Predictive device, predictive method and predictive program

Abstract

One objective of the present disclosure is to provide a predictive device, a predictive method, and a predictive program, each capable of predicting the required replacement time of an oil check valve installed in a vehicle. The predictive device, predictive method, and predictive program according to the present disclosure serve to predict the replacement time of an oil check valve provided in an oil circuit system of a vehicle.The predictive device comprises: a sensing section that acquires first information regarding the operating history of the oil check valve and second information regarding the oil temperature profile of the oil in the oil circulation system from a storage element; and a calculation section that calculates an indicator regarding the required replacement time of the oil check valve based on the first and second pieces of information using a pre-trained classification model.

Description

Technical field

[0001] The present disclosure relates to a predictive device, a predictive method, and a predictive program. State of the art

[0002] Generally, an oil check valve (hereinafter also referred to as an "OCV") that reverses the flow path of the engine oil is located in a vehicle's oil circuit system. This type of oil check valve is located downstream of an oil cooler in the oil circuit system and is used as a diverter valve to direct the oil in the oil circuit system into an oil jet channel leading to an oil jet that cools the piston of an engine (see, for example, PTL 1). Cited documents Patent literature

[0003] PTL 1 Japanese Disclosure Document No. 2014-98345 Summary of the invention; Technical task

[0004] In this type of oil check valve, a resin material (for example, a polyamide-based resin) is used as the coil holder element (Bobin element) of a solenoid coil that actuates the oil check valve. Such resin materials generally degrade under heat.

[0005] If the resin material in the oil check valve deteriorates, oil can leak from the valve. If the damage caused by the oil leak worsens, there is a risk of oil leaking both inside and outside the vehicle, necessitating premature replacement of the oil check valve.

[0006] Currently, however, the occurrence of an oil leak from the oil check valve can only be determined by testing the actual product in a testing institute or similar, and it is difficult to predict in advance when the leak will occur.

[0007] The present invention was made with regard to the problems mentioned above. One objective of the present invention is to provide a predictive device, a predictive method, and a predictive program, each of which is capable of predicting the time at which a replacement of an oil check valve installed in a vehicle is necessary. Solution to the task

[0008] The present disclosure, which can achieve the above-described goal, includes the following: A device for predicting the required replacement time of an oil check valve provided in an oil circuit system of a vehicle, wherein the predictive device comprises the following: a data acquisition section that acquires from a storage element first information regarding the operating history of the oil check valve and second information regarding the oil temperature profile of the oil in the oil circuit system; and a calculation section that calculates an indicator regarding the required replacement time of the oil check valve based on the first and second pieces of information using a pre-trained classification model.

[0009] In another aspect, the present revelation includes the following: A method for predicting the required replacement time of an oil check valve provided in an oil circuit system of a vehicle, wherein the predictive method comprises the following: a processing step to capture initial information regarding the operating history of the oil check valve and second information regarding the oil temperature profile of the oil in the oil circuit system from a storage element; and a processing step to calculate an indicator regarding the required replacement time of the oil check valve based on the first and second pieces of information using a pre-trained classification model.

[0010] In another aspect, the present revelation includes the following: A predictive program that causes a computer to perform a prediction of the required replacement time of an oil check valve provided in an oil circuit system of a vehicle, wherein the predictive program comprises the following: a processing step to capture initial information regarding the operating history of the oil check valve and second information regarding the oil temperature profile of the oil in the oil circuit system from a storage element; and a processing step to calculate an indicator regarding the required replacement time of the oil check valve based on the first and second pieces of information using a pre-trained classification model. Advantageous effects of the invention

[0011] With the predictive device according to the present invention, it is possible to predict the required replacement time of an oil check valve installed in a vehicle. Brief description of the drawings Fig. Figure 1 shows an exemplary embodiment of a vehicle configuration; Fig. 2 shows an exemplary embodiment of the overall configuration of an oil conveying device; Fig. Figure 3 shows an example of how to configure an ECU; Fig. Figure 4 shows an example of oil temperature profile data in the oil circuit system; Fig. Figure 5 shows an example of operating history data of an oil check valve; Fig. Figure 6 shows the results of an investigation into the correlation between vehicles with and without replacement of an oil check valve and the respective parameters; Fig. Figure 7 shows the results of a market survey on the replacement frequency or non-replacement frequency (i.e., occurrence or non-occurrence of a fault) of oil check valves and the installation types in vehicles; Fig. Figure 8 shows an example of the configuration of a classification model; Fig. 9 is a flowchart showing an example of how the ECU operates; Fig. Figure 10 is a schematic representation of the flowchart from Fig. 9; Fig. Figure 11 shows an embodiment of a display screen shown on a display unit of the ECU; and Fig. Figure 12 shows an exemplary embodiment of a transition display screen for the probability of error occurrence. Description of embodiments

[0012] Preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. In this description and the drawings, components with essentially the same function are designated with the same reference numerals, and repeated descriptions are omitted. <Gesamtkonfiguration des Fahrzeugs>

[0013] The following describes the configuration of a vehicle (hereinafter referred to as "vehicle C") according to an embodiment of the present invention and a predictive device arranged in vehicle C.

