Information processing apparatus

Abstract

The invention provides an information processing device capable of extracting, from raw data, data suitable for analyzing the lubrication history of a power transmission device mounted on a vehicle. The information processing apparatus includes: a first step (S110) of setting a plurality of time windows for cutting out data from raw data using a physical quantity related to a lubrication history of the power transmission apparatus as a feature quantity; a second step (S120) for cutting out data from the original data; a third step (S130) for calculating the frequency distribution of the feature quantity obtained from the raw data and the frequency distribution of the feature quantity obtained from the extracted data; and a fourth step (S140 and S150) for determining whether the original data and the extracted data are similar or not using the frequency distribution. The information processing apparatus repeatedly executes the search processing of the first to fourth steps by changing the settings of the plurality of time windows, and extracts extracted data similar to the original data.

Description

Technical Field

[0001] This invention relates to an information processing device. Background Technology

[0002] Patent Document 1 discloses an information processing device that reduces the size of the data used for analysis by compressing the raw data. The raw data used for analysis is data collected over a predetermined period using sensors mounted on a vehicle.

[0003] The information processing device disclosed in Patent Document 1 compresses data by extracting data acquired at the time point when a certain vehicle speed is reached and data acquired at the time point when the vehicle speed inflection point from the original data.

[0004] Existing technical documents

[0005] Patent documents

[0006] Patent Document 1: Japanese Patent Application Publication No. 2008-108247 Summary of the Invention

[0007] The problem the invention aims to solve

[0008] Data acquired from multiple sensors mounted on the vehicle can be used to analyze whether lubrication of components in the vehicle's power transmission system has been adequate or insufficient. If suitable lubrication history data can be extracted from the raw data, the lubrication history can be analyzed in a shorter time than when using the raw data. The aforementioned information processing device uses vehicle speed, position information, and driving data at any given time. This device establishes a correspondence between vehicle speed and driving position to extract vehicle speed variation patterns and sets an operating schedule for the engine and motor to minimize fuel consumption. However, this information processing device cannot extract suitable lubrication history data from the raw data for analyzing the components in the vehicle's power transmission system.

[0009] Solution for solving the problem

[0010] An information processing apparatus for solving the aforementioned problem extracts a portion of data from raw data collected continuously for a predetermined period using multiple sensors mounted on a vehicle, and extracts data for analyzing the lubrication history of components in the power transmission system mounted on the vehicle. This information processing apparatus includes a processing circuit. The processing circuit performs a first step by setting multiple time windows to extract data from a portion of the raw data, such that the sum of all time windows is shorter than the predetermined period. The processing circuit performs a second step by extracting data from the raw data through the multiple time windows. The processing circuit uses the rotational speed of a rotating body that lubricates the components by rotating and splashing lubricating oil stored in the housing of the power transmission system as a characteristic quantity. The processing circuit performs a third step by calculating the frequency distribution of the characteristic quantity obtained from the raw data and the frequency distribution of the characteristic quantity obtained from the extracted data, which is obtained by combining all data extracted through the multiple time windows. The processing circuit performs a fourth step by using the frequency distribution to determine whether the raw data and the extracted data are similar. The processing circuit repeatedly executes steps 1 to 4 by changing the settings of the multiple time windows. The information processing device extracts extracted data similar to the original data.

[0011] An information processing apparatus for solving the aforementioned problem extracts a portion of data from raw data collected continuously for a predetermined period using multiple sensors mounted on a vehicle, and extracts data for analyzing the lubrication history of components in the power transmission system of the vehicle. This information processing apparatus includes a processing circuit. The processing circuit performs a first step by setting multiple time windows to extract data from a portion of the raw data, such that the sum of all time windows is shorter than the predetermined period. The processing circuit performs a second step by extracting data from the raw data through the multiple time windows. The processing circuit uses the rotational speed of an oil pump that discharges lubricating oil to lubricate the components as a characteristic quantity. The processing circuit performs a third step by calculating the frequency distribution of the characteristic quantity obtained from the raw data and the frequency distribution of the characteristic quantity obtained from the extracted data, which is obtained by combining all data extracted through the multiple time windows. The processing circuit performs a fourth step by using the frequency distribution to determine whether the raw data and the extracted data are similar. The processing circuit repeatedly performs steps 1 to 4 by changing the settings of the multiple time windows. The aforementioned information processing device extracts extracted data that is similar to the original data.

[0012] The effects of the invention

[0013] Using the aforementioned information processing device, it is possible to extract data from raw data that is suitable for analyzing the lubrication history of components in the power transmission system of a vehicle. Attached Figure Description

[0014] Figure 1 This is a schematic diagram illustrating the relationship between a data center, a vehicle, and an information processing terminal as one embodiment of an information processing device.

[0015] Figure 2 This is a cross-sectional view of a power distribution mechanism, which is an example of a power transmission device.

[0016] Figure 3 This is a graph representing a portion of the raw data related to the lubrication history of the components in the power distribution mechanism, (a) representing the shift in the rotational speed of the differential gear ring, (b) representing the shift in the rotational speed of the sun gear, (c) representing the vehicle tilt angle, and (d) representing the rotational speed of the second oil pump.

[0017] Figure 4 It is a flowchart representing the processing flow performed by the processing circuitry of a data center.

[0018] Figure 5 It is the frequency distribution of the rotational speed of the differential gear ring in the original data.

[0019] Figure 6 It is the frequency distribution of the rotational speed of the second oil pump in the original data.

