ENGINE TESTING METHOD, ENGINE TESTING PROGRAM, AND ENGINE TESTING APPARATUS

Abstract

To solve the problem in an engine test in which a test pattern is created by manually adjusting an operating boundary through trial and error while detecting engine abnormality to result in an increases in number of man-hours.SOLUTION: A computer performs processes of: obtaining a first test pattern; inputting a first manipulated variable into a mathematical model based on the first test pattern to monitor a first monitored parameter as a simulation of an engine test; holding the first manipulated variable until the first monitored parameter becomes less than a first threshold when the first monitored parameter exceeds the first threshold; creating a history of the first manipulated variable in the simulation as a second test pattern; inputting a second manipulated variable into a real engine based on the second test pattern to monitor a second monitored parameter; holding the second manipulated variable until the second monitored parameter becomes less than a second threshold when the second monitored parameter exceeds the second threshold; and obtaining time series data for the second manipulated variable and a controlled variable.SELECTED DRAWING: Figure 7

Description

[Technical field]

[0001] The present invention relates to an engine testing method, an engine testing program, and an engine testing apparatus. [Background technology]

[0002] It is known that it is possible to build highly accurate engine models by using transient operation data of automobile engines. In engine tests using transient operation, the operating variables input to the engine are comprehensively changed over time, and tests are conducted under various conditions. Therefore, before the tests, the range of operating variables that will not cause the engine to enter an abnormal state is searched for. Summary of the Invention [Problem to be solved by the invention]

[0003] However, the search for the range of the manipulated variables is performed during steady-state operation, and differences in the operating boundaries may occur due to the effects of the system's dead time and time constants during transient operation. Therefore, the operating boundaries are manually adjusted through trial and error while detecting engine abnormalities such as deterioration of exhaust gas and misfires, and test patterns are repeatedly created. However, if this trial and error takes a long time, there is a problem that the overall man-hours for engine testing, including pre-test preparation, will increase.

[0004] In one aspect, an object of the present invention is to provide an engine testing method, an engine testing program, and an engine testing apparatus that can perform an engine test with fewer testing man-hours. [Means for solving the problem]

[0005] In the first proposal, a computer obtains a first test pattern in which an operating variable used in an engine test changes over time, and, as a simulation of the engine test, inputs the first operating variable based on the first test pattern into a mathematical model that uses the test pattern as an input and expresses a time-series response of the engine, and monitors at least one of an excess air ratio, an intake manifold pressure and temperature, an exhaust manifold pressure and temperature, and a maximum pressure rise rate in a cylinder, which are obtained by inputting the first operating variable into the mathematical model, as a first monitoring parameter for an engine abnormality, and when the first monitoring parameter exceeds a first threshold value, the computer monitors the first monitoring parameter. the first manipulated variable is held until the data of the first manipulated variable becomes less than a first threshold, a history of the first manipulated variable in the simulation is created as a second test pattern, at least one of an excess air ratio, an intake manifold pressure and temperature, an exhaust manifold pressure and temperature, and a maximum pressure rise rate in the cylinder, which are obtained by inputting the second manipulated variable based on the second test pattern into the real engine, is monitored as a second monitoring parameter, and when the second monitoring parameter exceeds the second threshold, the second manipulated variable is held until the second monitoring parameter becomes less than the second threshold, and a process of acquiring time series data of the second manipulated variable and the controlled amount is executed. Effect of the Invention

[0006] In one aspect, engine testing can be performed with fewer test man-hours. [Brief description of the drawings]

[0007] [Figure 1] FIG. 1 is a diagram showing an example of the configuration of an engine testing system according to this embodiment. [Diagram 2] FIG. 2 is a diagram showing an example of a Chirp signal. [Diagram 3] FIG. 3 is a diagram showing an example of a mathematical model according to the present embodiment. [Figure 4] FIG. 4 is a diagram showing another example of the mathematical model according to the present embodiment. [Diagram 5]FIG. 5 is a diagram showing an example of measurement positions of the monitoring parameters according to the present embodiment. [Figure 6] FIG. 6 is a diagram showing an example of a combination of the monitoring parameters and the manipulated variables according to the present embodiment. [Figure 7] FIG. 7 is a diagram showing an example of the determination of the monitoring parameters according to the present embodiment. [Figure 8] FIG. 8 is a flowchart showing an example of the flow of a simulation process by the engine testing device 100 according to this embodiment. [Figure 9] FIG. 9 is a flowchart showing an example of the flow of engine test processing by the information processing device 200 according to the present embodiment. [Figure 10] FIG. 10 is a diagram showing an example of a hardware configuration of an engine testing device 100 according to this embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Hereinafter, examples of the engine test method, the engine test program, and the engine test device according to the present embodiment will be described in detail with reference to the drawings. Note that the present embodiment is not limited to these examples. Also, each example can be appropriately combined within a range that does not cause a contradiction.

[0009] [Overall configuration example] The configuration of an engine testing system according to this embodiment will be described with reference to Fig. 1. Fig. 1 is a diagram showing an example of the configuration of an engine testing system according to this embodiment. As shown in Fig. 1, the engine testing system includes an engine testing device 100 and an engine 300. The engine testing device 100 and the engine 300 are connected to each other so as to be able to communicate with each other.

[0010] The engine testing device 100 may be an engine controller that controls the engine 300. The engine testing device 100 performs an engine test by inputting an operational variable for controlling the engine 300 to the engine 300 based on a test pattern. The test pattern is, for example, a Chirp signal or an Amplitude-modulated Pseudo Random Binary Sequences (APRBS) signal that indicates a time-series change in the operational variable. The test pattern is stored in a test pattern 121.

