TWI409385B - A reciprocating engine swivel number calculating device and a reciprocating engine control device - Google Patents

Abstract

Provided is an apparatus for calculating the number of revolutions of a reciprocating engine, wherein a plurality of signals are generated per one crank rotation and the actual number of revolutions of the reciprocating engine is calculated on the basis of the plurality of signals. A fluctuation component removal means is provided to remove explosion-dependent fluctuation components of the reciprocating engine contained in the plurality of signals in accordance with a crank phase angle, the number of revolutions, and the amount of fuel injected, when the actual number of revolutions is calculated.

Description

Reciprocating engine revolution number calculation device and reciprocating engine control device

The present invention relates to an apparatus for detecting the number of engine revolutions of a reciprocating engine.

In general, the number of engine revolutions is detected based on a pulse signal generated in synchronization with the rotation of the crankshaft. A reciprocating engine that operates at a high number of revolutions of several thousand RPM or more, even if the pulse signal detected from each revolution of the crankshaft is used to determine the number of engine revolutions, since the sampling interval is very short for the control reaction, it does not occur in reactivity. problem. In addition, a reciprocating engine that operates at a low number of revolutions of several hundred RPM or less has a number of pulses occurring in one revolution of the crankshaft because one pulse of the pulse is delayed in the control reaction in one revolution. Need to calculate the number of engine revolutions.

However, since the reciprocating engine converts the reciprocating motion of the piston into the rotational motion of the crankshaft, the moment of inertia associated with the cranking is varied by a period of roughly twice the number of revolutions. In addition, since the reciprocating motion of the piston includes a compression and an explosion stroke, there is also a variation corresponding to the period. Therefore, when a plurality of pulse signals are generated in one rotation of the crankshaft, and the number of revolutions is calculated based on this, the calculated number of revolutions includes such periodic pulsations. Therefore, when the target number of revolutions is set and the actual number of revolutions of the crankshaft is feedback-controlled, an unnecessary command is output due to the influence of the pulsation, and the fuel supply is unstable, and the controllability and operability are deteriorated. In response to such a problem, a structure has been proposed in which the number of revolutions is sampled and maintained by the explosion period of the engine (Patent Document 1).

[Previous Technical Literature] [Patent Literature]

[Patent Document 1] Japanese Special Fair 3-24581

However, as described in Patent Document 1, when the number of revolutions is sampled and held by the pulsation cycle of the reciprocating engine, there is a problem that the reactivity is impaired in the sudden load fluctuation, and the control performance cannot be achieved. Further, although the method of calculating the number of revolutions using the moving average to remove the above-described pulsation is also considered, the problem of the reaction lag still occurs. It is an object of the present invention to accurately calculate the number of revolutions that remove the pulsating effects of the reciprocating engine.

The reciprocating engine of the reciprocating engine of the present invention generates a plurality of signals in one rotation of the crank, and calculates the actual number of revolutions of the reciprocating engine based on the plurality of signals, and is characterized by having a variation component removing means, which is calculated At the actual number of revolutions, the explosion fluctuation component of the reciprocating engine included in the plurality of signals is removed in accordance with the crank phase angle, the number of revolutions, and the fuel injection amount.

The plurality of signals are generated from each fixed crank angle, and the explosion fluctuation component is removed by calculating a correction coefficient corresponding to the crank phase angle in the angular velocity calculated based on the plurality of signals. Further, it is preferable that the fluctuation component removing means removes the explosion fluctuation component from the calculated number of revolutions in accordance with the crank phase angle. The correction coefficient is set to remove the value of the inertia fluctuation component in addition to the explosion variation component. The angular velocity is estimated from the phase angle, the number of revolutions, and the fuel injection amount, and the correction coefficient is obtained based on the estimated angular velocity. At this time, the correction coefficient should be the reciprocal of the normalized estimated angular velocity of the estimated angular velocity so that the average value becomes 1. Further, the number of revolutions for estimating the angular velocity is obtained by, for example, using a moving average value of the angular velocity calculated based on a plurality of signals or performing any one of the values of the first-order lag filtering.

