A method, apparatus, equipment and vehicle for predicting motor torque.

Abstract

The application provides a motor torque prediction method, device, equipment and vehicle, which are used for solving the problem of battery pack overcharge and overdischarge caused by the delayed response of the vehicle controller due to the CAN network transmission delay. The motor torque prediction method comprises the following steps: sending a request torque to a motor controller; acquiring an actual motor speed output by the motor controller after processing the request torque; predicting an initial estimated motor torque corresponding to each of a plurality of preset delays according to the actual motor speed and the request torque; the specific time length of the plurality of preset delays is different; and selecting an optimal value from the plurality of initial estimated motor torques as an estimated value of the current execution torque of the motor according to the actual motor speed and the plurality of initial estimated motor torques.

Description

Technical Field

[0001] This application relates to the field of vehicle control, specifically to a method, device, equipment, and vehicle for predicting motor torque. Background Technology

[0002] New energy vehicles contain a battery pack. At normal temperatures and low voltage levels (State of Charge, SOC), the battery pack has limitations on charging and discharging power. Especially under extreme operating conditions (i.e., at low temperatures and low SOC), the charging and discharging power limitations become even more stringent, requiring more precise power control. Slight overcharging and over-discharging can easily occur. Overcharging refers to the battery pack's charging power exceeding its maximum permissible charging power, while over-discharging refers to the battery pack's discharging power exceeding its maximum permissible discharging power. The root cause of overcharging and over-discharging problems is that the battery pack's operating capacity is insufficient under extreme operating conditions, preventing it from charging and discharging properly.

[0003] Under transient conditions, power calculation imbalances can easily occur. Due to the CAN or CANFD transmission delay in the information exchange between multiple subsystems, the torque limiting of the motor is not timely, the torque control scheme of the motor is unreasonable, and ultimately the engine power control is not accurate enough, which can easily lead to overcharging and over-discharging problems.

[0004] To address overcharging and over-discharging issues, besides improving the battery pack's overcharging and over-discharging capabilities through material selection and design (i.e., hardware optimization), multi-subsystem optimization can also be implemented from an architectural and control perspective. For distributed domain control systems, transmission latency is a significant influencing factor, primarily comprising three components: First, the ADC sampling and processing by the Battery Management System (BMS) before analysis and calculation, followed by transmission via the CAN bus to the Vehicle Control Unit (VCU), introduces a certain delay. Second, the VCU parses chassis torque requirements, battery pack torque limit requirements, and motor power limit requirements, and sends control torque to the Motor Controller Unit (IPU), a process with bus transmission latency. Third, the IPU receives the control torque request from the VCU and parses the actual required execution torque based on the motor status. This execution torque is usually not equal to the control torque sent by the VCU, and the IPU feeds it back to the VCU, with CAN transmission latency still present during the VCU's reception process. Therefore, the information received by the VCU is generally data from some time ago, leading to a delay in identifying overcharging and over-discharging issues compared to the time the BMS issues a fault warning.

[0005] Previous solutions to this problem often involved limiting battery usage to prevent charging and discharging power from exceeding limits. However, this approach failed to unleash the actual working capacity of the battery, resulting in resource waste. For range-extended vehicles, the situation is slightly more complex. The vehicle control unit (VCU) also analyzes the generator power demand and sends engine start-stop commands. During engine start-up, additional power is consumed and there is a generation delay. The VCU then issues a generator power request. After the engine starts, the VCU controls the generator to determine the required generator speed, allowing the generator to participate in the engine speed regulation process. The generator adjusts torque execution in real-time using a closed-loop system based on the range extender's torsional vibration, and feeds this torque back to the VCU. The information sent from the motor controller IPU to the vehicle controller VCU has a transmission delay of 10-20ms. When there is a severe low-pass filtering effect when the information sent by the motor controller IPU is sent, the delay received by the vehicle controller VCU can even reach 40-50ms. Accurate torque information is lost, resulting in delayed judgment and response. The vehicle controller VCU cannot distinguish the actual power of each electrical component. Under extreme conditions, it cannot make the best torque arbitration, which can easily lead to overcharging and over-discharging problems. Summary of the Invention

[0006] This invention provides a method, apparatus, device, and vehicle for predicting motor torque, which solves the problem of delayed response of the vehicle controller (VCU) due to CAN network transmission delay, ultimately leading to overcharging and over-discharging of the battery pack.

