JPS61201846A - Control device for air-fuel ratio of multicylinder engine - Google Patents

Abstract

PURPOSE:To eliminate control of learning of each cylinder, by a method wherein, by means of a correction factor determined from one air-fuel ratio control desired value representing all cylinders, a new control desired value at each cylinder accompanied by a change in an operating condition is calculated. CONSTITUTION:A fluctuation in combustion is detected by combustion fluctuation detecting means 8a-8d located to their respective cylinder of an engine 1, and based on outputs therefrom, the air-fuel ratio of fuel-air mixture fed to each cylinder according to a change in an operating state is controlled to a desired value at each cylinder. In which case, learning value by a learning means 14, a correction factor setting means 15 is provided for comparing, in a given operating range, an average air-fuel ratio control desired value of an arbitrary one cylinder or each cylinder with an air-fuel ratio control desired value of each cylinder and setting a correction factor ki by means of which a new air-fuel ratio control desired value of each cylinder to a learning value is calculated. This enables calculation of a new air-fuel ratio control desired value of each cylinder by means of one learning map.

Description

【発明の詳細な説明】 （産業上の利用分野） 本発明は、多気筒エンジンの空燃比制御装置に関し、特
に各気筒毎の燃焼変動に雄づき各気筒毎に運転状態の変
化に伴って空燃比をυＪｒＩＡするようにしたｂのの改
良に関する。DETAILED DESCRIPTION OF THE INVENTION (Industrial Application Field) The present invention relates to an air-fuel ratio control device for a multi-cylinder engine. This invention relates to an improvement of b by changing the fuel ratio to υJrIA.

（従来の技術） 従来、多気筒エンジンの空燃比制′ＯＪ＋装置として、
特開昭５９−４６３５２号公報に開示されるように、各
気筒毎に燃焼変動（各気筒毎の図示平均有効圧のサイク
ル間＊ｅ及び図示平均有効圧の気筒間較差など）を検出
づる燃焼変動検出手段を設け、該各燃焼変動検出手段の
出力つまりサイクル間変動及び気筒間較差などに基づき
各気筒毎に経年空化や燃焼変化などによる運転状態の変
化に伴って各気筒に供給される混合気の空燃比を目標値
に制御することにより、空燃比を可及的にリーンに設定
して燃費率を低く維持しながら、エンジンラフネスの発
生を精緻に抑制するようにしたものが提案されている。(Prior art) Conventionally, as an air-fuel ratio control 'OJ+ device for a multi-cylinder engine,
As disclosed in Japanese Unexamined Patent Application Publication No. 59-46352, a combustion system that detects combustion fluctuations for each cylinder (inter-cycle *e of indicated mean effective pressure for each cylinder, inter-cylinder range of indicated mean effective pressure, etc.) Fluctuation detection means are provided, and based on the output of each combustion fluctuation detection means, that is, inter-cycle fluctuations, inter-cylinder differences, etc., the fuel is supplied to each cylinder in accordance with changes in operating conditions due to age-related depletion, combustion changes, etc. A system has been proposed that precisely suppresses the occurrence of engine roughness by controlling the air-fuel ratio of the air-fuel mixture to a target value, thereby setting the air-fuel ratio as lean as possible and maintaining a low fuel efficiency rate. ing.

（発明が解決しようとする問題点） しかるに、上記従来の多気筒エンジンの空燃比制御装置
では、運転状態の変化に伴って各気筒毎の新しい空燃比
制御目標値を枠出設定する場合、金気Ｉｌ！５＠に学習
制御しているため、各気筒ｆＱに運転状態の変化に応じ
た学習マツプが必要となり、ＣＰＩＪなどのコントロー
ルユニットにおけるメモリ容量（ＲＡＭ）が非常に大き
なものとなり、実施づる上で不利であった。(Problems to be Solved by the Invention) However, in the above-mentioned conventional air-fuel ratio control device for a multi-cylinder engine, when setting a new air-fuel ratio control target value for each cylinder in response to changes in operating conditions, it is difficult to set a new air-fuel ratio control target value for each cylinder. Mind you! Since learning control is performed at 5@, each cylinder fQ requires a learning map that responds to changes in operating conditions, and the memory capacity (RAM) in control units such as CPIJ becomes extremely large, which is disadvantageous in implementation. Met.

