Multi-layer interconnect chip design method

Through the multi-layer interconnected chip design method, the chip layout scheme map is built and the chip layout mining network is used to solve the problem of low optimization reliability in chip layout design, and the optimization reliability of chip layout scheme is improved.

CN118627456BActive Publication Date: 2024-11-01CHENGDU HONGXUN MICROELECTRONICS TECH CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411045283.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-01
Publication Date
2024-11-01
Estimated Expiration
2044-08-01

AI Technical Summary

Technical Problem

The reliability of defect optimization in existing chip layout design is not high. Due to the designer's knowledge reserve, it is difficult to ensure the reliability of chip layout solution optimization.

Method used

The multi-layer interconnected chip design method is adopted to determine the defective chip layout information and the target chip layout plan, perform solution vectorization operations, build a chip layout plan map, and determine the relevant defective chip layout plan and its optimization plan to improve the reliability of optimization.

Benefits of technology

The reliability of chip layout solution optimization is improved, and by building a knowledge graph and chip layout mining network, the certainty of related defective chip layout solutions and the reliability of optimization solutions are enhanced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118627456B_ABST
    Figure CN118627456B_ABST
Patent Text Reader

Abstract

The multi-layer stacked interconnect chip design method provided in the present application relates to the field of chip design technology. In the present application, first, the defective chip layout information and the target chip layout scheme are determined; secondly, the target chip layout scheme is vectorized and the target chip layout description vector is output; then, based on the target chip layout description vector and the defective chip layout description vectors corresponding to multiple defective chip layout schemes, a chip layout scheme map is output; finally, based on the chip layout scheme map, the relevant defective chip layout schemes are determined among the multiple defective chip layout schemes, and the optimized chip layout scheme corresponding to the relevant defective chip layout scheme is used as the basis for optimizing the defect problems existing in the target chip layout scheme. Based on the above content, the problem of low reliability of defect optimization of chip layout schemes existing in the prior art can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of chip design, and in particular to a method for designing a multi-layer interconnected chip. Background Art

[0002] In the process of chip production, chip layout is a complex and time-consuming task. At present, it is generally done by manual layout design, then modeling and simulation, and then corresponding improvements or optimizations are made for the problems that arise in the simulation process. However, in the process of improvement or optimization, it is generally based on the knowledge reserves of the corresponding designers themselves. In this way, due to the limitations of personal knowledge reserves (such as small amount of knowledge reserves and difficult to ensure reliability), when optimizing the defects of the chip layout solution, it is easy to have low reliability problems. Summary of the invention

[0003] In view of this, the purpose of the present application is to provide a multi-layer interconnect chip design method to improve the problem of low reliability of defect optimization of chip layout schemes existing in the prior art.

[0004] To achieve the above objectives, this application adopts the following technical solutions:

[0005] A multi-layer interconnect chip design method, comprising:

[0006] Determine defective chip layout information and a target chip layout scheme, wherein the defective chip layout information includes multiple defective chip layout schemes under multiple chip layout defect types and defective chip layout description vectors corresponding to the multiple defective chip layout schemes, each of the defective chip layout schemes corresponds to a defective chip layout description vector, the defective chip layout information and the target chip layout scheme are both used to describe the layout status of the corresponding multi-layer stacked interconnected chip, the defective chip layout scheme refers to a scheme with defect problems, the defect problems belong to the corresponding chip layout defect types, the chip layout defect types at least include electromagnetic isolation defects and heat dissipation defects, the electromagnetic isolation defects refer to electromagnetic interference between multi-layer stacked wirings, the defective chip layout scheme also has a corresponding optimized chip layout scheme, the optimized chip layout scheme refers to a scheme formed after optimizing the defect problems existing in the defective chip layout scheme, and the target chip layout scheme belongs to a scheme with defect problems and needs to be optimized;

[0007] Performing a scheme vectorization operation on the target chip layout scheme, and outputting a target chip layout description vector corresponding to the target chip layout scheme;

[0008] Performing knowledge graph determination processing based on the target chip layout description vector and the defective chip layout description vectors corresponding to the multiple defective chip layout schemes, and outputting a corresponding chip layout scheme graph, wherein the chip scheme members in the chip layout scheme graph include a first chip scheme member and multiple second chip scheme members, the first chip scheme member corresponds to the target chip layout scheme, and each of the second chip scheme members corresponds to a defective chip layout scheme;

[0009] Based on the chip layout scheme map, a related defective chip layout scheme corresponding to the target chip layout scheme is determined among the multiple defective chip layout schemes, and an optimized chip layout scheme corresponding to the related defective chip layout scheme is used as a basis for optimizing the defect problem existing in the target chip layout scheme.

[0010] In a preferred embodiment of the present application, in the above-mentioned multi-layer stacked interconnect chip design method, the step of performing a scheme vectorization operation on the target chip layout scheme and outputting a target chip layout description vector corresponding to the target chip layout scheme includes:

[0011] Based on the target chip layout scheme, performing relationship determination processing on each chip component in the corresponding multi-layer stacked interconnected chip to form a corresponding plurality of chip component relationship bodies, wherein each of the chip component relationship bodies includes two chip components and a layout relationship between the two chip components, wherein the chip components include at least one of a signal line, a power plane, a ground layer, a substrate, and a heat dissipation structure, and the layout relationship includes at least one of a layout position relationship, a layout electrical relationship, a layout thermal relationship, and a layout mechanical relationship;

[0012] Traversing the plurality of chip component relationship bodies to form a chip component traversal link;

[0013] Determine any chip component in the multi-layer stacked interconnected chip as a target chip component;

[0014] Acquire a first adjacent chip component relationship body of the target chip component from the plurality of chip component relationship bodies, and obtain a first chip component description vector of the target chip component based on candidate chip component description vectors corresponding to the chip components included in the first adjacent chip component relationship body and candidate layout relationship description vectors corresponding to the layout relationship;

[0015] Acquire a second adjacent chip component relationship body of the target chip component from the chip component traversal link, and obtain a second chip component description vector of the target chip component based on the optimized chip component description vector of the chip component and the optimized layout relationship description vector of the layout relationship included in the second adjacent chip component relationship body;

[0016] Based on the first chip component description vector and the second chip component description vector, a target chip component description vector of the target chip component is obtained, and based on the candidate layout relationship description vectors of the layout relationship included in the first adjacent chip component relationship body and the second adjacent chip component relationship body, a target layout relationship description vector of the layout relationship is obtained;

[0017] Based on the target chip component description vector and the target layout relationship description vector, a target chip layout description vector corresponding to the target chip layout solution is determined.

[0018] In a preferred embodiment of the present application, in the above-mentioned multi-layer stacked interconnect chip design method, the step of obtaining a first adjacent chip component relationship body of the target chip component from the multiple chip component relationship bodies, and obtaining a first chip component description vector of the target chip component based on the candidate chip component description vector corresponding to the chip component included in the first adjacent chip component relationship body and the candidate layout relationship description vector corresponding to the layout relationship, comprises:

[0019] Acquire a first adjacent chip component relationship body of the target chip component from the multiple chip component relationship bodies, and aggregate candidate chip component description vectors corresponding to the chip components included in the first adjacent chip component relationship body and candidate layout relationship description vectors corresponding to the layout relationship to form an adjacent chip component aggregation vector corresponding to the first adjacent chip component relationship body;

[0020] Based on a predetermined adjacency relationship characterization vector, the adjacent chip component aggregation vector is mapped, and the formed mapped chip component description vector is parameter adjusted to form a corresponding focus parameter distribution;

[0021] Based on the adjacent chip component aggregation vectors of all first adjacent chip component relationship bodies with the same chip relationship pointing and the focusing parameter distribution, a first chip component description vector of the target chip component pointed to in the chip relationship is obtained, wherein the chip relationship pointing is used to reflect the layout direction corresponding to the layout relationship of the target chip component in the chip component relationship body, and the layout direction includes another chip component in the chip component relationship body pointing to the target chip component or the target chip component pointing to another chip component in the chip component relationship body.

[0022] In a preferred embodiment of the present application, in the above-mentioned multi-layer interconnection chip design method, the step of mapping the adjacent chip component aggregation vector based on the predetermined adjacency relationship characterization vector, and performing parameter adjustment processing on the formed mapped chip component description vector to form a corresponding focus parameter distribution includes:

[0023] Determine D adjacency relationship characterization vectors, and, based on the D adjacency relationship characterization vectors, respectively map the adjacent chip component aggregation vectors, and perform parameter adjustment processing on the formed mapped chip component description vectors to form D focus parameter distributions, wherein D is greater than 0;

[0024] The step of obtaining the first chip component description vector of the target chip component in the chip relationship pointing to the target chip component based on the adjacent chip component aggregation vectors of all first adjacent chip component relationship bodies with the same chip relationship pointing to the target chip component and the focusing parameter distribution comprises:

[0025] Based on the adjacent chip component aggregation vectors of all first adjacent chip component relationship bodies with the same chip relationship orientation and the D focusing parameter distributions, D first chip component description vectors of the target chip component at the chip relationship orientation are obtained;

[0026] The D first chip component description vectors are fused to form a first chip component description vector of the target chip component in the direction of the chip relationship.