[0007] The technical solution of this invention is as follows:

[0008] On the one hand, this application also provides a method for predicting motor torque, the method comprising:

[0009] Send a torque request to the motor controller;

[0010] Obtain the actual motor speed output by the motor controller after processing the requested torque;

[0011] Based on the actual speed of the motor and the requested torque, the initial estimated torque of the motor is predicted after several preset delays; the specific durations of the multiple preset delays are different.

[0012] Based on the actual speed of the motor and multiple initial estimated torques of the motor, the optimal value is selected from the multiple initial estimated torques of the motor as the estimated value of the current execution torque of the motor.

[0013] Preferably, the step of selecting an optimal value from the multiple initial estimated torques of the motor as the current operating torque of the motor, based on the actual speed of the motor and multiple initial estimated torques of the motor, includes:

[0014] The initial estimated torque of the multiple motors is converted into the initial estimated speed of the multiple motors;

[0015] Calculate the speed difference between the initial estimated speed of each motor and the actual speed of the motor;

[0016] The initial estimated torque of the motor corresponding to the smallest absolute value of multiple speed differences is taken as the optimal value and used as the estimate of the motor's current operating torque.

[0017] Preferably, multiple preset delays are determined in advance through offline simulation tests using data transmitted from internal motor quantities and the CAN bus in the power domain.

[0018] Preferably, the step of predicting the initial estimated torque of the motor after multiple preset delays, based on the actual speed of the motor and the requested torque, includes:

[0019] Calculate the product of the actual speed of the motor and the preset gain coefficient corresponding to each preset delay;

[0020] Each of the aforementioned products is then converted into torque compensation amounts;

[0021] The torque compensation amount and the requested torque are added together to obtain the initial expected torque of the motor after each preset delay.

[0022] Preferably, the method further includes:

[0023] Based on the current operating torque of the motor, the optimal torque arbitration is performed on all electrical components in the vehicle.

[0024] On the other hand, this application also provides a motor torque prediction device, the motor torque prediction device comprising:

[0025] The sending module is used to send a torque request to the motor controller;

[0026] The actual motor speed acquisition module is used to acquire the actual motor speed output by the motor controller after processing the requested torque;

[0027] The motor initial estimated torque prediction module is used to predict the motor initial estimated torque after multiple preset delays based on the actual speed of the motor and the requested torque; the specific durations of the multiple preset delays are different.

[0028] The current execution torque determination module is used to select an optimal value from the multiple initial estimated torques of the motor as the torque value of the current execution torque of the motor, based on the actual speed of the motor and multiple initial estimated torques of the motor.

[0029] Preferably, the current torque determination module includes:

[0030] The motor initial estimated speed conversion unit is used to convert the multiple motor initial estimated torques into multiple motor initial estimated speeds;

[0031] The difference calculation unit is used to calculate the speed difference between the initial estimated speed of multiple motors and the actual speed of the motors.

[0032] The current execution torque determination unit is used to determine the optimal value of the initial estimated torque of the motor corresponding to the smallest absolute value of multiple speed differences, and use it as the estimated value of the current execution torque of the motor.

[0033] Preferably, multiple preset delays are determined in advance through offline simulation tests using internal motor parameters and data transmitted on the CAN bus.

[0034] On the other hand, this application also provides a control device, including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the motor torque prediction method described above.

[0035] On the other hand, this application also provides a vehicle including the aforementioned motor torque prediction device.

[0036] The beneficial effects of this invention are as follows:

[0037] Due to the delay in CAN transmission, the actual torque output by the motor controller after processing the requested torque is not the true torque of the motor at the current acquisition time. This application predicts the current torque of the motor at the current acquisition time by delaying the actual torque of the motor. Consequently, the vehicle controller can identify the motor's torque exceeding the limit in advance and perform optimal torque arbitration based on the actual power of each electrical component, thereby reducing the probability of overcharging and over-discharging of the battery pack. Attached Figure Description

[0038] Figure 1 This is a flowchart illustrating the motor torque prediction method in the embodiments of this application;

[0039] Figure 2 This is a flowchart illustrating step S3 in an embodiment of this application;

[0040] Figure 3 This is a flowchart illustrating step S4 in an embodiment of this application;

[0041] Figure 4 This is a schematic diagram of the power domain CAN bus and nodes in the embodiments of this application;

[0042] Figure 5 This is a schematic diagram illustrating the principle of motor torque prediction in the embodiments of this application;

[0043] Figure 6 This is a schematic diagram of the torque prediction function considering dynamic delay in an embodiment of this application.

