JP2008076083A - Setting method of sampling period, and signal processing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To ensure irregular change of a sampling period, and to suppress precisely an influence of pulsation of an objective signal on a sampling result. <P>SOLUTION: In A/D conversion start processing executed for each reference sampling period [4ms], while changing a period element acquired in arrangement order for each execution of processing from a waiting time table wherein each period element is arrayed in random order, the period element is acquired, and the acquired period element is set as a waiting time WT from finish of the processing until start of an A/D converter. Hereby, the sampling period is set individually for each sampling by using a difference between a period element arrayed in the XI th number of a waiting time calculation table and a period element arrayed in the (XI+1) th number as a value added to the reference sampling period. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

    The present invention relates to a sampling period setting method for sampling performed when an analog signal is converted into a digital signal, and a signal processing apparatus employing the setting method.

    Many sensors are installed in the vehicle to detect the driving situation. Most sensors output the detection result as an analog signal. The sensor signal output as an analog signal is converted into a digital signal through an analog / digital (A / D) converter, and then subjected to processing in a microcomputer that controls various kinds of vehicle control. The A / D conversion of the sensor signal is performed by sampling the voltage or current of the sensor signal at a constant cycle. In many cases, high-frequency random noise (white noise) due to the influence of surrounding electromagnetic waves is superimposed on the sensor signal, so the noise component of the sensor signal is removed by a hardware filter (CR filter, etc.) or software filter. Is performed prior to the A / D conversion.

    On the other hand, in an air flow meter or the like that detects the intake air amount (intake flow rate) of the engine, the level (output voltage, output current, etc.) of the sensor signal may pulsate. For example, in the case of an air flow meter, the intake air in the engine is intermittently performed, and the air flow rate in the intake passage increases and decreases in a constant cycle, so the sensor signal level (voltage value, current value, etc.) It pulsates at a cycle according to the rotation speed. The sampling timing in the A / D conversion is synchronized with such pulsation, and the detection accuracy may deteriorate. That is, if the sampling period of the sensor signal coincides with the pulsation period or becomes an integral multiple of the pulsation period, the level of the sensor signal is sampled only with the same phase in the pulsation period. For example, in the case of sampling a sensor signal of an air flow meter at a cycle of “4 milliseconds” in a 4-cylinder engine that performs intake at intervals of “180 ° CA”, if the engine rotation speed is near “7200 rpm”, the cycle of the intake pulsation And the sampling cycle of the sensor signal of the air flow meter coincide with each other. In such a case, only the peak portion or valley portion of the pulsation of the sensor signal is sampled, and the detection result of the intake air amount deviates from the average intake flow rate in the intake passage, resulting in air-fuel ratio control and drivability. It will have an undesirable effect.

    As countermeasures against the influence of such pulsation, the time constant of the filter is increased to reduce the pulsation of the sensor signal in advance before A / D conversion, or the sampling period is shortened so that the sampling period becomes the pulsation period. It is conceivable to increase the engine speed to synchronize. However, if the time constant of the filter is increased, a sudden change in the intake air amount cannot be detected. In addition, shortening the sampling cycle increases the processing load on the microcomputer related to signal capture.

    Further, Patent Document 1 proposes a countermeasure for dynamically changing the sampling period so that the influence of the pulsation is canceled together with the change of the pulsation period. Specifically, the period of the intake pulsation is obtained from the engine rotation speed, and the sampling period is set so that the phase of the pulsation period of the sensor signal to be sampled is shifted by a half phase for each sampling. For this purpose, the sampling period may be set to “1/2” times the pulsation period. In this case, however, the sampling time interval becomes too short in the high engine speed range, and the signal is captured by the microcomputer. Such processing load becomes excessive. Therefore, as the engine speed increases, the sampling period is relatively increased to “½” times, “3/2” times, “5/2” times,... According to such setting of the sampling period, the phase of the pulsation period of the sensor signal to be sampled is shifted by a half phase for each sampling, and the peak part and the valley part of the pulsation are alternately sampled. The influence can be countered. However, the sampling cycle setting method has a problem that the sampling cycle changes greatly depending on the engine speed.

    In Patent Document 2, two A / D conversion channels farthest from the microcomputer are assigned for processing the sensor signal of the air flow meter, and a time difference is provided for each signal acquisition process of the microcomputer. It is described that sampling is performed once. Even in such a case, it is possible to prevent only the peak portion or the valley portion of the pulsation from being sampled. In addition, in this case, since the sampling is performed twice in one process related to signal capture, the increase in processing load can be kept small compared to a case where the sampling period is simply shortened. However, assigning two A / D conversion channels for capturing the sensor signal of the air flow meter can be employed only when the number of A / D conversion channels is sufficient. In recent years, the time required for A / D conversion per channel has been shortened to about “10 microseconds” by improving the performance of the A / D converter. Even if they are assigned, it is difficult to secure a sufficient sampling time difference.

Therefore, as seen in Patent Document 3, a technique for suppressing the influence of pulsation by changing the sampling period irregularly has been proposed. In this document, the value of the lower-order bits of the sampling value of the sensor signal and the value of the lower-order bits of the counter (free-run counter) that measures the sampling period are used as pseudo-random numbers. Then, by setting the sum of the irregular periodicity obtained by multiplying the pseudo random number obtained from such values by a constant and the fixed periodicity that is a constant as the sampling period, the irregular change is realized. ing.
JP-A-63-97856 JP-A-2-241948 JP 2000-276991 A

    If the technique described in Document 3 can change the sampling cycle irregularly as intended, it is certainly possible to avoid synchronization of the sampling cycle with periodic pulsations. However, the method described in the document 3 has a problem in the random number generation method used for irregularly changing the sampling period, and the sampling period may not be irregularly changed as intended.

