CN117932979A - Automobile engine oil life assessment prediction method based on big data - Google Patents
Automobile engine oil life assessment prediction method based on big data Download PDFInfo
- Publication number
- CN117932979A CN117932979A CN202410331144.9A CN202410331144A CN117932979A CN 117932979 A CN117932979 A CN 117932979A CN 202410331144 A CN202410331144 A CN 202410331144A CN 117932979 A CN117932979 A CN 117932979A
- Authority
- CN
- China
- Prior art keywords
- engine oil
- oil data
- data
- time
- mileage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000010705 motor oil Substances 0.000 title claims abstract description 407
- 238000000034 method Methods 0.000 title claims abstract description 49
- 230000005856 abnormality Effects 0.000 claims abstract description 154
- 230000002159 abnormal effect Effects 0.000 claims abstract description 27
- 238000001514 detection method Methods 0.000 claims abstract description 4
- 239000003921 oil Substances 0.000 claims description 78
- 238000010606 normalization Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000002123 temporal effect Effects 0.000 claims description 5
- 238000007637 random forest analysis Methods 0.000 claims description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000007306 functionalization reaction Methods 0.000 description 1
- 230000001050 lubricating effect Effects 0.000 description 1
- 238000005461 lubrication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003647 oxidation Effects 0.000 description 1
- 238000007254 oxidation reaction Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Lubrication Details And Ventilation Of Internal Combustion Engines (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to an automobile engine oil life assessment and prediction method based on big data, which comprises the following steps: acquiring an engine oil data set, and acquiring the driving mileage abnormality degree and the service time abnormality degree of each engine oil data through abnormality detection; then obtaining the real abnormal degree of each engine oil data; obtaining the real driving mileage of each engine oil data according to the using time, the driving mileage and the real abnormality degree of each engine oil data; obtaining the accuracy of the real driving mileage of each engine oil data according to the real abnormality degree, the using time and the driving mileage of each engine oil data, and obtaining the final abnormality degree of each engine oil data; thereby obtaining normal engine oil data in the engine oil data set; and according to the normal engine oil data, predicting the service life of the engine oil of the automobile. According to the invention, when the service life of the automobile engine oil is predicted according to the plurality of automobile engine oil data, abnormal data in the automobile engine oil data are removed, so that the predicted result is more similar to real data.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an automobile engine oil life assessment and prediction method based on big data.
Background
Automotive engine oils play an important role in engine lubrication, cooling, sealing and cleaning. However, as the service life of the engine oil increases, the performance of the engine oil gradually decreases, so that the engine oil cannot effectively protect the engine, and even damages the engine. Therefore, accurate assessment and prediction of oil life becomes critical. In the prediction process, a large amount of vehicle and engine oil operation data needs to be collected and analyzed, and the accuracy of the collected data directly influences the final prediction accuracy, so that noise reduction treatment is generally required to be performed on the collected engine oil data.
The engine oil data for predicting the service life of the engine oil of the automobile mainly comprises the service time and the driving distance of the engine oil, and because the driving distance and the service time of the engine oil of the automobile are in large connection, the driving distance or the service time of the engine oil data can not be used independently to obtain the real abnormal degree of the engine oil data, so that when the noise is removed according to the driving distance or the service time of the engine oil data, the noise data with high abnormal degree can be contained in the data after the noise removal, and the service life of the engine oil of the automobile predicted according to the data after the noise removal can be inaccurate.
Disclosure of Invention
The invention provides an automobile engine oil life assessment prediction method based on big data, which aims to solve the existing problems.
The invention relates to a big data-based automobile engine oil life assessment and prediction method which adopts the following technical scheme:
one embodiment of the invention provides a big data-based automobile engine oil life assessment prediction method, which comprises the following steps:
acquiring an engine oil data set of engine oil with the same brand and the same model; each engine oil data in the engine oil data set corresponds to a driving mileage and a service time;
abnormality detection is carried out on the driving mileage and the service time of all the engine oil data respectively, and the driving mileage abnormality degree and the service time abnormality degree of each engine oil data are obtained;
Obtaining the real abnormality degree of each engine oil data according to the abnormality degree of the driving mileage of each engine oil data, the abnormality degree of the using time, the difference between the driving mileage of the engine oil data with the same using time and the difference between the using time of the engine oil data with the same driving mileage;
Obtaining the real driving mileage of each engine oil data according to the using time, the driving mileage and the real abnormality degree of each engine oil data;
obtaining the accuracy of the real driving mileage of each engine oil data according to the real abnormality degree, the using time and the driving mileage of each engine oil data;
Performing curve fitting according to the service time, the real driving mileage and the accuracy of the real driving mileage of each engine oil data to obtain the final abnormality degree of each engine oil data;
Obtaining normal engine oil data in the engine oil data set according to the final abnormal degree of each engine oil data;
And according to the normal engine oil data, predicting the service life of the engine oil of the automobile.
