The present invention pertains to combustion engines, and in particular to a method for the control of a combustion engine according to the preamble of claim 1. The invention also relates to a system and a vehicle, as well as a computer program and a computer program product, which implement the method according to the invention.

Background of the invention The background description below constitutes a background description for the invention, and thus need not necessarily constitute prior art technology.

In relation to vehicles in general, and at least to some extent heavy goods vehicles in particular, development is constantly ongoing in the quest for fuel efficiency and reduced exhaust emissions. Due to e.g. increased government interests concerning pollution and air quality in e.g. urban areas, emission standards and regulations have been drafted in many jurisdictions. When heavy goods vehicles are driven, such as cargo vehicles, buses and similar, vehicle economy has over time gained an increasing impact on profitability in the business where the vehicle is used. The main expenditure items for the day to day operation of a vehicle consist, apart from the cost of acquisition of the vehicle, of the vehicle

driver's salary, costs of repair and maintenance, and fuel for driving the vehicle. Thus it is, within each one of these areas, important to attempt to reduce the costs to the extent possible .

In addition to economic/environmental aspects, according to the above, there are also additional aspects which should be taken into consideration in the construction of vehicles. For

RECORD COPY TRANSLATION example, driver comfort is important, perhaps in particular in heavy goods vehicles, and much work is also often invested in the driver environment. This includes work relating to noise comfort, i.e. minimising/optimising of primarily unwanted sound/noise which the driver is subjected to when driving the vehicle, where loud or otherwise disturbing noise may have a negative impact on the driver's driving of the vehicle, e.g. by causing stress and/or tiredness.

Another aspect consists of the noise which the vehicle emits in its environment, i.e. how the vehicle's progress is experienced sonically in the environment where the vehicle is driven. For example, there may be laws and regulations also in this regard, regulating permitted noise emissions from

vehicles . Summary of the invention

One objective of the present invention is to provide a method to control a combustion engine. This objective is achieved with a method according to claim 1.

The present invention pertains to a method for the control of a combustion engine, where said combustion engine comprises at least one combustion chamber and elements for the supply of fuel to said combustion chamber, and wherein the combustion in said combustion chamber occurs in combustion cycles. The method is characterised in that: - during a first part of a first combustion cycle, at least one first parameter value with respect to a physical quantity for combustion in said combustion chamber is determined,

- based on said first parameter value, a representation of a resulting pressure amplitude is estimated, during a subsequent part of said first combustion cycle, after said first part of said first combustion cycle and in said combustion chamber, such as a maximum pressure amplitude, and

- based on said estimated pressure amplitude, subsequent combustion is controlled, where said subsequent combustion consists of a subsequent part of said first combustion cycle, after said first part of said first combustion cycle.

As mentioned above, the noise which is generated when driving a vehicle and which many times to a great extent is considered to be unwanted noise, is an important parameter not only in an effort to achieve a good driver environment, but also from the point of view of the environment in which the vehicle is being driven .

As is known, there are many sources of sound and noise in vehicles, and a main source consists of the combustion engine. The noise to which a combustion engine gives rise depends to a great extent on the combustion in the combustion engine's combustion chamber, and primarily on the manner in which pressure changes during combustion. The noise arising depends at least partly on the maximum pressure amplitude, i.e. the maximum pressure, which arises during combustion. Noise also arises as a result of pressure changes, and in particular when the pressure rises quickly.

According to the present invention, the combustion is

controlled with respect to the pressure level which arises during combustion, e.g. through a control action aiming to limit the maximum pressure which may arise during a combustion (combustion cycle) .

According to one embodiment, the manner in which the pressure changes during the combustion is also controlled, in

particular during an ongoing pressure increase, and in

particular by a control action which aims to limit the maximum speed of pressure change arising at combustion. The control of the combustion may be arranged to be carried out individually for each cylinder, and the combustion may be controlled for a subsequent combustion cycle based on

information from one or several previous combustion cycles. According to one embodiment, a representation of the maximum pressure amplitude which is expected to result during a combustion cycle is predicted, so that the combustion in a subsequent combustion cycle is controlled based on this estimation, and the control action in the subsequent

combustion cycle may be adapted to avoid e.g. an unwanted high pressure amplitude.

According to one embodiment, an ongoing combustion is

controlled during a combustion cycle, so that the invention provides control of an ongoing combustion process, where the control action may be carried out during an ongoing

combustion, with the objective to e.g. prevent an unwanted high pressure amplitude from arising.

Control according to the present invention may be achieved by, during a first part of a combustion cycle, e.g. when the combustion cycle has been started, determining a parameter value relating to a physical quantity for the combustion, such as a pressure prevailing in the combustion chamber. Based on this parameter value, e.g. the prevailing pressure, an

expected maximum pressure (maximum pressure amplitude) for a subsequent part of said first combustion cycle may then be predicted through an estimation, where the combustion during a subsequent part of the combustion cycle may be controlled with respect to the expected maximum pressure amplitude. Said first parameter value thus constitutes a representation of an actually prevailing condition for said physical quantity at a time/crank angle position when said first combustion cycle has been initiated, and, according to one embodiment, at a time/crank angle position when the combustion of fuel has been initiated. Said parameter value corresponding to said first parameter value may also be arranged to be determined at a number of times/crank angle positions after the combustion of fuel has been initiated during said first combustion cycle, for use in connection with control during said first

combustion cycle. According to one embodiment, an expected maximum pressure increase speed is also estimated, so that control may also occur with respect to the latter. Said first parameter value may be determined through the use of sensor elements such as pressure sensor elements.

The combustion may e.g. be controlled by determining an injection strategy for application at a subsequent injection during the combustion cycle, where an expected maximum

pressure amplitude may be estimated at the determination of an injection strategy, where an injection strategy - such as one injection strategy of several injection strategies - may be selected, where an injection strategy is selected which is not expected to result in an unwanted pressure development during the combustion. For example, an injection strategy may be selected which is expected to result in a maximum pressure amplitude below some applicable threshold for the maximum pressure, where such threshold amounts to an applicable maximum pressure which e.g. is expected to result in a noise level emitted, which in its turn is below some applicable noise level, or meets another criterion relating to emission of noise.

