Oil level monitoring system

Abstract

The invention provides an oil level monitoring system. A method of monitoring a fluid level in a power unit relies on a neural network model. The model is used to predict lateral acceleration during an air intake event. These predictions are compared with measurements of lateral acceleration. When the measured and predicted values are not well correlated, the controller takes a corrective action.

Description

Technical Field

[0001] The present disclosure relates to a method for monitoring fluid level in a vehicle power unit. More specifically, the present disclosure relates to a method based on machine learning. Background Art

[0002] In a vehicle's power unit, fluid is actively circulated to provide cooling and lubrication. An electric pump is used to circulate the fluid from the oil sump to various outlets strategically located throughout the power unit. In some cases, lateral acceleration can move the fluid in the oil sump so that the fluid inlet is exposed to air instead of fluid. This is more common if the fluid level is low due to a leak. Measuring the fluid level directly requires additional parts whose reliability may not be perfect. Summary of the invention

[0003] A vehicle includes a power unit, an electric pump, and a controller. The power unit has an oil pan containing a fluid. The power unit may be an electric drive unit including a motor and a gearbox. The electric pump is configured to pump the fluid from an inlet in the oil pan to a plurality of outlets. The controller is programmed to detect changes in fluid level and respond accordingly. The controller measures lateral acceleration. The controller also uses a model to predict lateral acceleration based on power consumption of the electric pump. The controller may only consider lateral acceleration in one direction. The model may be a neural network model that has been trained using experimental data obtained using a predetermined amount of fluid. The model may also use measured vehicle speed. In response to a correlation measure between the measured lateral acceleration and the predicted lateral acceleration being less than a threshold, the controller restricts flow to a subset of the plurality of outlets to ensure that flow is maintained to a higher priority outlet. The controller may also send a signal to an operator.

[0004] A power unit for a vehicle includes a housing, an electric pump, and a controller. The housing defines an oil pan containing a fluid. The electric pump is configured to pump the fluid from an inlet in the oil pan to a plurality of outlets. The controller is programmed to detect changes in fluid level and respond accordingly. The controller measures lateral acceleration, vehicle speed, and power consumption of the electric pump. The controller then uses a model to predict lateral acceleration based on the power consumption and the vehicle speed. The controller may only consider lateral acceleration in one direction. The model may be a neural network model trained using experimental data, the experimental data being obtained using a predetermined amount of fluid. In response to a correlation measure between the measured lateral acceleration and the predicted lateral acceleration being less than a threshold, the controller restricts flow to a subset of the plurality of outlets to ensure that flow is maintained to a higher priority outlet. The controller may also send a signal to an operator. The power unit may also include a traction motor and a gearbox.

[0005] A method for detecting abnormal fluid level includes comparing predicted lateral acceleration with measured lateral acceleration. Fluid is pumped from an oil pan to multiple outlets using an electric pump. Power consumption and lateral acceleration of the electric pump are measured. Air induction events are identified based on changes in the power consumption. During the air induction event, a model is used to calculate a predicted lateral acceleration based on the measured power consumption. The model can be a neural network model. The model can also use measured vehicle speed. In response to a correlation measure between the measured lateral acceleration and the predicted lateral acceleration being less than a threshold, flow is restricted to a subset of the multiple outlets to ensure that flow is maintained to a higher priority outlet. A signal can also be sent to an operator. The method can include training the model using experimental data obtained using a predetermined amount of fluid in the oil pan. The method can only consider lateral acceleration in one direction. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Figure 1 A block diagram of a hybrid electric vehicle is shown.

[0007] Figure 2 Is suitable for Figure 1 A side cross-section of a power unit in a vehicle.

[0008] Figure 3A , Figure 3B and Figure 3C yes Figure 2 End views of the power unit in various conditions.

[0009] Figure 4 is Figure 2 Graphs of several parameters during an air induction event in a power unit.

[0010] Figure 5 It is used to build models to predict Figure 1 Flowchart of the lateral acceleration in a vehicle.

[0011] Figure 6 Is to use Figure 5 Models are constructed and monitored in response to fluid loss Figure 1 A flow chart of a method of measuring a fluid level in a vehicle. DETAILED DESCRIPTION

[0012] Embodiments are described herein. However, it should be understood that the disclosed embodiments are merely examples and that other embodiments may take various and alternative forms. The drawings are not necessarily drawn to scale. Some features may be enlarged or minimized to show details of particular components. Therefore, the specific structural details and functional details disclosed herein should not be interpreted as limiting, but merely as a representative basis for teaching those skilled in the art.

[0013] The various features shown and described with reference to any one of the accompanying drawings may be combined with features shown in one or more other drawings to produce embodiments not explicitly shown or described. The combinations of features shown provide representative embodiments for typical applications. However, for specific applications or implementations, various combinations and modifications of features consistent with the teachings of the present disclosure may be desired.

