Method for controlling the drive power of a working machine, electrified drive train for a working machine and working machine
The electrified drive train with adaptive power reduction using a neural network addresses the issue of insufficient power supply to auxiliary drives by predicting future power demands, ensuring efficient power distribution in working machines.
Patent Information
- Application Number
- DE102024207675
- Authority / Receiving Office
- DE · DE
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-08-13
- Publication Date
- 2026-02-19
AI Technical Summary
Existing methods for operating an electric drive train of a working machine fail to consider the operating point of one electric motor when the other is in operation, leading to insufficient power supply to auxiliary drives.
An electrified drive train with independent electric motors for the drive system and power take-off (PTO) uses a neural network to adaptively reduce drive power when auxiliary power exceeds a predefined value, considering factors like driving speed and power requirements, utilizing a fully connected neural network (FCNN) trained with operating point-specific data.
Ensures sufficient power is always available for auxiliary units by proactively adjusting drive power based on predicted power needs, enhancing power management efficiency and reliability.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
[0001] The invention relates to a method for controlling the drive power of a working machine according to the preamble of claim 1, an electrified drive train for a working machine according to the preamble of claim 8 and a corresponding working machine.
[0002] It is standard practice for construction machinery, and especially agricultural vehicles, to be equipped with power-split transmissions. In a power-split transmission, the drive power of a single engine, usually a diesel engine, is distributed as needed between the main drive and the required auxiliary drives.
[0003] The auxiliary power take-offs (APOs) on such vehicles are often also known as PTOs (Power Take-Offs). An APO can, for example, be designed as a power take-off shaft, which can be used to drive an implement.
[0004] Electrically powered agricultural vehicles are also known in the prior art; these vehicles have one or more electric motors for their drive system or power take-offs instead of an internal combustion engine. Electrically powered vehicles offer a number of advantages over conventionally powered vehicles, such as lower maintenance and operating costs.
[0005] In this context, DE 10 2014 214 477 A1 discloses a power take-off (PTO) for a commercial vehicle for a continuously variable drive of a power take-off shaft. The PTO comprises a PTO gearbox with a gearbox input shaft and a gearbox output shaft, wherein a mechanical variator is arranged on the gearbox input shaft and at least one transmission stage is arranged between the gearbox input shaft and the gearbox output shaft.
[0006] JPS 616 026 A1 describes a method for improving the working efficiency of an agricultural vehicle by monitoring the engine speed. The engine speed changes according to load fluctuations during operation, thus controlling the vehicle's speed accordingly. When a load is applied to a power take-off (PTO) shaft and an axle, causing the engine speed to drop below a predetermined value, the main transmission automatically downshifts one gear to reduce the load. If the engine speed exceeds the predetermined value after the load reduction, the main transmission automatically returns to its original gear.
[0007] DE 10 2019 214 351 A1 describes a method for operating a drive train of a working machine, wherein the drive train comprises a working drive and a travel drive. In a first speed range of the working machine, the working drive is driven by a first electric motor and the travel drive is driven by a second electric motor. In a second speed range of the working machine, the travel drive is additionally or exclusively driven by the first electric motor.
[0008] However, the known methods for operating an electric drive train of a working machine are disadvantageous in that, when operating a drive train with separately operated electric motors for the travel drive and the working drive, the operating point of the other electric motor is not taken into account.
[0009] One object of the invention is to propose an improved method for controlling the drive power of a working machine.
[0010] This problem is solved according to the invention by the method for controlling the drive power of a working machine according to claim 1. Advantageous embodiments and further developments of the invention are described in the dependent claims.
[0011] The invention relates to a method for controlling the drive power of a working machine, wherein the working machine has a drive with a first electric motor and a power take-off with a second electric motor, wherein the first electric motor provides the drive power and wherein the second electric motor provides a power take-off.
[0012] The invention describes a method for controlling the drive power of a working machine. The drive power is provided by a first electric motor, which is assigned to the drive system. In addition, the working machine also has a power take-off (PTO), for example, a power shaft, which is driven by a power take-off supplied by a second electric motor.
[0013] The drive system and the power take-off together constitute the drive train of the machine. Since the first and second electric motors are electric motors, the drive train is accordingly an electrified drive train. The machine itself is therefore also an electrified machine.
