DDR package structure optimization method and system

Abstract

This application provides a method and system for optimizing DDR packaging structures. In this method, physical field performance simulation is performed on the first target packaging parameters of the target DDR to quantify the influence of these parameters on physical field performance. Then, it is determined whether the simulation data satisfies the set of physical field constraints to locate the first unbalanced physical field that does not meet the constraints. Inter-field coupling analysis is then performed on this first unbalanced physical field to obtain a first unbalanced physical field combination. Furthermore, the combination of the first unbalanced physical field combination and the coupling analysis of the first target packaging parameters determines a set of packaging adjustment regions. This allows the region division for packaging optimization to be associated with the coupling between multiple physical fields, rather than optimizing a single physical field or packaging region in isolation. This avoids cross-field coupling conflicts and reduced structural reliability caused by optimizing a single physical field or region, effectively improving the overall physical field performance of the DDR packaging structure.

Description

Technical Field

[0001] This application relates to the field of semiconductor packaging technology, and in particular to a method and system for optimizing DDR packaging structure. Background Technology

[0002] With the rapid development of the digital economy, the demand for DDR (Double Data Rate) memory is gradually shifting towards high speed, high capacity, and miniaturization, which places more stringent requirements on the design of DDR package structures. However, current optimization methods for DDR memory package structures typically focus on optimizing a single physical field. For example, to improve the integrity of signal transmission within DDR memory, micro-bump layouts that shorten signal paths or impedance-matched wiring designs are often used. To address memory heat dissipation, buried copper layers or external metal heat sinks are added inside the package. To achieve memory miniaturization, the linewidth or bump spacing of the RDL (Redistribution Layer) is reduced. However, these isolated optimization methods are difficult to reconcile. For instance, while buried copper layers can effectively conduct heat, the spatial overlap between the copper layer and the signal layer can easily cause electromagnetic coupling, leading to increased signal jitter and ultimately reducing signal integrity. Furthermore, when reducing the micro-bump spacing to decrease signal delay, the increased bump density exacerbates thermal stress concentration, leading to solder joint fatigue failure and affecting the reliability of the package structure. Therefore, current DDR is limited by its packaging method, resulting in certain performance bottlenecks. Summary of the Invention

[0003] Based on the above problems, in order to solve the performance bottleneck problem of DDR due to the limitation of packaging method, this application provides a method and system for optimizing DDR packaging structure.

[0004] The embodiments of this application disclose the following technical solutions: In a first aspect, embodiments of this application provide a method for optimizing a DDR packaging structure, the method comprising: Physical field performance simulation is performed based on the first target package parameters of the target DDR to obtain multiple sets of first physical field performance simulation data; the first physical field performance simulation data is used to quantify the influence relationship of the first target package parameters on physical field performance; Determine whether the simulation data of the first physical field performance of each group meets the pre-set set of physical field constraints; If it is determined that any of the first physical field performance simulation data cannot satisfy the set of physical field constraints, an inter-field coupling analysis is performed based on the first unbalanced physical field that does not satisfy the set of physical field constraints to obtain a first unbalanced physical field combination; the first unbalanced physical field combination is used to characterize physical fields whose coupling strength with the first unbalanced physical field is greater than the coupling strength threshold. Based on the first imbalanced physical field combination and the first target packaging parameters, a set of packaging adjustment regions for the target DDR is determined through coupling analysis. Based on the set of encapsulation adjustment regions and the first combination of unbalanced physical fields, the first target encapsulation parameters are optimized in a hierarchical and collaborative manner to obtain the second target encapsulation parameters.

[0005] In one possible implementation, the step of determining a set of package adjustment regions for the target DDR based on the coupled analysis of the first imbalanced physical field combination and the first target package parameters includes: Based on the first target packaging parameters, a packaging region analysis is performed to determine multiple target packaging regions for the target DDR. Calculate the comprehensive field response score of each target packaging region relative to the first unbalanced physical field combination; the comprehensive field response score is used to characterize the influence relationship of all packaging parameters in the region on the physical field performance when they act together; as well as, Calculate the field coupling-mediated score of each target packaging region relative to the first unbalanced physical field combination; the field coupling-mediated score is used to characterize the influence relationship of all packaging parameters in the region on the cross-field coupling strength when they act together; The set of encapsulation adjustment regions is constructed based on the comprehensive score of the field response and the field coupling-mediated score of each target encapsulation region relative to the first unbalanced physical field combination.

