JP4666428B2 - Numerical analysis method and apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a method, an apparatus, and a storage medium for predicting and analyzing a physical quantity of a structure or event that changes unsteadily.
[0002]
[Prior art]
Numerical analysis methods such as the finite element method, boundary element method, difference method, or finite volume method are widely used as methods for predicting physical quantities in the target region in a time-varying unsteady state. It is applied to simulation. These are the target areas, for example, the shape of the product or structure to which the load is applied, the area where the gas or fluid with the velocity or temperature distribution flows, or the combination thereof, and the state quantity of the analysis target is the pressure, temperature Calculated as physical quantities such as displacement, stress, strain, and speed. According to the above numerical analysis method, the target region is made into a plurality of minute elements having a one-dimensional shape such as a beam element, a two-dimensional shape such as a triangle or a quadrangle, or a three-dimensional shape such as a triangular pyramid, a triangular prism, or a hexahedron. After dividing and defining the initial state of each microelement, unknown state quantities such as pressure, temperature, displacement, stress, strain, speed, etc. of microelement at arbitrary time tn while advancing the time step by constant or variable time increment Can be calculated using a computer.
[0003]
For example, when analyzing the inflow process of the filling material (fluid) in the mold cavity (flow path), such as the filling process in injection molding, the above method is used as a method for predicting the process in which the filling area gradually expands in the mold cavity. A numerical analysis method such as Here, by predicting how the filling area spreads, filling pressure rise, and shear stress, the material injection position with good filling balance and cavity gap dimensions are simulated on the computer, and the optimum before making the mold actually. By considering the mold design, the product development period and development cost can be reduced.
[0004]
[Problems to be solved by the invention]
When the above numerical analysis method is applied to a complicated shape such as injection molding of electronic connector parts, for example, it is necessary to reduce the size of the microelements in order to accurately represent the shape, which inevitably increases the number of microelements. . Further, in recent years, when product design is performed using a three-dimensional CAD utilized in design, and numerical calculation is performed using this data, automatic division is often performed using software for dividing into minute elements called a preprocessor. At this time, the shape of the microelements collapses.For example, if the triangular microelements become obtuse triangles, the analysis accuracy decreases. There is a tendency for the number of microelements to increase.
[0005]
For such an increase in the number of microelements, the amount of unknown variables increases, so the analysis time increases. In particular, unsteady analysis that needs to be repeated in the time direction is completed within the practical analysis time. Often becomes difficult.
[0006]
On the other hand, for example, in the analysis of the injection molding process, the pressure at the filled position increases monotonically with time, while the flow front portion changes every moment during the filling and the pressure changes rapidly. Many. In this way, there is often a state in which a severe change in the state quantity occurs in a part of the analysis region and a relatively monotonous change in the other part in the analysis other than the injection molding process. In the case where the entire region having a different degree of change in the state quantity is collectively analyzed, the time interval used for the analysis needs to be set finely based on the region where the change is rapid.
[0007]
In view of such circumstances, the present invention aims to provide a predictive analysis method for a physical phenomenon that can be easily performed in a practical time even for unsteady numerical analysis that handles a large-scale region with a large number of minute elements. And
[0008]
[Means for Solving the Problems]
In order to achieve the above object, according to the present invention, the analysis area of the analysis object is divided into a plurality of minute elements by changing the shape of the injection molded product without changing the minute element size by the analysis condition input device. A step of setting, a region A including a minute element whose state quantity changes rapidly with time and a region B including a minute element whose state quantity changes slowly over time in the analysis region of the analysis model; A step of obtaining a state quantity by using a calculation method with a small temporal calculation load for B, and setting a boundary condition based on the state quantity B, and a calculation with a large temporal calculation load for the region A What is claimed is: 1. A physical phenomenon prediction analysis method comprising a step of obtaining a state quantity by a technique and setting it as a state quantity A, and storing a state quantity analysis result combining the areas A and B in a data storage device. It is subjected.
