JPH04231647A - Air-fuel ratio controller for internal combustion engine - Google Patents

Air-fuel ratio controller for internal combustion engine

Info

Publication number
JPH04231647A
JPH04231647A JP40909990A JP40909990A JPH04231647A JP H04231647 A JPH04231647 A JP H04231647A JP 40909990 A JP40909990 A JP 40909990A JP 40909990 A JP40909990 A JP 40909990A JP H04231647 A JPH04231647 A JP H04231647A
Authority
JP
Japan
Prior art keywords
air
fuel ratio
fuel
injection amount
amount
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP40909990A
Other languages
Japanese (ja)
Inventor
Masataka Osawa
正敬 大澤
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Toyota Central R&D Labs Inc
Original Assignee
Toyota Central R&D Labs Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Toyota Central R&D Labs Inc filed Critical Toyota Central R&D Labs Inc
Priority to JP40909990A priority Critical patent/JPH04231647A/en
Publication of JPH04231647A publication Critical patent/JPH04231647A/en
Pending legal-status Critical Current

Links

Landscapes

  • Electrical Control Of Air Or Fuel Supplied To Internal-Combustion Engine (AREA)

Abstract

PURPOSE:To prevent the air-fuel ratio control lag by aptly correcting the fuel injection amount against a fuel delay presenting complicated behaviors. CONSTITUTION:The basic fuel injection amounts qw, the change rate DELTAqw of qw, the change rates DELTA<2>qw of change rates, the number of revolutions Ne, the intake pressure Ps, and the cooling-water temperature THW are inputted to an input layer of a neural network, and each of these factors is multiplied by weights gammaij<12>, gammaj<23>, and the products thus obtained are added together to determine a correction amount Rw. The weights are corrected according to a deviation between a target air-fuel ratio and an actual air-fuel ratio. The basic injection amount of fuel and the correction amount Rw are added together to determine an injection amount of fuel to thereby control the air-fuel ratio.

Description

【発明の詳細な説明】[Detailed description of the invention]

【0001】0001

【産業上の利用分野】本発明は内燃機関の空燃比制御装
置に係り、特に機関燃焼室に吸入される吸入空気量に関
連した物理量を予測しかつニューラルネットワークを用
いた非線型学習制御によって空燃比を制御する内燃機関
の空燃比制御装置に関する。
[Field of Industrial Application] The present invention relates to an air-fuel ratio control device for an internal combustion engine, and in particular, it predicts a physical quantity related to the amount of intake air taken into an engine combustion chamber and performs air-fuel ratio control using nonlinear learning control using a neural network. The present invention relates to an air-fuel ratio control device for an internal combustion engine that controls a fuel ratio.

【0002】0002

【従来の技術】内燃機関の空燃比学習制御装置として、
特開昭63−285239号公報に記載された装置があ
る。この装置は、吸気圧力または吸入空気量の変化率と
機関回転速度とに応じて定められた燃料噴射量の増減マ
ップを用いて、加減速時の燃料噴射量の補正を行ってい
る。また、加減速時の空燃比の基準値からの最大ずれ量
に応じて学習によって増減マップの値を修正している。
[Prior Art] As an air-fuel ratio learning control device for an internal combustion engine,
There is an apparatus described in Japanese Patent Application Laid-Open No. 63-285239. This device corrects the fuel injection amount during acceleration and deceleration using a fuel injection amount increase/decrease map determined according to the rate of change in intake pressure or intake air amount and engine rotational speed. Furthermore, the values of the increase/decrease map are corrected by learning according to the maximum deviation amount of the air-fuel ratio from the reference value during acceleration/deceleration.

【0003】上記従来の技術では、吸気圧力または吸入
空気量の変化率に応じて燃料噴射量を補正しているが、
燃料遅れに対する燃料噴射量の補正はこの変化率だけで
は正確に規定できず、変化率の変化率等を用いた複雑な
補正が必要である。従って、上記従来の技術では複雑な
挙動を示す燃料遅れに対する燃料噴射量の補正を適切に
行うことができない、という問題がある。また、増減マ
ップの修正を空燃比の最大ずれ量を用いて行っているだ
けで、空燃比ずれ量の時々刻々の値を用いて修正してい
ないため、内燃機関の機台間差(機差)や経時変化等に
よる空燃比ずれを補正できない、という問題がある。す
なわち、空燃比のずれ量は、吸気系容積による気筒内へ
の吸入空気量の吸入遅れ、燃料のインテークマニホール
ド内壁面等への付着、液面流れ等を原因とする燃料遅れ
により生ずるが、この空燃比のずれ量の時々刻々の値お
よびずれの波形パターンには、種々の遅れがどのように
寄与しているかの情報が含まれている。しかしながら、
上記のように最大ずれ量を用いると、この情報が有効に
使用されず、空燃比が正確に制御されなくなる。
[0003] In the above conventional technology, the fuel injection amount is corrected according to the rate of change in intake pressure or intake air amount.
Correction of the fuel injection amount for the fuel delay cannot be accurately defined based only on this rate of change, and requires complicated correction using the rate of change of the rate of change. Therefore, the conventional technique described above has a problem in that it is not possible to appropriately correct the fuel injection amount for the fuel delay that exhibits complicated behavior. In addition, since the increase/decrease map is corrected only by using the maximum air-fuel ratio deviation amount and not by using the momentary value of the air-fuel ratio deviation amount, the internal combustion engine difference between machines (machine difference ), air-fuel ratio deviations due to changes over time, etc. cannot be corrected. In other words, the amount of deviation in the air-fuel ratio is caused by a delay in the intake air amount into the cylinder due to the volume of the intake system, fuel adhesion to the inner wall of the intake manifold, fuel delay due to liquid level flow, etc. The momentary value of the air-fuel ratio deviation amount and the waveform pattern of the deviation include information on how various delays contribute. however,
If the maximum deviation amount is used as described above, this information will not be used effectively and the air-fuel ratio will not be accurately controlled.