The method according to the present invention may e.g. be implemented with the help of one or several FPGA (Field- Programmable Gate Array) circuits, and/or one or several ASIC (application-specific integrated circuit) circuits, or other types of circuits which may handle the desired calculation speed .

Further characteristics of the present invention and

advantages thereof will be described in the detailed

description of example embodiments set out below and in the enclosed drawings.

Brief description of drawings

Fig. 1A shows schematically a vehicle in which the present invention may be used. Fig. IB shows a control device in the control system for the vehicle shown in Fig. 1A.

Fig. 2 shows the combustion engine in the vehicle shown in

Fig. 1A in more detail.

Fig. 3 shows an example embodiment according to the present invention.

Fig. 4 shows an example of an estimated pressure track for a combustion, and an actual pressure track up to a first crank angle position.

Figs. 5A-B show an example of regulation in situations with more than three injections.

Fig. 6 shows an example of MPC .

Detailed description of embodiments

Fig. 1A shows a diagram of a driveline in a vehicle 100 according to an embodiment of the present invention. The driveline comprises one combustion engine 101, which in a customary manner, via an output shaft on the combustion engine 101, usually via a flywheel 102, is connected to a gearbox 103 via a clutch 106.

The combustion engine 101 is controlled by the engine's control system via a control device 115. Likewise, the clutch 106, which may consist of e.g. an automatically controlled clutch, as well as the gearbox 103 are controlled by the vehicle's control system with the help of one or more

applicable control devices (not shown) . Naturally, the

vehicle's driveline may also be of another type, such as a type with a conventional automatic gearbox, or a type with a manual gearbox, etc.

An output shaft 107 from the gearbox 103 operates the driving wheels 113, 114 in a customary manner via the end gear and driving shafts 104, 105. Fig. 1A shows only one shaft with driving wheels 113, 114, but in a customary manner the vehicle may comprise more than one shaft equipped with driving wheels, or one or more extra shafts, such as one or more support shafts. The vehicle 100 also comprises an exhaust system with an after-treatment system 200 for customary treatment

(purification) of exhaust emissions resulting from combustion in the combustion chamber (e.g. cylinders) of the combustion engine 101.

Further, combustion engines in vehicles of the type shown in Fig. 1A are often equipped with controllable injectors in order to supply the desired amount of fuel at the desired point in time in the combustion cycle, such as at a specific piston position (crank angle degree) in the case of a piston engine, to the combustion engine's combustion chamber. Fig. 2 shows schematically an example of a fuel injection system for the combustion engine 101 exemplified in Fig. 1A. The fuel injection system consists of a so-called Common Rail system, but the invention is equally applicable in other types of injection systems. Fig. 2 shows only one

cylinder/combustion chamber 201 with a piston 203 active in the cylinder, but the combustion engine 101 consists, in the present example, of a six-cylinder combustion engine, and may generally consist of an engine with any number of cylinders/combustion chambers, such as any number of

cylinders/combustion chambers in the range 1-20 or even more. The combustion engine also comprises at least one respective injector 202 for each combustion chamber (cylinder) 201. Each injector is thus used for injection (supply) of fuel in a respective combustion chamber 201. Alternatively, two or more injectors per combustion chamber may be used. The injectors 202 are individually controlled by respective actuators (not shown) arranged at the respective injectors, which, based on received control signals, e.g. from the control device 115, control the opening/closing of the injectors 202.

The control signals for the control of the actuators'

opening/closing of the injectors 202 may be generated by some applicable control device, such as, in this example, by the engine control device 115. The engine control device 115 thus determines the amount of fuel which actually is to be injected at any given time, e.g. based on prevailing operating

conditions in the vehicle 100. The injection system shown in Fig. 2 thus consists of a so- called Common Rail system, which means that all injectors (and therefore all combustion chambers) are supplied with fuel from a common fuel conduit 204 (Common Rail) , which, with the use of a fuel pump 205, is filled with fuel from a fuel tank (not shown) at the same time as the fuel in the conduit 204, also with the help of the fuel pump 205, is pressurised to a certain pressure. The highly pressurised fuel in the common conduit 204 is then injected into the combustion engine's 101 combustion chamber 201 when the respective injector 202 is opened. Several openings/closings of a specific injector may be carried out during one and the same combustion cycle, whereby several injections may thus be carried out during the combustion of one combustion cycle. Further, each combustion chamber is equipped with a respective pressure sensor 206, for sending of signals regarding a prevailing pressure in the combustion chamber to e.g. the control device 115. The

pressure sensor may e.g. be piezo-based and should be fast enough to be able to send crank angle resolved pressure signals, such as at every 10th, every 5th or every crank angle degree or with another suitable interval, e.g. more

frequently . With the help of a system of the type shown in Fig. 2, the combustion during a combustion cycle in a combustion chamber may to a large extent be controlled, e.g. with the use of multiple injections, where the times and/or duration of the respective injections may be controlled, and where data from e.g. the pressure sensors 206 may be taken into consideration in connection with this control action.

According to one embodiment of the present invention, the combustion for a subsequent combustion cycle is controlled based on a previous combustion cycle, i.e. the calculation from a previous combustion cycle is used in the control of a subsequent combustion cycle. According to one embodiment of the invention, e.g. the injection times and/or duration of the respective injections and/or injected fuel amount during an ongoing combustion cycle are adapted, based on data from the ongoing combustion cycle.