本発明はかかる点に鑑みてなされたものであり、その目
的とするところは、任意の１つの気筒の空燃比制御目標
値もしくは各気筒の空燃比制御目標１＋１１の平均値と
の比較をもとに補正係数を求め、この補正係数により運
転状態の変化に伴う各気筒毎の新しい空燃比ｆｉ制御目
標値を算出することにより、空燃比制御目標値の算出時
間の理れなどを招くことなく、各気筒毎の学習制御を不
要にして＝ｌントロールユニットのメモリ容量を少なく
することにある。The present invention has been made in view of the above, and its purpose is to compare the air-fuel ratio control target value of any one cylinder or the average value of air-fuel ratio control targets 1+11 of each cylinder. By calculating a new air-fuel ratio fi control target value for each cylinder according to changes in operating conditions using this correction coefficient, the calculation time for calculating the air-fuel ratio control target value is not distorted. The purpose is to eliminate the need for learning control for each cylinder, thereby reducing the memory capacity of the control unit.

（問題点を解決するための手段） 上記の目的を達成するため、本発明の解決手段は、第１
図に示すように、各気筒ｆＱに燃焼変動を検出する燃焼
変動検出手段８８〜８ｄを設け、該各燃焼変シｊ検出手
段８ａ〜８ｄの出ツノに塁づいて各気筒毎に運転状態の
変化に応じＣ多気筒に供給される混合気の空燃比を目標
値に制御Ｊ゛るようにした多気筒エンジンの空燃比制御
装置において、任意の１つの気筒の空燃比制御目標値も
しくは各気筒の空燃比制御目標値の平均値を学習プる学
習手段１４を設けるとともに、該学と手段１４による学
習値と各気筒の空燃比制御目標値とを所定運転域で比較
し、学習値に対する各気筒毎の新しい空燃比制御目標値
を粋出させる補正係数ｋｉを設定する補正係数設定手段
１５を設ける構成としたものである。(Means for solving the problem) In order to achieve the above object, the solving means of the present invention is as follows:
As shown in the figure, combustion fluctuation detection means 88 to 8d for detecting combustion fluctuations are provided in each cylinder fQ, and the operating state of each cylinder is determined based on the output horns of the combustion fluctuation detection means 8a to 8d. In an air-fuel ratio control device for a multi-cylinder engine that controls the air-fuel ratio of an air-fuel mixture supplied to multiple cylinders to a target value according to changes in the air-fuel ratio of an arbitrary one cylinder or each cylinder. A learning means 14 for learning the average value of the air-fuel ratio control target values of The configuration is such that a correction coefficient setting means 15 is provided for setting a correction coefficient ki for determining a new air-fuel ratio control target value for each cylinder.

（作用） 上記の構成により、本発明では、各気筒毎に運転状態の
変化に伴って新たな空燃比制御目標値を求める場合、学
習手段で学習した学習値としての任意の１つの気筒の空
燃比制御目標値もしくは各気筒の空燃比制御目標値の平
均値と、各気筒の空燃比制御目標値とを所定運転域で比
較して、金気ＷＩ　ｉｎの補正係数を求めたのち、上記
学習値をこの各気Ｉ！！５毎の補正係数で補正して各気
筒毎の新しい空燃比制御目標値がｑ出されるので、各気
筒ｆｕの学習制御を要さずに、各気筒毎の新たな空燃比
制御目標値が１つの学習マツプでもってかつその陣出時
間に遅れなどな〈従来と同様のスピードでしって得られ
ることになる。(Function) With the above configuration, in the present invention, when determining a new air-fuel ratio control target value for each cylinder in accordance with a change in the operating state, the air-fuel ratio control value of any one cylinder can be used as the learning value learned by the learning means. The fuel ratio control target value or the average value of the air-fuel ratio control target value of each cylinder is compared with the air-fuel ratio control target value of each cylinder in a predetermined operating range to find the correction coefficient for the fuel ratio WI in, and then the above-mentioned learning is performed. Value this value! ! Since the new air-fuel ratio control target value for each cylinder is corrected with a correction coefficient of 5, the new air-fuel ratio control target value for each cylinder can be calculated by With a single learning map, there is no delay in starting time, and you will be able to learn at the same speed as before.