【0004】0004

【発明が解決しようとする課題】本発明は上記問題点を
解決すべくなされたもので、複雑な挙動を示す燃料遅れ
に対する燃料噴射量補正を、基本噴射量、基本噴射量の
変化率、この変化率の変化率、更には機関回転速度、吸
気圧力、冷却水温等の機関運転状態量に重みを加えて演
算し、加減速時等の過渡時の空燃比を適切な値に保持す
ることができると共に、空燃比ずれ量の時々刻々の値を
用いて補正要素を修正して機関の機差や経時変化等によ
る空燃比の制御ずれを防止することができる内燃機関の
空燃比制御装置を提供することを目的とする。
SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and it is an object of the present invention to correct the fuel injection amount for the fuel delay, which exhibits complicated behavior, by adjusting the basic injection amount, the rate of change of the basic injection amount, and the basic injection amount. It is possible to maintain the air-fuel ratio at an appropriate value during transients such as acceleration and deceleration by calculating the rate of change and adding weight to engine operating state variables such as engine speed, intake pressure, and cooling water temperature. Provided is an air-fuel ratio control device for an internal combustion engine that can prevent air-fuel ratio control deviations due to engine machine differences, changes over time, etc. by correcting correction elements using momentary values of air-fuel ratio deviation amounts. The purpose is to

【0005】[0005]

【課題を解決するための手段】上記目的を達成するため
に本発明は、機関回転速度を検出する回転速度検出手段
と、機関の暖気状態量を検出する暖気状態検出手段と、
機関に吸入される吸入空気量に関連した物理量を検出す
る物理量検出手段と、機関の排気から排気空燃比を検出
する空燃比検出手段と、前記機関回転速度及び前記物理
量に基づいて基本噴射量を演算する基本噴射量演算手段
と、前記機関回転速度と前記暖気状態量と前記吸入空気
量に関連した物理量と前記基本噴射量と該基本噴射量の
変化率と該変化率の変化率とを入力とし燃料供給遅れに
対応した前記基本噴射量を補正する補正量を各入力に重
みを加えて演算する補正量演算手段と、前記排気空燃比
と目標空燃比との差に応じて前記補正量演算手段の重み
を修正する重み修正手段と、前記基本噴射量と前記補正
量とによって燃料噴射量を求め該燃料噴射量によって空
燃比を制御する空燃比制御手段と、を含んで構成したも
のである。
[Means for Solving the Problems] In order to achieve the above object, the present invention provides a rotation speed detection means for detecting the engine rotation speed, a warm state detection means for detecting the warm state quantity of the engine,
a physical quantity detecting means for detecting a physical quantity related to the amount of intake air taken into the engine; an air-fuel ratio detecting means for detecting an exhaust air-fuel ratio from the exhaust gas of the engine; and a basic injection amount based on the engine rotation speed and the physical quantity. inputting a basic injection amount calculation means to be calculated, the engine rotational speed, the warm-up state quantity, the physical quantity related to the intake air amount, the basic injection amount, the rate of change of the basic injection amount, and the rate of change of the rate of change; correction amount calculation means for calculating a correction amount for correcting the basic injection amount corresponding to the fuel supply delay by adding weight to each input; and correction amount calculation means for calculating the correction amount according to the difference between the exhaust air-fuel ratio and the target air-fuel ratio. The fuel injection device is configured to include weight correction means for correcting the weight of the means, and air-fuel ratio control means for determining the fuel injection amount based on the basic injection amount and the correction amount and controlling the air-fuel ratio based on the fuel injection amount. .

【0006】[0006]

【作用】本発明の基本噴射量演算手段は、機関回転速度
及び吸入空気量に関連した物理量に基づいて、機関燃焼
室に吸入される吸入空気量に応じた基本噴射量を演算す
る。機関燃焼室に吸入される吸入空気量に関連した物理
量としては、スロットル弁と機関燃焼室との間の吸気管
内の吸入空気の圧力(吸気圧力)またはスロットル弁の
上流側を通過する吸入空気量がある。補正量演算手段は
、基本燃料噴射量を補正するための燃料遅れに対応した
補正量を、機関回転速度、暖機状態量、機関に吸入され
る吸入空気量に関連した物理量、基本噴射量、基本噴射
量の変化率、この変化率の変化率を入力とし、各入力に
重みを加えて演算する。重み修正手段は、排気空燃比と
目標空燃比との時々刻々の差に応じて補正量演算手段の
重みを修正する。そして、基本噴射量と補正量とによっ
て燃料噴射量を求め、この燃料噴射量によって空燃比を
制御する。このように、基本噴射量を補正しているため
、基本噴射量に複雑な挙動の燃料遅れに対応した補正成
分が付加され、吸気管内に噴射された燃料が燃焼室内に
吸入されるまでの輸送遅れや吸気管内壁付着量による遅
れ等を補正することができる。また、時々刻々の排気空
燃比と目標空燃比との差に応じて重みを修正しているた
め、多様な位相進み成分が必要とされる補正量と、排気
空燃比と目標空燃比との差との時間的な対応がとられ、
機関の状態変化により変化した燃料遅れに合致した補正
パターンを重みを修正することにより形成可能となり、
内燃機関の機差や時間経過による内燃機関の状態変化に
対する補正を行うことができる。
[Operation] The basic injection amount calculating means of the present invention calculates a basic injection amount corresponding to the amount of intake air taken into the engine combustion chamber, based on physical quantities related to the engine rotational speed and the amount of intake air. Physical quantities related to the amount of intake air taken into the engine combustion chamber include the pressure of intake air in the intake pipe between the throttle valve and the engine combustion chamber (intake pressure), or the amount of intake air passing upstream of the throttle valve. There is. The correction amount calculation means calculates the correction amount corresponding to the fuel delay for correcting the basic fuel injection amount by calculating the engine rotation speed, the warm-up state quantity, the physical quantity related to the amount of intake air taken into the engine, the basic injection amount, The rate of change of the basic injection amount and the rate of change of this rate of change are input, and calculations are performed by adding weights to each input. The weight correction means corrects the weight of the correction amount calculation means in accordance with the momentary difference between the exhaust air-fuel ratio and the target air-fuel ratio. Then, the fuel injection amount is determined from the basic injection amount and the correction amount, and the air-fuel ratio is controlled using this fuel injection amount. In this way, since the basic injection amount is corrected, a correction component corresponding to the complex behavior of fuel delay is added to the basic injection amount, and the transport of the fuel injected into the intake pipe until it is sucked into the combustion chamber. It is possible to correct the delay and the delay due to the amount of adhesion on the inner wall of the intake pipe. In addition, since the weight is modified according to the difference between the exhaust air-fuel ratio and the target air-fuel ratio from time to time, various phase advance components are required to correct the amount of correction and the difference between the exhaust air-fuel ratio and the target air-fuel ratio. A time-wise response was taken to
By modifying the weights, it is possible to form a correction pattern that matches the fuel delay that changes due to engine state changes.
It is possible to make corrections for machine differences in the internal combustion engine and changes in the state of the internal combustion engine over time.