[0014] The predictive device according to the present embodiment is configured by an engine control unit (ECU) which actuates an oil check valve installed in the vehicle, the predictive device predicting the required replacement time of the oil check valve. However, the predictive device does not necessarily have to be installed in the vehicle, but can, for example, also be a management computer located outside the vehicle.

[0015] Fig. Figure 1 shows an exemplary embodiment of the configuration of vehicle C.

[0016] Vehicle C, for example, is a large vehicle with the installation structure Ca. Vehicle C includes an engine Cb and an oil supply device Cbb, which supplies oil to the engine Cb for cooling the piston.

[0017] Fig. Figure 2 shows an embodiment of the overall configuration of the oil conveying device Cbb. Fig. Figure 3 shows an example of the configuration of the ECU 100.

[0018] The oil supply device Cbb includes an oil pan 2, an oil circulation system 3, an oil nozzle channel 6, an oil pump 10, an oil cooler 20, an oil filter 30, an oil check valve 40, an oil temperature sensor 50 and the ECU 100.

[0019] The oil pump 10 draws in the oil stored in the oil pan 2 and delivers it under pressure into the oil circulation system 3. The oil pump 10 is, for example, a mechanical oil pump that is driven synchronously with the rotation of the crankshaft of the engine Cb.

[0020] The oil circuit system 3 is a circulation path for the oil and circulates the oil delivered by the oil pump 10. In the oil circuit system 3, the oil filter 30, the oil cooler 20, and the oil check valve 40 are arranged sequentially downstream. The oil circuit system 3 is connected to the oil nozzle channel 6 at the position of the oil check valve 40 – the oil nozzle channel 6 is an oil channel that leads to an oil nozzle which cools the piston of the engine Cb.

[0021] The oil cooler 20 cools the oil in the oil circuit system 3 and pumps it to the oil check valve 40.

[0022] The oil check valve 40 switches the flow path of the oil coming from the oil cooler 20 between the side of the oil nozzle channel 6 and the return side of the oil circuit system 3. More precisely, the oil check valve 40 actuates an integrated solenoid valve in response to a control signal from the ECU 100 to switch the flow path of the oil in the oil circuit system 3.

[0023] Since the configuration of the oil check valve 40 corresponds to the known configuration from the prior art, its description is omitted in the present specification.

[0024] The oil check valve 40, for example, comprises a spool valve which is arranged to be axially movable back and forth in an interior space provided within a housing. The oil check valve 40 actuates the spool valve by means of the solenoid valve integrated with the spool valve and thereby switches the oil flow path in the oil circuit system 3 between the side of the oil nozzle channel 6 and the return side of the oil circuit system 3.

[0025] In the oil check valve 40, a resin material (for example, a polyamide-based resin) is used as the coil material of a coil that forms the solenoid valve. This coil material also serves as a sealing material for the oil in the oil check valve 40.

[0026] The oil check valve 40 is electrically connected to the ECU 100 and is actuated and controlled by a control current supplied by the ECU 100. In this case, when no control current is applied to the solenoid valve in the oil check valve 40, the oil check valve 40 directs the oil supplied by the oil cooler 20 to the side of the oil nozzle channel 6. Conversely, when control current is applied to the solenoid valve in the oil check valve 40, the solenoid valve is actuated to move the spool valve inside the oil check valve 40 to one side, thereby directing the oil coming from the oil cooler 20 to the return side of the oil circuit system 3. That is, the oil check valve 40 according to the present embodiment is controlled such that when current is applied (i.e., in the current-ON state), the oil nozzle is deactivated, and when current is deactivated (i.e., in the current-OFF state), the oil nozzle supply is activated.

[0027] The oil temperature sensor 50 is arranged in the oil circuit system 3 and continuously measures the oil temperature in the oil circuit system 3. In the present embodiment, the oil temperature sensor 50 is arranged downstream of the oil cooler 20 in the oil circuit system 3 and measures the temperature of the oil flowing through the oil check valve 40. The oil temperature sensor 50 is electrically connected to the ECU 100 and transmits the measured oil temperature to the ECU 100.

[0028] The ECU 100 is a computer which includes, for example, as main components a CPU 101, a ROM 102, a RAM 103, an external storage medium (e.g. a flash memory) 104, a communication unit (e.g. an Internet-connected communication module) 105, an input unit (e.g. keyboard or mouse) 106, a display unit (e.g. liquid crystal display) 107 and the like.

