JPH0311415B2 - - Google Patents

Description

[Detailed description of the invention] The present invention relates to a method for calculating remaining fuel amount.

While driving a car, it is often necessary to predict the possible travel distance from the current remaining fuel amount and fuel consumption rate.
However, conventional fuel gauges can only provide approximate values. Conventional fuel gauges mainly measure the height of the liquid level using a float in the fuel tank and display the remaining amount of fuel in an analog manner, so even if the accuracy is low, it is sufficient. In systems that accurately measure the amount of remaining fuel and display it digitally, the sampled value fluctuates irregularly due to vibrations, tilting, etc. of the vehicle body. For this reason, the display is stabilized, for example, by calculating the average value over a certain period of time. However, even this method is not satisfactory for the purpose of predicting travel distance. This is because the amount of remaining fuel calculated using this method is the moving time average of the decreasing amount, so there was a drawback that it was always higher than the current value.

The present invention provides a method for calculating the amount of remaining fuel that improves on these drawbacks.

In electronically controlled fuel injection (EFI),
The amount of fuel injected is controlled by the opening time of the injection nozzle. That is, the injection nozzle has only two values, open and closed, and there is no intermediate state.

Therefore, the amount of fuel consumed can be obtained by integrating the opening time of the injection nozzle.

The present invention focuses on this point and obtains a signal proportional to fuel consumption by integrating the pulse duration of the electrical signal (τ signal) of the opening time of the injection nozzle of the electronically controlled fuel injection device (EFI). It is something.
Figure 1 shows the EFI injection nozzle opening τ signal and this τ
The consumption pulse waveform obtained by ANDing with the clock pulse to measure the duration of the signal is shown. Here, the clock pulse is a signal having a period short enough to measure the pulse width of the τ signal.

Since the fuel consumption amount is proportional to the integrated value of the pulse width of the τ signal, and the pulse width of the τ signal is proportional to the number of consumption pulses, the fuel consumption amount can be obtained by integrating the number of consumption pulses.

Since a car repeatedly stops and starts, the correlation between the value of the remaining fuel level sensor and mere elapsed time is not very good, but the value of the remaining fuel level sensor and the integrated value of the number of consumption pulses mentioned above have a good linear correlation. show.

The present invention provides a method of estimating the remaining fuel amount sensor value and the integrated value of the consumption pulse number using linear regression to calculate the current remaining amount of fuel.

According to the present invention, the analog value of the remaining fuel amount sensor is digitally converted and inputted to the arithmetic and control unit CPU, while the consumption pulses are integrated, and the output for each constant amount of consumption is linearly correlated with the output of the remaining fuel amount sensor. A fuel remaining amount calculation method is provided in which formulas are sequentially determined and the current remaining fuel amount is calculated.

According to the fuel remaining amount calculation method according to the present invention, the gradually decreasing remaining fuel amount is not determined by a simple moving average of the past, but the remaining fuel amount is determined by τ
The current amount of fuel remaining is estimated by calculating the regression with the fuel consumption calculated from the signal, and it is not a mere temporal moving average, so the current remaining amount can be determined.

