Capacitance identification and mesh optimization method and system for integrated passive device simulation

Abstract

The application provides a capacitor identification and mesh optimization method and system for integrated passive device simulation. The method comprises the steps of determining a target dielectric layer in an integrated passive device model; determining a projection overlapping area of each pair of upper and lower metal contact surfaces having a projection overlapping relationship on the plane to which the target dielectric layer belongs and an area of a bounding box thereof; marking the projection overlapping area with a ratio of its own area to the area of the bounding box greater than a ratio threshold value as a target projection overlapping area; identifying the related entity structure of the target projection overlapping area as a capacitor; and performing mesh encryption setting on the related area of the capacitor. The system comprises various functional modules corresponding to the above steps. According to the application, the IPD capacitor encryption area can be automatically found through physical characteristics, the capacitor identification and mesh setting are integrated, the simulation stage adaptive process is accelerated, and the calculation efficiency is significantly improved while the simulation accuracy is ensured.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of EDA model layout simulation, and more particularly to a capacitance identification and grid optimization method and system for integrated passive device simulation. BACKGROUND

[0002] For the simulation of capacitance in IPD (Integrated Passive Device), the setting of grid size is the core of simulation accuracy and efficiency, which directly determines the reliability of the simulation results and the calculation cost.

[0003] The grid size defines the degree of discretization of the structure by the simulation engine. For IPD capacitors, which are extremely sensitive to electromagnetic field distribution, the necessity of grid size is reflected in the following three aspects:

[0004] Simulation accuracy guarantee: The grid is the basis for field solving. A sufficiently fine grid is a prerequisite for accurately capturing the electric field distribution in the dielectric layer and the physical field such as conductor edge current crowding and skin effect, which directly determines the calculation reliability of key parameters such as capacitance (C), quality factor (Q), and self-resonant frequency (SRF).

[0005] Resource trade-off: The grid size setting constitutes a direct trade-off between simulation computing resources and result accuracy. Too sparse will lead to distorted solving, introducing unacceptable errors and making the results invalid. Although too dense can improve accuracy, it will cause a sharp increase in computing load, consuming a large amount of time and memory, and resulting in low efficiency.

[0006] Convergence basis: Reasonable grid setting is a necessary condition for iterative solvers to achieve numerical convergence and obtain stable and reliable results.

[0007] In actual simulation, adaptive grid refinement is the most important method. The solver will first perform a preliminary solution on a relatively coarse grid, and then automatically refine the grid in areas with dramatic field changes (such as dielectric layers, metal edges, and port vicinity) based on the gradient of electric or magnetic field energy or other error estimation criteria, and then perform a second solution. This process is repeated several times until the difference between the results of the two consecutive solutions meets the preset convergence criteria. On the basis of adaptive grid, for known key areas, manual setting can be performed in advance to speed up the adaptive process. IPD capacitors are sensitive to electromagnetic field distribution, and their grid refinement can be pre-set to shorten the adaptive process. In order to realize the grid refinement pre-setting of IPD capacitors, the specific position of IPD capacitors in IPD needs to be known in advance, which involves the capacitance finding link. The existing capacitance finding methods mainly include the following:

[0008] Method one, based on design rules, that is, pre-defining the geometric rules of the capacitor, such as layer stacking, spacing, area, etc., and identifying according to the design rule checking tool.

[0009] Method two, extracting capacitor element information from the schematic diagram or netlist, and realizing matching through circuit-version correspondence.

[0010] Method three, establishing a standard capacitor template for pattern matching, and using image recognition and feature matching to identify.

[0011] Method four, based on machine learning, using a trained model to identify the capacitor structure.

[0012] Method five, based on attribute labeling, manually or automatically adding special attribute labels during design, and identifying the capacitor by parsing the attributes in the file.

[0013] However, the above-mentioned capacitor finding methods have the following disadvantages: method one is heavily dependent on pre-defined geometric rules, and cannot flexibly identify novel or non-standard capacitor structures, with high false positive rate. Method two must have complete circuit schematic diagram information, and cannot be used for pure physical layout verification. Method three is tedious to maintain, and requires creating and maintaining a large template library for capacitors of different shapes and sizes, with large calculation amount and difficulty in adapting to structural deformation. Method four is highly dependent on a large amount of high-quality labeled data to train the model, with large consumption of computing resources. Method five requires manual addition of labels during design, which is easy to miss and cannot handle unlabeled historical designs. SUMMARY

[0014] Therefore, the present application provides a capacitor identification and grid optimization method and system for integrated passive device simulation.