[0029] The functions of the ECU 100 described below are realized, for example, by the CPU 101 referencing a processing program as well as various data stored in ROM 102, RAM 103, external storage medium 104, etc. The processes executed by the CPU 101 correspond to the functions of a "communication section" and a "calculation section" according to the embodiment of the present invention.

[0030] The external storage medium 104 stores the data D1 of a previously trained classification model, oil temperature history data D2 of the oil in the oil circuit system 3, operating history data D3 of the oil check valve 40 and driving history data D4 of the vehicle C.

[0031] The oil temperature profile data D2 of the oil in the oil circuit system 3, the operating history data D3 of the oil check valve 40, and the driving profile data D4 of the vehicle C are data that are continuously stored in the external storage medium 104 while the vehicle C is in motion. The data D1 of the trained classification model, for example, is a classification model pre-trained through a learning process with adapted parameters (details of which are described below). <Grundprinzip der Vorhersage des erforderlichen Austauschzeitpunkts des Ölrückschlagventils>

[0032] The ECU 100 according to the present embodiment has a function for predicting the required replacement time of the oil check valve 40 (i.e., for predicting when the oil check valve 40 needs to be replaced). First, the basic principle of a method for predicting the replacement time of the oil check valve 40 by the ECU 100 is described.

[0033] The inventors of the present invention recognized that quantifying the thermal damage to the coil material of the oil check valve 40 is important for accurately predicting the replacement time of the oil check valve 40, and have investigated influencing factors that cause this thermal damage to the coil material. This is because the oil loss from the coil material is mainly caused by the thermal degradation of the resin material from which the coil material is made.

[0034] As a result, it was determined that the thermal damage to the coil material of the oil check valve 40 is composed of (1) thermal damage due to the ambient temperature around the coil material and (2) thermal damage due to the coil wire.

[0035] More precisely, (1) the thermal damage due to the ambient temperature around the coil material refers to thermal damage caused by the temperature increase of the surroundings due to the oil temperature in the oil circuit system 3 in which the oil check valve 40 is used. In the oil check valve 40, the coil material is located in close proximity to the oil in the oil circuit system 3, which is why the ambient temperature around the coil material is approximately equal to the oil temperature. Such an ambient temperature around the coil material causes the thermal degradation of the coil material to progress at this temperature.

[0036] The total amount of heat damage that the coil material experiences due to this ambient temperature (hereinafter referred to as "total amount of heat damage") can be quantified from the frequency of occurrence of the respective oil temperatures in the oil circuit system 3 from the start of use of the oil check valve 40 up to a reference time for assessing the thermal damage.

[0037] Fig. Figure 4 shows an example of the oil temperature profile data D2 of the oil in the oil circuit system 3. In practice, the oil temperature profile data D2 are time-resolved temperature data that are continuously recorded by the oil temperature sensor 50. Fig. Figure 4 shows data in which the temporal temperature data have been converted into frequencies of occurrence for certain oil temperatures from the beginning of the use of the oil check valve 40 (e.g. from delivery of vehicle C) until the current time.

[0038] As in Fig. As shown in Figure 4, the ECU 100, according to the present embodiment, calculates the frequency of occurrence for each oil temperature of the oil in the oil circuit system 3 from the time-resolved temperature data (oil temperature profile data D2) continuously recorded by the oil temperature sensor 50 and thus derives the total amount of heat damage that the coil material experiences due to the ambient temperature.

[0039] In this context, the rate of thermal degradation of a component generally depends on the ambient temperature of that component. This rate of degradation can be determined using the Arrhenius equation. Therefore, to more accurately calculate the total amount of heat damage to the coil material due to the ambient temperature, the ECU 100, according to the invention, converts the oil temperature into a "b. heat stress coefficient" according to the Arrhenius equation, calculates a heat stress quantity (i.e., a heat damage quantity) for each oil temperature by multiplying the "a. frequency" of the oil temperature by the "b. heat stress coefficient" (a × b), and determines the total amount of heat damage by summing the obtained values. This total amount of heat damage is then fed into the classification model described below as an explanatory variable.In the present embodiment, the oil temperature of the polyamide-based resin used as the coil material of the oil check valve 40 was converted into the “b. heat load coefficient” using the Arrhenius equation.

[0040] (2) The thermal damage caused by the coil wire is thermal damage caused by the heat generated by the solenoid coil, which is wound around and fixed to the coil material of the oil check valve 40. The oil check valve 40 is actuated each time the flow path of the oil circuit system 3 is switched. The solenoid coil in the oil check valve 40 generates heat with each operation due to the current flowing through it. This heat generated by the solenoid coil directly heats the coil material, thereby causing thermal damage to the coil material.

[0041] Therefore, the degree of thermal damage to the coil material depends on the heat generation time of the solenoid coil, that is, on the operating time of the oil check valve 40. Furthermore, the heat generation of the solenoid coil is greatest when switching from the current-OFF state to the current-ON state. Thus, the degree of heat generation of the solenoid coil is also related to the number of actuations of the oil check valve 40 (i.e., how often the oil check valve 40 has been actuated).