DESCRIPTION OF THE PREFERRED EMBODIMENTS The method for calculating the remaining amount of fuel according to the present invention will be explained with reference to the drawings. FIG. 2 shows its configuration diagram. An analog output regarding the remaining fuel amount from the fuel remaining amount sensor 1 is A/D converted by an A/D converter 2 and inputted to an arithmetic and control device 3. On the other hand, the consumption pulse is obtained by ANDing the τ signal from the EFI fuel injection electric signal (τ signal) generator 4 and the clock signal, and the pulse is integrated by the counter 5. The counter 5 performs calculations every time a certain amount is reached. A consumption integration signal i is output to the control device 3. The arithmetic and control unit 3 reads the remaining fuel amount data Y _i of the remaining fuel amount sensor 1 through the A/D converter 2 every time this consumption amount accumulation signal i is input, and reads this consumption amount accumulation signal i and the remaining fuel amount data. Y _i is sequentially recorded in the random access memory RAM6. The arithmetic and control unit 3 receives the consumption integration signal i and the remaining fuel amount data.
Using the least squares method from the value of Y _i , coefficients a and b of a linear regression line y=a·i+b that approximates the relationship between the consumption integrated signal i and the remaining fuel amount data Y _i are determined. The consumption integrated signal i output by the consumption pulse counter 5, the remaining fuel amount data Y i obtained from the remaining fuel amount sensor ₁ , and 1 determined from this i and Y _i
The relationship of the following regression line y=a·i+b is shown in FIG. In FIG. 3, the horizontal axis indicates fuel consumption. Since the consumption accumulation signal i is output at fixed consumption intervals, if the number of outputs of i is plotted on the horizontal axis, it will be at fixed intervals as shown in FIG. Further, the vertical axis indicates the remaining amount of fuel. The value of the remaining fuel amount data Y _i obtained from the remaining fuel amount sensor 1 for the consumption integrated signal i is calculated by the linear regression line y=
Distributed above and below a・i+b. From the linear regression line thus obtained, the calculated value y _o corresponding to the current time (i=N) is obtained as the current fuel remaining amount and displayed on the display 7. Similarly, in the same way as sequentially calculating the moving average, determine the total number of consumption integration signals i for performing the linear regression calculation, and when the latest i・Y _i is input,
The oldest i.Y _i is deleted, linear regression calculations are performed sequentially, and the current remaining fuel amount y is determined and displayed on the display 7.

In this case, the variance around the linear regression line is calculated, and the standard deviation σ is determined. Limits such as ±2σ and ±3σ are set to exclude abnormal values of the value Y _i of the fuel remaining amount sensor.

According to the fuel remaining amount calculation method of the present invention, instead of using the inaccurate value Y _i of the fuel remaining amount sensor as it is, it is calculated based on the correlation with fuel consumption from the EFI τ signal.
Since the following regression line is obtained and the current remaining fuel amount y is obtained from this, the remaining fuel amount can be determined with high accuracy. If the average fuel consumption rate Km/ of this car is known, it is possible to estimate the drivable distance from the present moment onwards with considerable accuracy based on this average fuel consumption rate and the calculated remaining fuel amount y.

[Brief explanation of the drawing]

Fig. 1 is a consumption pulse waveform diagram shown by ANDing the EFI τ signal and the clock pulse, Fig. 2 is a configuration diagram of a device related to the method for calculating the remaining fuel amount according to the present invention, and Fig. 3 is an integration of the τ signal. A correlation diagram showing the distribution of pulse i for each fixed amount and fuel gauge value Y _i is shown. In the figure, 1 is the fuel remaining amount sensor, 2 is the A/D
Converter, 3 is an arithmetic and control unit, 4 is a fuel injection electric signal generator, 5 is a counter, 6 is a RAM, and 7 is a display.

Claims (1)

[Claims] 1. Electrical signals of the opening time of the injection nozzle in an electronically controlled fuel injection device are integrated to obtain an integrated consumption signal i every time a certain amount of fuel is consumed, and this integrated consumption signal i is obtained. Remaining fuel amount data Y _i is obtained from the remaining fuel amount sensor at each time, and the above consumption integration signal i and the remaining fuel amount data are
Find the coefficients a and b of the linear regression line y=a・i+b using Y _i , and use this linear regression line y=a・i+b to find the remaining fuel amount _{y o} at the current consumption integrated signal i=N. Characteristic fuel remaining amount calculation method. 2 Remaining fuel amount data corresponding to consumption integration signal i
When finding the linear regression line y from Y _i , y=a・i
Calculate the variance of the remaining fuel amount data Y _i around +b, find the limit for adopting the remaining fuel amount data Y _i from the standard deviation value, and calculate the remaining fuel amount data that is outside the limit.
2. The fuel remaining amount calculation method according to claim 1, characterized in that the value of Y _i is excluded.