JP2023094269A - Data processing device, data processing method and calculation model generation system - Google Patents

Abstract

To provide a data processing device, a data processing method and a calculation model generation system that select measurement data of an abrasion loss of a tire and can efficiently generate a database.SOLUTION: A data processing device 20 comprises: a data acquisition unit 22a; a threshold setting unit 22b; a continuity determination unit 22c; and a database generation unit 22d. The data acquisition unit 22a acquires data associated with an abrasion loss of a tire in a predetermined period. The threshold setting unit 22b sets a lower limit side threshold value and an upper limit side threshold value for the data acquired by the data acquisition unit 22a. The continuity determination unit 22c determines whether or not change in the predetermined period is continuous with data outside a range between the lower limit side threshold value and the upper limit side threshold value that is set by the threshold setting unit 22b. The database generation unit 22d generates an abrasion loss database 23a excluding data with a negative determination result made by the continuity determination unit 22c.SELECTED DRAWING: Figure 4

Description

The present invention relates to a data processing apparatus and data processing method for processing data on the amount of wear of tires mounted on a vehicle, and an arithmetic model generation system for generating an arithmetic model for estimating the amount of wear of tires.

In general, tire wear progresses according to running conditions, running distance, and the like. In recent years, devices such as devices that display the measured pressure and temperature by attaching a sensor for measuring the pressure and temperature of the tire to the tire have been commercialized.

Patent Literature 1 describes a conventional abnormal data detection method for detecting abnormal data from tire usage history data. This abnormal data detection method collects data related to tire usage history, and then selects a plurality of correlated items from the collected tire usage history items, such as mileage and wear amount. Extract a dataset that is a combination of After that, abnormal data is detected from the extracted data set by machine learning algorithms such as Isolation Forest. When detecting abnormal data, whether or not the data set is abnormal is determined based on a discriminant model in which a plurality of data sets previously determined to be normal are constructed as training data.

JP 2019-074927 A

In the abnormal data detection method described in Patent Document 1, a discriminant model must be learned in advance based on training data, and the discriminant model must be constructed according to the input variables of the discriminant model. There is a problem that the processing becomes complicated and the processing load increases. In addition, the measurement value of the amount of wear, which is the teacher data used for learning the computational model for estimating the amount of tire wear, may have measurement errors depending on the skill level of the operator. There was a probability that a stone or the like would be caught in the gap, causing a measurement error.

SUMMARY OF THE INVENTION The present invention has been made in view of such circumstances, and its object is to provide a data processing apparatus and a data processing method capable of selecting tire wear amount measurement data and efficiently generating a database. and to provide an arithmetic model generation system.

A data processing apparatus according to one aspect of the present invention includes a data acquisition unit that acquires data on the amount of wear of a tire mounted on a vehicle in a predetermined period; a threshold setting unit for setting a threshold; and continuity for determining whether or not the data outside the range of the lower threshold and the upper threshold set by the threshold setting unit continuously change during the predetermined period. A determination unit and a database generation unit that generates a wear amount database by excluding data for which a result of determination by the continuity determination unit is negative.

Another aspect of the invention is a data processing method. The data processing method includes a data acquisition step of acquiring data relating to the amount of wear of a tire mounted on a vehicle for a predetermined period of time, and threshold setting of setting a lower threshold value and an upper threshold value for the data acquired by the data acquisition step. a continuity determination step of determining whether or not the change in the data outside the range of the lower threshold and the upper threshold set in the threshold setting step is continuous in the predetermined period; and a database generating step of generating a wear amount database by excluding data for which the determination result of the property determining step was negative.

Another aspect of the invention is a computational model generation system. The computational model generation system has the above-described data processing device and a learning-type computational model that calculates tire wear based on the input information. A wear amount calculation unit that calculates the wear amount of the tire by input, and compares the wear amount data included in the wear amount database generated by the data processing device with the wear amount calculated by the wear amount calculation unit. and a learning processing unit for learning the arithmetic model.

ADVANTAGE OF THE INVENTION According to this invention, the measurement data of the wear amount of a tire can be selected and a database can be efficiently produced|generated.

