place after the decimal point in the above equation.

The feature expression of the fault descriptor vector after the weight is calculated is as follows:

vector participle	Weight of
		Cam shaft	0.286
Position of	0.191
		Cylinder	0.048
Timing of	0.048
		Transition of	0.048

(1) Search arrangement cell word:

as shown in fig. 1, when the user inputs the fault description T, the fault description T is first subjected to word vector decomposition to segment the complete sentence into a combination of a plurality of cell words. Then, word segmentation search is carried out in the cell word bank aiming at each cell word. If so, calculating the weight of the cell word; if not, storing the cell words which are not searched in the newly added industry cell word bank, searching the rest cell words, and calculating the probability.

And calculating weights according to the cell words searched in the cell word bank, and arranging the cell words from large to small according to the weights.

(2) And (3) analysis results:

taking the automobile fault as an example:

the fault is described as:

P0021A camshaft position (Cylinder Row 2) -over-timing Advance

And (3) searching each cell word in the cell word bank, calculating the weight, and arranging the cell words from large to small according to the weight:

characteristic item	Weight of
		Cam shaft	0.286
Position of	0.191
		Actuator device	0.095
Sensor with a sensor element	0.048
		Conversion	0.048
Response to	0.048
		Transition of	0.048
Circuit arrangement	0.048
		Cylinder	0.048
Regulating	0.048
		Exhaust of gases	0.048
Timing of	0.048

(3) Using discriminant analysis:

the word vector sequence of the fault description T is T ═ (w1, w2, w3,.., wn), and the sequence of each cellular word in the fault description T after the weight is calculated is S ═ (S1, S2, S3.., sn). The calculation formula of the word vector classification matching maximum probability is as follows: w ═ argmaxP (S | T).

Grading according to the syntactic rules that the word vector is classified and matched with the maximum probability from high to low:

classification	Rules
		First stage	Cam shaft
Second stage	Position of
		Three-stage	Timing of

(4) Decision tree analysis

And establishing a decision tree aiming at the result of cellular word grading according to the syntactic rule.

Classification model 1

As shown in FIG. 2, the camshaft is in one stage, the position, switching and regulation are in two stages, and the sensor, the actuator, the response, the timing and the exhaust are in three stages. The corresponding solutions are respectively as follows: the sensor corresponds to a camshaft position sensor, an engine wire harness and an engine control unit, the actuator corresponds to the engine wire harness, the engine control unit and a camshaft adjuster, and the timing corresponds to the engine wire harness, the engine control unit, the camshaft adjuster, a camshaft tensioner and a timing chain, and so on. The solution pointed by the thick arrow indicates that the probability of replacing the fitting is much higher than that pointed by other thin arrows, and the probability is preset in the decision tree model according to a large amount of maintenance data. Therefore, through carrying out word vector decomposition on the fault description of camshaft position (cylinder row 2) -timing over-advance and searching in a cell word bank, three cell words of the camshaft, the position and the timing are obtained, and through calculating the weight of the cell words and arranging the cell words according to a syntactic rule, the camshaft is located at the first level, the position is located at the second level and the timing is located at the third level. Through decision classification of the decision tree model, the most likely maintenance solution is to replace or maintain the camshaft tensioner.

Classification model 2

And (3) performing word vector decomposition on a fault description statement of 'camshaft position sensor ═ sensor unreliable signal' and searching in a cell word bank to obtain three cell words of a camshaft, a position and a sensor. And calculating the weight of the cell words according to the three cell words, and arranging the cell words according to a syntactic rule to obtain that the cam shaft is positioned at the first level, the position is positioned at the second level, and the sensor is positioned at the third level. And then, the measured value of the component is combined to judge the accessory needing to be replaced. First, whether the wire plug has poor contact or not and the needle head is bent are checked. If so, the wire plug needs to be replaced. If not, entering a next-level decision tree and checking whether the voltage supply of the camshaft sensor is between 4.5 and 5.5V. If so, the signal voltage of the engine control unit is checked to see if it is between 4.5-5.5V. If otherwise the engine wiring harness needs to be replaced. By analogy, a maintenance solution is given by combining decision classification of the decision tree model with component measurement values.

The embodiment can remotely and quickly identify the fault of the equipment and accurately provide an effective maintenance solution. The detection standards are unified, and errors caused by manual judgment are avoided. And a large amount of manpower and material resources are saved.