[0020] Explanation of reference numerals in the attached figures

[0021] 10…vehicle, 20…hybrid power system, 30…engine, 31…first electric generator, 32…second electric generator, 40…power distribution mechanism, 41…housing, 42…planetary gear unit, 43…planet carrier, 44…ring, 45…sun gear, 46…pinion, 47…ring gear, 49…reduction gear, 53…differential ring gear, 56…first oil pump, 57…second oil pump, 80…vehicle control unit, 90…communication device, 400…communication network, 500…data center, 510…processing circuit, 520…storage device, 530…communication device, 600…information processing terminal, 610…processing circuit, 620…storage device, 630…communication device. Detailed Implementation

[0022] <First Implementation>

[0023] The following is for reference Figures 1-5 The first embodiment of the information processing device will be described.

[0024] <Structure of Information Processing Systems>

[0025] Figure 1This describes the structure of the information processing system. The information processing system includes a data center 500 equipped with information processing devices, an information processing terminal 600, multiple vehicles 10, and a communication network 400. The data center 500 can communicate with the multiple vehicles 10 and the information processing terminal 600 via the communication network 400.

[0026] <Data Center 500 Structure>

[0027] like Figure 1 As shown, the data center 500 includes a processing circuit 510, a storage device 520, and a communication device 530. The processing circuit 510 is an information processing device, equipped with a CPU that executes processing according to a program and a ROM storing the program. The storage device 520 is capable of storing a large amount of data. The communication device 530 performs wired or wireless communication via a communication network 400. The communication device 530 includes hardware such as a network adapter, various communication software, or combinations thereof.

[0028] <Structure of Information Processing Terminal 600>

[0029] like Figure 1 As shown, the information processing terminal 600 includes a processing circuit 610, a storage device 620, and a communication device 630. The processing circuit 610 includes a CPU that executes processing according to a program and a ROM storing the program. The storage device 620 is capable of storing a large amount of data. The communication device 630 performs wired or wireless communication via a communication network 400. The communication device 630 includes hardware such as a network adapter, various communication software, or combinations thereof.

[0030] Information processing terminal 600 is, for example, a personal computer.

[0031] <Structure of Vehicle 10>

[0032] Vehicle 10 is equipped with a communication device 90. The communication device 90 transmits data acquired by vehicle 10 to data center 500 via communication network 400. Vehicle 10 is equipped with a hybrid power system 20 and a vehicle control unit 80. The hybrid power system 20 includes an engine 30, a first electric generator 31, a second electric generator 32, and a power distribution mechanism 40. The vehicle control unit 80 includes a first control device 81 for controlling engine 30 and a second control device 82 for controlling the first electric generator 31 and the second electric generator 32. The vehicle control unit 80 is equipped with multiple sensors for data collection. The first control device 81 is equipped with a CPU for controlling the operating state of engine 30. The first control device 81 controls engine 30 based on data collected by the sensors. The second control device 82 is equipped with a CPU for controlling the first electric generator 31 and the second electric generator 32. The second control device 82 controls the first electric generator 31 and the second electric generator 32 based on data collected by the sensors. One example of the data collected by the vehicle control unit 80 is the crankshaft speed, the electric generator speed, the speed of the gears and shafts constituting the power distribution mechanism 40, and the temperature of the lubricating oil.

[0033] <Structure of Power Distribution Mechanism 40>

[0034] The power distribution mechanism 40 is a power transmission device that transmits power generated by the engine 30, the first electric generator 31, and the second electric generator 32 to the drive wheels of the vehicle 10. For example... Figure 2 As shown, the power distribution mechanism 40 has a planetary gear unit 42, a reduction gear 49, a motor gear 52 and a differential gear ring 53 inside the housing 41. Figure 2 The power distribution mechanism 40 shown is used, for example, in a front-wheel drive vehicle with a hybrid powertrain 20 at the front of the vehicle. The planetary gear unit 42 includes a sun gear 45, three pinions 46, a planet carrier 43, and a ring 44. The sun gear 45 is located at the center of the planetary gear unit 42. The sun gear 45 is connected to a first electric generator 31. The three pinions 46 are arranged around the sun gear 45, supported by the planet carrier 43. The ring 44 has a gear ring 47 on its inner circumferential surface and an output gear 48 on its outer circumferential surface. The rotation axis of the sun gear 45, the rotation axis of the planet carrier 43, the rotation axis of the ring 44, and the engine output shaft 33 are on the same straight line. The engine output shaft 33 is the output shaft of the engine 30. The output of the engine 30 is input to the planet carrier 43.

[0035] exist Figure 2The diagram shows axes S1 to S4. Axis S1 is the axis through which the sun gear 45, planet carrier 43, ring 44, and engine output shaft 33 pass. Axis S2 is the axis through which the reduction gear 49 passes. Axis S3 is the axis through which the output shaft of the second electric generator 32 and the motor gear 52 pass. The motor gear 52 is fixed to the output shaft of the second electric generator 32. Axis S4 is the axis through which the differential gear ring 53 passes.

[0036] The output of the engine 30, input to the planetary carrier 43, is distributed from the pinion 46 to the sun gear 45 and the ring gear 47. The first electric generator 31 is, for example, a rotary electric motor used for both power generation and propulsion. The torque distributed to the ring gear 47 drives the ring 44 to rotate, which in turn drives the output gear 48. The output gear 48 meshes with the reduction gear 50 of the reduction gear 49 mounted on the S2 shaft. Furthermore, the reduction gear 50 also meshes with the motor gear 52 of the second electric generator 32 mounted on the S3 shaft. The second electric generator 32 is, for example, a rotary electric motor used for propulsion. Figure 2 As shown, the reduction gear 49, in addition to the large reduction gear 50, also includes a small reduction gear 51. The small reduction gear 51 meshes with a differential gear ring 53 mounted on the S4 shaft. The driving force of the differential gear ring 53 is transmitted to the drive wheel via the differential device and the drive shaft. According to the above structure, the power distribution mechanism 40 uses the reduction gear 49 to integrate the outputs of the engine 30 and the first electric generator 31 on the S1 shaft and the output of the second electric generator 32 on the S3 shaft into a single power source, which is then transmitted to the drive wheel.

