JP2008069772A - Optimization method of fuel injection nozzle for internal combustion engine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an optimization method of a fuel injection nozzle for an internal combustion engine having a nozzle body 1 and a needle valve 3 axially movable in a hole 2 of the nozzle body against a closing force, wherein the needle valve 3 is provided with a valve disc sealing surface 4 on an end face thereof nearer to a combustion engine and the valve disc sealing surface 4 moves in cooperation with a valve seat surface 6 of the nozzle body 1 so as to control a transverse area of a flow across at least one injection opening 5 opened to the combustion chamber of the internal combustion engine. <P>SOLUTION: In the optimization method, any number of independent parameters is used for geometrically forming a fuel injection nozzle. A mathematical technique of the particle swarm optimization (PSO) is also used. An optimum reference obtained based on the above is reflected in a structural configuration of the fuel injection nozzle. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The invention relates to a method for optimizing a fuel injection nozzle of an internal combustion engine according to the preamble of claim 1 of the claims.

  As is well known, the fuel injection nozzle is composed of a nozzle body and a nozzle needle valve that can move in the axial direction against the closing force in the hole of the nozzle body. The nozzle needle valve is sealed at its end face on the combustion chamber. The valve body sealing surface cooperates with the valve seat surface of the nozzle body to control the cross-sectional area of the flow through at least one injection opening into the combustion chamber of the internal combustion engine. The fuel injection nozzle is incorporated in a nozzle holder, and a fuel injection pipe is connected to the upper end of the nozzle holder as is well known. Fuel acts on the inclined shoulder of the needle valve in the pressure chamber inside the nozzle body and opens the nozzle needle valve against the closing force generally generated by hydraulic or spring force.

  A normal fuel injection nozzle is, for example, a so-called hole type nozzle as known from Patent Document 1, and this hole type nozzle is preferably incorporated in the engine so as to be directly injected. In a hole type nozzle in which a hole is formed in the form of a blind hole in the nozzle body, usually, 12 or more holes are provided in the nozzle dome. The final formation of the hole nozzle is generally determined by engine experiments. In that case, various nozzle changes relating to the number of holes, the hole diameter, the hole length, the angle between the holes, and the angle of the hole with respect to the cylinder axis, for example, must be examined for effects on output, fuel consumption rate and generation of harmful substances.

  The formation of the fuel injection nozzle has a decisive influence on engine combustion. In particular, the number, diameter, position and geometry of the nozzle holes and the high pressure in the blind holes are important for the formation of the jet flow in the hole nozzle. In particular, as blind hole volume increases, hydrocarbon (HC) and particle (eg, soot) generation and fuel consumption increase. It is also known that an excessive blind hole volume causes severe contamination of engine lubricant and severe wear on the cylinder guide surface.

  Furthermore, the fuel injection nozzle is heavily loaded mechanically, thermally and hydraulically. Thus, for example, due to functional requirements in the nozzle dome and needle valve seat, structural standards for component strength must be observed.

  Mathematical methods for the design of fuel injectors are already known, in which case the geometry has a system in the variable space, for example in the form of an objective function (x) consisting of the various parameters mentioned above, In the variable space, the objective function value is obtained repeatedly and repeatedly by a plurality of calculation processes. In this case, in each calculation process, the expansion of the probability quantity based on the stochastic differential equation is calculated for the objective function (x). The best objective function value and its parameters obtained with this mathematical method are used for the design of the fuel injection nozzle.

Such sequential approaches are often unfit for purpose because of their very complex interrelationships.
German Patent Application Publication No. 198441192

  An object of the present invention is to provide a method for optimizing a fuel injection nozzle of an internal combustion engine, in which the shape of the fuel injection nozzle of the internal combustion engine is optimized at least with respect to a reduction in the generation of harmful substances, an improvement in wear behavior and a reduction in fuel consumption. There is.

  This problem is solved by the fuel injection nozzle optimization method for an internal combustion engine according to claim 1.

  The present invention presents a completely new fuel injection nozzle optimization method that allows any type of fuel injection nozzle to be optimized with respect to individual requirements and ambient conditions.

  Special perforated nozzles with a large number of injection openings, single row perforated (FIG. 1) or double row perforated (FIG. 2) are considered in each embodiment.

  The algorithm executed is a multi-criteria version of the particle swarm optimization algorithm (Particle Swarm Optimization) developed by the applicant. The particle swarm algorithm is a probability optimization method similar to the natural world developed from the field of artificial intelligence. This algorithm is based on a population of particles (a possible parameter set) that interact with each other during movement in the exploration space, like birds in swarms (see: 1995). IEEE International Conference on Neutral Network, J. Kennedy and R. Everhart, “Particle Swarm Optimization Process” published on pages 1942–1948).

