FR3048454A1 - DEVICE AND METHOD FOR BALANCING A MULTI-YELLOW THERMAL MOTOR - Google Patents

Abstract

A method of balancing a multicylinder heat engine, comprising the following steps: • driving the heat engine by means of an electric motor speed-controlled on the speed (V) of the heat engine, • measurement, for each cylinder, of an elementary corrective torque (CCE) associated with said cylinder, exerted by the electric motor on the heat engine, • determination, for each cylinder, of a correction of the mass of fuel to be injected, able to compensate for said elementary corrective torque (CCE ).

Description

The present invention relates to the balancing of a multicylinder heat engine.

In a multicylinder heat engine, each cylinder contributes, during an angular sector of the cycle, to provide an elementary torque. The addition of these elementary couples contributes to producing a torque resulting from the heat engine. So that the motor rotates "round", and produce a resultant torque regular and angularly balanced, it is appropriate that the elementary couples from all the cylinders are substantially equal. However, various parameters among which: the volumetric efficiency, the compression ratio, the fuel mass injected, may differ from one cylinder to another and cause variations causing different elementary couples.

In order to correct these differences, it is known to observe the operating thermal engine in order to estimate the elementary torque of each cylinder, conventionally by means of tooth time measurements of a crankshaft sensor, and to compensate for them by varying the amount of fuel injected into each cylinder accordingly. It is thus possible to balance a heat engine.

Such an approach, however, has at least two disadvantages. The first disadvantage is that the correction is carried out in post-processing, and that during an initial learning phase the heat engine must operate for a certain time without correction, thus being unbalanced. This can be damaging for the engine itself. In addition this degrades the feeling of the driver. Another disadvantage, related to the use of tooth times to estimate torque variations, is that this approach can not be used on all engine operating ranges. The invention proposes an alternative that does not have these disadvantages. An electric motor is used to drive the engine. The electric motor is slaved in speed on the speed of the engine. It follows that the electric motor produces a corrective torque in order to correct the torque variations. This balances the heat engine.

An elementary corrective torque associated with each cylinder is exerted by the electric motor on the engine. This elementary corrective torque is advantageously measured. A correction of the mass of fuel to be injected, compensating for said elementary corrective torque, is determined for each cylinder.

According to another characteristic, this correction can be applied during an injection step.

According to another characteristic, the measurement of the elementary correction pairs is deduced from a control current of the electric motor.

According to another characteristic, the measurement of the elementary correction pairs comprises the following steps: • measurement of a resulting corrective torque, • equation of an angular range corresponding to an engine cycle in a number of sectors equal to the number of cylinders, • association to each cylinder of the sector in which said cylinder is propellant, the elementary corrective torque associated with a cylinder being equal to the resulting corrective torque in the associated sector and zero in the other sectors.

According to another characteristic, the correction is determined as a function of the elementary corrective torque and possibly of at least one operating parameter of the motor.

According to another characteristic, the determined correction is recorded and indexed as a function of the elementary correction torque and, if appropriate, of said at least one operating parameter of the engine, for later reuse.

According to another characteristic, a learning phase, comprising the training, measuring and determining steps, is applied continuously or on demand or only in specific operating zones or only in an initial phase until all the desired corrections are determined.

According to another characteristic, an exploitation phase, comprising the injection step, is applied continuously.

According to another characteristic, the electric motor is an electric machine of a hybrid thermal / electric vehicle.

According to another characteristic, the electric motor is no longer used for balancing, when the elementary corrective pairs are substantially compensated by the corrections. The invention also relates to a device comprising an electronic control unit and means for implementing the steps of a method according to the invention. The invention also relates to a vehicle comprising such a device. FIG. 1 shows a diagram indicating the target fuel mass for each cylinder; FIG. 2 presents a diagram indicating the elementary torque for each cylinder; FIG. 3 presents a diagram indicating the elementary corrective torque for each cylinder. FIG. 4 shows a diagram indicating the fuel mass correction for each cylinder; FIG. 5 presents a comparative time diagram of the torque, corrective torque and speed quantities. Other features, details and advantages of the invention will emerge more clearly from the detailed description given below.