[0037] Lubrication of the components in the power distribution mechanism 40

[0038] The components in the power distribution mechanism 40 are lubricated by lubricating oil within the housing 41 of the power distribution mechanism 40. Lubrication of the components is achieved through the splashing of lubricating oil by rotating components immersed in lubricating oil, the dispersion of lubricating oil from rotating components coated with lubricating oil, and the distribution of lubricating oil supplied by an oil pump. Lubricating oil is stored in the lower part of the housing 41. Figure 2 The single-dotted line indicates the level of lubricating oil stored in the casing 41. For example... Figure 2 As shown, the lower part of the gear of the differential gear ring 53 is immersed in lubricating oil. Therefore, when the differential gear ring 53 rotates, the gear inside the housing 41 is lubricated by the splashed lubricating oil. On the other hand, the first oil pump 56 and the second oil pump 57 draw in the lubricating oil stored in the lower part of the housing 41 and deliver the lubricating oil to the supply path. The suction ports of the first oil pump 56 and the second oil pump 57 for drawing in lubricating oil are located at... Figure 2The two oil pumps are located on the right side of the partition wall 55 provided in the housing 41. The two oil pumps deliver lubricating oil drawn in from the intake port to the lubricating oil supply path and distribute it to the various components of the power distribution mechanism 40. For example, the first oil pump 56 is configured to be connected to the pump drive gear 54 that meshes with the differential gear ring 53. In this case, when the rotational speed of the differential gear ring 53 is high, the distribution of lubricating oil by means of the first oil pump 56 becomes intense. The second oil pump 57 is, for example, an oil pump connected to the engine output shaft 33 of the engine 30. In this case, when the rotational speed of the engine 30 is high, the distribution of lubricating oil by means of the second oil pump 57 becomes intense. Furthermore, the rotating components of the power distribution mechanism 40, which are lubricated by splashing or dispersing lubricating oil, cause the lubricating oil adhering to these components to scatter within the housing 41 as they rotate. Therefore, when the rotational speed of the rotating components is high, the scattering of lubricating oil becomes intense. Therefore, the lubrication status of each component of the power distribution mechanism 40 is affected by the splashing of lubricating oil by the differential gear ring 53, the distribution of lubricating oil via the first oil pump 56 and the second oil pump 57, and the splashing of lubricating oil by each rotating component.

[0039] Data Extraction

[0040] Information processing terminal 600 is used to analyze the lubrication history of components in power distribution mechanism 40, an example of a power transmission device mounted in a vehicle. Lubrication history analysis refers to using data acquired by multiple sensors mounted in the vehicle to analyze whether lubrication of the components in the power transmission device has been adequate or insufficient. When analyzing the lubrication history, information processing terminal 600 sends an instruction to data center 500. Upon receiving the instruction, processing circuit 510 of data center 500 analyzes a portion of the vast amount of data stored in storage device 520 of data center 500. Depending on the purpose of the analysis, data to be used is selected from the vast amount of data stored in storage device 520. This data includes data on physical quantities related to the lubrication history of the components collected using multiple sensors mounted in vehicle 10. These physical quantities are called characteristic quantities. Processing circuit 510 uses characteristic quantities to analyze the lubrication history of the components in the power distribution mechanism 40 of a specific vehicle 10. The number of characteristic quantities used for analysis can be one or more.

[0041] To perform this analysis, the processing circuit 510 utilizes a large amount of data collected over a prolonged period. This analysis involves a massive amount of computation, thus requiring a considerable amount of time.

[0042] Therefore, it is considered to extract extractable data that captures the overall characteristics of the original data from a large amount of data. If such extractable data can be extracted, the processing circuit 510 can perform analysis in a shorter time by using the extracted data. For example, in the case of analyzing the lubrication history of the aforementioned components after 100,000 hours of operation, the processing circuit 510 extracts extractable data equivalent to 20,000 hours from the original data of 100,000 hours.

[0043] Figure 3 The original data representing the characteristic quantities associated with the power distribution mechanism 40. Figure 3 The raw data shown is a portion of the data for 100,000 hours of a single vehicle. Figure 3 The raw data shown includes four characteristic quantities. The first is the rotational speed of the rotating body that lubricates the aforementioned components by splashing lubricating oil stored in the housing 41 of the power distribution mechanism 40 through rotation. In this embodiment, this rotational speed is the rotational speed of the differential gear ring 53. The second is the rotational speed of the second rotating body that lubricates the aforementioned components by scattering adhering lubricating oil through rotation within the housing 41 of the power distribution mechanism 40. In this embodiment, the rotational speed of the second rotating body is the rotational speed of the sun gear 45. The third is the tilt angle of the vehicle 10. The fourth is the rotational speed of the second oil pump 57, which serves as an oil pump for discharging lubricating oil that lubricates the aforementioned components.

[0044] Figure 3 (a) represents the rotational speed of the differential gear ring 53 in 100,000 hours of data. Figure 3 (b) represents the rotational speed of the sun gear 45 in 100,000 hours of data. Figure 3 (c) represents the tilt angle of vehicle 10. Figure 3 (d) indicates the rotational speed of the second oil pump 57.