[0015] According to a first aspect of the present application, a capacitor identification and grid optimization method for integrated passive device simulation is provided, which comprises the following steps:

[0016] determining a target dielectric layer in the integrated passive device model that meets the thickness requirement;

[0017] determining all upper metal contact surfaces and all lower metal contact surfaces corresponding to the target dielectric layer, and determining the area of the projection overlap region of each pair of upper metal contact surfaces and lower metal contact surfaces having a projection overlap relationship on the plane to which the target dielectric layer belongs, and the area of the bounding box of the projection overlap region;

[0018] calculating the ratio of the area of each projection overlap region to the area of the bounding box of the projection overlap region, and if a ratio is greater than a predetermined ratio threshold, marking the corresponding projection overlap region as a target projection overlap region;

[0019] determining the upper metal contact surface and the lower metal contact surface corresponding to the target projection overlap region, and identifying the structure composed of the corresponding entity of the upper metal contact surface, the corresponding entity of the lower metal contact surface and the target dielectric layer as a capacitor;

[0020] defining the region of the bounding box of the target projection overlap region as an XY region, defining the range of the upper and lower surfaces of the target dielectric layer as a Z region, and setting a grid encryption for the three-dimensional region commonly defined by the XY region and the Z region.

[0021] Optionally, the step of determining the target dielectric layer meeting the thickness requirement in the integrated passive device model comprises:

[0022] discretizing all geometric structures in the integrated passive device model, and dividing all geometric structures into a dielectric group and a metal group according to the material properties of the geometric structures;

[0023] traversing each dielectric geometric structure in the dielectric group: determining the maximum length of the dielectric geometric structure in the X direction, the maximum length of the dielectric geometric structure in the Y direction and the maximum length of the dielectric geometric structure in the Z direction according to the bounding box of the dielectric geometric structure, and determining the minimum of the three as the thickness of the dielectric geometric structure, and if the thickness of the dielectric geometric structure meets the requirement, dividing the dielectric geometric structure into the dielectric group to be processed.

[0024] Optionally, the step of determining all upper metal contact surfaces and all lower metal contact surfaces corresponding to the target dielectric layer comprises:

[0025] determining the maximum face, the center point of the maximum face and the vector of the longest side of the target dielectric geometric structure in the dielectric group to be processed;

[0026] determining the maximum face normal vector of the target dielectric geometric structure according to the maximum face of the target dielectric geometric structure;

[0027] traversing each metal geometric structure in the metal group to determine all metal contact surfaces of the target dielectric geometric structure;

[0028] determining all metal contact surfaces parallel to the target dielectric geometric structure according to the normal vector of each metal contact surface of the target dielectric geometric structure and the maximum face normal vector of the target dielectric geometric structure.

[0029] Optionally, the step of determining the area of the projection overlap region of each pair of upper metal contact surface and lower metal contact surface having a projection overlap relationship on the plane to which the target dielectric layer belongs comprises:

[0030] create a first independent surface identical to the largest surface of the target medium geometry, and perform coordinate system normalization on the first independent surface so that the first independent surface is located in the XY plane of the global coordinate system;

[0031] traverse all metal contact surfaces parallel to the target medium geometry: create a second independent surface identical to the metal contact surface, and perform coordinate system normalization on the second independent surface;

[0032] for all second independent surfaces after coordinate system normalization, move the second independent surfaces that are not in the XY plane of the global coordinate system so that all second independent surfaces are located in the XY plane of the global coordinate system;

[0033] According to the DFS algorithm, obtain the overlap of each second independent surface with other second independent surfaces;

[0034] Determine whether each second independent surface and other second independent surfaces having an overlap relationship are upper surfaces or lower surfaces, if there are at least two upper surfaces, integrate the corresponding at least two upper surfaces, and if there are at least two lower surfaces, integrate the corresponding at least two lower surfaces;

[0035] If there are both upper surfaces and lower surfaces in each second independent surface and other second independent surfaces having an overlap relationship, perform intersection operation on the upper surfaces and the lower surfaces to obtain a projection overlap region.

[0036] Optionally, the step of calculating the ratio of the area of each projection overlap region to the area of the bounding box of the projection overlap region, and if a ratio is greater than a predetermined ratio threshold, marking the corresponding projection overlap region as a target projection overlap region includes:

[0037] For each projection overlap region obtained, obtain the bounding box of each projection overlap region;

[0038] Determine the ratio of the area of each projection overlap region to the area of the corresponding bounding box, and if the ratio is greater than 0.8, mark the corresponding projection overlap region as a target projection overlap region.

[0039] According to a second aspect of the present application, a capacitance identification and mesh optimization system for integrated passive device simulation is provided, which includes the following functional modules:

[0040] A medium layer screening module is configured to determine a target medium layer in an integrated passive device model that meets the thickness requirement;

[0041] an area obtaining module, configured to determine all upper metal contact surfaces and all lower metal contact surfaces corresponding to the target medium layer, and determine an area of a projection overlapping region of each pair of upper metal contact surface and lower metal contact surface having a projection overlapping relationship on a plane to which the target medium layer belongs, and an area of a bounding box of the projection overlapping region;

[0042] an area ratio obtaining and judging module, configured to calculate a ratio of the area of each projection overlapping region to the area of the bounding box of the projection overlapping region, and if a ratio is greater than a predetermined ratio threshold, mark the corresponding projection overlapping region as a target projection overlapping region;

[0043] a capacitance identifying module, configured to determine the upper metal contact surface and the lower metal contact surface corresponding to the target projection overlapping region, and identify a structure composed of the corresponding upper metal contact surface, the corresponding lower metal contact surface and the target medium layer as a capacitance;

[0044] a grid encryption module, configured to define an area of the bounding box of the target projection overlapping region as an XY area, define a range defined by upper and lower surfaces of the target medium layer as a Z area, and perform grid encryption setting on a three-dimensional area commonly defined by the XY area and the Z area.