[0044] Figure 7 This is a schematic flowchart of the motor torque prediction device in the embodiments of this application. Detailed Implementation

[0045] The method of the present invention will be further described below with reference to the embodiments and accompanying drawings. This embodiment is implemented based on the technical solution of the present invention, and provides detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0046] It is known that for a vehicle distributed domain control system, its power balance equation can be expressed as follows (1):

[0047] P M (t k )+P GE (t k )+P Eng (t k )+P A (t k ) = P B (t k (1)

[0048] P M (t k P represents the actual operating power of the motor at time k. GE (t k P represents the actual power output of the generator at time K. Eng (t k P represents the actual power output of the engine at time k. A (t k P represents the auxiliary component and resistance power at the Kth time point. B (t k Let P be the actual power output of the battery pack at time k. A positive power output indicates discharging, and a negative power output indicates charging. For pure electric vehicles, the engine (Eng) and generator (GE) will not be included, i.e., P in equation (1) above is... Eng (t k ) and P GE (t k (This does not exist.)

[0049] For the vehicle control unit (VCU), let the current time be t. k Due to transmission delay, we can only know the past moment t.k-N And in the future t k+M The actual power of each subsystem is shown in equation (2) below:

[0050] P M (t k-N )+P GE (t k+1 )+P Eng (t k+M )+P A (t k )≠P B (t k-1 (2)

[0051] P M (t k-N P represents the actual operating power of the motor at the kN-th time point. GE (t k+1 P represents the actual power output of the generator at time K+1. Eng (t k+M P represents the actual power output of the engine at time point K+M. A (t A P represents the auxiliary component and resistance power at the Kth time point. B (t k-1 Let be the actual power output of the battery pack at time k-1. Since the time points are not at the same time, the power balance cannot satisfy the equation. Before optimization, this difference Δ0 is expressed as:

[0052] Δ0=P M (t k-N )+P GE (t k+1 )+P Eng (t k+M )+P A (t k )-P B (t k-1 (3)

[0053] To reduce this power difference and improve control robustness and accuracy, part of the torque strategy of the motor controller IPU is moved forward to the vehicle controller VCU. This allows us to know the current operating torque of the motor in advance, and thus the optimized power balance relationship can be expressed as:

[0054] Δ1=P M (t k+1 )+P GE (t k+1 )+P Eng (t k+M )+P A (t k)-P B (t k-1 (4)

[0055] Whether it's a pure electric vehicle or a range-extended electric vehicle, as long as the inequality |Δ1|<|Δ0| is satisfied, it means that the power balance or torque capability limitation has been improved. Therefore, the vehicle controller (VCU) can further reduce the probability of battery pack overcharging and over-discharging by setting the motor's torque limit in advance before sending the next torque request to the motor controller (IPU).

[0056] Based on the above ideas, this application provides a motor torque prediction method for a distributed domain control system. This motor torque prediction method is applied to the vehicle controller. By identifying the current execution torque of the motor at the current acquisition time in advance in the vehicle controller, the overcharging and over-discharging problems of the battery pack caused by changes in motor torque and CAN transmission delay are avoided.

[0057] Reference Figure 1 The method includes:

[0058] S1, send a torque request to the motor controller;

[0059] S2, Obtain the actual motor speed output by the motor controller after processing the requested torque;

[0060] S3, based on the actual speed of the motor and the requested torque, predict the initial estimated torque of the motor after multiple preset delays; the specific durations of the multiple preset delays are different.

[0061] S4. Based on the actual speed of the motor and the multiple initial estimated torques of the motor, select the optimal value from the multiple initial estimated torques of the motor as the estimated value of the current execution torque of the motor.

[0062] Combination Figure 4 For the aforementioned step S1, the vehicle controller VCU will determine the requested torque based on the vehicle gear position, chassis power requirements, torque limits required for battery management, motor operating status, efficiency, NVH performance requirements, etc., and send the requested torque to the motor controller IPU.

[0063] Combination Figure 4 For step S2 mentioned above, the motor controller IPU will calculate the actual motor execution torque and the actual motor speed based on the requested torque, the vehicle torque arbitration result, the battery capacity limit, the motor operating status, and the transmission system torsional vibration requirements.