【0007】また、スロットル弁の開度を検出し、この
スロットル弁の開度および前記物理量に基づいて将来の
物理量を予測する予測手段を更に設け、前記予測手段で
予測された将来の物理量に基づいて基本噴射量を演算す
る構成とすることもできる。多気筒内燃機関の場合、機
関1回転に1回あるいは複数回全気筒同時に燃料を噴射
する場合と、各気筒独立に燃料を噴射する場合とがある
。全気筒同時噴射の場合は、噴射された燃料の全てが特
定気筒に吸入されるわけではなく、他の気筒にも吸入さ
れるため、予測すべき将来の時点は実験に基づいて定め
るのが好ましい。一方、各気筒独立噴射の場合は、噴射
された燃料が特定気筒にのみ吸入されるため、予測すべ
き将来の時点は、この特定気筒に吸入される吸入空気量
が確定する時点、すなわち特定気筒の吸気弁開期間中の
吸気弁閉弁時期に近い時期がよい。予測手段で予測され
た将来の物理量に基づいて基本燃料噴射量を演算するこ
とによって基本噴射量を機関の燃焼室に吸入される吸入
空気量に応じて精度良く供給することができる。
[0007] Furthermore, there is further provided a prediction means for detecting the opening degree of the throttle valve and predicting a future physical quantity based on the opening degree of the throttle valve and the physical quantity, and predicting a future physical quantity based on the future physical quantity predicted by the prediction means. It is also possible to have a configuration in which the basic injection amount is calculated by using the basic injection amount. In the case of a multi-cylinder internal combustion engine, there are cases where fuel is simultaneously injected into all cylinders once or multiple times per engine revolution, and cases where fuel is injected into each cylinder independently. In the case of simultaneous injection in all cylinders, not all of the injected fuel is inhaled into a specific cylinder, but also into other cylinders, so it is preferable to determine the future point to be predicted based on experiments. . On the other hand, in the case of independent injection for each cylinder, the injected fuel is inhaled only into a specific cylinder, so the future point in time that should be predicted is the point in time when the amount of intake air that will be inhaled into this specific cylinder is determined, that is, the specific cylinder The best timing is close to the intake valve closing timing during the intake valve opening period. By calculating the basic fuel injection amount based on the future physical quantity predicted by the prediction means, the basic injection amount can be supplied with high accuracy according to the amount of intake air taken into the combustion chamber of the engine.

【0008】[0008]

【実施例】以下図面を参照して本発明の実施例を詳細に
説明する。本実施例は、4気筒内燃機関に本発明を適用
し、4気筒独立に燃料噴射を行うことによって空燃比を
制御するようにしたものである。また、本実施例では機
関燃焼室に吸入される吸入空気量に関連した物理量とし
て吸気圧力を用いている。図1に示すように、4気筒内
燃機関26にはスロットルボディ10およびインテーク
マニホールド20を備えた吸気管、エキゾーストマニホ
ール32を備えた排気管34が連結されている。スロッ
トルボディ10には、機関26の燃焼室に吸入される吸
入空気量を調節するスロットル弁12が取付けられてい
る。このスロットル弁12には、スロットル弁12の開
度を検出してスロットル弁開度に比例した信号を出力す
るスロットル開度センサ14が取付けられている。イン
テークマニホールド20には、インテークマニホールド
20内の吸入空気の絶対圧力(吸気圧力)を検出する半
導体式の吸気圧センサ16および吸入空気の温度を検出
する吸気温センサ18が取付けられている。そしてイン
テークマニホールド20の各枝管24には、各気筒に対
応させて燃料噴射弁22がそれぞれ取付けられている。 また排気管34のエキゾーストマニホールド下流側には
、排気空燃比を検出するO2センサ等で構成された空燃
比センサ36が取付けられている。また機関26には、
暖機状態を代表する量である機関冷却水温度を検出する
冷却水温センサ28およびカムシャフトの回転を検出し
て機関回転速度信号を出力するカムポジションセンサ3
0が取付けられている。なお、機関オイル温で暖機状態
を代表してもよい。
DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In this embodiment, the present invention is applied to a four-cylinder internal combustion engine, and the air-fuel ratio is controlled by independently injecting fuel into the four cylinders. Further, in this embodiment, intake pressure is used as a physical quantity related to the amount of intake air taken into the engine combustion chamber. As shown in FIG. 1, the four-cylinder internal combustion engine 26 is connected to an intake pipe including a throttle body 10 and an intake manifold 20, and an exhaust pipe 34 including an exhaust manifold 32. A throttle valve 12 is attached to the throttle body 10 to adjust the amount of intake air taken into the combustion chamber of the engine 26. A throttle opening sensor 14 is attached to the throttle valve 12 to detect the opening of the throttle valve 12 and output a signal proportional to the throttle valve opening. Attached to the intake manifold 20 are a semiconductor-type intake pressure sensor 16 that detects the absolute pressure (intake pressure) of intake air within the intake manifold 20, and an intake air temperature sensor 18 that detects the temperature of the intake air. A fuel injection valve 22 is attached to each branch pipe 24 of the intake manifold 20 in correspondence with each cylinder. Further, an air-fuel ratio sensor 36 composed of an O2 sensor or the like that detects the exhaust air-fuel ratio is attached to the exhaust pipe 34 on the downstream side of the exhaust manifold. Also, in engine 26,
A cooling water temperature sensor 28 that detects the engine cooling water temperature, which is a quantity representative of the warm-up state, and a cam position sensor 3 that detects the rotation of the camshaft and outputs an engine rotation speed signal.
0 is attached. Note that the warm-up state may be represented by the engine oil temperature.