As mentioned above, the noise which the operation of a

combustion engine generally produces depends to a large extent on the combustion in the combustion engine's combustion chamber, and in particular on the manner in which the pressure changes during the combustion. According to the invention, the combustion is controlled primarily with regard to the maximum pressure permitted to arise in the combustion chamber during the combustion. According to one embodiment, the maximum pressure derivative at combustion is also controlled, i.e. the maximum speed with which the pressure changes, and in

particular pressure increases. Fig. 3 shows an example method 300, according to the present invention, where the method according to the present example is arranged to be carried out by the engine control device 115 shown in Figs. 1A-B.

In general, control systems in modern vehicles consist of a communication bus system consisting of one or more

communications buses to connect a number of electronic control devices (ECUs) , such as the control device, or controller, 115, and various components arranged on the vehicle. According to prior art, such a control system may comprise a large number of control devices, and the responsibility for a specific function may be distributed among more than one control device.

For the sake of simplicity, Figs. 1A-B show only the control device 115, in which the present invention is implemented in the embodiment shown. The invention may, however, also be implemented in a control device dedicated to the present invention, or wholly or partly in one or several other control devices already existing in the vehicle. Considering the speed at which calculations according to the present invention are carried out, the invention may be arranged to be implemented in a control device which is especially adapted for real time calculations of the type described below. The implementation of the present invention has shown that e.g. ASIC and FPGA solutions are suitable for and cope well with calculations according to the present invention.

The function of the control device 115 (or the control

device (s) at which the present invention is implemented), according to the present invention, may, apart from depending on sensor signals from the pressure sensor 202, e.g. depend on signals from other control devices or sensors. Generally, control devices of the type shown are normally arranged to receive sensor signals from different parts of the vehicle, as well as from different control devices arranged on the

vehicle .

Control is often controlled by programmed instructions. These programmed instructions typically consist of a computer program, which, when it is executed in a computer or control device, causes the computer/control device to carry out the desired control action, as a method step in the process according to the present invention.

The computer program usually consists of a computer program product, where the computer program product comprises an applicable storage medium 121 (see Fig. IB), with the computer program stored on said storage medium 121. Said digital storage medium 121 may e.g. consist of any from the following group: ROM (Read-Only Memory), PROM (Programmable Read-Only Memory), EPROM (Erasable PROM), Flash, EEPROM (Electrically Erasable PROM), a hard disk unit, etc., and may be set up in or in combination with the control device, where the computer program is executed by the control device. By changing the computer program's instructions, the vehicle's behaviour may thus be adjusted in a specific situation.

An example control device (control device 115) is shown schematically in Fig. IB, and the control device in turn may comprise a calculation unit 120, which may consist of e.g. a suitable type of processor or microcomputer, e.g. a circuit for digital signal processing (Digital Signal Processor, DSP) , one or several FPGA (Field-Programmable Gate Array) circuits or one or several circuits with a predetermined specific function (Application Specific Integrated Circuit, ASIC) . The calculation unit 120 is connected to a memory unit 121, which provides the calculation unit 120 with e.g. the stored program code and/or the stored data which the calculation unit 120 needs in order to be able to carry out calculations. The calculation unit 120 is also set up to store interim or final results of calculations in the memory unit 121.

Further, the control device is equipped with devices 122, 123, 124, 125 for receipt and sending of input and output signals. These input and output signals may contain waveforms, pulses, or other attributes, which may be detected as information for processing by the calculation unit 120 by the devices 122, 125 for the receipt of input signals. The devices 123, 124 for sending output signals are arranged to convert the calculation result from the calculation unit 120 into output signals for transfer to other parts of the vehicle's control system and/or the component (s) for which the signals are intended. Each one of the connections to the devices for receipt and sending of input and output signals may consist of one or several of the following; a cable, a data bus, such as a CAN (Controller Area Network) bus, a MOST (Media Oriented Systems Transport) bus, or any other bus configuration; or of a wireless connection.

Reverting to the method 300 shown in Fig. 3, the method begins at step 301, where it is determined whether the control according to the invention of the combustion process should be carried out. The control according to the invention may e.g. be arranged to be carried out continuously as soon as the combustion engine 101 is started. Alternatively, the control action may be arranged to be carried out e.g. as long as the combustion engine's combustion is not to be controlled

according to some other criterion. For example, there may be situations where it is desirable that the control action is carried out based on factors other than emitted noise

primarily. According to one embodiment, simultaneous control of the combustion is carried out with respect to the noise emitted during combustion, and at least one additional control parameter. For example, a weighing up may be carried out, where the control parameters' prioritisation on fulfilment of a desired control result may e.g. be arranged to be controlled according to some suitable cost function.

The method, according to the present invention, thus consists of a method for the control of the combustion engine 101, while the combustion takes place in said combustion chamber 201 in combustion cycles. According to prior art, the term combustion cycle is defined as the steps comprised in a combustion in a combustion engine, such as a two-stroke engine's two strokes and a four-stroke engine's four strokes. The term also includes cycles where no fuel is actually injected, but where the combustion engine is still operated with a certain engine speed, such as with the vehicle's driving wheels via the driveline in e.g. dragging. That is to say, even if no injection of fuel is carried out, a combustion cycle is still completed for e.g. every two revolutions (for four-stroke engines), or e.g. every revolution (two-stroke engines), which the combustion engine's output shaft rotates. The same applies to other types of combustion engines. In step 302 it is determined whether a combustion cycle has been or will be started, and where this is the case, the method continues to step 303 while a parameter i representing an injection number is set equal to one.

In step 303 an injection schedule/control alternative which is expected to result in a desired pressure development during the combustion cycle is determined, such as an injection schedule which is expected to limit the maximum pressure amplitude in the combustion chamber during the combustion cycle's combustion.