[0042] This means that the total amount of heat damage suffered by the coil material due to the heat generated by the coil wire can be quantified from the total operating time and the total number of actuations of the oil check valve 40 from the start of use of the oil check valve 40 up to a reference time for assessing the thermal damage.

[0043] Fig. Figure 5 shows an example of the operating history data D3 of the oil check valve 40. The operating history data D3 are time-resolved data regarding the ON / OFF state of the oil check valve 40. According to the present embodiment, the oil check valve 40 is in the ON state when the oil nozzle is switched off (e.g., at idle) and in the OFF state when the oil nozzle is activated (e.g., while the vehicle is in motion).

[0044] As in Fig. As shown in Figure 5, according to the present embodiment, the ECU 100 calculates the total operating time and the total number of actuations of the oil check valve 40 from the start of use (e.g., upon delivery of vehicle C) to the current time from the time-resolved data on the ON / OFF state of the oil check valve 40, and these values ​​are used as a reference for the total amount of heat damage to the coil material caused by the coil wire. That is, according to the present embodiment, the ECU 100 uses the actuation frequency (total operating time and total number of actuations) of the oil check valve 40 as an explanatory variable for the classification model described below.

[0045] Fig. Figure 6 shows the results of a market study on the correlation between vehicles with and without oil check valve replacement and various parameters. This study examined the extent to which it is possible to differentiate between vehicles with and without oil check valve replacement based on various parameters. In each representation of Fig. Section 6 compares two variables with different parameters and shows the correlation between the extent of each parameter and vehicles with and without oil check valve replacement. The term "damage / mileage" in Fig. 6 denotes a value obtained by dividing the total amount of heat damage due to the ambient temperature by the total distance traveled.

[0046] Out of Fig. 6 shows that the total amount of heat damage caused by the ambient temperature, the total operating time of the oil check valve and the total number of actuations of the oil check valve are strongly correlated with the replacement or non-replacement of the oil check valve (i.e. the occurrence or non-occurrence of a fault).

[0047] Furthermore, it shows Fig. 6, that the total distance traveled by the vehicle is also strongly correlated with the replacement or non-replacement of the oil check valve (i.e., the occurrence or non-occurrence of a fault).

[0048] In addition, the inventors of the present invention have found that there is also a strong correlation between the frequency of replacement or non-replacement of the oil check valve (i.e., the occurrence or non-occurrence of a fault) and the vehicle type, in particular the installation type of the vehicle.

[0049] Fig. Figure 7 shows the results of a market survey on the replacement frequency or non-replacement frequency of the oil check valve (i.e., the occurrence or non-occurrence of a fault) and the vehicle installation type. The replacement rate in Fig. Figure 7 represents the ratio of replacement to non-replacement of the oil check valve of various vehicles after a certain period of time from the start of use of the oil check valve (here: after three and a half years).

[0050] As from Fig. As can be seen in Figure 7, the oil check valve ages relatively quickly in vehicles that frequently undergo repeated starting and stopping cycles, such as refuse collection vehicles. This is because the engine in such vehicles overheats more easily, and the oil temperature in the oil circuit system 3 rises more readily due to the frequent starting and stopping. Furthermore, the heat generated by the solenoid coil of the oil check valve also increases in such vehicles as a result of the frequent starting and stopping cycles.

[0051] Furthermore, the market research revealed that the point at which oil check valve replacement is deemed necessary also varies depending on the vehicle type. For example, in refrigerated vehicles, where external oil leakage is unacceptable, the oil check valve is replaced at an early stage as soon as any deterioration of its external appearance is observed—even before any actual oil loss from the check valve is detected. This means that differentiating the replacement time based on vehicle type is essential for predicting the oil check valve replacement time and informing the user when replacement is required.

[0052] Examples of installation types in vehicles include refuse collection vehicles, refrigerated and deep-freeze vehicles, concrete mixer trucks, concrete pump trucks, tank trucks, cleaning vehicles and cab-over vehicles.

[0053] As described above, the required replacement time of the oil check valve 40 depends strongly on (1) the total amount of heat damage to the oil check valve 40 due to the ambient temperature around the coil material, and (2) the total amount of heat damage to the oil check valve 40 due to the heat generated by the coil wire. That is, by using this information as explanatory variables and by building a classification model using machine learning based on market-acquired training data, the replacement time of the oil check valve 40 can be predicted.

[0054] In addition, it is proposed in this context to further improve the prediction accuracy by adding the total distance traveled by vehicle C from the start of use of the oil check valve 40 and the installation type in vehicle C as supplementary factors to the explanatory variables.

