JPH01311218A - Analytic data processor - Google Patents

Abstract

PURPOSE:To acquire spectrum data with high accuracy by setting plural points on a background assumed with a measurement output. CONSTITUTION:Parameters required to carry out algorithm for determining the background from the measurement data are set, a data graph 3B of principal points on a background curve is searched by the algorithm, and searched data are connected to generate a polygonal background line. Then the polygonal background line is smoothed by a proper smoothing method to generate a smooth curved background function graph 3C. Then the background value is subtracted from the measurement data and the result is used to draw a spectrum as shown by a graph 3D. Consequently, the values of only peaks which project are displayed clearly, so the peak are easily analyzed and the high- accuracy spectrum data is obtained.

Description

[Detailed description of the invention] (Industrial application field) The present invention relates to an analytical data processing device.

(Conventional Ik method) In normal spectroscopic analysis or chromatography, measurement records include background, and by removing this background from the measurement data, the actual spectrum containing only true analytical information is Displaying the data is very effective in analyzing measurement data.

To remove background in a data processing device, a method is used in which a background function is set and subtracted from raw measurement data. Conventionally, the method of setting a background function at the knee is to visualize the raw measurement data, assume an appropriate background, specify some appropriate points on the background line, and create a polygonal background function. Because of this assumption, there was a problem in that the background would be quite different from the actual one, and the obtained spectrum would be quite different from the actual spectrum.

(Problems to be Solved by the Invention) An object of the present invention is to make it possible to obtain highly accurate real spectrum data.

(Means for Solving the Problem) 1l11 constant data from an analyzer is obtained in the form of expressing measurement output as a function of wavelength and time as independent variables. For convenience, the following description will be made with independent variables such as wavelength and time on the y-axis and the measurement output on the y-axis.

In the analytical data processing device, when a measurement output y is obtained for an independent variable 4 means for calculating Y, a function for calculating a function of x, Y representing a 1L smooth curve that connects the plurality of sets of x, Y values based on the plurality of sets of x, )′ given above, and the memory. Read each x and y of the measurement data from the contents of the measurement data, calculate the Y value of the "-" function for the X value of the same measurement data, and remove the function Y value calculated from each y value of the measurement data. It is equipped with an arithmetic control means having V1 function.

(For 11゛) The device of the present invention has a function of creating a smooth curve that connects a plurality of set points. , the background can be removed by subtraction processing.

The present invention focuses on the fact that the error becomes smaller by averaging it, and attempts to remove the setting error included in the background setting by correcting the background curve from a polygonal line to a smooth curve. be. That is,
When setting multiple points on the background, there is a setting error, but by creating a smooth curve that connects the set points up and down, the above setting errors are averaged and removed. Any method can be used to create a smooth curve connecting the points of , and various methods can be used.

Also, multiple sets of X used for background functions.

Although the method for selecting the value of y is arbitrary, it is possible to set a background function by assuming a background on raw measurement data and specifying multiple points on the assumed background. It is also possible to automatically select the minimum point of the spectral curve by using a computer.

(Example) FIG. 1 shows an example of the present invention.
2 is an analyzer, and 2 is an interface for communicating signals between the analyzer 1 and the CPU 3. The CPU 3 controls the analyzer 1 and processes the measured data. 4 is a memory that stores the measured data. 5 is a display device that displays the measured data and results. 6 is a computer that displays various data and commands.
This is a keyboard for inputting to PU3.

FIG. 2 shows a flowchart of data processing.

Set the parameters necessary to execute the algorithm that determines the background from the measured data (a), and use this algorithm to find the data at the main points on the background curve [see Figure 3 C9 (b), Create a polygonal background line by connecting the retrieved data.

Smoothing the polygonal background line into a smooth curve using an appropriate smoothing method, such as the least S-multiplication method, creates a smooth curved background function [Fig. (1) Subtract the above background value from the measurement data (e).

When a spectrum is drawn using the subtracted data, it becomes as shown in the third diagram, where only the peaks stand out and the sizes of the peaks are clearly displayed, making it easy to analyze the peaks.

An example of a method for determining the background will be described.

First, the minimum point is found from the measured data. This is detected as the point where the sign of the difference between two adjacent measured values is reversed from negative to positive. Next, the peaks on both sides of one minimum value are detected. If the interval between the two is narrow and the peak values on both sides are high, the minimum value can be seen as the tails of the peaks on both sides overlapping and being higher than the background level, so such minimum points are excluded. . Therefore, h/d is calculated from the interval d between the peaks on both sides of the -* local minimum value and the ratio of the remaining peak heights obtained by subtracting the local minimum value from the peak heights on both sides, and this calculated h/d When exceeds a preset value, that minimum point is excluded.

The above-mentioned value of 2N is the parameter set in step (a) in the flowchart of FIG.

Finally, connect the remaining minimum points as shown above to create a single broken line. This is part of the flowchart in Figure 2.
The background curve is a broken line of steps.

As an example of smoothing, when the number of minimum points set as described above is 4 or more, which is relatively small, an n-1 degree polynomial is assumed for the number n of points, and n −1 coefficients are determined from the data of n points using the least squares method.

Since the final purpose of the present invention is to obtain a smooth curved background function with few errors, the method itself is arbitrary. It is also possible to obtain a smooth curved background with little error by drawing a background curve on the display screen using a light ben or similar device based on the observer's experience, and reading the drawn background data with a computer. be.

(Effects of the Invention) According to the present invention, it has become possible to more accurately display only the peaks due to the characteristics of the sample, making analysis accurate and easy.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of an embodiment of the present invention, Fig. 2 is a flowchart of data processing, and Fig. 3 is a spectral diagram showing the data processing process. is the spectral diagram of the base point of the found peak, and is the same I
AC is the background diagram after smoothing processing, same L
21D is a spectrum diagram of the measurement data after background removal.

Claims (1)

[Claims] When a measurement output y is obtained for an independent variable x, a memory for storing the measurement data x, y, and a plurality of sets of x, Y are provided as a background for the measurement data x, y. means, a function for calculating an x, Y function representing a smooth curve connecting the plurality of sets of x, Y values based on the plurality of sets of x, Y given above, and a function for calculating the above function from the contents of the memory. An analytical data processing device comprising: arithmetic control means having a function of calculating a value obtained by removing a background from each y value of measurement data by subtracting it as a background.