[0011] FIG. 2 is a diagram showing an example of a Chirp signal. In FIG. 2, the horizontal axis of the graph represents time, and the vertical axis represents the value of the manipulated variable. As shown in FIG. 2, the Chirp signal continuously changes the frequency components over time, making it possible to realize a highly comprehensive test of the test pattern with trigonometric function characteristics. The Chirp signal is calculated, for example, using the following formula (1).

[0012]

number

[0013] The manipulated variables are engine speed, fuel injection amount, turbine opening, EGR (Exhaust Gas Recirculation) valve opening, ITH (Intake Throttle) valve opening, etc. Therefore, test patterns such as Chirp signals are created for each manipulated variable.

[0014] In addition, since the test pattern is pattern data indicating time-series changes in the operation variables as shown in FIG. 2, the engine testing device 100 changes operation variables such as the engine speed and the fuel injection amount based on the test pattern and inputs them to the engine 300 to control the engine 300.

[0015] Furthermore, the engine testing device 100 performs a simulation of an engine test by inputting a first manipulated variable based on the test pattern into a mathematical model that uses the test pattern as an input and expresses a time-series response of the engine. Furthermore, a test pattern correction unit 114 of the engine testing device 100 corrects the test pattern based on the result of the simulation. The corrected test pattern is stored in a test pattern 121. By using the corrected test pattern for an engine test of the engine 300, it is possible to perform more accurate monitoring and control of the engine 300, and to perform the engine test with less test man-hours. Furthermore, the mathematical model is stored in a mathematical model 122.

[0016] A mathematical model according to this embodiment is illustrated. FIG. 3 is a diagram showing an example of the mathematical model according to this embodiment. The mathematical model shown in FIG. 3 is a Hammerstein-Wiener model. The Hammerstein-Wiener model expresses a system by decomposing an input / output relationship into linear and nonlinear elements. The Hammerstein-Wiener model is composed of three blocks, expresses dynamics using a linear transfer function, and gives the input and output of the linear system using a nonlinear function.

[0017] Fig. 4 is a diagram showing another example of a mathematical model according to the present embodiment. The mathematical model shown in Fig. 4 is a machine learning model using LSTM (Long Short Term Memory). The LSTM model has a structure in which each unit in the hidden layer of an RNN (Recurrent Neural Network) is replaced with a memory called an LSTM Block and three gates. This makes it possible to learn long-term dependence in addition to short-term dependence, and to control the engine in the engine test simulation with higher accuracy.

[0018] Each parameter in the LSTM model shown in FIG. 4 is calculated using the following equations (2) to (7).

[0019]

number

[0020]

number

[0021]

number

[0022]

number

[0023]

number

[0024]

number

[0025] Here, s is the sigmoid function, b is the bias, W is the input weight, U is the regression weight, and f and g are the hyperbolic tangent functions (tanh).

[0026] The engine testing device 100 inputs the operation variables based on the test pattern to the mathematical model shown in Fig. 3 or 4, performs a simulation of the engine test, and creates a history of the operation variables input in the simulation as a new test pattern. This allows the test pattern to be corrected. In the engine test simulation, the engine testing device 100 controls the operation variables based on the monitoring parameters of the engine in the simulation, which are obtained by inputting the operation variables to the mathematical model. The monitoring parameters will be described in more detail.

[0027] The data acquisition unit 111 of the engine testing device 100 acquires monitoring parameters of the engine in the simulation and the engine 300, which are obtained by inputting the operation variables to the mathematical model and the engine 300, and controlled amounts for the operation variables. The monitoring parameters are parameters that are monitored to prevent engine abnormalities from occurring. Specifically, the monitoring parameters are, for example, the excess air ratio, the pressure and temperature of the intake manifold, the pressure and temperature of the exhaust manifold, and the maximum pressure rise rate in the cylinder. The data acquisition unit 111 can store the operation variables input to the mathematical model in the operation variable history 124. The data acquisition unit 111 can also store the operation variables input to the engine 300 and the controlled amounts acquired from the engine 300 in the test data history 125 as history data of the engine test.

[0028] FIG. 5 is a diagram showing an example of the measurement position of the monitoring parameters according to the present embodiment. For example, as shown in FIG. 5, the intake manifold pressure and temperature, which are the monitoring parameters, are measured in the intake passage at the inlet of the intake manifold in the internal combustion engine. The maximum pressure rise rate in the cylinder, which is also a monitoring parameter, is measured in the cylinder of the engine 300, i.e., in the cylinder. The exhaust manifold pressure and temperature and excess air ratio, which are also monitoring parameters, are measured in the exhaust passage that is combined by the exhaust manifold in the internal combustion engine. In the case of a simulation of an engine test using a mathematical model, the monitoring parameters are obtained as output data by inputting the operation variables into the mathematical model as monitoring parameters of the engine in the simulation.

[0029] Furthermore, the first threshold determination unit 112 of the engine testing device 100 determines whether or not the monitoring parameters are within the range of the first threshold in the engine test simulation. The first threshold is an upper limit value or a lower limit value that is preset in the threshold value 123 for each monitoring parameter so that the engine in the simulation does not go into an abnormal state. Therefore, the first threshold value is not a threshold value at which the engine in the simulation goes into an abnormal state, but a threshold value for warning that the engine may go into an abnormal state if the operation variable continues to be changed based on the test pattern.

[0030] Further, the second threshold determination unit 115 of the engine testing device 100 determines whether or not the monitoring parameter is within the range of the second threshold during the engine test of the engine 300. The second threshold is an upper limit value or a lower limit value that is set in advance for the threshold value 123 for each monitoring parameter so that the engine 300 does not enter an abnormal state. Therefore, the second threshold value is not a threshold value at which the engine 300 enters an abnormal state, but a threshold value for warning that an abnormal state may occur if the manipulated variable continues to be changed based on the test pattern.

[0031] The monitoring parameters are determined in advance for each operation variable and monitored. FIG. 6 is a diagram showing an example of a combination of the monitoring parameters and the operation variables according to the present embodiment. FIG. 6 is an example, but for example, the air excess ratio of the monitoring parameters is monitored when any of the fuel injection amount, the turbine opening, the EGR valve opening, and the ITH valve opening is operated. On the other hand, the maximum pressure rise rate in the cylinder of the monitoring parameters is monitored when the fuel injection amount is operated. Strictly speaking, since the operation amount of each operation variable can affect all the monitoring parameters, when the engine testing device 100 inputs each operation variable into the mathematical model or the engine 300 and operates it, it is also possible to monitor all the monitoring parameters. However, as shown in FIG. 6, by setting in advance a combination of monitoring parameters that have a large effect on the operation variables and dividing the monitoring parameters for each operation variable, it is possible to perform more accurate monitoring and control of the mathematical model and the engine 300.

[0032] FIG. 7 is a diagram showing an example of the determination of the monitoring parameters according to the present embodiment. FIG. 7 shows the time transition of the operational variables input to the mathematical model and the engine 300, and the monitoring parameters obtained from the mathematical model and the engine 300. In FIG. 7, a first threshold and a second threshold are shown as the thresholds of the monitoring parameters in order to collectively explain the monitoring parameters obtained from the mathematical model and the engine 300. In the example of FIG. 7, the first threshold is a threshold for the first monitoring parameter obtained from the mathematical model, and the second threshold is a threshold for the second monitoring parameter obtained from the engine 300. Therefore, the first threshold and the second threshold may be different values.

[0033] As shown in FIG. 7, when the monitoring parameter exceeds the first threshold or the second threshold (at time t1), the engine testing device 100 holds the manipulated variable until the monitoring parameter becomes less than the first threshold (at time t2). In this way, the engine testing device 100 controls the engine in the simulation using the mathematical model and the engine 300 so as not to enter an abnormal state. Note that, although FIG. 7 shows that one manipulated variable is held for a monitoring parameter, multiple manipulated variables may be held for one monitoring parameter. Furthermore, the engine testing device 100 can monitor multiple monitoring parameters and hold one or multiple manipulated variables corresponding to each of the multiple monitoring parameters, and the manipulated variables to be held may be based on priorities that are preset for the multiple monitoring parameters.

[0034] 7, the lower limit values ​​of the monitoring parameters are set as the first and second threshold values, but the upper limit values ​​may be set, or both the upper and lower limit values ​​may be set. Furthermore, the settings of the upper and lower limit values ​​of the monitoring parameters may be different for each monitoring parameter.

[0035] In this way, the manipulated variable determination unit 113 of the engine testing device 100 determines the next manipulated variable within a range in which the monitoring parameters do not cause an abnormal state for the engine in the simulation using the mathematical model or the engine 300, and controls the manipulated variable. Then, the engine testing device 100 inputs the determined manipulated variables to the mathematical model or the engine 300, and controls the engine in the simulation or the engine 300. Note that the manipulated variables determined for the monitoring parameters may be based on, for example, the combination shown in Fig. 6. Furthermore, the manipulated variable determination unit 113 can store the manipulated variables input to the mathematical model and the engine 300 as history data in the manipulated variable history 124 and the test data history 125, respectively.

[0036] The engine 300 is an actual engine of an automobile. The engine 300 is operated by the operation variables input by the engine testing device 100. The operation variables input to the engine 300 are operation variables based on a history of operation variables input in a simulation of an engine test using a mathematical model, that is, a corrected test pattern. Therefore, in FIG. 1, the operation variables input to the engine 300 are the final values.

[0037] Furthermore, the engine 300 returns controlled variables and monitoring parameters for the manipulated variables, which are obtained by inputting the manipulated variables, to the engine testing device 100. Strictly speaking, rather than returning each piece of data to the engine testing device 100, each piece of data is acquired by the engine testing device 100.

[0038] [Processing flow] Next, a flow of a simulation process of an engine test performed by the engine testing device 100 will be described with reference to Fig. 8. Fig. 8 is a flowchart showing an example of a flow of a simulation process performed by the engine testing device 100 according to this embodiment. The simulation process shown in Fig. 8 is started at any timing.

[0039] First, the engine testing device 100 acquires a first test pattern stored in the test pattern 121, and starts a simulation by inputting an operational variable to a mathematical model based on the first test pattern (step S101). A processing loop is started by step S101.

[0040] Next, the engine testing device 100 acquires each monitoring parameter as a response from the mathematical model to which the manipulated variables are input in step S101 (step S102).

[0041] Next, the engine testing device 100 determines whether or not the acquired monitoring parameter is less than a first threshold for each monitoring parameter (step S103). Note that the determination in step S103 may be whether or not the monitoring parameter is equal to or less than the first threshold.

[0042] If all the monitoring parameters are less than the first threshold (step S103: Yes), the engine testing device 100 acquires the manipulated variables corresponding to the current test time from the first test pattern (step S104). The acquired manipulated variables are input to the mathematical model as the next manipulated variables. If it is within the test time, the process returns to step S102, and the process is repeated until the end of the test time. In addition, the engine testing device 100 stores the manipulated variables to be input to the mathematical model for each loop in the manipulated variable history 124 as history data of the manipulated variables.

[0043] After the test time has elapsed, the engine testing device 100 creates a modified test pattern, i.e., a second test pattern, from the history data of the manipulated variables stored in the manipulated variable history 124 (step S106). After step S106 is executed, the engine test simulation process shown in FIG. 8 ends.