  The purpose of particle swarm optimization (PSO) is generally to find the optimal point of the search function, that is, the objective function (x), which has a maximum or minimum value depending on the definition. Can be presented. However, instead of a certain local optimum (generally as in the case of numerical methods of the type mentioned at the beginning), a global optimum of the entire search space, ie the analysis space, is found.

  The parameters (particles) of the PSO are randomly distributed at the beginning of the calculation and / or determined and distributed over the entire analysis space, so that this parameter has a corresponding position in the variable space. Random and / or fixed velocity vectors are also assigned to the particles for initialization.

  For the next step of the optimization algorithm, each particle is specifically positioned at the state of the neighboring particle and its best position so far. From the individual best solutions for each individual particle, the group best solution is selected by a comparison operation. As such, the group follows the direction of the best positioned particles as a whole.

  The positioning of the particles in the analysis space is n-dimensional depending on the number of unknowns in the reference number / independent parameters / objective function (x).

  The objective function is formulated such that a plurality of objective quantities having unique weights are integrated therein. This is called a quality function. The target amount corresponds to a concrete requirement that can be grasped mathematically, for example, the blind hole volume to be minimized in the fuel injection nozzle, and the blind hole volume to be minimized has a functional relationship with the geometric parameter. .

  This optimization algorithm has been found to be very quick, fit for purpose and therefore very effective.

  Another advantage is that since particle swarm optimization does not use derivatives, it can be used for almost any function that is itself discontinuous. That is, it is very stable.

  In summary, PSO is a robust and fast optimization method that serves to find global optimal points in almost any multidimensional mathematical function. By formulating a quality function having a weighted target quantity, a multiple criterion formula is determined.

  The algorithm requires a series of settings that affect its behavior. These set amounts are determined based on conventional experience values.

  Advantageous embodiments of the invention can be understood from the description of the dependent claims and the following description. Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings, but the present invention is not limited thereto.

  The present invention comprises a nozzle body (1) and a nozzle needle valve (3) which can move in the axial direction against the closing force in the hole (2) of the nozzle body (1). The valve body sealing surface (4) has a valve body sealing surface (4) at its combustion chamber side end face, and controls the cross-sectional area through which the at least one injection opening (5) into the combustion chamber of the internal combustion engine controls. The invention relates to a method for optimizing a fuel injection nozzle of an internal combustion engine, which cooperates with the valve seat surface (6) of the nozzle body (1). Using the selection, the particle swarm optimization (PSO) mathematical method is used to reflect the best-sought criteria in the structural shape of the fuel injection nozzle.

In a particularly advantageous manner, the optimization method described above with PSO comprises the following steps. That is,
Determine predetermined ambient conditions (nozzle type and nozzle size) for the geometry, eg the required nozzle flow-through and nozzle pressure stage;
-Technical requirements in the form of objective quantities of objective function (x) for any number of independent parameters such as maximum pressure, blind hole volume or minimum nozzle needle seat angle in the nozzle body hole (blind hole) ( Know how much to optimize.
-Deriving independent parameters (degrees of freedom) of technical requirements from definitions of target quantities such as nozzle needle valve seat angle, nozzle needle valve seat lower diameter, nozzle tip angle, injection angle or nozzle needle valve stroke.
-Determine the parameter limits in the form of a predetermined numerical range.
-Mathematically weight the objective quantity of the technical requirements in the quality function.

  The quality function is obtained from a mathematical formulation of requirements arising from geometry, fluid pressure, mechanisms, etc.

  FIG. 1 shows a first embodiment of the method according to the invention. In this case, the injection nozzle to be optimized is a hole type nozzle, and the needle valve (3) guided so as to be movable in the axial direction against the closing force in the hole (2) of the nozzle body (1) is blindly connected. A nozzle dome (7) for closing the hole (2) with respect to the combustion chamber, and the needle valve (3) has a valve body sealing surface (4) on its combustion chamber side end surface. (4) cooperates with the conical valve seat surface (6) of the blind hole (2), and a plurality of injection openings (5) protrude from the nozzle dome (7) toward the combustion chamber of the internal combustion engine.