[0009]
In the present invention, the temporal calculation load is the number of operations generated in the CPU of the computer, and the calculation time can be shortened as the calculation load is smaller, that is, the number of operations is smaller. In the region A where the state quantity changes drastically, an analysis with a high calculation load is performed in order to perform an accurate analysis, and for the region B, a high-speed calculation is performed by a calculation method with a small calculation load, thereby improving the overall analysis efficiency. it can. The setting of the boundary condition of the region A represents that the value of the state amount B is a known state amount on the boundary of the region A with respect to the state amount of the portion where the region A and the region B are in contact with each other.
[0010]
According to another aspect of the present invention, the analysis region of the analysis object is divided into a plurality of microelements without changing the microelement size of the divided injection molded product shape by the analysis condition input device, and the analysis model is constructed And a step of setting a region A including a minute element whose state quantity changes rapidly and a region B including a minute element whose state quantity changes gradually in the analysis region of the analysis model, A step of obtaining a state quantity for each long time increment Δtm to obtain a state quantity B, and a step of obtaining a state quantity for each region A based on the state quantity B and obtaining a state quantity A for a short time increment Δtn. A physical phenomenon prediction / analysis method is provided, in which a state quantity analysis result obtained by combining the possessed area A and the area B is stored in a data storage device.
[0011]
Here, the time when the analysis of the region A is performed is the Δtn interval, the time when the entire region including the region A and the region B is analyzed is the Δtm interval. The entire region can be analyzed at intervals, and only the region A can be analyzed at Δtn intervals. By this method, the analysis of the entire region becomes Δtm intervals, which are coarse time increments, and only the analysis of region A is performed at Δtn intervals, so the time required for the analysis can be greatly reduced.
[0013]
As a method of approximately calculating the state quantity Pn of the region B at tn, for example, the state quantity at the last two steps tm-2 and tm-1 (tm-1−tm-2 = Δtm) in which the entire region has been analyzed. The prediction can be made by extrapolating from Pm-2 and Pm-1 by the following equation (1).
Pn = Pm-1 + {(Pm-1-Pm-2) / (tm-1-tm-2)} * (tn-tm-1) (1)
In addition, as a method for predicting Pn in the region B, the above equation uses a linear equation with respect to tn, but approximates with higher accuracy using a quadratic or higher order function, an exponential function, a logarithmic function, or the like. Is also possible.
[0014]
In the above method, the analysis region is divided into two areas A and B. However, it is also possible to divide the analysis region into three or more and set different time increments for each.
[0015]
In addition, the time increments Δtn and Δtm do not necessarily have to be real time, and may be increments of a completion rate until the analysis is completed. For example, when analyzing an injection molding filling process, The ratio of the filling volume to the molded product volume may be defined as a filling rate, and the increment of the filling rate until the filling is completed may be defined as the time increment. Or, in the case of obtaining deformation or stress for a load in structural analysis, the load is virtually gradually increased from a zero load state, and the method for obtaining the deformation or stress at the point when the finally added load value is reached In the above, the time increment may be an increment of the ratio with respect to the finally added load.
[0016]
Further, according to another preferred aspect of the present invention, the analysis object is a flow path through which a fluid flows, the state quantity is the pressure of the fluid, and the region A is a fluid interior near the free surface of the fluid. There is provided a method for predicting and analyzing a physical phenomenon, wherein the region B is a part or all of a region inside the fluid and not included in the region A.
[0017]
For example, in the analysis of the process in which a material is filled in a mold in injection molding, the object to be analyzed is a void portion (cavity) of the mold, and the portion in contact with air at the flow front portion of the fluid material is a free surface. is there. Usually, at time t after the start of filling, the filled portion of the mold gap portion becomes the analysis region. In numerical analysis, an analysis region is defined by being divided into a plurality of minute elements. The minute elements include three-dimensional elements such as a one-dimensional element such as a cylinder, a two-dimensional element such as a triangle or a quadrangle, a triangular pyramid, a triangular prism, and a hexahedron. Dimensional elements are used. A region composed of minute elements in a filled portion in contact with the free surface is referred to as the vicinity of the free surface and is referred to as region A.