[0045] Optionally, the medium layer screening module is further configured to:

[0046] discretize all geometric structures in the integrated passive device model, and divide all geometric structures into a medium group and a metal group according to material properties of the geometric structures;

[0047] traverse each medium geometric structure in the medium group: determine a maximum length of the medium geometric structure in an X direction, a maximum length of the medium geometric structure in a Y direction and a maximum length of the medium geometric structure in a Z direction according to a bounding box of the medium geometric structure, and determine a minimum of the three as a thickness of the medium geometric structure, and if the thickness of the medium geometric structure meets a requirement, divide the medium geometric structure to a medium group to be processed.

[0048] Optionally, the area obtaining module is further configured to:

[0049] determine a maximum face, a maximum face center point and a vector of a longest side of a target medium geometric structure in the medium group to be processed;

[0050] determine a maximum face normal vector of the target medium geometric structure according to the maximum face of the target medium geometric structure;

[0051] traverse each metal geometric structure in the metal group to determine all metal contact surfaces of the target medium geometric structure;

[0052] Determine all metal contact surfaces parallel to the target medium geometry according to the normal vector of each metal contact surface of the target medium geometry and the maximum surface normal vector of the target medium geometry.

[0053] Optionally, the area acquisition module is further configured to:

[0054] Create a first independent surface identical to the maximum surface of the target medium geometry, and perform coordinate system normalization on the first independent surface so that the first independent surface is located in the XY plane of the global coordinate system.

[0055] Traverse all metal contact surfaces parallel to the target medium geometry: create a second independent surface identical to the metal contact surface, and perform coordinate system normalization on the second independent surface.

[0056] For all second independent surfaces after coordinate system normalization, move the second independent surfaces that are not in the XY plane of the global coordinate system so that all second independent surfaces are located in the XY plane of the global coordinate system.

[0057] According to the DFS algorithm, obtain the overlap of each second independent surface and other second independent surfaces.

[0058] Determine whether each second independent surface and other second independent surfaces having an overlap relationship are upper surfaces or lower surfaces, if there are at least two upper surfaces, integrate the corresponding at least two upper surfaces, and if there are at least two lower surfaces, integrate the corresponding at least two lower surfaces.

[0059] If each second independent surface and other second independent surfaces having an overlap relationship have both upper surfaces and lower surfaces, perform intersection operation on the upper surfaces and the lower surfaces to obtain the projection overlap region.

[0060] Optionally, the area ratio acquisition and determination module is further configured to:

[0061] For each projection overlap region obtained, obtain the bounding box of each projection overlap region.

[0062] Determine the ratio of the area of each projection overlap region to the area of the corresponding bounding box, and if the ratio is greater than 0.8, mark the corresponding projection overlap region as a target projection overlap region.

[0063] The beneficial effects of the present application are:

[0064] The method for capacitance identification and grid optimization facing integrated passive device simulation of the application automatically finds the IPD capacitance encryption area through physical characteristics, does not rely on external information or preset rules, does not need to maintain a large template library and manual manual marking, automatically adapts to various processes and styles, and can efficiently and adaptively intelligently identify the area needing to be encrypted based on the physical layout only. The capacitance identification and grid setting are completed integrally, the adaptive process of the simulation stage is accelerated, the calculation efficiency is significantly improved while the simulation precision is ensured.

[0065] The system for capacitance identification and grid optimization facing integrated passive device simulation of the application belongs to the same general inventive concept as the method for capacitance identification and grid optimization facing integrated passive device simulation, has at least the same beneficial effects as the method for capacitance identification and grid optimization facing integrated passive device simulation, and the beneficial effects will not be repeated here.

[0066] Other features and advantages of the application will be described in detail in the following specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0067] The application can be better understood by reference to the following description taken in conjunction with the accompanying drawings, in which like or similar elements are designated with identical reference numerals in all the figures.

[0068] Figure 1 An implementation flowchart of the method for capacitance identification and grid optimization facing integrated passive device simulation according to the embodiment of the application is shown;

[0069] Figure 2 A specific operation flowchart of the method for capacitance identification and grid optimization facing integrated passive device simulation according to the embodiment of the application is shown;

[0070] Figure 3 A schematic diagram of the DFS algorithm constructing an adjacency list according to the embodiment of the application is shown;

[0071] Figure 4 A flowchart of the DFS algorithm finding a connected group according to the embodiment of the application is shown;

[0072] Figure 5 A structural block diagram of the system for capacitance identification and grid optimization facing integrated passive device simulation according to the embodiment of the application is shown. DETAILED DESCRIPTION

[0073] For those skilled in the art to have a more full understanding of the technical solutions of the present application, in the following, exemplary embodiments of the present application will be described more fully and in greater detail with reference to the accompanying drawings. Obviously, one or more embodiments of the present application described below are only one or more of the specific manners in which the technical solutions of the present application can be implemented, and are not exhaustive. It should be understood that the technical solutions of the present application can be implemented in other manners belonging to the same general inventive concept without being limited by the exemplary described embodiments. Based on one or more embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts should belong to the scope of protection of the present application.

