JP5726574B2 - Element analysis system, element analysis method, and element analysis program - Google Patents

Description

  The present invention relates to an element analysis system, an element analysis method, and an element analysis program for analyzing an element having an element structure composed of a plurality of element shapes.

  Today, devices for realizing various functions are being studied. In particular, MEMS (Micro Electro Mechanical Systems) are devices in which mechanical element parts, sensors, actuators, electronic circuits, and the like are integrated on a single silicon substrate, glass substrate, organic material, or the like. In such a MEMS, a mechanical component and an electronic component are often mixed.

  In order to evaluate the structure of such a MEMS, an analysis support system for generating an equivalent circuit for a structure composed of electrical components and mechanical components and analyzing the operation of the structure using the equivalent circuit is provided. (See Patent Document 1). In the technique described in this document, a structure is designed using components stored in a 3D model data storage unit of a simulation server. Then, the simulation server creates a machine connection graph and a dual graph using the geometric connection information. In addition, the dual graph is used to connect the machine terminals of another equivalent circuit, and the electrical connection graph is created using the geometric connection information to connect the electrical terminals of the individual equivalent circuit.

JP 2010-97475 A (first page, FIG. 1)

  When manufacturing elements having the same structure, manufacturing variations may occur depending on the elements. In particular, in MEMS, since a mechanical element is included, characteristic variation may occur in mechanical characteristics and electrical characteristics depending on the size and shape of the element. In this case, it is also possible to analyze the cause of characteristic variation by observing and comparing the structures of a plurality of elements using a scanning electron microscope (SEM). However, SEM observation takes time and effort. In order to observe the internal structure of the element, destructive inspection using a cross-sectional SEM or the like is necessary.

  The present invention provides an element analysis system, an element analysis method, and an element analysis program for efficiently evaluating each element shape in an element composed of a plurality of element shapes.

In order to solve the above-mentioned problems, the invention according to claim 1 is characterized in that measured value storage means for recording the output characteristics of an element having an element structure composed of element shapes for each element, and the element shape of the element structure. An element analysis system comprising model generation means for generating an equivalent circuit model represented by a model parameter converted into an equivalent circuit, and control means connected to the input / output means, wherein the control means includes the measured value For each element recorded in the storage means, for the model parameter of the equivalent circuit model generated by the model generation means, the measured value of the output characteristic corresponding to the element recorded in the measured value storage means corresponds to the element. by applying the equivalent circuit model, means for calculating the parameter values of the model parameters of the equivalent circuit model, the parameters for each of the model parameters Means for calculating the statistics, using the statistics, and summarized in that a section for outputting an evaluation result of the corresponding element shape model parameters.

According to a second aspect of the present invention, in the element analysis system according to the first aspect, the control means specifies an element shape in which the variation of the model parameter is larger than a reference value as an evaluation result of the element shape corresponding to the model parameter. And output it.

  The invention according to claim 3 is the element analysis system according to claim 1 or 2, wherein the measured value storage means records output characteristics of a plurality of elements formed on one substrate, The gist is that the control means calculates the in-plane variation on the substrate for each element shape from the output characteristics recorded in the measured value storage means, and outputs the element shape having a large in-plane variation.

  According to a fourth aspect of the present invention, in the element analysis system according to any one of the first to third aspects, when the control means acquires a new element structure, the element shape of the element structure is determined. The gist is to predict the output characteristics of an element having a new element structure using statistical values.

  According to a fifth aspect of the present invention, in the element analysis system according to the fourth aspect, the control means creates a parameter generation function for calculating an output value from a variation distribution of the statistical value of the parameter value, and the control means The gist is to predict the statistical value of the output characteristic of the element including the new element shape using the output value generated by this parameter generation function.