【0009】燃料噴射弁22の各々は駆動回路44に接
続されている。この駆動回路44は燃料噴射量信号を出
力するマイクロコンピュータ54に接続されている。そ
して、マイクロコンピュータ54には、スロットル開度
センサ14、吸気圧センサ16、吸気温センサ18、冷
却水温センサ28、カムポジションセンサ30、空燃比
センサ36及び燃料噴射弁22が接続されている。
[0009] Each of the fuel injection valves 22 is connected to a drive circuit 44. This drive circuit 44 is connected to a microcomputer 54 that outputs a fuel injection amount signal. The microcomputer 54 is connected to a throttle opening sensor 14, an intake pressure sensor 16, an intake temperature sensor 18, a cooling water temperature sensor 28, a cam position sensor 30, an air-fuel ratio sensor 36, and a fuel injection valve 22.

【0010】このマイクロコンピュータ54は、スロッ
トル開度センサ14、吸気圧センサ16、吸気温センサ
18等から入力される信号に基づいて燃料噴射量を演算
し、燃料噴射量信号を出力するもので、RAM、ROM
、CPU、およびこれらを接続するバス等で構成されて
いる。
The microcomputer 54 calculates a fuel injection amount based on signals input from the throttle opening sensor 14, intake pressure sensor 16, intake temperature sensor 18, etc., and outputs a fuel injection amount signal. RAM, ROM
, a CPU, and a bus that connects these.

【0011】このマイクロコンピュータ54を機能によ
って分割した機能ブロックで説明する。吸気圧力予測手
段38は、機関回転速度Ne、吸気圧力Ps、スロット
ル開度θを用いて以下の(2) 式に基づいて各気筒の
吸気圧力の予測値Qを演算する。なお、吸気温THAを
更に付加して予測してもよい。基本噴射量演算手段40
は、吸気圧力予測手段38で演算された吸気圧力の予測
値Q、機関回転速度Ne、吸気温THA、冷却水温TH
W、および目標空燃比A/F0 に基づいて以下の(4
) 式に従って排気空燃比A/Fを目標空燃比A/F0
 にするための基本燃料噴射量を各気筒毎に演算する。 入力変換手段42は、基本燃料噴射量を補正するための
補正量を演算する非線形学習手段52への入力、すなわ
ち吸気圧力、機関回転速度、冷却水温、基本噴射量qw
、基本噴射量の変化率Δqw、この変化率の変化率Δ2
 qwを各気筒毎に演算するとともに、非線形学習手段
52への入力を演算するための状態量を記憶する。なお
、第W気筒に対するΔqw、Δ2 qwは以下の(1)
 式で与えられる。
The microcomputer 54 will be explained using functional blocks divided by function. The intake pressure prediction means 38 calculates a predicted value Q of the intake pressure of each cylinder based on the following equation (2) using the engine rotational speed Ne, intake pressure Ps, and throttle opening θ. Note that the prediction may be made by further adding the intake air temperature THA. Basic injection amount calculation means 40
are the predicted value Q of the intake pressure calculated by the intake pressure prediction means 38, the engine rotation speed Ne, the intake air temperature THA, and the cooling water temperature TH.
Based on W and the target air-fuel ratio A/F0, the following (4
) According to the formula, the exhaust air-fuel ratio A/F is set to the target air-fuel ratio A/F0.
The basic fuel injection amount for each cylinder is calculated for each cylinder. The input conversion means 42 inputs inputs to the nonlinear learning means 52 that calculates a correction amount for correcting the basic fuel injection amount, that is, intake pressure, engine rotation speed, cooling water temperature, and basic injection amount qw.
, rate of change of basic injection amount Δqw, rate of change of this rate of change Δ2
qw is calculated for each cylinder, and state quantities for calculating input to the nonlinear learning means 52 are stored. In addition, Δqw and Δ2 qw for the W cylinder are as follows (1)
It is given by Eq.

【0012】   Δqw(kw)=qw(kw)−qw(kw−1)
  Δ2 qw(kw)=Δqw(kw)−Δqw(k
w−1)…(1) ただし、wは気筒を示しw=1のと
き第1気筒#1、w=2のとき第3気筒#3、w=3の
とき第4気筒#4、w=4のとき第2気筒#2を示す。
Δqw(kw)=qw(kw)−qw(kw−1)
Δ2 qw (kw) = Δqw (kw) − Δqw (k
w-1)...(1) However, w indicates a cylinder, and when w=1, the first cylinder #1, when w=2, the third cylinder #3, when w=3, the fourth cylinder #4, and w= 4 indicates the second cylinder #2.

【0013】非線形学習手段52は、7素子の入力層、
L素子の中間層及び1素子の出力層から成るニューラル
ネットワークを備えると共に、減算手段50で演算され
た目標空燃比A/F0 と排気空燃比A/Fとの差に基
づいて重みを演算する演算手段を備え、演算された重み
と入力変換手段42で演算された複数の入力量とから、
補正量を演算する。加算手段46は基本燃料噴射量と補
正量を加算することによって燃料噴射量を求め、この燃
料噴射量に対応する燃料噴射量信号を駆動回路44に入
力する。駆動回路44は、燃料噴射量信号に基づいて燃
料噴射弁22を各気筒独立に制御する。
The nonlinear learning means 52 includes an input layer of seven elements,
A calculation comprising a neural network consisting of an intermediate layer of L elements and an output layer of one element, and calculating weights based on the difference between the target air-fuel ratio A/F0 calculated by the subtraction means 50 and the exhaust air-fuel ratio A/F. from the calculated weight and the plurality of input quantities calculated by the input conversion means 42,
Calculate the correction amount. The adding means 46 calculates the fuel injection amount by adding the basic fuel injection amount and the correction amount, and inputs a fuel injection amount signal corresponding to this fuel injection amount to the drive circuit 44. The drive circuit 44 independently controls the fuel injection valves 22 for each cylinder based on the fuel injection amount signal.