1 is a block diagram showing the functional configuration of a computational model generation system according to an embodiment; FIG. It is a block diagram which shows the functional structure of an in-vehicle measuring device. FIG. 4 is a schematic diagram for explaining wear amount estimation and learning of a computing model; It is a block diagram which shows the functional structure of a data processor. 4 is a flow chart showing a procedure of database generation processing by the data processing device; 4 is a chart showing the results of wear amount estimation of an arithmetic model learned from measured wear amount data.

BEST MODE FOR CARRYING OUT THE INVENTION The present invention will be described below based on preferred embodiments with reference to FIGS. 1 to 6. FIG. The same or equivalent constituent elements and members shown in each drawing are denoted by the same reference numerals, and duplication of description will be omitted as appropriate. In addition, the dimensions of the members in each drawing are appropriately enlarged or reduced for easy understanding. Also, in each drawing, some of the members that are not important for explaining the embodiments are omitted.

(embodiment)
FIG. 1 is a block diagram showing the functional configuration of a computational model generation system 100 according to an embodiment. The computational model generation system 100 includes a tire wear amount measuring device 60, an in-vehicle measurement device 70, a weather information server device 80, a computational model generation device 10, and a data processing device 20. Generate.

The arithmetic model generation device 10 receives vehicle measurement information such as vehicle speed and position information, and tire measurement information measured by the tires 7 from an on-vehicle measurement device 70 mounted on the vehicle via a communication network 9 such as the Internet. get. The computational model generation device 10 acquires weather information from the weather information server device 80 . Further, the arithmetic model generation device 10 acquires each data in the wear amount database of the tire 7 generated by the data processing device 20 based on the wear amount data of the tire 7 measured by the tire wear amount measuring device 60 .

The data processing device 20 removes the data of the wear amount of the tire 7 measured by the tire wear amount measuring device 60, the data whose numerical value is out of the distribution range of the other wear amount data due to the measurement error, etc. Generate a quantity database. The calculation model generation device 10 uses each data in the wear amount database generated by the data processing device 20 as teacher data to make the calculation model 13a learn, thereby improving the accuracy of wear amount estimation by the calculation model 13a.

The tire wear amount measuring device 60 directly measures the depth of grooves provided in the tread of the tire 7 over a predetermined period (several months to several years) to acquire the wear amount of the tire 7 . The tire wear amount measuring device 60 transmits the measured wear amount data of the tire 7 to the data processing device 20 via the communication network 9 . A tire operator may measure the depth of each groove using a measuring instrument, a camera, or visually, and the tire wear amount measuring device 60 may store measurement data input by the operator. Further, the tire wear amount measuring device 60 may be a dedicated device that measures the depth of the groove by a mechanical or optical method and stores the wear amount.

Specifically, for example, when the tire has four grooves, the tire wear amount measuring device 60 measures at four points in the width direction, and furthermore, measures the same groove in the circumferential direction, for example, at intervals of 120°, at three points. Measure in place. Thereby, the uneven wear data in the width direction or the circumferential direction of the tire is also stored in the tire wear amount measuring device 60 . Note that the tire wear amount measuring device 60 may indirectly measure the depth of the groove by calculation from information on the running distance and the number of rotations and speed of the tire, since the diameter of the tire changes due to wear of the tire. In addition, the direct measurement of the groove depth may be combined with the prediction by calculation from the running distance and the number of revolutions/speed of the tire.

FIG. 2 is a block diagram showing the functional configuration of the in-vehicle measuring device 70. As shown in FIG. The in-vehicle measurement device 70 includes a vehicle measurement section 71 , a tire measurement section 72 , an information acquisition section 73 and a communication section 74 . Each part in the in-vehicle measurement device 70 can be realized by electronic elements such as a CPU of a computer, mechanical parts, etc. in terms of hardware, and realized by computer programs etc. in terms of software. It depicts the functional blocks implemented by Therefore, those skilled in the art will understand that these functional blocks can be realized in various forms by combining hardware and software.

The vehicle measurement unit 71 has a speedometer 71a, a GPS receiver 71b, and an acceleration sensor 71c mounted on the vehicle. The speedometer 71a measures the speed of the vehicle. The GPS receiver 71b measures the current position information (latitude, longitude and altitude) of the vehicle. The acceleration sensor 71c measures the acceleration of the vehicle in three axial directions.