[0045] Data extraction is achieved by using multiple time windows to extract data from the raw data. Figure 3 In this example, as an example of multiple time windows, dashed lines represent the first time window W_1, the second time window W_2, and the third time window W_3. The start and end points of each time window are set to ensure they do not overlap. In this example, 20,000 hours of data are extracted. Therefore, the start and end points of each time window are set to a total length of 20,000 hours, calculated by summing the durations of all time windows.

[0046] Data center 500 search represents the setting of the start and end points of each time window for the data extraction pattern used to extract features captured from the overall raw data. Data center 500 stores the information of the data extraction pattern used to extract the above-mentioned data in storage device 520. The stored data extraction pattern information is the information on the settings of each time window discovered through the search.

[0047] The processing circuit 510 extracts data from the raw data based on the information of the cut-out mode stored in the storage device 520.

[0048] <Search Processing in Cutout Mode>

[0049] Figure 4 This is a flowchart illustrating a series of processes related to the search process for the cut-out mode. This series of processes is executed by the processing circuit 510 of the data center 500 according to a program. The following shows an example of the processing circuit 510 using two characteristic quantities: a first characteristic quantity and a second characteristic quantity. In this case, the first characteristic quantity is the rotational speed of the differential gear ring 53. The second characteristic quantity is the rotational speed of the sun gear 45.

[0050] like Figure 4 As shown, the processing circuit 510 acquires raw data in step S100. The raw data is a subset of data selected from the vast amount of data stored in the storage device 520 of the data center 500 according to the purpose of analysis.

[0051] The raw data used to analyze the lubrication history of the components in the power distribution mechanism 40 of a vehicle 10 is the data of the object vehicle 10 selected from the massive data of multiple vehicles 10.

[0052] Next, in step S110, the processing circuit 510 sets multiple time windows to extract data from the raw data. Figure 3 In the example shown, the duration of each time window is all equal. Figure 3 As shown, the data cut out by each cutout window is the data of each feature quantity in the same period.

[0053] Each time step S110 is executed, the processing circuit 510 randomly sets the number of time windows, the start point of each time window, and the end point of each time window. At this time, the processing circuit 510 sets each time window in a manner that prevents them from overlapping. Thus, the processing circuit 510 randomly sets multiple time windows in such a way that the sum of the periods of all time windows becomes a preset period. The processing circuit 510 may also, in the processing of step S110, as follows: Figure 3 As shown, the duration of each time window is fixed to a constant, and multiple time windows are set. Alternatively, in step S110, the processing circuit 510 can also set multiple time windows by fixing the number of time windows to a constant value.

[0054] In this way, multiple time windows are set through the processing in step S110, thereby determining the data extraction mode from the original data. After determining the extraction mode, the processing circuit 510 causes the processing to proceed to step S120.

[0055] In step S120, the processing circuit 510 extracts data from the original data using a determined extraction mode. That is, in step S120, the processing circuit 510 extracts data from the original data using multiple pre-defined time windows. Then, the processing circuit 510 combines all the data extracted through the multiple time windows to create the extracted data.

[0056] Next, in step S125, the processing circuit 510 classifies the data of the first feature contained in the original data into multiple partitions based on the data of the second feature when the first feature is collected. Similarly, the processing circuit 510 classifies the data of the first feature contained in the extracted data into multiple partitions based on the data of the second feature, in a manner corresponding to the partitions of the original data.

[0057] In the processing of step S130, the processing circuit 510 calculates the frequency distribution of the first feature of the original data and the frequency distribution of the first feature of the extracted data for each partition based on the original data classified into multiple partitions and the data of the first feature of the extracted data.

[0058] The data for the first feature is classified into multiple levels, and the frequency distribution represents the distribution of the number of data points in each level. Since the total frequency of the first feature differs between the original and extracted data, a simple comparison of the frequency distributions of the original and extracted data is not possible. In the case of extracting 20,000 hours of data from 100,000 hours of original data, the total frequency of the extracted data is approximately one-fifth of the total frequency of the original data. In this case, by multiplying the frequency of each level of the extracted data by five, a frequency distribution with the same total frequency as the original data can be obtained. Regardless of the method described above, even when calculating the relative frequency distribution as the frequency distribution of the original and extracted data, it is possible to compare the distribution of data in the original data and the distribution of data in the extracted data. The relative frequency distribution represents the percentage of the total frequency for a given level.

[0059] Figure 5This represents the frequency distribution of the rotational speed of the differential gear ring 53, obtained from the raw data of vehicle 10, as the first characteristic quantity. In this frequency distribution, zero rotational speed is taken as the smallest level, and the rotational speeds are divided into m levels from "1" to "m". When the rotational speed of the sun gear 45, as the second characteristic quantity, is divided into multiple zones, the following calculation is performed for each zone of the second characteristic quantity: Figure 5 The frequency distribution of the rotational speed of the differential gear ring 53 in the original data is shown. For example, it can be classified into a partition where the rotational speed of the sun gear 45 is less than a predetermined speed and a partition where the rotational speed of the sun gear 45 is greater than or equal to the predetermined speed. In this case, two frequency distributions are calculated: one for the rotational speed of the differential gear ring 53 when the rotational speed of the sun gear 45 is less than the predetermined speed and the other for the rotational speed of the differential gear ring 53 when the rotational speed of the sun gear 45 is greater than or equal to the predetermined speed. Furthermore, corresponding to the above partitions and levels, the frequency distribution of the rotational speed of the differential gear ring 53 is also calculated for each partition based on the rotational speed of the sun gear 45 for the extracted data.