Further, the method for obtaining the real abnormality degree of each engine oil data according to the difference between the abnormality degree of the driving mileage of each engine oil data and the abnormality degree of the using time, the driving mileage of the engine oil data with the same using time and the difference between the using times of the engine oil data with the same driving mileage comprises the following specific steps:
Obtaining the mileage abnormality degree of each engine oil data according to the difference between the mileage abnormality degree of each engine oil data and the mileage of the engine oil data with the same use time;
obtaining the time abnormality degree of each engine oil data according to the difference between the use time abnormality degree of each engine oil data and the use time of the engine oil data of the same driving mileage;
And obtaining the real abnormality degree of each engine oil data according to the mileage abnormality degree of each engine oil data and the time abnormality degree of each engine oil data.
Further, the mileage abnormality degree of each engine oil data is obtained according to the difference between the mileage abnormality degree of each engine oil data and the mileage of the engine oil data with the same use time, and the specific formula is as follows:
Wherein: indicating the mileage abnormality degree of the ith engine oil data,/> Mileage representing ith engine oil data,/>Representing the mileage of the (v) th oil data identical to the use time of the (i) th oil data,/>Representation and/>The same time of use of the individual oil dataMileage abnormality degree of individual oil data/>Representing the number of oil data in the oil data set having the same use time as the i-th oil data,/>As a function of absolute value.
Further, the time abnormality degree of each engine oil data is obtained according to the difference between the time abnormality degree of each engine oil data and the time of use of the engine oil data of the same driving mileage, and the specific formula is as follows:
Wherein: Indicating the degree of temporal abnormality of the ith oil data,/> Represents the/>Time of use of individual oil data,/>Representation and/>The driving mileage of the individual engine oil data is the same/>Time of use of individual oil data,/>Representation and/>The driving mileage of the individual engine oil data is the same/>Degree of abnormality in use time of individual oil data,/>Representing the number of the engine oil data which has the same driving mileage as the ith engine oil data in the engine oil data set,/>As a function of absolute value.
Further, the actual abnormality degree of each engine oil data is obtained according to the mileage abnormality degree of each engine oil data and the time abnormality degree of each engine oil data, and the specific formula is as follows:
Wherein: Represents the/> True degree of abnormality of individual oil data,/>Represents the/>Mileage abnormality degree of individual oil data/>Represents the/>Degree of abnormality in use time of individual oil data,/>Indicating the mileage abnormality degree of the ith engine oil data,/>Indicating the degree of temporal abnormality of the ith oil data.
Further, the obtaining the real driving mileage of each engine oil data according to the using time, the driving mileage and the real abnormality degree of each engine oil data comprises the following specific steps:
optionally selecting one engine oil data in the engine oil data set, marking the engine oil data as target engine oil data, and marking the service time of the target engine oil data as the service time of the target engine oil data ；
According to the service time, the driving mileage and the real abnormality degree of the target engine oil data, the real driving mileage of the target engine oil data is obtained, and the corresponding specific calculation formula is as follows:
Wherein the method comprises the steps of Representing the real driving mileage of target engine oil data,/>Indicating that the y-th service time in the engine oil data set is/>Mileage of oil data of (1)/>Indicating that the y-th service time in the engine oil data set is/>True degree of abnormality of oil data of (1)/>The number of oil data with the use time x in the oil data set is represented.
Further, the accuracy of the actual driving mileage of each engine oil data is obtained according to the actual abnormality degree, the using time and the driving mileage of each engine oil data, and the method comprises the following specific steps:
Calculating the ratio of the average value of the real abnormal degrees of all the engine oil data with the use time x in the engine oil data set to the average value of the real abnormal degrees of all the engine oil data in the engine oil data set, calculating the normalized value of the ratio, calculating the difference value of the normalized value of the ratio, subtracting 1, calculating the normalized value of the inverse proportion of the standard deviation of the driving mileage of all the engine oil data with the use time x in the engine oil data set, and recording the product of the difference value and the normalized value of the inverse proportion as the accuracy of the real driving mileage of the target engine oil data.
Further, the curve fitting is performed according to the usage time, the actual driving mileage and the accuracy of the actual driving mileage of each engine oil data, so as to obtain the final abnormality degree of each engine oil data, which comprises the following specific steps:
in the engine oil data set, taking the accuracy of the real driving mileage of each engine oil data as a fitting weight factor, and performing curve fitting by using a weighted least square method according to the driving time and the real driving mileage of each engine oil data to obtain a fitting error corresponding to each engine oil data;
the fitting error of each engine oil data is recorded as the final degree of abnormality of each engine oil data.