[0055] Even though the total amount of heat damage due to the ambient temperature around the coil material, the total amount of heat damage due to the heat generated by the coil wire, and the total distance traveled by the vehicle C overlap with respect to the total amount of heat damage to the coil material, these elements can be represented as individual feature variables in the classification model generated by machine learning. Furthermore, the classification model can adjust the degree of influence of these elements on the eventual need to replace the oil check valve (i.e., the probability of failure) during the learning process.

[0056] In the present embodiment, a neural network is used as the classification model D1, which is capable of processing large amounts of data and solving nonlinear classification problems well.

[0057] Fig. Figure 8 shows an example of the configuration of the classification model D1.

[0058] Based on the aforementioned elements, the classification model D1 according to the present embodiment calculates a probability of failure of the oil check valve 40 at a predetermined reference time in order to predict the replacement time of the oil check valve 40 (for example, the time at which the failure of the oil check valve 40 occurs).

[0059] More precisely, the classification model D1 according to the present embodiment uses as explanatory variables (a) the total distance traveled by vehicle C, (b) the total operating time of the oil check valve 40, (c) the total number of actuations of the oil check valve 40, (d) the total amount of heat damage due to the ambient temperature caused by the oil temperature in the oil circuit system 3, and (e) the installation type of vehicle C. That is, the classification model according to the present embodiment has input elements in an input layer that receive these elements. Elements (a) to (d) are entered as sum values ​​from the start of use of the oil check valve 40 until the determination reference time.

[0060] Furthermore, the classification model D1 according to the present embodiment includes an output element in an output layer that outputs the probability of a fault occurring at a predetermined reference time from the start of use of the oil check valve 40. That is, the classification model D1 according to the present embodiment outputs the probability of a fault occurring at a predetermined reference time as an indicator for the required replacement time of the oil check valve.

[0061] In classification model D1 according to the present embodiment, the number of layers in the hidden layer is limited to one for the sake of computational effort reduction. The ReLU function is used as the activation function for the hidden layer, and the sigmoid function as the activation function for the output layer.

[0062] The classification model D1 was trained beforehand using training data collected from the market. The training data consists of a dataset in which the historical data of points (a) to (e), stored in a memory element or similar for each vehicle, are linked to the correct output data regarding whether or not the oil check valve has been replaced. The training dataset used was undersampled to ensure a 1:1 ratio between the number of defective and intact vehicles. A defective vehicle is defined as one in which the oil check valve has been replaced.

[0063] During the learning process, the historical data from points (a) to (e) of the training data are fed into the input layer, and the output layer outputs whether the vehicle should be classified as intact or faulty. The result is then checked for accuracy, and the network parameters (i.e., weighting coefficient and bias) are updated. Specifically, an error (i.e., a loss function) is calculated between the correct value (here: 1 or 0) and the output value. This error is propagated back from the output layer through the layers, and the network parameters (i.e., weighting coefficient and bias) are adjusted to approximate the correct value.

[0064] The data of the classification model D1, trained in this way through the learning process, are stored in the external storage medium 104 of the vehicle C. That is, in this case, the external storage medium 104 stores model data relating to the input layer, the intermediate layer, and the output layer of the neural network, as well as the network parameters adjusted by the learning process (i.e., weighting coefficients and bias). <Spezifische Verarbeitung zur Vorhersage des Austauschzeitpunkts des Ölrückschlagventils>

[0065] The specific processing by which the ECU 100 predicts the required replacement time of the oil check valve 40 according to the present embodiment is described below.

[0066] Fig. Figure 9 is a flowchart illustrating an example of the operation of the ECU 100. Fig. Figure 10 is a schematic representation of the flowchart from Fig. 9.

[0067] The basic concept of the flowchart in Fig. 9 is as follows.

[0068] Based on the oil temperature history data D2 of the oil in the oil circuit system 3, the ECU 100, according to the present embodiment, estimates the total amount of heat damage to the oil check valve 40 caused by the ambient temperature increase due to the oil temperature – from the time the oil check valve 40 is put into operation until a future determination reference time (“future determination reference time” here refers to a time for evaluating the probability of failure of the oil check valve 40; this definition applies continuously below). Additionally, based on the operating history data D3 of the oil check valve 40, the ECU 100 estimates the actuation frequency (i.e., the total amount of heat damage to the oil check valve 40) of the oil check valve 40 from the time the oil check valve 40 is put into operation until a future determination reference time.Similarly, the ECU 100 uses the vehicle C's driving history data D4 to estimate the total distance traveled by vehicle C from the time the oil check valve 40 is put into operation until a future reference time.

[0069] These estimation calculations can be carried out, for example, by using the ratio between the time span from the start of use of the oil check valve 40 until the present and the time span from the start of use of the oil check valve 40 until the future determination reference time (here: the ratio of the total distances traveled), as in Fig. Figure 10 shows the ECU 100 estimating each total value for the period from the start of use of the oil check valve 40 to the future determination reference time by extrapolating the corresponding total value from the start of use of the oil check valve 40 to the present, e.g., using a time ratio.

