Patent Grant
10830165
This Patent application claims priority from Italian Patent Application No. 102018000001107 filed on Jan. 16, 2018, the entire disclosure of which is incorporated herein by reference.
The present invention relates to the field of methods and systems for the identification of disturbances and for the consequent control of the internal engine to suppress such disturbances.
Internal combustion engines include one or more pistons associated with as many cylinders. The relative reciprocating motion is transformed into a rotary motion by means of a known crank mechanism.
Given a predetermined engine capacity, an advantage of fractioning said capacity is the easy balancing of the alternating and rotating masses.
The disadvantage of a multi-cylinder engine obviously lies in the higher cost.
With respect to this cost, it is important to ensure an optimum engine balancing under operating conditions, especially when the engine includes six or more cylinders and is intended for a prestigious as well as a sporty vehicle.
The internal combustion engine balancing depends on several factors:
An efficient balancing of an internal combustion engine also allows reducing the vibrations it transmits to the relative supports that connect the internal combustion engine to the body of the vehicle and to the transmission.
A lower intensity of the vibrations allows reducing or simplifying the flywheel with an overall reduction of the mass of the vehicle. This reduction, in turn, saves fuel and improves the vehicle performance.
The object of the present invention is to provide a balancing system designed to compensate or at least reduce any imbalances in the rotation of an internal combustion engine equipped with a plurality of cylinders.
The basic idea of the present invention is to detect an imbalance of the torque delivered by a single cylinder with respect to the remaining cylinders by means of a frequency analysis of a first signal representative of the rotation speed of the drive shaft when this speed is substantially constant. In particular, said first signal is converted into a corresponding signal in the frequency domain and a predetermined frequency band is identified. When the amplitude of said second signal in said predetermined frequency band exceeds a predetermined threshold, then a phase of said second signal is calculated in said second predetermined band.
This identification is carried out so as to save only the half-order frequency, namely the one obtained for a frequency equal to half the nominal frequency corresponding to the rotation speed of the drive shaft.
In other words, only a so-called band pass filter tuned to the half-order frequency can be used.
This phase identifies a specific cylinder that causes said imbalance, namely that introduces a disturbance.
It is clear that when it is stated that the speed of the engine is substantially constant this is true with the exception of the unavoidable imbalances due to the crank mechanisms and of the possible presence of a disturbance caused by a cylinder.
When there is a positive ignition engine, one or more strategies for correcting said imbalance or disturbance might be implemented by varying the ignition advance of one or more cylinders.
On the other hand, when the engine is a diesel cycle engine, it is not possible to intervene on the advances, since the ignition of the mixture is spontaneous, but it is possible to intervene on the supply of one or more cylinders of the internal combustion engine.
Thanks to the present invention, it is possible to obtain a better balance of the internal combustion engine, bringing said amplitude of said second signal back into said predetermined band below said predetermined threshold.
The engine rpm signal is generally present in the engine control unit, therefore no variation of the existing hardware is required.
Moreover, the method object of the present invention can advantageously be implemented in a processing unit designed to control the internal combustion engine modified/programmed to be a further object of the present invention.
The claims describe preferred variants of the invention forming an integral part of the present description.
Further objects and advantages of the present invention will become clear from the following detailed description of an embodiment thereof (and of its variants) and from the annexed drawings, given purely by way of an illustrating and non-limiting example, in which:
The same reference numbers and letters in the figures identify the same elements or components.
In the present description, the term “second” component does not imply the presence of a “first” component. These terms are in fact used only for clarity's sake and are not to be meant in a restrictive way.
According to the present invention, the internal combustion engine comprises a plurality of cylinders. For example,
The cylinders are connected to a single drive shaft CS to which a so-called phonic wheel is rigidly associated, to which a sensor S is in turn associated that is able to measure an (angular) speed of the drive shaft. Generally, it uses the Hall effect, but other technologies can be used.
An ECU (Engine Control Unit) processing unit is operatively connected to said angular speed sensor S.
Disturbance Detection
The processing unit is programmed to acquire the signal generated by said sensor and to convert it into a signal in the frequency domain, for example by the Fourier transform.
They are preferably processed acquisition ranges of said signal in which the same signal is averagely constant.
A first ideal analysis condition can coincide with the idle condition of the internal combustion engine.
Other ideal conditions can be identified when the angular speed of the drive shaft is averagely constant over a predetermined time range, for example, when the vehicle is traveling at a constant speed.
A nominal frequency is defined based on the average angular rotation speed of the drive shaft, for example, 850 rpm corresponds to a nominal frequency of 14.17 Hz.
It is defined as “half-order frequency” that frequency corresponding to half the nominal frequency, e.g. 14.17/2=7.08≈7 Hz. It is sometimes called “half-order harmonic”.