[0027] In a preferred embodiment of the present application, in the above-mentioned multi-layer stacked interconnect chip design method, the step of obtaining the first chip component description vector of the target chip component at the chip relationship pointing to the adjacent chip component aggregation vector of all first adjacent chip component relationship bodies with the same chip relationship pointing to and the focusing parameter distribution includes:

[0028] Multiplying the focusing parameter distribution and the adjacent chip component aggregation vector, and outputting a corresponding adjacent chip component focusing vector;

[0029] Aggregate the adjacent chip component focusing vectors of all first adjacent chip component relationship bodies with the same chip relationship pointing, and perform an excitation mapping operation on the aggregated vectors to obtain the first chip component description vector of the target chip component at the chip relationship pointing.

[0030] In a preferred embodiment of the present application, in the above-mentioned multi-layer stacked interconnect chip design method, the step of obtaining the target chip component description vector of the target chip component based on the first chip component description vector and the second chip component description vector, and the step of obtaining the target layout relationship description vector of the layout relationship based on the candidate layout relationship description vector of the layout relationship included in the first adjacent chip component relationship body and the second adjacent chip component relationship body, comprises:

[0031] Aggregating the first chip component description vector and the second chip component description vector to obtain a corresponding aggregated chip component description vector;

[0032] Outputting a target chip component description vector of the target chip component based on the aggregated chip component description vector and the candidate chip component description vector of the target chip component;

[0033] The candidate layout relationship description vector corresponding to the layout relationship included in the first adjacent chip component relationship body and the candidate layout relationship description vector corresponding to the layout relationship included in the second adjacent chip component relationship body are merged to output a target layout relationship description vector of the layout relationship.

[0034] In a preferred embodiment of the present application, in the above-mentioned multi-layer stacked interconnect chip design method, before the step of obtaining the second adjacent chip component relationship body of the target chip component in the traversal link from the chip component, and obtaining the second chip component description vector of the target chip component based on the optimized chip component description vector of the chip component and the optimized layout relationship description vector of the layout relationship included in the second adjacent chip component relationship body, the step of performing a scheme vectorization operation on the target chip layout scheme and outputting the target chip layout description vector corresponding to the target chip layout scheme also includes:

[0035] Utilizing a chip layout mining network, optimizing candidate chip component description vectors corresponding to chip components included in the chip component traversal link and candidate layout relationship description vectors corresponding to layout relationships, and outputting optimized chip component description vectors corresponding to the chip components and optimized layout relationship description vectors corresponding to the layout relationships; or

[0036] Using a first chip layout mining network, the candidate chip component description vectors corresponding to the chip components included in the chip component traversal link are optimized, and the optimized description vectors corresponding to the chip components are output; and using a second chip layout mining network, the candidate layout relationship description vectors corresponding to the layout relationship included in the chip component traversal link are optimized, and the optimized layout relationship description vectors corresponding to the layout relationship are output.

[0037] In a preferred embodiment of the present application, in the above-mentioned multi-layer stacked interconnect chip design method, the step of determining, based on the chip layout scheme map, a related defective chip layout scheme corresponding to the target chip layout scheme among the multiple defective chip layout schemes, and using the optimized chip layout scheme corresponding to the related defective chip layout scheme as a basis for optimizing the defect problem existing in the target chip layout scheme includes:

[0038] Determine a movement possibility parameter between chip solution members in the chip layout solution map, wherein the movement possibility parameter is determined based on the corresponding chip layout description vector, and the movement possibility parameter is used to reflect the possibility of moving from one chip solution member to another chip solution member during the member movement process;

[0039] According to the movement possibility parameters between the chip solution members in the chip layout solution map, multiple chip solution members in the chip layout solution map are subjected to member movement processing, and according to the corresponding order relationship of the chip solution members passed during the member movement processing, multiple chip solution members in the chip layout solution map are arranged to form a corresponding ordered set of chip solution members;

[0040] Determine A set types included in the ordered set of chip solution members and the chip solution members included in each set type, where A is greater than 0;

[0041] According to the A set types and the chip solution members included in each set type, the ordered set of chip solution members is subjected to a first clustering operation and a second clustering operation, and first clustering size information of the first clustering operation and second clustering size information of the second clustering operation are obtained, wherein the first clustering operation is used to cluster members between types, and the second clustering operation is used to cluster members within a type;

[0042] According to the first cluster size information and the second cluster size information, target cluster size information for clustering the ordered set of chip solution members is obtained;

[0043] According to the goal of reducing the target cluster size information, optimizing the A set types and the chip solution members included in each set type;

[0044] When the target cluster size information is not greater than the predetermined reference cluster size information, determine the target set type corresponding to the first chip solution member, and mark the chip solution members in the target set type to be marked as chip solution members in a first chip solution member cluster, and form a first chip solution cluster to which the target chip layout solution belongs based on the chip layout solutions corresponding to each chip solution member in the first chip solution member cluster, wherein the first chip solution cluster includes B defective chip layout solutions, and B is greater than 0;

[0045] Based on the chip layout defect type corresponding to each defective chip layout scheme in the B defective chip layout schemes, obtaining C chip layout defect types corresponding to the first chip scheme cluster, where C is greater than 0 and C is less than or equal to B;

[0046] According to the solution quantity of the defective chip layout solutions corresponding to any one of the C chip layout defect types in the B defective chip layout solutions, obtaining a solution quantity ratio corresponding to any one of the chip layout defect types in the first chip solution cluster;

[0047] Marking the chip layout defect types whose solution amount ratio is not less than a predetermined reference ratio among the C chip layout defect types, so as to obtain the chip layout defect type corresponding to the target chip layout solution;

[0048] Based on the chip layout defect type corresponding to the target chip layout scheme, a corresponding related defective chip layout scheme is determined among the multiple defective chip layout schemes, and an optimized chip layout scheme corresponding to the related defective chip layout scheme is used as a basis for optimizing the defect problem existing in the target chip layout scheme.

[0049] In a preferred embodiment of the present application, in the above-mentioned multi-layer stacked interconnect chip design method, the steps of performing a first clustering operation and a second clustering operation on the ordered set of chip solution members according to the A set types and the chip solution members included in each set type, and obtaining first clustering size information of the first clustering operation and second clustering size information of the second clustering operation include:

[0050] Determine a type movement possibility parameter between any one of the A type of sets and other type of sets, and output corresponding A type movement possibility parameters, wherein the other type of sets are A-1 type of sets other than the any one type of sets;

[0051] Based on the A types of movement possibility parameters, determining corresponding first cluster size information;

[0052] Determine the distribution parameters of each chip solution member in any one of the set types, and determine the in-set clustering size information of any one of the set types according to the distribution parameters of each chip solution member in the any one of the set types and the type movement possibility parameters between the any one of the set types and other set types, and output the in-set clustering size information of the corresponding A set types;

[0053] Based on the intra-cluster size information of the A set types, corresponding second cluster size information is obtained.

[0054] In a preferred embodiment of the present application, in the above-mentioned multi-layer stacked interconnect chip design method, the step of optimizing the A set types and the chip solution members included in each set type according to the goal of reducing the target cluster size information includes:

[0055] Optimizing the A types of movement possibility parameters according to the goal of reducing the target cluster size information;

[0056] According to the optimized A category movement possibility parameters, the number of chip solution members corresponding to each set category in the optimized A set categories is obtained;

[0057] According to the order relationship of each chip solution member in the ordered set of chip solution members from front to back and the number of chip solution members included in each set type in the optimized A set types, the ordered set of chip solution members is aggregated to form A set types.

[0058] The multi-layer stacked interconnect chip design method provided by the present application first determines the defective chip layout information and the target chip layout scheme; secondly, the target chip layout scheme is vectorized and the target chip layout description vector is output; then, based on the target chip layout description vector and the defective chip layout description vectors corresponding to the multiple defective chip layout schemes, the chip layout scheme map is output; finally, based on the chip layout scheme map, the relevant defective chip layout scheme is determined among the multiple defective chip layout schemes, and the optimized chip layout scheme corresponding to the relevant defective chip layout scheme is used as the basis for optimizing the defective problem existing in the target chip layout scheme. Based on the above content, the defective chip layout scheme related to the target chip layout scheme can be determined, so that when optimizing the defective problem of the chip layout scheme, the optimized chip layout scheme corresponding to the defective chip layout scheme can be used as a reference (i.e., similar problems are optimized and processed similarly), so that the reliability of the optimization is improved, and therefore, the problem of low reliability of defect optimization of chip layout schemes existing in the prior art can be improved. In addition, by constructing a knowledge map, the reliability of determining the relevant defective chip layout scheme can be further improved, so that the reliability of the optimized chip layout scheme used as a reference can also be higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings.

[0060] Figure 1 A structural block diagram of an electronic device provided in an embodiment of the present application.

[0061] Figure 2 A schematic flow chart of a multi-layer interconnect chip design method provided in an embodiment of the present application.

[0062] Figure 3 A schematic diagram showing the chip relationship orientation provided in an embodiment of the present application.

[0063] Figure 4 A block diagram of a multi-layer interconnected chip design device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0065] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0066] like Figure 1 As shown, an embodiment of the present application provides an electronic device, wherein the electronic device may include a memory, a processor, and a multi-layer stacked interconnect chip design device.

[0067] In detail, the memory and the processor are electrically connected directly or indirectly to achieve data transmission or interaction. For example, the memory and the processor can be electrically connected through one or more communication buses or signal lines. The multi-layer stacked interconnected chip design device includes at least one software function module stored in the memory in the form of software or firmware. The processor is used to execute an executable computer program stored in the memory, for example, the software function modules and computer programs included in the multi-layer stacked interconnected chip design device, so as to implement the multi-layer stacked interconnected chip design method provided in the embodiment of the present application.

[0068] Optionally, the memory may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), etc.