[0006] In one possible implementation, constructing the set of encapsulation adjustment regions based on the comprehensive score of the field response of each of the target encapsulation regions relative to the first imbalanced physical field combination and the field coupling-mediated score includes: The high field response interval and the high field coupling interval are determined by ranking the field response comprehensive score and the field coupling-mediated score of each target encapsulation region relative to the first unbalanced physical field combination; the high field response interval and the high field coupling interval are the intervals whose score ranking is higher than the ranking threshold. The target encapsulation region that falls simultaneously into the high-field response range and the high-field coupling range is determined as the main conflict optimization region. as well as, The target encapsulation region that falls only into the high field response range or the high field coupling range is determined as the secondary conflict adaptation region; as well as, The target encapsulation region that does not fall within the high field response range and the high field coupling range is determined as the fine-tuning region; Based on the main conflict optimization region, the secondary conflict adaptation region, and the fine-tuning region, the encapsulation adjustment region set is constructed.

[0007] In one possible implementation, the step of performing hierarchical collaborative optimization of the first target encapsulation parameters based on the encapsulation adjustment region set and the first imbalanced physical field combination to obtain the second target encapsulation parameters includes: Based on the first unbalanced physical field combination, a coupled simulation is performed to obtain a cross-physical field coupling model for the first unbalanced physical field combination; the cross-physical field coupling model is used to characterize the influence of the change in the packaging parameters within a unit amplitude on the physical field performance; Performance simulations were performed based on the cross-physics coupling model to determine several main region encapsulation parameters whose SOBOL exponents are greater than an exponent threshold within the main conflict optimization region. The SOBOL exponent is used to characterize the sensitivity of the encapsulation parameters to the physical field performance. The main region encapsulation parameters are parameters within the first target encapsulation parameters. Based on the target constraint conditions corresponding to the first unbalanced physical field combination in the physical field constraint condition set, the parameters of multiple main region encapsulation parameters are optimized to obtain multiple main region optimization parameters. Based on multiple main region optimization parameters, the auxiliary region encapsulation parameters in the auxiliary conflict adaptation region are optimized to obtain auxiliary region optimization parameters. Based on the field response comprehensive score, the fine-tuning region encapsulation parameters in the fine-tuning region are optimized to obtain fine-tuning region optimization parameters. The main region optimization parameters, the auxiliary region optimization parameters, and the fine-tuning region optimization parameters are determined as the second target encapsulation parameters.

[0008] In one possible implementation, the step of performing parameter-assisted optimization on the secondary region encapsulation parameters within the secondary conflict adaptation region based on multiple primary region optimization parameters to obtain secondary region optimization parameters includes: Calculate the robust solution of the main region corresponding to multiple main region optimization parameters; Based on the robust solution of the main region and the encapsulation parameters of the auxiliary region, a nonlinear correlation model between the main conflict optimization region and the auxiliary conflict adaptation region is constructed according to the field coupling-mediated score corresponding to each of the two. The nonlinear correlation model is used to characterize the interaction relationship between each of the main region optimization parameters and each of the auxiliary region encapsulation parameters. Based on the aforementioned nonlinear correlation model, parameter constraint analysis is performed to obtain a collaborative constraint factor; the collaborative constraint factor is used to calibrate the parameter adjustment range of the auxiliary region encapsulation parameters. Based on the high-field response interval or high-field coupling interval into which the secondary conflict adaptation region falls, the secondary region encapsulation parameters are optimized according to the nonlinear correlation model and the cooperative constraint factor to obtain the secondary region optimization parameters.

[0009] In one possible implementation, the parameter optimization of multiple main region encapsulation parameters is performed based on the target constraint conditions corresponding to the first imbalanced physical field combination in the physical field constraint condition set, to obtain multiple main region optimization parameters, including: Determine the hard and soft constraints in the target constraints; The hard constraints are converted into parameter optimization boundaries, and the soft constraints are transformed into functions to obtain the target optimization function for all the encapsulated parameters of the main regions. Based on the NSGA-II algorithm, the optimal parameter solution set is obtained by multi-objective optimization of multiple main region encapsulation parameters through the parameter optimization boundary and the objective optimization function; the NSGA-II algorithm is a multi-objective optimization algorithm. Determine the robust solution in the optimal parameter solution set, and determine the encapsulation parameter values ​​in the robust solution as the main region optimization parameters.

[0010] In one possible implementation, the step of optimizing the fine-tuning region encapsulation parameters within the fine-tuning region based on the comprehensive field response score to obtain optimized fine-tuning region parameters includes: Extract the field response score of the encapsulation parameter for each of the fine-tuning regions; the field response score is used to characterize the influence of a unit amplitude change of a single encapsulation parameter within the region on the physical field performance; Based on the field response score of the encapsulation parameter of each fine-tuning region, a target fine-tuning parameter is determined; the target fine-tuning parameter is the parameter whose field response score is less than the fine-tuning score threshold. The target fine-tuning parameters are optimized based on the field response score to obtain the optimized parameters for the fine-tuning region; the preset parameter fine-tuning range is determined based on the process calibration parameters of the target DDR.

[0011] In one possible implementation, the step of optimizing the target fine-tuning parameters based on the field response score to obtain the optimized parameters for the fine-tuning region includes: Obtain the packaging process parameters of the target DDR; Based on the packaging process parameters of the target DDR, a process deviation analysis is performed to determine the parameter adjustment range for the target fine-tuning parameters, and the fine-tuning step size for the target fine-tuning parameters is determined based on the field response score; the field response score is negatively correlated with the fine-tuning step size; Based on the parameter adjustment range and the fine-tuning step size, the target fine-tuning parameters are optimized to obtain the optimized parameters for the fine-tuning region.

[0012] Secondly, embodiments of this application provide a DDR packaging structure optimization system, the system comprising: The performance simulation module is used to perform physical field performance simulation based on the first target package parameters of the target DDR, and obtain multiple sets of first physical field performance simulation data; the first physical field performance simulation data is used to quantify the influence relationship of the first target package parameters on the physical field performance; The determination module is used to determine whether the first physical field performance simulation data of each group meets the preset set of physical field constraints. The first coupling analysis module is used to perform inter-field coupling analysis on a first unbalanced physical field that does not meet the set of physical field constraints when it is determined that any of the first physical field performance simulation data cannot meet the set of physical field constraints, and obtain a first unbalanced physical field combination; the first unbalanced physical field combination is used to characterize physical fields whose coupling strength with the first unbalanced physical field is greater than the coupling strength threshold. The second coupling analysis module is used to perform coupling analysis based on the first imbalance physical field combination and the first target packaging parameters to determine the set of packaging adjustment regions for the target DDR. The parameter optimization module is used to perform hierarchical collaborative optimization of the first target encapsulation parameters based on the encapsulation adjustment region set and the first imbalance physical field combination to obtain the second target encapsulation parameters.

[0013] In one possible implementation, the second coupling analysis module is specifically used for: Based on the first target packaging parameters, a packaging region analysis is performed to determine multiple target packaging regions for the target DDR. Calculate the comprehensive field response score of each target packaging region relative to the first unbalanced physical field combination; the comprehensive field response score is used to characterize the influence relationship of all packaging parameters in the region on the physical field performance when they act together; as well as, Calculate the field coupling-mediated score of each target packaging region relative to the first unbalanced physical field combination; the field coupling-mediated score is used to characterize the influence relationship of all packaging parameters in the region on the cross-field coupling strength when they act together; The set of encapsulation adjustment regions is constructed based on the comprehensive score of the field response and the field coupling-mediated score of each target encapsulation region relative to the first unbalanced physical field combination.

[0014] Compared with existing technologies, this application has the following beneficial effects: This application provides a method and system for optimizing DDR packaging structures. In this method, physical field performance simulation is performed on the first target packaging parameters of the target DDR to quantify the influence of parameters on physical field performance. Then, it is determined whether the simulation data satisfies the set of physical field constraints to locate the first unbalanced physical field that does not meet the constraints. Inter-field coupling analysis is then performed on the first unbalanced physical field to obtain the first unbalanced physical field combination, thus realizing the identification of coupling relationships between physical fields. This approach is not limited to a single physical field and optimizes packaging parameters based on the inter-field coupling relationship. Furthermore, the combination of the first unbalanced physical field combination and the coupling analysis of the first target packaging parameters determines the set of packaging adjustment regions. This allows the region division for packaging optimization to be associated with the coupling between multiple physical fields, rather than optimizing a single physical field or packaging region in isolation. Finally, based on the set of packaging adjustment regions and the first unbalanced physical field combination, hierarchical collaborative optimization is performed on the first target packaging parameters to achieve collaborative optimization and adjustment of multiple physical fields and multiple packaging regions. This avoids cross-field coupling conflicts and reduced structural reliability caused by optimization of a single physical field or region, effectively improving the overall physical field performance of the DDR packaging structure. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A flowchart illustrating a DDR packaging structure optimization method provided in this application embodiment; Figure 2 A flowchart illustrating a method for determining a set of encapsulation adjustment regions provided in an embodiment of this application; Figure 3 This is a schematic diagram illustrating a hierarchical collaborative optimization process for the first target encapsulation parameters, provided as an embodiment of this application. Figure 4 This is a schematic diagram of a DDR packaging structure optimization system provided in an embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with specific embodiments and accompanying drawings. It should be particularly noted that the embodiments described in this application are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0018] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0019] As mentioned earlier, current optimization methods for DDR memory packaging structures typically focus on optimizing a single physical field. For example, to improve the integrity of signal transmission within DDR memory, micro-bump layouts that shorten signal paths or impedance-matched wiring designs are commonly used. To address memory heat dissipation, buried copper layers or external metal heat sinks are added inside the package. To achieve memory miniaturization, RDL linewidth or bump spacing is reduced. However, these isolated optimization methods are difficult to reconcile. For instance, while buried copper layers effectively conduct heat, the spatial overlap between the copper layer and the signal layer can easily induce electromagnetic coupling, leading to increased signal jitter and ultimately reducing signal integrity. Conversely, when reducing micro-bump spacing to decrease signal delay, the increased bump density exacerbates thermal stress concentration, causing solder joint fatigue failure and affecting the reliability of the packaging structure. Therefore, current DDR memory, constrained by its packaging methods, suffers from certain performance bottlenecks.

[0020] To address this issue, this application provides a method and system for optimizing DDR packaging structures. In this method, physical field performance simulation is performed on the first target packaging parameters of the target DDR to quantify the influence of parameters on physical field performance. Then, it is determined whether the simulation data satisfies the set of physical field constraints to locate the first unbalanced physical field that does not meet the constraints. Inter-field coupling analysis is then performed on the first unbalanced physical field to obtain a first unbalanced physical field combination, thus identifying the coupling relationships between physical fields. This approach is not limited to a single physical field and optimizes packaging parameters based on the inter-field coupling relationships. Furthermore, a set of packaging adjustment regions is determined by combining the first unbalanced physical field combination and the first target packaging parameter coupling analysis. This allows the region division for packaging optimization to be associated with the coupling between multiple physical fields, rather than optimizing a single physical field or packaging region in isolation. Finally, hierarchical collaborative optimization is performed on the first target packaging parameters based on the set of packaging adjustment regions and the first unbalanced physical field combination. This achieves collaborative optimization and adjustment of multiple physical fields and multiple packaging regions, avoiding cross-field coupling conflicts and reduced structural reliability caused by optimizing a single physical field or a single region, effectively improving the overall physical field performance of the DDR packaging structure.

[0021] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0022] See Figure 1 The figure is a flowchart illustrating a DDR packaging structure optimization method provided in an embodiment of this application, specifically including the following steps: S101: Perform physical field performance simulation based on the first target package parameters of the target DDR to obtain multiple sets of first physical field performance simulation data; the first physical field performance simulation data is used to quantify the influence relationship between the first target package parameters and the physical field performance.

[0023] The first step in optimizing the packaging structure of a target DDR is to perform physical field performance simulation using the target DDR's primary packaging parameters. This simulation begins by determining the core physical field types to be simulated based on the target DDR's packaging design specifications and application scenarios. These core physical fields primarily include electromagnetic fields, thermal fields, and structural stress fields—key physical fields in DDR packaging. A corresponding physical field performance simulation model is then built using finite element method (FEM) tools, and the primary target packaging parameters are imported into the model. These primary target packaging parameters include core design parameters for DDR packaging such as RDL linewidth, bump spacing, buried copper layer thickness, dielectric constant of the dielectric layer, and solder joint layout. Subsequently, multiple different combinations of values ​​are set for each packaging parameter to simulate typical operating conditions of DDR, such as different operating temperatures, load currents, and operating frequencies. Performance simulations are then performed on individual physical fields, taking into account the initial coupling characteristics between fields. After completing the initial performance simulation, the physical field performance indicators corresponding to each parameter combination are collected, including electromagnetic field signal jitter and impedance matching, thermal field chip junction temperature and temperature distribution uniformity, stress field package warpage and solder joint thermal stress, etc., ultimately forming multiple sets of first physical field performance simulation data. In one possible implementation, the collected data can also be simultaneously quantized and calibrated to allow the data to more intuitively reflect the specific impact of changes in package parameters on physical field performance, ensuring the quantitative correlation of the data.

[0024] The purpose of this step is to provide a quantitative analysis basis for DDR packaging optimization. On the one hand, multiple sets of first-physics performance simulation data can clearly characterize the mapping relationship between the first target packaging parameters and physical field performance. For example, by analyzing the simulation results, it can be clearly seen that a 10% reduction in bump pitch will lead to a 15dB increase in electromagnetic coupling noise, or that a 5-micron increase in copper layer thickness will reduce the junction temperature by 8°C, thus identifying the key parameters in the packaging parameters that play a dominant role in performance. On the other hand, the physical field performance simulation data generated based on the packaging parameters provides a data benchmark for verification with physical field constraints in subsequent steps. By comparing specific constraints (such as a predetermined maximum junction temperature), unsatisfactory packaging parameters or combinations of packaging parameters can be quickly located, and then cross-field coupling analysis can be carried out for the unbalanced physical field.

[0025] S102: Determine whether the simulation data of the first physical field performance of each group meets the pre-set set of physical field constraints.

[0026] After obtaining the first physical field performance simulation data, the system combines it with a pre-defined set of physical field constraints to determine whether the current first target packaging parameters of the target DDR will negatively affect the physical field performance of the target DDR. In this embodiment, if the first physical field performance simulation data can satisfy all the constraints in the pre-defined set of physical field constraints, it indicates that the current packaging parameters of the DDR will not negatively affect the physical field performance. Conversely, it proves that the current packaging parameters of the DDR need to be adjusted.

[0027] In this embodiment, the set of physical field constraints is formulated by combining the product design specifications of DDR packaging, the rated operating conditions of DDR in actual applications, and extreme operating conditions. This set is a performance constraint standard for physical fields such as electromagnetic fields, thermal fields, and structural stress fields, and is also the core basis for determining whether the physical field performance of DDR packaging meets the standards. This set is based on hard constraints with no room for flexible adjustment, and represents the minimum performance that DDR packaging must meet. Different DDR models (such as DDR4 and DDR5) will adapt and adjust specific thresholds according to their own high-speed and high-capacity design requirements. Specifically, electromagnetic fields and thermal fields are used as examples in the set of physical field constraints. Examples of electromagnetic field constraints include signal jitter ≤50ps, impedance matching degree ≥90%, and electromagnetic radiation value ≤30dBμV / m. These constraints are used to ensure the integrity of DDR signal transmission and electromagnetic compatibility. Examples of thermal field constraints include chip operating junction temperature ≤125℃, maximum internal temperature difference of the package ≤20℃, and heat dissipation efficiency ≥85%. These constraints are used to adapt to the heat dissipation requirements of DDR under high-speed operation and avoid performance degradation due to excessive temperature. In addition, the set of physical field constraints also covers the key performance indicators of each physical field, ensuring that all dimensions for judging the physical field performance of DDR packaging are covered.

[0028] The comparison process between the simulation data of the first physical field performance and the set of physical field constraints first requires preprocessing multiple sets of simulation data. On one hand, performance index data from different physical fields and units of measurement are normalized to unify the quantification dimensions. On the other hand, the simulation data is validated to remove abnormal data caused by deviations in the simulation conditions, ensuring that the data participating in the comparison truly reflects the physical field performance corresponding to the encapsulation parameters. Then, a group-by-group, index-by-index comparison method is adopted. Each performance index of the electromagnetic field, thermal field, and structural stress field in each set of simulation data is compared with the corresponding index threshold in the set of physical field constraints to clarify the satisfaction status of individual indices. After this, a comprehensive judgment is made on each set of data. If any physical field index in a set of simulation data fails to meet the corresponding threshold in the set of constraints, the entire set of simulation data is determined to fail to meet the constraint requirements. Only when all physical field indices in a set of data meet the constraint thresholds is the set of physical field constraints considered to be satisfied. Finally, the comparison results of all groups are summarized, and the groups that do not meet the constraints, the corresponding abnormal indicators, and the specific degree of deviation of the indicators from the threshold are marked, so as to provide data basis for locating the first unbalanced physical field and the inter-field coupling analysis in the subsequent process.

[0029] S103: If it is determined that any of the first physical field performance simulation data cannot satisfy the set of physical field constraints, perform inter-field coupling analysis based on the first unbalanced physical field that does not satisfy the set of physical field constraints to obtain a first unbalanced physical field combination; the first unbalanced physical field combination is used to characterize physical fields whose coupling strength with the first unbalanced physical field is greater than the coupling strength threshold.

[0030] When it is determined that the performance simulation data of the first physical field cannot meet the set of physical field constraints, the first step is to locate the first unbalanced physical field that does not meet the constraints. This is the physical field whose performance indicators directly fail to reach the thresholds corresponding to the set of physical field constraints. Common examples include the junction temperature of the thermal field chip exceeding the limit or the electromagnetic field signal jitter exceeding the standard. Based on this, further inter-field coupling analysis needs to be carried out based on this first unbalanced physical field to determine the physical fields with strong coupling relationships with the first unbalanced physical field, ultimately forming the first unbalanced physical field combination.

[0031] It is important to note that in the first imbalanced physical field combination, besides the first imbalanced physical field itself, the physical fields corresponding to the first imbalanced physical field in the combination may be those that satisfy the constraint conditions themselves but have a high coupling strength with the first imbalanced physical field. That is, the first imbalanced physical field combination does not only include the first imbalanced physical field itself, but also other physical fields whose performance indicators meet the requirements of the physical field constraint set, but whose calculated coupling strength with the first imbalanced physical field exceeds the coupling strength threshold. For example, if the thermal field is the first imbalanced physical field and its junction temperature exceeds the limit, while the electromagnetic field's signal jitter, impedance matching, and other indicators meet the constraints, but simulation calculations show that changes in the thermal field temperature significantly exacerbate the electromagnetic coupling between the fields, and the coupling strength between the two physical fields exceeds the pre-set coupling strength threshold, then the electromagnetic field in the target DDR will still be included in the first imbalanced physical field combination. While these physical fields currently meet performance standards, their strong coupling with the first unbalanced physical field means that performance anomalies in the first unbalanced physical field can have a significant potential negative impact. Optimizing only the first unbalanced physical field can easily lead to performance degradation in these highly coupled physical fields, ultimately causing optimization failure. Therefore, incorporating these physical fields into the combination allows for a holistic consideration of the DDR packaging physical field imbalance problem, defining the critical physical field range for subsequent coupling analysis and hierarchical collaborative optimization, thereby avoiding cross-field performance conflicts caused by optimizing a single physical field.