[0044] On the other hand, if any of the monitoring parameters exceeds the first threshold (step S103: No), the engine testing device 100 fixes (holds) the manipulated variables at their previous values ​​(step S105). Here, the previous values ​​of the manipulated variables are, for example, the most recent manipulated variables input to the mathematical model. The manipulated variables to be held may be the manipulated variables corresponding to the monitoring parameters that exceed the first threshold, as shown in the combinations in FIG. 6, or may be all the manipulated variables.

[0045] After step S105 is executed, if it is within the test time, the process returns to step S102, the hold of the manipulated variables is released, and the process is repeated until the end of the test time. On the other hand, if the test time has ended, the engine testing device 100 executes step S106, and after executing step S106, the engine test simulation process shown in FIG. 8 ends.

[0046] Next, the flow of engine test processing by the engine testing device 100 will be described with reference to Fig. 9. Fig. 9 is a flowchart showing an example of the flow of engine test processing by the engine testing device 100 according to this embodiment. The engine test processing shown in Fig. 9 is started at any timing after the engine 300 is started.

[0047] First, the engine testing device 100 acquires from the engine 300 the current manipulated variables, i.e., the initial values ​​of the manipulated variables input to the engine 300, and the controlled variables obtained by inputting the manipulated variables (step S201). As shown in Fig. 9, after step S201, a test pattern corrected by the information processing device 200 based on the simulation result of the engine test, i.e., a second test pattern, is acquired, and a loop of engine testing and processing is started.

[0048] Next, the engine testing device 100 acquires each monitoring parameter and controlled variable obtained by inputting the manipulated variables from the engine 300 (step S202). Note that in the first step S202 immediately after the start of the loop, the controlled variables have already been acquired in step S201, so there is no need to acquire the controlled variables again. In addition, the engine testing device 100 stores the controlled variables acquired in step S201 or step S202 in the test data history 125 as history data of the engine test.

[0049] Next, the engine testing device 100 determines whether or not the acquired monitoring parameter is less than a second threshold for each monitoring parameter (step S203). Note that the determination in step S203 may be whether or not the monitoring parameter is equal to or less than the second threshold.

[0050] If all the monitoring parameters are less than the second threshold value (step S203: Yes), the engine testing device 100 acquires the operation variables corresponding to the current test time from the second test pattern (step S204). The acquired operation variables are input to the engine 300 as the next operation variables. If it is within the test time, the process returns to step S202, and the process is repeated until the end of the test time. On the other hand, if the test time has ended, the engine test process shown in FIG. 9 ends. The engine testing device 100 stores the operation variables input to the engine 300 for each loop in the test data history 125 as engine test history data.

[0051] On the other hand, if any of the monitoring parameters exceeds the second threshold (step S203: No), the engine testing device 100 fixes (holds) the manipulated variables at their previous values ​​(step S205). Here, the previous values ​​of the manipulated variables are, for example, the most recent manipulated variables input to the engine 300. Note that the manipulated variables to be held may be the manipulated variables corresponding to the monitoring parameters that exceed the second threshold, as shown in the combinations in Fig. 6, or may be all the manipulated variables.

[0052] After step S205 is executed, if it is within the test time, the process returns to step S202, the hold of the manipulated variables is released, and the process is repeated until the end of the test time. On the other hand, if the test time has ended, the engine test process shown in FIG. 9 ends.

[0053] As described above, the engine testing device 100 acquires a first test pattern in which an operating variable used in an engine test changes over time, and, as a simulation of an engine test, inputs the first operating variable based on the first test pattern into a mathematical model that uses the test pattern as an input and expresses a time-series response of the engine, and monitors at least one of the excess air ratio, the pressure and temperature of the intake manifold, the pressure and temperature of the exhaust manifold, and the maximum pressure rise rate in the cylinder, which are obtained by inputting the first operating variable into the mathematical model, as a first monitoring parameter for an engine abnormality, and when the first monitoring parameter exceeds a first threshold value, The first manipulated variable is held until the monitoring parameter becomes less than a first threshold value, a history of the first manipulated variable in the simulation is created as a second test pattern, and at least one of the excess air ratio, the intake manifold pressure and temperature, the exhaust manifold pressure and temperature, and the maximum pressure rise rate in the cylinder obtained by inputting the second manipulated variable to the engine 300 based on the second test pattern is monitored as a second monitoring parameter, and when the second monitoring parameter exceeds the second threshold value, the second manipulated variable is held until the second monitoring parameter becomes less than the second threshold value, and time series data of the second manipulated variable and the controlled quantity is obtained.

[0054] In this way, the engine testing device 100 controls the manipulated variables based on the monitoring parameters obtained by inputting the manipulated variables based on the test pattern corrected by simulating the engine test using a mathematical model to the engine 300. This eliminates trial and error when creating the test pattern, and enables the engine test to be performed with fewer man-hours.

[0055] In addition, the process of acquiring a first test pattern executed by the engine testing device 100 includes a process of acquiring, as the first test pattern, a Chirp signal or an APRBS signal indicating a time series change in a first operating variable, and the process of generating a second test pattern executed by the engine testing device 100 includes a process of generating, as the second test pattern, a Chirp signal or an APRBS signal indicating a time series change in a second operating variable.

[0056] This allows the engine testing device 100 to perform a highly comprehensive engine test.

[0057] Moreover, the engine testing device 100 sets an upper limit or a lower limit of a first monitoring parameter as a first threshold value, and sets an upper limit or a lower limit of a second monitoring parameter as a second threshold value.

[0058] This allows the engine testing device 100 to control the engine 300 with higher accuracy so as to prevent the engine 300 from going into an abnormal state.

[0059] Moreover, the engine testing device 100 sets both the upper limit value and the lower limit value of the first monitoring parameter as the first threshold value, and sets both the upper limit value and the lower limit value of the second monitoring parameter as the second threshold value.