    First, consider the case where the value of the lower bits of the sampling value of the sensor signal is used as a random number. Even if the detection amount of the sensor has hardly changed, if the sampling period changes irregularly, the phase of the pulsation period in which sampling is performed becomes indefinite.Therefore, there is irregularity in the value of the lower bits of the sampling value of the sensor signal. appear. Therefore, in such a case, it is possible to change the sampling period irregularly by using the value of the lower bit as a random number. However, it is assumed that a sampling cycle coincident with the pulsation cycle is accidentally set at a certain time. The phase of the pulsation cycle in which sampling is performed at this time is the same as in the previous sampling. Therefore, there is a possibility that the sampling value of the sensor signal at this time is hardly different from the previous sampling. As a result, there is a possibility that the sampling period is fixed to be almost constant. On the other hand, the same can occur when the value of the lower bit of the free counter is used. In other words, when randomness in a closed system is used as a random number, there is a possibility that a feedback relationship will occur between the sampling period and the state quantity that is the basis of the random number, and a regular sampling period may be set. Cannot be denied.

    In recent years, adaptation of various engine control parameters is performed in a simulation environment without using an actual machine. The adaptation in the simulation environment is often performed in a state where the input parameters such as the intake air amount are set to a constant value or a calculated value artificially generated using a relatively simple function. If this happens, there is a higher possibility that the sampling period will have regularity. Then, the adaptation is performed under such inappropriate sensor signal capturing operation, and there is a possibility that the inappropriate adaptation is performed.

    The present invention has been made in view of such circumstances, and the problem to be solved is to ensure irregular fluctuation of the sampling period and to more accurately suppress the influence of the pulsation of the target signal on the sampling result. An object of the present invention is to provide a sampling period setting method and a signal processing device that can be used.

    In order to solve the above-described problem, in the invention according to claim 1, in the sampling period setting method for setting the sampling period when sampling an analog signal, a table in which a plurality of periodic elements are arranged in random order is created in advance. In addition, the periodic elements are acquired from the table in the order of arrangement for each sampling, and the sampling period is individually set for each sampling according to the acquired periodic elements.

    Here, the periodic element means a variable value used for calculation when setting the sampling period. That is, the sampling period is obtained through operations such as addition, subtraction, multiplication, and division using the periodic element. Of course, the periodic element itself can be directly set to the sampling period.

    In the above method, since the sampling period is individually set for each sampling according to the periodic elements arranged in random order in the table, the sampling period is irregularly changed, and the pulsation period of the analog signal to be sampled adversely affects the sampling result. Can not be given. Since the pattern of irregular change of the sampling period at this time is uniquely determined by the arrangement of the periodic elements in the table, the artificially programmed irregularity is given to the sampling period. For this reason, there is no room for contingent elements to add irregularity to the sampling period, such as when using the sampling value of the sensor signal or the value of the low-order bit of the free-run counter. A situation in which regularity is lost cannot occur. That is, it is possible to reliably ensure that the sampling period actually changes irregularly without needing to verify. In addition, whether under actual usage conditions or in a simulation environment that simplifies the transition pattern of input parameters, the sampling cycle can always be irregularly changed in a constant pattern, giving the simulation results high reliability. You will be able to

    In addition, since the sampling period can be set entirely on software, it can be realized at a lower cost than when a dedicated random number generator is provided. Incidentally, when the sampling period is irregularly changed using the random number generated by the random number generator, it is impossible to predict in what pattern the sampling period actually changes. Can be predicted. This is advantageous in situations such as when pre-verifying whether sampling is actually performed properly or when performing analysis when a problem occurs. That is, when the sampling cycle is irregularly changed using random numbers generated each time, it is extremely difficult to reproduce the sampling operation. However, according to the method of the present invention, the sampling operation can be easily reproduced faithfully.

    In such a sampling period setting method according to claim 1, in claim 2, the sampling period is set to be the sum of the variable component obtained according to the periodic element obtained from the table and the reference sampling period which is a fixed value. It is set. If it is desired to change the sampling period within a certain range around the reference value, setting the sampling period in this manner makes it easy to set numerical values such as periodic elements arranged in the table.

    Further, according to claim 3, in the sampling period setting method according to claim 2, the i-th array element arranged in the table is “TBL [i]”, and the index incremented every sampling is “XI”. Then, the difference between the XIth periodic element TBL [XI] arranged in the table and the (XI + 1) th periodic element TBL [XI + 1] is added to the reference sampling period which is a fixed value. The sampling period is set so as to be the value obtained. In such a case, the average value (expected value) of the sampling period matches the reference sampling period. Further, according to claim 4, if the range of possible values of each periodic element arranged in the table is set to “0” or more and less than the reference sampling period, once for each desired period (window). Sampling is surely performed.

    If the sampling period is irregularly changed as described above, it is possible to suitably avoid the influence of the synchronization between the pulsation period of the analog signal to be sampled and the sampling period on the sampling result. However, in a situation where there is no possibility of such synchronization in the first place, there is no need to intentionally change the sampling period irregularly. Therefore, as in the method according to claim 5, in the sampling period setting method according to claims 2 to 4, when the sampling period is fixed to the reference sampling period, the sampling period is synchronized with the pulsation period of the analog signal. If it is determined whether or not to synchronize, and it is determined to be in synchronization, the sum of the variable component and the reference sampling period is set as the sampling period. If not, the sampling period is set as the reference sampling period. Even if it is set, it is possible to avoid the influence of the synchronization on the sampling result.

