JPS6236562B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a final value prediction circuit that predicts the final value of a measurement signal that changes with a first order lag time constant. More specifically, the present invention relates to a final value prediction circuit that predicts the final value of a measurement signal whose first-order lag time constant is not constant.

When measuring a signal that changes with a first-order lag time constant or performing control based on the measurement results, it is necessary to predict the final value of the measured signal in order to improve the responsiveness of measurement or control. be done. Conventionally, a first-order lead circuit with the same time constant was used to predict the final value, but since such a first-order lead circuit has a fixed time constant, it can be used only when the first-order lag time constant remains constant. can only be applied. The time constant of the primary advance circuit must be adaptively changed in response to a measurement value whose time constant varies, but no suitable means has been available to date to make this possible.

An object of the present invention is to provide a final value prediction circuit that can be applied to a measurement signal whose first-order lag time constant varies.

According to the present invention, the logarithm of the difference between the measured signal and its predicted final value is obtained, and the predicted final value is feedback-controlled so that the logarithm becomes a linear function of time.

Hereinafter, the present invention will be explained in detail with reference to the drawings. First, before explaining the apparatus according to the embodiment of the present invention, the principle of the present invention will be explained.

Measurement signal x varying with first-order lag time constant T
(t) can be expressed by the following equation.

x(t)=x _O e〓〓〓〓〓+x∞(1−e〓〓〓
〓) (1) However, X _O is the initial value when t=0, and x∞ is the final value when t=∞. If the relationship of equation (1) is illustrated, it will be as shown in FIG. 1, and it will be either curve a or b depending on the magnitude relationship between the initial value x _O and the final value x∞.

Now, for the case of curve a, the predicted final value is x
∞, and taking the natural logarithm of the difference between this and the measurement signal x(t), we get ln(x∞−x(t)) =ln{(x∞−x _O ) e -x
∞)} (2) Here, if x∞=x∞, equation (2) becomes ln(x∞−x(t))=−t/T+ln(x∞−x _O )( 3), and a linear function of time is obtained.

From this, conversely, if we can control the predicted final value x∞ so that the natural logarithm of the difference between it and the current value of the measured signal becomes a linear function of time, then
It can be said that the predicted final value at that time matches the actual final value.

On the other hand, regarding curve b, the equation equivalent to equation (2) is as follows: ln(x(t)−x∞)=ln{(x _O −x∞)e〓〓〓〓+( x∞−x
∞)} This equation is expressed as x∞ and x on the left side in equation (2).
(t), the first term x∞ and x _O on the right side, and the second term x∞ and x∞ are simply exchanged, so the curve a
The same can be said for the case of .

An embodiment of the present invention based on this principle is constructed as shown in FIG. In FIG. 2, 1 is a logarithmic conversion circuit, 2 is a second order differential circuit with respect to time, and 3 is a high gain amplifier circuit. The logarithmic conversion circuit 1 logarithmically converts the difference between the current value x(t) of the measurement signal and the output signal x∞ of the amplifier circuit 3, and this logarithmically converted value is second-order differentiated by the differentiator 2. The differential value by the differentiating circuit 2 is amplified by the amplifier circuit 3, fed back to the input of the logarithmic conversion circuit 1, and output as the predicted final value x∞ of the measurement signal. The polarity of the transfer function of the closed loop consisting of the logarithmic conversion circuit 1, the differentiation circuit 2, and the amplification circuit 3 is determined to be negative, and thereby the circuit system shown in Fig. 2 is considered to be a circuit system with self-balancing property. Ru.

When this circuit system is in equilibrium, the amplifier circuit 3
The input signal of is at its minimum value. Amplifier circuit 3
The gain of is sufficiently high that its input signal can be effectively considered as zero. Since the input signal of the amplifier circuit 3 is the output signal of the second-order differentiator circuit 2 with respect to time, the fact that this is zero means that the signal before second-order differentiation, that is, the output signal of the logarithmic conversion circuit 1, is the first-order time-related signal. It means that it is a function.

Therefore, based on the above-mentioned principle, the output signal x∞ of the amplifier circuit 3 in the balanced state is equal to the measurement signal x
This corresponds to the final value x∞ of (t). In other words, the predicted value of the final value x∞ has been obtained.

As shown in FIG. 1, the direction of change in the measurement signal x(t) can be either an increase or a decrease depending on the magnitude relationship between the initial value x _O and the final value x∞. In either case, the difference between the predicted final value and the current value of the measurement signal must be a signal of the same polarity (eg, positive polarity) in order to be able to determine the logarithm of the difference with the predicted final value. The predicted final value is larger than the current value when the direction of change of the measured signal is increasing, and is smaller than the current value when the direction of change of the measured signal is decreasing, so the difference in both cases has the same polarity. In order to do so, we must calculate the difference in one case in the opposite way to the other.

Furthermore, as the method for calculating the difference is reversed in this way, it is necessary to maintain the self-balanced property of the loop by reversing the polarity of the closed loop transfer function.

An embodiment with such means is shown in FIG. In FIG. 3, the same parts as in FIG. 2 are represented by the same symbols, but in addition, the input path for the difference signal and the output path for the amplified signal are each branched into two, and one path is connected to a polarity inverting circuit 4. , 5 are inserted, and two interlocking switches 6 and 7 are used to switch the route.
The switches 6 and 7 are switched by the comparator 8 based on the polarity of the time differential value of the measurement signal, that is, the direction of change of the measurement signal. If the direction of change is positive, the switches 6 and 7 are switched to the contact 1 side. However, if the direction of change is negative, the contact 2 side is turned on.

With such a circuit, when the measurement signal increases, i.e. x∞x(t), x∞−x
(t) is input to the logarithmic conversion circuit 1, and if the measurement signal decreases, that is, x∞x(t), x(t) - x∞ is input to the logarithmic conversion circuit 1, and in either case A signal of positive polarity is also inputted to the amplifier circuit 3, and the polarity of the output of the amplifier circuit 3 is switched in accordance with the change in the direction of action of x, thereby maintaining the negative feedback condition.

FIG. 4 shows a measured value processing circuit as an application example of such a final value prediction circuit. In Figure 4,
The input path of the measurement signal is branched into two, a final value prediction circuit 10 is inserted in one of them, and either the signal that has passed through the final value prediction circuit 10 or the signal that has not passed through the final value prediction circuit 10 can be selectively output by switching the switch 11. It is becoming more and more common. The switching of the switch 11 calculates the difference between the signal that has passed through the final value prediction circuit 10 and the signal that has not passed through the final value prediction circuit 10.
This is done by comparing the difference with the reference value V〓 by the comparator 12, and when the difference value is larger than the reference value V〓, the switch 11 is turned to the 2 side to select the predicted final value, and the difference is When the value of is smaller than the reference value V〓,
The switch 11 is turned to the 1 side to select the measured value itself. The reference value V〓 is set to a small value, so that when the current value of the measured value is far from the final value, the predicted final value is output, and the current value of the measured value matches or is close to the final value. When the value is reached, the measured value itself is output.

As described above, the present invention calculates the logarithm of the difference between the measured signal and its predicted final value, and performs feedback control on the predicted final value so that the logarithm becomes a linear function of time.

Therefore, according to the present invention, it is possible to obtain a final value prediction circuit that can be applied not only to measurement signals with a constant first-order lag time constant, but also to measurement signals with varying first-order lag time constants.

[Brief explanation of the drawing]

FIG. 1 is a graph showing the time change of a measurement signal having a first-order lag time constant, FIG. 2 is a conceptual block diagram of an embodiment of the present invention, and FIG. 3 is a conceptual diagram of another embodiment of the present invention. The block diagram, FIG. 4, is a conceptual block diagram of an application example of the present invention. 1... Logarithmic conversion circuit, 2... Second order differential circuit, 3
...Amplification circuit, 4, 5...Polarity inversion circuit, 6,7
...Switch, 8...Comparator.

Claims (1)

[Claims] 1. A logarithmic conversion circuit for calculating the logarithm of the difference between the current value of the measured signal and its predicted final value, and controlling the predicted final value based on the output signal of this logarithmic conversion circuit so that this output signal becomes a linear function of time. and a feedback control circuit for predicting a final value.