Further, the revolution number calculation device of the present invention generates a plurality of signals in one rotation of the crank, and calculates the actual number of revolutions of the reciprocating engine based on the plurality of signals, and is characterized in that the variable component removal means is provided, and the actual rotation is calculated. In the case of a number, the inertia fluctuation component of the reciprocating engine included in the plurality of signals is removed in accordance with the crank phase angle.

Further, when the time required for the crank to rotate once is constant, the plurality of signals may be formed to be generated at regular time intervals in accordance with the inertia fluctuation. At this time, the plurality of signals are generated by the sensor detecting a plurality of detected portions provided along the circumferential direction of the rotating body that is integrally rotated with the crank. When the time required for one rotation of the crank is constant, the signals are generated at regular intervals, and the detected portions are arranged at intervals that are not equal intervals in accordance with the inertia fluctuation.

The reciprocating engine control device of the present invention is characterized in that it has any one of the above-described revolution number calculating devices.

The ship of the present invention is characterized by comprising the above-described reciprocating engine control device.

In addition, the method for calculating the number of revolutions of the reciprocating engine of the present invention generates a plurality of signals in one rotation of the crank, and calculates the actual number of revolutions of the reciprocating engine based on the plurality of signals, which is characterized by calculating the actual number of revolutions. The explosion fluctuation component of the reciprocating engine included in the plurality of signals is removed in accordance with the crank phase angle, the number of revolutions, and the fuel injection amount.

With the present invention, the number of revolutions that remove the pulsation effect of the reciprocating engine can be accurately calculated.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

The first figure is a block diagram showing the structure of a control system of a low speed reciprocating engine according to a first embodiment of the present invention. In addition, in the first figure, only one cylinder is shown, but in general, a plurality of cylinders (for example, ten cylinders or less), the details of the number of cylinders and the number of common revolutions will be described later.

The control system 10 adjusts the fuel supply amount to the reciprocating engine 11 and controls the speed thereof to perform PID control in such a manner as to maintain the set target number of revolutions. The reciprocating engine 11 is a conventional diesel engine, and the crankshaft 12 is coupled to the piston 17 via a crank 13, a link 14, a crosshead 15, and a piston rod 16, and the piston 17 is supplied to the combustion chamber 19 from the fuel injection nozzle 18. The explosion of the fuel reciprocates within the sleeve 20 and imparts a rotational force to the crankshaft 12. Further, when the exhaust valve 21 is opened, the combustion gas is discharged from the exhaust port 22, and new air is supplied from the ventilating port 23.

A flywheel 24 is provided on the crankshaft 12, and a tooth portion 25 in which teeth are formed at a predetermined pitch is provided at a peripheral portion of the flywheel 24. Further, the crank angle sensor 26 is disposed at a position close to the tooth portion 25. The crank angle sensor 26 is provided with a non-contact switch, an encoder, or the like, and generates a pulse signal, for example, when each tooth of the tooth portion 25 passes. Further, the tooth portion 25 and the crank angle sensor 26 may be provided with a mechanism for detecting a reference position during rotation. For example, when using a non-contact switch, it is considered to set the width of one tooth to be larger than the other teeth. Further, when the encoder is used, in addition to the tooth portion 25, one projection may be provided on the peripheral portion of the flywheel 24, and a mechanism for detecting the projection may be provided.

The pulse signal generated by the crank angle sensor 26 is input to the number-of-turns calculation unit 28 provided in the control unit 27. The number-of-turns calculation unit 28 calculates the angular velocity and the phase angle of the crankshaft 12 from the input pulse signal, and calculates the number of revolutions of the crankshaft 12 by the method described later, and outputs it to the operation amount calculation unit 29. The operation amount calculation unit 29 calculates the operation amount (fuel supply amount) of the fuel pump 30 based on the number of revolutions input, and the fuel pump 30 drives the pump in accordance with the operation amount, supplies it to the fuel injection nozzle 18, and ejects it to the combustion chamber 19 at a specified timing. fuel.

Further, the operation amount calculated by the operation amount calculation unit 29 is also transmitted to the number-of-turns calculation unit 28. In other words, the number-of-turns calculation unit 28 executes the calculation of the number of revolutions in the present embodiment based on the angular velocity, the phase angle, the number of revolutions, the amount of operation (fuel supply amount), and the data recorded in the memory unit 31 corresponding to the values. Processing (details are described later). In the first example, the amount of operation is fed back to the number-of-turns calculation unit 28 as the amount of fuel supply. However, in order to feed back a more accurate fuel supply amount, a sensor may be provided in the fuel pump 30 or the like. The amount of fuel supplied, and the measured fuel supply is fed back.