    On the other hand, if the pulsation pattern changes depending on the situation, the pattern of irregular change of the sampling cycle that is optimal for suppressing the influence also changes. In that respect, in the invention according to claim 6, in the sampling period setting method according to any one of claims 1 to 5, as the table used for setting the sampling period, a plurality of different arrangement modes of periodic elements are used. The table used for setting the sampling cycle is switched according to the sampling execution status. For this reason, it is possible to change the pattern of the irregular change of the sampling period in accordance with the change of the pulsation pattern according to the situation. For example, when sampling the sensor output of an intake air amount detection sensor or an intake pressure detection sensor, the sensor output pulsates in accordance with the intake pulsation. Since the period of the intake pulsation changes depending on the number of cylinders of the engine or the engine speed, the pulsation pattern of the sensor output changes according to the number of cylinders of the engine and the engine speed. Therefore, when the sensor output of the intake air amount detection sensor or the intake pressure detection sensor is to be sampled, the table is switched according to the number of engine cylinders, the engine speed range, or both. The pattern of the irregular change in the sampling period can be suitably changed in accordance with the change in the pulsation pattern of the sensor output.

    The period arranged in the table used in such a sampling period setting method may be obtained through verification using a simulation environment or a compatible environment, as described in claim 7.

    Note that the above sampling period setting method can be applied to sampling of a sensor signal of an intake air amount detection sensor of an engine, as described in claim 8. Of course, in addition to this, for sampling of an arbitrary analog signal, the pulsation is superimposed on the analog signal to be sampled, and the sampling result may be adversely affected when the pulsation is synchronized with the sampling period. The sampling period setting method of the invention can be similarly applied.

    On the other hand, in order to solve the above-mentioned problem, in the invention according to claim 9, in the signal processing device that periodically samples an analog signal and converts it into a digital signal, a table in which a plurality of periodic elements are arranged in random order; A sampling period setting unit that acquires periodic elements from the table in the order of arrangement for each sampling and individually sets the sampling period for each sampling according to the acquired period is provided.

    In the above configuration, since the sampling period is set according to the periodic elements arranged in the order of the table, the sampling period is irregularly changed. The pattern of irregular change of the sampling period at this time is uniquely determined by the arrangement of the periodic elements of the table, that is, artificially programmed, and the sampling period is actually not necessary to verify. It can be assured that it changes irregularly. In addition, the sampling cycle can be irregularized at a lower cost than when a dedicated random number generator is provided, and the sampling operation can be faithfully reproduced. .

    The sampling period setting means of such a signal processing device has a sampling period so as to be the sum of a variable component obtained in accordance with the periodic element acquired from the table and a reference sampling period which is a fixed value. It can also be configured to perform individual settings. In such a case, when it is desired to change the sampling period within a certain range around the reference value, it is possible to more easily set the numerical values of the periodic elements arranged in the table.

    Further, the sampling period setting means of the signal processing device according to claim 9 is set such that the i th periodic element arranged in the table is “TBL [i]” and incremented for each sampling as described in claim 11. When the index is “XI”, the XIth periodic element TBL [XI] arranged in the table and the (XI + 1) th periodic element TBL [XI + 1] in the reference sampling period which is a fixed value. If the sampling period is set so as to be a value obtained by adding the difference between the two, the average value (expected value) of the irregularly changed sampling periods is matched with the reference sampling period. Further, according to the twelfth aspect, if the range of possible values of each periodic element of the table is set to “0” or more and less than the reference sampling period, the sampling is surely performed once in the window. Will be able to.

    Note that, as described above, in a situation where there is no possibility of synchronization between the pulsation period of the analog signal to be sampled and the sampling period, the sampling period does not need to be irregularly changed. And determining means for determining whether or not the fixed sampling period is synchronized with the pulsation period of the analog signal when the sampling period is fixed to the reference sampling period, and the sampling period setting means determines the determination. When it is determined that the means is “synchronized”, the sampling period is individually set for each sampling so as to be the sum of the variable component and the reference sampling period. Otherwise, the sampling period is set to the reference sampling period. It is good also as setting to fixed uniformly.

    Further, in the invention described in claim 14, as the table used for setting the sampling period in the signal processing apparatus according to claims 10-13, a plurality of tables having different arrangement modes of the periodic elements are provided, and the sampling period setting is performed. Table switching means for switching the table used by the means for setting the sampling cycle in accordance with the execution state of sampling is provided. In such a case, the irregular change pattern of the sampling period can be changed in accordance with the change of the pulsation pattern according to the sampling situation. For example, if the analog signal to be sampled is the sensor signal of the sensor for detecting the intake air amount of the engine, the sensor signal pulsates due to the influence of intake air pulsation caused by intermittent intake of air in each cylinder of the engine. And the pulsation pattern (cycle) changes depending on the number of cylinders and the rotational speed of the engine. Accordingly, as in claim 15, the table switching means switches the table in accordance with the number of cylinders of the engine to be applied, or in accordance with claim 16, the table switching means is in the engine speed range. If the table is switched accordingly, it is possible to change the pattern of the irregular change of the sampling period in accordance with the change of the pulsation pattern of the sensor signal of the intake air amount detection sensor.