In order to control a heat engine, it is determined, for each cylinder, a fuel mass reference MF to inject, typically by a motor control unit, or ECU for "Engine Control Unit" in English. This setpoint is, in a known manner, determined according to various parameters, such as the position of the accelerator pedal, the load, the engine speed, etc.

An example of such a fuel mass set point MF is illustrated, as a function of a cylinder number, here between 0 and 3 and repeating itself periodically, in FIG. 1.

This MF fuel mass setpoint is used to control fuel injection. However, because of several disturbing factors, the same mass of fuel MF injected does not necessarily produce the same elementary torque CE in a cylinder. The disturbing factors are mainly caused by dispersions that can be spatial and / or temporal. Thus the volumetric efficiency can vary from one cylinder to another due to dimensional or behavioral variations from one cylinder to another. The compression ratio may also vary from one cylinder to another depending on the nature or composition of the fuel and / or temperature and / or pressure in a cylinder. The mass of fuel actually injected may still be different from its MF setpoint due to dimensional or behavioral differences from one injector to another.

This leads to a dispersion of the elementary couples CE. FIG. 2 illustrates the elementary pairs CE obtained, as a function of a cylinder number, by applying the fuel mass instructions MF of FIG. 1.

The dispersion of the elementary couples CE causes an imbalance of the engine. Thus the resulting torque sum of the individual elementary couples CE is not regular, even though the cylinders are angularly equilateral around the crankshaft and their contributions are temporally equidistributed, even in the case where the MF instructions are identical.

In order to compensate for this imbalance, the heat engine is driven by means of an electric motor, which electric motor is advantageously controlled in speed on a setpoint equal to the average speed of the heat engine.

The dispersion of the elementary pairs CE respectively produced by each cylinder, results in an oscillation of the angular velocity V of the engine around its mean value. The servocontrol of the electric motor in speed, in order to smooth said angular velocity V to a mean value, exerts on the heat engine a corrective torque CC. This correction torque CC can be distributed angularly on the different cylinders, with a correction elementary pair CCE associated with each cylinder.

FIG. 3 illustrates the CCE elementary correction pairs as a function of the cylinder numbers produced by the speed-controlled electric motor. Because of the speed control these corrective pairs CCE compensate for the differences between the elementary couples CE of FIG. 2 and "respond" to these elementary couples CE. Thus, the elementary corrected pairs, defined for each cylinder as the sum of the elementary torque CE and the elementary correction torque CCE, are substantially equal from one cylinder to the other.

The only application of the electric motor controlled by the heat engine balances the heat engine. This balancing is substantially immediate, since it is effective as soon as the electric motor operates and the speed control converges. The invention goes however further. It adds a step of determining, for each cylinder, a correction FAC of the fuel mass MF to be injected, able to compensate for said elementary correction pair CCE. During this step, for each cylinder, a correction FAC of the fuel mass MF is determined.

It was seen that the disturbing factors were numerous. Most of them (volumetric efficiency, compression ratio, etc.) are not controllable. The easiest parameter to control is the mass of fuel injected, which is then retained to correct disturbances, from all causes.

A FAC correction is determined in such a way that a corrected fuel mass, equal to the MF fuel mass initially determined by the ECU (and such that the cylinder produces a CE elementary torque) weighted FAC correction, ie MF x FAC, causes said cylinder to produce a corrected EC + CCE torque. As soon as the FAC corrections are known for at least one operating zone of the engine, the method can add a step applying said FAC corrections, in said zone. In such a step, the method injects into each cylinder, a fuel mass MF corrected for said FAC correction, a corrected fuel mass equal to MF x FAC. It follows that each cylinder then produces a corrected elementary pair CE + CEC. It follows that the new pairs produced by the cylinders are substantially equal, and that the heat engine is balanced.

The electric motor can always be present and active. However, since the speed V of the heat engine is now smooth, the speed control produces substantially no corrective action.

For the process to work and to determine FAC corrections, the engine must be running. The combustion engine may, as desired, be started before or after the electric motor.

