Patent Grant
10126203
The present application claims priority to European Patent Application No. 14174739.4 filed Jun. 27, 2014, the entirety of the disclosures of which are expressly incorporated herein by reference.
Not Applicable.
The present invention relates to the field of the monitoring system of the vehicular component, and in particular to a system for detecting an air filter condition.
Software functions for the detection of clogged air filters are known.
Usually, the detection is performed by comparison of the pressure drop over the filter with a certain, maximally permitted threshold value. If the threshold is permanently being exceeded by the measured pressure drop, an error is detected and appropriate countermeasures will be taken by the control unit, e.g. a warning lamp, reduction of engine power, etc.
Sometimes, such detection strategy is not affordable.
Situations are known, depending on engine layout and air flow, where the variation of pressure drop does not change sensitively from an optimal condition to a rather clogged condition. For example the pressure drop at the air filter could be about 10-15 mbar for a new filter cartridge that reaches about 50 mbar for a severely clogged condition.
In addition, due to the strong variability of the ambient pressure, from low pressure of about a few hundred at high altitude, to high pressure of about 1100 mbar at the sea level, these sensors must have very high accuracy for a large pressure variation.
Automotive pressure sensors have typical accuracy of 2% of the full scale, i.e. here about 20 mbar×2 sensors, i.e. 40 mbar in the worst case. Thus the error extent is comparable with the pressure drop change. Thus, it is impossible to detect a clogged filter reliably with such known approach.
Therefore it is the main object of the present invention to provide a system for detecting an air filter condition, in particular for combustion engines, affordable even in those situations where the pressure drop variation between two extremes conditions, namely new and clogged, is really limited.
The main principle of the invention is to base the condition estimation of the air filter on a linear regression calculation of pressure drop values at the filter versus corresponding squared air volume flow crossing the air filter. Thus, the angular coefficient of the linear regression or a function construed on it is compared with one or more thresholds in order to recognize said operative condition of the air filter.
It is clear that either the angular coefficient or a function of it represent one numeric value.
According to a preferred embodiment of the invention, the angular coefficient is used to calculate the filter pressure drop at a predefined highest theoretical volume flow through the air filter, and compare this corresponding pressure drop with a threshold pressure drop to determine the operative filter condition. Therefore, said pressure drop, calculated on said angular coefficient and on said predefined highest theoretical volume flow, is a numeric value obtained as a function of said angular coefficient value, preliminarily calculated.
The first object of the present invention is a method for detecting an air filter condition, in particular for combustion engines.
Such method is capable to give a reliable picture of the clogging condition of the engine air flow filter.
According to an embodiment of the invention, the method is also suitable to recognize the improper removal or the tampering of the air filter, or, as alternative, a leakage through the engine intake manifold.
Another object of the present invention is a system implementing the abovementioned method.
Another object of the present invention is a combustion engine comprising such system.
A further object of the present invention is a terrestrial vehicle comprising such engine.
These and further objects are achieved by means of the attached claims, which describe preferred embodiment of the invention, forming an integral part of the present description.
The invention will become fully clear from the following detailed description, given by way of a mere exemplifying and non limiting example, to be read with reference to the attached drawing figures, wherein:
The same reference numerals and letters in the figures designate the same or functionally equivalent parts.
According to the present invention, the term “second element” does not implies the presence of a “first element”, first, second, etc. are used only for improving the clarity of the description and should not be interpreted in a limiting way.
According to the present invention, the variation of the pressure drop Δp over the air filter versus the variation of mean gas velocity through the same filter is considered. For a given filter geometry, the gas velocity can also be expressed through the air volume flow {dot over (V)}. While V indicates the volume, the dot above the same symbol indicates the time derivative of the air volume, or simply air flow {dot over (V)}.
According to the present invention, it has been found that dependency of pressure drop Δp over the air filter on the squared volume flow {dot over (V)}2 is linear, and the two values are interrelated through a factor k:
Δp=k·{dot over (V)}2 Eq. 1
Factor k corresponds to the angular coefficient of a linear regression calculated on said interrelated values Δp, {dot over (V)}2. It includes all information regarding the degree of frictional loss through the flow restriction; here, it either reflects the degree of filter clogging through pollution, or indicates a not properly working filter if less than expected for a clean filter cartridge.
If k or f(k) exceeds a predefined upper threshold ThU, namely when k or f(k)>ThU: YES, then the air filter is acknowledged as clogged, and preferably a DFC (ECU internal diagnostic fault code) is set after some error debouncing.
If, on the contrary, k or f(k) is less than a predefined lower threshold ThL, a DFC for a possibly wrong, or not correctly mounted or removed or tampered filter cartridge is set, or for a leaking air supply system. Also in this case, the DFC is preferably set after some error debouncing.
By setting ThL<Th1< . . . <Thn<ThU, n≥2, an arbitrary number of different interim thresholds can be defined, which allow determination of the state of the air cleaner with higher resolution. As an example, with three different thresholds ThL, Th1 and ThU, thanks to the present method it is possible to distinguish the following conditions:
ThL<k<Th1: filter is acknowledged as new and/or in good conditions;
In addition, the change of k or f(k) within a small time interval can be exploited to derive additional information. If, referring to above example, the k factor suddenly passes from one of the following conditions Th1<k<ThU or ThU<k to the condition k<ThL, such event is interpreted as a tampering approach: the filter has probably been removed to get rid of the previous errors. Thus, preferably the ECU turns the engine into a recovery mode, with a limited power/torque/speed.
As alternative approach, k can also serve as a direct measure for the state or clogging degree of the filter. The actual k value represents a percentage value between a predefined minimum kMin and a predefined maximum kMax:
This approach leads to a comparison of a function of k with the thresholds kMin and Kmax.
The continuously determined filter clogging degree can be used to apply protective countermeasures, like an analogous maximal power reduction, or a threshold logic like the one described above.
Now, we show the application of the present invention to another function of k. As example it can be calculated maximal pressure drop □pmax over air cleaner, which would be present at predefined, maximally expected air flow through the filter {dot over (V)}max:
Δpmax=k·{dot over (V)}max2 Eq. 3
Both a threshold approach and a continuous approach like in the abovementioned examples works also for □pmax=f(k).
Thus embodiment is the most robust with respect to the other approaches according to the present invention.
According to a preferred embodiment of the present invention, the pressure drop over air cleaner is continuously measured at the sample rate offered by the electronic controller, for example every 10 ms.
The pressure drop over air cleaner is for example measured by a differential pressure sensor over air cleaner, or by an arrangement of absolute pressure sensors up- and downstream air cleaner. In the second case, the pressure difference (drop) is calculated by subtracting the measured absolute pressure after air cleaner (before an optional turbo charger) from an ambient pressure value, acquired by another sensor on-board the vehicle, for example embedded in the ECU.
The air volume flow {dot over (V)} through the air cleaner is either directly measured by means of an appropriate measuring device in the air flow path, or calculated by the electronic controller, based on other physical measurements like e.g. pressures, temperatures, lambda values, and predefined data, like e.g. engine geometry.
The raw data can be permanently stored within the ECU for later evaluation, for example, if they are in a useful volume flow range. According to a preferred embodiment of the invention the applicable range [{dot over (V)}min, {dot over (V)}max] of air flows or alternatively [{dot over (V)}min2, {dot over (V)}max2] of squared air flows is subdivided into a predefined number of adjacent intervals, for example 8 of the same size, marked through indices 0 to 7. Then, the couples of values are acquired (Δp,{dot over (V)}2) and grouped on the basis of the respective interval, 0-7, belonging the air flow value and associated to the respective interval. For convenience, such couples are called as “data-couple(s)” in the following.
When an air flow value lies outside such applicable range of air flows, namely when it is lower or higher than the applied min/max values, respectively, it is preferably disregarded or recorded for different uses, namely to recognize abnormal conditions.
In the context of the present invention “applicable range” means a range of air-flow defined by the kind of combustion engine considered within its functioning range in normal conditions.
The aforementioned grouping of the data-couples can be advantageously recorded within two data-matrixes having size (at least) n×8. When an applicable minimum number of data-couples has been updated, a linear regression calculation is performed in order to calculate the k factor:
where n is the number of updated data-couples, {dot over (
In order to save storage memory, just two 1×8 arrays can be implemented, by storing, step by step, only said {dot over (
Advantageously, the subdivision of the applicable range of air flows into, preferably 8, “discrete intervals” confers high calculation robustness to the method.
However, a different number of intervals can be accounted for or a different mathematical strategy, per se known, that does not require such range subdivision can be implemented.
The memory arrays can be reset periodically, for example at any engine switching-off, and also after every calculation of k according to Equ. 4, or function of k like in Equ. 3. After that, the data are considered as “used” and reset, i.e. overwritten by the next new set of data.
Especially, in this latter specific case, several tests have been carried out and it has been found that the Equ. 4 could lead to unreliable results with huge errors when the air flow measurements are too close together, this could happens in specific application where the engine is called to run almost stationary.
Thus, the abovementioned range subdivision permits to inhibit the calculation of the Equ. 4, when the acquired values of air-flows are not sufficiently spread. Preferably, the method comprises a step to check that at least 4-5 of the 8 groups associated to the corresponding intervals contain at least one data-couple.
This invention can be implemented advantageously in a computer program comprising program code means for performing one or more steps of such method, when such program is run on a computer. For this reason, the patent shall also cover such computer program and the computer-readable medium that comprises a recorded message, such computer-readable medium comprising the program code means for performing one or more steps of such method, when such program is run on a computer.
Many changes, modifications, variations and other uses and applications of the subject invention will become apparent to those skilled in the art after considering the specification and the accompanying drawings which disclose preferred embodiments thereof. All such changes, modifications, variations and other uses and applications which do not depart from the scope of the invention are deemed to be covered by this invention.
Further implementation details will not be described, as the man skilled in the art is able to carry out the invention starting from the teaching of the above description.
| Number | Date | Country | Kind |
|---|---|---|---|
| 14174739 | Jun 2014 | EP | regional |
| Number | Name | Date | Kind |
|---|---|---|---|
| 8574355 | Jang | Nov 2013 | B2 |
| 8613792 | Ragland | Dec 2013 | B2 |
| 20050247194 | Kang | Nov 2005 | A1 |
| 20110146246 | Farman | Jun 2011 | A1 |
| 20130192208 | Shibata | Aug 2013 | A1 |
| 20150198076 | Li | Jul 2015 | A1 |
| 20150254958 | Sherman | Sep 2015 | A1 |
| 20150275738 | Van Nieuwstadt | Oct 2015 | A1 |
| 20150354480 | Dudar | Dec 2015 | A1 |
| Number | Date | Country |
|---|---|---|
| 10260784 | Jul 2004 | DE |
| 2604818 | Jun 2013 | EP |
| 2620202 | Jul 2013 | EP |
| Number | Date | Country | |
|---|---|---|---|
| 20160018288 A1 | Jan 2016 | US |