（実施例） 以下、本発明の実施例を第２図以下の図面に基づいて説
明する。(Example) Hereinafter, an example of the present invention will be described based on the drawings from FIG. 2 onwards.

第２図は本発明を燃焼噴射式４気筒エンジンに適用した
実施例を示す。同図において、１は直列に配列された４
つの気筒を右するエンジン、２はよりｌ端がエアクリー
ナ３を介して大気に間口してエンジン１に吸気を供給す
るための吸気通路であって、該吸気通路２の下流側には
吸気拡大室４が設けられ、該吸気拡大室４からは各気筒
に対応して４つの独立した独立吸気通路２ａ〜２ｄに分
岐されて各々対応する各気筒に連通されている。上記各
独立吸気通路２ａ〜２ｄにはそれぞれ燃焼噴射弁５ａ〜
５ｄが配設されており、各燃料噴射弁５ａ・−５ｄは燃
料供給通路６を介して燃焼ポンプ７に接続されていて、
該燃料ポンプ７から圧送される燃料を各燃料ｖＩＳ射弁
５ａ〜５ｄから所定のタイミングで各独立吸気通路２ａ
〜２ｄに噴射供給して、各気筒に所定空燃比の混合気を
供給するようになされている。FIG. 2 shows an embodiment in which the present invention is applied to a combustion injection type four-cylinder engine. In the same figure, 1 is 4 arranged in series.
The engine has two cylinders on the right, and 2 is an intake passage whose l end opens to the atmosphere via an air cleaner 3 to supply intake air to the engine 1, and on the downstream side of the intake passage 2 there is an intake expansion chamber. 4, and the intake expansion chamber 4 is branched into four independent intake passages 2a to 2d corresponding to each cylinder, each of which communicates with the corresponding cylinder. Each of the independent intake passages 2a to 2d has combustion injection valves 5a to 5a, respectively.
5d are arranged, and each fuel injection valve 5a-5d is connected to a combustion pump 7 via a fuel supply passage 6,
The fuel pumped from the fuel pump 7 is transferred from each fuel vIS injection valve 5a to 5d to each independent intake passage 2a at a predetermined timing.
to 2d to supply a mixture at a predetermined air-fuel ratio to each cylinder.

そして、８ａ〜８ｄは各気筒毎にｇ２４ノられ、各気筒
の最大燃焼圧力Ｐ　ｗａｘにより各気筒の燃焼変動を検
出する燃焼変動検出手段としての筒内圧力センサ、９は
吸入空気量を検出する１アフローセンリ”、１０はクラ
ンク角によりエンジン回転数を検出する回転数センナで
あって、これら各センサ８ａ〜８ｄ、９．１０の出力は
上記各燃料噴射弁５ａ〜５ｄを駆動制御するＣＰＵより
なるコントロールユニット１１に入力されている。また
、該コントロール１１にはスロットル弁の開１立信号（
エンジン負荷信号）ＳＯおよび排気ガス還流量を制御す
る還流制御弁の量弁信号ＳＶ等が入力されている。8a to 8d are in-cylinder pressure sensors g24 for each cylinder, which serve as combustion fluctuation detection means for detecting combustion fluctuations in each cylinder based on the maximum combustion pressure Pwax of each cylinder; and 9, a cylinder pressure sensor that detects the amount of intake air. 1A flow sensor, 10 is a rotation speed sensor that detects the engine rotation speed based on the crank angle, and the output of each of these sensors 8a to 8d, 9.10 is made up of a CPU that drives and controls each of the fuel injection valves 5a to 5d. It is input to the control unit 11.The control 11 also has a throttle valve opening signal (
The engine load signal (SO) and the amount valve signal SV of the recirculation control valve that controls the amount of exhaust gas recirculation are input.