[0069] Furthermore, the processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a system on chip (SoC), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0070] Understandably, Figure 1The structure shown is for illustration only, and the electronic device may also include Figure 1 More or fewer components as shown, or with Figure 1 The different configurations shown, for example, may also include a communication unit for exchanging information with other electronic devices.

[0071] Combination Figure 2 The embodiment of the present application also provides a multi-layer interconnected chip design method applicable to the above electronic device, wherein the method steps defined in the process related to the multi-layer interconnected chip design method can be implemented by the electronic device.

[0072] The following will Figure 2 The specific process shown is explained in detail.

[0073] Step S110 , determining defective chip layout information and a target chip layout solution.

[0074] In an embodiment of the present application, the electronic device can determine defective chip layout information and a target chip layout scheme. The defective chip layout information includes multiple defective chip layout schemes under multiple chip layout defect types (exemplarily, one chip layout defect type corresponds to one or more defective chip layout schemes, such as chip layout defect type 1 corresponds to defective chip layout scheme X and defective chip layout scheme Y, and chip layout defect type 2 corresponds to defective chip layout scheme Z, defective chip layout scheme W and defective chip layout scheme D) and defective chip layout description vectors corresponding to the multiple defective chip layout schemes (such as vectorizing the defective chip layout scheme to obtain a corresponding description vector, or feature mining the defective chip layout scheme, and characterizing the mined features in the form of a vector, that is, obtaining a corresponding description vector). Each of the defective chip layout schemes corresponds to a defective chip layout description vector, and the defective chip layout scheme and the target chip layout scheme are both used to describe the layout status of the corresponding multi-layer stacked interconnected chip (exemplarily, the corresponding multi-layer stacked interconnected chip model can be obtained by three-dimensionally modeling the multi-layer stacked interconnected chip, and the chip layout scheme can be used to describe the multi-layer stacked interconnected chip model). The defective chip layout scheme refers to a scheme with defective problems, which belong to the corresponding chip layout defect types, and the chip layout defect types at least include electromagnetic isolation defects (electromagnetic interference can cause signal distortion and deformation of signal waveforms, so electromagnetic isolation defects are relatively important defects and need to be focused on), heat dissipation defects (wherein, when the temperature of the integrated circuit chip is high, if it is not cooled in time, the signal-to-noise ratio of the device will be greatly reduced, reducing the performance of the chip; long-term operation in a high temperature environment is likely to cause device aging or even failure; high temperature can also cause bonding lift, bond wire breakage and solder joint fatigue, etc., and uneven temperature distribution can even cause chip breakage, so heat dissipation defects are relatively important defects and need to be focused on), and the electromagnetic isolation defect refers to electromagnetic interference between multi-layer stacked wiring. The defective chip layout scheme also has a corresponding optimized chip layout scheme, which refers to a scheme formed after optimizing the defective problem existing in the defective chip layout scheme (the specific formation method is not limited, for example, it can be manually optimized based on corresponding experts, or it can be automatically optimized through some neural networks), and the target chip layout scheme belongs to a scheme with defective problems and needs to be optimized, that is, a chip layout scheme to be optimized.

[0075] Step S120 , performing a scheme vectorization operation on the target chip layout scheme, and outputting a target chip layout description vector corresponding to the target chip layout scheme.

[0076] In an embodiment of the present application, after determining the target chip layout scheme, the electronic device may perform a scheme vectorization operation on the target chip layout scheme and output a target chip layout description vector corresponding to the target chip layout scheme. Exemplarily, feature mining may be performed on the target chip layout scheme, and then the result of the feature mining may be represented in the form of a vector.

[0077] Step S130 , performing knowledge graph determination processing based on the target chip layout description vector and the defective chip layout description vectors corresponding to the multiple defective chip layout solutions, and outputting a corresponding chip layout solution graph.

[0078] In an embodiment of the present application, after obtaining the target chip layout description vector, the electronic device can perform knowledge graph determination processing based on the target chip layout description vector and the defective chip layout description vectors corresponding to the multiple defective chip layout schemes, and output the corresponding chip layout scheme graph. Among them, the chip scheme members in the chip layout scheme graph include a first chip scheme member and multiple second chip scheme members, the first chip scheme member corresponds to the target chip layout scheme, and each of the second chip scheme members corresponds to a defective chip layout scheme. In other words, each chip layout scheme can be used as a chip scheme member, and the corresponding chip layout description vector can be used as the attribute information of the chip scheme member. In addition, the correlation between chip scheme members can be determined based on the similarity (such as cosine similarity) between the corresponding chip layout description vectors, and chip scheme members with high correlation (such as similarity greater than a preset similarity) can be connected through graph edges or associated line segments to characterize their association relationship.

[0079] Step S140, based on the chip layout scheme map, determining a related defective chip layout scheme corresponding to the target chip layout scheme from among the multiple defective chip layout schemes, and using the optimized chip layout scheme corresponding to the related defective chip layout scheme as a basis for optimizing the defect problem existing in the target chip layout scheme.

[0080] In an embodiment of the present application, after forming the chip layout scheme map, the electronic device can determine, based on the chip layout scheme map, a related defective chip layout scheme corresponding to the target chip layout scheme among the multiple defective chip layout schemes, and use the optimized chip layout scheme corresponding to the related defective chip layout scheme as a basis for optimizing the defect problem existing in the target chip layout scheme.

[0081] Based on the above content, the defective chip layout scheme related to the target chip layout scheme can be determined, so that when optimizing the defective problem of the chip layout scheme, the optimized chip layout scheme corresponding to the defective chip layout scheme can be used as a reference (i.e., similar problems are optimized similarly, etc.), so that the reliability of the optimization is improved. Therefore, the problem of low reliability of defective chip layout scheme optimization in the prior art can be improved. In addition, by constructing a knowledge graph, the reliability of determining the relevant defective chip layout scheme can be further improved, so that the reliability of the optimized chip layout scheme used as a reference can also be higher.

[0082] It should be noted that for the above step S120, the specific manner of performing the scheme vectorization operation on the target chip layout scheme is not limited and can be selected according to actual needs.

[0083] For example, in an alternative implementation, in order to improve the characterization capability of the output target chip layout description vector for the target chip layout solution, the above-mentioned step S120 may further include step S121, step S122, step S123, step S124, step S125, step S126 and step S127, and the specific contents of each step are described as follows.

[0084] Step S121 : Based on the target chip layout solution, a relationship determination process is performed on each chip component in the corresponding multi-layer stacked interconnect chip to form a corresponding plurality of chip component relationship bodies.

[0085] In an embodiment of the present application, the relationship determination processing can be performed on each chip component in the corresponding multi-layer stacked interconnected chip based on the target chip layout scheme to form a corresponding plurality of chip component relationship bodies. Each of the chip component relationship bodies includes two chip components and a layout relationship between the two chip components, wherein the chip components include at least one of a signal line, a power plane, a ground layer, a substrate, and a heat dissipation structure, and the layout relationship includes at least one of a layout position relationship, a layout electrical relationship, a layout thermal relationship, and a layout mechanical relationship. For example, if chip component 1 is welded to chip component 2, the corresponding chip component relationship body is "chip component 1, welding, chip component 2". For another example, if chip component 2 is electrically connected to chip component 3, the corresponding chip component relationship body is "chip component 2, electrical connection, chip component 3". For another example, if chip component 2 is heat transferred to chip component 4, the corresponding chip component relationship body is "chip component 2, heat transfer, chip component 4". In other words, "welding", "electrical connection", and "heat transfer" refer to layout relationships.

[0086] Step S122: traverse the plurality of chip component relationship bodies to form a chip component traversal link.

[0087] In an embodiment of the present application, after forming the multiple chip component relationship bodies, the multiple chip component relationship bodies can be traversed to form a chip component traversal link, such as "chip component 1, welding, chip component 2, electrical connection, chip component 3".

[0088] Step S123: determine any chip component in the multi-layer stacked interconnected chip as a target chip component.

[0089] In the embodiment of the present application, any chip component in the multi-layer interconnected chip can be determined as a target chip component. Exemplarily, each chip component in the multi-layer interconnected chip can be determined as a target chip component in sequence or in parallel, so as to determine the corresponding target chip component description vector and target layout relationship description vector through subsequent steps.

[0090] Step S124, obtaining the first adjacent chip component relationship body of the target chip component from the multiple chip component relationship bodies, and obtaining the first chip component description vector of the target chip component based on the candidate chip component description vector corresponding to the chip component included in the first adjacent chip component relationship body and the candidate layout relationship description vector corresponding to the layout relationship.