The invention according to claim 6 is an actual value storage means for recording the output characteristics of an element having an element structure composed of element shapes for each element, and model parameters obtained by converting the element shape of the element structure into an equivalent circuit. A method for analyzing an element using an element analysis system comprising model generation means for generating an equivalent circuit model and a control means connected to an input / output means, the control means comprising: For each element recorded in the actual value storage means, for the model parameter of the equivalent circuit model generated by the model generation means, the actual value of the output characteristic corresponding to the element recorded in the actual value storage means is stored in the element. by applying the corresponding equivalent circuit model, and calculating a parameter value of the model parameters of the equivalent circuit model, para for each of the model parameters Calculating a statistical value of over data value, using the statistics to be required to execute and outputting the evaluation result of the corresponding element shape model parameters.

The invention according to claim 7 is an actual value storage means for recording output characteristics of an element having an element structure composed of element shapes for each element, and model parameters obtained by converting the element shape of the element structure into an equivalent circuit. A program for analyzing an element using an element analysis system including a model generation unit that generates an equivalent circuit model and a control unit connected to an input / output unit, the control unit including: For each element recorded in the actual value storage means, for the model parameter of the equivalent circuit model generated by the model generation means, the actual value of the output characteristic corresponding to the element recorded in the actual value storage means is stored in the element. by applying the corresponding equivalent circuit model, it means for calculating the parameter values of the model parameters of the equivalent circuit model, for each of the model parameters It means for calculating a statistical value of the parameter values, using the statistical values, and summarized in that to function as means for outputting the evaluation result of the element shape corresponding to the model parameters.

(Function)
According to the first, sixth, and seventh aspects of the present invention, the control unit is configured to measure actual value storage means for the model parameters of the equivalent circuit model generated by the model generation unit for each element recorded in the actual measurement value storage unit. The parameter value of the model parameter of the equivalent circuit model is calculated by applying the actually measured value of the output characteristic corresponding to the element recorded in the above to the equivalent circuit model corresponding to the element . Next, a statistical value of the parameter value is calculated for each model parameter. Then, the evaluation result of the element shape corresponding to the model parameter is output using the statistical value. Thereby, the variation in element shape can be evaluated from the actually measured output characteristics, and this evaluation result can be provided.

According to the invention described in claim 2, the control means, as the evaluation result of the corresponding element shape model parameters, the variation of the model parameters, and outputs the identifying larger element shapes than the reference value. Thereby, it is possible to grasp element shapes having large variations.

  According to the invention described in claim 3, the control means calculates the in-plane variation on the substrate for each element shape from the output characteristics recorded in the actual value storage means, and selects the element shape having a large in-plane variation. Output. Thereby, the variation | correspondence corresponding to the position of the element arrange | positioned on a board | substrate can be grasped | ascertained.

  According to the invention described in claim 4, when the new element structure is acquired, the control means predicts the output characteristics of the element having the new element structure using the statistical value of the element shape constituting the element structure. . Accordingly, it is possible to predict the output characteristics of the element structure including the new element shape in consideration of the variation in the element shape evaluated based on the actually measured output characteristics.

  According to the fifth aspect of the present invention, the control means generates a parameter generation function for calculating an output value based on a variation distribution of the statistical values of the parameter values, and the control means outputs the output generated by the parameter generation function. Using the value, the statistical value of the output characteristic of the element including the new element shape is predicted. Thereby, it is possible to generate a variation state in a pseudo manner using the parameter generation function.

  According to the present invention, each element shape can be efficiently evaluated in an element composed of a plurality of element shapes.

The system schematic of embodiment of this invention. Explanatory drawing of the relationship with the element structure in the embodiment of this invention, an output characteristic, and an equivalent circuit. It is explanatory drawing of the data recorded on each memory | storage part of embodiment of this invention, Comprising: (a) is an actual measurement characteristic data memory | storage part, (b) is a parameter value memory | storage part, (c) is a statistical value data memory | storage part, (D) is explanatory drawing of the data recorded on the parameter production | generation part. Explanatory drawing of the process sequence of this embodiment.