[0014] The drive system and the power take-off (PTO) preferably have no mechanical coupling; that is, the drive system power cannot be supplied to the PTO, and vice versa. Instead, the drive system power can be supplied exclusively to the drive system, and the PTO power can be supplied exclusively to the PTO. Thus, both drives, the drive system and the PTO, can be operated independently of each other.
[0015] According to the invention, it is now provided that the drive power is reduced when the auxiliary power exceeds a predefinable value.
[0016] This means that while there is no mechanical drive coupling, there is a functional coupling between the power supply of the drive system and the power supply of the power take-off (PTO). If the PTO power exceeds a predefined value, the drive power is reduced accordingly, for example, to a predefined value or by a predefined value.
[0017] The predefinable value of the auxiliary power output is preferably 80%, 90% or 95% of the maximum available auxiliary power output.
[0018] The invention is based on the understanding that the required power take-off (PTO) output is also influenced by the driving speed of the machine and thus by the available drive power. The faster the machine travels, the less time the PTO must take to perform the required work, so the power requirement of the PTO increases with the driving speed.
[0019] For example, the power take-off (PTO) can drive a sickle bar mower, a hay turner, or a potato harvester. The faster the machine travels, the faster the PTO must operate, and therefore the more power is required.
[0020] In some situations, this can lead to a situation where the maximum available power take-off (PTO) output is no longer sufficient to supply the PTO.
[0021] By reducing the drive power according to the invention when the auxiliary power exceeds the predefinable value, it can be ensured that sufficient auxiliary power is always available to operate the auxiliary power unit in accordance with the respective requirement.
[0022] According to a preferred embodiment of the invention, the drive power is adaptively reduced as the auxiliary power exceeds the predefinable value.
[0023] In this case, the reduction of the drive power should not be fixed or one-off by a specific value or to a specific value, but rather the reduction should be adaptive depending on how far the applied auxiliary power exceeds the predefinable value.
[0024] The drive power is preferably reduced continuously or in stages as the auxiliary power exceeds the predeterminable value.
[0025] This allows the reduction of the drive power to be adapted to the respective situation, and the maximum possible drive power can still be provided in the respective situation.
[0026] According to a further preferred embodiment of the invention, it is provided that the auxiliary power is determined by means of a neural network.
[0027] As has been shown, using a neural network allows for a very reliable determination of the power take-off (PTO) output at any time.
[0028] A fully connected neural network (FCNN) is preferably used, as it has comparatively low requirements for the computing power of a processor that executes the neural network.
[0029] According to a particularly preferred embodiment of the invention, it is provided that a gradient of the auxiliary power is also determined by means of the neural network.
[0030] Determining the gradient of the power take-off (PTO) power, i.e., the change in PTO power over time, allows the neural network to predict with a high degree of probability the PTO power will be in the immediate future and to reduce the drive power in advance.
[0031] The determined gradient is then also advantageously used to reduce the drive power, especially if the neural network recognizes that the auxiliary power in the immediate future, based on the currently determined auxiliary power and the gradient, will exceed the specified value in the immediate future.
[0032] According to a further particularly preferred embodiment of the invention, it is provided that a digital vehicle model is executed by the neural network, wherein the neural network is trained once using operating point-specific power measurement data of the drive system or the auxiliary power unit.
[0033] The vehicle model is preferably a so-called data-driven model or regression model, which is particularly suitable for determining the power take-off power or the gradient of the power take-off power by interpolating previously trained power measurement data.
[0034] For training purposes, a large number of measurement data are preferably recorded at different operating points of the auxiliary drive using suitable sensors, for example, temperature sensors, torque sensors, speed sensors, current sensors or voltage sensors.
[0035] Measurement data is particularly preferably acquired at operating points that lie in the outer edge of the operating spectrum, i.e. at particularly high and particularly low temperatures, speeds, torques, electrical currents and voltages.
[0036] In the context of the invention, "training" refers not only to the creation of the vehicle model by measuring different operating points of the power take-off, but also to the uploading of an already created vehicle model to an identical or similar machine. In other words, the vehicle model is created only once for a selected machine and then copied to the corresponding control units of identical or similar machines. This has the advantage that each machine does not need to be individually equipped with the comparatively expensive sensors, and, more importantly, the corresponding vehicle model does not need to be created individually for each machine, which is comparatively time-consuming.
[0037] According to a particularly preferred embodiment of the invention, it is provided that the power measurement data are each assigned to an electric current, an electric voltage, an ambient temperature, a rotor temperature and / or a stator temperature of the first electric motor.
[0038] This enables the neural network to reliably determine the respective auxiliary drive power and the gradient of the auxiliary drive power.
[0039] Advantageously, the electric current, electric voltage, ambient temperature, rotor temperature and / or stator temperature of the second electric motor are also recorded and made available to the neural network in order to train the neural network.
[0040] According to a further particularly preferred embodiment of the invention, it is provided that the neural network is executed on a vehicle control unit.
[0041] This offers the advantage that a vehicle control unit, which is already integrated into the machine to control and regulate the power take-off or the drive system, provides the computing power required to execute the neural network. Therefore, an additional processing unit can be advantageously dispensed with.
[0042] The invention further relates to an electrified drive train for a working machine, comprising a drive system with a first electric motor and a power take-off with a second electric motor, wherein the first electric motor is configured to provide drive power and wherein the second electric motor is configured to provide power take-off.
[0043] The drive train according to the invention is characterized in that the drive train is designed to reduce the driving drive power when the auxiliary power exceeds a predefinable value.
[0044] The drive train according to the invention thus enables the execution of the method according to the invention and leads to the advantages already described.
[0045] The invention also relates to a working machine comprising a drive train according to the invention. Thus, the advantages of the invention also extend to the working machine according to the invention.
[0046] The working machine is preferably an agricultural vehicle, in particular an agricultural tractor.
[0047] The invention is explained below by way of example with reference to embodiments shown in the figures.
[0048] They show: Fig. 1. An exemplary and schematic embodiment of a method according to the invention for controlling the drive power of a working machine in the form of a flowchart, Fig. 2. An exemplary and schematic embodiment of an electrified drive train 10 according to the invention for a working machine, comprising a drive 11 with a first electric motor 12 and a power take-off 13 with a second electric motor and Fig. 3. An exemplary and schematic representation of another possible embodiment of an electrified drive train 10 according to the invention for a working machine.
[0049] Identical objects, functional units, and comparable components are designated across all figures using the same reference symbols. These objects, functional units, and comparable components are identical in their technical characteristics unless explicitly or implicitly stated otherwise in the description.
[0050] Fig. Figure 1 shows, by way of example and schematically, a possible embodiment of a method according to the invention for controlling the drive power of a working machine in the form of a flowchart. The working machine is, for example, an agricultural tractor.
[0051] The working machine comprises a drive unit 11 with a first electric motor 12 and a power take-off unit 13 with a second electric motor 14.
[0052] In a first process step 101, the first electric motor 12 provides the drive power, while the second electric motor 14 simultaneously provides a secondary drive power in process step 102.
[0053] In process step 103, operating parameters such as the electric current and the electric voltage at the second electric motor 14 are provided to a neural network 20, so-called Fully Connected Neuronal Network 20, which is executed on a vehicle control unit 15.
[0054] The auxiliary power is determined from these operating parameters in step 104 using the neural network 20.
[0055] In step 105, a gradient of the auxiliary power is also determined, also using the neural network 20.
[0056] In step 106, the neural network 20 now recognizes that the power output of the auxiliary drive will soon exceed a predefinable value.
[0057] Therefore, in step 107, the drive power is proactively and adaptively reduced depending on the expected exceedance of the predefined value, so that the predefined value is not exceeded. Due to the reduction in drive power, the machine travels comparatively slower than before, so that the power take-off (PTO) output is sufficient to perform the work to be done per unit of time.
[0058] Fig. Figure 2 shows an exemplary and schematic embodiment of an electrified drive train 10 according to the invention for a working machine, comprising a drive 11 with a first electric motor 12 and a power take-off 13 with a second electric motor 14.
[0059] The first electric motor 12 is designed to provide the drive power and the second electric motor 14 is designed to provide the auxiliary power.
[0060] The powertrain 10 also includes a vehicle control unit 15, for example a driving strategy control unit 15, on which a neural network 20 is executed.
[0061] If the neural network 20 detects that the auxiliary power exceeds a predefined value, the drive power is automatically reduced.
[0062] Fig. Figure 3 shows, by way of example and schematically, another possible embodiment of an electrified drive train 10 according to the invention for a working machine.
[0063] The drive train 10 comprises a first electric motor 12, which is assigned to a drive 11 and a second electric motor 14, which is assigned to a power take-off 13.
[0064] Furthermore, the powertrain 10 also includes a neural network 20. The neural network 20 is trained, for example, with measurement data from the second electric motor 14, speeds specified by the driver for the power take-off 13, and with measurement data of the mechanical power provided at the power take-off 13.
[0065] The measurement data of the second electric motor 14 include, for example, a measured electric current, a measured electric voltage and a measured temperature of the second electric motor 14.
[0066] Furthermore, the neural network 20 is also trained, for example, with measurement data from a working hydraulic system, whereby the working hydraulic system is driven by the auxiliary drive 13.
[0067] The neural network 20 is thus able to determine the power take-off power and the gradient of the power take-off power in a power take-off module 21 during the operation of the drive train 10.
[0068] From the determined auxiliary power, the determined gradient of the auxiliary power and a drive power provided by the first motor 12, a drive power module 22 of the neural network 20 is then checked to see if the auxiliary power will soon exceed the specified value.
[0069] The speed control module 23 then reduces the drive power of the drive train 10 as needed. Reference sign 10 Powertrain 11 Drive system 12 first electric motor 13 Power take-off 14 second electric motor 15 Vehicle control unit, driving strategy control unit 20 neural network, fully connected neural network 21 Power take-off module 23 Speed control module 101 Providing the drive power 102 Providing the power take-off 103 Providing the operating parameters of the second electric motor 104 Determining the power take-off 105 Determining the gradient 106 Exceeding the predeterminable value 107 Reducing the drive power QUOTES INCLUDED IN THE DESCRIPTION
[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature
[0000] DE 10 2014 214 477 A1
[0005] JPS 616 026 A1
[0006] DE 10 2019 214 351 A1
[0007]
Claims
[1] Method for controlling the drive power of a working machine, wherein the working machine has a drive (11) with a first electric motor (12) and a power take-off (13) with a second electric motor (14), wherein the first electric motor (12) provides the drive power (101) and wherein the second electric motor (14) provides a power take-off (102), characterized by , that the drive power is reduced (107) when the auxiliary power exceeds a predeterminable value (106). [2] Method according to claim 1, characterized by , that the drive power is adaptively reduced (107) as the auxiliary power exceeds the predefinable value (106). [3] Method according to at least one of claims 1 and 2, characterized by , that the auxiliary power is determined using a neural network (20) (104). [4] Method according to claim 3, characterized by, that a gradient of the auxiliary power is also determined using the neural network (105). [5] Method according to at least one of claims 3 and 4, characterized by , that a digital vehicle model is executed by the neural network (20), wherein the neural network (20) is trained once using operating point-specific power measurement data. [6] Method according to claim 5, characterized by , that the power measurement data are each assigned to an electric current, an electric voltage, an ambient temperature, a rotor temperature and / or a stator temperature of the first electric motor (12). [7] Method according to at least one of claims 3 to 6, characterized by , that the neural network (20) is executed on a vehicle control unit (15). [8] Electrified powertrain (10) for a working machine, comprising a drive system (11) with a first electric motor (12) and a power take-off (13) with a second electric motor (14), wherein the first electric motor (12) is designed to provide drive power and wherein the second electric motor (14) is designed to provide a power take-off, characterized by , that the drive train (10) is designed to reduce the drive power when the auxiliary power exceeds a predefinable value. [9] Drive train (10) according to claim 8, characterized by , that the drive train (10) is configured to perform a method according to at least one of claims 1 to 7. [10] Working machine comprising a drive train (10) according to at least one of claims 8 and 9.
Citation Information
Patent Citations
Auxiliary drive of a commercial vehicle for a drive shaft
DE102014214477A1
Method for operating a drive train of a working machine, drive train for a working machine and working machine
DE102019214351A1
Methods for operating a vehicle
DE102019220196A1
Hybrid drive system for a working machine and method for controlling such a system
DE102023101526A1
Load control device in farm tractor
JP1986006026A