[0060] This allows the engine testing device 100 to control the engine 300 with higher accuracy so as to prevent the engine 300 from going into an abnormal state.

[0061] In addition, the process of holding a first operating variable executed by the engine testing device 100 includes a process of holding one first operating variable for one first monitoring parameter, and the process of holding a second operating variable executed by the engine testing device 100 includes a process of holding one second operating variable for one second monitoring parameter.

[0062] This allows the engine testing device 100 to monitor and control the engine 300 with higher accuracy.

[0063] In addition, the process of holding a first operating variable executed by the engine testing device 100 includes a process of holding a plurality of first operating variables for one first monitoring parameter, and the process of holding a second operating variable executed by the engine testing device 100 includes a process of holding a plurality of second operating variables for one second monitoring parameter.

[0064] This allows the engine testing device 100 to monitor and control the engine 300 with higher accuracy.

[0065] In addition, the process of holding a first operating variable performed by the engine testing device 100 includes a process of holding the first operating variable based on a first priority for the first monitoring parameter, and the process of holding a second operating variable performed by the engine testing device 100 includes a process of holding the second operating variable based on a second priority for the second monitoring parameter.

[0066] This allows the engine testing device 100 to monitor and control the engine 300 with higher accuracy.

[0067] Furthermore, the process of inputting the first manipulated variable into the mathematical model, which is executed by the engine testing device 100, includes a process of inputting the first manipulated variable into the Hammerstein-Wiener model as the mathematical model.

[0068] This allows the engine testing device 100 to perform a more accurate simulation of an engine test.

[0069] In addition, the process of inputting a first operational variable into a mathematical model executed by the engine testing device 100 includes a process of inputting the first operational variable into a mathematical model constructed by any one of a DNN, an RNN, and an LSTM.

[0070] This allows the engine testing device 100 to perform a more accurate simulation of an engine test.

[0071] [system] The information including the processing procedures, control procedures, specific names, various data and parameters shown in the above documents and drawings may be changed arbitrarily unless otherwise specified. In addition, the specific examples, distributions, numerical values, etc. described in the embodiments are merely examples and may be changed arbitrarily.

[0072] Furthermore, the specific form of distribution or integration of the components of each device is not limited to that shown in the figure. For example, the manipulated variable determination unit 113 of the engine testing device 100 may be distributed to multiple processing units, or the first threshold determination unit 112 and the second threshold determination unit 115 of the engine testing device 100 may be integrated into one processing unit. In other words, all or some of the components may be functionally or physically distributed or integrated in any unit depending on various loads and usage conditions. Furthermore, all or any part of the processing functions of each device may be realized by a CPU and a program analyzed and executed by the CPU, or may be realized as hardware using wired logic.

[0073] [Hardware] Fig. 10 is a diagram showing an example of a hardware configuration of an engine testing device 100 according to this embodiment. As shown in Fig. 10, the engine testing device 100 includes a communication unit 100a, a storage device 100b, a memory 100c, and a processor 100d. The units shown in Fig. 10 are connected to each other via a bus or the like.

[0074] The communication unit 100a is a network interface card or the like, and communicates with other information processing devices, etc. The storage device 100b stores programs and data for operating each function of the engine testing device 100 shown in FIG.

[0075] The processor 100d reads out a program for operating each function of the engine testing device 100 shown in Fig. 1 from the storage device 100b etc. Then, the processor 100d loads the read out program into the memory 100c, thereby executing a process for realizing each function of the engine testing device 100 shown in Fig. 1.

[0076] The engine testing device 100 can also realize each function by reading a program that operates each function of the engine testing device 100 shown in Fig. 1 from a recording medium using a medium reading device and executing the program. Note that the program in this other embodiment is not limited to being executed by the engine testing device 100. For example, the present invention may be similarly applied to a case where another information processing device executes a program or a case where these execute a program in cooperation with each other.

[0077] 1 may be distributed via a network such as the Internet. The program may be recorded on a computer-readable recording medium such as a hard disk drive (HDD), a solid state drive (SSD), a flexible disk (FD), a CD-ROM, a magneto-optical disk (MO), or a digital versatile disk (DVD), and may be read out from the recording medium and executed by the computer.

[0078] The following supplementary notes are further disclosed regarding the embodiments including the above examples.

[0079] (Appendix 1) A first test pattern is obtained in which an operating variable used in an engine test changes over time; inputting a first operation variable based on the first test pattern into a mathematical model representing a time series response of an engine using a test pattern as an input, as a simulation of the engine test; monitoring at least one of an excess air ratio, an intake manifold pressure and temperature, an exhaust manifold pressure and temperature, and a maximum pressure rise rate in a cylinder, which are obtained by inputting the first operation variable into the mathematical model, as a first monitoring parameter for an engine abnormality; if the first monitoring parameter exceeds a first threshold, holding the first manipulated variable until the first monitoring parameter becomes less than the first threshold; creating a history of the first manipulated variable in the simulation as a second test pattern; monitoring at least one of the excess air ratio, the pressure and temperature of the intake manifold, the pressure and temperature of the exhaust manifold, and the maximum pressure rise rate in the cylinder, as a second monitoring parameter, which is obtained by inputting a second operating variable based on the second test pattern into the real engine; When the second monitoring parameter exceeds a second threshold, the second manipulated variable is held until the second monitoring parameter becomes less than the second threshold, and time series data of the second manipulated variable and the controlled variable is acquired. An engine testing method, characterized in that the processing is executed by a computer.