Further, according to the final abnormality degree of each engine oil data, obtaining normal engine oil data in the engine oil data set, including the following specific steps:
Normalizing the final abnormal degree of all the engine oil data by using a minimum maximum normalization method to obtain a normalized value of the final abnormal degree of each engine oil data;
And recording the engine oil data with the normalized value of the final abnormality degree smaller than the preset abnormality threshold as normal engine oil data.
Further, the method for predicting the service life of the engine oil of the automobile according to the normal engine oil data comprises the following specific steps:
and inputting all normal engine oil data in the engine oil data set into the random forest model, outputting the predicted value of the engine oil life, and finishing the engine oil life prediction.
The technical scheme of the invention has the beneficial effects that: acquiring an engine oil data set of engine oil with the same brand and the same model; each engine oil data in the engine oil data set corresponds to a driving mileage and a service time; abnormality detection is carried out on the driving mileage and the service time of all the engine oil data respectively, and the driving mileage abnormality degree and the service time abnormality degree of each engine oil data are obtained; obtaining the real abnormality degree of each engine oil data according to the difference between the abnormality degree of the driving mileage of each engine oil data and the abnormality degree of the using time, the difference between the driving mileage of the engine oil data with the same using time and the difference between the using time of the engine oil data with the same using time, wherein when the real abnormality degree of each engine oil data is obtained, the real abnormality degree of each engine oil data is more accurate according to the driving mileage of the engine oil data with the same using time as each engine oil data in the engine oil data set and the using time of the engine oil data with the same driving mileage as each engine oil data in the engine oil data set; obtaining the real driving mileage of each engine oil data according to the using time, the driving mileage and the real abnormality degree of each engine oil data; obtaining the accuracy of the real driving mileage of each engine oil data according to the real abnormality degree, the using time and the driving mileage of each engine oil data; performing curve fitting according to the service time, the real driving mileage and the accuracy of the real driving mileage of each engine oil data, wherein the real driving mileage of each engine oil data is closer to the real situation, so that the curve fitting is more accurate, and the final abnormality degree of each engine oil data is closer to the real situation; according to the final abnormal degree of each engine oil data, obtaining the normal engine oil data in the engine oil data set, wherein the obtained normal engine oil data is more accurate because the final abnormal degree of each engine oil data is more accurate; because the obtained normal engine oil data is more accurate, the prediction result of the service life of the engine oil of the automobile is more accurate.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of the steps of the method for estimating and predicting the service life of the engine oil of the automobile based on big data.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to the specific implementation, structure, characteristics and effects of the method for estimating and predicting the service life of the engine oil of the automobile based on big data according to the invention by combining the attached drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the big data-based automobile engine oil life assessment and prediction method provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for estimating and predicting the service life of engine oil of an automobile based on big data according to an embodiment of the invention is shown, the method includes the following steps:
Step S001: acquiring an engine oil data set of engine oil with the same brand and the same model; each engine oil data in the engine oil data set corresponds to a driving mileage and a service time; and respectively detecting the driving mileage and the using time of all the engine oil data in an abnormal way, and acquiring the driving mileage abnormality degree and the using time abnormality degree of each engine oil data.
The service life of the engine oil of the automobile is mainly represented by the service time and the driving mileage of the engine oil, so that an untreated engine oil data set containing a plurality of engine oil data is obtained by collecting the service time and the driving mileage of the engine oil of the same brand and the same model at the existing automobile repair communication website, wherein each engine oil data contains the service time and the driving mileage, and the engine oil data is the engine oil data in the engine oil data set. And in the untreated oil data set if the time of use of one oil data occurs less than the time of use of the other oil data in the untreated oil data setOr the mileage of one oil datum occurs less frequently than the mileage of the other oil datum in the untreated oil datum setThen the oil data is removed from the untreated oil data set to obtain an oil data set. Preset/>, in this embodimentOther values may be set in other embodiments, and this embodiment is not limited thereto.
The service life of the engine oil of the automobile is mainly represented by the driving mileage and the service time, and the same service life is generally achieved for the same type of engine oil of the same type, namely the service time or the driving mileage of the engine oil of the same type is generally similar, and at the moment, the abnormal degree of the driving mileage and the abnormal degree of the service time of each engine oil data can be obtained according to the corresponding driving mileage and the deviation of the service time of each engine oil data.
The driving mileage abnormality degree and the using time abnormality degree of the engine oil data are obtained as follows:
respectively inputting the driving mileage and the using time corresponding to all the engine oil data in the LOF algorithm, and outputting the driving mileage abnormality degree of each engine oil data reflected by the driving mileage and the using time And degree of usage time abnormality/>. The LOF algorithm is also called local anomaly factor algorithm, and is a known technique, and the specific method is not described here. What needs to be described is: the LOF algorithm evaluates the abnormality degree by calculating the LOF value of each data point, namely the LOF value of the driving range of each engine oil data, namely the driving range abnormality degree of each engine oil data, and the LOF value of the using time of each engine oil data, namely the using time abnormality degree of each engine oil data.
Step S002: and obtaining the real abnormality degree of each engine oil data according to the abnormality degree of the driving mileage and the abnormality degree of the using time of each engine oil data, the difference between the driving mileage of the engine oil data with the same using time and the difference between the using times of the engine oil data with the same driving mileage.
Because the automobile engine oil can age and oxidize along with the time in the actual automobile engine oil use process, the lubricating performance of the automobile engine oil is reduced, so that the driving mileage of the automobile engine oil is reduced, and the performance of the automobile engine oil is accelerated to be reduced due to the increase of the driving mileage of the automobile engine oil, so that the service life of the automobile engine oil is reduced. Because the driving mileage of the engine oil of the automobile has a larger relation with the use time, the real abnormal degree of the engine oil data cannot be obtained through the driving mileage or the use time of the engine oil alone, and the real abnormal degree of the engine oil data needs to be obtained according to the driving mileage and the use time of the engine oil data at the same time.
In the actual use process of the automobile engine oil, the same use time reflects similar aging and oxidation characteristics, so that the driving mileage corresponding to the engine oil data with the same use time is generally similar; the same oil of the same driving range experiences the same loss, resulting in consistent reduction of the oil performance, and thus, the same driving range of the oil data generally has similarity in corresponding usage time.
Therefore, the actual abnormality degree of the engine oil data needs to be respectively related to the deviation of the corresponding use time of the engine oil data of the same driving mileage and the deviation of the corresponding driving mileage of the same use time data, and the actual abnormality degree of each engine oil data is obtained by combining the driving mileage abnormality degree and the use time abnormality degree of each engine oil data. Wherein the larger the deviation of the use time between the single engine oil data and other engine oil data of the same driving mileage is, the larger the real abnormality degree of the single engine oil data is; the greater the deviation of the driving mileage between the single oil data and other oil data whose oil usage time is the same, the greater the degree of true abnormality of the single oil data. Meanwhile, the greater the initial degree of abnormality of the other data is, the more unreliable the deviation between the other data and the single engine oil data is, namely the lower the degree of abnormality expression of the single engine oil data is; and the greater the initial degree of abnormality reflected by the driving mileage or the usage time of the single engine oil data, the lower the degree of performance of the driving mileage deviation corresponding to the same usage time or the usage time deviation corresponding to the same driving mileage on the single engine oil data abnormality.
Accordingly, the true degree of abnormality of the individual oil data is determined. First, theThe true degree of abnormality of the individual oil data is/>The calculation formula is as follows:
Wherein: Represents the/> True degree of abnormality of individual oil data,/>、/>Respectively represent the/>The driving mileage abnormality degree and the usage time abnormality degree of the individual engine oil data. /(I)Indicating the mileage abnormality degree of the ith engine oil data,/>Mileage representing ith engine oil data,/>Representing the mileage of the (v) th oil data identical to the use time of the (i) th oil data,/>Representation and/>The same time of use of the individual oil dataMileage abnormality degree of individual oil data/>The number of the oil data which has the same service time as the ith oil data in the oil data set is represented; /(I)Represents the degree of temporal abnormality of the ith engine oil data, wherein/>Represents the/>Time of use of individual oil data,/>Representation and the firstThe driving mileage of the individual engine oil data is the same/>Time of use of individual oil data,/>Representation and/>The driving mileage of the individual engine oil data is the same/>Degree of abnormality in use time of individual oil data,/>Representing the number of the engine oil data which has the same driving mileage as the ith engine oil data in the engine oil data set,/>As a function of absolute value.
When the service time or the driving distance of part of the engine oil data in the engine oil data set is different from that of other engine oil data in the engine oil data set, the true abnormality degree of the engine oil data is not calculated, and the engine oil data does not participate in final life prediction.
The first step is obtained by the above operationTrue degree of abnormality of individual oil data, in terms of/>The individual engine oil data are taken as examples to obtain the actual abnormality degree of all the engine oil data.
Step S003: and obtaining the real driving mileage of each engine oil data according to the using time, the driving mileage and the real abnormality degree of each engine oil data.
When the service life of the engine oil of the automobile is predicted, the service life is mainly obtained under different service times, and the corresponding service life has a certain trend along with the change of the service time, namely the service life of the engine oil of the automobile has a distribution relation about the service time, and the final abnormality degree of all engine oil data is represented by the deviation of the distribution of all engine oil data and the distribution relation about the service time of the service life.
Note that, there may be a plurality of oil data corresponding to the same usage time among the oil data, that is, the correspondence relationship between the driving range and the usage time is not clear, so the distribution relationship between the driving range and the usage time cannot be directly determined. Therefore, it is first necessary to determine the correspondence between the driving range and the use time.
Specifically, first, a true driving range corresponding to each engine oil data in the current engine oil data is determined. Optionally selecting one engine oil data in the engine oil data set, marking the engine oil data as target engine oil data, and marking the service time of the target engine oil data as service timeThe greater the actual abnormality degree of the plurality of engine oil data with the use time x of the engine oil data set, the lower the expression degree of the corresponding driving mileage to the current use time corresponding to the actual driving mileage.
Accordingly, the real driving mileage of the target engine oil data is determined to beThe calculation formula is as follows:
Wherein the method comprises the steps of Representing the real driving mileage of target engine oil data,/>Indicating that the y-th service time in the engine oil data set is/>Mileage of oil data of (1)/>Indicating that the y-th service time in the engine oil data set is/>True degree of abnormality of oil data of (1)/>The number of oil data with the use time x in the oil data set is represented.
And carrying out the operation on each engine oil data in the engine oil data set to obtain the actual driving mileage of each engine oil data in the engine oil data set.
Step S004: and obtaining the accuracy of the real driving mileage of each engine oil data according to the real abnormality degree, the using time and the driving mileage of each engine oil data.
Further, a distribution relation of the driving mileage with respect to the using time is determined according to the using time of the engine oil data and the obtained real driving mileage corresponding to each using time. The real driving mileage corresponding to each using time is generally fitted directly to obtain the distribution relation of the driving mileage about the using time, but the accuracy of the obtained real driving mileage directly influences the accuracy of the fitting to obtain the distribution relation of the driving mileage about the using time.
The accuracy of each obtained real driving mileage directly depends on the difference between all driving mileage corresponding to each using time and the real abnormality degree of engine oil data corresponding to each using time, and the greater the difference between the driving mileage corresponding to the specific using time is, the lower the accuracy of the real driving mileage which is jointly reflected is; meanwhile, the greater the abnormality degree of the engine oil data corresponding to the driving mileage corresponding to the using time is relative to the abnormality degree of the whole engine oil data, the lower the accuracy of the real driving mileage which is jointly reflected by the engine oil data is.
Accordingly, the accuracy of determining the true driving range of the target engine oil data isThe calculation formula is as follows:
Wherein the method comprises the steps of Accuracy of true mileage representing target engine oil data,/>Mean value representing true abnormality degree of all engine oil data with use time x in engine oil data set,/>A mean value representing the true degree of abnormality of all the oil data in the oil data set; /(I)Representing the driving mileage standard deviation of all the engine oil data with the using time x in the engine oil data set; normalizing the data value to be in the [0,1] interval as a linear normalization function; /(I) The present embodiment uses/>, as an exponential function based on natural constantsTo present inverse proportion relation and normalization processing, and the implementer can set inverse proportion function and normalization function according to actual situation.
Step S005: and performing curve fitting according to the service time, the real driving mileage and the accuracy of the real driving mileage of each engine oil data to obtain the final abnormality degree of each engine oil data.
Further, based on the actual driving mileage corresponding to each engine oil data obtained in the steps and the accuracy of the actual driving mileage of each engine oil data, the distribution relation of the driving mileage with respect to the use time is determined. Specifically, the accuracy of the real driving mileage corresponding to each using time is taken as a fitting weight factor for all the real driving mileageFitting was performed. Wherein the fitting method directly adopts the existing weighted least square method to output a fitting expression/>。
The method comprises the following steps: and in the engine oil data set, taking the accuracy of the real driving mileage of each engine oil data as a fitting weight factor, and performing curve fitting by using a weighted least square method according to the driving time and the real driving mileage of each engine oil data to obtain a fitting curve, thereby further obtaining a fitting error corresponding to each engine oil data. What needs to be described is: the horizontal axis of the fitted curve fitted by the curve is the running time of the engine oil data, and the vertical axis is the actual running mileage of the engine oil data. The fitting error of each engine oil data is recorded as the final degree of abnormality of each engine oil data. The fitting error is the difference between the fitted real driving mileage and the real driving mileage.
The least square method is a known technique, and the embodiment does not describe the least square method, wherein the fitting error is an output feature word of the least square method, and the method for obtaining the fitting error is a known technique, and the embodiment does not describe the fitting error.
Step S006: obtaining normal engine oil data in the engine oil data set according to the final abnormal degree of each engine oil data; and according to the normal engine oil data, predicting the service life of the engine oil of the automobile.
Through the process, the final abnormality degree of each engine oil data in the engine oil data set is obtained, and the prediction of the engine oil data is completed according to the final abnormality degree of each engine oil data. The specific acquisition method is as follows:
Normalizing the final abnormal degree of all the engine oil data by using a minimum maximum normalization method to obtain a normalized value of the final abnormal degree of each engine oil data;
Then preset the abnormal threshold value The engine oil data with the final abnormal degree normalized value smaller than the preset abnormal threshold value is recorded as normal engine oil data;
And outputting all the normal engine oil data to a random forest model, and outputting the predicted value of the engine oil life, so as to realize accurate prediction of the engine oil life of the automobile.
In the present embodiment, an abnormality threshold is presetOther values may be set in other embodiments, and the present embodiment is not limited. It should be noted that the maximum and minimum normalization method and the random forest model are known techniques, and this embodiment is not described.
The present invention has been completed.
In summary, the above embodiments are only preferred embodiments of the present invention, and are not intended to limit the present invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. The automobile engine oil life assessment and prediction method based on big data is characterized by comprising the following steps of:
acquiring an engine oil data set of engine oil with the same brand and the same model; each engine oil data in the engine oil data set corresponds to a driving mileage and a service time;
abnormality detection is carried out on the driving mileage and the service time of all the engine oil data respectively, and the driving mileage abnormality degree and the service time abnormality degree of each engine oil data are obtained;
Obtaining the real abnormality degree of each engine oil data according to the abnormality degree of the driving mileage of each engine oil data, the abnormality degree of the using time, the difference between the driving mileage of the engine oil data with the same using time and the difference between the using time of the engine oil data with the same driving mileage;
Obtaining the real driving mileage of each engine oil data according to the using time, the driving mileage and the real abnormality degree of each engine oil data;
obtaining the accuracy of the real driving mileage of each engine oil data according to the real abnormality degree, the using time and the driving mileage of each engine oil data;
Performing curve fitting according to the service time, the real driving mileage and the accuracy of the real driving mileage of each engine oil data to obtain the final abnormality degree of each engine oil data;
Obtaining normal engine oil data in the engine oil data set according to the final abnormal degree of each engine oil data;
And according to the normal engine oil data, predicting the service life of the engine oil of the automobile.
2. The method for estimating and predicting the life of engine oil of an automobile based on big data as set forth in claim 1, wherein the obtaining the true abnormality degree of each engine oil data based on the difference between the abnormality degree of the driving mileage and the abnormality degree of the use time of each engine oil data, the difference between the driving mileage of the engine oil data of the same use time and the use time of the engine oil data of the same use time includes the steps of:
Obtaining the mileage abnormality degree of each engine oil data according to the difference between the mileage abnormality degree of each engine oil data and the mileage of the engine oil data with the same use time;
obtaining the time abnormality degree of each engine oil data according to the difference between the use time abnormality degree of each engine oil data and the use time of the engine oil data of the same driving mileage;
And obtaining the real abnormality degree of each engine oil data according to the mileage abnormality degree of each engine oil data and the time abnormality degree of each engine oil data.
3. The method for estimating and predicting the life of engine oil of an automobile based on big data as claimed in claim 2, wherein the obtaining the mileage abnormality degree of each engine oil data based on the difference between the mileage abnormality degree of each engine oil data and the mileage of the engine oil data of the same usage time comprises the following specific formulas:
Wherein: indicating the mileage abnormality degree of the ith engine oil data,/> Indicating the driving range of the ith engine oil data,Representing the mileage of the (v) th oil data identical to the use time of the (i) th oil data,/>Representation and/>The same time of use of the individual oil dataMileage abnormality degree of individual oil data/>Representing the number of oil data in the oil data set having the same use time as the i-th oil data,/>As a function of absolute value.
4. The method for estimating and predicting the service life of engine oil of an automobile based on big data as set forth in claim 2, wherein the obtaining the time abnormality degree of each engine oil data based on the difference between the time abnormality degree of each engine oil data and the time of use of the engine oil data of the same mileage comprises the following specific formulas:
Wherein: Indicating the degree of temporal abnormality of the ith oil data,/> Represents the/>The use time of the individual oil data,Representation and/>The driving mileage of the individual engine oil data is the same/>Time of use of individual oil data,/>Representation and/>The driving mileage of the individual engine oil data is the same/>Degree of abnormality in use time of individual oil data,/>Representing the number of the engine oil data which has the same driving mileage as the ith engine oil data in the engine oil data set,/>As a function of absolute value.
5. The method for estimating and predicting the service life of the engine oil of the automobile based on big data as claimed in claim 2, wherein the actual abnormality degree of each engine oil data is obtained according to the mileage abnormality degree of each engine oil data and the time abnormality degree of each engine oil data, comprising the following specific formulas:
Wherein: Represents the/> True degree of abnormality of individual oil data,/>Represents the/>Mileage abnormality degree of individual oil data/>Represents the/>Degree of abnormality in use time of individual oil data,/>Indicating the mileage abnormality degree of the ith engine oil data,Indicating the degree of temporal abnormality of the ith oil data.
6. The method for estimating and predicting the service life of the engine oil of the vehicle based on big data as set forth in claim 1, wherein the obtaining the actual driving distance of each engine oil data according to the usage time, the driving distance and the actual abnormality degree of each engine oil data comprises the following specific steps:
optionally selecting one engine oil data in the engine oil data set, marking the engine oil data as target engine oil data, and marking the service time of the target engine oil data as the service time of the target engine oil data ；
According to the service time, the driving mileage and the real abnormality degree of the target engine oil data, the real driving mileage of the target engine oil data is obtained, and the corresponding specific calculation formula is as follows:
Wherein the method comprises the steps of Representing the real driving mileage of target engine oil data,/>Indicating the y-th service time in the engine oil data set asMileage of oil data of (1)/>Indicating that the y-th service time in the engine oil data set is/>True degree of abnormality of oil data of (1)/>The number of oil data with the use time x in the oil data set is represented.
7. The method for estimating and predicting the service life of the engine oil of the vehicle based on big data as set forth in claim 6, wherein the obtaining the accuracy of the actual driving range of each engine oil data according to the actual abnormality degree, the using time and the driving range of each engine oil data comprises the following specific steps:
Calculating the ratio of the average value of the real abnormal degrees of all the engine oil data with the use time x in the engine oil data set to the average value of the real abnormal degrees of all the engine oil data in the engine oil data set, calculating the normalized value of the ratio, calculating the difference value of the normalized value of the ratio, subtracting 1, calculating the normalized value of the inverse proportion of the standard deviation of the driving mileage of all the engine oil data with the use time x in the engine oil data set, and recording the product of the difference value and the normalized value of the inverse proportion as the accuracy of the real driving mileage of the target engine oil data.
8. The method for estimating and predicting the service life of the engine oil of the vehicle based on big data according to claim 1, wherein the curve fitting is performed according to the service time, the actual driving range and the accuracy of the actual driving range of each engine oil data, so as to obtain the final abnormality degree of each engine oil data, comprising the following specific steps:
in the engine oil data set, taking the accuracy of the real driving mileage of each engine oil data as a fitting weight factor, and performing curve fitting by using a weighted least square method according to the driving time and the real driving mileage of each engine oil data to obtain a fitting error corresponding to each engine oil data;
the fitting error of each engine oil data is recorded as the final degree of abnormality of each engine oil data.
9. The method for estimating and predicting the service life of engine oil of an automobile based on big data as claimed in claim 1, wherein the step of obtaining the normal engine oil data in the engine oil data set according to the final degree of abnormality of each engine oil data comprises the following specific steps:
Normalizing the final abnormal degree of all the engine oil data by using a minimum maximum normalization method to obtain a normalized value of the final abnormal degree of each engine oil data;
And recording the engine oil data with the normalized value of the final abnormality degree smaller than the preset abnormality threshold as normal engine oil data.