【0014】次に、上記マイクロコンピュータ54によ
って燃料噴射量を演算するための演算ルーチンを詳細に
説明する。図2は、燃料噴射演算ルーチンを示すもので
、このルーチンは、図9に示す演算タイミングで180
°CA(クランク角)毎に実行される。ステップ100
において機関回転速度Ne、吸気圧力Ps、スロットル
開度θ、吸気温THA、冷却水温THW、排気空燃比A
/F、目標空燃比A/F0 を取り込み、ステツプ11
1で機関回転速度Ne、吸気圧力Psにより現在の運転
状態が存在する領域を決定する。次のステップ102に
おいて現在の運転状態が図3に示す予測領域のオーバー
ラップ領域(斜線で示す部分)に属しているかを判断す
る。予測領域は、回転速度Neと吸気圧力Psとによっ
て複数の領域に分割されており、各領域の境界はオーバ
ーラップさせてオーバーラップ領域が設けられている。 ステップ102でオーバーラップ領域でないと判断され
たときは、ステップ104において現在の運転状態が属
する領域Sの時刻(kw+2)の時点、すなわち吸気弁
開期間中の吸気弁閉弁時期に近い時時期での第W気筒の
吸気圧力の予測値Qs(kw+2)を以下の(2) 式
に従って演算する。ただし、Sは図3の領域の番号を示
す整数である。
Next, a calculation routine for calculating the fuel injection amount by the microcomputer 54 will be explained in detail. FIG. 2 shows a fuel injection calculation routine, and this routine is executed at 180 at the calculation timing shown in FIG.
Executed every °CA (crank angle). step 100
At engine speed Ne, intake pressure Ps, throttle opening θ, intake temperature THA, cooling water temperature THW, exhaust air-fuel ratio A
/F, target air-fuel ratio A/F0 is taken in, step 11
1, the region in which the current operating state exists is determined based on the engine rotational speed Ne and the intake pressure Ps. In the next step 102, it is determined whether the current driving state belongs to the overlap region (shaded area) of the prediction region shown in FIG. The prediction region is divided into a plurality of regions depending on the rotational speed Ne and the intake pressure Ps, and the boundaries of each region overlap to provide an overlapping region. If it is determined in step 102 that the region is not an overlap region, then in step 104 the time (kw+2) of the region S to which the current operating state belongs, that is, the time close to the intake valve closing timing during the intake valve open period is determined. The predicted value Qs(kW+2) of the intake pressure of the W-th cylinder is calculated according to the following equation (2). However, S is an integer indicating the number of the area in FIG.

【0015】   Qs(kw+2)=Ps(kw)+a1 (Ps(
kw)−Ps(kw−2))+b1 (θ(kw)−θ
(kw−2))+b2 (θ(kw−1)−θ(kw−
3))+c1 (Ne(kw)−Ne(kw−1))+
c2 (Ne(kw−1)−Ne(kw−3))…(2
) なお、(2) 式において吸気温THAを更に加え
て予測値を求めてもよい。ステップ108において領域
Sの予測値Qs(kw+2)を第W気筒の基本燃料噴射
量qwを演算するための予測値Qとしてステップ116
へ進む。
Qs(kw+2)=Ps(kw)+a1 (Ps(
kw)-Ps(kw-2))+b1 (θ(kw)-θ
(kw-2))+b2 (θ(kw-1)-θ(kw-
3))+c1 (Ne(kw)-Ne(kw-1))+
c2 (Ne(kw-1)-Ne(kw-3))...(2
) Note that the predicted value may be obtained by further adding the intake air temperature THA to equation (2). In step 108, the predicted value Qs (kW+2) of the region S is set as the predicted value Q for calculating the basic fuel injection amount qw of the W cylinder, and in step 116
Proceed to.

【0016】ステップ102でオーバーラップ領域と判
断されたときは、ステップ110においてオーバーラッ
プ領域に属する領域m、n(m、nは領域の番号を示す
)の時刻(kw+2)の時点の予測値Qm(kw+2)
、Qn(kw+2)を上記(2) 式に基づいて演算す
る。ステップ112において現在の運転状態が領域m、
領域nに属している度合に対応する重み、すなわち適合
度wm、wnを演算する。この重みwm、wnは、図4
に示すように領域2と領域3とがオーバラップしている
ときには、w2 、w3 となる。そして、ステップ1
14において以下の(3) 式に示す加重平均によって
第W気筒の基本燃料噴射量qwを演算するための予測値
Qを演算する。
When it is determined in step 102 that the area is an overlapping area, in step 110, the predicted value Qm of areas m and n (m and n indicate area numbers) belonging to the overlapping area at time (kw+2) is calculated. (kW+2)
, Qn(kw+2) are calculated based on equation (2) above. In step 112, the current operating state is in the area m,
Weights corresponding to the degree of belonging to region n, that is, degrees of suitability wm and wn are calculated. These weights wm and wn are shown in FIG.
When region 2 and region 3 overlap as shown in FIG. 2, w2 and w3 are obtained. And step 1
14, a predicted value Q for calculating the basic fuel injection amount qw of the W cylinder is calculated by the weighted average shown in the following equation (3).

【0017】[0017]

【数1】[Math 1]

【0018】ただし、wm、wnは0≦wm、wn≦1
と正規化する。次のステップ116では時刻(kw+2
)での予測値Qと、現時刻kwでの機関回転速度Ne、
吸気温THA、冷却水温THW、および目標空燃比A/
F0 とに基づいて燃料遅れに対応した第W気筒の基本
燃料噴射量qwを演算する。
[0018] However, wm and wn are 0≦wm, wn≦1
and normalize it. In the next step 116, the time (kw+2
) and the engine rotational speed Ne at the current time kW,
Intake air temperature THA, cooling water temperature THW, and target air-fuel ratio A/
The basic fuel injection amount qw of the W-th cylinder corresponding to the fuel delay is calculated based on F0.

【0019】   qw=f(Q、Ne、THA、THW、A/F0 
)…(4) ただしfは関数を表す。
qw=f(Q, Ne, THA, THW, A/F0
)...(4) where f represents a function.