[0070] Fig. Figure 10 shows an example for each estimated value at a future point in time (hereinafter referred to as the “+30,000 km point”) when vehicle C has traveled an additional +30,000 km, given that vehicle C’s current total mileage is 100,000 km. In this case, each estimated value at the +30,000 km point is calculated as 1.3 times (= 130,000 km / 100,000 km) the value recorded in the timeline at the current point in time, in accordance with the ratio of the total mileages. Fig. For example, the total number of actuations of the oil check valve 40 at the +30,000 km point is estimated at 1,300 actuations (1,000 × 1.3), the total operating time of the oil check valve 40 at the +30,000 km point at 6,500 hours (5,000 × 1.3), and the total amount of heat damage due to the oil temperature at the +30,000 km point at 130,000 (100,000 × 1.3).

[0071] Subsequently, using the trained classification model D1, the ECU 100 calculates the probability of failure of the oil check valve 40 at the future determination reference time based on these elements.

[0072] Furthermore, the ECU 100 calculates the probability of failure occurring for several future points in time by projecting the reference point over time and indicates the point in time (among these points in time) at which the probability of failure exceeds a predetermined threshold (here: 50%) – this point in time is determined as the replacement point for the oil check valve 40. In the present embodiment, the total distance traveled by vehicle C is used as the reference point.

[0073] The flowchart is shown below according to Fig. 9 described in detail.

[0074] In step S1, the ECU 100 records the total distance traveled by vehicle C from the driving history data D4 stored in the external storage medium 104 and sets the total distance traveled as the input value for the classification model D1.

[0075] In step S1, the total distance traveled by vehicle C, used as input for the classification model D1, corresponds to the total distance traveled by vehicle C from the start of use of the oil check valve 40 until the future determination reference time; however, in the first loop iteration, the total distance traveled by vehicle C from the start of use of the oil check valve 40 until the current time is provisionally set as the input value.

[0076] In step S2, the ECU 100 records the total number of actuations and the total operating time of the oil check valve 40 from the operating history data D3 of the oil check valve 40, which is stored in the external storage medium 104, and sets the total number of actuations and the total operating time as input values ​​for the classification model D1. In this case, the ECU 100 determines how, with respect to Fig. 5 described, the total operating time and the total number of actuations from the start of use of the oil check valve 40 (e.g. from delivery of vehicle C) to the current time based on the time-resolved data (operating history data D3) about the ON / OFF state of the oil check valve 40.

[0077] In step S2, the total number of actuations and total operating time of the oil check valve 40, set as input values ​​for the classification model D1, correspond to the total number of actuations and total operating time of the oil check valve 40 from the start of its use until the future determination reference time. However, in the first loop iteration, the total number of actuations and total operating time of the oil check valve 40 from the start of its use until the current time are provisionally set as input values.

[0078] In step S3, the ECU 100 determines the total amount of heat damage to the oil check valve 40, caused by the oil temperature (ambient temperature) in the oil circuit system 3, from the oil temperature profile data D2 of the oil in the oil circuit system 3 stored in the external storage medium 104 and sets the total amount of heat damage as the input value for the classification model D1. As already mentioned in connection with Fig. As described in section 4, the ECU 100 calculates the frequency of occurrence for each oil temperature in the oil circuit system 3 from the time-resolved temperature data (oil temperature profile data D2) continuously recorded by the oil temperature sensor 50. The ECU 100 then calculates the amount of heat damage for each oil temperature by multiplying the “a. frequency” by the “b. heat stress coefficient”, which is obtained by applying the Arrhenius equation to the oil temperature (a × b), and sums the values ​​to determine the total amount of heat damage to the oil check valve 40.

[0079] In step S3, the total heat damage amount due to the oil temperature (ambient temperature) in oil circuit system 3, set as the input value for the classification model D1, corresponds to the total heat damage amount due to the oil temperature (ambient temperature) in oil circuit system 3 during the period from the start of use of the oil check valve 40 until the future determination reference time. However, in the first loop iteration, the total heat damage amount due to the oil temperature (ambient temperature) in oil circuit system 3 from the start of use of the oil check valve 40 until the current time is provisionally used as the input value.

[0080] In step S4, the ECU 100 detects the vehicle type C and sets this type as the input value for the trained classification model D1. For example, the user selects and enters one of the following vehicle types: refuse collection vehicle, refrigerated and deep-freeze vehicle, concrete mixer, concrete pump truck, tank truck, cleaning vehicle, or cab-over vehicle.

[0081] In step S5, the ECU 100 calculates the probability of failure of the oil check valve 40 at the determination reference time (in the first loop iteration: current time) based on the trained classification model D1 and the input values ​​set in steps S1 to S4. Specifically, the ECU 100 performs forward propagation processing of the trained classification model D1 to calculate the probability of failure of the oil check valve 40 based on the input values ​​set in steps S1 to S4.