[0080] (Appendix 2) The process of acquiring the first test pattern includes a process of acquiring, as the first test pattern, a Chirp signal or an Amplitude-modulated Pseudo Random Binary Sequences (APRBS) signal indicating a time-series change of the first manipulated variable; The engine testing method according to claim 1, wherein the process of generating the second test pattern includes a process of generating, as the second test pattern, a Chirp signal or an APRBS signal indicating a time-series change in the second manipulated variable.

[0081] (Additional Note 3) As the first threshold value, an upper limit value or a lower limit value of the first monitoring parameter is set; As the second threshold, an upper limit value or a lower limit value of the second monitoring parameter is set. 2. The engine testing method according to claim 1, wherein the processing is executed by the computer.

[0082] (Additional Note 4) As the first threshold value, both an upper limit value and a lower limit value of the first monitoring parameter are set; As the second threshold, both the upper limit value and the lower limit value of the second monitoring parameter are set. 2. The engine testing method according to claim 1, wherein the processing is executed by the computer.

[0083] (Additional Note 5) The process of holding the first manipulated variable includes a process of holding one of the first manipulated variables for one of the first monitoring parameters, The engine testing method described in Appendix 1, characterized in that the process of holding the second operating variable includes a process of holding one of the second operating variables for one of the second monitoring parameters.

[0084] (Additional Note 6) The process of holding the first manipulated variable includes a process of holding a plurality of the first manipulated variables for one of the first monitoring parameters, The engine testing method described in Appendix 1, characterized in that the process of holding the second operating variable includes a process of holding a plurality of the second operating variables for one of the second monitoring parameters.

[0085] (Supplementary Note 7) The process of holding the first manipulated variable includes a process of holding the first manipulated variable based on a first priority for the first monitoring parameter, 2. The engine testing method of claim 1, wherein the process of holding the second operating variable includes a process of holding the second operating variable based on a second priority for the second monitoring parameter.

[0086] (Appendix 8) The engine testing method according to appendix 1, characterized in that the process of inputting the first operational variable into the mathematical model includes a process of inputting the first operational variable into a Hammerstein-Wiener model as the mathematical model.

[0087] (Appendix 9) The engine testing method described in Appendix 1, characterized in that the process of inputting the first operational variable into the mathematical model includes a process of inputting the first operational variable into the mathematical model constructed by any one of a Deep Neural Network (DNN), a Recurrent Neural Network (RNN), and a Long Short Term Memory (LSTM).

[0088] (Appendix 10) A first test pattern is obtained in which an operating variable used in an engine test changes along a time series; inputting a first operation variable based on the first test pattern into a mathematical model representing a time series response of an engine using a test pattern as an input, as a simulation of the engine test; monitoring at least one of an excess air ratio, an intake manifold pressure and temperature, an exhaust manifold pressure and temperature, and a maximum pressure rise rate in a cylinder, which are obtained by inputting the first operation variable into the mathematical model, as a first monitoring parameter for an engine abnormality; if the first monitoring parameter exceeds a first threshold, holding the first manipulated variable until the first monitoring parameter becomes less than the first threshold; creating a history of the first manipulated variable in the simulation as a second test pattern; monitoring at least one of the excess air ratio, the pressure and temperature of the intake manifold, the pressure and temperature of the exhaust manifold, and the maximum pressure rise rate in the cylinder, as a second monitoring parameter, which is obtained by inputting a second operating variable based on the second test pattern into the real engine; When the second monitoring parameter exceeds a second threshold, the second manipulated variable is held until the second monitoring parameter becomes less than the second threshold, and time series data of the second manipulated variable and the controlled variable is acquired. An engine testing program that causes a computer to execute processing.

[0089] (Appendix 11) The process of acquiring the first test pattern includes a process of acquiring, as the first test pattern, a Chirp signal or an Amplitude-modulated Pseudo Random Binary Sequences (APRBS) signal indicating a time-series change of the first manipulated variable; 11. The engine test program according to claim 10, wherein the process of generating the second test pattern includes a process of generating, as the second test pattern, a Chirp signal or an APRBS signal indicating a time-series change in the second manipulated variable.

[0090] (Appendix 12) As the first threshold, an upper limit value or a lower limit value of the first monitoring parameter is set; As the second threshold, an upper limit value or a lower limit value of the second monitoring parameter is set. 11. The engine testing program according to claim 10, wherein the engine testing program causes the computer to execute a process.

[0091] (Additional Note 13) As the first threshold value, both an upper limit value and a lower limit value of the first monitoring parameter are set; As the second threshold, both the upper limit value and the lower limit value of the second monitoring parameter are set. 11. The engine testing program according to claim 10, wherein the engine testing program causes the computer to execute a process.

[0092] (Additional Note 14) The process of holding the first manipulated variable includes a process of holding one of the first manipulated variables for one of the first monitoring parameters, The engine testing program according to claim 10, wherein the process of holding the second operating variable includes a process of holding one of the second operating variables for one of the second monitoring parameters.

[0093] (Additional Note 15) The process of holding the first manipulated variable includes a process of holding a plurality of the first manipulated variables for one of the first monitoring parameters, The engine testing program described in Appendix 10, characterized in that the process of holding the second operating variable includes a process of holding a plurality of the second operating variables for one of the second monitoring parameters.

[0094] (Supplementary Note 16) The process of holding the first manipulated variable includes a process of holding the first manipulated variable based on a first priority for the first monitoring parameter; 11. The engine testing program according to claim 10, wherein the process of holding the second operating variable includes a process of holding the second operating variable based on a second priority for the second monitoring parameter.

[0095] (Appendix 17) The engine testing program according to appendix 10, characterized in that the process of inputting the first operating variable into the mathematical model includes a process of inputting the first operating variable into a Hammerstein-Wiener model as the mathematical model.