[0091] In an embodiment of the present application, after determining the target chip component, the first adjacent chip component relationship body of the target chip component can be obtained from the multiple chip component relationship bodies (exemplarily, the first adjacent chip component relationship body can refer to the chip component relationship body including the target chip component), and the first chip component description vector of the target chip component can be obtained based on the candidate chip component description vector corresponding to the chip component included in the first adjacent chip component relationship body and the candidate layout relationship description vector corresponding to the layout relationship. Exemplarily, the candidate chip component description vector corresponding to the chip component can be obtained by mapping the description information of the chip component (such as name, function, location, material, shape, size and other attribute information), and exemplarily, it can be obtained by training a word embedding model (such as Word2Vec, a word embedding technology developed by Google, which learns the vector representation of words by training a neural network model). For example, "a four-corner support heat sink structure using AlSiC material" can be mapped to:

[0092] Use: [0.12, -0.45, 0.67, ..., 0.32];

[0093] AlSiC material: [-0.25, 0.31, -0.15, ..., -0.02];

[0094] of: [0.08, -0.29, 0.53, ..., -0.18];

[0095] Four corners: [0.41, 0.12, -0.09, ..., 0.27];

[0096] Support: [-0.37, 0.09, 0.62, ..., -0.03];

[0097] heat sink: [0.19, 0.36, -0.24, ..., 0.45];

[0098] Structure: [-0.05, 0.17, -0.08, ..., -0.29];

[0099] Then, the word vectors of each word can be combined to form a candidate chip component description vector. For example, the candidate chip component description vector of "four-corner support heat sink structure using AlSiC material" is:

[0100] [ \text{Vec}(\text{“Four-corner support heat sink structure using AlSiC material”}) = \text{Vec}(\text{“adopted”}) + \text{Vec}(\text{“AlSiC material”}) + \text{Vec}(\text{“of”}) + \text{Vec}(\text{“four corners”}) + \text{Vec}(\text{“support”}) + \text{Vec}(\text{“heat sink”}) + \text{Vec}(\text{“structure”}) ].

[0101] Step S125, obtaining the second adjacent chip component relationship body of the target chip component from the chip component traversal link, and obtaining the second chip component description vector of the target chip component based on the optimized chip component description vector of the chip component and the optimized layout relationship description vector of the layout relationship included in the second adjacent chip component relationship body.

[0102] In an embodiment of the present application, the second adjacent chip component relationship body of the target chip component can be obtained from the chip component traversal link (for example, the chip component traversal link where the target chip component is located can be determined first, and then the chip component relationship body including the target chip component in the chip component traversal link can be determined as the second adjacent chip component relationship body), and the second chip component description vector of the target chip component can be obtained based on the optimized chip component description vector of the chip component and the optimized layout relationship description vector of the layout relationship included in the second adjacent chip component relationship body. The optimized chip component description vector can be formed by optimizing the candidate chip component description vector, and the optimized layout relationship description vector can be formed by optimizing the layout relationship description vector.

[0103] Step S126, based on the first chip component description vector and the second chip component description vector, obtain the target chip component description vector of the target chip component, and based on the candidate layout relationship description vector of the layout relationship included in the first adjacent chip component relationship body and the second adjacent chip component relationship body, obtain the target layout relationship description vector of the layout relationship.

[0104] In an embodiment of the present application, after obtaining the first chip component description vector and the second chip component description vector, a target chip component description vector of the target chip component can be obtained based on the first chip component description vector and the second chip component description vector. Furthermore, a target layout relationship description vector of the layout relationship can be obtained based on the candidate layout relationship description vector of the layout relationship included in the first adjacent chip component relationship body and the candidate layout relationship description vector of the layout relationship included in the second adjacent chip component relationship body. In this way, a target chip component description vector and a target layout relationship description vector related to the target chip component can be obtained.

[0105] Step S127: determining a target chip layout description vector corresponding to the target chip layout solution based on the target chip component description vector and the target layout relationship description vector.

[0106] In an embodiment of the present application, after obtaining the target chip component description vector and the target layout relationship description vector, the target chip layout description vector corresponding to the target chip layout scheme can be determined based on the target chip component description vector and the target layout relationship description vector. Exemplarily, the target chip component description vector and the target layout relationship description vector of the target chip component can be aggregated or fused (such as performing splicing and other processing) to obtain an aggregate description vector of the target chip component, and then, since each chip component can be used as a target chip component, an aggregate description vector of each chip component can be obtained. In this way, after the aggregate description vector of each chip component is convolved, the convolution processing results of each chip component can be aggregated or fused (such as performing splicing and other processing), and then, convolution processing can be performed again, etc., so as to obtain the target chip layout description vector corresponding to the target chip layout scheme.

[0107] It can be understood that, in the above-mentioned step S124, the specific method of obtaining the first chip component description vector is not limited and can be selected according to actual needs. For example, in an alternative embodiment, in order to fully mine the key information of the target chip component and make the final target chip layout description vector have better characterization ability, the above-mentioned step S124 can further include step S124a, step S124b and step S124c. The specific content of each step is as follows.

[0108] Step S124a, obtaining the first adjacent chip component relationship body of the target chip component from the multiple chip component relationship bodies, and aggregating the candidate chip component description vectors corresponding to the chip components included in the first adjacent chip component relationship body and the candidate layout relationship description vectors corresponding to the layout relationship to form an adjacent chip component aggregation vector corresponding to the first adjacent chip component relationship body; illustratively, the candidate chip component description vectors and the candidate layout relationship description vectors can be spliced ​​and processed, such as the adjacent chip component aggregation vector is {candidate chip component description vector, candidate layout relationship description vector}.

[0109] In an embodiment of the present application, a first adjacent chip component relationship body of the target chip component can be obtained from the multiple chip component relationship bodies, and the candidate chip component description vectors corresponding to the chip components included in the first adjacent chip component relationship body and the candidate layout relationship description vectors corresponding to the layout relationship are aggregated to form an adjacent chip component aggregation vector corresponding to the first adjacent chip component relationship body.

[0110] Step S124b, based on the predetermined adjacency relationship characterization vector, the adjacent chip component aggregation vector is mapped, and the formed mapped chip component description vector is parameter adjusted to form a corresponding focus parameter distribution.

[0111] In an embodiment of the present application, after obtaining the adjacent chip component aggregation vector, the adjacent chip component aggregation vector can be mapped based on a predetermined adjacency relationship characterization vector, and the formed mapping chip component description vector can be parameter adjusted to form a corresponding focus parameter distribution. The adjacency relationship characterization vector can be a weight parameter distribution, can be a network parameter in a neural network, and is determined during the training process of the neural network, that is, the multi-layer stacked interconnect chip design method provided in the embodiment of the present application can be implemented by the neural network, and the neural network can be obtained by training based on corresponding sample data, except that the sample chip layout scheme has a related defect chip layout scheme as a label, so that the neural network can be continuously iterated and optimized based on the label, thereby obtaining a reliable neural network.

[0112] Step S124c, based on the adjacent chip component aggregation vectors of all first adjacent chip component relationship bodies with the same chip relationship pointing and the focusing parameter distribution, obtain the first chip component description vector of the target chip component in the chip relationship pointing.

[0113] In an embodiment of the present application, the first chip component description vector of the target chip component at the chip relationship pointing can be obtained based on the adjacent chip component aggregation vectors of all first adjacent chip component relationship bodies with the same chip relationship pointing and the focusing parameter distribution. Wherein, the chip relationship pointing is used to reflect the layout direction corresponding to the layout relationship of the target chip component in the chip component relationship body, and the layout direction includes another chip component in the chip component relationship body pointing to the target chip component or the target chip component pointing to another chip component in the chip component relationship body. As described above, and combined with Figure 3 When chip component 2 is used as the target chip component, the chip relationship directions of "chip component 1, welding, chip component 2" and "chip component 2, electrical connection, chip component 3" are different; the chip relationship directions of "chip component 2, electrical connection, chip component 3" and "chip component 2, heat transfer, chip component 4" are the same; exemplarily, as described above, the chip relationship directions may be two, and therefore, the first chip component description vector at the two chip relationship directions can be obtained, and thus, the two first chip component description vectors can be further fused, such as averaging, weighted summation, convolution after splicing, etc., to form the first chip component description vector of the target chip component.

[0114] It is understandable that, in an alternative implementation, the accuracy of the obtained first chip component description vector is further improved, and the above step S124b may include the following contents:

[0115] First, D adjacency relationship characterization vectors can be determined. Secondly, based on the D adjacency relationship characterization vectors, the adjacency chip component aggregation vectors can be mapped respectively, and the formed mapping chip component description vector can be parameter-adjusted to form D focusing parameter distributions, wherein D is greater than 0 and is an integer, for example, it can be equal to 1, 2, 3, etc.; illustratively, the D adjacency relationship characterization vectors can be multiplied with the adjacency chip component aggregation vector respectively (or, the adjacency relationship characterization vector and the adjacency chip component aggregation vector can be superimposed or spliced, and then, the superimposed or spliced ​​result can be activated, and the specific activation function is not limited) to form a mapping chip component description vector, and then, the mapping chip component description vector can be normalized by functions such as softmax to obtain D focusing parameter distributions.

[0116] It is understandable that, in an alternative implementation, after obtaining D focusing parameter distributions, the above step S124c may include the following contents:

[0117] First, based on the adjacent chip component aggregation vectors of all first adjacent chip component relationship bodies with the same chip relationship pointing to them and the D focusing parameter distributions, the D first chip component description vectors of the target chip component at the chip relationship pointing to them can be obtained (exemplarily, when there are multiple first adjacent chip component relationship bodies with the same chip relationship pointing to them, the adjacent chip component aggregation vectors of all first adjacent chip component relationship bodies with the same chip relationship pointing to them can be averaged first, and then the obtained mean vectors can be multiplied by the D focusing parameter distributions respectively, that is, weighting of the mean vector based on the focusing parameter distribution is achieved to mine out important information therein);

[0118] Secondly, the D first chip component description vectors can be fused (exemplarily, the D first chip component description vectors can be superimposed or spliced, and then the superposition or splicing results can be activated, and the specific activation function is not limited) to form the first chip component description vector of the target chip component in the direction of the chip relationship.

