JP2001101156A - System for automatically estimating optimum simulation parameter - Google Patents

Abstract

PROBLEM TO BE SOLVED: To objectively judge the agreement between raw data and simulated data. SOLUTION: A system provided with a comparing data display window 1, a parameter inputting dialog window 2 and a simulation calculation module 3 is also provided with a parameter variation setting dialog window 4 for setting up a start valued, variable step width and an end value and preparing a parameter variation table and a data comparing module 5 for evaluating similarity between a simulation calculation result and comparing data and constituted so as to estimate an optimum simulation parameter by performing simulation calculation for all conditions of the parameter variation table and evaluating similarity between the calculation result and the comparing data.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to electron spin resonance (E).
The present invention relates to a system for automatically estimating an optimal simulation parameter applicable to a measurement device or the like.

[0002]

2. Description of the Related Art Conventionally, in order to perform a structural analysis of an object to be measured from data obtained by ESR or the like, a simulation program corresponding to the object to be measured such as a solid, a liquid, and a gas is prepared, and many parameter values are set. The simulation calculation was performed by changing the data, the raw data was compared with the data obtained by the simulation calculation, an appropriate parameter set having a high similarity was determined, and the structural analysis was performed.

The simulation program includes, for example, parameters such as hyperfine coupling constant, nuclear spin number, nuclear spin type, total radical number, total radical type, central magnetic field,
A predetermined function calculation (simulation calculation) based on parameters such as a magnetic field sweep width, a simulation function for calculating each peak of a waveform, for example, a contribution ratio (weight) of a Gaussian function, a Lorentz function, and a line width at the time of peak calculation. Is a program that is used to select and determine parameters by comparing the obtained waveform with the waveform obtained by measurement, and to perform a structural analysis from the determined parameters. . At this time, different functions are used as functions for generating a waveform depending on a solid, a liquid, a gas, and the like.

FIG. 3 is a diagram for explaining a parameter estimation system using a conventional simulation program.
In the figure, reference numeral 1 denotes a comparison data display window,
The comparison data (raw data) A obtained by the measurement and the data B obtained by the simulation calculation are displayed.
Reference numeral 2 denotes a dialog window for inputting parameters. On this screen, the values of parameters 1 to n are set. The simulation calculation module 3 performs a simulation calculation based on the parameters set in the parameter dialog window 2 and outputs a calculation result B to the comparison data window 1. The screen transition between the comparison data display window 1 and the parameter input dialog window 2 is performed by well-known screen switching or multi-window display. While frequently changing each parameter value in the parameter input dialog window 2, the user compares the raw data displayed on the screen with the simulation data to determine an appropriate parameter.

[0005]

As described above, in the conventional parameter estimation system, the raw data and the simulation data displayed on the screen are changed while frequently changing each parameter value in the parameter input dialog window. The appropriate parameters are determined by comparison.
In fact, it relied on the user's expertise and intuition, and it was extremely difficult to completely match the simulation data with the raw data. In addition, confirmation of the degree of coincidence relies on visual observation on a screen, which is not supported by numerical values, and there is a possibility that a possibility of existence of a more appropriate parameter set may be overlooked because an assumption is made.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and objectively determines the consistency between raw data and simulation data to prevent leakage of an optimal parameter set. The purpose is to be able to calculate automatically up to.

[0007]

According to the present invention, a simulation calculation is performed using a comparison data display window, a parameter input dialog window for inputting a plurality of parameters, and the input parameters, and a calculation result is obtained. Parameter estimating system comprising: a simulation calculation module that outputs a parameter to a comparison data display window, and further sets a start value, a variable step width, and an end value for each parameter, and creates a parameter variable table. A dialog window and a data comparison module that evaluates the similarity between the result calculated by the simulation calculation module and the comparison data are provided, and the simulation calculation is performed for all the conditions of the parameter variable table,
The method is characterized in that a similarity between a simulation calculation result and comparison data is evaluated to estimate an optimal simulation parameter.

Further, the present invention is characterized in that parameter sets are listed in order from the one with the highest similarity, and a dialog window for displaying the conditions and data for each parameter set is provided.

[0009]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a system for automatically estimating optimum simulation parameters according to the present invention. The same reference numerals as those in FIG. The comparison data display window 1 is the raw data A for comparison.
And a calculation result B obtained by the simulation calculation. The parameter input dialog window 2 is a screen that transits from the window 1 or is a window displayed over the window 1 for setting and inputting the parameters 1 to n used for the simulation calculation. When each parameter is double-clicked on this screen, a parameter change setting dialog window 4 is displayed. When a start value, a step width, and an end value of each parameter are set on this screen, all possible values are set for the parameter change. Created as a table. This table contains parameters 1 to
It may be created for each n, or one table may be created for parameters 1 to n. Each value of the parameter variable table created in this way is automatically taken into the simulation calculation module 3 and simulation calculation is performed. The simulation calculation module 3 performs a simulation calculation for all combinations of the values of the parameters 1 to n, that is, for all the conditions of the variable table.

The data evaluation module 5 determines the similarity between each data calculated by the simulation calculation module 3 and the raw data for comparison. A well-known method such as pattern matching is used to determine the similarity. Then, the parameter group that yielded the best result and its data are displayed in the comparison data display window 1. Note that when searching for the best parameter group, for example, a preliminary search is performed to roughly set the step width to roughly estimate the optimal range, and then a method of estimating the optimal parameter by reducing the step width is used. What is necessary is just to apply about a parameter.

The parameter sets obtained in this way are listed in order from the one with the highest similarity, and are displayed as a parameter list in the order of similarity 1, 2,... N as shown in FIG. Double-click to display the conditions and data at that time. By storing this parameter list in a storage device, it can be used for the next experiment. Also, when the test object is determined, the starting value, the step width, the end value, etc. for each parameter are determined appropriately by several simulations. By applying this to the calculation of, the calculation time can be greatly reduced.

[0012]

As described above, according to the present invention, it is possible to objectively determine the similarity between the data calculated by simulation and the raw data for comparison. Leakage can be prevented, and calculation can be automatically performed up to an optimum parameter set, so that parameters can be estimated very efficiently and labor can be saved.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a system for automatically estimating optimum simulation parameters according to the present invention.

FIG. 2 is a diagram illustrating a parameter list.

FIG. 3 is a diagram illustrating a parameter estimation system using a conventional simulation program.

[Explanation of symbols]

1 ... Comparison data display window, 2 ... Parameter input dialog window, 3 ... Simulation calculation module, 4 ... Parameter variable setting dialog window, 5 ... Data evaluation module.

Claims (2)

[Claims] 1. A comparison data display window, a parameter input dialog window for inputting a plurality of parameters, a simulation calculation is performed using the input parameters, and a calculation result is output to the comparison data display window. A simulation parameter estimating system including a simulation calculation module that performs a starting value, a variable step width,
A parameter change setting dialog window for setting an end value and creating a parameter change table; and a data comparison module for evaluating a similarity between the result calculated by the simulation calculation module and the comparison data. A simulation calculation is performed for all of the conditions, and a similarity between the simulation calculation result and the comparison data is evaluated to estimate an optimum simulation parameter. 2. A method according to claim 1, further comprising the step of listing parameter sets in descending order of similarity and providing a dialog window for displaying conditions and data for each parameter set. Estimation system.