Next, the conditions (number of cylinders and number of revolutions) required to perform the calculation of the number of revolutions in the present embodiment will be briefly described.

The angular velocity variation of the reciprocating engine 11 when the crankshaft 12 is rotated, as explained in the prior art, is due to the variation of the moment of inertia (hereinafter referred to as inertial variation) due to the reciprocating motion of the piston 17 or the like, and the compression due to the combustion cycle. (Deceleration), explosion (acceleration), the change in the pressure in the cylinder (hereinafter referred to as the explosion change). Since the influence of the inertia variation is smaller than that of the explosion, the following is explained based on the conditions of the explosion.

Note that for one cylinder, the explosion of the two-stroke engine produces a maximum value change in one revolution (360°), and the explosion variation of the four-stroke engine produces one-time maximum in two revolutions (720°). change. Therefore, the two-stroke engine of the n-cylinder varies every 360/n (degrees), and the four-stroke engine varies every 720/n (degrees).

Usually, the angular velocity change caused by the explosion variation in one cylinder is mainly generated in a specified angular range (angular velocity variation region) before and after the top dead center. The angular width Φ of the angular velocity variation region varies depending on the deflection of the crankshaft, the axial distance between the large end and the small end of the connecting rod, the compression ratio, and the like, but is usually about 120° 60° before and after the top dead center. When the number of cylinders is increased, the angular velocity fluctuation region between the front and rear cylinders overlaps, and the fluctuation in each cylinder is eliminated and suppressed. When the degree of influence of the explosion variation inversely proportional to the degree of overlap of the angular velocity variation region Φ is α, the influence degree α can be expressed as the offset angle of the explosion timing between the cylinders (two strokes: 360/n, four strokes: 720) /n) divided by the angular velocity variability angle Φ. That is, α=360/n/Φ in the two-stroke engine and α=720/n/Φ in the four-stroke engine.

Empirically, by suppressing the explosion fluctuation component by multi-cylinderization, it is high when α < 0.33, and the revolution number calculation processing of this embodiment is not required. However, when α ≧ 0.33, the influence of the explosion fluctuation component occurs at the angular velocity, and the revolution number calculation processing of the present embodiment is effective. In particular, when α ≧ 0.35, the revolution number calculation processing of the present embodiment is effective.

For example, when Φ = 120°, the two-stroke engine is α ≧ 0.33 for 9 cylinders or less, and α ≧ 0.33 for 18 cylinders or less, and the revolution number calculation processing of the present embodiment is effective.

In addition, in the case of digital control, the sampling interval Ts in the measurement needs to be sufficiently smaller than the interval Ta of the A/D conversion (in this case, the interval of the D/A conversion is also the same), and it is usually required to be Ta>5‧Ts and should be Ta. >10‧Ts. Further, the interval A of the A/D (D/A) conversion is closely related to the reaction lag of the number of revolutions of the controlled amount and the reaction lag of the fuel injection amount (fuel supply amount) of the operation amount, and the interval Ta needs to be delayed. For a short time interval. When the hysteresis of the controlled quantity and the manipulated quantity is represented by the time constants Td and Tm of the first-order lag, it must be Td>10‧Ta and Tm>Ta, and should be Td>50‧Ta and Tm>5‧Ta.

Therefore, for digital control, the sampling interval Ts must satisfy Td>50‧Ts and Tm>5‧Ts, and should satisfy Td>500‧Ts and Tm>50‧Ts.

In addition, when the sampling interval (pulse interval) Ts coincides with the fluctuation period of the controlled amount (angular velocity), or if it is larger than the filter, the same effect as the filter is performed, and no change occurs in the measurement result of the controlled amount. . At this time, the interval between explosions in the n-cylinder two-stroke engine is every 360/n (degrees), and in the four-stroke engine every 720/n degrees. In addition, when the number of revolutions is R (rpm), since the crankshaft performs 360‧R/60 (degrees) revolution per second, the angular velocity variation period caused by the explosion is 60/n/R (seconds) in the two-stroke engine, four 120/n/R (seconds) in the trip engine.