[0119] It is understandable that, in an alternative implementation, the above step S124c may include the following contents:

[0120] First, the focusing parameter distribution and the adjacent chip component aggregation vector may be multiplied to output the corresponding adjacent chip component focusing vector, that is, the adjacent chip component aggregation vector may be weighted based on the focusing parameter distribution to mine out important information therein;

[0121] Secondly, the adjacent chip component focusing vectors of all first adjacent chip component relationship bodies with the same chip relationship pointing to them can be aggregated (such as splicing, superposition, etc.), and the aggregated vectors can be subjected to excitation mapping operations (also known as activation processing, and the parameters of the specific activation function can be formed as network parameters of the neural network during the training process) to obtain the first chip component description vector of the target chip component at the chip relationship pointing to them.

[0122] Among them, in the above-mentioned step S125, the specific method of "obtaining the second chip component description vector of the target chip component based on the optimized chip component description vector of the chip component included in the second adjacent chip component relationship body and the optimized layout relationship description vector of the layout relationship" can be the same as the method of "obtaining the first chip component description vector of the target chip component based on the candidate chip component description vector corresponding to the chip component included in the first adjacent chip component relationship body and the candidate layout relationship description vector corresponding to the layout relationship", which will not be repeated here.

[0123] Furthermore, in order to make the obtained optimized chip component description vector and optimized layout relationship description vector more reliable and improve the characterization capability of the second chip component description vector, before the above-mentioned step S125, step S120 may further include the following contents:

[0124] A chip layout mining network (such as the neural network described above) may be used to optimize candidate chip component description vectors corresponding to chip components included in the chip component traversal link and candidate layout relationship description vectors corresponding to layout relationships, and output optimized chip component description vectors corresponding to the chip components and optimized layout relationship description vectors corresponding to the layout relationships; or

[0125] Using a first chip layout mining network, the candidate chip component description vectors corresponding to the chip components included in the chip component traversal link are optimized, and the optimized description vectors corresponding to the chip components are output; and using a second chip layout mining network, the candidate layout relationship description vectors corresponding to the layout relationship included in the chip component traversal link are optimized, and the optimized layout relationship description vectors corresponding to the layout relationship are output; wherein the first chip layout mining network and the second chip layout mining network can be part of the aforementioned neural network.

[0126] It can be understood that, in an alternative implementation, in order to make the result of optimizing the vector have higher reliability, when optimizing the candidate chip component description vector and the candidate layout relationship description vector together by using the chip layout mining network, it can include:

[0127] First, the chip components in the multi-layer stacked interconnected chip corresponding to the target chip layout scheme can be arbitrarily combined to form a corresponding plurality of pending chip component relationship bodies; wherein each pending chip component relationship body includes two chip components and a layout relationship between the two chip components, and the layout relationship can be randomly generated;

[0128] Secondly, based on the multiple chip component relationship bodies, the multiple pending chip component relationship bodies can be marked to form a first type of pending chip component relationship body and a second type of pending chip component relationship body, wherein the layout relationship in any one of the first type of pending chip component relationship bodies is correct, and the layout relationship in any one of the second type of pending chip component relationship bodies is wrong;

[0129] Then, the chip layout mining network can be used to optimize the candidate chip component description vectors corresponding to the chip components included in the to-be-determined chip component relationship body (including convolution operations, weighted operations based on weight parameters, and superposition based on bias parameters, etc.), to obtain a corresponding first optimized description vector, and then, based on the first optimized description vector, the layout relationship between the chip components included in the to-be-determined chip component relationship body is predicted to obtain a corresponding first predicted layout relationship;

[0130] Afterwards, the error (such as edit distance, etc.) between the layout relationship between the chip components included in the pending chip component relationship body and the corresponding first predicted layout relationship may be calculated;

[0131] Furthermore, the network parameters of the chip layout mining network may be optimized and adjusted in the direction of reducing the error corresponding to the first type of chip component relationship bodies to be determined and increasing the error corresponding to the second type of chip component relationship bodies to be determined (the condition for the adjustment to be terminated may be that a preset number of times is reached or the error reduction amplitude converges, etc.), to obtain a new chip layout mining network;

[0132] Furthermore, the new chip layout mining network can be used to optimize candidate layout relationship description vectors corresponding to the correct layout relationship between chip components included in the chip component relationship body to be determined, to obtain a corresponding second optimized description vector, and then, based on the second optimized description vector, the layout relationship between chip components included in the chip component relationship body to be determined is predicted, to obtain a corresponding second predicted layout relationship;

[0133] Finally, the network parameters of the new chip layout mining network can be optimized and adjusted in the direction of reducing the error between the second predicted layout relationship and the corresponding correct layout relationship (the condition for the end of adjustment may be reaching a preset number of times or the error reduction amplitude converges, etc.), to obtain an updated chip layout mining network, and, using the updated chip layout mining network, optimize the candidate chip component description vectors and the candidate layout relationship description vectors together.

[0134] It is understandable that, in the above step S126, the specific method of obtaining the target chip component description vector and the target layout relationship description vector is not limited and can be selected according to actual needs. For example, in an alternative implementation, the above step S126 may include the following:

[0135] First, the first chip component description vector and the second chip component description vector can be aggregated to obtain a corresponding aggregate chip component description vector; illustratively, the first chip component description vector and the second chip component description vector can be superimposed or spliced ​​to obtain a corresponding aggregate chip component description vector; it should be noted that in the aforementioned steps, after obtaining the first chip component description vector and the second chip component description vector corresponding to each chip relationship, they can be not merged first, so, in the current step, the first chip component description vector and the second chip component description vector corresponding to each chip relationship can be aggregated separately to obtain the aggregate chip component description vector corresponding to each chip relationship. As described above, two aggregate chip component description vectors can be obtained, and then the two aggregate chip component description vectors are averaged or weighted summed to obtain the final aggregate chip component description vector;

[0136] Secondly, based on the aggregate chip component description vector and the candidate chip component description vector of the target chip component, a target chip component description vector of the target chip component can be output; illustratively, the aggregate chip component description vector and the candidate chip component description vector of the target chip component can be added to output a corresponding target chip component description vector;

[0137] Then, the candidate layout relationship description vector corresponding to the layout relationship included in the first adjacent chip component relationship body and the candidate layout relationship description vector corresponding to the layout relationship included in the second adjacent chip component relationship body can be fused to output the target layout relationship description vector of the layout relationship; exemplarily, the candidate layout relationship description vector corresponding to the layout relationship included in the first adjacent chip component relationship body and the candidate layout relationship description vector corresponding to the layout relationship included in the second adjacent chip component relationship body can be superimposed or spliced, and then the superposition or splicing result can be activated to obtain the corresponding target layout relationship description vector.

[0138] It should be noted that for the above step S140 , the specific method of determining the optimized chip layout solution is not limited and can be selected according to actual needs.

[0139] For example, in an alternative implementation, in order to determine a more reliable layout scheme for a relevant defective chip so that the corresponding optimized chip layout scheme has higher reliability when used as a basis for optimizing the defect problem, the above-mentioned step S140 may further include step S141, step S142, step S143, step S144, step S145, step S146, step S147, step S148, step S149, step S151 and step S152, and the specific content of each step is described as follows.

[0140] Step S141, determining the movement possibility parameters between chip solution members in the chip layout solution map.

[0141] In an embodiment of the present application, a movement possibility parameter between chip solution members in the chip layout solution map can be determined. The movement possibility parameter is determined based on the corresponding chip layout description vector, and the movement possibility parameter is used to reflect the possibility of moving from one chip solution member to another chip solution member during the member movement process. Exemplarily, the cosine similarity between the corresponding chip layout description vectors can be calculated, and then the cosine similarity can be normalized to obtain the corresponding movement possibility parameter.

[0142] Step S142, performing member movement processing on multiple chip solution members in the chip layout solution map according to movement possibility parameters between chip solution members in the chip layout solution map, and arranging multiple chip solution members in the chip layout solution map according to the corresponding sequence relationship of chip solution members passed during the member movement processing to form a corresponding ordered set of chip solution members.

[0143] In the embodiment of the present application, after determining the movement possibility parameter, multiple chip solution members in the chip layout solution map can be subjected to member movement processing (such as random walk) according to the movement possibility parameter between chip solution members in the chip layout solution map, and multiple chip solution members in the chip layout solution map can be arranged according to the corresponding order relationship of the chip solution members passed during the member movement processing to form a corresponding ordered set of chip solution members, such as chip solution member 1→chip solution member 3→chip solution member 5→chip solution member 7→chip solution member 2. In addition, the determination of the first chip solution member can be arbitrary, and the first chip solution member (i.e., the chip solution member corresponding to the target chip layout solution) can also be used as the first chip solution member.

[0144] Step S143, determining the A set types included in the ordered set of chip solution members and the chip solution members included in each set type.

[0145] In an embodiment of the present application, after forming the ordered set of chip solution members, the A set types included in the ordered set of chip solution members and the chip solution members included in each set type can be determined, where A is greater than 0 and is an integer. Exemplarily, the number of set types included in the ordered set of chip solution members can be determined based on the number of chip layout defect types included in the defective chip layout information, that is, the number of chip layout defect types included in the defective chip layout information is equal to the number of set types included in the ordered set of chip solution members. If the defective chip layout information includes A chip layout defect types, then the ordered set of chip solution members includes A set types. The number of chip solution members included in each set type can be determined based on the number of defective chip layout schemes under each chip layout defect type included in the defective chip layout information. For example, the defective chip layout information may include defective chip layout solutions under three chip layout defect categories, namely, 7 defective chip layout solutions under the first chip layout defect category, 16 defective chip layout solutions under the second chip layout defect category, and 12 defective chip layout solutions under the third chip layout defect category. The chip solution member ordered set may be divided into three set categories according to the arrangement order of each chip solution member in the chip solution member ordered set, wherein the first set category includes 7 chip solution members, the second set category includes 16 chip solution members, and the third set category includes 12 chip solution members. Among them, the number of chip solution members in the set category of the first chip solution member increases by 1. For example, if the first chip solution member is divided into the first set category, the number of chip solution members in the first set category increases by 1, so that the first set category includes 8 chip solution members.