[0074] Embodiments: Figure 1 The implementation flowchart of the capacitance identification and mesh optimization method for integrated passive device simulation of the embodiments of the present application is shown. Referring to Figure 1 , the capacitance identification and mesh optimization method for integrated passive device simulation of the embodiments of the present application includes the following steps:

[0075] Step S100, determining a target dielectric layer in the integrated passive device model that meets the thickness requirement;

[0076] Step S200, determining all upper metal contact surfaces and all lower metal contact surfaces corresponding to the target dielectric layer, and determining the area of the projection overlap region of each pair of upper metal contact surface and lower metal contact surface having a projection overlap relationship on the plane of the target dielectric layer, and the area of the bounding box of the projection overlap region;

[0077] Step S300, calculating the ratio of the area of each projection overlap region to the area of the bounding box of the projection overlap region, and if a ratio is greater than a predetermined ratio threshold, marking the corresponding projection overlap region as a target projection overlap region;

[0078] Step S400, determining the upper metal contact surface and the lower metal contact surface corresponding to the target projection overlap region, and identifying the structure composed of the corresponding entity of the upper metal contact surface, the corresponding entity of the lower metal contact surface and the target dielectric layer as a capacitance;

[0079] Step S500, defining the area of the bounding box of the target projection overlap region as an XY region, defining the range limited by the upper and lower surfaces of the target dielectric layer as a Z region, and setting a mesh encryption for the three-dimensional region commonly limited by the XY region and the Z region.

[0080] Specifically, in the embodiment of the present application, after the capacitance is identified, a grid encryption process is needed for the relevant three-dimensional region, which is a part of the three-dimensional structure layer to which the target medium layer belongs, and which is jointly defined by the region of the bounding box of the target projection overlapping region and the thickness of the target medium layer.

[0081] Further, the operation of determining the target medium layer in the integrated passive device model that meets the thickness requirement in step S100 of the embodiment of the present application further includes:

[0082] Discretizing all geometric structures in the integrated passive device model, and dividing all geometric structures into a medium group and a metal group according to the material properties of the geometric structures;

[0083] Traversing each medium geometric structure in the medium group: determining the maximum length of the medium geometric structure in the X direction, the maximum length of the medium geometric structure in the Y direction, and the maximum length of the medium geometric structure in the Z direction according to the bounding box of the medium geometric structure, and determining the minimum of the three as the thickness of the medium geometric structure, and if the thickness of the medium geometric structure meets the requirement, dividing the medium geometric structure into the medium group to be processed.

[0084] Further, the operation of determining all upper metal contact surfaces and all lower metal contact surfaces corresponding to the target medium layer in step S200 of the embodiment of the present application further includes:

[0085] Determining the maximum face, the center point of the maximum face, and the vector of the longest side of the target medium geometric structure in the medium group to be processed;

[0086] Determining the maximum face normal vector of the target medium geometric structure according to the maximum face of the target medium geometric structure;

[0087] Traversing each metal geometric structure in the metal group to determine all metal contact surfaces of the target medium geometric structure;

[0088] Determining all metal contact surfaces parallel to the target medium geometric structure according to the normal vector of each metal contact surface of the target medium geometric structure and the maximum face normal vector of the target medium geometric structure.

[0089] Further, the operation of determining the area of the projection overlapping region of each pair of upper metal contact surface and lower metal contact surface having a projection overlapping relationship on the plane to which the target medium layer belongs in step S200 of the embodiment of the present application further includes:

[0090] Creating a first independent face identical to the maximum face of the target medium geometric structure, and performing coordinate system normalization processing on the first independent face to make the first independent face located in the XY plane of the global coordinate system;

[0091] Traverse all metal contact surfaces parallel to the target medium geometry structure: create a second independent surface identical to the metal contact surface, and perform coordinate system normalization on the second independent surface;

[0092] For all second independent surfaces after coordinate system normalization, move the second independent surfaces not in the global coordinate system XY plane to make all second independent surfaces in the global coordinate system XY plane;

[0093] According to the DFS algorithm, obtain the overlap of each second independent surface and other second independent surfaces;

[0094] Determine whether each second independent surface and other second independent surfaces having an overlap relationship are upper surfaces or lower surfaces, if there are at least two upper surfaces, integrate the corresponding at least two upper surfaces, and if there are at least two lower surfaces, integrate the corresponding at least two lower surfaces;

[0095] If there are both upper surfaces and lower surfaces in each second independent surface and other second independent surfaces having an overlap relationship, perform intersection operation on the upper surfaces and the lower surfaces to obtain a projection overlap region.

[0096] Further, in step S300 of the embodiment of the present application, the operation of calculating the ratio of the area of each projection overlap region to the area of the bounding box of the projection overlap region, and if a ratio is greater than a predetermined ratio threshold, marking the corresponding projection overlap region as a target projection overlap region further includes:

[0097] For each projection overlap region obtained, obtain the bounding box of each projection overlap region;

[0098] Determine the ratio of the area of each projection overlap region to the area of the corresponding bounding box, and if the ratio is greater than 0.8, mark the corresponding projection overlap region as a target projection overlap region.