In other words, in the two-stroke engine, when the value of the sampling interval Ts obtained by the digital control is 60/n/R (second) or less, the calculation of the number of revolutions of the present embodiment is effective, and the four-stroke engine is at 120/n/R. (seconds) is valid when it is below. In other words, the number of common revolutions R of the reciprocating engine to be applied to the revolution number calculation processing of the present embodiment is 60/n/Ts (rpm) or less in the two-stroke engine, and 120/n/ in the four-stroke engine. Ts (rpm) or less.

For example, in a two-stroke six-cylinder engine, when Td=10 seconds and Tm=1 seconds, the digital control system is constrained to Ts<1/5, and should be Ts<1/50. At this time, Ts≒1/5 R is about 50 rpm or less, and when Ts ≒ 1/50 is 500 rpm or less.

Next, a specific example of the revolution number calculation processing of the present embodiment will be described with reference to the second to seventh figures. The second graph shows a time-series variation of the angular velocity variation of the crankshaft at a certain number of revolutions in a seven-cylinder engine, with the horizontal axis corresponding to time and the vertical axis corresponding to normalized angular velocity. Further, the angular velocity is normalized so that the average value thereof is 1.

As shown in the second figure, even if the number of revolutions of the engine is constant and the average value of the angular velocity is constant, the angular velocity is subject to fluctuations in inertia and explosion fluctuations in a certain period. In the past, since the angular velocity of the engine revolution is multiplied by the specified coefficient, as shown in the second figure, when the angular velocity is pulsated by the influence of inertia fluctuation and explosion fluctuation, the number of engine revolutions that should be fixed is calculated as The angular velocity changes with the same period and changes.

In general, the fluctuation of the angular velocity ω due to the inertia fluctuation and the explosion fluctuation is basically determined only by the phase angle θ of the crank, the number of revolutions N, and the fuel supply amount Q. Therefore, the angular velocity including the fluctuation due to the inertia fluctuation and the explosion fluctuation is estimated from the phase angle θ of the crank, the number of revolutions N, and the fuel supply amount Q, and when the angular velocity ω of the fluctuation component is actually measured, the angular velocity ω can be obtained. The number of revolutions of the pulsation caused by the inertia fluctuation and the explosion fluctuation is roughly removed. Further, at this time, the number of revolutions N used for estimating the angular velocity is, for example, a moving average of the measured angular velocity ω or a smoothed value of the first-order lag filtering, or a value obtained by the previous process. Further, the fuel supply amount Q uses, for example, the value of the last fuel supply amount.

Further, the time constant for smoothing by the first-order lag filter is preferably set to a value that is sufficiently shorter than the time constant (tens of seconds) of the reaction of the number of crank revolutions (for example, 10% or less of the reaction time constant of the number of crank revolutions). ). However, the time constant of the first-order lag filter can also be simply set to a time constant of about 2 seconds.

When the estimated value of the angular velocity in the number of revolutions N and the fuel supply amount Q is Ω(θ; N, Q), the average value is Ω _m (= 2πN), and the measured angular velocity ω (θ) in the phase θ In the phase θ, the average value of the angular velocity ω _m (θ) showing the value of ω(θ) in the phase θ is expressed as ω _m (θ)=ω(θ)‧Ω _m (N)/ Ω(θ; N, Q) (due to ω: ω _m = Ω: Ω _m ). Therefore, when f(θ)=Ω _m (N)/Ω(θ; N, Q) is obtained, the effective revolution number Ne(θ) of the pulsation component is roughly obtained for the measured angular velocity ω(θ) (hereinafter referred to as The number of effective revolutions is Ne(θ)=ω _m (θ)/(2π)=ω(θ)‧f(θ)/(2π)[rad/sec].

The third figure covers the crank angle phase 360° section showing the normalized angular velocity Ω(θ) of a certain number of revolutions N and the fuel supply amount Q, which is included in the seven-cylinder two-stroke engine including the inertia variation and the pulsation caused by the explosion fluctuation. Ωm change and its reciprocal f(θ) change graph. Further, the horizontal axis is defined by the TDC of a certain cylinder as the center (0), and the phase angle (180°) between the TDC and the BDC is set to 1. In addition, in the third figure, the normalized angular velocity Ω(θ)/Ωm is indicated by a solid line, and the reciprocal is indicated by a broken line. In the present embodiment, the correction coefficient f(θ) at the time of calculation of the engine revolution number is shown.