[0014] Figure 1 A vehicle powertrain configuration of a hybrid vehicle 10 is shown. Power is provided by an internal combustion engine 12 and an electric traction motor 14. At a given time, power may be provided by the engine 12, by the traction motor 14, or by a combination of the two. The engine 12 drives the front wheels 16 and 18 via a transaxle 20. The traction motor 14 drives the rear wheels 22 and 24 via a reduction gear arrangement 26 and a rear differential 28. The traction motor 14 and the reduction gear arrangement may be housed in a common housing 30 and are collectively referred to as a power unit. The power unit may also include a differential. The combination of an internal combustion engine and a transaxle may also be referred to as a power unit. The vehicle may have only a single power unit, or may have multiple power units of the same type.

[0015] Figure 2 3 is a side view of the power unit. A housing 30 defines an oil sump that contains a fluid 32. An electric pump 34 circulates the fluid by drawing the fluid from an inlet 36 in the oil sump and distributing the fluid to a set of outlets 38. The outlets are positioned so that the fluid provides cooling and lubrication to components of the traction motor 14 and the reduction gear arrangement 26. The fluid then drains back into the oil sump due to gravity.

[0016] Figure 3A is an end view of the power pack with the fluid filled at nominal level and the power pack not subject to any lateral acceleration. It should be noted that the inlet 36 is not necessarily centered. Figure 3B The case where the power unit is subjected to a lateral acceleration to the right is shown. The lateral acceleration may be caused by the vehicle turning a corner, the vehicle driving on a banked road, or some combination of the two. In this case, air may be drawn into the inlet 36. Due to the offset position of the inlet 36, a lateral acceleration to the left will not cause air ingestion. When the fluid volume is at the nominal level, air ingestion events are very rare and short in duration, so that brief interruptions in cooling or lubrication flow are not a problem. Figure 3CA situation is shown where the fluid level has dropped from the nominal level. This could be due to a slow leak, for example. With lower fluid levels, air ingestion occurs at lower levels of right lateral acceleration. As a result, air ingestion events occur more frequently and tend to last longer, increasing the likelihood of cooling or lubrication interruptions.

[0017] Figure 4 An air ingestion event is graphically shown as a function of time. Since the density of air is negligible relative to the fluid, air ingestion causes a reduction in the electrical power required by the electric pump. It should be noted that a lateral acceleration to the left (indicated by a negative value) does not cause air ingestion. When the lateral acceleration exceeds a certain value, the pump power begins to drop. It continues to drop as the fluid currently in the line is pumped through the line. When the lateral acceleration decreases, the pump power returns to its normal level. The inventors have discovered that vehicle speed affects the relationship between lateral acceleration and pump power.

[0018] A machine learning model, such as a neural network model, may be created that predicts lateral acceleration during an air ingestion event based on pump power and vehicle speed. The cause and effect relationships of the neural network model may be the opposite of the cause and effect relationships of the physical system. The inventors have found that such a neural network model accurately predicts lateral acceleration when trained using data obtained at a specific fluid level. The inventors have also found that when the fluid level changes enough to require corrective action, the predicted lateral acceleration does not correlate well with the observed lateral acceleration. Figure 5 and Figure 6 A method for monitoring fluid level in a power unit of a vehicle based on these findings is presented.

[0019] Figure 5 is a flow chart of a method for creating a neural network model. The method may be performed when a vehicle is first placed in service. Alternatively, it may be performed on a vehicle representative of a vehicle in which fluid level monitoring will be performed. At 40, the oil pan is filled to a nominal level. At 42, the vehicle is driven through a driving cycle in which multiple air ingestion events of varying severity and duration will occur. While the vehicle is driven at 42, parameters such as vehicle speed, lateral acceleration, and pump power are measured and recorded at 44. At 46, the data is processed to identify air ingestion events and a separate record is created for each event. Each record indicates the vehicle speed, lateral acceleration, and pump power measured over time from the start of the event. This processing may occur in real time as the data is collected, or may be performed in a post-processing step. To facilitate event identification, a subset of air ingestion events may be utilized rather than the entire event. As Figure 4As shown, the model may utilize only the portion of each event after the time at which the pump power begins to increase. At 48, a neural network model or other data-trained type of model is trained using the data obtained at 46.

[0020] Figure 6 Flow chart of a process performed by a controller associated with a power unit. The process is performed at regular intervals, such as once every 100 milliseconds. At 60, pump power is measured directly or indirectly. For example, pump power can be measured indirectly by measuring current and voltage and multiplying the values. At 62, the process determines whether an air ingestion event or a modeled portion of an air ingestion event is ongoing. The controller can identify the start of a new event by increasing pump power. The controller can infer that an event is ongoing based on detecting the start of an event at a previous time and not yet identifying the end of the event. If no air ingestion event is ongoing at 62, the process ends.