【0020】次のステップ118では、第W気筒の基本
燃料噴射量qw等に基づいて燃料遅れに対応した補正量
を演算する。すなわち、図5に示すように、ステツプ1
19において基本噴射量の変化率、すなわち1階差分及
びこの変化率の変化率、すなわち2階差分を演算する。 次に図5で示すようなニューラルネットワークの前記基
本噴射量、該基本噴射量の1階および2階差分ならびに
機関回転速度Ne、吸気圧力Ps、冷却水温THWに対
応する重み   Γr(kw)=  γij12(kw),γj 2
3(kw)…(5) (ただし、i=1、2、・・・7
,j=1、2、・・・Lである。)を取込み、ステップ
121で以下の(6) 式に従って補正量Rwを演算す
る。
In the next step 118, a correction amount corresponding to the fuel delay is calculated based on the basic fuel injection amount qw of the W cylinder, etc. That is, as shown in FIG.
In step 19, the rate of change in the basic injection amount, ie, the first-order difference, and the rate of change of this rate of change, ie, the second-order difference, are calculated. Next, the basic injection amount of the neural network as shown in FIG. 5, the first and second differences in the basic injection amount, and the weights Γr (kw) = γij12 corresponding to the engine rotational speed Ne, intake pressure Ps, and cooling water temperature THW. (kw), γj 2
3 (kw)...(5) (however, i=1, 2,...7
, j=1, 2, . . . L. ), and in step 121, the correction amount Rw is calculated according to the following equation (6).

【0021】すなわち入力層i素子から中間層j 素子
への重みをγij12、中間層j素子から出力素子への
重みをγj 23として、中間層の各素子jへの入力を
Uj、中間層の出力をhj、出力層の出力、すなわち補
正量をRwとすれば、補正量Rwは次のようになる。
That is, the weight from the input layer i element to the intermediate layer j element is γij12, the weight from the intermediate layer j element to the output element is γj23, the input to each element j in the intermediate layer is Uj, and the output of the intermediate layer is Let hj be hj, and the output of the output layer, that is, the correction amount, be Rw, the correction amount Rw will be as follows.

【0022】[0022]

【数2】[Math 2]

【0023】ここで、f(x)は図7に示すような各素
子の入出力を表す関数であり、f(x)=1/(1+e
xp(−x))を用いる。またγ7j12、γL 23
は入出力関数f(x)におけるしきい値の学習値である
Here, f(x) is a function representing the input and output of each element as shown in FIG. 7, and f(x)=1/(1+e
xp(-x)). Also γ7j12, γL 23
is the learning value of the threshold value in the input/output function f(x).

【0024】また、ktは、入出力関数f(x)が単位
出力[0,1]の範囲しかとらないため、実物理量に変
換する係数である。
Further, kt is a coefficient for converting into an actual physical quantity since the input/output function f(x) takes only the range of unit output [0, 1].

【0025】次のステップ124では、第W気筒の基本
燃料噴射量qwに補正量Rwを加算することにより第W
気筒の燃料噴射量を演算し、所定の燃料噴射タイミング
になった時点で第W気筒に対応する燃料噴射弁22から
燃料を噴射させる。
In the next step 124, the W-th cylinder is adjusted by adding the correction amount Rw to the basic fuel injection amount qw of the W-th cylinder.
The fuel injection amount for each cylinder is calculated, and at a predetermined fuel injection timing, fuel is injected from the fuel injection valve 22 corresponding to the W-th cylinder.

【0026】図8は、図9に示す学習タイミングで18
0°CA毎に実行される学習ルーチンを示すものである
。ステップ150において以下の式に従って目標空燃比
A/F0 と排気空燃比A/Fとの差δ(kw)を演算
すると共に、前回演算された重みΓr(kw−1)(=
Δγij12(kw−1),Δγj 23(kw−1)
)を取り込む。そしてステップ152において以下の(
7) 式に従って現在時刻(時刻kw)の重みΓr(k
w)を計算する。ここでδ(kw)=A/F0 −A/
Fとする。
FIG. 8 shows 18 at the learning timing shown in FIG.
This shows a learning routine that is executed every 0° CA. In step 150, the difference δ (kw) between the target air-fuel ratio A/F0 and the exhaust air-fuel ratio A/F is calculated according to the following formula, and the previously calculated weight Γr (kw-1) (=
Δγij 12 (kw-1), Δγj 23 (kw-1)
). Then, in step 152, the following (
7) The weight Γr(k
Calculate w). Here δ(kw)=A/F0 −A/
Let it be F.

【0027】出力層の重みΔγj 23(kw)は、The weight of the output layer Δγj 23 (kw) is


0028】
[
0028

【数3】[Math 3]

【0029】中間層の重み修正量Δγij12(kw)
[0029] Intermediate layer weight correction amount Δγij12 (kw)
teeth

【0030】[0030]

【数4】[Math 4]

【0031】ただし、dqw(kw´)はi=1のとき
qw i=2のときΔqw i=3のときΔ2 qw i=4のときNe i=5のときPs i=6のときTHW i=7のとき1.0 であり、dqw(kw´)は時刻kwにおいて測定され
た空燃比に影響を及ぼす時刻kw´の該状態量である。
However, dqw (kw') is qw when i=1, Δqw when i=2, Δ2 when i=3, Δ2 when qw i=4, Ne when i=5, Ps when i=6, THW i= 7, it is 1.0, and dqw(kw') is the state quantity at time kw' that affects the air-fuel ratio measured at time kw.

【0032】上式中ε、αは正の定数であり、前記の重
みの修正を効率的に行なう所定の値とし、特にαは重み
の過度な修正を防止するため1以下とすることが好まし
い。
In the above formula, ε and α are positive constants, and are set to predetermined values for efficiently modifying the weights. In particular, α is preferably set to 1 or less to prevent excessive modification of the weights. .

【0033】そして、ステップ154において重みΓr
(kw)を記憶する。図10に、上記実施例の空燃比の
変化を従来例と比較して示す。
Then, in step 154, the weight Γr
(kw). FIG. 10 shows changes in the air-fuel ratio in the above embodiment in comparison with the conventional example.