The tire measuring section 72 has a temperature sensor 72a and a pressure sensor 72b. The temperature sensor 72a and the pressure sensor 72b are arranged in the air valve of the tire 7 mounted on the vehicle, or are firmly wrapped around the wheel with a belt or the like and fixed, and measure the temperature and air pressure of the tire 7. The temperature sensor 72a may be arranged on the inner liner of the tire 7 or the like.

The information acquisition unit 73 acquires vehicle measurement information (speed, position information, acceleration, etc.) measured by the vehicle measurement unit 71 and tire measurement information (tire temperature, air pressure, etc.) measured by the tire measurement unit 72 . The information acquisition unit 73 associates measured time information or acquired time information with each measurement data included in the vehicle measurement information and the tire measurement information. The information acquisition unit 73 transmits the vehicle measurement information and the tire measurement information together with the time information associated with each measurement data from the communication unit 74 to the computational model generation device 10 .

If the vehicle is equipped with a device such as a digital tachometer, the information acquisition unit 73 may acquire vehicle speed, acceleration, position information, and the like collected by the device. The communication unit 74 communicates with the communication network 9 by wireless communication such as WiFi (registered trademark), and transmits the vehicle measurement information, the tire measurement information, and the time information acquired by the information acquisition unit 73 to the computation model via the communication network 9. Send to the generation device 10 .

Returning to FIG. 1, the weather information server device 80 provides weather information for various places. The weather information provided by the weather information server device 80 is information including rainfall amounts, snowfall amounts, snowfall amounts, temperatures, hours of sunshine, and the like in various places. The computational model generation device 10 acquires weather information for the location where the vehicle is traveling from the weather information server device 80 .

The arithmetic model generation device 10 includes a communication unit 11 , a vehicle information acquisition unit 12 , a wear amount calculation unit 13 , a storage unit 14 and a learning processing unit 15 . Each part in the computational model generation device 10 can be realized by electronic elements such as a CPU of a computer, mechanical parts, etc. in terms of hardware, and realized by computer programs etc. in terms of software. It depicts the functional blocks realized by cooperation. Therefore, those skilled in the art will understand that these functional blocks can be realized in various forms by combining hardware and software.

The communication unit 11 is connected to the communication network 9 by wireless or wired communication, and communicates with the communication unit 74 of the in-vehicle measurement device 70 . The communication unit 11 also communicates with the weather information server device 80 via the communication network 9 .

The vehicle information acquisition unit 12 acquires vehicle measurement information (speed, position information, acceleration, etc.) and tire measurement information (tire temperature, air pressure, etc.) transmitted from an in-vehicle measurement device 70 mounted on the vehicle. The vehicle information acquisition unit 12 calculates and acquires the travel distance of the vehicle based on the vehicle measurement information.

The vehicle information acquisition unit 12 can calculate and acquire the travel distance based on the position information of the vehicle measurement information. Further, the travel distance of the vehicle may be calculated based on the speed data in the vehicle measurement information and the time data associated with the data. That is, by multiplying the speed data arranged in chronological order by the time difference up to the next point in time, the traveling distance of the vehicle can be calculated.

The vehicle information acquisition unit 12 does not need to calculate the mileage by itself if the information on the mileage of the vehicle is provided from the vehicle or an external device for vehicle management or the like. may be obtained.

The vehicle information acquisition unit 12 outputs the acquired travel distance to the wear amount calculation unit 13 . The vehicle information acquisition unit 12 outputs the acquired tire measurement information (tire temperature, air pressure, etc.) to the wear amount calculation unit 13 . When the wear amount calculation unit 13 estimates the amount of tire wear based on an arithmetic model that uses vehicle acceleration as an input factor, the vehicle information acquisition unit 12 outputs the acceleration data in the vehicle measurement information to the wear amount calculation unit 13. .

The vehicle information acquisition unit 12 also acquires from the storage unit 14 data used for estimating the wear amount of the tire 7 among the vehicle specification data 14 a and the tire specification data 14 b and outputs the data to the wear amount calculation unit 13 . The storage unit 14 is a storage device configured by, for example, an SSD (Solid State Drive), hard disk, CD-ROM, DVD, etc., and stores data provided in advance regarding specifications of various vehicles and tires 7. .