The following points will be clarified in FIG. 1 and FIG. That is,
• As the predetermined peripheral conditions, peripheral conditions that can be freely set expressing the other restrictions related to the nozzle type, nozzle size and application are given.
Technical requirements include: a) maximum pressure in the blind hole (resulting in the best mixture formation), b) minimum blind hole volume (resulting in soot and HC generation and reduction in fuel consumption), c) nozzle hole (5) Minimum wall thickness between (due to required strength), d) Minimum nozzle needle seat angle (due to wear of nozzle needle seat), e) Blind hole (2) injection opening (5) as possible The size of at least one, multiple or all of the deep positions (due to hydrodynamic requirements) is known.
As independent parameters, a) valve seat angle (σ) of nozzle needle valve (3), b) nozzle needle tip angle (α), c) needle valve stroke (h), d) nozzle hole diameter (DL), e ) Blind hole diameter (DE), f) Nozzle needle valve seat lower diameter (DA), g) Blind hole angle (ε), h) Injection angle (κ), i) Injection point injection point (hSP), k) The blind hole depth (hS), l) number of nozzle holes, m) nozzle hole diameter (DL), at least one, a plurality or all of the sizes are given.
• The determined parameter limit value includes a numerical range or a discrete value.
• All technical requirements (see above) are standardized and incorporated into the quality function, and a) the distance between the needle tip of the nozzle needle and the bottom of the blind hole (2) must not be less than the minimum, b) The width of the nozzle needle valve seat must not be less than the minimum value, c) The wall thickness between the nozzle holes (5) must not be less than the minimum value, d) The cross-sectional area at the blind hole inlet at the maximum stroke of the needle valve Through-flow cross-sectional area at the lower edge of the nozzle needle valve seat must be greater than e) The difference between the valve seat half angle (δ / 2) and the blind hole angle (ε) must not be less than the value to be determined, f) Blind The hole angle (ε) must be smaller than the nozzle needle tip half angle (α / 2), g) The distance between the upper edge of the nozzle hole and the inlet edge of the blind hole inside the blind hole must not be less than the value to be determined H) The tip angle (α) of the nozzle needle is Must be greater than or equal to the seating angle (σ), i) less than the requirement that the nozzle needle seat lower diameter (DA) must be greater than or equal to the blind hole diameter (DE) And have one condition.

  That is, the method according to the present invention includes the minimization of the quality function by a mathematical optimization algorithm that simultaneously derives the best parameters of the fuel injection nozzle.

  The illustrated embodiment optimized in accordance with the present invention shows a single row perforated nozzle in the example of FIG. 1 and a double row perforated nozzle in FIG. It has such optimized parameters. That is, a nozzle needle valve seat angle (σ) of about 84 °, a nozzle needle tip angle (α) of about 100 °, a nozzle needle valve stroke of about 0.5 mm, a nozzle hole diameter (DL) of about 0.4 mm, 3 mm blind hole diameter (DE), about 3.5 mm nozzle needle valve seat lower diameter (DA), 0 ° blind hole angle (ε) about nozzle needle valve axis, about 75 ° injection angle (κ), about It has a blind hole depth (hS) of 0.3 mm, an ejection point (hSP) of about 0.1 mm, and the number of nozzle holes = 13.

  With the nozzle optimized according to the implemented weight of the technical requirements, the following improvements were obtained compared to the conventional nozzle. That is, the blind hole pressure was increased by about 10%, the blind hole volume was reduced by about 60%, and the wall thickness between the nozzle holes was reduced by about 7%.

  The so-developed fuel injection nozzles often exhibit optimum values of the opposing parameters for the technical requirements with respect to the formulated requirements. This ensures that an internal combustion engine equipped with the illustrated fuel injection nozzle is excellent in terms of minimal toxic generation, slight wear and minimal fuel consumption. At the same time, the development process of the illustrated fuel injection nozzle is tailored and therefore quick and reliable.

  The optimization method shown is generally suitable for all high-dimensional optimization purposes that are mathematically determined.

1 is a cross-sectional view of a fuel injection nozzle or hole nozzle optimized according to the present invention. FIG. 2 is a cross-sectional view of a hole nozzle optimized according to the present invention with a plurality of rows, here two rows of jet openings, extending circularly in the nozzle dome.