[0082] In step S6, the ECU 100 determines whether the probability of failure of the oil check valve 40 is equal to or greater than a threshold value (for example, 50%). If the probability of failure of the oil check valve 40 is below the threshold value (S6: NO), the ECU 100 proceeds to step S7.

[0083] In step S7, the ECU 100 adds a distance traveled at a predefined interval (here: +30,000 km) to the total distance traveled by vehicle C and updates the input data of the trained classification model D1 set in steps S1 to S3.

[0084] In step S7, the ECU calculates 100, as in connection with Fig. 10 described, after adding +30,000 km to the total distance traveled by vehicle C, the total number of actuations of the oil check valve 40 at the +30,000 km point, the total operating time of the oil check valve 40 at the +30,000 km point and the total amount of heat damage due to the oil temperature at the +30,000 km point.

[0085] After step S7, the ECU 100 returns to step S1 and sets the calculated values ​​at the +30,000 km mark as new input values ​​for the trained classification model D1 (steps S1, S2, S3). Subsequently, in step S5, the ECU 100 recalculates the probability of failure of the oil check valve 40 and, in step S6, checks whether the probability of failure is equal to or greater than the threshold (e.g., 50%).

[0086] In this way, the ECU 100 repeats the loop processing of steps S1 to S7 until the failure probability of the oil check valve 40 becomes equal to or greater than the threshold (i.e., +30,000 km are added to the total mileage of vehicle C each time). If the failure probability in step S6 is equal to or greater than the threshold (e.g., +50%) (S6: YES), the ECU 100 proceeds to step S8.

[0087] In step S8, the ECU 100 displays the time at which the probability of failure of the oil check valve 40 has become equal to or greater than the threshold (in this case: the total distance traveled by vehicle C when the probability of failure of the oil check valve 40 becomes equal to or greater than the threshold) on the display unit 107 as the replacement time of the oil check valve 40.

[0088] Fig. Figure 11 shows an example of a display screen shown on the display unit 107 of the ECU 100.

[0089] The display screen according to the present embodiment comprises: an input unit m1 for the user to select the installation type of the vehicle C, an input unit m2 for selecting the data to be read out (i.e. the history data D2, D3 and D4) from the external storage medium 104, and an input unit m3 for entering the total distance traveled at the time of the last replacement of the oil check valve 40.

[0090] Furthermore, if the user enters the elements and then selects the m4 analysis execution button on the display screen, the ECU 100 executes the processing according to the flowchart in Fig. 9 off. The ECU 100 then displays the result based on the processing of the flowchart in Fig. 9 calculated replacement time of the oil check valve 40 in the result display area m5 of the display screen. [Effect]

[0091] As described above, the device for predicting the replacement time of an oil check valve according to the present embodiment comprises: a data acquisition section that acquires first information regarding the operating history of the oil check valve and second information regarding the oil temperature profile of the oil in the oil circuit system from a storage element, and a calculation section that calculates an indicator for the replacement time of the oil check valve based on the first and second pieces of information using a previously trained classification model.

[0092] The predictive device according to the present embodiment makes it possible to suggest the replacement time of an oil check valve in advance. This allows the component to be replaced before a fault occurs in the oil check valve, thus preventing a sudden failure of the oil check valve.

[0093] Furthermore, the predictive device of the present embodiment enables, in particular, the prediction of the replacement time of the oil check valve using a simple classification model based on a neural network. This is advantageous insofar as the prediction calculation can be performed without high computational effort. (Variation 1)

[0094] In the embodiment described above, one aspect was explained in which only a single point in time (the total distance traveled by vehicle C), at which the probability of failure of the oil check valve 40 is equal to or greater than the threshold, is displayed as the replacement time of the oil check valve 40. However, it may be helpful for the user to know the trend of the probability of failure of the oil check valve 40 for future points in time.

[0095] From this perspective, the ECU 100 can display the progression of the fault probability of the oil check valve 40 for future points in time on the display unit 107. For example, the ECU 100 shows the change in the fault probability of the oil check valve 40 as a function of the time elapsed since the current time.

[0096] Fig. Figure 12 shows an example of a transition display screen for the probability of error occurrence. If such a modification is implemented, it is recommended to refer to step S7 of the flowchart in Fig. The distance to be added to the total distance traveled by vehicle C should be short. This allows the interval between the plotted points of the course of the fault probability of the oil check valve 40 to be reduced for future times. (Variation 2)

[0097] In the embodiment described above, a neural network was given as an example of the classification model D1; however, a different model can also be used as classification model D1. For example, a support vector machine (SVM), a Bayesian classifier, or an ensemble model can also be used as classification model D1. Furthermore, the classification model can be formed by combining several different classifier types. (Variation 3)

[0098] In the embodiment described above, the total amount of heat damage to the oil check valve 40 caused by the ambient temperature increase resulting from the oil temperature from the start of use of the oil check valve 40 until the future reference time is estimated based on the oil temperature profile data D2 of the oil in the oil circuit system 3. This is calculated using the ratio of the time elapsed from the start of use of the oil check valve 40 to the present to the time elapsed from the start of use of the oil check valve 40 until the future reference time (in the description above: ratio of total driving distances). However, the estimation calculation can be adapted in various ways. For example, in the case of seasonal fluctuations in the frequency of use of the vehicle C or in the frequency of certain oil temperatures, a calculation formula can be used that takes such variations into account.