[0099] The method for identifying and optimizing the grid of the integrated passive device simulation of the embodiment of the present application will be described in more detail based on a specific example.

[0100] Figure 2 The specific operation flowchart of the method for identifying and optimizing the grid of the integrated passive device simulation of the embodiment of the present application is shown. Referring to Figure 2 The specific operation of the method for identifying and optimizing the grid of the integrated passive device simulation of the embodiment of the present application includes the following steps:

[0101] Step 1, initially collect the discrete medium and metal geometry information.

[0102] Step 1.1, traverse all the geometry structures in the integrated passive device model, considering that some of the geometry structures may be subjected to Unite operation, which may cause inconsistent thickness, separate the BodyID of the geometry structure subjected to Unite operation with SeparateBody command to obtain multiple separated BodyId, if the structure material is medium, add it to DieBodies, if the structure material is metal, add it to ConductorBodies.

[0103] Step 2, find the medium layer with the required thickness, and calculate the related data of the medium thin layer.

[0104] Step 2.1, traverse DieBodies, calculate the bounding box of discrete BodyId, obtain the maximum point and minimum point of the bounding box, and the minimum value of the difference of XYZ three dimensions is recorded as DiffMin, if DiffMin meets the set medium layer thickness, add the data pair of BodyId and DiffMin to DieInfoList.

[0105] Step 2.2, traverse DieInfoList, for each BodyId and DiffMin data pair, find each face of the geometry structure through BodyId, calculate the size of the face and find the MaxFaceId of the largest face and the face center point FaceCenter. Find all edges through FaceId, calculate the length of the edge and the vector of the longest edge EdgeVector. Add the five data of BodyId and DiffMin data pair, MaxFaceId, FaceCenter, EdgeVector to the structure body to DieSizeInfoList.

[0106] Step 3, traverse DieSizeInfoList, find the metal surface in contact with the medium layer, and judge whether the upper and lower metal contact surfaces of the medium layer have intersection, after fault tolerance filtering of the intersection area, obtain the encryption area.

[0107] Step 3.1, find the normal vector DieFaceNormal of MaxFaceId in the structure body.

[0108] Step 3.2, traverse the Conductor Bodies from Step 1, find all the faces of the metal geometry by BodyId, judge whether each face is in contact with the dielectric layer, if so, further judge whether the normal vector of the face is parallel to the DieFaceNormal, if so, add the FaceId of the metal face into ParallelFaceList, ParallelFaceList contains all the metal upper and lower surfaces which are in contact with the current dielectric layer and parallel to the DieFaceNormal.

[0109] Step 3.3, use the CreateObjectFromFace command to create a new independent face MaxFaceIdNew which is exactly the same as MaxFaceId, use the inverse matrix of Matrix4d to normalize the face in the coordinate system, which can also be understood as pose normalization, to prepare for the subsequent bounding box calculation. The normal vector of the normalized face is consistent with the direction of the world coordinate system, and the center point of the face is (0, 0, 0). In this pose, it is convenient to perform subsequent overlap area search and operation.

[0110] Step 3.4, if ParallelFaceList is not empty, traverse ParallelFaceList. Calculate the normal vector Normal of ParallelFace, and use the CreateObjectFromFace command to create a new independent face ParallelFaceNew which is exactly the same as ParallelFace. Use the inverse matrix of Matrix4d to normalize the independent face in the coordinate system. Add all the normalized faces to XYParalledFaceList, and add the normal vectors to XYParalledFaceNormalList. Since ParallelFaceList may contain the upper and lower surfaces of the metal, in order to facilitate subsequent intersection area search, move the metal faces which are not in the XY plane of the global coordinate system to make all the faces lie in the same plane. The method to calculate the Move parameter is: calculate the center point of each normalized face, if the Z value ZPosition of the center point is not zero, move the face in the Z direction according to the vector (0, 0, -ZPosition), at this time all the metal faces in contact with the dielectric layer are located in the same plane.

[0111] Step 3.5, use DFS algorithm to find whether there is an intersection for all the metal faces moved to the same plane.

[0112] Step 3.5.1, XYParalledFaceList has N faces, which are regarded as N nodes, and the spatial relationship between the faces is converted into the node relationship of the graph. If two faces are in contact with each other, an undirected edge is established between the nodes corresponding to the two faces. A two-dimensional data structure Graph[N][N] is constructed to represent the adjacency list. For XYParalledFaceList, a loop is performed to judge whether the second, third, and Nth faces are in contact with the first face. If nodes P and Q are in contact with node 1, nodes P and Q are added to Graph[1] in turn, and node 1 is added to Graph[P] and Graph[Q]. Then, starting from the second face, the third, fourth, and Nth faces are judged for contact with the second face. If they are in contact, the corresponding nodes are added to Group. After the face from N-1 to the last face is judged, the adjacency list of the contact relationship of all faces is constructed, and Graph[x] stores the indices of all other faces connected to face x, as shown in Figure 3

[0113] Step 3.5.2, create an array Visited[N] to represent whether each node has been visited, and traverse each face.