[0064] In addition, the chassis controller calculates the required driving force, yaw rate, and wheel slip ratio of the entire vehicle based on braking status and motor operating status. The battery management system (BMS) calculates safe current and safe voltage thresholds based on the power requirements of various electrical components (including vehicle accessories) and ambient temperature to avoid overcharging and over-discharging. The vehicle control unit (VCU) also combines energy management results to calculate the engine's required power and the generator's target torque. The engine and generator controllers perform start-stop control and speed regulation control to operate at the target power and speed, and the generator plays a role in suppressing torsional vibration during speed regulation. Each controller sends its operating status to the CAN network, sharing operating status information. The gateway plays a role in bidirectional transmission and sharing of information across the CAN network, forwarding the status and operating information of other controllers across the network in real time.

[0065] In step S3 of this application, multiple preset delays are determined in advance through offline simulation tests using data transmitted on the motor's internal quantities and the CAN bus of the power domain.

[0066] In the initial offline simulation, it is assumed that the actual speed and torque of the motor are precisely known. The motor controller IPU sends data to the CAN bus at fixed intervals. The actual torque and speed of the motor read by the vehicle controller VCU are delayed. To account for this delay, it is essential to solve the torque prediction problem caused by the delay in sending and receiving data. Secondly, it is also assumed that the transfer function H(s) of the motor torsional vibration active control is known, specifically for the following system:

[0067]

[0068] Where s is a complex variable in the Laplace transform; b2, b1, b0 and a2, a1, a0 are the coefficients of the transfer function, which are constants representing specific properties of the system.

[0069] The focus of offline simulation is to optimize and improve the control capabilities of the vehicle control unit (VCU). Therefore, a motor torque prediction module is added to the existing VCU to simulate the actual system operation. This motor torque prediction module provides phase compensation. More specifically, using known transfer functions and measured data, open-loop analysis of the causal relationships of the dynamic system is performed to capture phase delay parameters. This involves simulating CAN transmission delay and the differences in task cycles between different domain controllers in closed-loop simulation. Then, in actual closed-loop applications, these identified and known delays are removed, thereby achieving accurate prediction of the actual torque of the motor, improving the accuracy of calculating the actual power consumption of the drive motor, and further improving the overcharge and over-discharge control of the battery, reducing the frequency of problems.

[0070] In offline simulation, accurate identification and modeling of input delay are crucial. This embodiment considers acquiring data from internal motor quantities and the CAN bus in the power domain, typically the maximum value of the sender and receiver cycles. The strategy for considering decimal or non-integer cycle delays in modeling involves interpolating the delays of two adjacent integer cycles. For example, if the sender cycle is Tspan1 and the receiver cycle is Tspan2, and Tspan1 > Tspan2, then the actual input delay should be Tspan1. If the total input and output delay is known to be Tspan1 + Tspan2, then the input delay can be estimated as 0.5 * (Tspan1 + Tspan2). When the input delay Tspan1 is 1.5T, where T is the VCU task cycle, the input delay calculation method, expressed in the Z-domain, is as follows:

[0071]

[0072] During the offline parameter identification process, the above formula needs to be used as the input delay. After the offline parameter identification is completed, the delay module needs to be removed during closed-loop operation because the actual hard-wired signal already contains the delay information.

[0073] For sampling reception delay (also known as output delay), the identification method for reception delay and input delay is the same. The average value is also calculated by collecting the internal quantity of the drive motor controller and comparing it with the CAN signal. Using the input delay information, the average sampling reception delay can be obtained when the total delay is known.

[0074] During offline simulation, analysis was performed at equal time intervals. A slight discrepancy was observed between transient delay information and average delay during closed-loop operation. This difference primarily stemmed from variations in the CAN transceiver mechanism and the task cycle of the drive motor current loop. To resolve this discrepancy, a more refined technical solution can be implemented. For the vehicle control unit (VCU), the current input is known. Data acquired from the CAN bus requires receive delay correction processing for accurate simulation. If receive delay is not processed, and data "containing delay information"—i.e., output signal channels with causal relationships—is used directly, then receive delay calculation is unnecessary. In other words, as the data sender, the VCU only cares about input delay, aiming to estimate the torque with input delay.

[0075] After identifying the fixed phase delay, a main calculation module can be established. Typically, the fixed delay of the main calculation module can be directly obtained through testing. For example, if the minimum delay unit is known to be T, the average test results yield Tspan = 1.5T = 0.015s. Figure 6Establishing trial modules involves analyzing the model and its stability at a certain multiple of the delay. If the fixed delay of the main module is T, then only the stability of trial modules of 2T, 3T, ..., N*T needs to be analyzed. Typically, 1 to 5 trial modules are sufficient for analysis. Stability analysis involves fine-tuning the parameters of each trial module to ensure that it meets both accuracy and stability requirements during offline simulation. In particular, the error feedback gains can be different; the main difference between the main module and the trial modules lies in the feedback gain settings.

[0076] Through the offline simulation described above, the number and specific duration of the preset delays in step S3 can be determined. For example, in this embodiment of the application, the number of preset delays is set to 5, namely 10ms, 20ms, 30ms, 40ms and 50ms.

[0077] Reference Figure 2 In this embodiment of the application, step S3, which predicts the initial estimated torque of the motor after multiple preset delays based on the actual speed of the motor and the requested torque, includes:

[0078] S31, calculate the product of the actual speed of the motor and the preset gain coefficient corresponding to each preset delay;

[0079] S32, convert each of the aforementioned products into torque compensation amounts respectively;

[0080] S33, add the torque compensation amount and the requested torque respectively to obtain the initial expected torque of the motor after each preset delay.

[0081] The specific value of the preset gain coefficient corresponding to the preset delay can be set according to actual needs.

[0082] In this embodiment, the control principle and detailed scheme for torque prediction can be found in [reference needed]. Figure 5 It displays the relationship between motor torque and time, explaining the principle of predictive analysis. The solid line represents the prediction result from the vehicle controller (VCU), which is ahead in time phase, while the dashed line represents the actual result received by the VCU, which is behind in time phase. For example, at times T1 and T2, torque information is marked with three different shapes: a triangle represents the motor's current execution torque predicted by the VCU at time T2; a rhombus represents the actual execution torque received by the VCU at time T1; and a pentagon represents the actual execution torque received by the IPU from the motor controller at time T2.

[0083] After predicting the initial estimated motor torque after multiple preset delays in step S3, step S4 selects the initial estimated motor torque with the smallest error as the output. Specifically, referring to... Figure 3 Step S4 includes:

[0084] S41, convert the initial estimated torque of the multiple motors into the initial estimated speed of the multiple motors;

[0085] S42, calculate the speed difference between the initial estimated speed of the multiple motors and the actual speed of the motors;

[0086] S43, the initial estimated torque of the motor corresponding to the smallest absolute value of multiple speed differences is taken as the optimal value, and used as the estimated value of the current execution torque of the motor.

[0087] After obtaining an estimate of the current operating torque of the motor, in this embodiment of the application, the method further includes:

[0088] The vehicle control unit (VCU) adjusts the final driving smoothness requirements based on driver needs, chassis requirements, and the estimated current torque of the predicted motor, and issues commands for engine power and generator speed, thus forming coordinated control of various subsystems.

[0089] Because the CAN delay has been compensated, the current operating torque of the motor can be predicted more quickly, and the torque difference of the motor can be known in advance, which is more conducive to controlling the overall power consumption balance.

[0090] On the other hand, refer to Figure 7 This application also provides a motor torque prediction device, the motor torque prediction device comprising:

[0091] The sending module 101 is used to send a torque request to the motor controller;

[0092] The actual motor speed acquisition module 102 is used to acquire the actual motor speed output by the motor controller after processing the requested torque;

[0093] The motor initial estimated torque prediction module 103 is used to predict the motor initial estimated torque after multiple preset delays based on the actual speed of the motor and the requested torque; the specific durations of the multiple preset delays are different.

[0094] The current execution torque determination module 104 is used to select an optimal value from the multiple initial estimated torques of the motor as the torque value of the current execution torque of the motor, based on the actual speed of the motor and multiple initial estimated torques of the motor.

[0095] Preferably, the current execution torque determination module 104 includes:

[0096] The motor initial estimated speed conversion unit 1041 is used to convert the multiple motor initial estimated torques into multiple motor initial estimated speeds;

[0097] The difference calculation unit 1042 is used to calculate the speed difference between the initial estimated speed of multiple motors and the actual speed of the motors.

[0098] The current execution torque determination unit 1043 is used to determine the optimal value of the initial estimated torque of the motor corresponding to the smallest absolute value of multiple speed differences, and use it as the estimated value of the current execution torque of the motor.

[0099] Preferably, multiple preset delays are determined in advance through offline simulation tests using internal motor parameters and data transmitted on the CAN bus.

[0100] On the other hand, this application also provides a control device, including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the motor torque prediction method described above.

[0101] On the other hand, this application also provides a vehicle including the aforementioned motor torque prediction device.

[0102] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0103] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0104] It should also be noted that, in this document, the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are used for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or component referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, relational terms such as "first" and "second" are used to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations, nor should they be construed as indicating or implying relative importance. Moreover, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements does not include those elements, but also includes other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes the element.

[0105] The technical solution provided by this invention has been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this invention. The descriptions of the above embodiments are only for the purpose of helping to understand this invention, and the content of this specification should not be construed as a limitation of this invention. Furthermore, for those skilled in the art, there will be different forms of changes in the specific implementation methods and application scope based on this invention. It is neither necessary nor possible to exhaustively list all implementation methods here, but obvious changes or modifications derived therefrom are still within the protection scope of this invention.

Claims

1. A method of motor torque prediction, characterized by, The motor torque prediction method comprises: sending a request torque to a motor controller; acquiring an actual motor speed output by the motor controller after processing the request torque; predicting, according to the actual motor speed and the request torque, motor initial estimated torques corresponding to a plurality of preset time delays respectively; the plurality of preset time delays have different specific time lengths; selecting, according to the actual motor speed and the plurality of motor initial estimated torques, an optimal value from the plurality of motor initial estimated torques as an estimated value of a current execution torque of the motor; the step of predicting, according to the actual motor speed and the request torque, motor initial estimated torques corresponding to a plurality of preset time delays respectively comprises: respectively calculating products of the actual motor speed and preset gain coefficients corresponding to each preset time delay; respectively converting the products into torque compensation amounts; respectively adding the torque compensation amounts and the request torque to obtain motor initial estimated torques corresponding to each preset time delay.

2. The electric motor torque prediction method of claim 1, wherein, the step of selecting, according to the actual motor speed and the plurality of motor initial estimated torques, an optimal value from the plurality of motor initial estimated torques as a current execution torque of the motor comprises: converting the plurality of motor initial estimated torques into a plurality of motor initial estimated speeds; calculating speed difference values of the plurality of motor initial estimated speeds and the actual motor speed respectively; selecting, as the optimal value, a motor initial estimated torque corresponding to a minimum absolute value of the plurality of speed difference values, and selecting the motor initial estimated torque as the estimated value of the current execution torque of the motor.

3. The electric motor torque prediction method of claim 1, wherein, The plurality of preset time delays are determined through offline simulation test of data transmitted through a motor internal quantity and a power domain CAN bus in advance.

4. The electric motor torque prediction method of claim 1, wherein, The method further comprises: conducting optimal torque arbitration for each electrical component on the vehicle in combination with the current execution torque of the motor.

5. A motor torque prediction device characterized by comprising: The motor torque prediction device comprises: a sending module configured to send a request torque to a motor controller; a motor actual speed acquisition module configured to acquire an actual motor speed output by the motor controller after processing the request torque; a motor initial estimated torque prediction module configured to predict, according to the actual motor speed and the request torque, motor initial estimated torques corresponding to a plurality of preset time delays respectively; the plurality of preset time delays have different specific time lengths; a current execution torque determination module configured to select, according to the actual motor speed and the plurality of motor initial estimated torques, an optimal value from the plurality of motor initial estimated torques as a torque value of a current execution torque of the motor; the step of predicting, according to the actual motor speed and the request torque, motor initial estimated torques corresponding to a plurality of preset time delays respectively comprises: respectively calculating products of the actual motor speed and preset gain coefficients corresponding to each preset time delay; respectively converting the products into torque compensation amounts; respectively adding the torque compensation amounts and the request torque to obtain motor initial estimated torques corresponding to each preset time delay.

6. The electric motor torque prediction apparatus according to claim 5, characterized by The current execution torque determination module comprises: a motor initial estimated speed conversion unit configured to convert the plurality of motor initial estimated torques into a plurality of motor initial estimated speeds; a difference calculation unit configured to calculate a speed difference between each of the plurality of initial estimated speeds and the actual speed of the motor; a current execution torque determination unit configured to determine the initial estimated torque corresponding to the minimum absolute value of the plurality of speed differences as the optimal value, and as an estimated value of the current execution torque of the motor.

7. The electric motor torque prediction apparatus according to claim 5, characterized by The plurality of preset time delays are determined through offline simulation tests on data transmitted through the motor internal quantity and the CAN bus.

8. A control device characterized by comprising: The motor torque prediction device comprises a processor, a memory, and a program or instruction stored in the memory and executable on the processor, and the program or instruction is executed by the processor to implement the steps of the motor torque prediction method according to any one of claims 1 to 4.

9. A vehicle characterized by comprising: The motor torque prediction device comprises the motor torque prediction device according to any one of claims 5 to 7.

Images