The vehicle specification data 14a includes information on vehicle performance such as manufacturer, vehicle name, vehicle model, vehicle weight, drive train, overall length, vehicle width, vehicle height, and maximum payload. The tire specification data 14b includes, for example, tire identification information, manufacturer, product name, tire size, tire width, aspect ratio, wear resistance performance, tire strength, static rigidity, dynamic rigidity, tire outer diameter, load index, Information about the performance of the tire 7 is included, such as date of manufacture. For example, an RFID is embedded in the tire 7, and when the wear amount is measured by the tire wear amount measuring device 60, the RFID is read and data that associates the tire identification information with the axial position in the vehicle is stored in the storage unit 14. can be The axial position of the tire 7 in the vehicle is changed by tire rotation, but even if the rotation history is not recorded, the correspondence between the tire identification information stored in the storage unit 14 and the axial position in the vehicle is maintained. By referring to the relationship, it is possible to determine the axial position where the tire 7 is mounted and to know when the axial position was changed. In addition, by storing in the storage unit 14 that the operation of changing the direction of the wheel in the tire 7 has been performed, it is possible to track whether each groove of the tire 7 is on the front or back side with respect to the wheel. can continue.

The wear amount calculator 13 has an arithmetic model 13 a and estimates the wear amount of the tire 7 . The arithmetic model 13a is a learning model that calculates the wear amount of the tire 7 based on the input information. FIG. 3 is a schematic diagram for explaining wear amount estimation and learning of the computation model 13a. The input data to the computation model 13a is roughly classified into each system of vehicle measurement information, tire measurement information, and other information.

Input data related to vehicle metrology information includes vehicle acceleration and distance traveled. The traveled distance is obtained by the vehicle information obtaining unit 12 as described above. Input data related to tire measurement information includes tire 7 temperature and air pressure. It should be noted that the acceleration of the vehicle is used as input data to the calculation model as appropriate.

The input data of the other information includes the road surface condition estimated based on the weather information, the maximum load of the vehicle included in the vehicle specification data 14a, the wear resistance performance of the tire 7 included in the tire specification data 14b, and the like. As the wear resistance performance of the tire 7, for example, a tire wear index value obtained by indexing the wear resistance performance of various tread formulations based on the Lambourn wear test, with the standard formulation being 100, is used.

The arithmetic model 13a uses a learning model such as a neural network, for example. The computational model 13a is constructed using, for example, a technique such as a DNN (Deep Neural Network) or a decision tree. Further, the calculation model 13a may be a multiple linear regression model for input information, for example, and may be model-generated by learning.

The learning processing unit 15 acquires each data in the wear amount database of the tire 7 generated by the data processing device 20, and uses the data as teacher data used for learning the computation model 13a. In the learning process of the computing model 13a, the wear amount of the tire 7 as output data is estimated by the computing model 13a based on the input information and compared with the teacher data.

The learning processing unit 15 compares the amount of wear of the tire 7 estimated by the arithmetic model 13a with the teacher data, newly sets various coefficients in the arithmetic process such as weighting in the arithmetic model 13a, and repeats updating of the model. to perform learning. Note that the learning processing unit 15 can use a known learning method such as gradient boosting. Also, for verification of the computational model 13a, a known verification method such as random data sampling or cross-validation can be used.

FIG. 4 is a block diagram showing the functional configuration of the data processing device 20. As shown in FIG. The data processing device 20 includes a communication section 21 , a data processing section 22 and a storage section 23 . Each part in the data processing device 20 can be realized by electronic elements such as a CPU of a computer, mechanical parts, etc. in terms of hardware, and by computer programs etc. in terms of software. It depicts the functional blocks implemented by Therefore, those skilled in the art will understand that these functional blocks can be realized in various forms by combining hardware and software.

The communication unit 21 is connected to the communication network 9 by wireless or wired communication, and communicates with the tire wear amount measuring device 60 . The storage unit 23 is a storage device configured by, for example, an SSD (Solid State Drive), hard disk, CD-ROM, DVD, etc., and stores a wear amount database 23a generated by the data processing unit 22. FIG. The data processing device 20 outputs each data in the generated wear amount database 23 a to the computational model generation device 10 , but may transmit each data to the computational model generation device 10 via the communication unit 21 .