[0060] Next, the processing circuit 510 in Figure 4 In step S140, the error between the frequency distribution of the first feature obtained from the original data and the frequency distribution of the first feature obtained from the extracted data is calculated for each of the multiple partitions based on the second feature. For example, the processing circuit 510 calculates the mean absolute error (MAE). The mean absolute error (MAE) is represented by the following mathematical formula 1.

[0061] [Formula 1]

[0062]

[0063] In the above mathematical formula 1, "n" is the total number of ranks in the frequency distribution. For example, in Figure 5 In the example shown, "n" is "m". "i" is the number that determines the rank in the frequency distribution. For example, if... Figure 5 In the example shown, "i" is the number from "1" to "m". "Y" is the frequency of the first feature in the corresponding level of the original data. "y" is the frequency of the first feature in the corresponding level of the extracted data.

[0064] As shown in the above mathematical formula 1, the processing circuit 510 calculates the sum of the errors of the frequencies in each level of the first characteristic quantity of the frequency distribution obtained from the original data and the frequency distribution obtained from the extracted data for each partition as the error.

[0065] When the error is calculated for all partitions, the processing circuit 510 proceeds to step S150. In step S150, the processing circuit 510 determines whether the calculated error for each partition is below a threshold. The threshold is a value used to determine whether extracted data with a frequency distribution close to the frequency distribution in the original data has been extracted using a set cutout mode. The size of this threshold is preset so that it can determine whether extracted data with a frequency distribution close to the frequency distribution in the original data has been extracted based on the error being below the threshold. This threshold can be determined as a value that differs for each partition.

[0066] In the process of step S150, if it is determined that the error of each partition is below the threshold (step S150: yes), the processing circuit 510 causes the process to proceed to step S160.

[0067] On the other hand, in the processing of step S150, if it is determined that any one of the errors is greater than the threshold (step S150: No), the processing circuit 510 returns the processing to step S110. That is, in order to reset the time window and extract the extracted data from the original data, the processing circuit 510 starts the processing of setting multiple new time windows.

[0068] In this way, the processing circuit 510 uses the frequency distribution of the first feature quantity partitioned by the second feature quantity to repeatedly perform the processing steps S110 to S150 until it can extract data similar to the original data.

[0069] As a result, in step S160, the processing circuit 510 stores in the storage device 520 a cut-out pattern in which the error of each partition is below a threshold. The extracted data cut out using this cut-out pattern is similar to the original data in the distribution of physical quantities related to lubrication history. Therefore, the extracted data described above can be used instead of the original data to analyze the lubrication history of the components in the power distribution mechanism 40.

[0070] <Function of the first embodiment>

[0071] The data center 500 of the information processing device in the first embodiment extracts a portion of the raw data collected continuously over a predetermined period using multiple sensors mounted on the vehicle 10. This data is extracted data used to analyze the lubrication history of the components in the power transmission device, i.e., the power distribution mechanism 40, mounted on the vehicle 10.

[0072] Data center 500 includes processing circuit 510. The raw data includes the rotational speed of the differential gear ring 53, which is the aforementioned rotating body, as a first characteristic. The raw data also includes the rotational speed of the sun gear 45, which is the aforementioned second rotating body, as a second characteristic. In this data center 500, processing circuit 510 performs a search process. The search process includes a first step (step S110), in which multiple time windows are set to extract a portion of the raw data's duration such that the sum of all time windows is shorter than the overall duration of the raw data. The search process includes a second step (step S120) to extract data from the raw data through the multiple time windows. The extracted data is obtained by combining all the data extracted from the raw data through the multiple time windows. The search process includes a third step (steps S125 and S130), in which the raw data is partitioned into multiple regions based on the second characteristic contained in the raw data, and the frequency distribution of the first characteristic obtained from the raw data is calculated for each region. On the other hand, the search process also includes a third step (steps S125 and S130), in which the extracted data is partitioned into multiple partitions corresponding to the multiple partitions of the original data by means of the second feature, and the frequency distribution of the first feature obtained from the extracted data is calculated for each partition. The exploration process includes a fourth step (steps S140 and S150), in which the error between the frequency distribution of the original data and the frequency distribution of the extracted data is calculated for each of the above-mentioned partitions, and it is determined whether the original data and the extracted data are similar. The processing circuit 510 performs a search process that repeatedly executes the processing of steps 1 to 4 by changing the settings of multiple time windows. As a result, the processing circuit 510 finds the data segmentation pattern where the error of each partition is below the threshold. Then, the processing circuit 510 stores the data segmentation pattern when extracting data similar to the original data in the storage device 520 (step S160).

[0073] The data center 500 uses the rotational speed of the differential gear ring 53, which acts as the rotating body, as a first characteristic quantity. Lubricating oil stored in the housing 41 of the power distribution mechanism 40 is splashed up by the rotating body to lubricate the constituent components. The higher the rotational speed of the rotating body, the more violently the lubricating oil in the housing 41 is splashed up. On the other hand, the data center 500 uses the rotational speed of the sun gear 45, which acts as the second rotating body, as a second characteristic quantity. The higher the rotational speed of the sun gear 45, which acts as the second rotating body, the more violently the lubricating oil is dispersed from the second rotating body. That is, the rotational speed of the second rotating body affects the lubrication history of the constituent components.

[0074] The aforementioned data center 500 uses the frequency distribution of the first feature, which is partitioned into multiple regions by the second feature, to determine the similarity between the original data and the extracted data. That is, the data center 500 uses the two features that influence lubrication history to extract data. Therefore, when analyzing lubrication history using the extracted data, analysis results that are close to those obtained using the original data can be obtained.