[0114] Step 3.5.3, find a node i that has not been visited, and perform DFS traversal from this node. A connected group Group is prepared in advance, and the core method is as follows: Step A, modify the access state Visited[i] of the node to visited, and add the node to Group; Step B, traverse all nodes connected to node i recorded in step 3.5.2. If the node has not been visited, repeat steps A and B. Until all nodes directly or indirectly connected to node i are recursively accessed. Thus, a connected group Group starting from node i is obtained, which contains all nodes that have been visited. Add Group to GroupList, as shown in Figure 4

[0115] Step 3.5.4, find the next unvisited node, and repeat step 3.5.3 until all nodes are visited. Finally, a number of connected groups GroupList are obtained.

[0116] Step 3.6, the connected groups represented by GroupList are the connected groups after the upper and lower surfaces of the metal are moved to the same plane. The actual overlapping area of the upper and lower surfaces of the IPD capacitor encryption region is only the overlapping area of the projections of the upper and lower surfaces. Therefore, the connected groups need to be further processed as follows:

[0117] ​​Step 3.6.1, traverse GroupList, for Group in it, assume the number of nodes is M, M is greater than 1, which means multiple faces contact each other, and it is necessary to further process.

[0118] Step 3.6.2, further traverse Group, through the normal vector stored in XYParalledFaceNormalList, it can be judged that the face is an upper surface or a lower surface, if the upper surface exists, it is recorded as UpBody, if multiple exist, Unite operation is performed to become a whole. Similarly, if the lower surface exists, it is recorded as DownBody, if multiple exist, Unite operation is performed to become a whole. After the traversal is completed, if the upper and lower surfaces exist, Intersection operation is performed to obtain the intersection region between the upper and lower surfaces, and is added to IntersectFaceList. After GroupList traversal is completed, several overlapping regions IntersectFaceList that satisfy the conditions of current medium having upper and lower metals and metal projection having overlap are obtained.

[0119] Step 3.7, traverse IntersectFaceList, because IntersectFace is not necessarily regular, it may be a wiring structure, and needs to be fault-tolerant and filtered: calculate the area of IntersectFace as Area1, calculate the bounding box of IntersectFace, obtain the maximum point Min and the minimum point Max of the bounding box, calculate the projection area Area2 of the bounding box on the medium, when the ratio of Area1 / Area2 is greater than 0.8, it is considered that the projection region is not a narrow structure such as wiring, and is a target capacitance encryption region. The projection region is two-dimensional, and the actual encryption region crosses the medium layer, so the actual encryption region is a three-dimensional region defined by the two-dimensional projection region of the bounding box on the medium and the thickness of the medium layer. The three-dimensional region is a region after coordinate system normalization, and needs to be multiplied by matrix Matrix4d to restore the original encryption region.

[0120] Step 4, set the encryption of the target region according to the user-specified MeshSize or MeshSizeDefault.

[0121] Specifically, in the embodiment of the application, when searching for an IPD capacitor, several associated parameters are involved:

[0122] The user-specified medium layer thickness is recorded as Thickness, and if the user does not set it, the default is 0.4 um.

[0123] User-specified Z-direction range to be searched (ZMin, ZMax), if the user does not specify, the default is all range, will search the entire EDA model.

[0124] User-specified Mesh size to be set is recorded as MeshSize, if the user does not specify, the default grid encryption value is the thickness of the found medium layer that meets the conditions multiplied by 20 recorded as MeshSizeAuto, which is proportional to the physical size, in line with the basic principles of electromagnetic simulation.

[0125] The capacitor identification and mesh optimization method for integrated passive device simulation of the embodiment of the application is based on the physical nature (metal-dielectric-metal) structure of the capacitor, does not depend on external information, is based on only geometry and material properties, is applicable to any process and design style, is adaptive to capacitors of various shapes and sizes, automatically completes the integration of identification and mesh setting, does not require manual operation, and improves the overall efficiency of design and simulation.

[0126] Correspondingly, on the basis of the capacitor identification and mesh optimization method for integrated passive device simulation proposed in the embodiment of the application, the embodiment of the application further proposes a capacitor identification and mesh optimization system for integrated passive device simulation.

[0127] Figure 5 The structural block diagram of the capacitor identification and mesh optimization system for integrated passive device simulation of the embodiment of the application is shown. Referring to Figure 5 The capacitor identification and mesh optimization system for integrated passive device simulation of the embodiment of the application comprises the following functional modules:

[0128] The medium layer screening module is used to determine the target medium layer that meets the thickness requirement in the integrated passive device model;

[0129] The area acquisition module is used to determine all upper metal contact surfaces and all lower metal contact surfaces corresponding to the target medium layer, and determine the area of the projection overlap region of each pair of upper metal contact surfaces and lower metal contact surfaces that have a projection overlap relationship on the plane to which the target medium layer belongs, and the area of the bounding box of the projection overlap region;

[0130] The area ratio acquisition and judgment module is used to calculate the ratio of the area of each projection overlap region to the area of the bounding box of the projection overlap region, and if a ratio is greater than a predetermined ratio threshold, the corresponding projection overlap region is marked as a target projection overlap region;

[0131] The capacitor identification module is used to determine the upper metal contact surface and the lower metal contact surface corresponding to the target projection overlap region, and identify the structure composed of the corresponding entity of the upper metal contact surface, the corresponding entity of the lower metal contact surface and the target medium layer as a capacitor;

[0132] The grid encryption module is used to define the region of the bounding box of the target projection overlapping region as an XY region, define the range defined by the upper and lower surfaces of the target medium layer as a Z region, and perform grid encryption setting on the three-dimensional region defined by the XY region and the Z region.