The seven-cylinder two-stroke engine has seven periodic angular velocity changes in the crank for one revolution, and the correction coefficient (reciprocal) changes in phase opposite to the angular velocity. In the example of the third figure, the correction factor changes in a form close to a sine wave.

In the example of the third figure, when the correction coefficient (reciprocal) f(θ) is approximated by a sine wave, the value f(θ) of the correction coefficient in the phase angle θ of the crank is expressed by the following formula.

f(θ)～A‧sin(B‧θ+C)+D

Among them, B is the coefficient determined by the number of cylinders and two strokes or four strokes. The example in the third figure is B=7. The amplitude A and the phase difference C are determined according to simulation or experiment, and the amplitude A mainly depends on the fuel supply amount (operation amount) Q and the number of revolutions N. The amplitude A is obtained, for example, by A=g(Q)‧h(N) which is a product of the functions g(Q) and h(N) obtained in advance. Further, a lookup table for the values of each of Q and N may be stored in the memory unit 31 (first map) or the like. The phase difference C also depends on the fuel supply amount (operation amount) Q and the number of revolutions N. However, the amount of change in the phase difference C is small and can be obtained by a method similar to the amplitude A. Further, the offset amount D corresponds to the average value Ωm and is D=1.

Further, since the correlation between the amplitude A and the phase difference C is low, it can be easily determined from the waveform of the correction coefficient f(θ) obtained by simulation or experiment.

The fourth figure shows a graph in which the correction coefficient f(θ) is approximated by the above sine wave, and the horizontal axis and the vertical axis are the same as those in the third figure. The fourth figure shows the correction coefficient f(θ) by a solid line and the value approximated by a sine wave by a broken line. As shown, in the seven-cylinder engine, the correction factor f(θ) is extremely accurately approximated by a sine wave.

The fifth figure shows the normalized angular velocity from the first figure, the correction coefficient f(θ) of the present embodiment is not used, and the number of revolutions (dashed line) calculated by the previous method and the present embodiment shown in the fourth figure are used. The time-varying change graph of the number of revolutions (solid line) calculated by the correction coefficient f(θ) of the approximate expression. That is, the number of revolutions of the dotted line (the number of revolutions of the previous method), if the angular velocity detected in the phase angle θ is ω(θ) [rad/sec], is 60 × ω (θ) / (2π) [rpm] to calculate. Further, the effective number of revolutions Ne calculated in the present embodiment is calculated as 60 × ω (θ) × f (θ) / (2π) [rpm].

Further, the angular velocity ω(θ) is calculated from the time interval of the pulse signal generated by the detection of the matching teeth in the crank angle sensor 26 and the tooth pitch (angle), and the phase angle is calculated from the pulse number of the reference pulse and the tooth pitch. . In the fifth diagram, the horizontal axis represents time (seconds), and the vertical axis represents the number of revolutions normalized from the average angular velocity to 1 and normalized.

It is also understood from the fifth figure that the calculation method of the number of revolutions using the correction coefficient f(θ) is significantly smaller than the previous method without using the correction coefficient f(θ), and the value is roughly equal to the average of the angular velocities. The number of revolutions in the value. Thereby, the pulsation caused by the inertia fluctuation and the explosion fluctuation is roughly removed from the number of revolutions.

Further, in the sixth diagram, the difference between the use of the correction coefficient f(θ) of the present embodiment and the use of the moving average is shown in order to remove the pulsation of the inertia fluctuation or the explosion fluctuation. The sixth figure shows the angular velocity ω(t) of the pulsation including the inertia fluctuation and the explosion fluctuation when the number of revolutions is reduced from 70 rpm to 66.5 rpm, the number of revolutions using the moving average of ω(t), and the use of the present embodiment. A time change diagram of the effective number of revolutions Ne of the correction coefficient f(θ), and the horizontal axis represents time and the vertical axis represents the number of revolutions [rpm].