The data processing unit 22 includes a data acquisition unit 22a, a threshold setting unit 22b, a continuity determination unit 22c, and a database generation unit 22d. The data acquisition unit 22 a acquires measurement data of the wear amount of the tire 7 mounted on the selected axial position (wheel position) of the vehicle from the tire wear amount measuring device 60 via the communication unit 21 . As described above, the tire wear amount measuring device 60 measures the wear amount of the tire 7 multiple times over a predetermined period (for example, several months to several years), and the data acquisition unit 22a collects all measured Get the wear amount data of The data acquisition unit 22a is capable of identifying a tire and a plurality of grooves of the tire in a predetermined period, and it is desirable to be able to acquire wear amount data for the same groove of the same tire in chronological order.

The threshold value setting unit 22b sets a lower limit side threshold value and an upper limit side threshold value for the data regarding the wear amount acquired by the data acquisition unit 22a. The threshold value setting unit 22b calculates a wear index value WI obtained by dividing the travel distance by the wear amount for each wear amount measurement data. The threshold value setting unit 22b may acquire data on the travel distance of the vehicle from the in-vehicle measurement device 70 or the arithmetic model generation device 10. FIG.

The threshold value setting unit 22b determines that the amount of wear of the tire 7 and the travel distance increase by 0.5 mm from one point in time of measuring the amount of wear to the next. , WI=1000/0.5=2000. The threshold value setting unit 22b can calculate the wear index value at each time point of wear amount measurement. As described above, when the tire wear amount measuring device 60, for example, has four grooves on the tire and measures the wear amount at four locations in the width direction, the threshold value setting unit 22b A wear index value is calculated for each wear amount. In addition to the four points in the width direction, the amount of wear may be measured at three points in the circumferential direction of the same groove at intervals of, for example, 120°.

The threshold value setting unit 22b calculates quartiles for all data of the wear index value by the quartile method, temporarily sets the first quartile to the first lower threshold, and sets the third quartile to Set to the second upper limit side threshold. The threshold value setting unit 22b calculates the average value WIa of the wear index value data within the range sandwiched between the lower limit side threshold and the upper limit side threshold.

The threshold value setting unit 22b calculates a ratio PA obtained by dividing all data of the wear index value WI by the average value WIa, and sets a predetermined range for the ratio PA. The predetermined range is, for example, a range in which the value of the ratio PA is 1/3 or more and 3 or less, where 1/3 is the second lower threshold and 3 is the second upper threshold. The predetermined range set by the threshold setting unit 22b is not limited to this, and the second lower limit threshold may be set to 1/4, and the second upper limit threshold may be set to 4 or the like. When calculating the average value of the wear index values WI, the threshold setting unit 22b and the continuity determination unit 22c may target a data group of wear index values in the same groove of the same tire, or a plurality of grooves of the same tire. A data group of wear index values in . Further, when calculating the average value of the wear index values WI, the threshold value setting unit 22b and the continuity determination unit 22c may calculate the average value for the data group of the wear index values for each axle. For example, in a large vehicle with three or more axles, the average value of the wear index values WI in all grooves of a plurality of tires arranged on the first front axle (steering axle) is calculated, and Calculate the average value of the wear index values WI in all the grooves of the plurality of tires arranged on the second (drive shaft), and calculate the wear in all the grooves of the plurality of tires arranged on the third rear axle (floating shaft). An average value of the index values WI may be calculated.

Data corresponding to the wear index value WI, which has a ratio PA smaller than the second lower limit threshold set by the threshold setting unit 22b, is temporarily recognized as inappropriate data caused by, for example, an error in measuring the amount of wear. . Further, data corresponding to the wear index value WI having a ratio PA larger than the second upper limit side threshold value set by the threshold value setting unit 22b is once recognized as inappropriate data. The threshold setting unit 22b outputs each piece of data recognized as inappropriate data to the continuity determination unit 22c.

The continuity determination unit 22c compares the data input from the threshold setting unit 22b with other data (i.e., data measured before and after) obtained with the same tire and with the same groove, It is determined whether or not the change in wear amount is continuous. As described above, the wear amount data is acquired multiple times within a predetermined period.

The continuity determination unit 22c calculates a ratio PB obtained by dividing the wear index value WI of other data acquired for the same tire and the same groove by the wear index value WI of the data input from the threshold setting unit 22b. A predetermined range is set for PB. The predetermined range is, for example, a range in which the value of the ratio PB is 1/3 or more and 3 or less. The predetermined range set by the continuity determination unit 22c is not limited to this, and may be set such that the value of the ratio PB is 1/4 or more and 4 or less.