Explanation of symbols

1 Nozzle body 2 holes (blind holes)
3 Nozzle needle valve 4 Valve sealing surface 5 Injection opening (nozzle hole)
6 Valve seat surface 7 Nozzle dome DL Nozzle hole diameter DE Blind hole diameter DA Nozzle needle valve seat lower diameter κ Injection angle h Nozzle needle valve stroke σ Valve seat angle α Nozzle needle tip angle ε Blind hole angle (related to injection nozzle axis)
hS Blind hole depth hSP Ejection point n Number of nozzle holes

Claims (4)

A nozzle body (1) and a needle valve (3) that can move in the axial direction against the closing force in the hole (2) of the nozzle body (1) are provided, and the needle valve (3) is on the combustion chamber side. The valve body sealing surface (4) has a valve body sealing surface (4) at the end face, and the valve body sealing surface (4) has a nozzle body ( In a method for optimizing a fuel injection nozzle of an internal combustion engine in cooperation with the valve seat surface (6) of 1)
Mathematical method of particle swarm optimization (Particle Swarm Optimization, PSO) is used for geometrical formation of fuel injection nozzles, utilizing any number of independent parameters, and the best sought criteria Is reflected in the structural shape of the fuel injection nozzle, the method for optimizing the fuel injection nozzle of the internal combustion engine. Particle swarm optimization (particle swarm optimization, PSO)
-Determine the predetermined ambient conditions (nozzle type and nozzle size) for the geometry, eg the required nozzle flow-through and nozzle pressure stage;
-Technical requirements in the form of objective quantities of objective function (x) for any number of independent parameters such as maximum pressure, blind hole volume or minimum nozzle needle seat angle in the nozzle body hole (blind hole) ( Know how much to optimize)
-Deriving independent parameters (degrees of freedom) of technical requirements from the definition of target quantities such as nozzle needle valve seat angle, nozzle needle valve seat lower diameter, nozzle tip angle, injection angle or nozzle needle valve stroke,
-Determine parameter limits in the form of a predetermined numerical range;
-Mathematically weighting the objective quantity of the technical requirements in the quality function,
The method of claim 1 including a process.   The injection nozzle to be optimized is a hole type nozzle, a needle valve (3) guided so as to be movable in the axial direction against the closing force in the hole (2) of the nozzle body (1), and a blind hole (2 ) And a nozzle dome (7) that closes the combustion chamber, and the needle valve (3) has a valve body sealing surface (4) on its combustion chamber side end surface, and the valve body sealing surface (4) In cooperation with the conical valve seat surface (6) of the blind hole (2), and a plurality of injection openings (5) project from the nozzle dome (7) toward the combustion chamber of the internal combustion engine. The method according to claim 1 or 2. ・ Predetermined peripheral conditions include peripheral conditions that can be freely set to express other limitations related to nozzle type, nozzle size, and application.
Technical requirements include: a) maximum pressure in the blind hole (resulting in the best mixture formation), b) minimum blind hole volume (resulting in soot and HC generation and reduction in fuel consumption), c) nozzle hole (5) Minimum wall thickness between (due to required strength), d) Minimum nozzle needle seat angle (due to wear of nozzle needle seat), e) Blind hole (2) injection opening (5) as possible The depth of at least one, multiple or all of the deep positions (due to hydrodynamic requirements),
As independent parameters, a) valve seat angle (σ) of nozzle needle valve (3), b) nozzle needle tip angle (α), c) needle valve stroke (h), d) nozzle hole diameter (DL), e ) Blind hole diameter (DE), f) Nozzle needle valve seat lower diameter (DA), g) Blind hole angle (ε), h) Injection angle (κ), i) Injection point injection point (hSP), k) A blind hole depth (hS), l) number of nozzle holes, m) nozzle hole diameter (DL), at least one, a plurality or all of the sizes are given.
The determined parameter limit value includes a numerical range or a discrete value;
• All technical requirements (see above) are standardized and incorporated into the Q function, and a) the distance between the nozzle needle tip and the bottom of the blind hole (2) must not be less than the minimum value, b) The width of the nozzle needle valve seat must not be less than the minimum value, c) The wall thickness between the nozzle holes (5) must not be less than the minimum value, and d) At the blind hole inlet at the maximum stroke of the needle valve The cross-sectional area must be greater than the break-through area at the lower edge of the nozzle needle valve seat. E) The difference between the valve seat half angle (δ / 2) and the blind hole angle (ε) must not be less than the value to be determined. F) The blind hole angle (ε) must be smaller than the nozzle needle half tip angle (α / 2), g) The value of the distance between the upper edge of the nozzle hole and the blind hole inlet edge inside the blind hole should be determined. H) Nozzle needle tip angle (α) is valve Must be greater than or equal to the angle (σ), i) at least one of the conditions that the nozzle needle seat lower diameter (DA) must be greater than or equal to the blind hole diameter (DE) Having
The method according to claim 3.

Images