Examples of frequency analysis of the angular speed of the internal combustion engine are given in
According to the present invention, the signal in the frequency domain is analysed, preferably by the Fourier transform, at said half-order frequency.
When the amplitude of the half-order frequency exceeds a predetermined settable threshold Th, then an imbalance condition of the internal combustion engine is detected.
It is therefore necessary to identify the cylinder causing this imbalance.
It is known that the Fourier transform is a function of the time domain with values in the complex plane. Therefore, it is possible to calculate the so-called phase spectrum of the function as arctan (Im(f)Re(f)) and in particular the phase fi(f) of the disturbance by substituting said half-order frequency to the variable f, e.g. 7 Hz if the engine is idling.
Once obtained a phase value according to the present invention, said phase is correlated with the cylinder causing this imbalance. In this regard, the plan is divided first into contiguous angular sectors numbered according to the same ignition sequence of the cylinders starting from a reference phase f (0), indicated in the Figure by the axis 0. An example is given in
Starting from said reference axis, the cylinder i is identified as the cause of this imbalance, which is the one whose angular sector comprises the aforesaid phase fi (f/2).
Preferably, the cylinder is identified when a counter exceeds a settable threshold, i.e. when the phase is stable after a predetermined number of iterations. This counter is reset every time the phase changes angular sector: in this way the identification takes place only after a sufficiently long time in order to reliably associate the calculated phase with the cylinder i.
If the operating parameters of the cylinders are identical for all the cylinders, i.e. if no previous correction has been made, then the procedure for the identification of the cylinder causing the aforesaid disturbance is considered terminated.
If, on the other hand, the engine operating parameters have been altered by a previous disturbance detection procedure followed by a disturbance suppression procedure, then it is possible to proceed in two ways:
Then a (new) disturbance correction procedure or a new reset procedure of any disturbance correction in progress can be started.
From what described above, it is clear that the initial problem may spontaneously end, e.g. due to impurities in the fuel that selectively hit a cylinder.
In such circumstances, a possible suppression strategy can cause a disturbance with an increase in the amplitude of the half-order frequency associated with a cylinder, which delivers, for example, more torque than the remaining cylinders.
It has been observed that this results in a disturbance given by the cylinder opposite the i-th cylinder, in the phase diagram shown in
If a previous correction made on the cylinder i were not taken into account, the procedure would then correct the torque delivered by the cylinder j. This might lead to the suppression of the imbalance, but it is however preferable to operate again on the cylinder i to guarantee an even distribution of the torque delivered by all the cylinders, i.e. avoiding that the cylinders i and j deliver a torque different from the one delivered by the others cylinders.
The second option b) is advantageous because it allows obtaining a quicker correction starting from the last correction carried out following the detection of the cylinder i as a cause of disturbance.
In short, the method for the identification of a torque delivery imbalance of an internal combustion engine equipped with two or more cylinders under constant speed operating conditions according to the present invention comprises in succession the following steps:
Subsequently, to detect the cylinder causing the imbalance, possibly to correct it or for a subsequent statistical processing, the following steps are carried out in succession:
Based on the type of internal combustion engine, the ECU processing unit can intervene on an operating parameter of a cylinder:
When the internal combustion engine is a positive ignition internal combustion engine and includes a post-treatment system of the exhausted gas allowing a lean combustion (lambda higher than 1), then either of the two or both of the previous strategies can be used in combination with each other.
The concept of ignition advance is known to the person skilled in the art and refers to reaching the PMS piston.
With regard to fuel injection, the timing of fuel injections and/or the total mass injected in each combustion cycle may be varied.
The effected correction aims at obtaining a variation of the torque delivered by a cylinder that is judged to be the cause of the disturbance. Said variation is made in a discrete manner, namely with predetermined and adjustable amplitude corrections of an operating parameter of said cylinder.
The amplitude of the half-order frequency is monitored regardless of the type of correction carried out. This provides a feedback on the exactness of the correction. If the amplitude is reduced, then the correction sign is correct, otherwise the correction sign is inverted until its amplitude is lower than said predetermined settable threshold Th. Therefore, the present suppression method is a closed loop control method.
In particular, it is acquired a first trend of said amplitude or a parameter representing it in a time range preceding a correction, then said correction is applied and it is acquired a second trend of said amplitude or a parameter representing it in a time range following said correction and comparing said first and second trends or the corresponding parameters. In particular, the amplitudes or the integrals of the trends are compared.
Therefore, according to the present invention, a disturbance suppression introduced by the previously identified cylinder is corrected by means of one of the following steps:
Preferably, the method is performed in closed loop and therefore also the following steps are carried out in succession:
It is clear that the correction is made in a recursive manner. Therefore, it is preferable that, if two consecutive inversions of the correction sign are obtained, then the algorithm can stall and it is preferable to reset any correction and restart from the beginning.