As another continuity determination method, the continuity determination unit 22c calculates an average value WIc of the wear index values WI of the other data acquired on the same tire and groove as the data temporarily recognized as inappropriate. do. A ratio PC is calculated by dividing data temporarily recognized as inappropriate by WIc, and a predetermined range is set for the ratio PC. The predetermined range is, for example, a range in which the value of the ratio PC is 1/3 or more and 3 or less. The predetermined range set by the continuity determination unit 22c is not limited to this, and may be set such that the value of the ratio PC is 1/4 or more and 4 or less.

The continuity determination unit 22c determines that the ratio PB is continuous if it is within the above-mentioned predetermined range, and determines that the input data recognized as inappropriate by the threshold setting unit 22b is appropriate data. Recertify. The continuity determination unit 22c determines that the ratio PB is not continuous if the ratio PB is outside the predetermined range, and recognizes that the input data recognized as inappropriate by the threshold setting unit 22b is still inappropriate data. do.

The continuity determination unit 22c determines the continuity of the data, and outputs the data for which the determination result is negative to the database generation unit 22d. The database generation unit 22d generates the wear amount database 23a by excluding the data input from the continuity determination unit 22c from the wear amount data. That is, the wear amount database 23a is obtained by excluding data whose continuity is denied by the continuity determination unit 22c from all the measured wear amount data acquired by the data acquisition unit 22a.

Next, the operation of the data processing device 20 will be described. FIG. 5 is a flowchart showing a procedure of database generation processing by the data processing device 20. As shown in FIG. The data acquisition unit 22a of the data processing device 20 acquires wear amount data of the tire 7 measured by the tire wear amount measuring device 60 (S1). The threshold value setting unit 22b calculates a wear index value WI for each piece of wear amount data acquired by the data acquisition unit 22a (S2).

The threshold value setting unit 22b sets a first lower threshold value and a first upper threshold value for the wear index value WI of each wear amount data based on the quartile method (S3). The threshold value setting unit 22b calculates the average value WIa of the data of the wear index values WI within the range between the lower threshold value and the upper threshold value (S4).

The threshold value setting unit 22b calculates a ratio PA obtained by dividing all data of the wear index value WI by the average value WIa (S5). The threshold value setting unit 22b temporarily recognizes data outside a predetermined range in the ratio PA of each data of the wear index value WI as inappropriate data, and outputs the data to the continuity determination unit 22c (S6).

The continuity determination unit 22c calculates a ratio PB obtained by dividing the wear index value WI of other data acquired for the same tire and same groove by the wear index value WI of the data temporarily recognized as inappropriate (S7). The continuity determination unit 22c determines whether the ratio PB is within a predetermined range (S8).

If it is determined in step S8 that the ratio PB is within the predetermined range (S8: YES), the database generation unit 22d causes the wear amount database 23a to include the data input from the continuity determination unit 22c (S9). , terminate the process. When step S8 determines that the ratio PB is not within the predetermined range (S8: NO), the database generation unit 22d excludes the data input from the continuity determination unit 22c from the wear amount database 23a (S10), End the process.

The data processing device 20 determines whether data outside the range of the lower limit side threshold and the upper limit side threshold is continuous with respect to the data regarding the wear amount of the tire 7 measured over a plurality of times in a predetermined period. If it is determined to be negative, the data is excluded from the database. As a result, the data processing device 20 can exclude inappropriate data, select wear amount measurement data of the tire 7, and efficiently generate a database.

The data processing device 20 calculates a wear index value WI obtained by dividing the travel distance of the vehicle by the wear amount for the wear amount data of the tire 7, and the wear index value WI is continuous for the data outside the range of the lower limit side threshold and the upper limit side threshold. determine if it is relevant. The amount of wear of the tire 7 increases as the travel distance of the vehicle increases. The data processing device 20 uses the wear index value WI obtained by dividing the travel distance of the vehicle between the time when the wear amount was measured and the time when the previous wear amount was measured in a predetermined period by the wear amount. can be suppressed.