[0099] Although the specific embodiments of the present invention have been described in detail above, these are to be understood merely as examples and do not limit the scope of the claims. The technology described in the claims comprises various modifications and variants of the embodiments described above. Industrial applicability

[0100] With the predictive device according to one aspect of the present invention, it is possible to predict the required replacement time of an oil check valve installed in a vehicle. 2 Oil pan 3 Oil circuit system 6 Oil nozzle channel 10 Oil pump 20 oil coolers 30 oil filters 40 Oil check valve 50 Oil temperature sensor 100 ECU 102 ROM 103 RAM 104 External storage medium 105 Communication unit 106 Input unit 107 Display unit C vehicle Approx. installation CB motor Cbb oil delivery device D1 classification model D2 Oil Temperature Trend Data D3 Operating history data D4 driving history data QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] JP 2014-98345

[0003]

Claims

[1] A predictive device for predicting the required replacement time of an oil check valve provided in an oil circuit system of a vehicle, the predictive device comprising: a data acquisition section that acquires from a storage element first information regarding the operating history of the oil check valve and second information regarding the oil temperature profile of the oil in the oil circuit system; and a calculation section that calculates an indicator regarding the required replacement time of the oil check valve based on the first and second pieces of information using a pre-trained classification model. [2] The predictive device according to claim 1, wherein the calculation section estimates, based on the first information, an actuation frequency of the oil check valve from a time at which the oil check valve begins to be used until a predetermined future time, estimates, based on the second information, an occurrence frequency of the oil temperature of the oil from the time at which the oil check valve begins to be used until the predetermined future time, and calculates the indicator at the predetermined future time by inputting values ​​of the actuation frequency of the oil check valve and the occurrence frequency of the oil temperature of the oil into the trained classification model. [3] The predictive device according to claim 2 wherein the calculation section calculates the indicator for several future points in time by projecting the predetermined future point in time forward and indicates one of the future points in time at which the indicator exceeds a predetermined threshold as the required replacement time of the oil check valve. [4] The predictive device according to claim 1: the detection section additionally captures third-party information regarding the vehicle's installation type. and the calculation section calculates the indicator based on the first, second and third pieces of information using the trained classification model. [5] The predictive device according to claim 1: the acquisition section additionally acquires fourth pieces of information regarding the vehicle's driving history from the storage element. and the calculation section calculates the indicator based on the first, second and fourth pieces of information using the trained classification model. [6] The predictive device according to claim 1 wherein the first information regarding the operating history of the oil check valve includes information about a total number of actuations of the oil check valve and a total operating time of the oil check valve. [7] The predictive device according to claim 1 wherein the second information regarding the oil temperature profile of the oil includes information about a frequency of occurrence for each of the oil temperatures of the oil. [8] The predictive device according to claim 7: wherein each of the oil temperatures is converted into a heat load coefficient of a coil material in the oil check valve using an Arrhenius equation, the heat load coefficient depending on the oil temperatures and wherein the oil temperature profile is used as information about the total amount of heat damage to the coil material, the heat damage being calculated based on the frequency of occurrence for each of the oil temperatures and the heat stress coefficient. [9] The prediction device according to claim 1 wherein the trained classification model is formed by a neural network. [10] The predictive device according to claim 1 wherein the indicator is a probability of failure of the oil check valve. [11] The predictive device according to claim 1 wherein the calculation section displays a curve of the indicator of the oil check valve as a function of the time elapsed since a current time. [12] A method for predicting the required replacement time of an oil check valve provided in an oil circuit system of a vehicle, the predictive method comprising: a processing step to capture initial information regarding the operating history of the oil check valve and second information regarding the oil temperature profile of the oil in the oil circuit system from a storage element; and a processing step to calculate an indicator regarding the required replacement time of the oil check valve based on the first and second pieces of information using a pre-trained classification model. [13] A predictive program that causes a computer to perform a prediction of the required replacement time of an oil check valve provided in an oil circulation system of a vehicle, wherein the predictive program comprises: a processing step to capture initial information regarding the operating history of the oil check valve and second information regarding the oil temperature profile of the oil in the oil circuit system from a storage element; and a processing step to calculate an indicator regarding the required replacement time of the oil check valve based on the first and second pieces of information using a pre-trained classification model.

Images