【0034】以上説明したように本実施例によれば、吸
入行程における吸気圧力(吸気弁閉弁時の吸気圧力)を
燃料噴射量を演算する時刻において現時刻および過去の
吸気圧力、回転速度、スロットル開度等から予測してい
るので、スロットル開度、回転速度が変化している過渡
状態においても正確に将来の吸気圧力を予測することが
できる。
As explained above, according to this embodiment, the intake pressure in the intake stroke (the intake pressure when the intake valve is closed) is calculated based on the current and past intake pressures, rotational speeds, and Since the prediction is made from the throttle opening, etc., the future intake pressure can be accurately predicted even in a transient state where the throttle opening and rotational speed are changing.

【0035】予測領域を複数の運転領域に分割して用意
し、各領域間をオーバーラップさせているため、各領域
内では簡単な線形式で予測値を求めることができ、かつ
オーバーラップ領域ではいくつかの線形式の和で予測値
を求めることができる。従ってオーバーラップ領域で別
の予測演算式を用いなくても済み、これにより、演算を
簡潔にすることができると共に演算式を必要最小限にす
ることができる。
[0035] Since the prediction region is divided into a plurality of operating regions and each region overlaps, the predicted value can be obtained in a simple linear form within each region, and in the overlap region The predicted value can be obtained by summing several linear forms. Therefore, there is no need to use a separate prediction calculation formula in the overlap region, which makes it possible to simplify the calculation and minimize the number of calculation formulas necessary.

【0036】また基本噴射量、基本噴射量の変化率、そ
の変化率、更には機関回転速度、吸気圧力、冷却水温等
の機関運転状態量を入力とし、該入力の非線形な組合せ
出力として得られる基本噴射量の補正量を求める非線形
補正演算手段を用いるため、複雑な挙動を示す燃料遅れ
のダイナックスを補償する動的な補正が可能となり燃料
の吸入遅れによる空燃比ずれを補正することができる。
In addition, the basic injection amount, the rate of change of the basic injection amount, the rate of change thereof, and engine operating state quantities such as engine speed, intake pressure, and cooling water temperature are input, and the output is obtained as a nonlinear combination of the inputs. Since a non-linear correction calculation means for determining the correction amount of the basic injection amount is used, it is possible to perform dynamic correction to compensate for the dynax of fuel delay which exhibits complicated behavior, and it is possible to correct the air-fuel ratio deviation due to the fuel intake delay.

【0037】測定された排気空燃比と目標空燃比との誤
差を用いて重みを逐次補正することは、誤差の時々刻々
の情報をもとに時間的に変化する補正量を出力する補正
モデルの重みを選択することに相当し、現在の機関運転
状態に応じた最適な重みの組み合わせが求められる。従
って、運転状態に応じた燃料補正を行うことができ、空
燃比ずれが補正できる。上記では、各気筒独立噴射につ
いて説明したが、全気筒同時噴射にも適用できる。この
場合には実験により定められた時点での吸気圧力を予測
すればよい。更に、本実施例では回転速度が急激に変化
する内燃機関に本発明を応用した例を示したが、本発明
はこれに限ることなく回転速度がほぼ一定である内燃機
関に応用することもできる。この場合機関回転速度の検
出は不要になることは言うまでもない。
Sequentially correcting the weight using the error between the measured exhaust air-fuel ratio and the target air-fuel ratio is a correction model that outputs a correction amount that changes over time based on momentary information about the error. This corresponds to selecting weights, and the optimum combination of weights is determined according to the current engine operating state. Therefore, the fuel can be corrected according to the operating condition, and the air-fuel ratio deviation can be corrected. Although the above description has been made regarding independent injection in each cylinder, the present invention can also be applied to simultaneous injection in all cylinders. In this case, it is sufficient to predict the intake pressure at a time determined by experiment. Furthermore, although this embodiment shows an example in which the present invention is applied to an internal combustion engine whose rotational speed changes rapidly, the present invention is not limited to this and can also be applied to an internal combustion engine whose rotational speed is approximately constant. . Needless to say, in this case, detection of the engine rotational speed becomes unnecessary.

【0038】[0038]

【発明の効果】以上説明したように本発明によれば、複
雑な挙動を示す燃料遅れに対する燃料噴射量の補正を、
燃料の挙動を考慮した動的な補正要素により行い、過渡
時の空燃比を適切な値に保持することができると共に、
非線形学習手段の出力としての補正量を空燃比のずれ量
の時々刻々の値を用いて修正しているため内燃機関の機
差や経時変化による空燃比ずれを補正することができる
、という効果が得られる。
[Effects of the Invention] As explained above, according to the present invention, the fuel injection amount can be corrected in response to a fuel delay that exhibits complicated behavior.
This is done using a dynamic correction element that takes into account the behavior of the fuel, making it possible to maintain the air-fuel ratio at an appropriate value during transient times, and
Since the correction amount as the output of the nonlinear learning means is corrected using the momentary value of the air-fuel ratio deviation amount, the effect is that it is possible to correct air-fuel ratio deviations due to internal combustion engine machine differences and changes over time. can get.

【図面の簡単な説明】[Brief explanation of the drawing]

【図1】本発明の実施例の空燃比制御装置を機能ブロッ
クで示すブロック図である。
FIG. 1 is a block diagram showing an air-fuel ratio control device according to an embodiment of the present invention using functional blocks.

【図2】燃料噴射量演算ルーチンを示す流れ図である。FIG. 2 is a flowchart showing a fuel injection amount calculation routine.

【図3】予測領域を説明するための線図である。FIG. 3 is a diagram for explaining a prediction area.

【図4】オーバーラップ領域の予測値を求めるための重
みを説明するための線図である。
FIG. 4 is a diagram for explaining weights for determining predicted values of overlapping regions.

【図5】図2の補正量演算ステップの詳細を示す流れ図
である。
FIG. 5 is a flowchart showing details of a correction amount calculation step in FIG. 2;

【図6】ニューラルネットワークを示すブロック図であ
る。
FIG. 6 is a block diagram showing a neural network.

【図7】各素子の入出力を表す関数の線図である。FIG. 7 is a diagram of a function representing input and output of each element.

【図8】重みの学習ルーチンを示す流れ図である。FIG. 8 is a flow chart showing a weight learning routine.

