JPH0570086B2 - - Google Patents

Description

[Detailed Description of the Invention] Technical Field of the Invention The present invention relates to a device for calculating the average period of a pulse signal output from a Karman vortex air flow sensor for measuring the amount of air flowing into an automobile engine or the like. be.

Prior Art and Problems Generally, in an internal combustion engine, the intake air amount is measured, and the fuel injection amount is controlled based on the measurement result so that the air-fuel ratio is constant. Various measurement devices have been proposed to measure the amount of intake air, but one of them is the Karman vortex air flow sensor (hereinafter referred to as the Karman sensor), which has excellent responsiveness.
is attracting attention. This Karman sensor utilizes the fact that when a vortex generator is placed on the intake side of an internal combustion engine, air vortices (Karman vortices) are generated in the vicinity at a frequency proportional to the air flow rate. It generates a pulse at
The period of this generated pulse signal is inversely proportional to the amount of incoming air at that time.

By the way, the air flow rate of an internal combustion engine fluctuates based on the cycles of intake, exhaust, etc.
As shown in FIG. 1, the greater the throttle valve opening, the greater the fluctuation range. Since the Kalman sensor has sufficient responsiveness to follow such fluctuations, for example, as shown in Fig. 12, when the throttle valve opening is large and the flow rate is large, this phenomenon occurs during the period when the flow rate is large. The period of pulses becomes extremely short, and the period of pulses that occur during periods of low flow rate becomes relatively long. Therefore, it is necessary to use some method to find the average pulse period.

A common technique for this purpose is to find the average period of each predetermined pulse. However, in this method, each fluctuation (first
When attempting to calculate the period of section 1 in Fig. 2), it is necessary to average every considerably large number of pulses, which has the disadvantage that responsiveness at low flow rates, where the period becomes long, is impaired. In particular, since there is variation in the number of pulses that appear in one fluctuation interval during large flow rates, it is possible to obtain a more accurate average value by calculating the average period for each of multiple fluctuation intervals during large flow rates. As a result, the response at small flow rates becomes increasingly poor.

Purpose of the Invention The present invention has been made in view of the above circumstances, and its purpose is to improve the calculation accuracy of the average period at times of large flow rates and also to improve the responsiveness at times of small flow rates. .

Principle of the Invention In general, in a Kalman sensor, when the fluctuation width is large during a large flow rate, as shown in Fig. 12, multiple pulses appear and a state with a large pulse period occurs every cycle of the fluctuation of the instantaneous air flow rate. appeared on.
Therefore, a first predetermined time period that is slightly shorter than the maximum period that appears at the time of a large flow rate is set, and the pulse period comparing means determines whether or not the period of the pulse signal is longer than this first predetermined time period. If you output a comparison result that is smaller when the flow rate is large, and a smaller comparison result when the flow rate is not high,
The large comparison results will be output in this order, so by monitoring this, it is possible to detect the break in fluctuations during large flow rates. Therefore, if the average is calculated when the breakpoint of the fluctuation is detected the number of times corresponding to the third predetermined value, it becomes possible to calculate the average for each of the plurality of fluctuations during a large flow rate. Further, since the period of the output pulse of the Kalman sensor at the time of a small flow rate is longer than the first predetermined time, the average If the number of pulses is calculated, it is possible to average the number of pulses corresponding to the second predetermined value. For example, if the second predetermined value is set to 1, it is also possible to calculate the average for each pulse. On the other hand, at medium flow rates, the pulse period is almost always shorter than the first predetermined time, so it is difficult to average the pulses at medium flow rates using only the above-mentioned configuration. However, this can be solved by calculating the average every time a number of pulses corresponding to the first predetermined value appear.

Configuration of the Invention FIGS. 1, 3, and 4 are explanatory diagrams of the configuration of the average calculation device of the present invention.

In the average calculation device shown in FIG. 1, the pulse number counting means 1 counts the number of pulses output from the Kalman sensor and outputs the counted value to the average calculation means 2. The pulse period measuring means 3 measures the period of the pulse output from the Kalman sensor, and the pulse period comparing means 4 compares each pulse period measured by the pulse period measuring means 3 with a first predetermined time, and determines the pulse period. A large comparison result is generated when the time is greater than the first predetermined time, and a small comparison result is generated otherwise. The average calculation means 2 satisfies the following conditions ~
Each time one of the above conditions holds true, the average period of the pulse signal is calculated based on the time since the end of the previous average calculation and the number of pulses.

When the pulse number counting means 1 has counted the number of pulses corresponding to the first predetermined value since the end of the previous average calculation, the pulse period comparison means 4 outputs a large comparison result the number of times corresponding to the second predetermined value. When the pulse period comparison means 4 outputs a small comparison result followed by a large comparison result, the third state is
The first predetermined time, which is the threshold of the period comparison means 4, is set slightly shorter than the longest pulse period (period tmax in FIG. 12) that appears at a large flow rate. . Further, the first predetermined value is a parameter that mainly determines the responsiveness at medium flow rates, and may be a fixed value or a variable value. If it is a fixed value, it needs to be set larger than the maximum number of pulses that will appear in one fluctuation section during a large flow rate. In addition, in the case of a variable value, it is desirable to set the value to be larger than the maximum number of pulses that would appear in one fluctuation section when the flow rate is large, and to set the value to be smaller as the flow rate becomes smaller. The second predetermined value is a parameter that mainly determines the responsiveness of averaging when the flow rate is small, and when the flow rate is small, averaging is performed for each pulse corresponding to the first predetermined value. The first predetermined value may be a fixed value, or may be as small as, for example, 1 when the flow rate is small, and may be a larger value, such as 5, as the flow rate becomes larger. When using a variable value, the increase is gradual;
It is desirable that the decrease be rapid. The third predetermined value may be fixed to, for example, 3, or may be a variable value that is large when the flow rate is large and becomes smaller as the flow rate is small. When using a variable value, it is desirable that the increase be gradual and the decrease be rapid.

Next, in order to make it easier to understand the first illustrated device, the second
A typical operation will be described below with reference to the example pulse signal waveform shown in the figure. In addition, in FIG. 2, m indicates the section to be averaged. Further, for convenience of explanation, the first predetermined value is 10, the second predetermined value is 3, and the third predetermined value is 3.

●At the time of large flow rate (see Figure 2 a) At the time of large flow rate, a state in which a pulse train with a short cycle is generated immediately after which a pulse with a long cycle (longer than the first predetermined time) is generated is continuously repeated. Therefore, the above condition is satisfied, and the average period is calculated every three fluctuations.

●At medium flow rate (see Figure 2b) At medium flow rate, a pulse train with a period shorter than the first predetermined time is generated. Therefore, the above condition is met and averaging is performed every 15 pulses.

●When the flow rate is small (see Figure 2c) When the flow rate is small, a pulse train with a period longer than the first predetermined time is generated. Therefore, the above condition is met and averaging is performed every three pulses.

●During sudden deceleration when a large flow rate suddenly changes to a small flow rate (see Figure 2 d) When a large flow rate suddenly changes to a small flow rate, in that transition region, the above conditions or conditions are met before the above conditions are met. To establish. Therefore, in this case, the average period is calculated from the time since the end of the previous average calculation and the number of pulses when the condition or is satisfied. For this reason, even though the throttle valve is actually in a fully closed state and a state in which not much air is introduced into the engine, there is a drawback that this cannot be immediately detected. That is, the averaging is delayed to some extent, and the final pulse period does not have much influence on the averaging result, resulting in somewhat poor responsiveness during sudden deceleration.

The invention shown in the configuration diagrams of FIGS. 3 and 4 further improves the drawbacks of the invention shown in FIG. 1.

The difference between the invention shown in FIG. 3 and the invention shown in FIG. 1 is that, in addition to the first average calculation means 2a which performs the same operation as the average calculation means 2 of FIG. The second average period calculation means 2b is provided which takes the latest pulse signal period as the average period.

The state in which the pulse period comparison means 4 outputs a small comparison result followed by a large comparison result has appeared at least once since the end of the previous average calculation, but it has still appeared the number of times corresponding to the second predetermined value. According to the present invention, as shown in FIG. Period P ₁ is calculated as the average period, and the period from the end of the previous average to immediately before pulse period P ₁ is ignored, so the transition to a small flow rate during sudden deceleration can be detected relatively quickly. I can do it.

The invention shown in FIG. 4 is different from the invention shown in FIG. 1 in that the invention shown in FIG. 4 is different from the invention shown in FIG.
Separately, a second pulse period comparing means produces a large comparison result when the period measured by the pulse period measuring means 3 is longer than a second predetermined time set to be longer than the first predetermined time. In addition to the first average calculation means 2a which operates in the same manner as the average calculation means 2 shown in FIG. The point is that an average period calculation means 2c is provided.

When a large comparison result is output from the second pulse period comparing means 4b According to the present invention, when the above condition is met as shown in FIG. 2 f, the immediately preceding pulse period P ₂
is calculated as an average period, so a transition to a small flow rate during sudden deceleration can be quickly detected. Note that the second predetermined time period is preferably a variable depending on the engine rotation speed, that is, a variable that increases in value as the engine rotation speed increases, in order to provide sufficient margin for the long period that appears at the time of a large flow rate.

FIG. 5 is a block diagram of main parts of an embodiment of the present invention, in which 11 is an internal combustion engine, 12 is an air cleaner, and 13 is an internal combustion engine.
is a Karman sensor, which sets its output signal a to "1" every time a Karman vortex is generated.
14 is a throttle chamber, 15 is an intake manifold, 16 is an electromagnetic fuel injector, 17 is a throttle valve that controls the flow of intake air, 18 is a crank angle sensor, 19 is a microprocessor, and 20 is a memory. , 21 is the output of the crank angle sensor 18 and other engine parameters such as cooling water temperature, intake temperature, throttle valve opening,
A data input section 22 is a data output section for inputting on/off information of various switches such as an idle switch to the microprocessor 19.

Intake air is supplied to each cylinder from the air cleaner 12 via the Karman sensor 13 and the throttle chamber 14, and from each branch of the intake malfold 15, and fuel is supplied to each cylinder from the fuel injector 1.
6 into the internal heat engine 11. Further, the intake air amount changes depending on the state of the internal combustion engine, and the output signal a also changes according to the change. The output signal a of the Kalman sensor 13 is sent to the microprocessor 1.
9 is input to the interrupt terminal INT, and an interrupt is generated, for example, at the rising edge of the output signal a.

FIG. 6 is a flowchart showing an example of interrupt processing by the microprocessor 19 when realizing the first illustrated device, and S1 to S20 indicate each step.
Note that the second predetermined value is set to 1 here.

When the microprocessor 19 receives an interrupt,
First, the time at that time is stored as the latest pulse time T1 (S1), and the pulse count number NN is incremented by +1 (S2). Next, the number of pulse counts
Determine whether NN is greater than or equal to a predetermined number (first predetermined number) (S3), and if it is greater than or equal to the predetermined number, step
The process moves to S13, and if it is less than the predetermined number, subtract the interrupt time T3 one pulse before from the latest pulse time T1 to obtain the current pulse period T2 (S4), and if T2 is greater than or equal to the predetermined time (first predetermined time), Determine whether or not there is one (S5). If T2 is longer than the predetermined time, the process moves to step S6, and if it is less than the predetermined time, the flag FLAG is set to "1" (S12), and the value of the latest pulse time T1 is set as the interrupt time T3 one pulse before. (S11), return to the main routine.

In step S6, it is determined whether the flag FLAG is set to "1" or not, and if it is set, the flag is set.
After setting FLAG to “0” (S7), the sum of the pulse count number N and the pulse count number NN is set as the pulse count number N (S8), and the pulse count number N
is greater than or equal to a predetermined value N ₃ (third predetermined value) (S9). If the value is greater than or equal to the predetermined value, the step
The process moves to S14, and if it is less than the predetermined value, the pulse count number NN is set to zero (S10), and the process returns to the main routine via step S11.

In addition, in step S13, the sum of the pulse count number N and the pulse count number NN is set as the pulse count number N, and then the inter-pulse time T4 to be averaged by subtracting the time T5 at the end of the previous averaging from the latest pulse time T1. is calculated (S14), and the average period T6 is calculated by dividing this T4 by the pulse count number N (S15). Then, the latest pulse time T1
is stored as time T5 at the end of averaging (S16),
The value of the predetermined number N ₁ (first predetermined number) is changed to the value obtained by adding the constant α to the current number of pulses N (S17). α
is set to about 2 to 4, for example. In addition, in order to limit the maximum value of N ₁ , N ₁ is compared with a predetermined value β (S18), and if it is larger than β, processing is performed to clamp it to the value of β (S19). When the average period has been calculated, the pulse count number N is cleared to zero in preparation for the next process (S20).

At the time of a large flow rate, when a pulse with a short period appears, the process moves from step S5 to step S12, and the flag FLAG is set to "1," and when a pulse with a long period appears, the process moves from step S5 to step S6. Since "1" is set in the flag FLAG in step S7, it is determined that this is the breakpoint of the fluctuation. Then, when a break in the fluctuation is detected, it is determined whether the number of pulses since the end of the previous average calculation process is greater than or equal to the predetermined value _N3 , so the predetermined value _N3 is set to If it is set to, for example, about 2.5 times the number of pulses within one fluctuation, it becomes possible to obtain the average value every three fluctuations.

Further, at the middle flow rate, the process moves from step S3 to step S13, so that the average is calculated for each pulse of the predetermined value _N1 .

When the flow rate is small, the flag FLAG is set to "0", so one pulse period T is set in step S5.
If it is determined that 2 is greater than the predetermined time, the process will proceed to step S13 via step S6.
In this embodiment, the average is calculated for each pulse.

FIG. 7 is a flowchart showing an example of interrupt processing by the microprocessor 19 when implementing the invention shown in FIG. Note that the second predetermined value is set to 1 here as well. The difference from the flowchart of FIG. 6 is that steps S30 to S32 are added. That is, in step S30, it is determined whether or not the current pulse period T2 is longer than a predetermined time _T2 (second predetermined time). The pulse period T2 is set as the average calculation period T6. After calculating the average, the flag FLAG is set to "0" and the process moves to step S16 (S32).

As shown in the flowchart of FIG. 8, the microprocessor 19 calculates the average air flow rate Q from the average pulse period T6 using the known relational expression Q=f ₁ (T6) in the main routine (S35).
In addition, according to the relational expression TM ₂ = f ₂ (engine speed) (a relational expression such that TM ₂ decreases as the engine speed increases),
Processing to change the value of TM ₂ is performed (S36).

FIG. 9 is an example of a flowchart when the microprocessor 19 realizes the invention shown in FIG. 3 at an interrupt, and S40 to S67 indicate each step. First of all,
The definitions of the labels used in this flowchart are shown below.

CAPT: Current interrupt time ZKAR: Previous interrupt time TQF: Final average period after annealing TAV: Basic average period TSIGMA: Added value of each pulse period since the end of the previous average calculation CSIGMA: At the end of the previous average calculation CKAR: Number of interrupts per fluctuation during high flow rate NMIN: Minimum number of pulses per fluctuation NMAX: Maximum number of pulses per fluctuation NAMASI: Number of fluctuations to be averaged during high flow rate expressed in number of pulses XKAR1: A flag that stores the Kalman sensor pulse cycle state XKAR2: A flag that stores the timing of TAV→TQF Basically, there is a relationship of NMIN≦CKAR≦NMAX. The details of the process will be explained below with reference to the flowchart shown in FIG.

Preprocessing When the microprocessor 19 receives an interrupt,
Previous interrupt time ZKAR and current interrupt time
The current pulse period is obtained from CAPT and stored in register Y and register D (S40). Next, change the current interrupt time CART to the previous interrupt time ZKAR
(S41) Next, the pulse period from the end of the previous average calculation is calculated (S42), and the number of interruptions within one fluctuation is calculated (S43). Next, the number of interruptions since the end of the previous average calculation is calculated (S44).

After performing the above preprocessing, averaging calculation processing is performed. The following cases will be explained separately.

At high flow rates At high flow rates, when a pulse train with a short period appears, the process moves from step S45 to step S63.
Flag XKAR1 is set to "1". When a long-period pulse appears, steps S45, S46, and S47
, and the flag XKAR1 is set to "1", so it is determined that it is a change break point. Note that the current pulse period is the first predetermined period T ₁ +1
The reason why the step S46 for determining whether or not it is longer than ms is added is to stabilize the operation by giving the comparison operation a hysteresis characteristic. When a break in fluctuation is detected, flag XKAR1 is set to “0”
Set the minimum number of pulses per fluctuation NMIN to 5,
Set the maximum number of pulses per fluctuation NMAX to 15.
If a fluctuation is detected multiple times, the number of interrupts CSIGMA from the end of the previous average calculation is set to a third predetermined value.
Step S51 to become bigger than NAMASI
In step S52, the average is calculated from the sum of each pulse period and the number of interruptions since the end of the previous average.

At medium flow rate At medium flow rate, the transition from step S45 to step S63 occurs at each interrupt, so the value of CKAR
When the value becomes larger than the value of NMAX, the average period is calculated from the sum of each pulse period since the end of the previous average and the number of interruptions.

When the flow rate is small When the flow rate is small, the process moves to step S46 → step → 47 → step S65 at each interrupt, so CKAR
CKAR when the value of becomes greater than the NMIN value
It is determined whether the value of is equal to the value of CSIGMA. At steady low flow rate, the value of CKAR and
Since the values of CSIGMA are equal, the process moves to step S52, and the average period is calculated from the sum of each pulse period since the end of the previous average and the number of interrupts.

At the time of sudden change from large flow rate to small flow rate (during sudden deceleration) In step S62 after detecting the fluctuation at large flow rate.
The value of CKAR is set to zero, but CSIGMA is not set to zero unless the average calculation process is performed, so when the flow transitions to a small flow rate and a pulse with a large period appears, the step goes through step S46 and step S65.
When the process moves to S66, it is determined that CKAR≠CSIGMA, and the value of register Y that stores the current pulse period is stored in register D (S67). That is, the latest pulse period is calculated as the average period.

Note that the following process is performed as post-processing after the average calculation process is completed.

Step S53: Some filtering is applied by calculating the average of the basic average period calculated this time and the basic average period calculated last time.

Step S54: Determine whether flag XKAR1 is "1" or not.

Step S56: When flag XKAR1 is “1”,
In other words, when the flow rate is medium, the value of NMIN is stored in register A.
Set to .

Step S55: When flag XKAR1 is “0”,
In other words, when the flow rate is small or large, CKAR becomes the value of NMIN.
The value obtained by adding the values of is set in register A.

Steps S57-S58: Set the value of register A to 15.
Clamp between NMAX-1 and store the value of register A to NMAX.

Step S59: Set flag XKAR2 to “1”,
Indicates that an average calculation has been performed.

Steps S60 to S62: TSIGMA, CSIGMA,
Clear the contents of CKAR to zero.

FIG. 10 is a flowchart showing an example of the main processing performed by the microprocessor 19 when realizing the invention shown in the third figure, and S70 to S85 indicate each step.

The microprocessor 19 first reads the flag
The content of XKAR2 is determined, and if it is "1", the average period has been calculated in the previous interrupt processing, so the processing from step S71 onwards is executed to calculate a new final average period TQF after smoothing. flag again
If XKAR2 is “0”, such processing is not performed and the final average period after smoothing calculated last time is used.
Calculate the amount of intake air using TQF. This step of calculating the amount of intake air is omitted from illustration.

In step S71, the average of the TAV value and the TQF value is set as a new TAV value (S71), and then the intake side pressure P is calculated by multiplying TAV by the engine speed NE (S72). Next, it is determined whether the value of P is equal to or greater than a predetermined value, for example, 610 mmHg (S73), and when it is equal to or greater than a predetermined value, that is, when the throttle valve is wide open, 5 is stored in register A (S76);
200 is stored in register B (S77). Also, in the following cases, 0 is stored in registers A and B (S74,
S75). Next, it is determined whether the value of register A is greater than or equal to NMIN, and if it is less than NMIN, step S80 is performed.
If the value is larger than NMIN+1, the value of NMIN+1 is stored in register A and the process moves to step S80. step
S80 sets the value of register A to NMIN. With these processes, the larger the flow rate, the more
Control is performed such that the value of NMIN gradually increases, and when the flow rate becomes small, the value of NMIN rapidly decreases to 0.

Next, it is determined whether the value of NAMASI is greater than or equal to the value of register B (S81), and if it is, the process proceeds to step S83.
If the value is less than , the value of NAMASI + 2 is stored in register B and the process moves to step S83. step
S83 sets the value of register B to the new NAMASI value. Through these processes, control is performed in which the value of NAMASI is gradually increased as the flow rate increases, and is rapidly decreased to 0 when the flow rate becomes small.

Step S84 is a step in which a new final average value after annealing is obtained by averaging the final average values after annealing n times before. Note that n is
It is a variable that changes depending on the value of NAMASI, for example
1 when NAMASI is 0-7, 2 when NAMASI is 8-15,
If it's 16-200, give it a 4.

Step S85 is a step in which the flag XKAR2 is set to "0" to indicate that the final average value after smoothing has been calculated.

According to the embodiments shown in FIGS. 9 and 10, the first predetermined number is divided into two, NMAX and NMIN, and NMAX is used when an interrupt is caused by a pulse with a cycle shorter than the first predetermined time, and NMIN is used for interrupts due to periodic pulses.
Therefore, in this embodiment, the average value is calculated when the pulse period longer than the first predetermined time continues a number of times corresponding to the value of NMIN, and the pulse period shorter than the first predetermined time corresponds to the value of NMAX. The average value is calculated when the number of repetitions continues. For this reason, small flow rate,
It is possible to further improve the responsiveness of averaging at medium flow rates.

Effects of the Invention As explained above, according to the present invention, when the flow rate is large, the average of the Kalman sensor output pulses is obtained for each variation, so even if there is a variation in the number of pulses for each variation, the influence of the variation can be ignored. This has the advantage that the average period can be determined with high accuracy without being affected. Furthermore, the average period can be determined for each number of pulses corresponding to the first predetermined value at medium flow rates, and for each number of pulses corresponding to the second predetermined value at low flow rates, By setting the value as small as possible, it becomes possible to obtain the average period with good responsiveness at both small and medium flow rates.

Furthermore, a state in which a large comparison result is output after a small comparison result from the pulse period comparing means has occurred at least once since the end of the previous average calculation, but only a number of times corresponding to the third predetermined value. When a plurality of large comparison results are successively outputted from the pulse period comparing means before the pulse period comparing means has not yet appeared, a second average period calculating means is provided that takes the latest pulse signal period as the average period, or a second pulse period comparing means that produces a large comparison result when the period measured by the pulse period measuring means is longer than a second predetermined time which is greater than the first predetermined time; and from this second pulse period comparing means. When a large comparison result is output, by providing a second average period calculation means that uses the latest pulse signal period as the average period, it is possible to average the small flow rate when the flow rate suddenly changes from a large flow rate to a small flow rate. This can be done quickly.

[Brief explanation of the drawing]

1, 3, and 4 are configuration explanatory diagrams of the average calculation device of the present invention, FIG. 2 is an explanatory diagram of its operation, FIG. 5 is a main part block diagram of an embodiment of the present invention, and FIG. 6 1 is a flowchart showing an example of the interrupt processing of the microprocessor 19 when realizing the device shown in FIG. 1, and FIGS. Flowchart showing an example; Flowchart showing an example of main processing performed by the microprocessor 19;
FIG. 9 is an example of a flowchart at the time of interrupt of the microprocessor 19 which realizes the invention shown in FIG.
FIG. 10 is a flowchart showing an example of the main processing performed by the microprocessor 19 when realizing the invention shown in FIG. A diagram showing the relationship, FIG. 12 is a diagram showing the instantaneous air flow rate at a large flow rate and the state of the output pulse of the Kalman sensor at that time. 11 is an internal combustion engine, 12 is an air cleaner, 13 is a Kalman sensor, 14 is a throttle chamber, 1
5 is an intake manifold, 16 is an electromagnetic fuel injector, 17 is a throttle valve, 19 is a microprocessor, and 20 is a memory.

Claims (1)

[Scope of Claims] 1. A device for calculating the average cycle of pulse signals output from a Karman vortex air flow sensor at timings related to the generation of Karman vortices, comprising: pulse number counting means for counting the number of pulse signals. , a pulse period measuring means for measuring the period of the pulse signal, the period measured by the pulse period measuring means is a first pulse period measuring means;
Pulse period comparison means that outputs a large comparison result when the time is greater than a predetermined time, and a small comparison result when it is not; average calculating means for calculating the average period of the pulse signal based on the average period of the pulse signal; when the number of pulses corresponding to the first predetermined value has been counted by the pulse number counting means since the end of the previous average calculation; When the comparison result is output a number of times corresponding to a second predetermined value; When the pulse period comparison means outputs a small comparison result followed by a large comparison result the number of times corresponding to a third predetermined value. 1. A device for calculating an average cycle of an output pulse of a Karman vortex type air flow sensor, characterized in that the average period calculation device is characterized in that the average period of the output pulse of a Karman vortex type air flow sensor is provided. 2. A device for calculating the average period of pulse signals outputted from a Karman vortex air flow sensor at timings related to the generation of Karman vortices, comprising: a pulse number counting means for counting the number of the pulse signals; and a period of the pulse signals. a pulse period measuring means for measuring a pulse period measuring means, the period measured by the pulse period measuring means is a first pulse period measuring means;
A pulse period comparison means that outputs a large comparison result when the time is longer than a predetermined time, and a small comparison result when it is not. a first average calculating means for calculating the average period of the pulse signal based on the pulse period comparison; when the pulse number counting means has counted the number of pulses corresponding to the first predetermined value since the end of the previous average calculation; and the pulse period comparison. When a large comparison result is outputted from the means a number of times corresponding to the second predetermined value, a state in which a large comparison result is outputted after a small comparison result from the pulse period comparison means becomes a third predetermined value. a second average period calculation means which takes the latest pulse signal period as the average period each time the following conditions are satisfied, and a large comparison result follows a small comparison result from the pulse period comparison means; Before the output state has appeared at least once since the end of the previous average calculation but has not yet appeared the number of times corresponding to the second predetermined value, a plurality of consecutive large comparison results are obtained from the pulse period comparison means. 1. A device for calculating an average cycle of output pulses of a Karman vortex type air flow rate sensor, characterized in that the average period calculation device includes a time when output pulses are output from a Karman vortex type air flow sensor. 3. A device for calculating the average period of pulse signals output from a Karman vortex air flow sensor at timings related to the generation of Karman vortices, comprising: a pulse number counting means for counting the number of the pulse signals; and a period of the pulse signals. a pulse period measuring means for measuring a pulse period measuring means, the period measured by the pulse period measuring means is a first pulse period measuring means;
a first pulse period comparing means which produces a large comparison result when the period is longer than the first predetermined time period, and a small comparison result when the period is not longer than the first predetermined time period; a second pulse period comparing means that produces a large comparison result when the period is longer than a predetermined time period; average calculation means for calculating the average of the cycles; when the number of pulses corresponding to a first predetermined value has been counted by the pulse number counting means since the end of the previous average calculation; and a great comparison from the first pulse period comparison means. When the result is output a number of times corresponding to the second predetermined value, a state in which the first pulse period comparing means outputs a small comparison result followed by a large comparison result corresponds to the second predetermined value. a second average period calculating means which sets the latest pulse signal period as the average period each time the following conditions are satisfied; and when a great comparison result is output from the second pulse period comparing means. A device for calculating the average period of Karman vortex air flow sensor output pulses, characterized by the following: