Anomaly detection system for electric pump systems

Abstract

The goal is to estimate signs of abnormality and identify the component showing signs of malfunction. [Solution] When it is estimated that there is a sign of abnormality in the electric pump system, if the first parameter is the upper or lower limit of the first normal range when the electric pump is normal, the notification device is controlled to determine that there is a sign of abnormality in the electric pump and to notify that there is a sign of abnormality in the electric pump. If the second parameter is the upper or lower limit of the second normal range when the battery is normal, the notification device is controlled to determine that there is a sign of abnormality in the power supply from the battery and to notify that there is a sign of abnormality in the power supply. If the third parameter is the upper or lower limit of the third normal range when the liquid state is normal, the notification device is controlled to determine that there is a sign of abnormality in the liquid and to notify that there is a sign of abnormality in the liquid.

Description

Technical Field

[0006] , , ,

[0005] , ,

[0001] The present disclosure relates to an abnormality estimation system for an electric pump system.

Background Art

[0002] Conventionally, as an abnormality estimation system for this type of electric pump system, there has been proposed one used for an electric pump system including an electric pump for pumping a liquid, which estimates an abnormality of the electric pump (see, for example, Patent Document 1). In this system, an abnormality of the electric pump is determined based on a determination value calculated from the current and voltage of the electric pump.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the above-described electric pump system, it has been recognized as an important problem to estimate signs of an abnormality before the abnormality occurs and identify components where signs of the abnormality are observed.

[0005] The main object of the electric pump system of the present disclosure is to estimate signs of an abnormality and identify components where signs of the abnormality are observed.

Means for Solving the Problems

[0006] The electric pump system of this disclosure employs the following means to achieve the above-mentioned main objective. The electric pump system of this disclosure is used in an electric pump system comprising an electric pump for pumping liquid and a battery for supplying power to the electric pump, and is an electric pump system abnormality estimation system for estimating signs of abnormality in the electric pump, comprising: a notification device for notifying information; a trained model obtained by machine learning, which takes as input variables a first parameter that reflects the discrepancy between the actual driving state of the electric pump and the driving request of the electric pump, a second parameter that reflects the power supplied from the battery to the electric pump, and a third parameter that reflects the state of the liquid, and as an output variable the presence or absence of signs of abnormality in the electric pump system; and estimates the presence or absence of signs of abnormality in the electric pump system based on the first, second, and third parameters, and when it is estimated that there are signs of abnormality in the electric pump system, The gist of this invention is a determination control device which controls the notification device to determine that there is a sign of abnormality in the electric pump and to notify the notification device when the first parameter is the upper or lower limit of the first normal range when the electric pump is functioning normally, determines that there is a sign of abnormality in the electric pump, and to notify the notification device that there is a sign of abnormality in the electric pump, when the second parameter is the upper or lower limit of the second normal range when the battery is functioning normally, determines that there is a sign of abnormality in the power supply from the battery, and to notify the notification device that there is a sign of abnormality in the power supply, when the third parameter is the upper or lower limit of the third normal range when the liquid is functioning normally, determines that there is a sign of abnormality in the liquid, and to notify the notification device that there is a sign of abnormality in the liquid. Here, "when each parameter (the first, second, and third parameters) is at the upper or lower limit of the normal range" includes not only when each parameter is at the upper or lower limit of the normal range, but also when it is close to the upper or lower limit of the normal range. [Brief explanation of the drawing]

[0007] [Figure 1] This is a schematic diagram of the drive system 20 equipped with the anomaly estimation system of the embodiment of the present disclosure. [Figure 2] This is an explanatory diagram illustrating how to create a pre-trained model. [Figure 3] This flowchart shows an example of a decision routine executed by the control device 40. [Modes for carrying out the invention]

[0008] Embodiments of this disclosure will be described with reference to the drawings. Figure 1 is a schematic diagram of a drive system 20 equipped with an abnormality estimation system according to an embodiment of this disclosure. The drive system 20 is mounted on an automobile equipped with a motor, engine, etc., and as shown in Figure 1, comprises a drive unit 22, a lubrication and cooling unit 30 as an electric pump system, a battery 34, and a control unit 40.

[0009] The drive unit 22 includes a power source that outputs power for driving, such as a motor or engine, and an automatic transmission that changes the speed of the power from the power source and transmits it to the axle connected to the drive shaft, which acts as an output component.