[0018]
In the analysis in the middle of filling in the injection molding process, the pressure value in the region B is approximated by the prediction method, and the pressure value in the unfilled portion is set to the atmospheric pressure, so that only the pressure value in the region A portion, that is, the flow front end portion is obtained. It can be analyzed as an unknown variable. For example, a plastic part having a complicated rib structure, such as a printer chassis, is defined by CAD, and automatically divided into a set of microelements composed of the three-dimensional elements by the method described in JP-A-10-255077. When analyzing, it is divided into about 100,000 small elements, and even if a workstation is used, the calculation time is often 20 hours or more. However, according to this method, most of the calculation is about 1000 to 2000 unknown variables. The analysis time can be reduced to about 2 hours.
[0019]
The analysis method can be applied to unsteady heat conduction analysis, fluid flow analysis, transient response structure analysis, large deformation structure analysis, impact analysis and the like in addition to the analysis of the injection molding process.
[0020]
Furthermore, according to another aspect of the present invention, an analysis model is constructed in which an analysis region of an analysis object is divided into a plurality of microelements without changing the microelement size of an injection molded product by using an analysis condition input device. Analysis data constructing means; area setting means for setting a region A including a minute element whose state quantity changes rapidly with time and a region B containing a minute element whose state quantity changes slowly over time in the analysis area of the analysis model; B region calculation means for obtaining a state quantity by obtaining a state quantity by a calculation method with a small temporal calculation load for the area B, and setting a boundary condition based on the state quantity B. A physical phenomenon characterized by having an A region calculation means obtained by a calculation method with a large calculation load and storing a state quantity analysis result combining the region A and the region B in a data storage device Predictive analysis apparatus is provided.
[0021]
According to another preferred aspect of the present invention, an analysis model is constructed in which an analysis region of an analysis object is divided into a plurality of microelements without changing the microelement size of the injection molded product by an analysis condition input device. Analysis data constructing means, and region setting means for setting a region A including a minute element whose state quantity changes rapidly with time and a region B including a minute element whose state quantity changes slowly over time in the analysis region of the analysis model For the region B, a state quantity is obtained for each long time increment Δtm, and is set as a state quantity B. Based on the state quantity B, the state quantity is obtained for each short time increment Δtn. There is provided an apparatus for predicting and analyzing physical phenomena, characterized in that a state quantity analysis result obtained by combining area A and area B can be stored in a data storage device.
[0026]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a flowchart for showing a schematic procedure of an embodiment when the present invention is applied to numerical analysis in the design of a structure.
[0027]
The numerical analysis method according to the present embodiment includes a data input / preparation step (steps 1 and 2), an analysis execution step (steps 3 to 8), and an analysis result display step (step 9) performed in normal numerical analysis. The analysis execution process is divided into steps 4 and 5 for performing normal analysis execution, and steps 6 and 7 for improving calculation efficiency.
[0028]
Furthermore, the analysis apparatus used in the present embodiment includes an analysis model construction unit 108, an initial condition setting unit 101, an analysis execution unit 105, a region AB setting unit 102, a time determination unit 103, and a state B state quantity calculation unit. 104, an arithmetic device 100 for performing arithmetic processing before and after analysis, comprising a time step advancing means 106 and an analysis result output means 107, a data storage device 110 comprising a memory, a hard disk, etc. for storing analysis data and software, a keyboard, a mouse, a digitizer There are an analysis condition input device 120 including a three-dimensional shape measuring device and the like, an output device 130 including a display, a printer, an optical modeling device, and the like. The configuration of the apparatus is shown in FIG. Each of these means is realized in the form of functions or subroutines on hardware constructed by the CPU and memory of an arithmetic unit (computer).
[0029]
【Example】
In the following embodiment, the purpose of explaining the basic method of implementation of the present invention is described, and the present invention is filled at the time of injection molding, taking as an example the injection molding into a rectangular flat cavity shown in FIG. The case where it applies to an analysis is demonstrated.
[0030]
[Example 1]
First, in step 1, the analysis condition input device 120 inputs shape data obtained by dividing the injection molded product shape into microelements and analysis conditions such as the viscosity of the molding material, the injection temperature, and the injection pressure. Here, as shown in FIG. 3, the shape of the plate-like injection-molded product was divided into quadrangular microelements. A gate portion 302 is input as a gate position where the material is injected. The input analysis condition data is stored in the memory and is stored in the data storage device 110 as necessary.
[0031]
Next, in step 2, the analysis condition data is read from the data storage device 110 into the memory of the arithmetic device 100, and the initial condition is set by the initial condition setting means 101. The analysis model can be constructed on the memory by reading the already created model on the memory or automatically creating it from CAD data representing the outer shape or the like. Here, the microelements at the gate position are defined as filled, and the other microelements are defined as unfilled. Further, a minute time increment (short time increment) Δtn and a coarse time increment (long time increment) Δtm are set, and the next overall analysis time is set to tm = Δtm. Here, Δtn = 0.01 sec and Δtm = 0.1 sec. Next, in step 2 ′, the analysis execution means 105 performs analysis up to 2Δtn = 0.02 sec, and the state quantities such as pressure, temperature, speed, etc. of the filled portion are calculated and stored in the data storage device 110. . The current time tn is tn = 2Δtn.
[0032]
Subsequently, in step 3, the region A and B setting means 102 extracts the region B consisting of the completely filled microelements and the region A consisting of the microelements in contact with B, and stores them in the data storage device 110. . FIG. 4 shows a state where the flow tip has advanced to the flow tip portion 401 as an example during filling, and shows a state where the region A 402 and the region B 403 have been extracted.
[0033]
Next, the current time tn is compared with the total analysis time tm. When tn ≧ tm, the process proceeds to step 4 where the analysis execution means 105 executes analysis for both the areas A and B, and the pressure, temperature, speed, etc. Is stored in the data storage device 110 as Pm, and the next entire analysis time is updated to tm = tm + Δtm in step 5. On the other hand, when tn <tm, the data storage device 110 calls the analysis result Pm−2 before 2Δtm and the result Pm−1 before Δtm to the arithmetic unit 100, and the region B calculation means 104 performs region B in step 6. Is calculated based on the following equation (2), for example.
Pn = Pm-1 + (Pm-1-Pm-2) / (tm-1-tm-2) * (tn-tm-1) (2)
FIG. 5 is a conceptual diagram showing how the pressure value Pn is calculated for the point 404 in the region B in FIG. Pn is calculated for all the minute elements in the region B and stored in the data storage device 110. FIG. 6 shows the pressure distribution calculated for the region B.
[0034]
Subsequently, in step 7, the analysis result of the region B is called by the arithmetic device 100 and is constrained as a boundary condition, and then the state execution amount Pn of the region A is obtained by the analysis execution means 105 and stored in the data storage device 110. Saved. FIG. 7 shows the pressure analysis result combining the region A and the region B. Here, as the analysis execution means 105, a method using a numerical analysis method such as a finite element method, a boundary element method, a difference method, or a finite volume method may be used.
[0035]
Further, it is determined whether or not all the microelements have been filled. If filling has not been completed, the current time tn is updated to tn + Δtn in step 8, and the process is repeated from step 3.
[0036]
When all the microelements have been filled, an analysis result is output by the analysis result output means 107 , and in step 9, the analysis result output device 130 outputs the analysis results such as pressure, temperature, speed, etc., as a contour display diagram and numerical data output. , Graph output, vector display, etc.
[0037]
[Example 2]
As an example of applying the physical phenomenon prediction analysis method and apparatus of the present invention to a practical product, an embodiment applied to an injection molded product shown in FIG. 8 will be described.
[0038]
First, in step 1, the analysis condition input device 120 inputs an analysis model in which the shape of the injection molded product is divided into minute elements, and analysis conditions such as the viscosity of the molding material, the injection temperature, and the injection pressure. Here, from the shape data shown in FIG. 8 defined by three-dimensional CAD, fine element division is performed using a finite element method preprocessor such as PATRAN (trade name of MSC. Software Corporation), for example. Divided into triangular pyramid-shaped microelements. Further, 902 parts in FIG. 9 were input as the gate position where the material was injected. The input analysis condition data is stored in the data storage device 110.
[0039]
Next, in step 2, the analysis condition data is read from the data storage device 110 into the arithmetic device 100, and the initial condition is set by the initial condition setting means 101. Here, the microelements at the gate position are defined as filled, and the other microelements are defined as unfilled.
[0040]
In the second embodiment, analysis control was performed based on the filling rate. First, the minute filling rate increment Δrn and the coarse filling rate increment Δrm are set, and the next overall analysis filling rate is set to rm = Δrm. Here, Δrn = 0.01% and Δrm = 1%. Next, in step 2 ′, the analysis execution means 105 executes an analysis with 2Δrn = 0.02%, and the state quantities such as the pressure, temperature, speed, etc. of the filled portion are calculated and stored in the data storage device 110. The The current filling rate rn is rn = 2.DELTA.rn.
[0041]
Subsequently, in step 3, the region A and B setting means 102 extracts the region B consisting of the completely filled microelements and the region A consisting of the microelements in contact with B, and stores them in the data storage device 110. . FIG. 10 shows a state where region A and region B are extracted as an example of filling.
[0042]
Next, the current filling rate rn is compared with the overall analysis filling rate rm. First, when rn ≧ rm, the process proceeds to step 4 where the analysis is executed by the analysis execution means 105 for both the areas A and B, and the pressure, temperature , and stored in the data storage device 110 the analysis result of the speed, and the like as Pm, the entire analysis packing ratio r m of the next is updated to rm + Δrm step 5. On the other hand, when rn <rm, the data storage device 110 calls the analysis result Pm−2 before 2Δrm and the result Pm−1 before Δrm to the arithmetic unit 100, and the region B calculation means 104 causes the region B Was calculated based on the following equation (3).
Pn = Pm-1 + (Pm-1-Pm-2) / (rm-1-rm-2) * (rn-rm-1) (3)
Pn is calculated for all the minute elements in the region B and stored in the data storage device 110.
Subsequently, in step 7, the analysis result of the region B is called by the arithmetic device 100 and is constrained as a boundary condition, and then the state execution amount Pn of the region A is obtained by the analysis execution unit 105 and stored in the data storage device 110. Saved.
[0043]
Based on the pressure values of the regions A and B calculated as described above, the flow velocity of the flow front end portion is calculated, and the flow front end is advanced. Then, it is determined whether or not the filling of all the microelements has been completed. If the filling has not been completed, the current filling rate rn is updated to rn + Δrn in step 8, and the process is repeated from step 3. FIG. 11 shows the progress state of the flow front end portion selected by repeating the analysis until the end of filling. The contour lines indicate the time from the start of injection when the flow front reaches each minute element portion, and a smooth filling analysis result is obtained by the above-described pressure approximation method.
[0044]
When all the microelements have been filled, an analysis result is output by the analysis result output means 107 , and in step 9, the analysis result output device 130 outputs the analysis results such as pressure, temperature, speed, etc., as a contour display diagram and numerical data output. , Graph output, vector display, etc.
[0045]
By this method, the analysis execution by the entire area including the areas A and B becomes an interval of the entire analysis filling rate increment Δrm, and can be performed at an interval of 1/100 with respect to the increment Δrn for executing the analysis of only the flow front end area A, The total analysis time can always be 10 minutes or less, compared to the time required for the analysis in the entire region, but it was 10 minutes or less, and the analysis results were almost the same.
[0046]
【The invention's effect】
According to a preferred aspect of the numerical analysis method and apparatus according to the present invention, the step of constructing an analysis model in which the analysis region of the analysis object is divided into a plurality of microelements, and the time of the state quantity in the analysis region of the analysis model A step of setting a region A including a minute element that changes drastically and a region B including a minute element whose temporal change in the state quantity is gradual, and obtaining a state quantity for the region B by a calculation method with a small temporal calculation load. Prediction of a physical phenomenon, comprising: a step of setting a state quantity B; and a step of setting a boundary condition based on the state quantity B and obtaining the region A by a calculation method with a large temporal calculation load An analysis method and apparatus are provided, and analysis time can be greatly reduced.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of an embodiment of the present invention.
FIG. 2 is an analysis device used in an example of an embodiment of the present invention.
FIG. 3 is a rectangular cavity used in Example 1 of the present invention.
4 is an example showing a state in the middle of analysis of the rectangular cavity of FIG. 3;
FIG. 5 is an example showing a pressure value approximating method in a filled portion of a rectangular cavity.
FIG. 6 is a pressure distribution recently calculated in a region A portion of a rectangular cavity.
FIG. 7 is a pressure distribution in which a rectangular cavity region A and region B are combined.