上記コントロールユニット１１は、上記各筒内圧カセン
リ８ａ〜８ｄの出力に甘づいて各気筒毎に運転状態の変
化に伴って各気筒に供給される混合気の空燃比がリーン
限界値（エンジンラフネスが生じない範囲での空燃比の
リーン側のｌａ）になるように各気筒毎の目標燃料噴Ｈ
ｆｆｉを判定する判定回路１２と、該判定回路１２の出
力を受け、各燃料噴射弁５８〜５ｄからの燃料噴射量を
制御する制御回路１３とを備えている。また、上記判定
回路１２には後述の学習マツプが具備されている。The control unit 11 adjusts the air-fuel ratio of the air-fuel mixture supplied to each cylinder to a lean limit value (engine roughness The target fuel injection H for each cylinder is set so that the air-fuel ratio is on the lean side (la) within the range where this does not occur.
It includes a determination circuit 12 that determines ffi, and a control circuit 13 that receives the output of the determination circuit 12 and controls the amount of fuel injected from each of the fuel injection valves 58 to 5d. Further, the determination circuit 12 is equipped with a learning map, which will be described later.

次に、上記コントロールユニット１１の判定回路１２の
作動を第３図に示すフローチャートにより説明する。先
ず、スデップＳＳ’？’空燃比のり−・ン限界移行に適
した運転状態を判断するべく定常運転状態であるか否か
を判別し、定常運転状態であるＹＥＳの場合には、ステ
ップＳ２において各気筒毎の最大燃焼圧力Ｐ　ｗａｘの
サイクル間変動σ１を下記式 ％式％） （ここでｎ：サイクル数（例えば１００サイクル）、ｐ
ｎ＋ａｘ：ｎサイクルでの最大燃焼圧力の平均値） Ｊ、り計算したのら、ステップＳ３でこのサイクル間変
動σ蟇が許容値σ０以下であるか否かを判別で−る。こ
の判別がσｉ≦σ、）のときには、サイクル間変動σｉ
が小さくて空燃比のリーン化がさらに可能ぐあると判断
して、ステップＳ４でその気筒に対する燃料の補正噴躬
邑ＱｏｉをΔＱだけ減損した１Ｉｆｌ−（Ｑｏｉ−ΔＱ
）に更新する一方、上記判別がσ１〉ＱＯのときには、
サイクル間変動σ；が大きくで空燃比がリーン限界に達
していると判断して、ステップＳｓでモの気筒に対する
補正噴射ｆｆ１ＱｏｉヲΔＱりＧｔ増ｆｆｉシ／；１１
　（Ｑｏｉ＋ＡＱ　）　ニ更新する。Next, the operation of the determination circuit 12 of the control unit 11 will be explained with reference to the flowchart shown in FIG. First of all, Sudepp SS'? 'In order to determine the operating state suitable for shifting to the air-fuel ratio limit, it is determined whether or not the operating state is a steady operating state. If YES in the steady operating state, the maximum combustion for each cylinder is determined in step S2. The cycle-to-cycle fluctuation σ1 of the pressure P wax is calculated using the following formula (% formula %) (where n: number of cycles (for example, 100 cycles), p
n+ax: average value of maximum combustion pressure over n cycles) After calculating J, it is determined in step S3 whether or not this inter-cycle fluctuation σ is less than the allowable value σ0. When this determination is σi≦σ,), the inter-cycle variation σi
is small, and it is determined that it is possible to make the air-fuel ratio leaner, and in step S4, the correction fuel injection Qoi for that cylinder is reduced by ΔQ, 1Ifl-(Qoi-ΔQ
), while when the above determination is σ1>QO,
It is determined that the inter-cycle fluctuation σ is large and the air-fuel ratio has reached the lean limit, and in step Ss, the correction injection for the cylinder is increased by ΔQ and Gt is increased.
(Qoi+AQ) D Update.