    The signal processing device according to any one of claims 9 to 14 is applied to a signal processing device used for analog / digital conversion of a sensor signal of an intake air amount detection sensor of an engine as described in claim 17. Can do. Of course, the present invention can be similarly applied to a signal processing apparatus used for A / D conversion of other signals.

    Hereinafter, an embodiment in which a sampling period setting method and a signal processing apparatus employing the setting method according to the present invention are embodied will be described with reference to FIGS. In this embodiment, the present invention is applied to a signal processing device that performs A / D conversion of a sensor signal of an intake air amount detection sensor in an in-vehicle control device, and a method of sampling the sensor signal at the time of the A / D conversion. It has become.

    First, an outline of a vehicle control device to which the present embodiment is applied will be described. A plurality of sensors and a plurality of actuators are connected to the vehicle control device. The vehicle control device generates a control signal for the actuator based on sensor signals input from these sensors, and outputs the generated control signal to the actuator, thereby adjusting the fuel injection amount and intake air to the engine. Control the amount.

    As shown in FIG. 1, the vehicle control apparatus 1 includes, as sensors, a VG sensor 10, which is a hot-wire air flow meter, a fuel temperature sensor 11, a water temperature sensor 12, an intake air temperature sensor 13, an oxygen sensor 14, and a rotation angle sensor. 15, a vapor flow sensor 16, an idle switch 17, and the like are connected. Further, an igniter 20, a fuel injection valve 21, an idle speed control valve (ISCV) 22, a purge control valve 23, an exhaust gas recirculation valve (EGRV) 24, and the like are connected to the vehicle control device 1 as actuators. .

    The VG sensor 10 is a sensor for detecting the intake air amount of the engine, and is a closed loop circuit for maintaining a constant temperature difference between the heat ray disposed in the intake passage and the heat ray and the air in the intake passage. And is configured. If the flow rate of air in the intake passage, that is, the amount of intake air is increased or decreased, the amount of heat taken away by the heat rays is also increased or decreased, and the current required for heating the heat rays is also increased or decreased. The VG sensor 10 outputs, to the vehicle control device 1, a sensor signal obtained by performing linearization or temperature compensation processing on the current flowing through the heat wire for such heating.

    On the other hand, the vehicle control device 1 includes a microcomputer 30 as its main part. The microcomputer 30 includes a central processing unit (MPU) 31, a ROM 32 storing various processing programs and data, a RAM 33 used as a work area, a backup RAM 34 for retaining data even after the engine is stopped, an input / output port 35, an input port 36 and output ports 37 to 41, and these are connected to the bus line BL.

    Further, the vehicle control device 1 includes buffers 50 to 53 and an A / D converter 54 to which the buffers 50 to 53 are connected and to the input / output port 35. The A / D converter 54 is a signal processing device that converts the sensor signal of each sensor input as an analog signal into a digital signal. Specifically, the A / D converter 54 takes in the sensor signal temporarily stored in the buffers 50 to 53 based on the conversion start signal input from the input / output port 35 and converts it into a digital signal. The digital signal is output to the input / output port 35. More specifically, the A / D converter 54 converts the sensor signal into a digital signal after the time (delay time) indicated by the conversion start signal has elapsed. In the vehicle control device 1, among the plurality of sensors described above, the VG sensor 10, the fuel temperature sensor 11, the water temperature sensor 12, and the intake air temperature sensor 13 are connected to the buffers 50 to 53, respectively.

    The vehicle control device 1 further includes a buffer 55, a comparator 56 connected to the buffer 55 and connected to the input port 36, and a waveform shaping circuit 57 connected to the input port 36. The comparator 56 and the waveform shaping circuit 57 shape the waveform of the input sensor signal, and then output the signal after waveform shaping to the input port 36. Among the plurality of sensors described above, the oxygen sensor 14 is connected to the buffer 55, and the rotation angle sensor 15 and the vapor flow sensor 16 are connected to the waveform shaping circuit 57, respectively. The idle switch 17 is also connected to the input port 36 through a buffer or the like, and an on / off signal is sent to the bus line BL.

    The vehicle control device 1 includes drive circuits 60 to 64 connected to the output ports 37 to 41, respectively. The above-mentioned igniter 20, fuel injection valve 21, ISCV 22, purge control valve 23, and EGRV 24 are connected to these drive circuits 60 to 64, respectively. In the vehicle control device 1, the command signals output from the output ports 37, 38, 39, and 40 are respectively connected to the igniter 20, the fuel injection valve 21, the ISCV 22, and the EGRV 24 via the drive circuits 60, 61, 62, 63, and 64. To be supplied to.

    Under such a configuration, the vehicle control device 1 generates a control signal to an actuator such as the igniter 20 based on each sensor signal input from the sensors 10 to 17, and sends this control signal to the drive circuits 60 to 64. By outputting, the actuator such as the igniter 20 is driven.

    Incidentally, the control signals input to these drive circuits 60 to 64 are generated by the microcomputer 30 based on the digital signal output from the A / D converter 54. In generating this control signal, the microcomputer 30 outputs the above-described conversion start signal to the A / D converter 54. Here, taking as an example a series of processing by the microcomputer 30 until the instantaneous flow rate average value used in generating the control signal from the sensor signal of the VG sensor 10 is calculated, analog / digital conversion (A (D conversion) will be described in detail. Here, as shown in FIG. 2, it is assumed that a sensor signal that pulsates at a period of “4 ms” is output from the VG sensor 10.

