FIELD OF THE DISCLOSURE

[0001] The present disclosure is related to a system and a method for calibrating a torque control function of a vehicle, in particular a feed-forward torque control function controlling an input torque applied by an engine or a motor to the driveline of the vehicle. BACKGROUND OF THE DISCLOSURE

[0002] Conventional vehicles comprise an engine or in case of an electric / hybrid vehicle (additionally) one or several electric motors for generating a torque. Said torque is input into a driveline (also called driving system) comprising e.g. a transmission and a power transfer system. The driveline transfers the torque to the driving wheels of the vehicle, when the vehicle accelerates or decelerates. When an input torque is applied to the driveline by the engine and/or motor (i.e. at acceleration or deceleration of the vehicle), vibrations may be additionally transmitted to the driveline. Furthermore also external vibration sources may cause vibrations in the driveline. In addition, vibrations may also cause further parts of the vehicle other than the driveline to vibrate.

[0003] In order to reduce such vibrations, it is known to use a torque control algorithm in the vehicle, e.g. a vibration reduction control algorithm through feedforward control logic. Such a torque reduction is in particular important in vehicles having no torque converters.

[0004] As also outlined in US20130325285, a known vibration reduction control algorithm may be implemented based on a control logic in which a feedforward control function outputs a motor command torque and a feedback control function calculates a vibration reduction torque for suppressing speed vibration extracted as motor sensor speed and motor model speed. A driving system transfer function outputs a final motor command torque obtained by summing the motor command torque and the vibration reduction torque.

[0005] In particular, the vibration reduction algorithm disclosed in US20130325285 includes optimized feedforward logic in which a request torque is divided into two or more different types of torques of which one of the two torques is provided two times with a time difference equal to half period from a vibration period of the driving system of the vehicle to reduce it's vibration; and a feedback logic in which information from the driving system is processed and added to the motor command torque of the optimized feedforward logic. The half period from a vibration of the driving system may be calculated by a motor velocity sensor measuring the motor velocity oscillations.

[0006] The vibration reduction algorithm described in US20130325285 is a combination of a feed-forward and a feed-back algorithm in order to reduce the vibration of the driving system. Additionally, the feed-forward algorithm is based on measurement of the actual system behavior, e.g. the motor velocity oscillations.

[0007] Conventionally, in order to calibrate a vibration reduction algorithm, it is necessary to carry out a manual calibration procedure. The process of calibrating the torque control function is an iterative process requiring several loops before the required result and system response target could be achieved, estimated to about 1000 man hours by a calibration engineer. During this time a dedicated test vehicle is required as well as a secured test area (track). Additionally to these requirements, the calibration quality is not reliable, as it is highly depending on the experience and expertise of the calibration engineer.

SUMMARY OF THE DISCLOSURE

[0008] Currently, it remains desirable to provide a system and method for reliably calibrating a torque control function of a vehicle in a time and cost efficient manner, in particular in a fully automated process.

[0009] Therefore, according to the embodiments of the present disclosure, it is provided a system for calibrating a torque control function of a vehicle having a driveline, wherein the torque control function controls an input torque applied to the driveline. The system comprises (or is configured to carry out/calculate) a mathematical representation configured to simulate in the time domain vehicle and/or driveline dynamics in response to an applied input torque.

The system is configured to:

- transform the mathematical representation into a frequency domain,

- define a desired response of the transformed mathematical representation in the frequency domain, - translate said desired response of the mathematical representation into a required input torque in the time domain, and

- calibrate the torque control function based on the required input torque.

[0010] Accordingly, the calibration technique is suitable for a fully automated process, desirably by a combination of simulation techniques and an analytical approach. Said simulation techniques may comprise simulating a physical model of the vehicle driveline and its vital components. Said analytical approach may identify the most suitable system torque for a pre-defined response criteria, e.g. using a Laplace transform function.

[0011] With the simulation techniques the physical system behavior can be simulated and the feed-forward algorithm can be calibrated fully off-line. A test vehicle may only be used for a final validation purpose.

[0012] The benefit of the system is a substantial reduction of the development resource in terms of man hours and requirements for a test vehicle or test track as well as an improvement of the calibration quality as mentioned above.

[0013] The input torque applied to the driveline may be generated by a torque generation unit, comprising e.g. an engine and/or one or several electric motors.

[0014] The mathematical function may enable simulation of the dynamic behavior of the vehicle under torque changes such as acceleration or deceleration.

[0015] The mathematical function may also be referred to as a mathematical simulation model simulating the vehicle and/or driveline dynamics. A simulation model of the vehicle may be created and connected to the ECU (Electronic Control Unit) control model.

[0016] The control model is not necessarily a model of the complete ECU but may rather be a model of one specific function running in the ECU. Both models (vehicle/driveline and control function) may run in the same software package but this is not mandatory.

[0017] The frequency domain may be a complex domain describing frequency and damping characteristics of the dynamic response of the mathematical representation.

[0018] The transformation of the mathematical representation may be a Laplace or Fourier transformation. [0019] The vehicle and/or driveline dynamics may represent characteristics of the driveline in operation and/or of at least one further part of the vehicle in operation other than the driveline. Examples of such other parts, e.g. suspension bushings of the vehicle, are listed below.

[0020] The driveline, as referred to in the present disclosure, may comprise the whole torque transmission path from the torque generation unit (desirably but not mandatorily excluding the torque generation unit) to the driving wheels of the vehicle (desirably including the wheels) or at least a part of it, e.g. the transmission and/or a power transfer system.

[0021] The vehicle and/or driveline dynamics may be also referred to as the dynamic behavior of the vehicle and/or the driveline under torque changes such as acceleration or deceleration.

[0022] The mathematical representation may represent at least one of:

- a vehicle road load formula,

- vehicle powertrain characteristics including driveline torsional characteristics,

- powertrain mounting system characteristics,

- suspension bushings characteristics, and

- wheel and/or tire characteristics.

[0023] The system may be configured to apply a linearization to the mathematical representation, before the system transforms the mathematical representation into a complex domain.

[0024] Said linearization may comprise removing discontinuities, variable gains.

[0025] The linearization may be a linear motion equation of nth order.

[0026] The vehicle may comprise a torque generation unit (also referred to as power unit). The torque control function may be configured to control characteristics of the torque generated by the torque generation unit. Said torque may be applied as input torque to the vehicle driveline.

[0027] The mathematical representation may have as input a variable representing the input torque applied to the vehicle's driveline. Additionally or alternatively the mathematical representation may have as response a variable representing the vehicle and/or driveline dynamics.

[0028] Said variable may be any mathematical expression comprising information about the driveline behaviour. [0029] The vehicle and/or driveline dynamics may represent at least one of:

- torque of wheel shafts of the vehicle,

- acceleration of the vehicle, and

- driveline dynamics and states, in particular speeds and accelerations of driveline components.

[0030] The system may translate said desired response into a required input torque of the mathematical representation in the time domain in two steps:

- In one step, the system may determine the required input torque based on the desired response in the frequency domain.

- In another step, the system may translate said required input torque into the time domain.

[0031] Alternatively, the system may translate the desired response into the time domain and determine the required input torque based on the desired response in the time domain.

[0032] The torque control function may be a feed-forward torque control function.

[0033] The input torque may be applied by an engine and/or a motor of the vehicle to the driveline. Accordingly, the torque generation unit may comprise the engine and/or a motor.

[0034] The system may comprise a data storage for storing the mathematical representation, and an electronic controller configured to execute the mathematical representation.

[0035] The vehicle may comprise an electronic control unit (ECU) which controls the torque control function in the vehicle. The system may comprise an electronic control unit (ECU) model simulating the ECU (or a part of the ECU) of the vehicle. Said ECU model may be configured to control the torque control function of the simulated vehicle. The simulated vehicle may be represented by the mathematical representation, i.e. the mathematical simulation model. Accordingly, the ECU model may control the simulated vehicle during the calibration process, e.g. according to predetermined instructions of accelerating and/or decelerating the simulated vehicle.

[0036] The present disclosure further relates to a method of calibrating a torque control function of a vehicle having a driveline. The torque control function controls an input torque applied to the driveline. A mathematical representation simulates in the time domain vehicle and/or driveline dynamics in response to an applied input torque. The method comprises the steps of:

- transforming the mathematical representation into a frequency domain,

- defining a desired response of the transformed mathematical representation in the frequency domain,

- translating said desired response of the mathematical representation into a required input torque in the time domain, and

- calibrating the torque control function based on the required input torque.

[0037] The method may comprise the step of simulating in the time domain vehicle and/or driveline dynamics in response to an applied input torque, desirably by using the mathematical representation.

[0038] The method may comprise further method steps which correspond to the functions of the system, as described above. The further method steps may be, as described below.

[0039] The frequency domain may be a complex domain describing frequency and damping characteristics of the dynamic response of the mathematical representation and/or the transformation of the mathematical representation may be a Laplace or Fourier transformation.

[0040] The mathematical representation may represent at least one of:

- a vehicle road load formula,

- vehicle powertrain characteristics including driveline torsional characteristics,

- powertrain mounting system characteristics,

- suspension bushings characteristics, and

- wheel and/or tire characteristics.

[0041] The method may comprise the further step of applying a linearization to the mathematical representation, desirably before the step of transforming the mathematical representation into a complex domain.

[0042] Said linearization may comprise removing discontinuities, variable gains.

[0043] The linearization may be a linear motion equation of nth order.

[0044] The torque control function may control characteristics of the torque generated by the torque generation unit. Said torque may be applied as input torque to the vehicle driveline. [0045] The mathematical representation may have as input a variable representing the input torque applied to the vehicle's driveline. Additionally or alternatively the mathematical representation may have as response a variable representing the vehicle and/or driveline dynamics.

[0046] Said variable may be any mathematical expression comprising information about the driveline behaviour.

[0047] The vehicle and/or driveline dynamics may represent at least one of:

- torque of wheel shafts of the vehicle,

- acceleration of the vehicle, and

- driveline dynamics and states, in particular speeds and accelerations of driveline components.

[0048] The step of translating said desired response into a required input torque of the mathematical representation in the time domain may comprise or consist of two steps:

- In one step, the system may determine the required input torque based on the desired response in the frequency domain.

- In another step, the system may translate said required input torque into the time domain.

[0049] Alternatively, the translating step may comprise the step of translating the desired response into the time domain and determining the required input torque based on the desired response in the time domain.

[0050] The torque control function may be a feed-forward torque control function.

[0051] The present disclosure further relates to a computer program comprising instructions for executing the steps of the method, when the program is executed by a computer.

[0052] It is intended that combinations of the above-described elements and those within the specification may be made, except where otherwise contradictory.

[0053] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure, as claimed.

[0054] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description, and serve to explain the principles thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Fig. 1 shows a schematic block diagram of a system according to embodiments of the present disclosure;

[0056] Fig. 2 shows two flow charts (a) and (b) schematically illustrating vehicle and/or driveline dynamics in response to an applied input torque in time and frequency domain according to embodiments of the present disclosure;

[0057] Fig. 3 shows three flow charts (a) to (c) schematically illustrating the methodology of determining calibration values for the torque control function according to embodiments of the present disclosure; and

[0058] Fig. 4 shows a flow chart illustrating the method of calibrating a torque control function according to embodiments of the present disclosure.

DESCRIPTION OF THE EMBODIMENTS

[0059] Reference will now be made in detail to exemplary embodiments of the disclosure, examples of which are illustrated in the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts.

[0060] Fig. 1 shows a block diagram of a system 10 according to embodiments of the present disclosure.

[0061] The system 10 may comprise an electronic controller 1, such as e.g. an electronic circuit, a processor (shared, dedicated, or group), a combinational logic circuit, a memory that executes one or more software programs, and/or other suitable components that provide the described functionality. In other words, system 10 may be a computer device. The electronic controller 1 may be configured to carry out the calibration method according to the present disclosure.

[0062] The system may further comprise a data storage 2 (i.e. a memory), which may store data, e.g. a computer program which when executed, carries out the calibration method according to the present disclosure. In particular, the system or the data storage may store software which comprises a mathematical representation 3 (e.g. a vehicle simulation model) according to the present disclosure. Desirably the system or the data storage may further store software (e.g. an ECU model) which comprises a torque control function 4 according to the present disclosure. [0063] In other words, the system 10 may comprise a vehicle simulation model 3 which simulates the driveline and/or vehicle characteristics of a vehicle, and an ECU model 4 of said vehicle which simulates the real Electronic Control Unit (ECU) controlling e.g. the power unit, driveline, and desirably other parts of said vehicle necessary for driving. The ECU model 4 may further simulate the torque generated by the power unit in response to a control command input into the power unit.

[0064] Fig. 2 shows two flow charts (a) and (b) schematically illustrating vehicle and/or driveline dynamics in response to an applied input torque in time domain (a) and frequency domain (b) according to embodiments of the present disclosure. In other words, the steps of flow chart (b) corresponds to those of flow chart (a) but have been transformed to the frequency (complex) domain. The methods illustrated by the flowcharts, in particular by flowchart (b), may be carried out by the system 10.

[0065] As shown in the first step of flowcharts (a) and (b), an acceleration command is input into an ECU model. Said acceleration command may e.g. correspond to the signal caused by the actuation of an acceleration pedal in a real vehicle.

[0066] The ECU model determines the simulated torque Tp(t) which would be created by a power unit (e.g. an engine and/or an electric motor) in response to the acceleration command.

[0067] The simulated torque Tp(t) may have different patterns, In particular, the simulated torque Tp(t) may comprise a change in torque with either positive or negative gradient.

[0068] Said torque is input into a simulation model (i.e. the mathematical representation) which in response may output simulated vehicle acceleration. Furthermore the simulation model outputs the simulated response Gx(t) of the vehicle, or at least of its driveline, in response to the torque input into the driveline. Said response may e.g. comprise vibrations of the driveline. In other words, the simulation model is created and connected to the ECU control model which enables simulation of the dynamic behavior of the vehicle under torque changes such as acceleration or deceleration.

[0069] Contents of the vehicle simulation model (virtual vehicle) may be a vehicle road load formula, a vehicle powertrain including torsional characteristics (driveshaft), powertrain mounting system, suspension bushings, and or the combination of wheel and tire.

[0070] The flowchart (b) corresponds to flowchart (a) but has been transformed to the frequency domain. Accordingly, the ECU model outputs a torque Tp(s) in the frequency domain which is input into the transformed mathematical representation. Said mathematical representation may be obtained by linearization and Laplace transformation (or alternatively e.g. by a Fourier transformation) of the mathematical representation of flowchart (a). Consequently, the transformed mathematical representation outputs a complex response Gx(s).

[0071] In other words, from the vehicle model a transfer function may be obtained after a linearization process and transformation to the complex Laplace domain. Hence, the transfer function (i.e. mathematical representation) describes the dynamics of the virtual (simulated) driveline and/or vehicle, however in the (complex) s-domain and not in the time domain.

[0072] The mentioned step of linearization may comprise removing discontinuities, variable gains, delays, etc. from the mathematical representation. Said linearization step may result in a linear motion equation of the nth order.

[0073] After the linearization step, the mathematical representation becomes a linear equation of motion of nth order. However, for each predefined condition a separate linear equation of the system state will be defined. The linearization itself is the process of converting the non-linear equations of motion (or equivalent mathematical/graphical representation in a software) into linear equations of motion of the system for that specific condition.

[0074] The mentioned step of Laplace transformation may comprise a domain change from time to a complex variable s. Said variable s may comprise frequency and damping. The ordinary differential equations (ODE) may become a polynomial with no derivatives.

[0075] The ODEs may be regarded as the model itself in its mathematical form. These are the equations of motion (or motion equations) which define a set of outputs as a function of another set of inputs, as well as the model states and its derivatives in time. The fact that it is a function of some variables and its derivatives makes them differential equations, which are difficult to solve. Laplace (or similar) transformation allows to represent this set of differential equations as ordinary equations (of a complex variable) with no derivatives, what is significantly easier to solve.

[0076] The complex response Gx(s) output by the transformed mathematical representation may be described as a combination of poles around the real axis Re(s) and the imaginary axis Im(s) in the s-plane.

[0077] The simulation model (desirably without the control unit model) may also be referred to as a plant. Said plant may be of any form (mathematical expression, software representation, non-linear or linear, in time domain or complex domain etc). For example, in the present case the 2-dimensional (s- plane) pole location representation is a way of visualizing the dynamic response of this plant to a given arbitrary input.

[0078] Fig. 3 shows three flow charts (a) to (c) schematically illustrating the methodology of determining the calibration values for the torque control function according to embodiments of the present disclosure.

[0079] Flow chart (a) shows the original plant response Gx(s) in complex domain (plant characterized as poles in s plane) and the response Gx(t) in time domain, comparable to flowcharts (a) and (b) of fig. 2.

[0080] In this flow chart (a) of fig. 3 the problem is visible that a continuously increasing and subsequently constant input torque Tp (cf. left diagram) in the time domain results in a undesired response Gx(t) (cf. right diagram) of the vehicle and/or driveline, as simulated by the (transformed) mathematical representation N(s) / D(s) .

[0081] The input torque shape in fig. 2 is just an example. Generally, a very dynamic or rapidly changing input (either positive or negative gradient) may result in a dynamic output, i.e. system response. That specific output is a result of the system including resonant frequencies (represented by the pole pairs located far from the real axis in s-plane) and which could be experienced as negative by the vehicle user. The purpose of this control function and methodology to pre-define is to become able to design the system response according to the requirements by modification of the input torque.

[0082] The (transformed) mathematical representation N(s) / D(s) is a representation of a transfer function with N as Nominator and D as Denominator in the s-domain. It is the transfer function representation of the transformed linear plant model. It may be a rational of two polynomials of s, the roots of D are the poles (values of s for which N/D=inf). It is desirably just a representation and there are other alternative ways of representing the concept.

[0083] The Denominator D(s) desirably provides information (for example by pole location) of the system's resonant frequencies and their independent damping rates.

[0084] In particular, the response Gx(t) may not increase continuously (but may comprise several local maxima) what may be due to the vibrations in the driveline, e.g. as a consequence of amplified frequencies due to the system dynamic properties, as explained previously.

[0085] For this reason, a desired response is defined in the complex domain, as it is shown in flow chart (b) of fig. 3. This may be achieved by rearranging the pole pairs (cf. second right diagram). The resulting response Gx(t) in the time domain (cf. right diagram) has a shape according to our requirements, (cf. left diagram).

[0086] In order to arrange the pole pairs, the methodology of pole placement may be used. This methodology may be based on the principle that if the damping ratio of each pole pair (resonant frequency, or vibration mode) is close to 1, there will be no (or small, or acceptable) overshoot, hence, oscillations, in the response. The damping ratio of each pole pair is related to its location in the s-plane. Changing this location means creating a different D part of the transfer function, i.e. changing the plant, which is only possible in a theoretical way.

[0087] As the plant N/D cannot be changed, an efficient way to avoid undesired oscillations on the output may be to remove from the input the frequencies that would otherwise be amplified by the plant. In other words, the purpose of the control function is to transform an arbitrary input torque shape (with a random frequency content) into a modified torque shape in which some of its frequencies have been removed, so that they are not amplified later by the plant. The resulting equivalent dynamic response is the same as that of a theoretical plant with custom pole locations.

[0088] As further shown in flow chart (c), said desired response can be translated through the mathematical representation N(s)/D(s) (i.e. the linearized function) to the corresponding input torque of the system. Once the required input torque has been defined, the calibration values of the (feedforward) torque control function may be obtained. [0089] Fig. 4 shows a flow chart illustrating the method of calibrating a torque control function according to embodiments of the present disclosure.

[0090] In step SI the mathematical representation f(t) is transformed into a frequency domain F(s). This may be done by a Laplace or a Fourier transformation, desirably with the preceding step of linearization.

[0091] In step S2 a desired response Gx(s) of the transformed mathematical representation is defined in the frequency domain. In other words, the optimum response of the simulation model (i.e. the mathematical representation) is calculated.

[0092] In step S3 said desired response Gx(s) of the mathematical representation is translated into a required input torque in the time domain. In other words, the input torque which is required to obtain said desired response Gx(s), is calculated. Step S3 may consist of two sub-steps. First, the required input torque Tp(s) in the frequency domain is determined. Subsequently, said torque Tp(s) in the frequency domain is transformed to the torque Tp(t) in the time domain.

[0093] Since the simulation model simulating the vehicle and/or the driveline (i.e. the mathematical representation) corresponds to the real vehicle and/or the driveline, it cannot be adapted. Therefore the input torque is adapted which influences the characteristics of the response of the vehicle and/or the driveline.

[0094] In step S4, the torque control function is calibrated based on the required input torque. In particular, the one or several calibration values (e.g. calibration map) may be determined based on the required input torque (Tp(s) or Tp(t)). Said value or values may be used to adapt (i.e. calibrate) the torque control function.

[0095] For example, the calibration parameter may be a calibration map which is used to obtain a maximum increase/decrease rate of Tp per calculation step in the ECU, as a function of the current Tp value. This means that the final calibration map may be obtained by representing the modified input torque rate (i.e. its derivative) as a function of the modified input torque.

[0096] Through this analytical approach the most suitable system torque for pre-defined response criteria can be identified in a simulation environment. By derivating the desired system input torque the calibration values (map) can be obtained. [0097] Throughout the description, including the claims, the term "comprising a" should be understood as being synonymous with "comprising at least one" unless otherwise stated. In addition, any range set forth in the description, including the claims should be understood as including its end value(s) unless otherwise stated. Specific values for described elements should be understood to be within accepted manufacturing or industry tolerances known to one of skill in the art, and any use of the terms "substantially" and/or "approximately" and/or "generally" should be understood to mean falling within such accepted tolerances.

[0098] Although the present disclosure herein has been described with reference to particular embodiments, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present disclosure.

[0099] It is intended that the specification and examples be considered as exemplary only, with a true scope of the disclosure being indicated by the following claims.