To reset any correction means referring to a set of reference parameters (set-point).
Based on the result of the comparison of the amplitudes related to the first range and to the second range, the correction continues by using the same correction sign or by inverting the correction sign until the amplitude of the half-order frequency is again below the threshold Th.
In other words, if the correction reduces the amplitude of the disturbance, it continues without inverting the correction sign, otherwise the sign is inverted and the correction is continued until the amplitude of the half-order frequency is reduced below the threshold Th.
According to a preferred variant of the disturbance suppression method, said first trend is low pass filtered in order to identify an improving or worsening trend of the correction action carried out with respect to the first trend acquired before the correction.
A similar comparison can be made by comparing the integrals of said first and second trends. Preferably, said integrals are obtained after the low pass filtering of both trends.
Later, various disturbance suppression techniques are described that operate on the only cylinder identified as disturbing and/or on the remaining cylinders.
In any case, the disturbance suppression provides a correction (ixa) of the torque delivered by said cylinder i and/or a correction (ixb, ixc) of the torque delivered by the remaining cylinders.
This torque correction can be positive or negative.
It is clear that, in the event of a disturbance correction, in order to decide to operate on the opposite cylinder or on the remaining cylinders rather than on the one identified by means of the aforesaid identification procedure, it is advisable to memorize (x) by relating the following information:
So, if it turns out that the identified cylinder is the one that had been previously corrected, its correction is reduced. In other words, a relative operating parameter is further corrected by starting the procedure with the inversion (xiv) of the sign of the last correction.
Once the amplitude of the disturbance has fallen below the threshold, the suppression procedure ends, maintaining the last correction, while the monitoring of the amplitude of the half-order frequency can continue.
It may happen that the causes that generated an imbalance end while a suppression strategy is still active.
In this case, it should be checked whether the cylinder causing the disturbance is not the cylinder opposite—in terms of angular sector—a cylinder that had been previously corrected. In this case, it is preferred, as described above, to operate on the cylinder that had been previously corrected by gradually reducing the correction previously made to a related control parameter.
The use of the flag is convenient for detecting when a correction procedure on a cylinder is completed or evolving.
Therefore, when step 15 reports “first cylinder i previously subjected to correction”, it means that its correction has been completed by setting the flag to zero.
According to the notation used, i=j or flag=1 represents a value assignment, while “==” represents a comparison.
Advance Control
Idling
When the disturbance detection procedure is performed at idle speed and the engine is a positive ignition engine, said change in the torque delivered by the cylinder i is obtained by correcting the ignition advance of the same cylinder as long as the engine is idling.
Therefore, this check is carried out integrally and exclusively as long as the engine is idling. Any correction of the operating parameters of the cylinders is cancelled by returning these parameters to their nominal values when the rotation speed of the engine exceeds a predetermined threshold such that the engine is not considered at idle speed.
Therefore, after the cylinder is identified, the advance is corrected to increase the torque delivered by the cylinder i and based on the obtained result the strategy continues
a. (ix) by further correcting without varying the sign of correction;
b. (xiv) by inverting the sign of correction and correcting further;
as long as the amplitude of the half-order frequency is below the threshold Th.
Out of experience, the ignition advance is initially increased.
In any case, the present method allows checking whether the intervention is correct and in case inverting the correction sign.
In the event that a second cylinder j starts to deliver less torque, the strategy is also able to intervene on the second cylinder by correcting the delivered torque.
As previously stated, if the second cylinder j is opposite the first cylinder i, it is preferable to operate backwards, i.e. by inverting the sign of the last correction made on the cylinder i until the amplitude of the half-order frequency falls below the threshold Th.
Alternatively, all previous corrections can be cancelled and the analysis can be restarted from the beginning.
Constant and Higher Than Idling Rotation Speed
Another ideal condition to detect a torque delivery imbalance occurs when the angular speed of the drive shaft is higher than the idling speed and is averagely constant over a predetermined time range.
This time range must be long enough to allow the acquisitions necessary for steps 10 and 17.
This control strategy, unlike the previous one, can be performed under any operating condition of the internal combustion engine.
Moreover, this control strategy can be combined with the previous one, which instead must be actuated at idling speed. This means that at idling speed the previous control strategy prevails, whereas the present control strategy works at engine speeds higher than the idling speed.
When the internal combustion engine is a positive ignition engine and it is not possible to operate on the fuel injection since the post-treatment system of the exhausted gas does not allow it, then the advance of the cylinder i or of the relative opposite cylinder j and/or of the remaining cylinders may be gradually varied.
If it is possible to increase the advance on the cylinder i, then the procedure corrects the advance of the cylinder i both in positive and in negative, as in the previous example.
If, however, it is not possible to increase the advance on the cylinder i, because for example the engine calibration is already at the limit, then the procedure reduces the advance of the cylinder j opposite the cylinder causing the disturbance.
The diagram of
Alternatively or in combination, it is possible to reduce the advance also on the remaining cylinders with respect to the cylinder i and j.
For example, the torque delivered by the cylinder no. 5 can be reduced by delta and the torque delivered by the remaining cylinders no. 2-no. 4, no. 6-no. 8 can be reduced by delta/h, where h is a parameter that can be calibrated, e.g. it can be equal to 2 or equal to the number N−1, where N represents the number of cylinders, which in the example is equal to 8.
For example, if the cylinder no. 1 (i) causes an imbalance, because for example it delivers less torque than the others, then the torque delivered by the cylinder no. 5 can be gradually reduced up to a certain reduction threshold.
Subsequently, the procedure operates on all the remaining cylinders by reducing their relative advance.
This not only reduces the amplitude of the imbalance, but also leads it to shift towards higher frequencies that are therefore more easily dampened.
It is clear that in a second correction procedure it may occur that the previously corrected cylinder no. 5 delivers less torque than the cylinder no. 1.
By starting the correction procedure its advance may be increased. In this case the procedure operates directly on the cylinder no. 5 and not on its opposite cylinder no. 1.
Therefore, with reference to
Injection Control
The fuel injection control can be advantageously performed in diesel cycle internal combustion engines as there are no stringent restrictions in the stoichiometry of the combustion or in the positive ignition engines with an ATS that allows them to operate even in conditions of lean mixture.
This control strategy can be applied both at idling speed and at other rpms.
The fuel injection can be corrected by checking the fuel microinjections, their timing and/or the total mass of fuel injected at each combustion cycle in the cylinder i and/or in the remaining cylinders.
Since additive correction strategies can be implemented both at idling speed and at other rpms for each cylinder, then it is preferable to always operate on the cylinder i when the disturbance can be suppressed by increasing the delivered torque.
Regarding the identification of the cylinder causing the disturbance, also in this case one of the aforesaid options a) or b) is applied.
Therefore, when, following a first disturbance suppression, it should turn out that the same suppression is the cause of a consequent disturbance leading to the cylinder j opposite the previously corrected cylinder, then according to option b) the correction previously made to the fuel injection map of said cylinder is gradually reduced.
The present invention can be advantageously implemented by means of a computer program, which comprises coding means for implementing one or more steps of the method, when this program is run on a computer. Therefore, the scope of protection extends to said computer program and also to computer readable means comprising a recorded message, said computer readable means comprising program coding means for implementing one or more steps of the method when said program is run on a computer.
Modifications to the embodiments of the described non-limiting example are possible without departing from the scope of the present invention, including all the equivalent embodiments for a person skilled in the art.
From the above description, the person skilled in the art is able to manufacture the object of the invention without introducing further construction details. The elements and features illustrated in the various preferred embodiments, including the drawings, may be combined with each other without however departing from the scope of protection of the present application. What has been described in the part relating to the state of the art only provides a better understanding of the invention and does not represent a declaration of existence of what has been described. Furthermore, if not specifically excluded in the detailed description, what has been described in the part relating to the state of the art is to be considered as an integral part of the detailed description.
| Number | Date | Country | Kind |
|---|---|---|---|
| 102018000001107 | Jan 2018 | IT | national |
| Number | Name | Date | Kind |
|---|---|---|---|
| 4301678 | Full | Nov 1981 | A |
| 20020148441 | Tuken | Oct 2002 | A1 |
| 20030089338 | Remele et al. | May 2003 | A1 |
| 20050229904 | Hagel et al. | Oct 2005 | A1 |
| 20070175443 | Schueler | Aug 2007 | A1 |
| 20080276697 | Ostman | Nov 2008 | A1 |
| 20090037060 | Carlhammar | Feb 2009 | A1 |
| 20100039076 | Trainer | Feb 2010 | A1 |
| 20140347043 | Chana | Nov 2014 | A1 |
| 20170122839 | Stola | May 2017 | A1 |
| 20170241369 | Jammoussi | Aug 2017 | A1 |
| 20170254726 | Das et al. | Sep 2017 | A1 |
| 20200040988 | Duan | Feb 2020 | A1 |
| Number | Date | Country |
|---|---|---|
| 2843219 | Mar 2015 | EP |
| Entry |
|---|
| Search Report for IT Application No. 2018000001107, search completed Sep. 7, 2018 (8 pages). |
| Number | Date | Country | |
|---|---|---|---|
| 20190218982 A1 | Jul 2019 | US |