[0146] Step S144, based on the A set types and the chip solution members included in each set type, the ordered set of chip solution members is subjected to a first clustering operation and a second clustering operation, and first clustering size information of the first clustering operation and second clustering size information of the second clustering operation are obtained.

[0147] In the embodiment of the present application, the chip solution members ordered set can be subjected to a first clustering operation and a second clustering operation according to the A set types and the chip solution members included in each set type, and first clustering size information of the first clustering operation and second clustering size information of the second clustering operation are obtained. The first clustering operation is used to cluster members between types, and the second clustering operation is used to cluster members within a type.

[0148] Step S145 , obtaining target cluster size information for clustering the ordered set of chip solution members according to the first cluster size information and the second cluster size information.

[0149] In an embodiment of the present application, after obtaining the first cluster size information and the second cluster size information, target cluster size information for clustering the ordered set of chip solution members can be obtained based on the first cluster size information and the second cluster size information.

[0150] Step S146, optimizing the A set types and the chip solution members included in each set type according to the goal of reducing the target cluster size information.

[0151] In an embodiment of the present application, after obtaining the target cluster size information, the A set types and the chip solution members included in each set type can be optimized according to the goal of reducing the target cluster size information. That is, if the target cluster size information is too large, it indicates that the clustering is too general or general, and the precision is not actually high. Therefore, it is necessary to optimize along the direction of reducing the target cluster size information. Exemplarily, the type movement possibility parameter between any set type and other set types can be adjusted according to the principle of reducing the target cluster size information, that is, the A type movement possibility parameters can be updated, that is, the A set types and the chip solution member quantity included in each set type can be updated; that is, the chip solution member quantity included in each set type in the optimized A set types can be determined based on the updated A type movement possibility parameters; then, according to the arrangement order of each chip solution member in the chip solution member ordered set from front to back, the chip solution member ordered set is re-divided into A set types according to the chip solution member quantity included in each set type in the optimized A set types. That is: according to the goal of reducing the target cluster size information, the movement possibility parameters of the A categories are optimized; according to the optimized movement possibility parameters of the A categories, the number of chip solution members corresponding to each set category in the optimized A set categories is obtained; according to the front-to-back order relationship of each chip solution member in the ordered set of chip solution members and the number of chip solution members included in each set category in the optimized A set categories, the ordered set of chip solution members is aggregated (i.e., re-divided) to form A set categories.

[0152] Step S147, when the target cluster size information is not greater than the predetermined reference cluster size information, determine the target set type corresponding to the first chip solution member, and mark the chip solution members in the target set type to mark them as chip solution members in the first chip solution member cluster, and based on the chip layout solutions corresponding to each chip solution member in the first chip solution member cluster, form a first chip solution cluster to which the target chip layout solution belongs.

[0153] In an embodiment of the present application, in the process of continuous iterative optimization, the current target cluster size information can be compared with the reference cluster size information, and then, when the target cluster size information is not greater than the predetermined reference cluster size information (i.e., it is reduced to a small enough size and the iterative optimization is stopped), the target set type corresponding to the first chip solution member (i.e., the set type to which the first chip solution member belongs) is determined, and the chip solution members in the target set type are marked as chip solution members in the first chip solution member cluster, and based on the chip layout schemes corresponding to each chip solution member in the first chip solution member cluster, the first chip solution cluster to which the target chip layout scheme belongs is formed. Among them, the first chip solution cluster includes B defective chip layout schemes, and B is greater than 0 and is an integer. In addition, the specific size of the reference cluster size information is not limited and can be selected according to actual needs.

[0154] Step S148 , obtaining C chip layout defect types corresponding to the first chip solution cluster based on the chip layout defect type corresponding to each of the B defective chip layout solutions.

[0155] In the embodiment of the present application, after the first chip solution cluster is formed, C chip layout defect types corresponding to the first chip solution cluster can be obtained based on the chip layout defect types corresponding to each defective chip layout scheme in the B defective chip layout schemes (i.e., the chip layout defect types corresponding to the B defective chip layout schemes are counted to obtain C chip layout defect types). C is greater than 0, and C is less than or equal to B.

[0156] Step S149: obtaining a solution quantity ratio corresponding to any one chip layout defect type in the first chip solution cluster according to the solution quantity of the defective chip layout solutions in the B defective chip layout solutions corresponding to any one chip layout defect type in the C chip layout defect types.

[0157] In the embodiment of the present application, after obtaining the C chip layout defect types corresponding to the first chip solution cluster, the solution quantity ratio corresponding to any one of the C chip layout defect types in the first chip solution cluster can be obtained according to the solution quantity of the defective chip layout solutions in the B defective chip layout solutions corresponding to any one of the C chip layout defect types. In other words, for each chip layout defect type in the C chip layout defect types, the quantity ratio of the defective chip layout solutions corresponding to the chip layout defect type in the B defective chip layout solutions can be counted.

[0158] Step S151 , marking the chip layout defect types whose solution amount ratio is not less than a predetermined reference ratio among the C chip layout defect types, so as to obtain the chip layout defect type corresponding to the target chip layout solution.

[0159] In the embodiment of the present application, after obtaining the solution amount ratio corresponding to each chip layout defect type, the chip layout defect types whose solution amount ratio is not less than a predetermined reference ratio (the specific value can be configured according to actual needs) among the C chip layout defect types can be marked to obtain the chip layout defect type corresponding to the target chip layout solution. Exemplarily, the chip layout defect type whose solution amount ratio is not less than the predetermined reference ratio and has the maximum value can be used as the chip layout defect type corresponding to the target chip layout solution.

[0160] Step S152, based on the chip layout defect type corresponding to the target chip layout scheme, determine a corresponding related defective chip layout scheme from the multiple defective chip layout schemes, and use the optimized chip layout scheme corresponding to the related defective chip layout scheme as a basis for optimizing the defect problem existing in the target chip layout scheme.

[0161] In an embodiment of the present application, after obtaining the chip layout defect type corresponding to the target chip layout scheme, a corresponding related defective chip layout scheme can be determined among the multiple defective chip layout schemes based on the chip layout defect type corresponding to the target chip layout scheme, and the optimized chip layout scheme corresponding to the related defective chip layout scheme can be used as a basis for optimizing the defect problem existing in the target chip layout scheme. For example, a defective chip layout scheme with the same chip layout defect type as that corresponding to the target chip layout scheme among the multiple defective chip layout schemes can be determined as a related defective chip layout scheme, and the optimized chip layout scheme corresponding to the related defective chip layout scheme can be used as a basis for optimizing the defect problem existing in the target chip layout scheme.

[0162] It is understandable that, in the above step S144, the specific method of obtaining the first cluster size information and the second cluster size information is not limited and can be selected according to actual needs. For example, in an alternative implementation, the above step S144 may include the following contents:

[0163] First, the type movement possibility parameters between any one of the A set types and other set types can be determined, and the corresponding A type movement possibility parameters can be output, wherein the other set types are A-1 set types other than the any one set type; illustratively, for each chip solution member in the any one set type, the number ratio of the chip solution member is determined (the number ratio is equal to the ratio between the number of the chip solution member and the number of all chip solution members), and the movement possibility parameter between the chip solution member and each other chip solution member in the other set types is determined, and then, the number ratio can be multiplied by each of the movement possibility parameters, and the multiplication results can be summed, and finally, the summation results corresponding to each chip solution member in the any one set type can be summed again to obtain the type movement possibility parameter corresponding to the any one set type, so that A type movement possibility parameters can be obtained;

[0164] Secondly, the corresponding first cluster size information can be determined based on the A type movement possibility parameters; exemplarily, the sum of the A type movement possibility parameters can be calculated, and then the ratio between each type movement possibility parameter and the sum is calculated, and the logarithm of the ratio is calculated, and then the ratio is used as a weighting coefficient to perform weighted summation on the results of the logarithm calculation, and the result of the weighted summation is inverted to obtain the corresponding first cluster size information;

[0165] Then, the distribution parameters of each chip solution member within any one of the set types can be determined, and, based on the distribution parameters of each chip solution member within any one of the set types and the type movement possibility parameters between the any one of the set types and other set types, the clustering size information within the set of the any one of the set types can be determined, and the clustering size information within the set of the corresponding A set types can be output; illustratively, the distribution parameter can be equal to the ratio of the number of the aforementioned chip solution members, based on which, the sum of the distribution parameters of each chip solution member within the any one of the set types can be calculated, and the sum of the sum and the type movement possibility parameter corresponding to the any one of the set types can be summed to obtain the corresponding sum result; on this basis, on the one hand, the any one of the set types can be calculated. The ratio between the type movement possibility parameter corresponding to the group type and the sum result is calculated, and the logarithm operation is performed on the ratio, and the product between the ratio and the result of the logarithm operation is inverted to obtain a first parameter; on the other hand, for each chip solution member in any one of the group types, the ratio between the distribution parameter of the chip solution member and the sum result is calculated, and the logarithm operation is performed on the ratio, and the ratio is used as a weighting coefficient to perform weighted summation calculation on the result of the logarithm operation, and the result of the weighted summation calculation is inverted to obtain a second parameter. Finally, the first parameter and the second parameter can be fused, such as adding, to obtain the in-group clustering size information corresponding to any one of the group types. In this way, the in-group clustering size information of A group types can be determined;

[0166] Finally, the corresponding second cluster size information may be obtained based on the intra-set cluster size information of the A set types; illustratively, the sum of the intra-set cluster size information of the A set types may be calculated to obtain the second cluster size information.