Generally, the supply of the amount of fuel, both with respect to quantity and manner of supply, i.e. the one or several fuel injections that are to be carried out during the combustion cycle are normally defined in advance, e.g. depending on the work (torque) which the combustion engine must carry out during the combustion cycle, since no change of the determined injection schedule is carried out during an ongoing combustion cycle according to prior art. Predetermined injection

schedules may e.g. exist in tables in the vehicle's control system for a large number of operating modes, such as

different engine speeds, different requested work, different combustion air pressures, etc., where tabulated data may e.g. be prepared by way of applicable tests/measurements during e.g. the development of the combustion engine and/or vehicle, so that the applicable injection schedule/control alternative may be selected based on prevailing conditions, and where the injection schedule may be selected or be adapted in advance e.g. in order to result in an expected maximum pressure amplitude, which is below a certain applicable pressure threshold .

These injection schedules/control alternatives may consist of the number of injections and their respective characteristics in the form of e.g. timing (crank angle position) of the start of the injection, the duration of the injection, the injection pressure etc., and thus may be stored for a large number of operating modes in the vehicle's control system, and e.g. be calculated/measured with the objective of resulting in a maximum pressure amplitude below a certain applicable

pressure. The injections may also be produced with the objective of meeting other targets as well, such as delivering a requested work, resulting in a certain maximum heat loss, a certain exhaust temperature, etc., where the injections may thus be produced based on a weighting of several parameters. According to the present embodiment, in step 303 such a predetermined injection schedule is therefore applied, where this predetermined injection schedule is thus selected, e.g. through table lookup, based on prevailing circumstances and a desired work to be achieved by the combustion engine, where the desired (requested) work is normally controlled

(requested) by a certain superior/other process, e.g. based on a request regarding driving power from the vehicle's driver and/or a cruise control system.

According to one embodiment, an injection schedule is

determined, which results in at least half of the desired work being achieved in order to ensure that the work done may not be controlled at too low a level.

According to one embodiment, the injection schedule is

determined entirely according to e.g. the calculations shown below, where e.g. different injection schedules defined in advance may be compared with each other in order to determine a most preferred injection schedule, but in the calculation example exemplified below, the calculations are, however, applied only after the injection has started during the combustion cycle. Since specific assumed conditions probably result in the same preferred injection schedule each time, it may be advantageous to select an injection schedule through some type of lookup before a combustion cycle, and thus to reduce the calculation load, so that calculation as set out below is thus carried out only after the injection has

started. In addition to the example below of how the injection schedule may be determined, other models with a similar function may alternatively be applied.

According to the present embodiment, in step 303 a pre-defined injection schedule at the start of the combustion cycle is thus determined, where control action according to the

invention is carried out only after the fuel injection has been started during a combustion cycle, e.g. only after at least one injection has been completed during the combustion cycle, or after one injection has at least been started. Fuel injection is thus normally carried out according to a predetermined schedule, where several injections may be arranged to be carried out during one and the same combustion cycle. This entails that the injections may be relatively short. For example, there are injection systems with 5-10 fuel injections/combustion, but the number of fuel injections may also be significantly greater, such as in the range of 100 fuel injections during one combustion cycle. The number of possible injections is controlled generally by the speed of the elements with which injection is carried out, i.e., in the case of a Common Rail system how fast the injectors may be opened and closed.

According to the present example, at least three fuel

injections inspi are carried out during one and the same combustion cycle, but as mentioned and as set out below, a greater number of injections may be arranged to be carried out, as well as only one.

The injection schedule is thus in the present example

determined in advance, with the objective of obtaining a pressure development which meets applicable criteria with respect to the maximum pressure amplitude arising during the combustion. A first injection inspi is carried out, and in step 304 it is determined whether said first injection inspi has been carried out, and if so, the method continues to step 305, where it is determined whether all the injections i have been carried out. Since this is not yet the case in the present example, the method continues to step 306 while i is

incremented by one for the next injection. Further, with the continuous use of the pressure sensor 206 y, such as with applicable intervals, e.g. every 0.1-10 crank angle degrees, the prevailing pressure in the combustion chamber is

determined.

The combustion process may generally be described with the pressure change in the combustion chamber which the combustion gives rise to. The pressure change during a combustion cycle may be represented by a pressure track, i.e. a representation of how the pressure in the combustion chamber varies/changes during the combustion. As long as the combustion progresses as expected, the pressure in the combustion chamber will be equal to that initially estimated, but as soon as the pressure deviates from the estimated pressure, the manner in which the pressure has changed, and thus very probably also the maximum pressure amplitude which will arise during the combustion, will also deviate from the estimated values. In addition, the subsequent part of the combustion cycle, and therefore the pressure change, will be impacted because changed

circumstances compared to the expected circumstances prevail in the combustion chamber at e.g. a subsequent injection.

If the combustion after the first injection inspi has thus proceeded just as expected, the conditions in the combustion chamber will correspond to the conditions intended for the injection, and likewise the hitherto resulting pressure change (the pressure track as set out below) in the combustion chamber will correspond to the expected pressure change up to this point. As soon as the conditions differ from the intended conditions, however, the pressure change during the combustion will deviate from the expected pressure change. Likewise, the subsequent part of the combustion will also be impacted since the conditions prevailing in the combustion chamber, e.g. with respect to pressure/temperature, at the next injection will not correspond to the expected conditions.

In practice, the actual pressure changes during the combustion (the pressure track) will also, for natural reasons, very probably deviate from the predicted pressure track during the course of the combustion, because of e.g. deviations from the modelled combustion. This is illustrated in Fig. 4, where a predicted pressure track 401 for an example injection schedule is shown (very schematically), i.e. the expected pressure track for the combustion chamber when the injection is carried out according to the selected injection profile. This

prediction of the pressure track may e.g. be carried out as described below.

Fig. 4 also shows an actual pressure track 402 up to the crank angle position φ ₁ , which constitutes the prevailing position after said first combustion has been carried out. In step 306, the pressure ρ _φ ι in the combustion chamber is determined with the use of the pressure sensor 206 after the first injection inspi has been carried out, at the crank angle position φ ₁ . Preferably, the pressure in the combustion chamber is

determined substantially continuously, e.g. at each crank angle degree, every tenth crank angle degree or with another suitable interval, during the entire combustion. As may be seen in Fig. 4, the actual pressure track up to cpi deviates from the estimated pressure track 401, and the actual pressure Pcpi also deviates at cpi from the estimated pressure ρ _φ ι _est according to the pressure track 401. The above means that the maximum pressure resulting to date also has deviated from the expected maximum pressure up to the crank angle position cpi .