[0075] <Effects of the first implementation method>

[0076] (1-1) Based on the aforementioned data center 500, it is possible to extract data from the raw data that is suitable for analyzing the lubrication history of the components in the power transmission device mounted on the vehicle 10.

[0077] (1-2) The processing circuit 510 of the data center 500 described above uses a physical quantity related to the lubrication history of the constituent components, which is different from the feature quantity contained in the original data, as a second feature quantity. In the third step, the processing circuit 510 partitions the original data into multiple data using the second feature quantity, calculates the frequency distribution of the feature quantity obtained from the original data for each partitioned data, and correspondingly partitions the extracted data into multiple data using the second feature quantity, calculates the frequency distribution of the feature quantity obtained from the extracted data for each partitioned data. For each data partitioned in the fourth step, the processing circuit 510 calculates the error between the frequency distribution of the original data and the frequency distribution of the extracted data. When the calculated error for each partition is below a threshold, the processing circuit 510 determines that the original data and the extracted data are similar.

[0078] The lubrication history of the components in the power distribution mechanism 40, which is a power transmission device, is affected by the difference in the second characteristic quantity. The data center 500, for each partition based on the second characteristic quantity, determines whether the error between the frequency distribution of the original data and the frequency distribution of the extracted data is below a threshold. Then, if the error in each partition is below the threshold, the original data and the extracted data are considered similar. Therefore, extracted data with a small error in the characteristic quantity can be extracted from any partition among the multiple partitions based on the second characteristic quantity. Thus, using the data center 500, extracted data can be extracted from the original data while also considering the influence of the second characteristic quantity on the lubrication history.

[0079] (1-3) The processing circuit 510 of the data center 500 described above uses the rotation speed of the second rotating body of the constituent component, which is lubricated by rotating within the housing 41 of the power distribution mechanism 40, which is a power transmission device, to cause the attached lubricating oil to scatter.

[0080] The higher the rotational speed of the second rotating body, the more violently the lubricating oil is dispersed from it. That is, the rotational speed of the second rotating body affects the lubrication history of the aforementioned components. The data center 500 uses the rotational speed of the second rotating body as a second feature quantity to determine the similarity between the original data and the extracted data. Therefore, the data center 500 extracts extracted data whose frequency distribution of feature quantities partitioned according to the rotational speed of the second rotating body is similar to the original data. Using the data center 500, extracted data can be extracted from the original data while also considering the influence of the rotational speed of the second rotating body on the lubrication history.

[0081] <Example of a modification to the first embodiment>

[0082] The first embodiment described above can be implemented with modifications as follows. The first embodiment and its modifications as described below can be combined with each other within a technically compatible range.

[0083] The data center 500 described above uses the rotational speed of the sun gear 45, a second rotating body within the housing 41 of the power distribution mechanism 40, as a second characteristic quantity. This rotation is used to lubricate the constituent components by causing the adhering lubricating oil to scatter. However, the second rotating body is not limited to the sun gear 45. For example, the second rotating body could also be the first electric generator 31 or the planet carrier 43.

[0084] In addition, the data center 500 can use the tilt angle of the vehicle 10 instead of the rotational speed of the second rotating body as the second characteristic quantity.

[0085] For example, the following shows an example of dividing the rotational speed of the differential gear ring 53, which is the first characteristic quantity mentioned above, into multiple zones according to the tilt angle of the vehicle 10. Figure 5 This represents the frequency distribution of the rotational speed of the differential gear ring 53 obtained from the raw data. In this frequency distribution, zero rotational speed is taken as the smallest level, and the rotational speed is divided into m levels from "1" to "m". When the vehicle 10 is divided into multiple partitions based on its tilt angle as the second characteristic quantity, the following calculation is performed for each partition of the second characteristic quantity: Figure 5 The frequency distribution of the aforementioned rotational speeds in the original data is shown. For example, the partitions can be set as three partitions: a partition where the tilt angle of vehicle 10 is positive, a partition where the tilt angle of vehicle 10 is zero, and a partition where the tilt angle of vehicle 10 is negative. When the tilt angle of vehicle 10 is positive, vehicle 10 is on an uphill slope. When the tilt angle of vehicle 10 is negative, vehicle 10 is on a downhill slope. Then, the frequency distribution of the aforementioned rotational speeds obtained from the original data is calculated for each of these three partitions. Furthermore, corresponding to the aforementioned partitions and levels, the same frequency distribution is calculated for the extracted data. Using the frequency distribution calculated in this way, similar determinations can be made between the original data and the extracted data.

[0086] If the level of the lubricating oil stored in the housing 41 of the power distribution mechanism 40 changes according to the tilt angle of the vehicle 10, the rotating body that splashes lubricating oil changes to a state of being immersed in lubricating oil. The immersion state of the rotating body affects the splashing of lubricating oil by the rotating body, and therefore the tilt angle of the vehicle 10 affects the lubrication history of the aforementioned components. The data center 500 uses the tilt angle of the vehicle 10 as a second feature quantity to determine the similarity between the original data and the extracted data. Therefore, the data center 500 extracts extracted data that is similar to the original data in terms of the frequency distribution of the first feature quantity, which is divided into multiple partitions based on the tilt angle of the vehicle 10. Using the data center 500, extracted data can be extracted from the original data while also considering the influence of the vehicle's tilt angle on the lubrication history.