次いで、各気筒の最大燃焼圧力Ｐ■ａｘ　ｉの気筒間較
差を求めるべ（、ステップＳ６で各気筒の最大燃焼圧力
の平均１ｎＰを下記の４気筒の場合の式％式％ よりＢ１算したのら、ステップＳ７で気ｖＲ１１１１差
（Ｐ−Ｐｍａｘｉ）が許容値ΔＰ以下であるか否かを判
別する。この判別が（Ｐ−Ｐｍａｘｉ）≦ΔＰのときに
は気［ｆｌ較差が小さいと判断してそのままステップＳ
　ｎに進む一方、（Ｐ−Ｐｓ＋ａｘ　ｉ　）　＞ΔＰの
とぎには気筒間較差が大きいと判断して燃料を増量すべ
く、ステップＳ８で補正噴射１１ＱｏｉをΔＱ′だけ増
量したｈｉ１ＣＱＯｉ＋ΔＱ’　　）に更新したのちス
テップＳ　ｕに移る。尚、上記ステップＳ＋の判別が定
常運転状態でないＮＯのときには、直ちにステップＳ９
に移って非定常判定凹路により加速時か減速時、あるい
は冷間時等を判定し、ステップＳ　１６でそのときの括
木噴ｔｇｍａｏを決定してステップＳ　ｎに進む。Next, calculate the inter-cylinder difference in the maximum combustion pressure Pax i of each cylinder (In step S6, calculate the average 1nP of the maximum combustion pressure of each cylinder by B1 from the following formula % formula % for 4 cylinders. Then, in step S7, it is determined whether or not the difference (P-Pmaxi) is less than or equal to the allowable value ΔP.If this determination is (P-Pmaxi)≦ΔP, it is determined that the difference in [fl] is small and the process is continued. Step S
While proceeding to step n, when (P-Ps+ax i )>ΔP, it was determined that the difference between the cylinders was large, and in order to increase the amount of fuel, the correction injection 11Qoi was updated to hi1CQOi+ΔQ' which was increased by ΔQ' in step S8. Afterwards, the process moves to step S u. Incidentally, if the determination in step S+ is NO that the steady operating state is not established, the process immediately proceeds to step S9.
Then, it is determined whether the engine is accelerating, decelerating, or cold by using the unsteady determination concave path, and in step S16, the exhaust gas tgmao at that time is determined, and the process proceeds to step Sn.

そして、ステップＳ　ｎで各気筒毎に基本噴ＤＩ　ｍＱ
ｏに上記の補正噴射１Ｑｏｉを加算して各気筒毎の燃料
噴射ＩＱｉを求め、ステップＳ　１２でこの信号をＩｌ
ｌ　１１回路１３に出力する。Then, in step Sn, the basic injection DI mQ is set for each cylinder.
The above correction injection 1Qoi is added to o to obtain the fuel injection IQi for each cylinder, and in step S12, this signal is
l11 Output to circuit 13.