    The microcomputer 30 activates the A / D converter 54 by executing an A / D conversion activation process, which will be described later, every “4 ms” execution cycle (reference timing value CP), and digitally converts the sensor signal that is an analog signal. Convert to signal. In executing the A / D conversion activation process, the microcomputer 30 refers to the waiting time calculation table TBL stored in the ROM 32. In the vehicle control apparatus 1 according to the present embodiment, as described above, the execution cycle of the A / D conversion activation process and the pulsation cycle of the sensor signal are the same by referring to the waiting time calculation table TBL by the microcomputer 30. Even if it exists, a sensor signal can be sampled suitably. First, the contents of the waiting time calculation table TBL will be described with reference to FIGS.

    As shown in FIG. 3, the waiting time calculation table TBL has, for example, a one-dimensional arrangement, and twelve periodic elements “Tad — 0” to “Tad — 11” are arranged in random order as the waiting time WT. This waiting time WT is obtained through verification using a simulation environment or a compatible environment. In the present embodiment, as shown in FIG. 4, these Tad_0 to Tad_11 are values obtained by dividing the difference between adjacent ones of Tad_0 to Tad_11 into 12 equal parts of the execution period of the A / D conversion activation process. Is the time required to be That is, Tad_0 to Tad_11 are 0, 0.333, 0.667, 1.000, 1.333, 1.667, 2.000, 2.333, 2.667, 3.000, 3.333, and 3.667 [ms], respectively. As will be described later, each of the periodic elements Tad_0 to Tad_11 of the waiting time calculation table TBL is used for calculating the sampling period Ts_n. In the waiting time calculation table TBL, these periodic elements Tad_0 to Tad_11 are arranged in random order as shown in FIG. 3 in order to irregularly change the sampling period Ts_n calculated using them. Here, the IXth periodic element arranged in the waiting time calculation table TBL is described as TBL [IX]. Note that “IX” is used as an index when acquiring the periodic element from the waiting time calculation table TBL, and the value thereof is any natural number from “0” to “11”.

Here, the processing procedure of the A / D conversion activation processing is shown as a flowchart in FIG. 5 and will be described with reference to this flowchart.
First, in step S10, the microcomputer 30 increments the value of the index IX, that is, adds “1” to the value. However, when the value of the index IX becomes larger than “11” (step S11: YES), the value is reset to “0”. That is, each time the A / D conversion activation process is executed, the microcomputer 30 sequentially increments the value of the index IX within the range of “0” to “11”.

    In step S13, the microcomputer 30 acquires the IXth periodic element (variable value TBL [IX]) of the above-described waiting time calculation table TBL as the waiting time WT based on the value of the index IX. In step S14, the microcomputer 30 determines whether or not the value of the waiting time WT is “0”. If it is determined in step S14 that the value of the waiting time WT is “0”, the microcomputer 30 causes the delay time in step S15 to execute A / D conversion of the sensor signal without waiting time. Is output to the A / D converter 54. Thereafter, the process is temporarily terminated.

    On the other hand, if it is determined in step S14 that the value of the waiting time WT is a value other than “0”, the microcomputer 30 determines that the conversion start signal having the waiting time WT as the delay time is A in step S16. / D converter 54. Thereafter, the process is temporarily terminated. Thereby, in the A / D converter 54, after the time indicated by the waiting time WT has elapsed from the output of the conversion start signal by the microcomputer 30, the A / D conversion of the sensor signal is executed.

    The interval from the (n−1) th A / D conversion to the nth A / D conversion of the sensor signal executed in accordance with the waiting time WT thus set, that is, the sensor signal sampling period Ts_n is: As shown in the following formula. Incidentally, the value of the index IX at the time of setting the n-th sampling cycle Ts_n is “1” from the remainder when the set number of sampling cycles n is divided by the total number of cycle elements “12” of the waiting time calculation table TBL. The value obtained by subtracting.

[Equation 1]

Ts_n = (4 [ms] −TBL [IX]) + TBL [IX + 1] (1)

FIG. 6 shows the timing of A / D conversion of the sensor signal by the A / D converter 54 together with the sensor signal pulsating with a period of “4 ms” using square symbols. As described above, the values of Tad_0 to Tad_11 stored in the waiting time calculation table TBL are rearranged so that the sampling periods Ts_n indicated by the relational expression are irregularly arranged. For this reason, the sampling period Ts_n indicated by the relational expression is not in regular order but has irregularity.

    Incidentally, as indicated by a circle symbol in FIG. 6, when the sampling period is fixed to “4 ms” as in the prior art, the A / D converter 54, when the pulsation of the sensor signal and the sampling period are synchronized, for example, Only the peak portion of the pulsation in the sensor signal is sampled. Incidentally, the engine speed at which the pulsation cycle is “4 ms” varies depending on the number of cylinders. For example, it is “7424 rpm” for a 4-cylinder engine, “4883 rpm” for a 6-cylinder engine, and “3662 rpm” for an 8-cylinder engine (both in the case of a 4-cycle engine. Each value is halved for a 2-cycle engine).

    When the A / D conversion through the A / D conversion activation process is completed, the microcomputer 30 generates an interrupt indicating the completion of the A / D conversion. When such an interrupt occurs, the microcomputer 30 executes the following A / D conversion completion processing. Next, the A / D conversion completion processing will be described with reference to the flowchart shown in FIG.

    As shown in FIG. 7, in step S <b> 20, the microcomputer 30 takes in the digital signal input from the A / D converter 54. In step S21, the microcomputer 30 obtains the instantaneous intake flow rate from the digital signal, and then in step S22, the instantaneous intake flow rate is stored in the circulation array (ring buffer). Thereafter, the process is temporarily terminated.