[0021] If it is determined at 62 that an air ingestion event is ongoing, then the vehicle speed is measured at 64. The vehicle speed measured at 64, the pump power measured at 60, and the time since the event started are used at 66 to calculate the vehicle speed using the pump power measured at 60. Figure 5 The process trains a model to predict lateral acceleration. At 68, the lateral acceleration is measured and recorded. The measured lateral acceleration may be unidirectional, meaning that only lateral acceleration to the left or right is considered. At 70, the controller determines whether the air ingestion event is complete. For example, completion may be determined by the pump power returning to its nominal level and the lateral acceleration measured at 68 being less than a threshold. If the air ingestion event has not been completed, the process ends.

[0022] After the air ingestion event is complete, the controller compares the measured lateral acceleration from 68 at each time point during the just completed event to the predicted lateral acceleration from 66. A correlation statistic (such as R-squared) is calculated at 72 to quantify the degree of agreement between the measured and predicted values. The correlation statistic is compared to a threshold at 74. If a high degree of agreement is indicated, the controller infers that the fluid level is close to the nominal level and sets the valve at 76 so that fluid is pumped to all outlets.

[0023] If, at 74, the controller finds that the measured lateral acceleration does not correlate well with the predicted lateral acceleration, the controller infers that the fluid level must have dropped significantly from the nominal level. In response, at 78, the controller notifies the operator and adjusts the valve at 80 so that fluid is pumped only to the high priority outlets. This ensures that the high priority outlets will receive adequate flow despite occasional interruptions due to air ingestion events. Which outlets are considered high priority may depend on the operating conditions. For example, at high ambient temperatures, outlets associated with providing cooling have the highest priority. In other cases, outlets associated with lubrication may have a higher priority.

[0024] exist Figure 6 In the illustrated embodiment, the controller adjusts the valve based on a single recent air intake event. In alternative embodiments, the controller may wait until a certain pattern occurs in a series of such events, or may gradually implement corrective action in stages.

[0025] The algorithms, methods or processes disclosed herein may be delivered to or implemented by a computer, a controller or a processing device, which may include any dedicated electronic control unit or a programmable electronic control unit. Similarly, the algorithms, methods or processes may be stored in various forms as data and instructions that can be executed by a computer or a controller, including but not limited to information permanently stored on a non-writable storage medium such as a read-only memory device and information that can be modified and stored on a writable storage medium such as an optical disk, a random access memory device or other magnetic and optical media. The algorithms, methods or processes may also be implemented as software executable objects. Alternatively, suitable hardware components (such as application specific integrated circuits, field programmable gate arrays, state machines or other hardware components or devices) or a combination of firmware, hardware and software components may be used to embody the algorithms, methods or processes in whole or in part.

[0026] Although exemplary embodiments are described above, these embodiments are not intended to describe all possible forms covered by the claims. The words used in the specification are descriptive rather than limiting, and it should be understood that various changes can be made without departing from the spirit and scope of these disclosed materials. For example, the terms "controller" and "multiple controllers" can be used interchangeably herein because the functionality of a controller can be distributed across several controllers / modules, which can all communicate via standard technologies.

[0027] As previously mentioned, the features of the various embodiments may be combined to form additional embodiments of the present invention that may not be explicitly described or shown. Although various embodiments may have been described as providing advantages or being preferred over other embodiments or prior art implementations with respect to one or more desired characteristics, it will be appreciated by those of ordinary skill in the art that one or more features or characteristics may be compromised to achieve desired overall system properties, depending on the specific application and implementation. These properties may include, but are not limited to, strength, durability, marketability, appearance, packaging, size, serviceability, weight, manufacturability, ease of assembly, and the like. Therefore, embodiments that are described as being less desirable than other embodiments or prior art implementations with respect to one or more characteristics are within the scope of the present disclosure and may be desirable for a particular application.

[0028] According to the present invention, a vehicle is provided, the vehicle having: a power unit having an oil pan containing a fluid; an electric pump configured to pump the fluid from an inlet in the oil pan to a plurality of outlets; and a controller programmed to: measure lateral acceleration; predict lateral acceleration based on a power consumption model of the electric pump; and restrict flow to a subset of the plurality of outlets in response to a correlation measure between the measured lateral acceleration and the predicted lateral acceleration being less than a threshold.

[0029] According to an embodiment, the power unit is an electric drive unit comprising a motor and a gearbox.

[0030] According to an embodiment, the model is a neural network model.

[0031] According to an embodiment, the model has been trained using experimental data associated with an oil sump containing a predetermined amount of fluid.