これに対し、運転状態の変化に伴って学習制御する場合
には、上記の各気筒の補正噴射ｆｉＱｏｉをちとにして
、ステップＳ　１３で各気筒の補正噴ｔＪＪｆｆｉの平
均値Ｑｏｉを計算したのちステップＳＨに移り、このス
テップＳ　１４の判別が定常運転状態でないＮＯのとき
に、ステップＳ　＋ｓにおいて躊本噴射ｍＱＯをＱｏｉ
だけ増量した値（Ｑｏ　＋Ｑｏｉ）に置き換えて、それ
をステップＳ　＋ｓで学習マツプに入力する１、また、
上記の補正噴射ｔｉｔの平均値Ｑｏｉの計棹後、ステッ
プＳ　＋ｙで補正係数ｋｉを設定すべき運転条イ′１を
満たしているか否か、例えば中負荷域でエンジン回転数
が１０００〜３０００　ｒｐｍでかつ吸気負圧が−４０
０〜−２００１１１１１−１（＋の運転域にあるか否か
を判別し、この判別がＹＥＳのときにのみステップＳ　
＋ａに移って、各気筒の燃料噴射Ｆａ（Ｑｏ　＋Ｑｏｉ
）とその各気筒の平均値（ＱＯ＋Ｑｏｉ）とを比較して
、その比から補正係数ｋｉ＝（Ｑｏ　＋　Ｑｏｉ）　／
　（ＱＯ−トＱｏｉ）を求め、それをステップＳ　１６
の学習マツプに入力する。尚、ステップＳ　１９で固定
マツプ（ＲＯＭ）により塞本噴躬ＭＱＯの初期値が、ま
た補正係数ｋｉの初期＊　（ｋｉ−１）がそれぞれセッ
トされており、ステップＳ２＠のエンジン運転開始のＹ
ＥＳの判別と同時に、これらの初期条件がステップＳ　
＋６の学習マツプに入力されている。On the other hand, when learning control is performed in accordance with changes in the operating state, the corrected injection fiQoi of each cylinder is calculated in step S13, and the average value Qoi of the corrected injection tJJffi of each cylinder is calculated in step S13. The process moves to SH, and when the determination in step S14 is NO that is not a steady operation state, the main injection mQO is changed to Qoi in step S+s.
1. Replace it with the value increased by (Qo + Qoi) and input it to the learning map in step S + s.
After calculating the average value Qoi of the correction injection tit mentioned above, in step S +y, it is determined whether or not the operating condition A'1 for setting the correction coefficient ki is satisfied, for example, when the engine speed is 1000 to 3000 rpm in the medium load range. And the intake negative pressure is -40
0 to -20011111-1 (determine whether or not it is in the + operating range, and only when this determination is YES, step S
Moving on to +a, each cylinder's fuel injection Fa (Qo +Qoi
) and the average value (QO + Qoi) of each cylinder, and from the ratio, the correction coefficient ki = (Qo + Qoi) /
(QO-tQoi) is obtained and it is processed in step S16.
input into the learning map. In addition, in step S19, the initial value of Ishimoto injection MQO and the initial value of the correction coefficient ki * (ki-1) are set in the fixed map (ROM), and Y at the start of engine operation in step S2@ is set.
At the same time as determining the ES, these initial conditions are
+6 has been input into the learning map.

そして、ステップＳ　１６の学習マツプに基づいて、ス
テップＳ２＋において各気筒毎の新しい基本噴射量ＱＯ
として上記ステップＳ　＋ｓで学習した基本噴射ｍＱｏ
に上記ステップＳ　＋ｓで算出した各気ｌ１ｇ１毎の補
正係数ｋｉを乗葬した値（Ｑｏ　・ｋｉ）に更新して、
ステップＳ　＋＋に戻ることを繰返す。Then, based on the learning map in step S16, a new basic injection amount QO for each cylinder is determined in step S2+.
The basic injection mQo learned in step S+s above is
is updated to a value (Qo ・ki) multiplied by the correction coefficient ki for each qi l1g1 calculated in step S+s above,
Repeat returning to step S++.

以Ｆの作動フローにおいて、ステップＳ　＋３〜Ｓ１５
により、それ以前のステップ８２〜Ｓ８で求められた各
気筒毎の補正噴射ｆｆ１Ｑｏｉから各気筒の補正噴ｔＡ
量の平均＋６０Ｏｉを求め、この伯から各気筒の空燃比
制御目標値の平均値に相当する。各気筒の燃料噴射量の
平均値（Ｑｏ　＋Ｑｏｉ）を学習するようにした学習手
段１４を構成している。また、ステップＳａｙ、Ｓ’ｓ
により、所定運転域において各気筒の燃料噴射量の平均
＊　（ＱＯ＋ｏｏｔ＞と各気筒の燃料噴ｔＪ４ｆｋ　（
Ｑｏ　＋Ｑｏｉ）とを比較することにより、つまり上記
学習手段１４の学習値と各気筒の空燃比制御目標値とを
比較することにより、学習値に対する各気筒毎の新しい
空燃比制御目標値を算出さゼる補正係数ｋｉを設定する
ようにした補正係数設定手段１５を構成している。In the following operation flow, steps S+3 to S15
Therefore, the correction injection tA for each cylinder is calculated from the correction injection ff1Qoi for each cylinder obtained in the previous steps 82 to S8.
The average amount +60Oi is calculated, and this ratio corresponds to the average value of the air-fuel ratio control target value for each cylinder. A learning means 14 is configured to learn the average value (Qo + Qoi) of the fuel injection amount of each cylinder. Also, step Say, S's
As a result, the average fuel injection amount of each cylinder * (QO+oot>) and the fuel injection amount of each cylinder tJ4fk (
By comparing the learned value of the learning means 14 and the air-fuel ratio control target value of each cylinder, a new air-fuel ratio control target value for each cylinder is calculated for the learned value. This constitutes a correction coefficient setting means 15 configured to set a correction coefficient ki for zero.

したがって、上記実施例においては、運転状態の変化に
伴う各気筒毎の新しい空燃比制御目標（直を設定Ｊる場
合、学習手段１４により学習した各気筒の空燃比制御目
標値の平均＋ｔ１（各気筒の目標燃料噴か１聞の平均値
）に対する補正係数ｋｉを求めて、この各補正係数ｋｉ
により各気筒毎の新しい空燃比Ｈｉｌ＋御目標値（各気
筒毎の新しい目標燃料鳴制吊）を算出するようにしたの
で、従来の如き各気筒毎の学習制御が不要となり、また
各気筒毎の空燃比制御目標値の算出時間に遅れなどな〈
従来と同様のスピードで算出されることになり、よって
ＣＰＵのメモリ容ｆｆｉ　（ＲＡＭ）を能力低下なく有
効に少なくすることができ、実用上有利なものとなる。Therefore, in the above embodiment, if a new air-fuel ratio control target (direct) is set for each cylinder due to a change in the operating state, the average of the air-fuel ratio control target values for each cylinder learned by the learning means 14 + t1 (each Find the correction coefficient ki for the target fuel injection of the cylinder (average value for one cycle), and calculate each correction coefficient ki.
Since the new air-fuel ratio Hil + control target value (new target fuel noise control value for each cylinder) is calculated for each cylinder, the conventional learning control for each cylinder is no longer necessary, and the There is no delay in calculating the air-fuel ratio control target value.
Since the calculation is performed at the same speed as in the past, the memory capacity ffi (RAM) of the CPU can be effectively reduced without deteriorating the performance, which is advantageous in practice.

尚、本発明は上記実施例に限定されるものではなく、そ
の他種々の変形例をも包含するものである。例えば、上
記実施例では燃料噴ｔＪＪｆｆｉの制御により空燃比制
御を行う場合について述べたが、吸入空気聞の制御によ
り空燃比ｔｉ制御を行う場合についても同様に適用でき
るものである。It should be noted that the present invention is not limited to the above-mentioned embodiments, but also includes various other modifications. For example, in the above embodiment, a case has been described in which the air-fuel ratio is controlled by controlling the fuel injection tJJffi, but the present invention can be similarly applied to a case in which the air-fuel ratio ti is controlled by controlling the intake air ratio.

また、上記実施例では学習手段１４の学習値として各気
筒の空燃比１ｔＩＩＪ御目標値の平均値を用いたが、そ
れに代わって任意の１つの気筒の空燃比制御目標値を学
習値とし、この学習値に対する各気筒毎の補正係数を設
定するようにしてもよく、上記実施例と同様の作用効果
を秦することができる。Further, in the above embodiment, the average value of the air-fuel ratio control target value of each cylinder is used as the learning value of the learning means 14, but instead, the air-fuel ratio control target value of any one cylinder is used as the learning value. A correction coefficient may be set for each cylinder with respect to the learning value, and the same effects as in the above embodiment can be achieved.