[0010] The lubrication and cooling device 30 is configured to cool and lubricate the drive unit 22 using hydraulic fluid. The lubrication and cooling device 30 includes an electric oil pump (electric pump) 32, a battery 34, and a warning light 38.

[0011] The electric oil pump 32 is powered by the battery 34 and pumps the hydraulic fluid stored in the oil pan 35 into the supply channel 36. The supply channel 36 supplies the hydraulic fluid from the electric oil pump 32 to the drive unit 22. The hydraulic fluid supplied to the drive unit 22 cools and lubricates the drive unit 22 and is then discharged back into the oil pan 35 through the return channel 37. The battery 34 is configured as a secondary battery. The warning light 38 is installed in the passenger compartment of the vehicle. The electric oil pump 32 and the warning light 38 are controlled by the control device 40.

[0012] The control device 40 is configured as a microprocessor centered on a CPU, and in addition to the CPU, it includes a ROM for storing processing programs, a RAM for temporarily storing data, a flash memory for storing data, and input / output ports (not shown). The control device 40 receives inputs via its input ports, including the pump rotation speed Np from a rotation speed sensor 32f for detecting the rotation speed of the electric oil pump 32, the oil temperature Toil from a temperature sensor 32t for detecting the temperature of the hydraulic oil circulating in the supply channel 36, the battery voltage Vb from a voltage sensor 34v for detecting the terminal voltage of the battery 34, and the battery current Ib from a current sensor 34a for detecting the current of the battery 34. The control device 40 outputs drive signals to the drive unit 22 and control signals to the warning light 38 via its output ports.

[0013] In the drive system 20 of this embodiment, the indicated rotational speed Np* is set based on the oil temperature Toil, the torque command Tp* for the electric oil pump 32 is set so that the pump rotational speed Np becomes the indicated rotational speed Np*, and the electric oil pump 32 is duty-controlled so that it drives the torque command Tp*. Through this control, the hydraulic fluid stored in the oil pan 35 is drawn up using the electric oil pump 32 and supplied to the supply passage 36, and the hydraulic fluid supplied to the drive unit 22 lubricates and cools the components that make up the drive unit 22, such as planetary gears and friction engagement elements. After lubricating and cooling the drive unit 22, the hydraulic fluid returns to the oil pan 35 through the return passage 37. In other words, the hydraulic fluid circulates in the order of oil pan 35, electric oil pump 32, drive unit 22, and oil pan 35.

[0014] In the drive system 20, the control device 40 stores in ROM a model trained by a neural network (machine learning) that uses a pump drive parameter (first parameter) P1, a power parameter (second parameter) P2, and an oil state parameter (third parameter) P3 as input variables (input to the input layer), and an abnormality prediction flag F indicating the presence or absence of abnormalities in the lubrication cooling device 30 as an output variable (output to the output layer). The pump drive parameter P1 is a parameter that reflects the discrepancy between the actual drive state of the electric oil pump 32 and the drive request of the electric oil pump 32. In the embodiment, the absolute value of the difference between the instructed rotation speed Np* and the pump rotation speed Np is used as the pump drive parameter P1. The power parameter P2 is a parameter that reflects the power supplied from the battery 34 to the electric oil pump 32. In the embodiment, the drive current of the electric oil pump 32 (battery current Ib detected by the current sensor 34a) is used as the power parameter P2. The oil state parameter P3 is a parameter that reflects the state of the hydraulic fluid. In this embodiment, the oil temperature Toil detected by the temperature sensor 32t is defined as the oil condition parameter P3. The abnormality warning flag F is set to a value of 1 when abnormal noise or a decrease in cooling performance indicating an abnormality in the electric pump system is detected, and is set to a value of 0 when no abnormal noise or decrease in cooling performance indicating an abnormality in the electric pump system is detected.

[0015] Figure 2 is an explanatory diagram illustrating how a trained model is created. In this machine learning process, the tester drives the vehicle multiple times on roads, test courses, and chassis dynamometers. The control device 40 acquires a dataset containing the pump drive parameter P1, power parameter P2, oil state parameter P3, and an anomaly prediction flag F multiple times at predetermined intervals while the vehicle is running. Once the control device 40 has acquired the number of datasets necessary to construct a highly accurate neural network, it creates a trained model using a neural network that takes the pump drive parameter P1, power parameter P2, and oil state parameter P3 as inputs to the input layer and the anomaly prediction flag F as an output from the output layer, as shown in Figure 2, and stores it in ROM. The reason for inputting the pump drive parameter P1 to the input layer is that even if the deviation between the actual drive state of the electric oil pump 32 and the drive request of the electric oil pump 32 is within the normal range (first normal range), if it is close to the upper or lower limit of the normal range, there is a high probability that an abnormality will occur in the electric oil pump 32, and a precursor to an abnormality in the electric pump system can be estimated. The reason for inputting the power parameter P2 to the input layer is that even if the drive current of the electric oil pump 32 is within the normal range (second normal range), if it is close to the upper or lower limit of the normal range, a precursor to an abnormality in the electric pump system can be estimated. The reason for inputting the oil state parameter P3 to the input layer is that even if the oil temperature Toil is within the normal range (third normal range), if it is close to the upper or lower limit of the normal range, a precursor to an abnormality in the electric pump system can be estimated. Therefore, using the trained model created in this way, a precursor to an abnormality in the electric pump system can be accurately predicted.

[0016] Next, we will describe the operation of a vehicle equipped with the abnormality estimation system of the embodiment configured in this way, particularly the operation when estimating signs of an abnormality in the electric pump system. Figure 3 is a flowchart of an example of a determination routine executed by the control device 40. This routine is executed repeatedly at predetermined intervals while the vehicle is running.

[0017] When this routine is executed, the CPU of the control device 40 inputs the current pump drive parameter P1, power parameter P2, and oil state parameter P3 (S100). Then, using the trained model stored in ROM beforehand, along with the pump drive parameter P1, power parameter P2, and oil state parameter P3, it sets the abnormality prediction flag F (S110). In this way, the CPU of the control device 40 sets the abnormality prediction flag F using a trained model created beforehand through machine learning, so the abnormality prediction flag F accurately reflects the signs of an abnormality occurring in the electric pump system.

[0018] Next, the CPU of the control device 40 determines whether the abnormality prediction flag F is valued at 1 (S120). If the abnormality prediction flag F is valued at 0, the CPU of the control device 40 determines that there is no indication of an abnormality occurring in the electric pump system and terminates this routine.

[0019] The CPU of the control device 40 determines in S120 that there is a sign of an impending malfunction in the electric pump system when the abnormality warning flag F is valued at 1. Based on the pump drive parameter P1, the power parameter P2, and the oil condition parameter P3, it identifies the cause of the abnormality warning flag F being valued at 1 and controls the warning light 38 to illuminate (S130), then terminates this routine. The identification of the cause of the abnormality warning flag F being valued at 1 is performed by determining whether the pump drive parameter P1, the power parameter P2, and the oil condition parameter P3 are at or close to the upper or lower limits of the normal range. When the pump drive parameter P1 is at or close to the upper or lower limit of the normal range, the CPU of the control device 40 identifies that the cause is a sign of an abnormality in the electric oil pump 32. Note that "the pump drive parameter P1 is at or close to the upper or lower limit of the normal range" also includes cases where the value is close to the upper or lower limit. The CPU of the control device 40 identifies that if the power parameter P2 is at the upper or lower limit of the normal range, the cause is an indication of an abnormality in the power supply from the battery 34. Note that "the power parameter P2 is at the upper or lower limit of the normal range" also includes cases where the value is close to the upper or lower limit. The CPU of the control device 40 identifies that if the oil condition parameter P3 is at or close to the upper or lower limit of the normal range, the cause is an indication of an abnormality in the condition of the hydraulic fluid. Note that "the oil condition parameter P3 is at the upper or lower limit of the normal range" also includes cases where the value is close to the upper or lower limit. By identifying the cause in this way, subsequent maintenance can be performed appropriately. In addition, by illuminating the warning light 38, the occupants can be notified of the signs of an abnormality in the electric pump system. Through these processes, the signs of an abnormality in the electric pump system are estimated and the parts showing signs of abnormality are identified.

[0020] In the above-described embodiment, the absolute value of the difference between the indicated rotational speed Np* and the pump rotational speed Np is used as the pump drive parameter P1. However, instead of the absolute value of the difference between the indicated rotational speed Np* and the pump rotational speed Np, or together with the absolute value of the difference between the indicated rotational speed Np* and the pump rotational speed Np, the absolute value of the difference between the torque command Tp* of the electric oil pump 32 and the actual driving torque may be used as the pump drive parameter P1. By doing so, it is possible to reflect in the abnormality prediction flag F the signs of abnormality of the electric pump system in which the absolute value of the difference between the torque command Tp* and the actual driving torque becomes large even though the difference between the indicated rotational speed Np* and the pump rotational speed Np is small. Thereby, it is possible to further improve the estimation accuracy of the signs of abnormality of the electric pump system.

[0021] In the above-described embodiment, the drive current of the electric oil pump 32 is used as the power parameter P2. However, instead of the drive current of the electric oil pump 32, or together with the drive current of the electric oil pump 32, the duty value when duty-controlling the electric oil pump 32 may be used as the power parameter P2. By doing so, it is possible to more appropriately reflect the signs of abnormality of the electric pump system in the abnormality prediction flag F. Thereby, it is possible to further improve the estimation accuracy of the signs of abnormality of the electric pump system.

[0022] In the above-described embodiment, the oil temperature Toil detected by the temperature sensor 32t is used as the oil state parameter P3. However, instead of the oil temperature Toil, or together with the oil temperature Toil, the estimated amount of foreign matter contained in the hydraulic oil may be used as the oil state parameter P3. By doing so, it is possible to reflect in the abnormality prediction flag F the factors leading to the abnormality of the electric oil pump 32. Thereby, it is possible to further improve the estimation accuracy of the signs of abnormality of the electric pump system.

[0023] In the above-described embodiment, as the electric pump system, an example is given of one including an electric oil pump 32 that pumps hydraulic oil. However, the electric pump system is not limited to one including the electric oil pump 32, and it may include an electric pump that pumps a liquid, such as an electric water pump that pumps cooling water into a flow path.

[0024] In the above-described embodiment, a warning lamp 38 is provided, and the warning lamp 38 is lit in S130. However, instead of the warning lamp 38 or together with the warning lamp 38, a device that notifies information, such as a speaker or a display, may be provided, and the information may be notified using these devices in S130.

[0025] The correspondence between the main elements of the embodiment and the main elements of the invention described in the column of means for solving the problems will be described. In the embodiment, the warning lamp 38 corresponds to the "notification device", and the control device 40 corresponds to the "determination control device".

[0026] Note that the correspondence between the main elements of the embodiment and the main elements of the invention described in the column of means for solving the problems is an example for specifically explaining the form for implementing the invention described in the column of means for solving the problems in the embodiment, and thus does not limit the elements of the invention described in the column of means for solving the problems. That is, the interpretation of the invention described in the column of means for solving the problems should be made based on the description in that column, and the embodiment is merely a specific example of the invention described in the column of means for solving the problems.

[0027] As described above, the embodiments for implementing the present disclosure have been described using the embodiments. However, the present disclosure is not limited to such embodiments, and it is of course possible to implement it in various forms without departing from the gist of the present disclosure.

Industrial Applicability

[0028] The present disclosure can be used in the manufacturing industry of an abnormal estimation system for an electric pump system and the like. [Explanation of Symbols]

[0029] 38 warning lights, 40 control devices.

Claims

[Claim 1] An abnormality estimation system for an electric pump system, which includes an electric pump for pumping liquid and a battery for supplying power to the electric pump, is used to estimate signs of abnormality in the electric pump, A notification device that broadcasts information, A trained model obtained by machine learning takes as input variables a first parameter that reflects the discrepancy between the actual driving state of the electric pump and the driving request of the electric pump, a second parameter that reflects the power supplied from the battery to the electric pump, and a third parameter that reflects the state of the liquid, and as an output variable the presence or absence of signs of abnormality in the electric pump system, and estimates whether or not there are signs of abnormality in the electric pump system based on the first, second, and third parameters, and when it is estimated that there are signs of abnormality in the electric pump system, if the first parameter is the upper or lower limit of the first normal range when the electric pump is normal, the electric A determination control device that determines if there are signs of abnormality in the power pump and controls the notification device to notify that there are signs of abnormality in the power pump; when the second parameter is at the upper or lower limit of the second normal range when the battery is normal, it determines if there are signs of abnormality in the power supply from the battery and controls the notification device to notify that there are signs of abnormality in the power supply; when the third parameter is at the upper or lower limit of the third normal range when the liquid is in a normal state, it determines if there are signs of abnormality in the liquid and controls the notification device to notify that there are signs of abnormality in the liquid; An anomaly detection system for an electric pump system equipped with the following features.

Images