[0167] Based on the above content, the above step S145 may include the following contents:

[0168] First, the product of the sum of the A category movement possibility parameters and the first cluster size information may be calculated to obtain a third parameter;

[0169] Then, the sum of the distribution parameters of each chip solution member in the arbitrary set type can be calculated, and the sum and the type movement possibility parameter corresponding to the arbitrary set type can be summed to obtain a corresponding sum result, and the sum result can be used as a weighting coefficient to weight (such as multiply) the cluster size information in the set corresponding to the arbitrary set type to obtain a corresponding weighted result, and the weighted results corresponding to each set type are summed to obtain a fourth parameter;

[0170] Finally, the sum of the third parameter and the fourth parameter may be calculated to obtain target cluster size information for clustering the ordered set of chip solution members.

[0171] Combination Figure 4 The embodiment of the present application also provides a multi-layer interconnected chip design device applicable to the above electronic device. The multi-layer interconnected chip design device may include a chip layout information determination module, a solution vectorization module, a knowledge graph determination module, and an optimization solution determination module.

[0172] The chip layout information determination module is used to determine the defective chip layout information and the target chip layout scheme, wherein the defective chip layout information includes multiple defective chip layout schemes under multiple chip layout defect types and defective chip layout description vectors corresponding to the multiple defective chip layout schemes, each of the defective chip layout schemes corresponds to a defective chip layout description vector, the defective chip layout information and the target chip layout scheme are both used to describe the layout status of the corresponding multi-layer stacked interconnected chip, the defective chip layout scheme refers to a scheme with a defect problem, and the defect problem belongs to the corresponding chip layout defect type, the chip layout defect type includes at least electromagnetic isolation defects and heat dissipation defects, and the electromagnetic isolation defect refers to electromagnetic interference between multi-layer stacked wirings, the defective chip layout scheme also has a corresponding optimized chip layout scheme, the optimized chip layout scheme refers to a scheme formed after optimizing the defect problem existing in the defective chip layout scheme, and the target chip layout scheme belongs to a scheme with defect problems and needs to be optimized. In an embodiment of the present application, the chip layout information determination module can be used to perform Figure 2 As shown in step S110, the relevant contents of the chip layout information determination module can refer to the above description of step S110.

[0173] The scheme vectorization module is used to perform a scheme vectorization operation on the target chip layout scheme and output a target chip layout description vector corresponding to the target chip layout scheme. In the embodiment of the present application, the scheme vectorization module can be used to perform Figure 2 As shown in step S120, for the relevant content of the scheme vectorization module, reference can be made to the above description of step S120.

[0174] The knowledge graph determination module is used to perform knowledge graph determination processing based on the target chip layout description vector and the defective chip layout description vectors corresponding to the multiple defective chip layout schemes, and output the corresponding chip layout scheme graph, wherein the chip scheme members in the chip layout scheme graph include a first chip scheme member and multiple second chip scheme members, the first chip scheme member corresponds to the target chip layout scheme, and each of the second chip scheme members corresponds to a defective chip layout scheme. In an embodiment of the present application, the knowledge graph determination module can be used to perform Figure 2 As shown in step S130, for relevant contents of the knowledge graph determination module, please refer to the previous description of step S130.

[0175] The optimization scheme determination module is used to determine, based on the chip layout scheme map, a relevant defective chip layout scheme corresponding to the target chip layout scheme among the multiple defective chip layout schemes, and to use the optimized chip layout scheme corresponding to the relevant defective chip layout scheme as a basis for optimizing the defect problem existing in the target chip layout scheme. In the embodiment of the present application, the optimization scheme determination module can be used to perform Figure 2 As shown in step S140, for the relevant contents of the optimization scheme determination module, refer to the above description of step S140.

[0176] In summary, the multi-layer stacked interconnect chip design method provided by the present application first determines the defective chip layout information and the target chip layout scheme; secondly, the target chip layout scheme is vectorized to output the target chip layout description vector; then, based on the target chip layout description vector and the defective chip layout description vectors corresponding to the multiple defective chip layout schemes, the chip layout scheme map is output; finally, based on the chip layout scheme map, the relevant defective chip layout scheme is determined among the multiple defective chip layout schemes, and the optimized chip layout scheme corresponding to the relevant defective chip layout scheme is used as the basis for optimizing the defective problem existing in the target chip layout scheme. Based on the above content, the defective chip layout scheme related to the target chip layout scheme can be determined, so that when optimizing the defective problem of the chip layout scheme, the optimized chip layout scheme corresponding to the defective chip layout scheme can be used as a reference (i.e., similar problems are optimized and processed similarly), so that the reliability of the optimization is improved, and therefore, the problem of low reliability of defect optimization of chip layout schemes existing in the prior art can be improved. In addition, by constructing a knowledge graph, the reliability of determining the relevant defective chip layout scheme can be further improved, so that the reliability of the optimized chip layout scheme used as a reference can also be higher.

[0177] In several embodiments provided in the embodiments of the present application, it should be understood that the disclosed device and method can also be implemented in other ways. The device and method embodiments described above are merely schematic, for example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the device, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0178] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0179] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, electronic device, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code. It should be noted that in this article, the term "include", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such process, method, article or device. Without more constraints, an element defined by the phrase "comprising a..." does not exclude the existence of other identical elements in the process, method, article or apparatus comprising the element.

[0180] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A multi-layer interconnect chip design method, It is characterized in that include: Determine defective chip layout information and a target chip layout scheme, wherein the defective chip layout information includes multiple defective chip layout schemes under multiple chip layout defect types and defective chip layout description vectors corresponding to the multiple defective chip layout schemes, each of the defective chip layout schemes corresponds to a defective chip layout description vector, the defective chip layout information and the target chip layout scheme are both used to describe the layout status of the corresponding multi-layer stacked interconnected chip, the defective chip layout scheme refers to a scheme with defect problems, the defect problems belong to the corresponding chip layout defect types, the chip layout defect types at least include electromagnetic isolation defects and heat dissipation defects, the electromagnetic isolation defects refer to electromagnetic interference between multi-layer stacked wirings, the defective chip layout scheme also has a corresponding optimized chip layout scheme, the optimized chip layout scheme refers to a scheme formed after optimizing the defect problems existing in the defective chip layout scheme, and the target chip layout scheme belongs to a scheme with defect problems and needs to be optimized; Performing a scheme vectorization operation on the target chip layout scheme, and outputting a target chip layout description vector corresponding to the target chip layout scheme; Performing knowledge graph determination processing based on the target chip layout description vector and the defective chip layout description vectors corresponding to the multiple defective chip layout schemes, and outputting a corresponding chip layout scheme graph, wherein the chip scheme members in the chip layout scheme graph include a first chip scheme member and multiple second chip scheme members, the first chip scheme member corresponds to the target chip layout scheme, and each of the second chip scheme members corresponds to a defective chip layout scheme; Based on the chip layout scheme map, a related defective chip layout scheme corresponding to the target chip layout scheme is determined among the multiple defective chip layout schemes, and an optimized chip layout scheme corresponding to the related defective chip layout scheme is used as a basis for optimizing the defect problem existing in the target chip layout scheme.

2. The multi-layer interconnect chip design method according to claim 1, It is characterized in that The step of performing a scheme vectorization operation on the target chip layout scheme and outputting a target chip layout description vector corresponding to the target chip layout scheme comprises: Based on the target chip layout scheme, performing relationship determination processing on each chip component in the corresponding multi-layer stacked interconnected chip to form a corresponding plurality of chip component relationship bodies, wherein each of the chip component relationship bodies includes two chip components and a layout relationship between the two chip components, wherein the chip components include at least one of a signal line, a power plane, a ground layer, a substrate, and a heat dissipation structure, and the layout relationship includes at least one of a layout position relationship, a layout electrical relationship, a layout thermal relationship, and a layout mechanical relationship; Traversing the plurality of chip component relationship bodies to form a chip component traversal link; Determine any chip component in the multi-layer stacked interconnected chip as a target chip component; Acquire a first adjacent chip component relationship body of the target chip component from the plurality of chip component relationship bodies, and obtain a first chip component description vector of the target chip component based on candidate chip component description vectors corresponding to the chip components included in the first adjacent chip component relationship body and candidate layout relationship description vectors corresponding to the layout relationship; Acquire a second adjacent chip component relationship body of the target chip component from the chip component traversal link, and obtain a second chip component description vector of the target chip component based on the optimized chip component description vector of the chip component and the optimized layout relationship description vector of the layout relationship included in the second adjacent chip component relationship body; Based on the first chip component description vector and the second chip component description vector, a target chip component description vector of the target chip component is obtained, and based on the candidate layout relationship description vectors of the layout relationship included in the first adjacent chip component relationship body and the second adjacent chip component relationship body, a target layout relationship description vector of the layout relationship is obtained; Based on the target chip component description vector and the target layout relationship description vector, a target chip layout description vector corresponding to the target chip layout solution is determined.

3. The multi-layer interconnect chip design method according to claim 2, It is characterized in that The step of acquiring a first adjacent chip component relationship body of the target chip component from the plurality of chip component relationship bodies, and obtaining a first chip component description vector of the target chip component based on a candidate chip component description vector corresponding to the chip component included in the first adjacent chip component relationship body and a candidate layout relationship description vector corresponding to the layout relationship, comprises: Acquire a first adjacent chip component relationship body of the target chip component from the multiple chip component relationship bodies, and aggregate candidate chip component description vectors corresponding to the chip components included in the first adjacent chip component relationship body and candidate layout relationship description vectors corresponding to the layout relationship to form an adjacent chip component aggregation vector corresponding to the first adjacent chip component relationship body; Based on a predetermined adjacency relationship characterization vector, the adjacent chip component aggregation vector is mapped, and the formed mapped chip component description vector is parameter adjusted to form a corresponding focus parameter distribution; Based on the adjacent chip component aggregation vectors of all first adjacent chip component relationship bodies with the same chip relationship pointing and the focusing parameter distribution, a first chip component description vector of the target chip component pointed to in the chip relationship is obtained, wherein the chip relationship pointing is used to reflect the layout direction corresponding to the layout relationship of the target chip component in the chip component relationship body, and the layout direction includes another chip component in the chip component relationship body pointing to the target chip component or the target chip component pointing to another chip component in the chip component relationship body.

4. The multi-layer interconnect chip design method according to claim 3, It is characterized in that The steps of mapping the adjacent chip component aggregation vector based on the predetermined adjacency relationship characterization vector, and performing parameter adjustment processing on the formed mapped chip component description vector to form a corresponding focus parameter distribution include: Determine D adjacency relationship characterization vectors, and, based on the D adjacency relationship characterization vectors, respectively map the adjacent chip component aggregation vectors, and perform parameter adjustment processing on the formed mapped chip component description vectors to form D focus parameter distributions, wherein D is greater than 0; The step of obtaining the first chip component description vector of the target chip component in the chip relationship pointing to the target chip component based on the adjacent chip component aggregation vectors of all first adjacent chip component relationship bodies with the same chip relationship pointing to the target chip component and the focusing parameter distribution comprises: Based on the adjacent chip component aggregation vectors of all first adjacent chip component relationship bodies with the same chip relationship orientation and the D focusing parameter distributions, D first chip component description vectors of the target chip component at the chip relationship orientation are obtained; The D first chip component description vectors are fused to form a first chip component description vector of the target chip component in the direction of the chip relationship.

5. The multi-layer interconnect chip design method according to claim 3, It is characterized in that The step of obtaining the first chip component description vector of the target chip component in the chip relationship pointing to the target chip component based on the adjacent chip component aggregation vectors of all first adjacent chip component relationship bodies with the same chip relationship pointing to the target chip component and the focusing parameter distribution comprises: Multiplying the focusing parameter distribution and the adjacent chip component aggregation vector, and outputting a corresponding adjacent chip component focusing vector; Aggregate the adjacent chip component focusing vectors of all first adjacent chip component relationship bodies with the same chip relationship pointing, and perform an excitation mapping operation on the aggregated vectors to obtain the first chip component description vector of the target chip component at the chip relationship pointing.

6. The multi-layer interconnect chip design method according to claim 2, It is characterized in that The step of obtaining a target chip component description vector of the target chip component based on the first chip component description vector and the second chip component description vector, and obtaining a target layout relationship description vector of the layout relationship based on a candidate layout relationship description vector of the layout relationship included in the first adjacent chip component relationship body and the second adjacent chip component relationship body, comprises: Aggregating the first chip component description vector and the second chip component description vector to obtain a corresponding aggregated chip component description vector; Outputting a target chip component description vector of the target chip component based on the aggregated chip component description vector and the candidate chip component description vector of the target chip component; The candidate layout relationship description vector corresponding to the layout relationship included in the first adjacent chip component relationship body and the candidate layout relationship description vector corresponding to the layout relationship included in the second adjacent chip component relationship body are merged to output a target layout relationship description vector of the layout relationship.

7. The multi-layer interconnect chip design method according to claim 2, It is characterized in that Before the step of acquiring the second adjacent chip component relationship body of the target chip component in the traversal link from the chip component, and obtaining the second chip component description vector of the target chip component based on the optimized chip component description vector of the chip component and the optimized layout relationship description vector of the layout relationship included in the second adjacent chip component relationship body, the step of performing a scheme vectorization operation on the target chip layout scheme and outputting the target chip layout description vector corresponding to the target chip layout scheme further includes: Utilizing a chip layout mining network, optimizing candidate chip component description vectors corresponding to chip components included in the chip component traversal link and candidate layout relationship description vectors corresponding to layout relationships, and outputting optimized chip component description vectors corresponding to the chip components and optimized layout relationship description vectors corresponding to the layout relationships; or Using a first chip layout mining network, the candidate chip component description vectors corresponding to the chip components included in the chip component traversal link are optimized, and the optimized description vectors corresponding to the chip components are output; and using a second chip layout mining network, the candidate layout relationship description vectors corresponding to the layout relationship included in the chip component traversal link are optimized, and the optimized layout relationship description vectors corresponding to the layout relationship are output.

8. The multi-layer interconnect chip design method according to any one of claims 1 to 7, It is characterized in that The step of determining, based on the chip layout scheme map, a related defective chip layout scheme corresponding to the target chip layout scheme among the multiple defective chip layout schemes, and using the optimized chip layout scheme corresponding to the related defective chip layout scheme as a basis for optimizing the defect problem existing in the target chip layout scheme includes: Determine a movement possibility parameter between chip solution members in the chip layout solution map, wherein the movement possibility parameter is determined based on the corresponding chip layout description vector, and the movement possibility parameter is used to reflect the possibility of moving from one chip solution member to another chip solution member during the member movement process; According to the movement possibility parameters between the chip solution members in the chip layout solution map, multiple chip solution members in the chip layout solution map are subjected to member movement processing, and according to the corresponding order relationship of the chip solution members passed during the member movement processing, multiple chip solution members in the chip layout solution map are arranged to form a corresponding ordered set of chip solution members; Determine A set types included in the ordered set of chip solution members and the chip solution members included in each set type, where A is greater than 0; According to the A set types and the chip solution members included in each set type, the ordered set of chip solution members is subjected to a first clustering operation and a second clustering operation, and first clustering size information of the first clustering operation and second clustering size information of the second clustering operation are obtained, wherein the first clustering operation is used to cluster members between types, and the second clustering operation is used to cluster members within a type; According to the first cluster size information and the second cluster size information, target cluster size information for clustering the ordered set of chip solution members is obtained; According to the goal of reducing the target cluster size information, optimizing the A set types and the chip solution members included in each set type; When the target cluster size information is not greater than the predetermined reference cluster size information, determine the target set type corresponding to the first chip solution member, and mark the chip solution members in the target set type to be marked as chip solution members in a first chip solution member cluster, and form a first chip solution cluster to which the target chip layout solution belongs based on the chip layout solutions corresponding to each chip solution member in the first chip solution member cluster, wherein the first chip solution cluster includes B defective chip layout solutions, and B is greater than 0; Based on the chip layout defect type corresponding to each defective chip layout scheme in the B defective chip layout schemes, obtaining C chip layout defect types corresponding to the first chip scheme cluster, where C is greater than 0 and C is less than or equal to B; According to the solution quantity of the defective chip layout solutions corresponding to any one of the C chip layout defect types in the B defective chip layout solutions, obtaining a solution quantity ratio corresponding to any one of the chip layout defect types in the first chip solution cluster; Marking the chip layout defect types whose solution amount ratio is not less than a predetermined reference ratio among the C chip layout defect types, so as to obtain the chip layout defect type corresponding to the target chip layout solution; Based on the chip layout defect type corresponding to the target chip layout scheme, a corresponding related defective chip layout scheme is determined among the multiple defective chip layout schemes, and an optimized chip layout scheme corresponding to the related defective chip layout scheme is used as a basis for optimizing the defect problem existing in the target chip layout scheme.

9. The multi-layer interconnect chip design method according to claim 8, It is characterized in that The step of performing a first clustering operation and a second clustering operation on the ordered set of chip solution members according to the A set types and the chip solution members included in each set type, and acquiring first clustering size information of the first clustering operation and second clustering size information of the second clustering operation includes: Determine a type movement possibility parameter between any one of the A type of sets and other type of sets, and output corresponding A type movement possibility parameters, wherein the other type of sets are A-1 type of sets other than the any one type of sets; Based on the A types of movement possibility parameters, determining corresponding first cluster size information; Determine the distribution parameters of each chip solution member in any one of the set types, and determine the in-set clustering size information of any one of the set types according to the distribution parameters of each chip solution member in the any one of the set types and the type movement possibility parameters between the any one of the set types and other set types, and output the in-set clustering size information of the corresponding A set types; Based on the intra-cluster size information of the A set types, corresponding second cluster size information is obtained.

10. The multi-layer interconnect chip design method according to claim 9, It is characterized in that The step of optimizing the A set types and the chip solution members included in each set type according to the goal of reducing the target cluster size information includes: Optimizing the A types of movement possibility parameters according to the goal of reducing the target cluster size information; According to the optimized A category movement possibility parameters, the number of chip solution members corresponding to each set category in the optimized A set categories is obtained; According to the order relationship of each chip solution member in the ordered set of chip solution members from front to back and the number of chip solution members included in each set type in the optimized A set types, the ordered set of chip solution members is aggregated to form A set types.