The measurement of the corrective torques, both for the resulting corrective torque and for the elementary corrective pairs, can be performed by any means. Thus in a trivial manner a torque meter can be disposed at any point in the drive chain between the heat engine and the electric motor.

According to an advantageous characteristic, if the electric motor allows it, the measurement of the elementary corrective pairs CCE and / or resulting CC is deduced from a control current of the electric motor. Thus, the control current of the electric motor, as produced by the controller performing the speed control, advantageously represents an image of the torque exerted by the electric motor on the heat engine, ie an image of the corrective torque CC, CCE.

According to a possible embodiment, the measurement of the elementary correction pairs CCE can be carried out according to the following steps. A resulting corrective torque CC, either on the angular range corresponding to an engine cycle, or the resulting correction torque DC exerted by the electric motor on the engine is measured by a torque meter or by the control current of the electric motor. The angular range is divided into angular sectors of the same size in a number of sectors equal to the number of cylinders. Each of the angular sectors thus obtained is associated with the cylinder whose propulsive phase is located in said angular sector. The propulsive phase is the explosion phase where the cylinder produces most of its elementary torque CE.

The resulting corrective torque CC is the superposition of the CCE elementary correction pairs associated with all the cylinders. The elementary correction pair CCE associated with a cylinder is equal to the resulting corrective torque CC in the angular sector associated with said cylinder and zero in the other sectors.

FAC correction can be determined by any method.

The correction FAC depends at least on the elementary correction torque CCE, and this according to a function advantageously increasing. This function can be determined theoretically by modeling the heat engine or experimentally bench.

The determination of the FAC correction can be refined by making it depend on one or more operating parameters of the engine. Thus, the function determining the FAC correction may depend, in addition to the elementary correction torque CCE, of a temperature, a pressure, the engine speed, the engine load, etc.

The function giving FAC correction as a function of the value of the elementary corrective couple CCE, and if necessary of at least one other parameter, can be expressed analytically, by means of a mapped chart or by a tabulated function.

The FAC correction is advantageously recorded and indexed according to its determination variables: the elementary correction torque CCE and, if appropriate, said at least one operating parameter of the engine. Thus, this FAC correction can be reused later, typically in the injection step.

With reference to FIG. 5 is illustrated a temporal division of the process which makes appear at least two phases. FIG. 5 shows a time diagram comprising comparatively the resultant motor torque C, superimposition of the elementary pairs CE of the different successive cylinders, the resulting corrective torque CC, superimposition of the elementary corrective pairs CCE of the different successive cylinders and the speed V of the heat engine. At the initial instant to start, the engine starts and operates by applying raw MF fuel masses, as initially determined by the engine control unit, to each of the cylinders. As seen previously, because of the different disturbances, this produces different elementary couples CE, inducing a speed V also disturbed. At the moment ti the electric motor and its servocontrol in speed are started. This has the effect, almost immediate, of balancing the engine. The enslaved electric motor produces a DC correction torque that compensates for the differences between the CE elementary couples. This also has the effect of smoothing the speed V.

During this phase, which can be called learning, the CCE elementary correction pairs are measured, then the FAC corrections are deduced. Depending on the number of parameters considered to define the FAC correction, it may be useful to go through different engine operating areas, in order to determine the FAC corrections for these zones.

During this learning phase, which continues for a moment ta, the first three steps of the process: training, measurement and determination, are performed. This learning phase can be applied continuously. It can still be applied on demand or in specific operating areas. Thus the learning phase is applied in an important way during an initial phase in order to determine a majority of FAC corrections. Then it can be put at least temporarily dormant. It can then be applied again, for example when the engine is operating in a particular zone, not yet explored, such as a high speed, in order to complete the determination of FAC corrections for this operating zone. Once all desired FAC corrections are determined, the learning phase can be definitively stopped, permanently, or for a time. This time can be, for example, the running cycle in progress. The learning is advantageously restarted periodically to take into account the various causes of drift: aging, components, etc. From time t2, FAC corrections are available, and can be applied, superimposed on the MF fuel masses initially calculated by the ECU. An operation phase, including the injection step can then be applied and see injected fuel masses corrected. The operation phase starts at time t2. It is then applied continuously until the engine stops. After t2, since the FAC corrections are applied during the injection step, the corrected torque is such that the different elementary couples CE are substantially equal to each other, from one cylinder to the other. The speed V is then equal to its average value. The electric motor produces substantially no corrective torque CC, CCE.

The electric motor can be a component dedicated to the balancing function of the engine. The only constraint is that it is capable of driving the heat engine, at least during the learning phase. However, it is an expensive component. Also, according to an advantageous embodiment, the invention is applied to a hybrid thermal / electric vehicle and the electric motor reuses the pre-existing electric machine of such a vehicle.

It has been seen that the exploitation phase extends from t2. The learning phase extends from ti to ta. These two phases can coexist, between t2 and ta. As soon as FAC corrections are available they can be used in operation. Coexistence is possible because an absence or inaccuracy of FCC corrections is of no consequence. The enslaved electric motor compensates for this absence / inaccuracy by producing a DC correction torque and ensuring balancing.

Also, in such a case, learning can continue or resume, in parallel, during operation. Because of the servo speed, this is automatic. As soon as a FAC correction is incorrect, the electric motor produces a correction torque and a new adapted FAC correction is determined, as necessary. Such a need may arise, for example, when the engine comes to operate in an operating zone not yet encountered. At the instant ta, it is eventually considered that the learning phase is over, because all FAC corrections are assumed to be determined. The electric motor and / or its servo are no longer necessary and can be deactivated. A deactivation is understood here electric (power off) and / or mechanical (uncoupling, disengagement). Disabling assumes that the motor is no longer used for balancing. However, it can be used for another function, such as its primary function of hybridization. The invention also relates to a device, typically an electronic control unit and / or computer, for example an engine control unit, including its program, capable of implementing such a balancing process. The invention also relates to a vehicle comprising such a device.

Claims (12)

1. A method of balancing a multicylinder heat engine, characterized in that it comprises the following steps: • driving of the engine by means of an electric motor speed-controlled speed (V) of the engine, • measuring, for each cylinder, an elementary corrective torque (CCE) associated with said cylinder, exerted by the electric motor on the heat engine, • determining, for each cylinder, a correction (FAC) of the fuel mass ( MF) to inject, able to compensate for said elementary correction torque (ECC). 2. Method according to claim 1, further comprising a following step: injection, into each cylinder, a fuel mass (MF) corrected for said correction (FAC). 3. Method according to one of claims 1 or 2, wherein the measurement of the elementary correction pairs (CEC) is deduced from a control current of the electric motor. 4. Method according to any one of claims 1 to 3, wherein the measurement of the elementary correction pairs (CCE) comprises the following steps: • measurement of a resulting corrective torque (CC), • equipartition of an angular range corresponding to a motor cycle in a number of sectors equal to the number of cylinders, • association with each cylinder of the sector in which said cylinder is propellant, the elementary corrective torque (CCE) associated with a cylinder being equal to the resulting corrective torque (CC) in the associated sector and zero in the other sectors. 5. Method according to any one of claims 1 to 4, wherein the correction (FAC) is determined as a function of the elementary correction torque (ECC) and possibly at least one operating parameter of the engine. 6. Method according to any one of claims 1 to 5, wherein the determined correction (FAC) is recorded, and indexed as a function of the elementary correction torque (CCE) and, if appropriate, of said at least one operating parameter of the engine, for be reused later. A method according to any one of claims 1 to 6, wherein a learning phase, comprising the training, measuring and determining steps, is applied continuously or on demand or only in specific operating areas or only in an initial phase until all desired corrections (FAC) are determined. The method of any of claims 2 to 7, wherein an exploitation portion, including the injecting step, is continuously applied. 9. A method according to any one of claims 1 to 8, wherein the electric motor is an electric machine of a hybrid thermal / electric vehicle. 10. Method according to any one of claims 1 to 9, wherein the electric motor is no longer used for balancing, when the elementary correction pairs (CEC) are substantially compensated by the corrections (FAC). 11. Device comprising an electronic control unit, and means for implementing a method according to any one of the preceding claims. 12. Vehicle comprising a device according to the preceding claim.