The above description is only for the purpose of creating a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can substitute or change the technical solution and the inventive concept of the present invention within the technical scope of the present invention.

Claims (8)

1. A fault code diagnosis vehicle work item and spare part retrieval method based on decision tree classification is characterized by comprising the following steps

Acquiring vehicle information data;

resolving the VIN code of the vehicle to obtain variables, wherein the variables comprise the engine displacement, the vehicle body type and the engine gearbox type which are obtained by resolving the VIN code;

thirdly, performing decision tree analysis on the spare part codes corresponding to the variables, finishing classification of the variable data to form spare part information, and establishing indexes to form a diagnosis knowledge base;

creating a language model, establishing a cell word bank, cutting words in the cell word bank to search cell words, arranging the cell words, and forming a diagnosis database of the fault codes corresponding to the work items by utilizing decision classification of a decision tree model;

and step five, associating the diagnosis database with the diagnosis knowledge base and establishing a main key.

2. The method for searching for the vehicle work items and the spare parts based on the fault code diagnosis of the decision tree classification as claimed in claim 1, further comprising a sixth step of identifying the fault code generated by the vehicle fault and performing classification search by analyzing variables obtained by the vehicle VIN code through keywords to obtain the work item and the spare part information.

3. The fault code diagnosis vehicle project and spare part retrieval method based on decision tree classification as claimed in claim 1 or 2, wherein in the fourth step, based on the historical records of the maintenance spare part table as data, the spare parts are classified through a decision tree model;

the spare part retrieval method comprises the following steps:

(1) distinguishing the same dimension of the maintenance spare part table according to the information entropy;

(2) the different dimensions of the maintenance spare part table are prioritized according to information gain;

(3) drawing a decision tree according to the priorities and the differentiation degrees divided in the steps (1) and (2);

(4) and inputting a regular VIN code, and outputting a spare part code under the vehicle type according to the VIN code and the obtained decision tree.

4. The method for fault code diagnosis vehicle work item and spare part search based on decision tree classification as claimed in claim 1 or 2, wherein the spare part code is associated with a price table of spare parts, and the spare part information comprises Chinese name, price and present use state of the spare part and applicable vehicle type information of the spare part.

5. The method for retrieving fault code diagnosis vehicle engineering and spare parts based on decision tree classification as claimed in claim 1, wherein the step four of creating a language model and establishing a cell lexicon comprises the steps of:

s1.1, collecting a professional fault description language;

s1.2, performing word vector decomposition on the professional fault description language.

6. The fault code diagnostic vehicle work item and spare part retrieval method based on decision tree classification as claimed in claim 5, wherein: the language model is created based on the assumption that the nth cell word occurs in relation to only the first n-1 cell words; the calculation formula of the occurrence weight of a fault description statement T is as follows:

P(T)＝P(w₁,w₂,w₃,…,w_n)

＝P(w₁)×P(w₂|w₁)×P(w₃|w₁,w₂)×…×P(w_n|w₁,w₂,…,w_n-1)

≈P(w₁)×P(w₂|w₁)×P(w₃|w₂)…P(w_n|w_n-1)；

wherein P (T) is the weight of the fault description statement T, P (w)_n|w₁,w₂,…,w_n-1) Is the weight of the nth cell word.

7. The method for retrieving the fault code diagnosis vehicle work item and spare part based on the decision tree classification as claimed in claim 1 or 2, wherein the step of word segmentation in the cell word bank in the fourth step for retrieving the cell words and the step of arranging the cell words comprises:

s2.1, performing word segmentation retrieval in the cell word bank aiming at a fault description statement T;

s2.2, if the cell words are searched, calculating the weight of the cell words;

s2.3, arranging the cell words according to the weight of the cell words.

8. The method for fault code diagnosis of vehicle work and spare parts based on decision tree classification as claimed in claim 7,

s2.2.1 calculating chi-square statistic for each of said cell words;

s2.2.2, taking the ith cell word with the highest chi-square statistic score, and calculating the occurrence frequency of the ith cell word in the jth fault description; the calculation method of the chi-square statistic score in the S2.2.1 step is as follows:

weight＝round((10×(1+lg(tf_ij))/(1+lg(l_j))),n)；

wherein tf is_ijThe number of times the ith cellular word appeared in the jth failure description, l_jN is an integer from 3 to 6 for the length of the jth fault description.

Images