10. The method for estimating and predicting the service life of the engine oil of the automobile based on big data as set forth in claim 1, wherein the step of predicting the service life of the engine oil of the automobile based on the normal engine oil data comprises the following specific steps:
and inputting all normal engine oil data in the engine oil data set into the random forest model, outputting the predicted value of the engine oil life, and finishing the engine oil life prediction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410331144.9A CN117932979A (en) | 2024-03-22 | 2024-03-22 | Automobile engine oil life assessment prediction method based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410331144.9A CN117932979A (en) | 2024-03-22 | 2024-03-22 | Automobile engine oil life assessment prediction method based on big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117932979A true CN117932979A (en) | 2024-04-26 |
Family
ID=90751126
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410331144.9A Pending CN117932979A (en) | 2024-03-22 | 2024-03-22 | Automobile engine oil life assessment prediction method based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117932979A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR19990018939U (en) * | 1997-11-14 | 1999-06-05 | 홍종만 | Vehicle consumable exchange time display control device |
| KR100639936B1 (en) * | 2005-08-30 | 2006-11-01 | 씨멘스 오토모티브 주식회사 | Method for detecting change time of engine oil of car |
| WO2020133033A1 (en) * | 2018-12-27 | 2020-07-02 | 潍柴动力股份有限公司 | Method and device for determining service life of lubricating oil |
| CN113902198A (en) * | 2021-10-13 | 2022-01-07 | 彩虹无线（北京）新技术有限公司 | Engine oil life prediction method and device, electronic equipment and storage medium |
| CN115909539A (en) * | 2022-10-24 | 2023-04-04 | 东风商用车有限公司 | Method and device for calculating performance loss of engine oil |
| CN116517659A (en) * | 2023-03-22 | 2023-08-01 | 重庆长安汽车股份有限公司 | Engine oil monitoring method and system and vehicle |
| WO2023238690A1 (en) * | 2022-06-10 | 2023-12-14 | 株式会社日立製作所 | System for predicting physical properties of compressor oil |
-
2024
- 2024-03-22 CN CN202410331144.9A patent/CN117932979A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR19990018939U (en) * | 1997-11-14 | 1999-06-05 | 홍종만 | Vehicle consumable exchange time display control device |
| KR100639936B1 (en) * | 2005-08-30 | 2006-11-01 | 씨멘스 오토모티브 주식회사 | Method for detecting change time of engine oil of car |
| WO2020133033A1 (en) * | 2018-12-27 | 2020-07-02 | 潍柴动力股份有限公司 | Method and device for determining service life of lubricating oil |
| CN113902198A (en) * | 2021-10-13 | 2022-01-07 | 彩虹无线（北京）新技术有限公司 | Engine oil life prediction method and device, electronic equipment and storage medium |
| WO2023238690A1 (en) * | 2022-06-10 | 2023-12-14 | 株式会社日立製作所 | System for predicting physical properties of compressor oil |
| CN115909539A (en) * | 2022-10-24 | 2023-04-04 | 东风商用车有限公司 | Method and device for calculating performance loss of engine oil |
| CN116517659A (en) * | 2023-03-22 | 2023-08-01 | 重庆长安汽车股份有限公司 | Engine oil monitoring method and system and vehicle |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111562108A (en) | Rolling bearing intelligent fault diagnosis method based on CNN and FCMC | |
| CN111222095B (en) | Rough difference judging method, device and system in dam deformation monitoring | |
| CN112414694B (en) | Equipment multistage abnormal state identification method and device based on multivariate state estimation technology | |
| CN112183947B (en) | Rail transit vehicle comprehensive evaluation method and device, electronic equipment and storage medium | |
| CN113836756B (en) | Intelligent monitoring method and system for annealing process of three-dimensional wound core transformer | |
| CN110457731B (en) | Method applied to vibration trend analysis of helicopter moving parts | |
| CN115422687A (en) | Service life prediction method of rolling bearing | |
| CN117591986B (en) | Real-time automobile data processing method based on artificial intelligence | |
| CN113743541B (en) | Method for predicting residual life of marine power system bearing based on degradation mode | |
| CN110084301B (en) | Hidden Markov model-based multi-working-condition process working condition identification method | |
| CN113071509B (en) | Fault detection method for wheel angle sensor of intelligent vehicle | |
| CN117932979A (en) | Automobile engine oil life assessment prediction method based on big data | |
| CN112380782A (en) | Rotating equipment fault prediction method based on mixed indexes and neural network | |
| CN116579125A (en) | Bearing remaining life prediction method, device, equipment and readable storage medium | |
| CN115508706A (en) | Lithium battery residual life prediction method considering capacity regeneration phenomenon | |
| CN112214934B (en) | Suspension system life prediction method based on multiple sensors and related device | |
| CN113283157A (en) | System, method, terminal and medium for predicting life cycle of intelligent stamping press part | |
| CN114088416A (en) | Fault diagnosis method and device for silencer of hydrodynamic retarder and computer equipment | |
| Morgan et al. | Predicting future states with $ n $-dimensional Markov chains for fault diagnosis | |
| CN118090078B (en) | Leakage online monitoring method for closed circulation water cooling system | |
| CN117874685B (en) | Pressure data abnormality early warning method for direct-current variable-frequency electric half ball valve | |
| CN118090078A (en) | Leakage online monitoring method for closed circulation water cooling system | |
| CN111832731B (en) | Multi-index monitoring method for representing uncertain state of oil and diagnosing faults | |
| CN114676384B (en) | Performance state percentile value estimation method based on particle filtering | |
| CN114063456B (en) | Fault prediction and early warning method using autoregressive model and Kalman filtering algorithm |
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 |