[0087] Furthermore, the data center 500 can also use the temperature of the lubricating oil that lubricates the aforementioned components and the rotational speed of the oil pump as a second characteristic quantity. When the temperature of the lubricating oil increases, the viscosity of the lubricating oil decreases, and the supply to the components changes.

[0088] • When the error of each partition based on the second feature is below the threshold, the data center 500 determines that the extracted data is similar to the original data. The data center 500 may also not use all the errors of each partition in the similarity determination. Instead, it can determine that the extracted data is similar to the original data by using only one or two partitions that have a significant impact on the lubrication history of the aforementioned components, provided that the errors of these partitions are all below the threshold.

[0089] In addition, Data Center 500 calculates the sum of errors for each partition based on the second feature. When the sum of these errors is below a threshold, it can be determined that the extracted data is similar to the original data.

[0090] <Second Implementation Method>

[0091] Next, refer to Figures 1-4 as well as Figure 6 The second embodiment will be described. It should be noted that the second embodiment differs from the first embodiment in that it uses only one characteristic quantity. This characteristic quantity is the rotational speed of the oil pump that discharges lubricating oil to lubricate the aforementioned components of the power distribution mechanism 40. In the following description, the differences from the first embodiment will be primarily explained. Detailed descriptions of components that are repeated in the first embodiment will be omitted. In the second embodiment, the information processing device for extracting data is also the processing circuit 510 of the data center 500.

[0092] like Figure 4 As shown, the processing circuit 510 performs the same series of processes as in the first embodiment.

[0093] In step S100, the processing circuit 510 acquires raw data for a specific vehicle 10. The raw data includes, as a characteristic quantity, the rotational speed of the oil pump that discharges lubricating oil to lubricate the aforementioned components. The rotational speed of the oil pump is, for example, the rotational speed of the second oil pump 57.

[0094] Next, in the processing of step S110, the processing circuit 510 targets... Figure 3 As shown in (d), the original data for the rotational speed of the second oil pump 57 described above is used to determine the cut-out mode by setting multiple time windows, similar to the first embodiment. The multiple time windows are set such that the sum of the periods of all time windows is shorter than the total period of the original data.

[0095] In step S120, the processing circuit 510, similarly to the first embodiment, uses multiple time windows to cut out data to create extracted data based on the cutting-out mode determined in step S110.

[0096] In the second embodiment, since the second feature quantity is not used, the processing of step S125 is not performed.

[0097] In step S130, the processing circuit 510 calculates the frequency distribution of the aforementioned feature quantities. The processing circuit 510 uses the rotational speed of the second oil pump 57 as the feature quantity, and calculates the frequency distribution of the feature quantities obtained from the original data and the frequency distribution of the feature quantities obtained from the extracted data. Figure 6 This represents the frequency distribution of the rotational speed of the second oil pump 57 obtained from the raw data of the vehicle 10 being analyzed. In this frequency distribution, zero rotational speed is taken as the smallest level, and the rotational speeds are divided into m levels from "1" to "m". The processing circuit 510 also calculates the frequency distribution corresponding to these levels for the extracted data. Figure 6 The frequency distribution is as shown.

[0098] Next, the processing circuit 510 in Figure 4 In step S140 of the process shown, similarly to the first embodiment, the error between the frequency distribution of the feature quantity obtained from the original data and the frequency distribution of the feature quantity obtained from the extracted data is calculated. The error can be calculated, for example, using the same formula as in the first embodiment for calculating the mean absolute error (MAE), namely, mathematical formula 1. If... Figure 6 In the example shown, "n" is "m". Similarly, "i" is the number from "1" to "m". "Y" is the frequency of the feature quantity in the corresponding level of the original data. "y" is the frequency of the feature quantity in the corresponding level of the extracted data. When calculating the error using the above mathematical formula 1, the processing circuit 510 causes the processing to proceed to step S150.

[0099] In step S150, the processing circuit 510 determines whether the calculated error is below a threshold. Similar to the first embodiment, the threshold is preset such that it can determine whether to extract data with a frequency distribution close to the frequency distribution in the original data based on the error being below the threshold. If the error is determined to be below the threshold in step S150 (step S150: Yes), the processing circuit 510 records the cut-out pattern in step S160, similar to the first embodiment. On the other hand, if the error is determined to be greater than the threshold (step S150: No), the processing circuit 510 returns to step S110, similar to the first embodiment.

[0100] In this way, the processing circuit 510 changes the settings of multiple time windows and repeats the processing steps S110 to S150 until it can extract data with an error below the threshold. As a result, similar to the first embodiment, in the processing of step S160, the cut-out mode with an error below the threshold is stored in the storage device 520.

[0101] <Function of the second implementation method>

[0102] The data center 500 of the information processing device in the second embodiment extracts a portion of the raw data collected continuously over a predetermined period using multiple sensors mounted on the vehicle 10. This data is extracted data used to analyze the lubrication history of the components in the power transmission device, i.e., the power distribution mechanism 40, mounted on the vehicle 10.

[0103] Data center 500 includes processing circuit 510. The raw data includes the rotational speed of the second oil pump 57 as a feature quantity. In data center 500, processing circuit 510 performs a search process. The search process includes a first step (step S110), in which multiple time windows are set to extract a portion of the raw data, such that the sum of all time windows is shorter than the overall period of the raw data. The search process includes a second step (step S120) to extract data from the raw data using the multiple time windows. The extracted data is obtained by combining all the data extracted from the raw data using the multiple time windows. The search process includes a third step (step S130) to calculate the frequency distribution of the feature quantity obtained from the raw data and the frequency distribution of the feature quantity obtained from the extracted data. The search process includes a fourth step (steps S140 and S150), in which the error between the frequency distribution of the raw data and the frequency distribution of the extracted data is calculated, and it is determined whether the raw data and the extracted data are similar. The processing circuit 510 performs a search process that repeatedly executes steps 1 to 4 by changing the settings of multiple time windows. Then, the processing circuit 510 stores the data extraction pattern that extracts data similar to the original data in the storage device 520 (step S160).