The threshold setting unit 22b of the data processing device 20 calculates the first quartile as the lower limit value and the third quartile as the upper limit value based on the quartile method, and the lower limit value is the first lower threshold, The upper limit is set as the first upper threshold. The threshold value setting unit 22b calculates an average value WIa for the data of the wear index value WI within the range of the first lower threshold value and the first upper threshold value, and divides each data by the average value WIa, and the ratio is within a predetermined range. is predetermined, the lower limit of the predetermined range is set as the second lower limit side threshold, and the upper limit is set as the second upper limit side threshold. As a result, even if the data is outside the range of the first lower threshold and the first upper threshold, the data processing device 20 can store the data in the wear amount database if it is within the range of the second lower threshold and the second upper threshold. 23a.

The data processing device 20 determines continuity with other data by the continuity determining unit 22c for data outside the range of the first lower threshold and the first upper threshold set by the threshold setting unit 22b, You may decide whether to include it in the wear amount database 23a or to exclude it. As a result, the data processing device 20 excludes, from the wear amount database 23a, data determined to have no continuity with other data among the data outside the range of the first lower limit side threshold and the first upper limit side threshold. be able to.

The computational model generation system 100 improves the accuracy of estimating the wear amount of the tire 7 by making the computational model 13a learn using the wear amount data in the wear amount database 23a generated by the data processing device 20 as teacher data. A computational model 13a can be generated.

FIG. 6 is a chart showing the result of wear amount estimation of the arithmetic model 13a learned from the measured wear amount data. The wear amount data used are those measured on actual vehicles. The computational model 13a was verified using a computational model based on a decision tree model and a computational model based on a neural network model as examples.

In FIG. 6, the RMSE (root-mean-square square root error) is shown. As for the result of the wear amount estimation by the computational model 13a, it can be seen that the RMSE value becomes lower and the estimation accuracy is improved when inappropriate data are excluded in both cases by the decision tree and the neural network.

(Modification)
In the above-described embodiment, the threshold setting unit 22b sets the first lower threshold and the first upper threshold based on the quartile method. You may set 1 lower limit side threshold and a 1st upper limit side threshold. Moreover, the threshold value setting unit 22b may use another statistical method for setting the second lower limit side threshold value and the second upper limit side threshold value. Further, the continuity determination unit 22c determines the continuity by setting a predetermined range for the ratio PB, but the continuity determination method is not limited to this.

Further, the threshold value setting unit 22b calculates the wear index value WI for each piece of wear amount data of the tire 7. It is not necessary to use the index value WI, and each threshold may be set for the amount of wear.

The data processing device 20 excludes data for which the continuity determination unit 22c determined that there is no continuity from the wear amount database 23a. The database generator 22d of the data processing device 20, for example, when the tire air pressure is lower than a predetermined threshold value and the amount of wear of the tire 7 is large, even if the data is determined to be non-continuous, , may be included in the wear amount database 23a.

The data processing device 20 may be configured as a device separate from the computational model generation device 10, or may be configured integrally with the computational model generation device 10 as one device.

Next, features of the data processing device 20, the data processing method, and the computational model generation system 100 according to the embodiment and modifications will be described.
The data processing device 20 includes a data acquisition section 22a, a threshold setting section 22b, a continuity determination section 22c, and a database generation section 22d. The data acquisition unit 22a acquires data regarding the amount of wear of the tire 7 mounted on the vehicle in a predetermined period. The threshold setting unit 22b sets a lower threshold and an upper threshold for the data acquired by the data acquisition unit 22a. The continuity determination unit 22c determines whether or not the data outside the range between the lower limit side threshold and the upper limit side threshold set by the threshold setting unit 22b changes continuously in a predetermined period. The database generation unit 22d generates the wear amount database 23a by excluding data for which the determination result of the continuity determination unit 22c was negative. As a result, the data processing device 20 can select wear amount measurement data of the tire 7 by excluding inappropriate data, and efficiently generate the wear amount database 23a.

Further, the data is a wear index value obtained by dividing the travel distance of the vehicle by the amount of wear. As a result, the data processing device 20 can suppress variations in data due to travel distance.

The threshold value setting unit 22b calculates the lower limit value and the upper limit value by the quartile method, and sets the lower limit value as the first lower threshold value and the upper limit value as the first upper threshold value. As a result, the data processing device 20 excludes, from the wear amount database 23a, data determined to have no continuity with other data among the data outside the range of the first lower limit side threshold and the first upper limit side threshold. be able to.