Since the pressure ρ _φ ι in the combustion chamber, after the first injection inspi has been carried out, differs from the corresponding estimated pressure ρ _φ ι _est at the crank angle position cpi, the conditions in the combustion chamber at the time of the next consecutive injection insp2 will differ from the predicted conditions, so that the subsequent combustion will also differ from the predicted combustion, if the

previously determined injection schedule were still used.

Thus, it is not at all certain that the desired limitation of the maximum pressure amplitude will be achieved during the combustion cycle. Therefore, it also not certain that it is the originally determined injection schedule which constitutes the most preferred injection schedule in an effort to achieve a combustion with a desired limitation of pressure amplitude.

In step 307 it is determined, whether the expected maximum pressure amplitude pmax_pred is expected to exceed some

applicable threshold value p_thres , where this may be determined in advance, and also to be arranged to vary depending on other circumstances such as current load, vehicle speed, etc. As long as this is not the case, the method reverts to step 304 for the performance of the next injection, and a new

estimation of p is then carried out. If, on the other hand, pmax_pred is expected to exceed p_thres , the method continues to step 308 in order to again determine an injection schedule, with the objective of controlling the pressure amplitude, e.g. with the objective of attempting to limit the pressure

amplitude to not exceeding p_thres . The control may e.g. be carried out according to the calculations shown below,

alternatively according to other applicable calculations with a similar objective, and be repeated as set out below during an ongoing combustion cycle, in order to, where needed, change the injection schedule during an ongoing combustion if the actually prevailing conditions in the combustion chamber differ from the predicted conditions, such as after each injection, or during an ongoing injection.

At the estimation of the expected maximum pressure amplitude, according to the invention, e.g. a model may be applied which describes the pressure change arising during the combustion. This model may be of different types, and consist of e.g. a computer driven model such as — = f (p ₀ id> ^u injection strategy > Ύ> ~^) > where p _oid constitutes the pressure at the previous

determination, i _n j _ect i _{on strategy} constitutes the control signal, i.e. the injection schedule,

Y constitutes generally the heat capacity ratio, i.e.

C C

Y =—— = — , where C„ and/or C, are generally available and

C _v C _p -R ^p

tabled for various molecules, and since the combustion

chemistry is previously known, these tabled values may be used jointly with the combustion chemistry in order thus to

calculate each molecule's (e.g. water, nitrogen, oxygen, etc.) impact on e.g. the total C _p value, so that this may be determined for the calculations above with a good accuracy, in advance or during e.g. ongoing combustion. Alternatively, C _p and/or C _v may be approximated in a suitable manner. dV

— constitutes the combustion chamber's volume change over time, which e.g. may be determined with the help of V(cp).

V(<p) constitutes the combustion chamber's volume as a function of the crank angle, may advantageously be available tabled in the control system's memory, or be calculated in an applicable dv dv

manner, where — may also be calculated, and thus — as well,

dq> dt

by multiplication with the combustion engine's prevailing engine speed. dp

Thus, the rate of change of the pressure — may be represented by such a model, which may be prepared by determining the

dp

result for a large number of input parameters, where — may then be tabled for a large number of circumstances, such as different loads, engine speeds, air pressure, etc., as is known to a person skilled in the art. By then accumulating

(integrating) ^ over time, the pressure p in the combustion chamber may be estimated, and by also determining, for each

dp

determination of —, the pressure p and by comparing the obtained pressure p with the previously, during the estimation, obtained maximum pressure p , where the higher of these values is stored as the new maximum pressure, the maximum pressure p during the combustion may be estimated, whereby control may be carried out if it is determined during the control that the pressure is expected to exceed a threshold value. Another alternative, which constitutes the alternative applied in the present example, is the use of a physical model over the pressure's p change during combustion in the combustion chamber. This model may consist of an applicable model, and according to the present example, a heat release equation is applied as set out below.

Estimation of the pressure p variation during the combustion may then be carried out as set out below. The pressure

prevailing in the combustion chamber p may be determined with the use of said pressure sensor, where continuous sensor signals may provide measured values for the pressure p at intervals/crank angle degrees with an applicable frequency.

Further, e.g. when the pressure change rate is also

dp

considered, — may be estimated for the part of the combustion which has already lapsed, and where an actual maximum pressure change rate may be estimated for the part of the combustion which has already lapsed, based on actual pressure data.

The pressure change may be determined as a function of time, such as as set out above, but may also be expressed in

dp

crank angle degrees φ, i.e. — , which entails an elimination of the combustion engine speed dependency in calculations.

In the cases where the pressure change rate is also considered during the control, the desired maximum pressure change rate ^ may e.g. be stored for various engine speeds n in order to thus e.g. represent a desired pressure change over time.

dp

Alternatively, — may e.g. be determined as set out below and άφ

then multiplied by the combustion engine's engine speed n in dp

order to obtain — .

dt

The present invention strives to, as needed, actively reduce the maximum pressure amplitude in the combustion engine, which may be carried out by estimating the expected maximum pressure amplitude for a subsequent part of the combustion cycle, where e.g. a maximum expected pressure amplitude may be determined, where the combustion may be controlled with the objective to keep the maximum pressure amplitude below a certain applicable pressure amplitude.

This also means that the pressure amplitude may be estimated for several different scenarios for combustion, such as different injection schedules, where the respective injection schedules will give rise to a specific pressure track, e.g. the pressure track shown in Fig. 4, and thus also different maximum pressure amplitudes during the combustion.