[0096] (Appendix 18) The engine testing program according to appendix 10, characterized in that the process of inputting the first operational variable into the mathematical model includes a process of inputting the first operational variable into the mathematical model constructed by any one of a Deep Neural Network (DNN), a Recurrent Neural Network (RNN), and a Long Short Term Memory (LSTM).

[0097] (Appendix 19) A first test pattern is obtained in which an operating variable used in an engine test changes along a time series; inputting a first operation variable based on the first test pattern into a mathematical model representing a time series response of an engine using a test pattern as an input, as a simulation of the engine test; monitoring at least one of an excess air ratio, an intake manifold pressure and temperature, an exhaust manifold pressure and temperature, and a maximum pressure rise rate in a cylinder, which are obtained by inputting the first operation variable into the mathematical model, as a first monitoring parameter for an engine abnormality; if the first monitoring parameter exceeds a first threshold, holding the first manipulated variable until the first monitoring parameter becomes less than the first threshold; creating a history of the first manipulated variable in the simulation as a second test pattern; monitoring at least one of the excess air ratio, the pressure and temperature of the intake manifold, the pressure and temperature of the exhaust manifold, and the maximum pressure rise rate in the cylinder, as a second monitoring parameter, which is obtained by inputting a second operating variable based on the second test pattern into the real engine; When the second monitoring parameter exceeds a second threshold, the second manipulated variable is held until the second monitoring parameter becomes less than the second threshold, and time series data of the second manipulated variable and the controlled variable is acquired. An engine testing device comprising a control unit for executing processing.

[0098] (Appendix 20) The process of acquiring the first test pattern includes a process of acquiring, as the first test pattern, a Chirp signal or an Amplitude-modulated Pseudo Random Binary Sequences (APRBS) signal indicating a time-series change of the first manipulated variable; 20. The engine testing device of claim 19, wherein the process of generating the second test pattern includes a process of generating, as the second test pattern, a Chirp signal or an APRBS signal indicating a time-series change in the second manipulated variable.

[0099] (Additional Note 21) As the first threshold value, an upper limit value or a lower limit value of the first monitoring parameter is set; As the second threshold, an upper limit value or a lower limit value of the second monitoring parameter is set. 20. The engine testing device according to claim 19, wherein the control unit executes the processing.

[0100] (Additional Note 22) As the first threshold value, both an upper limit value and a lower limit value of the first monitoring parameter are set; As the second threshold, both the upper limit value and the lower limit value of the second monitoring parameter are set. 20. The engine testing device according to claim 19, wherein the control unit executes the processing.

[0101] (Supplementary Note 23) The process of holding the first manipulated variable includes a process of holding one of the first manipulated variables for one of the first monitoring parameters, 20. The engine testing apparatus of claim 19, wherein the process of holding the second operating variable includes a process of holding one of the second operating variables for one of the second monitoring parameters.

[0102] (Additional Note 24) The process of holding the first manipulated variable includes a process of holding a plurality of the first manipulated variables for one of the first monitoring parameters, 20. The engine testing apparatus of claim 19, wherein the process of holding the second operating variable includes a process of holding a plurality of the second operating variables for one of the second monitoring parameters.

[0103] (Supplementary Note 25) The process of holding the first manipulated variable includes a process of holding the first manipulated variable based on a first priority for the first monitoring parameter; 20. The engine testing apparatus of claim 19, wherein the process of holding the second operating variable includes a process of holding the second operating variable based on a second priority for the second monitoring parameter.

[0104] (Appendix 26) The engine testing apparatus of appendix 19, characterized in that the process of inputting the first operating variable into the mathematical model includes a process of inputting the first operating variable into a Hammerstein-Wiener model as the mathematical model.

[0105] (Appendix 27) The engine testing apparatus described in Appendix 19, characterized in that the process of inputting the first operational variable into the mathematical model includes a process of inputting the first operational variable into the mathematical model constructed by any one of a Deep Neural Network (DNN), a Recurrent Neural Network (RNN), and a Long Short Term Memory (LSTM).

[0106] (Appendix 28) A processor; a memory operatively connected to the processor; An engine testing apparatus comprising: obtaining a first test pattern in which an operating variable used in an engine test varies over time; inputting a first operation variable based on the first test pattern into a mathematical model representing a time series response of an engine using a test pattern as an input, as a simulation of the engine test; monitoring at least one of an excess air ratio, an intake manifold pressure and temperature, an exhaust manifold pressure and temperature, and a maximum pressure rise rate in a cylinder, which are obtained by inputting the first operation variable into the mathematical model, as a first monitoring parameter for an engine abnormality; if the first monitoring parameter exceeds a first threshold, holding the first manipulated variable until the first monitoring parameter becomes less than the first threshold; creating a history of the first manipulated variable in the simulation as a second test pattern; monitoring at least one of the excess air ratio, the pressure and temperature of the intake manifold, the pressure and temperature of the exhaust manifold, and the maximum pressure rise rate in the cylinder, as a second monitoring parameter, which is obtained by inputting a second operating variable based on the second test pattern into the real engine; When the second monitoring parameter exceeds a second threshold, the second manipulated variable is held until the second monitoring parameter becomes less than the second threshold, and time series data of the second manipulated variable and the controlled variable is acquired. An engine testing device that executes a process. [Explanation of symbols]

[0107] 100 Engine Test Equipment 100a Communications Department 100b storage device 100c Memory 100d Processor 111 Data Acquisition Section 112 First threshold value determination unit 113 Operational Variable Determination Unit 114 Test Pattern Correction Department 115 Second threshold judgment unit 121 Exam Pattern 122 Mathematical Model 123 Threshold 124 Manipulated Variable History 125 Test Data History 300 Engine

Claims

1. obtaining a first test pattern in which an operating variable used in an engine test varies in a time series; inputting a first operation variable based on the first test pattern into a mathematical model expressing a time series response of the engine using a test pattern as an input, as a simulation of the engine test; monitoring at least one of an excess air ratio, an intake manifold pressure and temperature, an exhaust manifold pressure and temperature, and a maximum pressure rise rate in a cylinder, which are obtained by inputting the first operation variable into the mathematical model, as a first monitoring parameter for an engine abnormality; if the first monitoring parameter exceeds a first threshold, holding the first manipulated variable until the first monitoring parameter becomes less than the first threshold; creating a history of the first manipulated variable in the simulation as a second test pattern; monitoring at least one of the excess air ratio, the pressure and temperature of the intake manifold, the pressure and temperature of the exhaust manifold, and the maximum pressure rise rate in the cylinder, as a second monitoring parameter, which is obtained by inputting a second operating variable based on the second test pattern into the real engine; When the second monitoring parameter exceeds a second threshold, the second manipulated variable is held until the second monitoring parameter becomes less than the second threshold, and time series data of the second manipulated variable and the controlled variable is acquired. An engine testing method, characterized in that the processing is executed by a computer.

2. The process of acquiring the first test pattern includes a process of acquiring, as the first test pattern, a Chirp signal or an Amplitude-modulated Pseudo Random Binary Sequences (APRBS) signal indicating a time-series change of the first manipulated variable; 2. The engine testing method according to claim 1, wherein the process of generating the second test pattern includes a process of generating, as the second test pattern, a Chirp signal or an APRBS signal indicating a time series change of the second manipulated variable.

3. setting an upper limit value or a lower limit value of the first monitoring parameter as the first threshold value; As the second threshold, an upper limit value or a lower limit value of the second monitoring parameter is set.

2. The engine testing method according to claim 1, wherein the processing is executed by the computer.

4. setting both an upper limit value and a lower limit value of the first monitoring parameter as the first threshold value; As the second threshold, both the upper limit value and the lower limit value of the second monitoring parameter are set.

2. The engine testing method according to claim 1, wherein the processing is executed by the computer.

5. the process of holding the first manipulated variable includes a process of holding one of the first manipulated variables for one of the first monitoring parameters; 2. The engine testing method according to claim 1, wherein the process of holding the second manipulated variable includes a process of holding one of the second manipulated variables for one of the second monitoring parameters.

6. the process of holding the first manipulated variable includes a process of holding a plurality of the first manipulated variables for one of the first monitoring parameters; 2. The engine testing method according to claim 1, wherein the process of holding the second manipulated variable includes a process of holding a plurality of the second manipulated variables for one of the second monitoring parameters.

7. the process of holding the first manipulated variable includes a process of holding the first manipulated variable based on a first priority for the first monitoring parameter; 2. The engine testing method according to claim 1, wherein the step of holding the second manipulated variable includes a step of holding the second manipulated variable based on a second priority for the second monitored parameter.

8. 2. The engine testing method according to claim 1, wherein the process of inputting the first operational variable into the mathematical model includes a process of inputting the first operational variable into a Hammerstein-Wiener model as the mathematical model.

9. 2. The engine testing method according to claim 1, wherein the process of inputting the first operational variable to the mathematical model includes a process of inputting the first operational variable to the mathematical model constructed by any one of a Deep Neural Network (DNN), a Recurrent Neural Network (RNN), and a Long Short Term Memory (LSTM).

10. obtaining a first test pattern in which an operating variable used in an engine test varies in a time series; inputting a first operation variable based on the first test pattern into a mathematical model expressing a time series response of the engine using a test pattern as an input, as a simulation of the engine test; monitoring at least one of an excess air ratio, an intake manifold pressure and temperature, an exhaust manifold pressure and temperature, and a maximum pressure rise rate in a cylinder, which are obtained by inputting the first operation variable into the mathematical model, as a first monitoring parameter for an engine abnormality; if the first monitoring parameter exceeds a first threshold, holding the first manipulated variable until the first monitoring parameter becomes less than the first threshold; creating a history of the first manipulated variable in the simulation as a second test pattern; monitoring at least one of the excess air ratio, the pressure and temperature of the intake manifold, the pressure and temperature of the exhaust manifold, and the maximum pressure rise rate in the cylinder, as a second monitoring parameter, which is obtained by inputting a second operating variable based on the second test pattern into the real engine; When the second monitoring parameter exceeds a second threshold, the second manipulated variable is held until the second monitoring parameter becomes less than the second threshold, and time series data of the second manipulated variable and the controlled variable is acquired. An engine testing program that causes a computer to execute processing.

11. obtaining a first test pattern in which an operating variable used in an engine test varies in a time series; inputting a first operation variable based on the first test pattern into a mathematical model expressing a time series response of the engine using a test pattern as an input, as a simulation of the engine test; monitoring at least one of an excess air ratio, an intake manifold pressure and temperature, an exhaust manifold pressure and temperature, and a maximum pressure rise rate in a cylinder, which are obtained by inputting the first operation variable into the mathematical model, as a first monitoring parameter for an engine abnormality; if the first monitoring parameter exceeds a first threshold, holding the first manipulated variable until the first monitoring parameter becomes less than the first threshold; creating a history of the first manipulated variable in the simulation as a second test pattern; monitoring at least one of the excess air ratio, the pressure and temperature of the intake manifold, the pressure and temperature of the exhaust manifold, and the maximum pressure rise rate in the cylinder, as a second monitoring parameter, which is obtained by inputting a second operating variable based on the second test pattern into the real engine; When the second monitoring parameter exceeds a second threshold, the second manipulated variable is held until the second monitoring parameter becomes less than the second threshold, and time series data of the second manipulated variable and the controlled variable is acquired. An engine testing device comprising a control unit for executing processing.

Images