[0104] The data center 500 uses the rotational speed of the second oil pump 57, which discharges lubricating oil from the components of the power distribution mechanism 40, as a characteristic quantity. As the pump's rotational speed increases, it discharges more lubricating oil. The data center 500 uses the frequency distribution of this characteristic quantity to determine the similarity between the raw data and the extracted data. The data center 500 extracts data whose frequency distribution of the characteristic quantity is similar to that of the characteristic quantity in the raw data. That is, the data center 500 extracts data whose rotational speed distribution of the second oil pump 57 is similar to that of the raw data. Therefore, when using the extracted data to analyze lubrication history, analysis results that are close to those obtained using the raw data can be obtained. Using the data center 500, data suitable for analyzing the lubrication history of the components in the power distribution mechanism 40 of the vehicle 10 can be extracted from the raw data.

[0105] <Effects of the second implementation method>

[0106] The second embodiment has the same effects as the first embodiment (1-1) as the following.

[0107] (2-1) Using the data center 500 described above, it is possible to extract data from the raw data that is suitable for analyzing the lubrication history of the components in the power transmission device mounted on the vehicle 10.

[0108] <Example of a modification to the second embodiment>

[0109] • Similarly to the first embodiment, the process of distinguishing feature values ​​by the second feature value can be performed, and the first feature value and the second feature value can be used to search for the cutting pattern.

[0110] <Example of Change>

[0111] The following elements are common to the modifications made in the above embodiments. The following modifications can be combined and implemented within a technically compatible framework.

[0112] In the above embodiment, an example of embodying the information processing device as a data center 500 is shown. In contrast, the information processing device can also be embodyed as an information processing terminal 600. Alternatively, the information processing device can also be embodyed as a control device for the vehicle 10.

[0113] The aforementioned data center 500 determines the similarity between the original and extracted data by calculating the error in the frequency distribution. Alternatively, the data center 500 can also determine the similarity between the original and extracted data without calculating the error. For example, statistical methods such as goodness-of-fit tests can be used to determine similarity when the difference between the original and extracted data is not significant.

Claims

1. An information processing apparatus that extracts a portion of data from raw data collected over a predetermined period using multiple sensors mounted on a vehicle, and extracts data for analyzing the lubrication history of components in a power transmission device mounted on the vehicle, wherein... The information processing device includes a processing circuit. The processing circuit changes the settings of multiple time windows and repeatedly executes the following steps 1 to 4 to extract data similar to the original data: Step 1, the multiple time windows are set such that the period obtained by summing the periods of all time windows is shorter than the predetermined period, to cut out a portion of the original data period; Step 2: Extract data from the original data using the multiple time windows; Step 3: The rotational speed of the rotating body that lubricates the constituent components by splashing the lubricating oil stored in the housing of the power transmission device through rotation is taken as a feature quantity. The frequency distribution of the feature quantity obtained from the original data and the frequency distribution of the feature quantity obtained from the extracted data are calculated. The extracted data is obtained by combining all the data cut out through the multiple time windows. And in step 4, the frequency distribution is used to determine whether the original data and the extracted data are similar.

2. An information processing apparatus that extracts a portion of data from raw data collected over a predetermined period using multiple sensors mounted on a vehicle, and extracts data for analyzing the lubrication history of components in a power transmission device mounted on the vehicle, wherein... The information processing device includes a processing circuit. The processing circuit changes the settings of multiple time windows and repeatedly executes the following steps 1 to 4 to extract data similar to the original data: Step 1, the multiple time windows are set such that the period obtained by summing the periods of all time windows is shorter than the predetermined period, to cut out a portion of the original data period; Step 2: Extract data from the original data using the multiple time windows; Step 3: The rotational speed of the oil pump that discharges lubricating oil to lubricate the constituent components is used as a feature quantity. The frequency distribution of the feature quantity obtained from the original data and the frequency distribution of the feature quantity obtained from the extracted data are calculated. The extracted data is obtained by combining all data cut out through the multiple time windows. And in step 4, the frequency distribution is used to determine whether the original data and the extracted data are similar.

3. The information processing apparatus according to claim 1 or 2, wherein, The processing circuit uses a physical quantity related to the lubrication history of the constituent component that is different from the feature quantity contained in the original data as a second feature quantity. In the third step, the original data is partitioned into multiple data using the second feature quantity. For each partitioned data, the frequency distribution of the feature quantity obtained from the original data is calculated. Corresponding to the partitioning of the multiple data of the original data, the extracted data is partitioned into multiple data using the second feature quantity. For each partitioned data, the frequency distribution of the feature quantity obtained from the extracted data is calculated. In the fourth step, for each partitioned data, the error between the frequency distribution of the original data and the frequency distribution of the extracted data is calculated. When the error in each partition is below a threshold, the original data is determined to be similar to the extracted data.

4. The information processing apparatus according to claim 3, wherein, The processing circuit uses the rotational speed of the second rotating body of the constituent component, which lubricates the component by causing the adhering lubricating oil to scatter within the housing of the power transmission device, as the second characteristic quantity.

5. The information processing apparatus according to claim 3, wherein, The processing circuit uses the vehicle's tilt angle as the second characteristic quantity.

Images