At estimation of the pressure track, a model of the combustion may be used, and, as is familiar to one skilled in the art, the combustion may be modelled according to equation (1) : dQ— ^-calibrate {Q fuel (1)

, where K _calibrate is used to calibrate the model. K _calibrate

consists of a constant which is usually in the range of 0-1, but may also be arranged to assume other values, and which is determined individually, cylinder by cylinder, or for a certain engine or engine type, and depends in particular on the design of the injector nozzles (spreaders) .

Q _fuel consists of the energy value for injected fuel amount, Q consists of the amount of energy burned. The combustion dQ is thus proportionate to the injected fuel amount, minus the hitherto consumed fuel amount. The combustion dQ may,

alternatively, be modelled with the use of another applicable model, where e.g. regard may be had also to other parameters. For example, the combustion may also constitute a function which depends on a model of turbulence arising when air/fuel is supplied, which may impact the combustion to different extents, depending on the amount of air/fuel supplied.

Regarding the fuel injections, these may e.g. be modelled as a sum of step functions:

U = The fuel flow measured as the supplied mass m at an injection k, i.e. how the fuel enters into the combustion chamber during the time window u when the injection is carried out, expressed as the time lapsed during the crank angle degree φ interval during which the injector is open, may be modelled for a specific injection k as: where m constitutes the injected fuel amount, and f (m) e.g. depends on the injection pressure, etc. f (m) may e.g. be measured or estimated in advance.

The energy value Q _LH v f° ^r the fuel, such as diesel or petrol, is generally specified, so that such a general specification may be used. The energy value may also be specifically

provided by e.g. the fuel manufacturer, or be approximated for e.g. a country or a region. The energy value may also be arranged to be estimated by the vehicle's control system. With the energy value, the equation (1) may be resolved and the heat release Q may be determined as the combustion progresses.

Further, with the use of a predictive heat release equation, the pressure change in the combustion chamber during the entire combustion may be estimated as:

, where γ constitutes the heat capacity ratio according to the above .

The pressure p in the combustion chamber may be obtained through integration of the equation (4) as follows: f C fdQ γ dV\ (Y - 1\

V = Pinitiat + j dp = _Pinitial +J(—-—p—) d<p (5) where Pinitial constitutes an initial pressure, which before the start of the combustion's compression step may e.g. consist of the ambient pressure for combustion engines without a turbo, or a prevailing combustion air pressure for an engine with a turbo. When the estimation is carried out at a later point in time during the combustion cycle, such as estimation in step 307 after an injection has been carried out, Pinitial ^ma Y

constitute the then prevailing pressure, as determined by the pressure sensor 206, i.e. ρ _φ ι in the present example. Thus both the pressure p (and also the pressure derivative) in the combustion chamber may be estimated for the entire combustion, i.e. an expected curve corresponding to the curve 401 in Fig. 4 may be estimated.

Thus, with the use of equation (4) p , either as a function of the crank angle or time, by multiplying with the engine speed as above, may be estimated for the entire remainder of the combustion cycle, or also for an entire combustion cycle if the estimation is carried out before the fuel injection is started, where p at each iteration of the equations 4-5 there may be a comparison to p_thres in order to determine whether the pressure p is expected to exceed p thres during the combustion. Thus, the actual maximum pressure which is expected to arise need not be estimated, but the estimation may, according to one embodiment, be interrupted as soon as it has been

concluded that p is expected to exceed pjthres during the combustion .

Alternatively, the maximum pressure which is expected to be achieved during the estimation may be obtained by carrying out the integration as long as p (k+l ) > p (k) , where k, k+1, etc. consist of consecutive points in time/crank angle positions. As long as the pressure rises, the integration is thus continued, while the integration may be interrupted when p

(k+l) < p (k) , since the pressure has then started to drop. The maximum pressure may then be compared with the threshold value p_thres .

If this is the case, the method continues, as set out above, to step 308 in order to determine a new injection strategy, since the control of the pressure in the combustion chamber may e.g. be carried out by controlling the fuel injection, and by carrying out an estimation of the pressure for a number of different injection schedules in step 308 with e.g. varying injection times and/or injection durations and/or number of injections and/or periods of time between injections, the estimated maximum pressure amplitudes for different injection alternatives may be compared, and thus an injection schedule may be determined which, if possible, entails that dp_thres is not achieved during the combustion, preferably with the additional condition that the desired work achieved on the combustion engine's output shaft is still obtained.

Thus, a work requested at combustion may also be determined, which work e.g. may be determined by some superior process which e.g. is responsible for the vehicle's propulsion, where the control may have as a condition that the resulting work at the combustion substantially corresponds to said requested work, or at least a partial amount thereof, such as at least half of the requested work.

Thus, in step 308, an injection schedule may be determined, such as an injection schedule among several defined injection schedules, where this injection schedule may be determined individually, cylinder by cylinder, based on sensor signals from at least one pressure sensor in the respective combustion chamber .

In relation to said injection schedule, there may be e.g.

several injection schedules defined in advance, where

calculations of the type described above may be carried out for each one of these available injection schedules.

Alternatively, the calculations may be carried out for the injection schedules which, for some reason, most probably are deemed to result in a desired low pressure amplitude. Hitherto the entire injection schedule for the remainder of the combustion has been evaluated, but the evaluation may also be arranged to be carried out for only the future injection after a previous injection, where subsequent injections may be handled gradually. The injection schedule selected in step 308 may thus consist of only the next injection.

When the injection schedule has been selected in step 308, the method reverts to step 304 for the completion of the next injection, so that this also gives rise to a combustion, and thus a heat release and a pressure track, and this will also probably deviate from the pressure track predicted in advance. This also means that the combustion, also at subsequent injections, will probably be impacted by prevailing conditions in the combustion chamber when the injection is started.