[0032] According to an embodiment, said measured lateral acceleration is unidirectional.

[0033] According to an embodiment, the controller is further configured to measure vehicle speed, and wherein the model is further based on vehicle speed.

[0034] According to an embodiment, the controller is further configured to issue a signal to an operator in response to the correlation measure being less than the threshold value.

[0035] According to the present invention, a power unit for a vehicle is provided, the power unit having: a controller programmed to restrict flow to a subset of a plurality of outlets in response to a correlation metric between a measured lateral acceleration and a predicted lateral acceleration being less than a threshold, the correlation metric being generated via a model using power consumption of an electric pump, the electric pump being configured to pump fluid from an inlet to the plurality of outlets.

[0036] According to an embodiment, the invention is also characterized by: a traction motor; and a gear box.

[0037] According to an embodiment, the model is a neural network model.

[0038] According to an embodiment, the model has been trained using experimental data obtained using a predetermined amount of fluid.

[0039] According to an embodiment, said measured lateral acceleration is unidirectional.

[0040] According to an embodiment, the controller is further configured to issue a signal to an operator in response to the correlation measure being less than the threshold value.

[0041] According to the present invention, a method for detecting abnormal fluid level includes: using an electric pump to pump fluid from an oil pan to multiple outlets; measuring the power consumption and lateral acceleration of the electric pump; identifying the start of an air induction event based on a change in the power consumption; during the air induction event, using a model to calculate a predicted lateral acceleration based on the measured power consumption; comparing the predicted lateral acceleration during the event with the measured lateral acceleration during the event; and restricting flow to a subset of the multiple outlets in response to a correlation measure between the measured lateral acceleration and the predicted lateral acceleration being less than a threshold.

[0042] In one aspect of the invention, the model is a neural network model.

[0043] In one aspect of the invention, the method comprises training the model using experimental data obtained using a predetermined amount of fluid in the oil sump.

[0044] In one aspect of the invention, said measured lateral acceleration is unidirectional.

[0045] In one aspect of the invention, the method comprises measuring vehicle speed, and wherein the model is further based on the measured vehicle speed.

[0046] In one aspect of the invention, the method includes issuing a signal to an operator in response to the correlation measure being less than the threshold value.

Claims

1. A vehicle, comprising: a power unit having an oil sump for containing a fluid; an electric pump configured to pump the fluid from an inlet in the oil sump to a plurality of outlets; and controller, the controller is programmed to Measuring lateral acceleration, using a model to predict lateral acceleration based on the power consumption of the electric pump, and Flow is restricted to a subset of the plurality of outlets in response to a correlation measure between the measured lateral acceleration and the predicted lateral acceleration being less than a threshold.

2. The vehicle of claim 1, wherein the power unit is an electric drive unit including a motor and a gearbox.

3. The vehicle of claim 1, wherein the model is a neural network model.

4. The vehicle of claim 3, wherein the model has been trained using experimental data associated with an oil sump containing a predetermined amount of fluid.

5. The vehicle of claim 1 wherein the measured lateral acceleration is unidirectional.

6. The vehicle of claim 1 wherein the controller is further configured to measure vehicle speed, and wherein the model is further based on vehicle speed.

7. A power unit for a vehicle, the power unit comprising: A controller is programmed to restrict flow to a subset of a plurality of outlets in response to a correlation metric between the measured lateral acceleration and the predicted lateral acceleration being less than a threshold, the correlation metric being generated via a model using power consumption of an electric pump configured to pump fluid from an inlet to the plurality of outlets.

8. The power unit according to claim 7, further comprising: Traction motor; as well as Gearbox.

9. The power unit of claim 7, wherein the model is a neural network model.

10. The power unit of claim 9, wherein the model has been trained using experimental data obtained using a predetermined amount of fluid.

11. A method for detecting abnormal fluid level, the method comprising: An electric pump is used to pump fluid from the oil sump to multiple outlets; measuring power consumption and lateral acceleration of the electric pump; identifying the onset of an air induction event based on the change in power consumption; during the air ingestion event, calculating a predicted lateral acceleration using a model based on the measured power consumption; comparing the predicted lateral acceleration during the event to the measured lateral acceleration during the event; as well as Flow is restricted to a subset of the plurality of outlets in response to a correlation measure between the measured lateral acceleration and the predicted lateral acceleration being less than a threshold.

12. The method of claim 11, wherein the model is a neural network model.

13. The method of claim 12, further comprising training the model using experimental data obtained using a predetermined amount of fluid in the oil sump.

14. The method of claim 11 further comprising measuring vehicle speed, and wherein the model is further based on the measured vehicle speed.

15. The method of claim 11, further comprising signaling an operator in response to the correlation metric being less than the threshold.

Images