【図9】燃料噴射量の演算タイミングと学習タイミング
とを示すタイミング図である。
FIG. 9 is a timing chart showing calculation timing and learning timing of fuel injection amount.

【図10】本実施例の空燃比の変化と従来の空燃比の変
化とを比較して示す線図である。
FIG. 10 is a diagram showing a comparison between the change in the air-fuel ratio of this embodiment and the change in the conventional air-fuel ratio.

【符号の説明】[Explanation of symbols]

16      吸気圧センサ 18      吸気温センサ 22      燃料噴射弁 28      冷却水温センサ 30      カムポジションセンサ36     
 空燃比センサ
16 Intake pressure sensor 18 Intake temperature sensor 22 Fuel injection valve 28 Cooling water temperature sensor 30 Cam position sensor 36
air fuel ratio sensor

Claims (1)

【特許請求の範囲】[Claims] 【請求項1】  機関回転速度を検出する回転速度検出
手段と、機関の暖気状態量を検出する暖気状態検出手段
と、機関に吸入される吸入空気量に関連した物理量を検
出する物理量検出手段と、機関の排気から排気空燃比を
検出する空燃比検出手段と、前記機関回転速度及び前記
物理量に基づいて基本噴射量を演算する基本噴射量演算
手段と、前記機関回転速度と前記暖気状態量と前記吸入
空気量に関連した物理量と前記基本噴射量と該基本噴射
量の変化率と該変化率の変化率とを入力とし燃料供給遅
れに対応した前記基本噴射量を補正する補正量を各入力
に重みを加えて演算する補正量演算手段と、前記排気空
燃比と目標空燃比との差に応じて前記補正量演算手段の
重みを修正する重み修正手段と、前記基本噴射量と前記
補正量とによって燃料噴射量を求め該燃料噴射量によっ
て空燃比を制御する空燃比制御手段と、から成る内燃機
関の空燃比制御装置。
1. A rotation speed detection means for detecting an engine rotation speed, a warm state detection means for detecting a warm state quantity of the engine, and a physical quantity detection means for detecting a physical quantity related to the amount of intake air taken into the engine. , an air-fuel ratio detection means for detecting an exhaust air-fuel ratio from the exhaust gas of the engine; a basic injection amount calculation means for calculating a basic injection amount based on the engine rotation speed and the physical quantity; A physical quantity related to the intake air amount, the basic injection amount, a rate of change of the basic injection amount, and a rate of change of the rate of change are input, and a correction amount for correcting the basic injection amount corresponding to the fuel supply delay is inputted. correction amount calculation means for calculating by adding a weight to the amount; weight modification means for modifying the weight of the correction amount calculation means according to the difference between the exhaust air-fuel ratio and the target air-fuel ratio; and the basic injection amount and the correction amount. An air-fuel ratio control device for an internal combustion engine, comprising: an air-fuel ratio control means for determining a fuel injection amount by determining a fuel injection amount and controlling an air-fuel ratio based on the fuel injection amount.
JP40909990A 1990-12-28 1990-12-28 Air-fuel ratio controller for internal combustion engine Pending JPH04231647A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP40909990A JPH04231647A (en) 1990-12-28 1990-12-28 Air-fuel ratio controller for internal combustion engine

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP40909990A JPH04231647A (en) 1990-12-28 1990-12-28 Air-fuel ratio controller for internal combustion engine

Publications (1)

Publication Number Publication Date
JPH04231647A true JPH04231647A (en) 1992-08-20

Family

ID=18518475

Family Applications (1)

Application Number Title Priority Date Filing Date
JP40909990A Pending JPH04231647A (en) 1990-12-28 1990-12-28 Air-fuel ratio controller for internal combustion engine

Country Status (1)

Country Link
JP (1) JPH04231647A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0996236A (en) * 1995-10-03 1997-04-08 Matsushita Electric Ind Co Ltd Vehicle control device and air-fuel ratio control device
JP2013072348A (en) * 2011-09-28 2013-04-22 Honda Motor Co Ltd Air-fuel ratio control device of internal combustion engine

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0996236A (en) * 1995-10-03 1997-04-08 Matsushita Electric Ind Co Ltd Vehicle control device and air-fuel ratio control device
JP2013072348A (en) * 2011-09-28 2013-04-22 Honda Motor Co Ltd Air-fuel ratio control device of internal combustion engine

Similar Documents

Publication Publication Date Title
EP1643101B1 (en) Intake air amount control device of internal combustion engine and control device
US5635634A (en) Method for calculating the air charge for an internal combustion engine with variable valve timing
KR100793116B1 (en) Cylinder inflow exhaust gas amount calculation system of internal combustion engine and intake passage inflow exhaust gas amount calculation system of internal combustion engine
US7295912B2 (en) Intake air volume controller of internal combustion engine
US5282449A (en) Method and system for engine control
US5954783A (en) Engine control system using combination of forward model and inverse model
US5427072A (en) Method of and system for computing fuel injection amount for internal combustion engine
JPH04231647A (en) Air-fuel ratio controller for internal combustion engine
JPH08121211A (en) Fuel control device for internal combustion engine
EP1645740B1 (en) Intake airvolume controller of internal combustion engine
US20060136115A1 (en) Control apparatus for internal combustion engine
JP2002309990A (en) Control device for internal combustion engine
JP2986843B2 (en) Air-fuel ratio control device for internal combustion engine
JP2564808B2 (en) Nonlinear feedback control method for internal combustion engine
JP6899416B2 (en) Internal combustion engine control device
US20230323830A1 (en) Method for adjusting a fuel mass to be injected
JP3011008B2 (en) Valve timing control device for internal combustion engine
JPH05231211A (en) Fuel injection control device for internal combustion engine
JPS62195438A (en) Working state detecting device for internal combustion engine
JPS6341634A (en) Air-fuel ratio controller for internal combustion engine
JPH01113546A (en) Intake-air temperature detecting device for internal combustion engine
JP3216299B2 (en) Air-fuel ratio control device for internal combustion engine
JPS61108857A (en) Method for controlling flow rate of exhaust reflux of internal-combustion engine
JPH04228855A (en) Output sensitivity correcting method for combustion pressure sensor
JPH1182077A (en) Intake control system of internal combustion engine