FIG. 8 is a CAD data shape of an injection molded product used in Example 2;
FIG. 9 is an example showing a state in which FIG. 8 is divided into minute elements.
10 is an example showing a state in the middle of filling in FIG. 9;
FIG. 11 shows the progress of the flow front during the injection molding of FIG.
[Explanation of symbols]
100: Arithmetic unit 101: Initial condition setting unit 102 in the arithmetic unit 102: Areas A and B setting unit 103 according to the degree of change in the state quantity 103: Time determination unit 104 that compares tn and tm in magnitude The approximate state quantity in region B Area B calculation means 105 for calculation: Analysis execution means 106 for executing analysis of the entire area A or areas A and B: Time step advance means 107: Analysis result output means 108: Analysis model construction means 110: Data storage device 120: Analysis Condition input device 130: Analysis result output device 301: Square minute element 302: Gate position 401: Flow tip portion 402: Region A
403: Region B
404: Pressure approximate calculation position 501: Already calculated pressure change 502: Pressure change near extrapolation 601: Pressure value 701 of region B portion 1: Pressure value 801 combining region A and region B 1: Injection molded product Triangular facet 901 representing surface shape: Triangular pyramid microelement 902 divided from injection molded product shape: Gate position 1001: Filled region B
1002: Region A which is the flow front portion
1003: Unfilled portion 1101: Filling time of minute element

Claims (5)

  The analysis area of the object to be analyzed is divided into a plurality of microelements without changing the microelement size of the injection molded product by the analysis condition input device, and the analysis model is constructed, and the state within the analysis area of the analysis model A step of setting a region A containing minute elements whose amount changes rapidly and a region B containing minute elements whose state amount changes slowly, and a state amount of the region B by a calculation method with a small temporal calculation load. And obtaining a state quantity B, and setting a boundary condition based on the state quantity B, obtaining a state quantity for the region A by a calculation method with a large temporal calculation load, and obtaining the state quantity A And storing a state quantity analysis result of the region A and the region B in a data storage device.   The analysis area of the object to be analyzed is divided into a plurality of microelements without changing the microelement size of the injection molded product by the analysis condition input device, and the analysis model is constructed, and the state within the analysis area of the analysis model A step of setting a region A including minute elements whose amount changes rapidly and a region B including minute elements whose state amount changes gradually; and for the region B, a state amount is obtained for each long time increment Δtm. The state A and the region B are combined with the step of obtaining the state amount B for the region A based on the state amount B and obtaining the state amount for each short time increment Δtn based on the state amount B. A method for predicting and analyzing a physical phenomenon, characterized in that a quantitative analysis result is stored in a data storage device.   The analysis object is a flow path through which a fluid flows, the state quantity is the pressure of the fluid, the region A is a region inside the fluid near the free surface of the fluid, and the region B is inside the fluid. The physical phenomenon prediction analysis method according to claim 1, wherein the physical phenomenon prediction analysis method is a part or all of a region not included in the region A.   Analysis data construction means for constructing an analysis model in which an analysis area of an analysis object is divided into a plurality of microelements without changing the microelement size of an injection molded product by an analysis condition input device, and an analysis area of the analysis model A region setting means for setting a region A including a minute element whose state quantity changes rapidly over time and a region B including a minute element whose state quantity changes slowly over time; A B area calculation means for obtaining a state quantity by calculating a state quantity, and setting the boundary condition based on the state quantity B, and calculating the A area by a calculation technique with a large temporal calculation load. And a physical quantity prediction / analysis apparatus characterized in that a state quantity analysis result combining the area A and the area B can be stored in a data storage device.   Analysis data construction means for constructing an analysis model in which an analysis area of an analysis object is divided into a plurality of microelements without changing the microelement size of an injection molded product by an analysis condition input device, and an analysis area of the analysis model A region setting means for setting a region A including a minute element whose state quantity changes rapidly and a region B including a minute element whose state quantity changes slowly; and for the region B, every long time increment Δtm A region calculation means for obtaining a state quantity to obtain a state quantity B, and an A area calculation means for obtaining a state quantity for each short time increment Δtn for the area A based on the state quantity B. An apparatus for predicting and analyzing physical phenomena, characterized in that a state quantity analysis result including region B can be stored in a data storage device.

Images