Thus in step 308, after a new subsequent injection has been completed, another new injection strategy for the remaining injections, alternatively for the subsequent injection, may be calculated with the help of the above equations, where the method then reverts to step 304 for the completion of the subsequent fuel injection, according to the new injection strategy which is calculated in step 308, while still taking the work to be achieved during combustion into consideration, which thus normally is controlled by some superior process, e.g. in response to a request for a certain driving force from the vehicle's driver or another function in the vehicle's control system, such as a cruise control function. The control may thus be arranged to be carried out after each injection i, and when all subsequent injections i have then been carried out, the method reverts from step 305 to step 301 to control a subsequent combustion cycle. According to one embodiment, the method is interrupted, however, as soon as the maximum

pressure of the combustion has been reached, which may be determined as set out below. The noise emitted during the combustion depends primarily on the pressure build-up and to a lesser degree on the subsequent pressure drop. For this reason, control may thus be interrupted when the combustion's maximum pressure has been reached.

In the above calculations, after each injection, the current pressure determination ρ _φ ι is used by using the pressure sensor 206 in the way Vinitiai described above, in order to again predict the maximum pressure amplitude in order to determine, where needed, a new injection schedule based on the now prevailing conditions in the combustion chamber, but now with data obtained further along in the combustion. That is to say, ρ _φ ι after the first combustion and similarly determined ρ _φ ι for subsequent injections, where thus Vinitiai changes at

calculations during the combustion cycle, and where the fuel injection is are adapted according to prevailing conditions after each injection, with the consequence that the injection schedule may change after each injection.

The present invention thus provides a method which adapts the combustion as the combustion progresses, and comprises

generally to control, based on a first parameter value, which is determined after a first part of the combustion has been completed, a subsequent part of the combustion during one and the same combustion cycle, where the combustion is controlled with respect to the maximum pressure during the combustion process . According to the above, the maximum pressure amplitude may thus be estimated for several different alternative injection schedules for the remaining injections, where an injection schedule which results in the most advantageous, e.g. the lowest, pressure amplitude may be selected when the subsequent injection is carried out. In cases where several injection schedules/control alternatives fulfil the applicable

conditions, other parameters may be used to select which of these are to be used. There may also be other reasons for simultaneously effecting control also on the basis of other parameters. For example, injection schedules may also be partly selected based on one or several of the perspectives pressure change rate, heat loss, exhaust temperature, work achieved in the combustion chamber, or nitrogen oxides

generated during the combustion as a further criterion, in addition to being selected based on pressure amplitude, where such determination may be carried out according to any of the parallel patent applications specified below. Specifically, the parallel application "METHOD AND SYSTEM FOR CONTROL OF A COMBUSTION ENGINE I" (Swedish patent application, application number: 1350506-0) shows a method to, based on an estimated maximum pressure change rate, control subsequent combustion.

Additionally, the parallel application "METHOD AND SYSTEM FOR CONTROL OF A COMBUSTION ENGINE II" (Swedish patent

application, application number: 1350507-8) shows a method to, during a first combustion cycle, control a subsequent part of combustion during said first combustion cycle, with respect to a temperature resulting in said subsequent combustion. Further, the parallel application "METHOD AND SYSTEM FOR

CONTROL OF A COMBUSTION ENGINE III" shows a method to, during a first combustion cycle, control combustion during a

subsequent part of said first combustion cycle with respect to a work achieved during the combustion.

Further, the parallel application "METHOD AND SYSTEM FOR

CONTROL OF A COMBUSTION ENGINE IV" shows a method to, during a first combustion cycle, control combustion during a subsequent part of said first combustion cycle, with respect to a

representation of a heat loss resulting during said

combustion .

Further, the parallel application "METHOD AND SYSTEM FOR

CONTROL OF A COMBUSTION ENGINE VI" shows a method to, during a first combustion cycle, estimate a first measure of nitrogen oxides resulting from combustion during said first combustion cycle, and to control the combustion during a subsequent part of said first combustion cycle based on said first measure.

According to the present invention, the combustion is thus adapted during ongoing combustion as needed, based on

deviations from the predicted combustion, and according to one embodiment an evaluation of the combustion is carried out each time an injection inspi has been completed, as long as further injections will be carried out.

According to the above described method, the injection

schedule at the start of the combustion cycle has been

determined based on tabulated values, but, according to one embodiment, the injection strategy may already before the fuel injection starts be determined in the manner described above, so that also the first injection is thus carried out according to an injection schedule determined as set out above. The control has hitherto been described in a manner where the characteristics for a subsequent injection are determined based on prevailing conditions in the combustion chamber after the previous injection. The control may, however, also be arranged to be carried out continuously, where pressure determinations may be carried out with the help of the

pressure sensor also during ongoing injection, and where the injection schedule may be calculated and corrected all the way until the next injection is initiated. Alternatively, even the ongoing injection may be impacted by calculated changes in the injection schedule, also in the cases where several shorter injections are carried out. For example, an ongoing injection may be interrupted if the pressure amplitude is too high. The injection may also consist of one single longer injection, where changes to the ongoing injection may be made

continuously, e.g. by way of so-called rate shaping, e.g. by changing the opening area of the injection nozzle and/or the pressure with which the fuel is injected, based on estimations and measured pressure values during the injection. Further, fuel supply during the combustion may comprise two fuel injections, where e.g. only the second or both injections are controlled, e.g. with the help of rate shaping. Rate shaping may also be applied in the event three or more injections are carried out. In relation to the injection strategies which should be evaluated, these may be devised in different ways. For

example, different distributions between injections may be evaluated, and e.g. an injected fuel amount may be

redistributed between subsequent injections, and/or the injection time may be changed for one or several subsequent injections, where potential limitations with respect to e.g. the minimum permitted duration or fuel amount for a fuel injection may be taken into consideration.

Instead of evaluating a number of specific injection

schedules, the method may be arranged to carry out e.g. the above calculations for a number of possible scenarios, where the calculations may be carried out for different injection durations/amounts/times for the different injections, with corresponding changes in released energy.

The more fuel injections carried out during a combustion cycle, the more parameters may be changed, while at the same time achieved work should be maintained. In the event of a large number of injections the control may therefore become relatively complex, since a large number of parameters may be varied and would thus need to be evaluated. For example, a very great number of injections may be arranged to be carried out during one and the same combustion cycle, such as ten, or even hundred or so injections.

In such situations, and also others as set out above, there may be several substantially equivalent injection strategies, which result in substantially the same maximum pressure amplitude, or which fulfil the requests/requirements regarding pressure amplitude. This introduces an unwanted complexity in the calculations.

According to one embodiment, a control action is applied where the injection nearest in time is considered to be a separate injection, and subsequent fuel injections are considered to be one single additional "virtual" injection, so that fuel may be distributed between these "two" injections in a manner

entailing that the maximum pressure during the first

combustion is not expected to exceed the desired levels. This is exemplified in Fig. 5A, where the injection 501 corresponds to inspi, as set out above, the injection 502 corresponds to insp ₂ , as set out above, and where the remaining injections 503-505 are treated as one single virtual injection 506, i.e. the injection 506 is treated as one injection with a fuel amount substantially corresponding to the total fuel amount for the injections 503-505, and where a distribution may be made between the injection 502 and the virtual injection 506. By proceeding in this manner, the shifting which occurs between insp ₂ and subsequent injections does not need to be distributed specifically between the injections 503-505, but the distribution at this stage is made between the injection 502 and the "virtual" injection 506, respectively.

When the injection 502 has then been completed the method is repeated as needed, exactly as above, with a new determination of an injection schedule in order to attempt, where needed, to reduce the pressure amplitude, but this time with the

injection 503 as a separate injection, see Fig. 5B, and the injections 504, 505 jointly constituting one virtual injection as the distribution is made, as set out above. In Fig. 5A the virtual injection 506 is constituted by three injections, but as is obvious, the virtual injection 506 may comprise, from the beginning, more than three injections, such as tens of injections or hundreds of injections, depending on how many injections that are planned to be carried out during the combustion cycle, so that the method is repeated until all the injections have been completed. According to one

embodiment, the method is interrupted, however, when the maximum pressure has been reached and the pressure in the combustion chamber begins to drop again, since the maximum pressure amplitude during the combustion may no longer be impacted . It is also possible to use e.g. MPC (Model Predictive Control) in the control according to the invention.

One example of MPC is shown in Fig. 6, where the reference curve 603 corresponds to the expected pressure development during the heat release during the combustion cycle, i.e. the result of equation (5) above for the selected injection schedule. The curve 603 may e.g. consist of a (lowest) level which may realistically be achieved during a combustion cycle for the maximum pressure at the given load and prevailing engine speed, and may e.g. be determined in advance, e.g. with applicable calculations and/or measurements on the engine type, so that these data may be stored in the control system's memory as functions of e.g. engine speed and load. This entails also that the combustion need not be controlled only toward a pressure prevailing at each time, but may also be arranged to be controlled toward an expected maximum pressure, such as e.g. the curve 603 in Fig. 6, where each injection may have as its objective to result in a combustion which

corresponds to the curve 603. The solid curve 602 up to the time k represents the actual development of the pressure which has arisen to date, and which has been calculated as set out above with the help of actual data from the crank angle resolved pressure

transmitter. The curve 601 represents the estimated, i.e. the expected, development for the pressure in the combustion chamber based on the predicted injection profile. Dashed injections 605, 606, 607 represent the predicted control signal, i.e. the injection profile which is expected to be applied, and 608, 609 represent already completed injections. The predicted injection profile is updated with applicable intervals, e.g. after each completed injection, in order to reach the final value sought and which is given by the

reference curve 603, and where the next injection is

determined based on prevailing conditions in relation to the estimated pressure development. Thus, the present invention provides a method which allows for a very good control of a combustion process, and which adapts the combustion during ongoing combustion, in order to achieve a combustion with controlled pressure change and associated noise emitted. According to the above, the combustion may be controlled during an ongoing combustion cycle. According to one

embodiment, the estimation is carried out, however, for a combustion cycle where a subsequent combustion cycle may then be controlled based on the estimation for the previous combustion cycle.

The invention has been exemplified above in a manner where a pressure sensor 206 is used to determine a pressure in the combustion chamber. As an alternative to using pressure sensors, instead one (or several) other sensors may be used, such as high-resolution ion current sensors, knock sensors or strain gauges, where the pressure in the combustion engine may be modelled with the use of sensor signals from such sensors. It is also possible to combine different types of sensors, e.g. in order to obtain a more reliable estimation of the pressure in the combustion chamber, and/or to use other applicable sensors, where the sensor signals are converted into corresponding pressures for use in control, as set out above .

Further, in the above description only the fuel injection has been adjusted. Instead of only controlling the amount of fuel supplied, the pressure during combustion may be arranged to be controlled with the help of e.g. exhaust valves, so that injection may be carried out according to a predetermined schedule, but where the exhaust valves are used to control the pressure in the combustion chamber. Further, control may be carried out with some applicable type of regulator, or e.g. with the help of state models and state feedback (e.g. linear programming, the LQG method or similar) .

The method according to the invention for the control of the combustion engine may also be combined with sensor signals from other sensor systems where the resolution of the crank angle level is not available, e.g. another pressure

transmitter, NOx sensors, NH3 sensors, PM sensors, oxygen sensors and/or temperature transmitters, etc., the input signals of which may e.g. be used as input parameters in the estimation of e.g. the maximum pressure, with the use of computer-driven models instead of models of the type described above .

Additionally, the present invention has been exemplified above in relation to vehicles. The invention is, however, also applicable in any vessels/processes where combustion control as per the above are applicable, e.g. watercrafts and

aircrafts with combustion processes as per the above.

It should also be noted that the system may be modified according to various embodiments of the method according to the invention (and vice versa) and that the present invention is in no way limited to the above described embodiments of the method according to the invention, but pertains to and

comprises all embodiments in the scope of the enclosed

independent claims.