In the sixth diagram, ω(t) is indicated by a broken line S1, and the number of revolutions obtained by using the moving average of ω(t) is represented by a curve S2. Further, the effective number of revolutions Ne calculated by the present embodiment is represented by a curve S3. As shown in the sixth figure, when the moving average is used, the change in the angular velocity ω(t) is delayed by the removal of the pulsating component from the obtained number of revolutions. Therefore, when the number of revolutions is used for feedback control, a reaction lag occurs in the speed control. On the other hand, the effective number of revolutions Ne of the present embodiment using the correction coefficient f(θ) removes the pulsation component and quickly follows the change in the average value of the angular velocity ω(t) without causing hysteresis.

In addition, when the number of cylinders such as the seven cylinders is large, the correction coefficient f(θ) can be approximated by a sine wave, but when the number of cylinders is small, it cannot be approximated by a sine wave. In this case, it is conceivable to distinguish the phase angle θ in the specified range, and each of the sections I approximates the correction coefficient fi(θ) of the inverse of the angular velocity using a polynomial or the like.

Further, in the seventh diagram, as in the fifth diagram, the result of dividing the crank once into five parts in one cylinder two-stroke engine and approximating the correction coefficient by n times in each section is shown. That is, the dotted line is the value of the number of revolutions when the correction is not performed, and the number of revolutions when the solid line is corrected.

As shown in the seventh figure, when the correction factor is used, the fluctuation of the calculated number of revolutions is greatly reduced. Further, in this case, each parameter of the approximate expression is stored in the memory unit 31 in accordance with the phase angle, the number of revolutions, and the fuel supply amount, and the number-of-turns calculation unit 28 calculates the number of revolutions based on the equivalent value selection formula.

As described above, when the control system of the low-speed reciprocating engine of the first embodiment is used, when the number of revolutions is calculated based on a plurality of pulse signals generated in one rotation of the crank, the inertia fluctuation and the explosion can be removed from the calculated number of revolutions. The impact of the change. Thereby, even in a large engine that maintains the number of revolutions at the set value and operates at a low speed, the height control reactivity can be maintained, and stable revolution number control can be realized.

Further, instead of using the approximation formula of the correction coefficient, the value of each phase angle, the number of revolutions, and the fuel supply amount (designated section) may be changed, and the value of the correction coefficient may be previously stored in the memory unit 31 as a lookup table. Further, when the variation is close to a sine wave, F(θ) of the opposite phase (cosine wave) of the approximate equation (sine wave) of Ω(θ)−Ωm at the time of the number of revolutions N and the fuel supply amount Q can be obtained, and The angular velocity variation is offset by the addition of ω(θ) and F(θ). In this case, Ne = 60 × (ω (θ) + F (θ)) / (2π) [rpm] is obtained.

Further, in the first embodiment, the influence of the inertia fluctuation and the explosion fluctuation is removed from the number of revolutions with reference to the phase angle of the crank, the number of revolutions, and the fuel supply amount. However, when it is not necessary to make correction according to the fuel supply amount, that is, when the influence of the explosion fluctuation is small, the number of revolutions can be calculated by correcting only the phase angle of the crank. At this time, since the fuel supply amount is not required to calculate the number of revolutions, the operation amount (fuel supply amount) is not fed back from the operation amount calculation unit 29 to the revolution number calculation unit 28. In addition, on the contrary, the influence of the inertia variation is smaller than the explosion fluctuation, and the correction can be offset only by considering the explosion variation.

Next, the second embodiment will be described with reference to the eighth embodiment. The second embodiment is not required to be corrected in accordance with the fuel supply amount, that is, in response to an explosion fluctuation such as a normal operation under a low load. As described above, the number of revolutions can be calculated only by the correction of the phase angle of the crank at this time. In the second embodiment, not every predetermined crank angle is generated by a pulse, but when the number of revolutions is constant, even if there is inertia fluctuation or the like, pulses are generated at regular intervals, and the control unit multiplies the fixed angular coefficient by the same angular velocity as before. To calculate the number of revolutions.

The eighth diagram is a block diagram showing the configuration of the control system of the low-speed reciprocating engine of the second embodiment, and similarly to the first diagram, only one cylinder is shown, but usually multiple cylinders (for example, 10 cylinders or less) ). In the same manner as in the first embodiment, the same reference numerals will be used, and the description thereof will be omitted.

In the first embodiment, the teeth are formed at the predetermined pitches on the tooth portions 25 provided on the peripheral portion of the flywheel 24. However, the teeth 34 of the flywheel 33 of the second embodiment are formed with teeth at unequal intervals. The interval between the teeth is mainly set corresponding to the inertia fluctuation. For example, even if the angular velocity of the crankshaft 12 (flywheel 33) fluctuates by inertia fluctuation, the pulse signal generated in the crank angle sensor 26 is generated at a certain time interval. It is formed at different intervals corresponding to the crank angle.

The pulse signal generated by the crank angle sensor 26 is input to the number-of-turns calculation unit 35 of the control device 32. The number-of-turns calculation unit 35 calculates the angular velocity from the pulse signal, calculates the number of revolutions by multiplying the fixed coefficient, and outputs it to the operation amount calculation unit 29. The operation amount calculation unit 29 calculates the operation amount of the fuel pump 30 based on the calculated number of revolutions, and outputs it to the fuel pump 30.

As described above, in the second embodiment, the control device does not apply the change, and the pulsation inherent to the engine that does not depend on the fuel supply amount, such as the inertia fluctuation, is removed from the number of revolutions.

Further, the object to be removed is not limited to the inertia fluctuation, and may be removed by the method of the second embodiment as long as the uniqueness corresponds to the fluctuation of the crank angle. Further, in the present embodiment, the unequal intervals are formed in accordance with the period of variation in the distance between the teeth, but a bar-shaped pattern corresponding to such a pitch (for example, a pattern drawn on the seal) may be provided on the periphery of the flywheel. The pattern is read by the sensor, and each revolution number generates a pulse signal of a certain interval. In addition, such structures may also be provided in the encoder rather than on the flywheel.

Further, the influence of the inertia fluctuation may be removed by the method of the second embodiment, and the explosion fluctuation may be corrected in accordance with the crank phase, the number of revolutions, and the fuel supply amount as in the first embodiment.

Further, the present invention is suitable for a marine engine or a land engine used as an engine or a generator in a factory, and the like, and is particularly useful for a large-sized, low-speed reciprocating engine such as a diesel engine. In addition, it is effective for a reciprocating engine that is used for a cylinder number of 10 or less cylinders and preferably 7 to 8 cylinders or less, and is usually used with a number of revolutions of 100 RPM or less and preferably 100 RPM or less.

In addition, the feedback control is exemplified by PID control, but the control method is not limited thereto, and can also be applied to modern control theory, applicable control, learning control, and the like. Further, each of the configurations described in the first embodiment and the second embodiment can be variously combined within the scope of integration.

Component symbol description

10. . . Control System

11. . . Reciprocating engine

12. . . Crankshaft

13. . . crank

14. . . link

15. . . Crosshead

16. . . Piston rod

17. . . piston

18. . . Fuel injection nozzle

19. . . Combustion chamber

20. . . Sleeve

twenty one. . . Exhaust valve

twenty two. . . exhaust vent

twenty three. . . Ventilation port

24,33. . . flywheel

25,34. . . Tooth

26. . . Crank angle sensor

27,32. . . Control device

28,35. . . Rotational number calculation unit

29. . . Operation amount calculation unit

30. . . Fuel pump

31. . . Memory department

The first figure is a control block diagram of a low speed reciprocating engine according to an embodiment of the present invention.

The second graph shows a time-series variation of the angular velocity variation of the crankshaft in a seven-cylinder engine, with the horizontal axis corresponding to time and the vertical axis corresponding to normalized angular velocity.

The third figure covers the crank phase angle 360° section showing the normalized angular velocity variation and its reciprocal variation diagram of the seven-cylinder two-stroke engine including the inertia variation and the pulsation caused by the explosion variation.

The fourth figure shows a diagram showing an example in which the above-described correction coefficient f(θ) is approximated by the above sine wave.

The fifth graph shows a temporal change graph of the number of revolutions (dashed line) calculated by the previous method and the number of revolutions (solid line) calculated using the correction coefficient f(θ).

The sixth graph shows a difference in reactivity when using the correction coefficient f(θ) of the present embodiment in order to remove the pulsation of the inertia fluctuation or the explosion fluctuation.