  Hereinafter, an embodiment in which an element analysis system, an element analysis method, and an element analysis program of the present invention are embodied will be described with reference to FIGS. In the present embodiment, evaluation of the characteristic variation of the electromechanical energy conversion element (MEMS) and prediction of the output characteristic considering the variation are performed. For this purpose, an element analysis apparatus 20 shown in FIG. 1 is used as an element analysis system.

  As shown in FIG. 2, an element structure in which a plurality of element shapes (a to e) are combined is assumed as a structure for which characteristic prediction is performed in the element analysis apparatus 20. A plurality of elements having such a structure are formed on the same substrate, and output characteristics (electrical characteristics and mechanical characteristics) of each element are measured. Then, the circuit element (model parameter) in the equivalent circuit model obtained by converting the element structure is subjected to statistical evaluation (in-plane variation) of the element structure on the substrate based on the parameter value of each element. Further, the output characteristics of the new device structure are predicted in consideration of statistical evaluation.

  The element analysis device 20 is a computer system that performs element characteristic evaluation processing and prediction processing. As shown in FIG. 1, an input unit 11 and an output unit 12 as input / output means are connected to the element analysis apparatus 20.

The input unit 11 is a user interface for inputting measured values of a plurality of elements whose output characteristics have been measured and structure information of elements to be evaluated. The input unit 11 is configured with a keyboard, a pointing device, and the like.

  The output unit 12 is a user interface that outputs the results of evaluation processing and characteristic prediction processing executed in the element analysis device 20. The output unit 12 includes a display.

  The element analysis device 20 includes a control unit 21, an actual measurement characteristic data storage unit 22, a parameter value storage unit 23, a statistical value data storage unit 24, a parameter generation unit 25, and an equivalent circuit generator 27.

  The control unit 21 evaluates the output characteristics of the element to be evaluated, and outputs an evaluation result of the element shape of this element. The control unit 21 functions as a control unit including a CPU, a RAM, a ROM, and the like (not shown), and performs processes described later (processes including an actual value registration stage, a model parameter calculation stage, a statistical processing stage, and a simulation stage). Then, by executing the element analysis program for this purpose, the control unit 21 functions as an actual measurement value registration unit 211, a model parameter calculation unit 212, a statistical processing unit 213, a simulation processing unit 214, and the like.

The measured value registration unit 211 executes a process of registering the measured output characteristic value of each element in the measured characteristic data storage unit 22.
The model parameter calculation unit 212 applies the output characteristics stored in the actual measurement characteristic data storage unit 22 to the equivalent circuit model generated by the equivalent circuit generator 27, and sets the circuit elements (model parameters) constituting the equivalent circuit model. A process for calculating the parameter value is executed.

  The statistical processing means 213 executes statistical processing for calculating the statistical value of the parameter value calculated for each model parameter. In this embodiment, an average value or a variance value is calculated as a statistical value. This statistical processing means 213 holds data relating to a reference value for specifying a model parameter having a large variation.

  The simulation processing unit 214 executes a simulation process related to other elements using the calculated statistical value. In the present embodiment, output characteristic prediction processing and device peripheral circuit design processing for an element having a new structure are executed. The simulation processing unit 214 uses an equivalent circuit model in a new element structure designed by combining element shapes in order to predict the output characteristics of the element having the new structure. Furthermore, the simulation processing unit 214 of this embodiment stores a peripheral circuit design table for designing peripheral circuits. In this peripheral circuit design table, conditions (peripheral circuit configuration) required for the peripheral circuit are recorded corresponding to the variation width of the output value of the output characteristics.

  The actual measurement characteristic data storage unit 22 functions as an actual measurement value storage unit. In the actual measurement characteristic data storage unit 22, as shown in FIG. 3A, an actual measurement characteristic record 220 regarding the output characteristics of each manufactured element is recorded. This actual measurement characteristic record 220 is recorded when the output characteristic of each element is actually measured and the actual value of the output characteristic is registered in the statistical evaluation process described later. In the actual measurement characteristic record 220, data relating to the element number and the output characteristic is recorded.

  In the element number data area, data relating to a number for identifying each element whose output characteristics are measured is recorded. For example, in the case of elements manufactured on the same substrate, the position (coordinate on the substrate) of each element can be specified by this element number.

  In the output characteristic data area, output characteristics actually measured in this element are recorded. In the output characteristics, for example, electrical characteristics such as “current-frequency characteristics” and mechanical characteristics such as “acceleration-output voltage characteristics” are recorded.

  As shown in FIG. 3B, the parameter value storage unit 23 stores a parameter value record 230 related to model parameters in the equivalent circuit model for each element whose output characteristics are actually measured. This parameter value record 230 is recorded when a model parameter is calculated in a statistical evaluation process described later. In this parameter value record 230, data relating to element numbers and parameter sets are recorded.

In the element number data area, data relating to a number for identifying each measured element is recorded.
In the parameter set data area, data relating to parameter values of circuit elements (model parameters) constituting an equivalent circuit model in the element structure of the element is recorded. For example, when the equivalent circuit model is composed of a plurality of circuit elements (inductors, capacitors, resistors, etc.), an induction value, a capacitance value, a resistance value, etc. are recorded.

  In the statistical value data storage unit 24, as shown in FIG. 3C, a statistical value record 240 for the model parameters of the measured element structure of the element is recorded. This statistical value record 240 is recorded when statistical evaluation processing described later is performed. In this statistical value record 240, data relating to model parameters and statistical values is recorded.

In the model parameter data area, data relating to an identifier for specifying each model parameter constituting an equivalent circuit for the element structure is recorded.
In the statistical value data area, data relating to the statistical value for the model parameter is recorded. In the present embodiment, data relating to the average value and the variance value is recorded as the statistical value.

  As shown in FIG. 3D, the parameter generation unit 25 records a parameter generation record 250 for the model parameters of the measured element structure. The parameter generation record 250 is recorded when a simulation process described later is performed. In this parameter generation record 250, data relating to model parameters and parameter generation functions is recorded.

In the model parameter data area, data relating to an identifier for specifying each model parameter constituting an equivalent circuit for the element structure is recorded.
In the parameter generation function data area, data related to a function for generating a statistical value for the model parameter in a pseudo manner is recorded. In the present embodiment, data relating to a random number generation function (random number generation means) for generating a distributed state represented by a statistical value is recorded.

  The equivalent circuit generator 27 functions as a model generation unit, and executes processing for generating an electrical equivalent circuit of the element structure based on the shape parameter of the element structure. For the generation of the equivalent circuit model, an analysis support method described in Patent Document 1 is used. Furthermore, the equivalent circuit generator 27 generates an analytical expression representing the output characteristics of the generated equivalent circuit. This analytical expression uses each circuit element constituting the equivalent circuit as a variable, and can calculate an output characteristic with respect to an input value of the element.

Next, with respect to the process of evaluating element variations using the element analysis apparatus 20 described above,
This will be described with reference to FIG. Here, description will be made in the order of statistical evaluation processing, evaluation output processing, and simulation processing.

(Statistical evaluation process)
First, output characteristics of a plurality of manufactured elements are measured. In this embodiment, a plurality of elements having the same structure are produced on one substrate, and the mechanical characteristics and electrical characteristics of each element are measured. Then, the measured value data of the specified output characteristic is recorded in a predetermined storage medium in association with the element number for specifying each measured element.

  In this case, the control unit 21 of the element analysis device 20 executes a registration process of the actual measurement value of the output characteristic (step S11). Specifically, the actual value registration unit 211 of the control unit 21 displays an output characteristic registration screen on the output unit 12. This output characteristic registration screen includes an element structure input field, an element number input field, and an output characteristic input field. Here, using the input unit 11, a shape parameter for specifying the element structure of the measured element is input to the element structure input field. Next, an element number for specifying each element is set in the element number input field. Then, an output characteristic corresponding to the element number is designated in the output characteristic input field. In this case, the actual measurement value registration unit 211 acquires the actual measurement value data of the output characteristic from the storage medium for each element number, and records it in the actual measurement characteristic data storage unit 22 in association with the element number.

  When the measured value data of all the elements to be evaluated are registered in the measured characteristic data storage unit 22, the control unit 21 of the element analysis apparatus 20 executes a process of generating model parameters of element shapes that realize the measured values (step). S12). Specifically, the model parameter calculation unit 212 of the control unit 21 acquires the shape parameters (shape elements, sizes, etc. constituting the element structure) set on the output characteristic registration screen. Next, the model parameter calculation means 212 supplies this shape parameter to the equivalent circuit generator 27.

  In this case, the equivalent circuit generator 27 temporarily stores the acquired shape parameter, and generates an equivalent circuit model corresponding to the shape parameter and an analytical expression of this equivalent circuit. Then, the equivalent circuit generator 27 supplies the generated equivalent circuit model and analytical expression to the model parameter calculation unit 212.

  Next, the model parameter calculation unit 212 applies the measured value of the output characteristic for each element to the equivalent circuit model and the analytical expression acquired from the equivalent circuit generator 27, so that the model parameter ( Circuit element) parameter values are calculated. Then, the model parameter calculation unit 212 records each calculated parameter value (parameter set) in the parameter value storage unit 23 in association with the element number.

  Next, the control part 21 of the element analysis apparatus 20 performs a statistical process (step S13). Specifically, the statistical processing means 213 of the control unit 21 acquires all parameter values for each model parameter from the parameter value storage unit 23. Next, the statistical processing means 213 calculates a statistical value (average value and variance value) of the parameter value for each model parameter by a known statistical processing method. Then, the statistical processing unit 213 records the calculated statistical value in the statistical value data storage unit 24 in association with each model parameter.

(Evaluation output processing)
Next, the evaluation output process will be described. This evaluation output process is executed when the above-described statistical evaluation process is completed.

Here, first, the control unit 21 of the element analysis apparatus 20 executes a process for specifying model parameters having a large variation (step S21). Specifically, the statistical processing means 21 of the control unit 21.
3 acquires a statistical value for each model parameter from the statistical value data storage unit 24. Then, the statistical processing means 213 rearranges the model parameters in order of decreasing standardized variance value obtained by dividing the variance value by the standard value for each parameter. As the standard value, a standard value predetermined for each parameter or a process capability index (Cpk) used in quality control or the like can be used.

  Next, the control unit 21 of the element analysis apparatus 20 performs a process for identifying the cause of variation (step S22). Specifically, the statistical processing means 213 of the control unit 21 specifies model parameters whose variation is larger than the reference value. Next, the statistical processing means 213 acquires a shape element corresponding to the identified model parameter from the equivalent circuit generator 27. Then, the statistical processing means 213 identifies this shape element as the cause of the variation.

  Next, the control part 21 of the element analysis apparatus 20 performs an evaluation output process (step S23). Specifically, the statistical processing means 213 of the control unit 21 displays an evaluation result screen on the output unit 12. This evaluation result screen includes a list showing average values, variance values, and causes (shape elements) for model parameters larger than the reference value.

(Simulation process)
Next, simulation processing will be described. This simulation process is executed when a process start instruction is input at the input unit 11.

  Here, first, the control unit 21 of the element analysis device 20 executes a process for generating a model parameter for reproducing a statistical value (step S31). Specifically, the simulation processing unit 214 of the control unit 21 acquires a statistical value (average value and variance value) for each model parameter from the statistical value data storage unit 24. Then, the simulation processing means 214 generates a random number generation function (parameter generation function) for calculating this statistical value. With this random number generation means, it is possible to pseudo-generate variations in model parameters that depend on variations in shape elements of the element structure. Then, the simulation processing unit 214 stores the random number generation unit in the parameter generation unit 25 in association with the model parameter.

  Next, the control unit 21 of the element analysis apparatus 20 executes a process for acquiring new structure information (step S32). Specifically, the simulation processing means 214 of the control unit 21 outputs an element structure setting screen at the output unit 12. In this element structure setting screen, each shape parameter input field in the element structure is provided. When the shape parameter of the element structure to be simulated is input to the shape parameter input field on the element structure setting screen using the input unit 11, the simulation processing unit 214 temporarily stores the shape parameter in the memory.

  Next, the control unit 21 of the element analysis apparatus 20 executes a model parameter calculation process corresponding to the element structure (step S33). Specifically, the simulation processing means 214 of the control unit 21 supplies the temporarily stored shape parameters to the equivalent circuit generator 27. In this case, the equivalent circuit generator 27 generates an equivalent circuit model and an analytical expression of this equivalent circuit in correspondence with the shape parameter. Then, the equivalent circuit generator 27 supplies the generated equivalent circuit model and analytical expression to the simulation processing unit 214.

Next, the simulation processing unit 214 acquires a parameter generation function (random number generation function) associated with the model parameter of the equivalent circuit model from the parameter generation unit 25. Then, the simulation processing means 214 performs pseudo-generation of model parameter variations by performing Monte Carlo simulation using this random number generation function. Then, the simulation processing unit 214 performs simulation of an equivalent circuit model using model parameters to which the pseudo-generated variation is applied. Thereby, the variation width of the output prediction characteristic based on the variation of the model parameter is calculated.

  Next, the control unit 21 of the element analysis apparatus 20 executes a design process using statistical values (step S34). Specifically, the simulation processing unit 214 of the control unit 21 uses the peripheral circuit design table to specify the peripheral circuit configuration necessary for allowing the output prediction characteristics including the variation width.

  Then, the simulation processing unit 214 displays a simulation result screen in the output unit 12. The simulation result screen includes the characteristic variation of the output prediction characteristic in the new structure and the peripheral circuit configuration in consideration of the characteristic variation.

According to this embodiment, the following effects can be obtained.
(1) In the present embodiment, the control unit 21 of the element analysis device 20 executes a process for registering the actual measurement value of the output characteristic (step S11). The control unit 21 of the element analysis apparatus 20 executes a process for generating a model parameter of an element shape that realizes an actual measurement value (step S12). Thereby, the model parameters of the equivalent circuit model can be specified for the actually manufactured element. Each element can be evaluated using this model parameter.

  (2) In this embodiment, the control part 21 of the element analysis apparatus 20 performs a statistical process (step S13). Thereby, the variation in the model parameters of the equivalent circuit model can be grasped.

  (3) In the present embodiment, the control unit 21 of the element analysis apparatus 20 executes a process for specifying model parameters having large variations (step S21). The control unit 21 of the element analysis apparatus 20 performs a process for identifying the cause of variation (step S22). And the control part 21 of the element analysis apparatus 20 performs an evaluation output process (step S23). When an element is actually manufactured, the shape and size of the element may vary depending on the manufacturing process and the like. Such variations also affect the output characteristics. On the other hand, variation cannot be evaluated with an equivalent circuit model specified from the element structure. Therefore, by using the correspondence relationship between the model parameters of the equivalent circuit model and the element shape, the variation in the element shape can be efficiently evaluated from the variation in the model parameter.

  (4) In this embodiment, the control part 21 of the element analysis apparatus 20 performs the production | generation process of the model parameter which reproduces a statistics value (step S31). In this case, a random number generation function for calculating the statistical value is generated. Thereby, it is possible to generate a pseudo variation state in a simulation using a random number generation function.

  (5) In the present embodiment, the control unit 21 of the element analysis apparatus 20 executes a new structure information acquisition process (step S32). And the control part 21 of the element analysis apparatus 20 performs the calculation process of the model parameter corresponding to an element structure (step S33). The control unit 21 of the element analysis apparatus 20 executes design processing using statistical values (step S34). Thereby, the statistical value of the output characteristic of the new element structure can be predicted in consideration of the variation. Then, it is possible to perform circuit design in consideration of this variation.

Moreover, you may change the said embodiment as follows.
In the above embodiment, the control unit 21 of the element analysis device 20 executes the evaluation output process (step S23). Here, as an evaluation result, an average value, a variance value, and a cause (shape element) are output for a model parameter larger than the reference value. The evaluation results in the evaluation output process are not limited to these. For example, for a plurality of elements manufactured on a substrate, the in-plane distribution of the variation may be output. In this case, the control unit 21 specifies the position on the substrate of each element based on the element number, and outputs an evaluation result corresponding to each position. Also, the in-plane distribution of the model parameters of each substrate is stored in the parameter value storage unit 23 so that the distribution difference between substrates in the same production lot or the distribution difference between substrates in different production lots is displayed. May be. Thereby, the variation in the manufacturing process can be evaluated. For example, the maintenance time of the manufacturing apparatus can be determined based on this evaluation (variation change tendency).

  Moreover, you may make it output the manufacturing process which produced each shape element. In this case, a data storage unit in which the manufacturing process is recorded in association with the shape element is prepared. And the manufacturing process of a shape element with large dispersion | variation is output. Thereby, when the variation is large, the manufacturing process can be reviewed.

  In the above embodiment, the control unit 21 of the element analysis apparatus 20 executes a new structure information acquisition process (step S32). Here, it is also possible to evaluate an element structure similar to the actually measured element structure. In this case, statistical values of actually measured output characteristics are used. In this case, an adjustment data storage unit that stores adjustment data for calculating the change tendency of the characteristic variation caused by the shape change is prepared in advance. Then, the model parameter modified based on the changed portion is calculated using the adjustment data recorded in the adjustment data storage unit with respect to the model parameter generated based on the actually measured value of the output characteristic.

  In the above embodiment, the control unit 21 of the element analysis device 20 executes a model parameter generation process for reproducing the statistical value (step S31). Here, a random number generation function for calculating a statistical value is generated. And the control part 21 of the element analysis apparatus 20 performs the calculation process of the model parameter corresponding to an element structure using this random number generation function (step S33). The variation used in the simulation process is not limited to the random number generated by the random number generation function, and the statistical value of the actually measured element may be used.

  In the above embodiment, the control unit 21 of the element analysis device 20 executes a design process using statistical values (step S34). Here, the configuration of the peripheral circuit considering the characteristic variation is determined. The design object is not limited to this. For example, an allowable value in the output characteristics of each element may be determined in advance, and the yield of elements that fall within this allowable range on the substrate may be calculated.

  In addition, when an element is manufactured over a substrate, characteristic variations may occur between the central region and the peripheral region. Therefore, an element region in which elements that fall within the allowable value are arranged may be specified based on the simulation result.

  Further, a simulation for changing the design value of the element shape may be performed so that the output characteristics fall within the allowable range. In this case, the shape parameter of the element shape having a large variation is changed by a predetermined value, and the simulation is repeated until the output characteristic in the equivalent circuit model falls within the allowable range. Thereby, it is possible to design an element structure in consideration of variations among elements.

  In the above embodiment, the model parameter calculation unit 212 applies the output characteristic stored in the measured characteristic data storage unit 22 to the equivalent circuit model generated by the equivalent circuit generator 27 to configure the equivalent circuit model. A process of calculating the parameter value of the circuit element (model parameter) is executed. The equivalent circuit generator 27 generates an equivalent circuit model using the analysis support method described in Patent Document 1. The method for calculating the parameter value is not limited to this, and it is also possible to use a known technique such as a nonlinear least square method or a genetic algorithm.

  In the above embodiment, the control unit 21 of the element analysis device 20 executes a process for generating a model parameter of an element shape that realizes an actual measurement value (step S12). In this case, the model parameter calculation unit 212 of the control unit 21 acquires the shape parameters (shape elements, sizes, etc. constituting the element structure) set on the output characteristic registration screen. In this case, an allowable range of parameter values may be determined in advance, and parameter values such as the size of the shape elements constituting the element structure may be determined so as to be within this range. Furthermore, the model parameters may be generated using weighting in the element shape. In this case, a weight value corresponding to the variation is registered for the element shape that is expected to have a large variation and the element shape that is expected to have a small variation. Then, a model parameter that matches the output characteristics is generated by giving priority to the element shape that is expected to have a small variation due to the weighting value. Thereby, the model parameter can be generated in consideration of the character of the element shape.

  DESCRIPTION OF SYMBOLS 11 ... Input part, 12 ... Output part, 20 ... Element analyzer, 21 ... Control part, 22 ... Actual measurement characteristic data storage part, 23 ... Parameter value storage part, 24 ... Statistical value data storage part, 25 ... Parameter generation part, 27, equivalent circuit generator, 211, measured value registration means, 212, model parameter calculation means, 213, statistical processing means, 214, simulation processing means.

Claims (7)

Actual value storage means for recording the output characteristics of an element having an element structure composed of element shapes for each element;
Model generating means for generating an equivalent circuit model represented by a model parameter obtained by converting the element shape of the element structure into an equivalent circuit;
An element analysis system comprising control means connected to the input / output means,
The control means is
For each element recorded in the actual measurement value storage means, for the model parameter of the equivalent circuit model generated by the model generation means, the actual value of the output characteristic corresponding to the element recorded in the actual measurement value storage means is the element. Means for calculating a parameter value of a model parameter of the equivalent circuit model by applying to an equivalent circuit model corresponding to
Means for calculating a statistical value of the parameter value for each model parameter;
An element analysis system comprising: means for outputting an evaluation result of an element shape corresponding to a model parameter using the statistical value. 2. The element analysis system according to claim 1, wherein the control unit specifies and outputs an element shape whose model parameter variation is larger than a reference value as an evaluation result of the element shape corresponding to the model parameter . In the actual measurement value storage means, output characteristics of a plurality of elements formed on one substrate are recorded,
The control means calculates an in-plane variation on the substrate for each element shape from the output characteristics recorded in the measured value storage means, and outputs an element shape having a large in-plane variation. 3. The element analysis system according to 1 or 2.   The said control means, when acquiring a new element structure, predicts the output characteristic of the element of a new element structure using the statistical value of the element shape which constitutes the above-mentioned element structure. 4. The element analysis system according to any one of 3. The control means creates a parameter generation function for calculating an output value from a variation distribution of the statistical value of the parameter value,
5. The element analysis system according to claim 4, wherein the control means predicts a statistical value of an output characteristic of the element including a new element shape by using an output value generated by the parameter generation function. . Actual value storage means for recording the output characteristics of an element having an element structure composed of element shapes for each element;
Model generating means for generating an equivalent circuit model represented by a model parameter obtained by converting the element shape of the element structure into an equivalent circuit;
A method for analyzing an element using an element analysis system comprising a control means connected to an input / output means,
The control means is
For each element recorded in the actual measurement value storage means, for the model parameter of the equivalent circuit model generated by the model generation means, the actual value of the output characteristic corresponding to the element recorded in the actual measurement value storage means is the element. Calculating a parameter value of a model parameter of the equivalent circuit model by applying to an equivalent circuit model corresponding to
Calculating a statistical value of the parameter value for each model parameter;
And a step of outputting an evaluation result of an element shape corresponding to the model parameter using the statistical value. Actual value storage means for recording the output characteristics of an element having an element structure composed of element shapes for each element;
Model generating means for generating an equivalent circuit model represented by a model parameter obtained by converting the element shape of the element structure into an equivalent circuit;
A program for analyzing an element using an element analysis system provided with a control means connected to an input / output means,
The control means;
For each element recorded in the actual measurement value storage means, for the model parameter of the equivalent circuit model generated by the model generation means, the actual value of the output characteristic corresponding to the element recorded in the actual measurement value storage means is the element. Means for calculating a parameter value of a model parameter of the equivalent circuit model by applying to an equivalent circuit model corresponding to
Means for calculating a statistical value of the parameter value for each model parameter;
An element analysis program that functions as means for outputting an evaluation result of an element shape corresponding to a model parameter using the statistical value.

Images