    Further, the microcomputer 30 calculates the average value of the instantaneous intake flow rate. Next, the instantaneous flow rate average value calculation process executed at this time is shown as a flowchart in FIG. 8, and the processing procedure for the instantaneous flow rate average value calculation process is listed below with reference to this flowchart.

    1. By dividing the ignition interval time by the execution period (reference timing) “4 ms” of the A / D conversion activation process, the integrated number is obtained (FIG. 7: step S30). Here, the ignition interval time refers to an ignition time interval between cylinders. For example, in the case of 4 cylinders, the time corresponding to 180 CA is the ignition interval time, and in the case of 6 cylinders, the time corresponding to 120 CA is the ignition interval time. In the case of 8 cylinders, the time corresponding to 90 CA is the ignition interval time. That is, the integrated number corresponds to the number of samplings within the ignition interval time and varies depending on the engine speed.

2. Of the instantaneous intake air flow rate stored in the circulation array, a sum is obtained for the values corresponding to the accumulated number from the latest data (FIG. 7: step S31).
3. By dividing the integrated value by the integrated number, an average value of the instantaneous intake flow rate is calculated (FIG. 7: Step S32).

    Thus, the digital signal generated by A / D converting the sensor signal of the VG sensor 10 is averaged through the execution of the instantaneous flow rate average value calculation process. For this reason, even if the sampling period is irregular as described above, there is no problem in calculating the fuel injection angle and the fuel injection amount in the engine control.

    Next, a specific example of the sampling cycle when the sensor signal of the VG sensor 10 is sampled by the vehicle control device 1 according to the present embodiment is shown in FIG. 9 and will be described in detail according to this specific example.

    As shown in the waiting time calculation table TBL of FIG. 3 described above, the waiting time WT acquired from the waiting time calculation table TBL when the index IX is “0” is Tad_4 (= 1.333 [ms]). The waiting time WT acquired from the waiting time calculation table TBL when the index IX is “1” is Tad_0 (= 0 [ms]). Similarly, the waiting time WT when the index IX is “2” is Tad_8 (2.667 [ms]), and the waiting time WT when the index IX is “3” is Tad_2 (0.667 [ms]).

    Therefore, the sampling cycle Ts_0 from the first A / D conversion to the second A / D conversion is 2.667 (= 4-1.333 +0) [ms]. The sampling period Ts_1 from the second A / D conversion to the third A / D conversion is 6.667 (= 4-0 + 2.667) [ms]. The sampling period Ts_2 until the second A / D conversion is 2.000 (= 4-2.667 +0.667) [ms]. As described above, the sampling period Ts_n is irregularly changed every sampling.

    In this embodiment, the vehicle control device functions as a signal processing device as part of its function. Further, in the present embodiment, in the A / D conversion activation process performed every reference sampling period, the result of obtaining the waiting time WT in the processes of steps S13 to S16 and starting the A / D converter after the waiting time WT. As described above, the sampling period is set by the sampling period setting means.

According to the sampling period setting method and the signal processing apparatus of the present embodiment described above, the following effects can be obtained.
(1) Since artificially programmed irregularities are added to the sampling period, there is no room for contingent elements in the provision of irregularities to the sampling period, and sampling is performed without bothering verification. It can be ensured that the period actually changes irregularly. Therefore, it is possible to reliably suppress the influence of the pulsation of the target signal on the sampling result by ensuring the irregular change of the sampling period.

    (2) Since the irregularly changing sampling period can be set only on software, the irregular sampling period can be changed at a lower cost than when using hardware such as a random number generator. can do.

(3) Since the faithful reproduction of the sampling operation can be easily performed, prior verification of the sampling operation and analysis when a problem occurs are facilitated.
(4) Regardless of the situation, the irregular change pattern is determined only by the arrangement of the periodic elements in the waiting time calculation table TBL, so that the actual use is possible even in a simulation environment in which the surrounding situation is simplified. Even under circumstances, the sampling period can be changed irregularly at a constant pattern. Therefore, a highly reliable simulation result can be obtained.

    (5) The sampling period is set as the sum of the variable component set according to the periodic element of the waiting time calculation table TBL and the fixed component (reference sampling period) that is a fixed value. The index XI that is incremented every sampling is used to obtain the XI-th and (XI + 1) -th periodic elements of the waiting time calculation table TBL, and the difference between them is used as the variable component. For this reason, the average value (expected value) of irregularly changing sampling periods can be matched with the reference sampling period, and sampling can be reliably performed once within a desired period (window). It also becomes. Further, it becomes easy to set numerical values for each periodic element of the waiting time calculation table TBL for irregularly changing the sampling period in such a manner.

(Other embodiments)
If the sampling period is irregularly changed as described above, it is possible to avoid synchronization between the pulsation period of the analog signal to be sampled and the sampling period. However, if it is known in advance that such synchronization may occur only under certain circumstances, the above synchronization can be achieved only by changing the sampling period irregularly only in situations where synchronization may occur. Can be avoided. Therefore, assuming that the sampling period is fixed to the reference sampling period, it is determined whether or not the current sampling period is in a state of being synchronized with the pulsation period of the sensor signal, and only when it is determined that the state is in a synchronized state. The sampling period may be changed irregularly, and if not, it may be fixed uniformly at the reference sampling period. In the case of the above embodiment, since the engine rotation speed at which the pulsation period of the sensor signal of the VG sensor 10 synchronizes with the reference sampling period (4 ms) is known in advance, the sampling period is irregularly changed only near the rotation speed. It is better to let them.

    On the other hand, if the generation state of the pulsation of the sensor signal changes depending on the situation, the pattern of the irregular change of the sampling period that can appropriately suppress the adverse effect of the pulsation of the sensor signal on the sampling result as described above also changes. In other words, if the pulsation pattern (cycle) changes depending on the situation, it is desirable to change the pattern of irregular change in the sampling cycle. However, in the above embodiment, the irregular change pattern of the sampling period is uniformly determined by the arrangement of the periodic elements in the waiting time calculation table TBL. Therefore, if a plurality of waiting time calculation tables TBL having different arrangements of periodic elements are provided and the table used for setting the sampling period is switched according to the situation, the irregular change of the sampling period is changed according to the change of the pulsation pattern. The pattern can be changed.

    For example, the synchronization between the pulsation period of the sensor signal of the VG sensor 10 and the reference sampling period may occur in a plurality of rotation speed regions. Since the pulsation cycle of the sensor signal of the VG sensor 10 is inversely proportional to the engine rotation speed, the pulsation cycle of the sensor signal differs in each of the plurality of rotation speed ranges. Therefore, even if the irregular change pattern of a specific sampling period can effectively suppress the adverse effects of pulsation in any of the rotation speed ranges where such synchronization occurs, other rotation speed ranges But it can't always be guaranteed that it will work in the same way. Therefore, a plurality of waiting time calculation tables that are set so that the sampling period irregularly changes with an effective pattern are individually provided for each rotation speed region where synchronization occurs. If the table used for setting the sampling period in each rotation speed region is switched, the adverse effect of pulsation can be effectively suppressed in any rotation speed region where synchronization occurs. More specifically, the engine rotation speed is acquired before the process of step S13 in the A / D conversion activation process of FIG. 5, and the waiting time WT is calculated in step S13 according to the acquired engine rotation speed. A process for selecting a waiting time table to be used is added. In this case, the process added here corresponds to the process of the table switching means.

    On the other hand, as described above, the engine rotation speed at which the pulsation cycle of the sensor signal of the VG sensor 10 and the reference sampling cycle are synchronized differs depending on the number of cylinders of the engine. As is clear from this, the occurrence of pulsation changes depending on the number of cylinders of the engine to be applied. Therefore, by providing a waiting time calculation table TBL with different arrangement patterns of periodic elements for each number of engine cylinders and switching the table used for setting the sampling period according to the number of engine cylinders to be applied, the number of cylinders can be changed. The signal processing apparatus can be provided with versatility that can be commonly applied to different engines.

    The sampling cycle setting method and the electronic processing apparatus adopting the setting method according to the present invention are not limited to the above-described embodiments, and can be further implemented, for example, as follows.

  The number of VG sensors 10 is arbitrary, and a plurality of VG sensors may be prepared, and these VG sensors may be A / D converted at a sampling period Ts_n at a time in the A / D converter 54. In this way, resource saving of the vehicle control device 1 can be achieved through reduction of the storage area of the ROM 32 and reduction of the processing load of the microcomputer.

    In the above embodiment, the microcomputer 30 sends a conversion start signal to the A / D converter 54 to execute A / D conversion after the waiting time WT determined from the waiting time calculation table TBL has elapsed. It was output. Thus, the method for executing the A / D conversion after the waiting time WT elapses is not limited to the method using the above-described waiting time calculation table TBL. For example, the value of the waiting time WT is embedded in the program, and the microcomputer 30 performs the above conversion to perform A / D conversion after the waiting time WT elapses by a conditional branch every time the A / D conversion starting process is executed. A start signal may be output. That is, in this case as well, the waiting time calculation table TBL does not exist in an independent form, but is only embedded in the processing program of the microcomputer 30, and is substantially the same as the above embodiment. It is.

    In the above embodiment, the difference between the XIth periodic element TBL [XI] arranged in the waiting time calculation table TBL and the (XI + 1) th periodic element TBL [XI] is used as the reference sampling period. The setting is performed so that the added value becomes the sampling period. However, when it is not necessary to match the average value (expected value) of the sampling period to a specific value (reference sampling period), the setting of the sampling period using the periodic element obtained from the waiting time calculation table, You may make it perform by another calculation aspect. In short, if the sampling period is obtained through a calculation using the periodic elements obtained in the order of arrangement from a table in which the periodic elements are arranged in random order, there is no room for contingent elements. Programmed irregularities can be added to the sampling period. For example, the periodic element acquired from the table can be used as a coefficient, and a product obtained by multiplying a fixed value can be set as the sampling period, or the acquired periodic element itself can be set as the sampling period.

    The present invention is not limited to the signal processing device that performs A / D conversion of the VG sensor in the in-vehicle control device, and the sampling period setting method for sampling related to the A / D conversion, and the pulsation The present invention can be applied to all devices that perform A / D conversion of an analog signal whose period varies, or general setting of a sampling period when sampling such an analog signal.

1 is a block diagram showing a schematic configuration of a vehicle control device applied to an embodiment of the present invention. The graph which shows an example of the sensor output of a VG sensor. The figure which shows an example of the waiting time calculation table memorize | stored in ROM of the vehicle control apparatus in the same embodiment. The figure which shows each value under the contrast with the execution time of a microcomputer, making the periodic element arranged in the waiting time calculation table into waiting time. The flowchart of the A / D conversion starting process which the embodiment employ | adopts. The graph which shows the sampling timing by a microcomputer together with transition of a sensor signal. The flowchart of the A / D conversion completion process which the embodiment employ | adopts. The flowchart of the instantaneous flow rate average value calculation process in the vehicle control apparatus to which the embodiment is applied. The graph which shows the sampling timing by a microcomputer together with transition of a sensor signal.

Explanation of symbols

    DESCRIPTION OF SYMBOLS 1 ... Vehicle control apparatus, 10 ... Hot wire type air flow meter (VG sensor), 11 ... Fuel temperature sensor, 12 ... Water temperature sensor, 13 ... Intake temperature sensor, 14 ... Oxygen sensor, 15 ... Rotation angle sensor, 16 ... Vapor flow sensor , 17 ... Idle switch, 20 ... Igniter, 21 ... Fuel injection valve, 22 ... Idle speed control valve, 23 ... Purge control valve, 24 ... Exhaust gas recirculation valve, 30 ... Microcomputer, 31 ... Central processing unit, 32 ... ROM, 33 ... RAM, 34 ... Backup RAM, 35 ... Input / output port, 36 ... Input port, 37-41 ... Output port, 50-53,55 ... Buffer, 54 ... A / D converter, 56 ... Comparator, 57 ... Waveform shaping circuit, 60 to 64 ... Drive circuit.

Claims (17)

A method of setting a sampling period when sampling an analog signal,
Create a table in which multiple periodic elements are arranged in random order in advance,
A method for setting a sampling period, comprising: acquiring periodic elements from the table in the order of arrangement for each sampling, and individually setting the sampling period for each sampling according to the acquired periodic elements. The sampling period setting method according to claim 1, wherein the sampling period is set to be a sum of a variable component obtained according to the periodic element acquired from the table and a reference sampling period that is a fixed value. When the periodic element arranged in the table in the table is “TBL [i]” and the index incremented every sampling is “XI”, the XI-th component in the table is set to the reference sampling cycle which is a fixed value. 2. The sampling according to claim 1, wherein the sampling period is set to be a value obtained by adding a difference between the periodic element TBL [XI] arranged in the first and the (XI + 1) th arranged periodic element TBL [XI + 1]. How to set the cycle. The sampling period setting method according to claim 3, wherein a range of possible values of each periodic element arranged in the table is set to "0" or more and less than the reference sampling period. When the sampling period is fixed to the reference sampling period, it is determined whether or not the sampling period and the pulsation period of the analog signal are in synchronization. When it is determined that the sampling period is in synchronization, the variable The sampling period setting method according to claim 2, wherein a sum of a component and the reference sampling period is set as a sampling period, and if not, a sampling period is set as the reference sampling period. The sampling period according to any one of claims 1 to 5, wherein a plurality of tables having different arrangement modes of periodic elements are prepared as the table, and the table used for setting the sampling period is switched according to a sampling execution state. Setting method. The sampling period setting method according to any one of claims 1 to 6, wherein the periodic elements arranged in the table are obtained through verification by a simulation environment or a compatible environment. The sampling period setting method according to any one of claims 1 to 7, wherein the sampling period setting method is applied to sampling of a sensor signal of an intake air amount detection sensor of an engine. In a signal processing apparatus that periodically samples an analog signal and converts it into a digital signal,
A table in which a plurality of periodic elements are arranged in random order;
Sampling period setting means for acquiring periodic elements from the table for each sampling in the order of arrangement and individually setting the sampling period for each sampling according to the acquired periodic elements;
A signal processing apparatus comprising: The sampling period setting unit performs the individual setting of the sampling period so as to be a sum of a variable component obtained according to a periodic element acquired from the table and a reference sampling period which is a fixed value. Signal processing device. When the periodic element arranged in the table in the table is “TBL [i]” and the index incremented every sampling is “XI”, the sampling period setting means sets the reference sampling period to a fixed value. The sampling period is set to a value obtained by adding the difference between the XIth periodic element TBL [XI] and the (XI + 1) th periodic element TBL [XI + 1] arranged in the table. The signal processing device according to claim 10. The signal processing apparatus according to claim 11, wherein a range of values that each periodic element of the table can take is set to be “0” or more and less than the reference sampling period. In the signal processing device according to any one of claims 10 to 12,
When the sampling period is fixed to the reference sampling period, it comprises a determination means for determining whether the fixed sampling period and the pulsation period of the analog signal are in a synchronized state,
The sampling period setting means individually sets the sampling period for each sampling so as to be the sum of the variable component and the reference sampling period when it is determined that the determination means is in the synchronized state; In some cases, the signal processing apparatus is characterized in that the sampling period is uniformly fixed to the reference sampling period. The signal processing device according to any one of claims 10 to 13,
As the table, a plurality of tables having different arrangement modes of periodic elements are provided,
A signal processing apparatus comprising: a table switching unit that switches a table used by the sampling cycle setting unit for setting the sampling cycle according to a sampling execution state. The analog signal is a sensor signal of an engine intake air amount detection sensor,
The signal processing device according to claim 14, wherein the table switching unit switches the table according to the number of cylinders of the engine to be applied. The analog signal is a sensor signal of an engine intake air amount detection sensor,
The signal processing device according to claim 14, wherein the table switching unit performs switching of the table according to a rotation speed range of the engine. The signal processing device according to any one of claims 9 to 14, wherein the signal processing device is used for analog / digital conversion of a sensor signal of a sensor for detecting an intake air amount of an engine.

Images