Further, the threshold value setting unit 22b calculates an average value for the data within the range of the lower limit value and the upper limit value calculated by the quartile method, and predetermines a predetermined range for the ratio obtained by dividing each data by the average value. , the lower limit of the predetermined range is defined as a second lower limit side threshold, and the upper limit is defined as a second upper limit side threshold. As a result, even if the data is outside the range of the first lower threshold and the first upper threshold, the data processing device 20 can store the data in the wear amount database if it is within the range of the second lower threshold and the second upper threshold. 23a.

The data processing method includes a data acquisition step, a threshold setting step, a continuity determination step, and a database generation step. The data acquisition step acquires data on the amount of wear of the tire 7 mounted on the vehicle for a predetermined period of time. The threshold setting step sets a lower threshold and an upper threshold for the data acquired by the data acquisition step. The continuity determination step determines whether or not the data outside the range of the lower limit side threshold and the upper limit side threshold set by the threshold setting step continuously change during the predetermined period. The database generation step generates the wear amount database 23a by excluding data for which the determination result of the continuity determination step was negative. According to this data processing method, it is possible to select measurement data of the amount of wear of the tire 7, excluding inappropriate data, and efficiently generate a database.

The computational model generation system includes the data processing device 20, the wear amount calculation unit 13, and the learning processing unit 15 described above. The wear amount calculation unit 13 has a learning type arithmetic model 13a that calculates the wear amount of the tire 7 based on the input information. Calculate the wear amount of The learning processing unit 15 compares the wear amount data included in the wear amount database 23a generated by the data processing device 20 with the wear amount calculated by the wear amount calculation unit 13, and causes the computation model 13a to learn. As a result, the computational model generation system can generate the computational model 13a that improves the accuracy of estimating the wear amount of the tire 7 .

The above has been described based on the embodiments of the present invention. Those skilled in the art will appreciate that these embodiments are illustrative and that various variations and modifications are possible within the scope of the claims of the present invention, and that such variations and modifications also fall within the scope of the claims of the present invention. It is about to be done. Accordingly, the description and drawings herein are to be regarded in an illustrative rather than a restrictive sense.

7 tire 13 wear amount calculator 13a computing model 15 learning processing unit
20 data processing device, 22a data acquisition unit, 22b threshold value setting unit,
22c continuity determination unit, 22d database generation unit,
23a Wear amount database, 100 Calculation model generation system.

Claims (6)

a data acquisition unit that acquires data relating to the amount of wear of a tire mounted on a vehicle in a predetermined period;
a threshold setting unit that sets a lower threshold and an upper threshold for the data acquired by the data acquisition unit;
a continuity determination unit that determines whether or not the data outside the range of the lower limit side threshold and the upper limit side threshold set by the threshold value setting unit changes continuously in the predetermined period;
a database generation unit that generates a wear amount database by excluding data for which the continuity determination unit has determined negative;
A data processing device comprising: 2. The data processing apparatus according to claim 1, wherein said data is a wear index value obtained by dividing the running distance of the vehicle by the amount of wear. 3. The threshold value setting unit calculates the lower limit value and the upper limit value by the quartile method, and sets the lower limit value as the lower limit side threshold and the upper limit value as the upper limit side threshold. Data processing apparatus as described. The threshold value setting unit calculates an average value for data within the range of the lower limit value and the upper limit value calculated by the quartile method, and predetermines a predetermined range for the ratio obtained by dividing each data by the average value. 3. The data processing apparatus according to claim 1, wherein a lower limit value of said predetermined range is set as a lower limit side threshold, and an upper limit is set as said upper limit side threshold. a data acquisition step of acquiring data regarding the amount of wear of a tire mounted on a vehicle for a predetermined period of time;
a threshold setting step of setting a lower threshold and an upper threshold for the data acquired in the data acquisition step;
a continuity determination step of determining whether or not the data outside the range of the lower limit side threshold and the upper limit side threshold set in the threshold setting step are continuous in the predetermined period;
a database generation step of generating a wear amount database by excluding data for which the determination result of the continuity determination step was negative;
A data processing method comprising: A data processing device according to any one of claims 1 to 4;
It has a learning type computational model for calculating the amount of tire wear based on the input information, and inputs information including at least the mileage of the vehicle into the computational model to calculate the amount of tire wear. Department and
a learning processing unit that compares the wear amount data included in the wear amount database generated by the data processing device with the wear amount calculated by the wear amount calculation unit to learn the arithmetic model;
A calculation model generation system characterized by comprising:

Images