Fig. 7 is a view showing the number of revolutions when the crank is divided into five parts in one cylinder two-stroke engine, the correction coefficient is approximated by n times in each section, and the number of revolutions is not corrected.

The eighth diagram is a block diagram showing the structure of the control system of the low speed reciprocating engine of the second embodiment.

10. . . Control System

11. . . Reciprocating engine

12. . . Crankshaft

13. . . crank

14. . . link

15. . . Crosshead

16. . . Piston rod

17. . . piston

18. . . Fuel injection nozzle

19. . . Combustion chamber

20. . . Sleeve

twenty one. . . Exhaust valve

twenty two. . . exhaust vent

twenty three. . . Ventilation port

twenty four. . . flywheel

25. . . Tooth

26. . . Crank angle sensor

27. . . Control device

28. . . Rotational number calculation unit

29. . . Operation amount calculation unit

30. . . Fuel pump

31. . . Memory department

Claims (13)

A reciprocating engine revolution number calculating device generates a plurality of signals in one crank rotation, and calculates an actual number of revolutions of the reciprocating engine based on the plurality of signals, and is characterized by having a variation component removing means, which is calculated in the foregoing At the actual number of revolutions, the explosion fluctuation component of the reciprocating engine included in the plurality of signals is removed in accordance with the crank phase angle, the number of revolutions, and the fuel injection amount. The apparatus for calculating a number of revolutions of a reciprocating engine according to claim 1, wherein the plurality of signals are generated from each of a certain crank angle, and the angular velocity calculated according to the plurality of signals is calculated by the operation corresponding to the crank The correction factor of the phase angle is used to remove the aforementioned explosion fluctuation component. The reciprocating engine revolution number calculating device according to claim 2, wherein the fluctuation component removing means removes the inertia fluctuation component from the calculated number of revolutions further corresponding to the crank phase angle. The apparatus for calculating a number of revolutions of a reciprocating engine according to claim 3, wherein the correction coefficient is set to remove a value of the inertia fluctuation component in addition to the explosion fluctuation component. A reciprocating engine revolution number calculating device according to claim 4, wherein the correction coefficient is obtained from the estimated angular velocity from the phase angle, the number of revolutions, and the fuel injection amount. The reciprocating engine of the reciprocating engine of claim 5, wherein the correction coefficient is a reciprocal of the normalized estimated angular velocity of the angular velocity estimated by the average value of 1. The reciprocating engine of the reciprocating engine of claim 6, wherein the number of revolutions for estimating the angular velocity is a moving average of angular velocities calculated based on the plurality of signals, or a first-order lag filtering is performed. It is obtained by any of the values. A reciprocating engine revolution number calculating device generates a plurality of signals in one crank rotation, and calculates an actual number of revolutions of the reciprocating engine based on the plurality of signals, and is characterized by having a variation component removing means, which is calculated in the foregoing At the actual number of revolutions, the inertia fluctuation component of the reciprocating engine included in the plurality of signals is removed in accordance with the crank phase angle. In the reciprocating engine of the reciprocating engine of claim 8, wherein the time required for the crank to rotate once is constant, the plurality of signals are generated at predetermined time intervals in accordance with the inertia fluctuation. The apparatus for calculating a number of revolutions of a reciprocating engine according to claim 9, wherein the plurality of signals are detected by the sensor by a plurality of detected portions along a circumferential direction of the body of revolution that is integrally rotated with the crank And generated. When the time required for one rotation of the crank is constant, the plurality of signals are generated at regular time intervals, and the detected portions are arranged at intervals that are unequal intervals corresponding to the inertia fluctuation. A reciprocating engine control device characterized by comprising the revolving number calculating device according to any one of the preceding claims. A ship characterized by having a reciprocating engine control device of claim 11 of the patent application. A method for calculating the number of revolutions of a reciprocating engine is to generate a plurality of signals in one rotation of the crank, and calculate the actual number of revolutions of the reciprocating engine based on the plurality of signals, which is characterized by calculating the actual number of revolutions, corresponding to the crank phase The angle, the number of revolutions, and the amount of fuel injection are excluded from the explosion fluctuation component of the reciprocating engine included in the plurality of signals.