[0133] The capacitance recognition and grid optimization system for integrated passive device simulation of the embodiment of the application automatically recognizes the IPD capacitance encryption region based on physical characteristics, does not depend on special markers or attributes, has good universality and flexibility, automatically performs empirical grid encryption on the found IPD capacitance, does not depend on manual intervention, and is integrated to complete recognition and grid setting, accelerates the adaptive process in the simulation stage, significantly improves the calculation efficiency while ensuring the simulation accuracy.

[0134] Although one or more embodiments of the present application have been described above, it should be understood by those of ordinary skill in the art that the present application can be implemented in any other form without departing from the spirit and scope of the present application. Therefore, the above-described embodiments are illustrative rather than limiting, and many modifications and substitutions are obvious to those of ordinary skill in the art without departing from the spirit and scope of the present application as defined by the appended claims.

Claims

1. A capacitance identification and mesh optimization method for simulation of integrated passive devices, characterized in that, include: Identify the target dielectric layer that meets the thickness requirements in the integrated passive device model; Determine all upper metal contact surfaces and all lower metal contact surfaces corresponding to the target dielectric layer, and determine the area of ​​the projected overlap region of each pair of upper and lower metal contact surfaces that have a projected overlap relationship on the plane to which the target dielectric layer belongs, as well as the area of ​​the bounding box of the projected overlap region. Calculate the ratio of the area of ​​each projection overlap region to the area of ​​the bounding box of the projection overlap region. If a ratio is greater than a predetermined ratio threshold, then mark the corresponding projection overlap region as the target projection overlap region. The upper and lower metal contact surfaces corresponding to the target projection overlap area are determined, and the structure formed by the entity corresponding to the upper metal contact surface, the entity corresponding to the lower metal contact surface, and the target dielectric layer is identified as a capacitor. The bounding box of the overlapping area of ​​the target projection is defined as the XY region, and the range defined by the upper and lower surfaces of the target medium layer is defined as the Z region. The three-dimensional region jointly defined by the XY region and the Z region is then meshed.

2. The capacitance identification and mesh optimization method for integrated passive device simulation according to claim 1, characterized in that, The step of determining the target dielectric layer that meets the thickness requirements in the integrated passive device model includes: All geometric structures in the integrated passive device model are discretized, and all geometric structures are divided into dielectric group and metal group according to the material properties of the geometric structures. Traverse each medium geometry in the medium group: determine the maximum length of the medium geometry in the X direction, the maximum length in the Y direction, and the maximum length in the Z direction based on the bounding box of the medium geometry, and determine the minimum of the three as the thickness of the medium geometry. If the thickness of the medium geometry meets the requirements, then classify the medium geometry into the medium group to be processed.

3. The capacitance identification and mesh optimization method for integrated passive device simulation according to claim 2, characterized in that, The step of determining all upper metal contact surfaces and all lower metal contact surfaces corresponding to the target dielectric layer includes: Determine the vector of the maximum surface, the center point of the maximum surface, and the longest side of the target medium geometry in the medium group to be processed; Determine the normal vector of the maximum surface of the target medium geometry based on the maximum surface of the target medium geometry. Traverse each metal geometry in the metal group to determine all metal contact surfaces of the target medium geometry; Based on the normal vector of each metal contact surface of the target medium geometry and the maximum surface normal vector of the target medium geometry, determine all metal contact surfaces parallel to the target medium geometry.

4. The capacitance identification and mesh optimization method for integrated passive device simulation according to claim 3, characterized in that, The step of determining the area of ​​the projected overlap region of each pair of upper and lower metal contact surfaces that have a projected overlap relationship on the plane of the target dielectric layer includes: Create a first independent surface that is exactly the same as the largest surface of the target medium geometry, and normalize the coordinate system of the first independent surface so that the first independent surface is located in the global coordinate system XY plane; Traverse all metal contact surfaces parallel to the target medium geometry: create a second independent surface that is identical to the metal contact surface, and normalize the coordinate system of the second independent surface; For all second independent surfaces after coordinate system normalization, perform a movement operation on the second independent surfaces that are not in the global coordinate system XY plane so that all second independent surfaces are in the global coordinate system XY plane. Based on the DFS algorithm, obtain the overlap between each second independent face and other second independent faces; Determine whether each second independent surface and other second independent surfaces that overlap with it are upper or lower surfaces. If there are at least two upper surfaces, perform an integration operation on the corresponding at least two upper surfaces. If there are at least two lower surfaces, perform an integration operation on the corresponding at least two lower surfaces. If each second independent surface and other second independent surfaces that overlap with it have both an upper surface and a lower surface, then the intersection operation of the upper surface and the lower surface is performed to obtain the projected overlapping region.

5. The capacitance identification and mesh optimization method for integrated passive device simulation according to claim 4, characterized in that, The step of calculating the ratio of the area of ​​each projected overlapping region to the area of ​​the bounding box of that projected overlapping region, and marking the corresponding projected overlapping region as the target projected overlapping region if a ratio is greater than a predetermined ratio threshold, includes: For each obtained projection overlap region, obtain the bounding box of each projection overlap region; Determine the ratio of the area of ​​each projected overlapping region to the area of ​​the corresponding bounding box. If the ratio is greater than 0.8, mark the corresponding projected overlapping region as the target projected overlapping region.

6. A capacitance identification and mesh optimization system for integrated passive device simulation, characterized in that, include: The dielectric layer screening module is used to identify target dielectric layers in the integrated passive device model that meet the thickness requirements; The area acquisition module is used to determine all upper metal contact surfaces and all lower metal contact surfaces corresponding to the target medium layer, and to determine the area of ​​the projected overlap region of each pair of upper metal contact surfaces and lower metal contact surfaces that have a projected overlap relationship on the plane to which the target medium layer belongs, as well as the area of ​​the bounding box of the projected overlap region. The area ratio acquisition and judgment module is used to calculate the ratio of the area of ​​each projection overlap region to the area of ​​the bounding box of the projection overlap region. If a ratio is greater than a predetermined ratio threshold, the corresponding projection overlap region is marked as the target projection overlap region. The capacitor identification module is used to determine the upper metal contact surface and the lower metal contact surface corresponding to the target projection overlap area, and to identify the structure formed by the entity corresponding to the upper metal contact surface, the entity corresponding to the lower metal contact surface, and the target dielectric layer as a capacitor. The mesh encryption module is used to define the bounding box of the overlapping area of ​​the target projection as the XY region, define the range defined by the upper and lower surfaces of the target medium layer as the Z region, and set the mesh encryption for the three-dimensional region jointly defined by the XY region and the Z region.

7. The capacitance identification and mesh optimization system for integrated passive device simulation according to claim 6, characterized in that, The media layer screening module is further used for: All geometric structures in the integrated passive device model are discretized, and all geometric structures are divided into dielectric group and metal group according to the material properties of the geometric structures. Traverse each medium geometry in the medium group: determine the maximum length of the medium geometry in the X direction, the maximum length in the Y direction, and the maximum length in the Z direction based on the bounding box of the medium geometry, and determine the minimum of the three as the thickness of the medium geometry. If the thickness of the medium geometry meets the requirements, then classify the medium geometry into the medium group to be processed.

8. The capacitance identification and mesh optimization system for integrated passive device simulation according to claim 7, characterized in that, The area acquisition module is further used for: Determine the vector of the maximum surface, the center point of the maximum surface, and the longest side of the target medium geometry in the medium group to be processed; Determine the normal vector of the maximum surface of the target medium geometry based on the maximum surface of the target medium geometry. Traverse each metal geometry in the metal group to determine all metal contact surfaces of the target medium geometry; Based on the normal vector of each metal contact surface of the target medium geometry and the maximum surface normal vector of the target medium geometry, determine all metal contact surfaces parallel to the target medium geometry.

9. The capacitance identification and mesh optimization system for integrated passive device simulation according to claim 8, characterized in that, The area acquisition module is further used for: Create a first independent surface that is exactly the same as the largest surface of the target medium geometry, and normalize the coordinate system of the first independent surface so that the first independent surface is located in the global coordinate system XY plane; Traverse all metal contact surfaces parallel to the target medium geometry: create a second independent surface that is identical to the metal contact surface, and normalize the coordinate system of the second independent surface; For all second independent surfaces after coordinate system normalization, perform a movement operation on the second independent surfaces that are not in the global coordinate system XY plane so that all second independent surfaces are in the global coordinate system XY plane. Based on the DFS algorithm, obtain the overlap between each second independent face and other second independent faces; Determine whether each second independent surface and other second independent surfaces that overlap with it are upper or lower surfaces. If there are at least two upper surfaces, perform an integration operation on the corresponding at least two upper surfaces. If there are at least two lower surfaces, perform an integration operation on the corresponding at least two lower surfaces. If each second independent surface and other second independent surfaces that overlap with it have both an upper surface and a lower surface, then the intersection operation of the upper surface and the lower surface is performed to obtain the projected overlapping region.

10. The capacitance identification and mesh optimization system for integrated passive device simulation according to claim 9, characterized in that, The area ratio acquisition and judgment module is further used for: For each obtained projection overlap region, obtain the bounding box of each projection overlap region; Determine the ratio of the area of ​​each projected overlapping region to the area of ​​the corresponding bounding box. If the ratio is greater than 0.8, mark the corresponding projected overlapping region as the target projected overlapping region.

Images