（発明の効果） 以上説明したように、本発明の多気筒エンジンの空燃比
制御装置によれば、運転状態の変化に伴う各気筒毎の新
たな空燃比制御目標値を、学習した任意の１つの気筒の
空燃比制御目標値もしくは各気筒の空燃比制御目標値、
の平均値と各気筒の空燃比制御目標値とを比較して求め
た各気筒毎の補正係数をもとに算出するようにしたので
、従来の如き各気筒毎の学問制御が不要で、１つの字型
マツプでもってかつその算出時間に遅れなどな〈従来と
同様のスピードでもって求めることができ、よってコン
トロールユニット（ＣＰＬＪ）のメモリ容ｆｆｌ　（Ｒ
ＡＭ）を有効に少な（することができ、実用上右利であ
る。(Effects of the Invention) As explained above, according to the air-fuel ratio control device for a multi-cylinder engine of the present invention, a new air-fuel ratio control target value for each cylinder due to a change in the operating state can be set using a learned arbitrary one. air-fuel ratio control target value for one cylinder or air-fuel ratio control target value for each cylinder,
Since the calculation is based on the correction coefficient for each cylinder obtained by comparing the average value of It has a square-shaped map and there is no delay in its calculation time (it can be calculated at the same speed as before, and therefore the memory capacity of the control unit (CPLJ) ffl (R
AM) can be effectively reduced (AM), which is of practical benefit.

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

第１図は本発明の構成を示すブロック図であφ。 第２図および第３図は本発明の実施例を示し、第２図は
全体概略構成図、第３図はコントロールユニットの判定
回路の作動フローを示すフローチャート図である。 １・・・エンジン、５８〜５ｄ・・・燃料噴射弁、８ａ
〜８ｄ・・・筒内圧力センサ、１１・・・コントロール
ユニット、１２・・・判定回路、１３・・・制御回路、
１４・・・学習手段、１５・・・補正係数設定手段。 ｊｊｕｊ：ｊ、’−，ｊ−夕 第１図 第２図FIG. 1 is a block diagram showing the configuration of the present invention. 2 and 3 show an embodiment of the present invention, FIG. 2 is a general schematic diagram, and FIG. 3 is a flowchart showing the operation flow of the determination circuit of the control unit. 1...Engine, 58-5d...Fuel injection valve, 8a
~8d... Cylinder pressure sensor, 11... Control unit, 12... Judgment circuit, 13... Control circuit,
14...Learning means, 15...Correction coefficient setting means. jjuj:j,'-,j-Evening Figure 1 Figure 2

Claims (1)

【特許請求の範囲】[Claims] （１）各気筒毎に燃焼変動を検出する燃焼変動検出手段
を設け、該各燃焼変動検出手段の出力に基づいて各気筒
毎に運転状態の変化に伴つて各気筒に供給される混合気
の空燃比を目標値に制御するようにした多気筒エンジン
の空燃比制御装置において、任意の１つの気筒の空燃比
制御目標値もしくは各気筒の空燃比制御目標値の平均値
を学習する学習手段と、該学習手段による学習値と各気
筒の空燃比制御目標値とを所定運転域で比較し、学習値
に対する各気筒毎の新しい空燃比制御目標値を算出させ
る補正係数を設定する補正係数設定手段とを設けたこと
を特徴とする多気筒エンジンの空燃比制御装置。(1) Combustion fluctuation detection means for detecting combustion fluctuations is provided for each cylinder, and based on the output of each combustion fluctuation detection means, the air-fuel mixture supplied to each cylinder is In an air-fuel ratio control device for a multi-cylinder engine configured to control an air-fuel ratio to a target value, a learning means for learning an air-fuel ratio control target value for any one cylinder or an average value of air-fuel ratio control target values for each cylinder; , correction coefficient setting means for comparing the learned value by the learning means and the air-fuel ratio control target value of each cylinder in a predetermined operating range, and setting a correction coefficient for calculating a new air-fuel ratio control target value for each cylinder with respect to the learned value; An air-fuel ratio control device for a multi-cylinder engine, characterized by comprising: