JPH02120918A - Method for linearizing nonlinear input - Google Patents

Abstract

PURPOSE:To obtain sufficient linearization accuracy in accordance with an input by connecting the optional number of linearizing tables in a memory area to select the number of break points. CONSTITUTION:When there are 6 tables #1 to #6 e.g., each table has 8 break points and the tables #1, #2 and #5 are connected, the table obtained by connecting the tables #1, #2, #5 has sections Ka to Kc corresponding to inputs Xa to Xc. Since each table has 8 break points 24 break points included in the 3 tables can be utilized, so that a more accurate linearizing method can be obtained. Consequently, the number of tables to be connected can be selected in accordance with an input and highly accurate linearization corresponding to the input can be simply attained.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a linear rice method for linearizing nonlinear input.

[Conventional technology]

There are various methods for linearizing nonlinear input, but recently there has been an increase in the number of methods that use software to perform linearization calculations. Among them, one of the methods that is relatively simple and has a short execution time is the table method. This prepares a memory area to store the linearization table,
In the linearization table above, set a certain number of breakpoints to linearize the nonlinear input,
Linearization is performed by performing linear interpolation calculations between corner points. Normally, linearization is achieved by using several such tables, setting each break point data, and specifying which table to use.

[Problem to be solved by the invention]

However, in the conventional method, since only one table can be selected, the accuracy of linearization is determined by the number of break points that table has, and depending on the input, sufficient linearization accuracy may not be obtained. Ta.

[Means to solve the problem]

In order to solve such problems, the present invention has a plurality of memory areas that store linearization tables for linearizing nonlinear input, and stores an arbitrary number of memory areas among the plurality of memory areas. The number of break points can be selected by connecting linearization tables.

[Operation] In the linear rice method for nonlinear input according to the present invention,
The tables are connected.

〔Example〕

Conventionally, the linear rice method for nonlinear input according to the present invention is
Previously, only one table could be selected per person, but by providing a function that allows tables to be connected (chained), it is now possible to select multiple tables for one input, and the characteristics of the input It is now possible to obtain the optimal linearization characteristics according to the Note that to select and specify a table, you only need to specify one of the connected tables, and then the system program itself determines whether or not the tables are connected, and if they are connected, Accurate linearization characteristics can be obtained by performing a linear calculation over the connected chifurs.

Next, an embodiment of the nonlinear input linearization method according to the present invention will be described in the case where the number of tables is 6, #1 to #6. It is assumed that each table has eight break points as shown in Table 1. This method has a function to connect each table, so for example, Table 2
As shown in #1. A case where #2 and #5 are connected will be explained. Note that #0 means that it is not connected to any table.

Figure 1 is an image diagram when tables are connected.
The horizontal axis is input X, and the vertical axis is output Y. X and Y are, for example, a voltage value and a flow rate value. Each chiffle has 8 break points, so #1 #2. The table that connects #5 shows the sections Ka, Kb, corresponding to the human power Xa, Xb, and Xc.
It becomes a chiful with Kc. As a result, in the conventional linear laser method, the number of break points in one table is 8.
Previously, only 24 break points of the three chifurs could be used, making it possible to perform a more accurate linear rice method.

FIG. 2 is a graph showing the -order compensation operation, and Y is calculated by the following formula.

・ ・ ・ ・ ・ (1) FIG. 3 is a flowchart for explaining a method for obtaining input X for −order interpolation calculation.

First, assume an interval (Step 1). The interval assumption is
In FIG. 1, the section Ka where the human power X is small may be changed to sections Kb and Kc where the human power X is large, and the section Kc where the input X is large may be changed to the sections Kb and Ka where the input X is small. First, an interval Ka is assumed, and it is determined whether the input X is within this interval. As shown in Fig. 1, if input X is between human power X2 and X3, an interval is determined for Ka (
Step 2). If the input X is in the interval Kc, the interval is 6'
Since the positive determination is not made, the process moves to step 3, where it is determined whether or not there is a connected chifur (step 3), and if there is, the section is assumed again. After the section is determined, the position of input X within the section is then determined. This may be determined, for example, in the same manner as the interval. That is, the input value may be determined from the minimum input value within the definite interval to the maximum input value, or from the maximum input value to the minimum input value. Position of human power X, e.g. 2nd position
As shown in the figure, when the distance between Xn and X(n-1) is determined, -order interpolation calculation is performed (step 4). If the interval is not determined and there is no concatenation table, out-of-interval processing is performed (step 2.3).

〔Effect of the invention〕

As explained above, the present invention has a plurality of memory areas that store linearization tables for linearizing nonlinear input, and stores linearization tables in any number of memory areas among the plurality of memory areas. By connecting and selecting the number of break points, the number of tables to be connected can be selected according to the input, so highly accurate linearization according to the input can be achieved with the same ease as before. Furthermore, since the table memory area can be used effectively, there is an effect that the degree of freedom is increased, such as the use of high-precision and low-precision data, and their combined use.

[Brief explanation of drawings]

Fig. 1 is an image diagram of a temporary table connection to explain an example of the linearization method of nonlinear input according to the present invention, Fig. 2 is a graph to explain linear interpolation calculation, and Fig. 3 is a diagram showing the determination of intervals and linearization. It is a flowchart for explaining interpolation calculation. output pY 7flY

Claims (1)

[Claims] It has multiple memory areas for storing linearization tables for linearizing nonlinear input, and selects the number of breakpoints by connecting linearization tables of an arbitrary number of memory areas among the multiple memory areas. A linear rice method for nonlinear input that is characterized by the following.