[0032] In practical analysis of inter-field coupling of the first unbalanced physical field, a quantitative analysis model of inter-field coupling strength adapted to the characteristics of DDR packaging needs to be built first. Quantitative indicators such as coupling coefficient and inter-field interference are introduced. The first unbalanced physical field is then subjected to pairwise coupling simulation analysis with all other physical fields involved in the DDR packaging, including electromagnetic fields, thermal fields, and structural stress fields. Combined with the actual operating parameters of DDR (such as load, temperature, and frequency), the interaction trends of different physical fields during changes in packaging parameters are simulated to calculate the actual coupling strength value between the first unbalanced physical field and each other physical field. Furthermore, the coupling strength threshold can be determined based on DDR packaging design standards and inter-field coupling influence judgment specifications, combined with the degree of influence of coupling on physical field performance in practical applications. This threshold serves as the basis for determining whether there is a strong coupling correlation between physical fields, providing a clear standard for subsequent screening of highly coupled physical fields. This embodiment does not limit the method for determining the coupling strength threshold, and therefore will not elaborate further.

[0033] S104: Based on the first imbalanced physical field combination and the first target packaging parameters, perform coupling analysis to determine the set of packaging adjustment regions for the target DDR.

[0034] After identifying the first imbalanced physical field combination that exists in the target DDR and has the risk of imbalance, this step aims to map the abstract coupling problem of the first imbalanced physical field combination to the actual packaging physical region of the target DDR. This transforms the optimization requirements at the physical field level into specific packaging region adjustment requirements, thus defining a clear operational scope for packaging parameter optimization. The packaging adjustment region set is a set of parameter optimization regions for the target DDR packaging. This set excludes packaging regions that are unrelated or have a very low correlation with the first imbalanced physical field combination, only including packaging regions whose parameter adjustments can directly or indirectly improve the field performance of the imbalanced physical field combination and reduce the inter-field coupling strength. The set consists of multiple packaging regions with significant coupling correlation to the imbalanced physical field combination, and its range and boundaries are precisely defined based on the quantitative results of coupling analysis, without ambiguous region divisions. In this embodiment, the determination of the packaging adjustment region set ensures that the packaging parameter optimization work focuses on the core region, avoiding parameter adjustments across the entire domain and improving the optimization efficiency of packaging parameters.

[0035] Next, with reference to the accompanying drawings of a specific process embodiment, we will introduce the process of determining the set of packaging adjustment regions by combining the first unbalanced physical field combination and the first target packaging parameters in this step.

[0036] See Figure 2 The figure is a flowchart illustrating a method for determining a set of encapsulation adjustment regions according to an embodiment of this application, specifically including the following steps: S1041: Based on the first target packaging parameters, perform packaging region analysis to determine multiple target packaging regions for the target DDR.

[0037] Before determining the set of packaging adjustment areas that need focused optimization, it is first necessary to identify all areas within the target DDR that require packaging, i.e., the target packaging areas of the target DDR. In determining the target packaging areas, it is first necessary to clarify the type, value specifications, and corresponding physical locations of various primary target packaging parameters. This clarifies the actual placement and function of each parameter within the DDR packaging structure. For example, parameters such as RDL linewidth and spacing correspond to the wiring layer (target packaging area), bump spacing and diameter correspond to the interconnect area, and buried copper layer thickness and distribution correspond to the heat dissipation area. Subsequently, based on the structural and functional zoning principles of DDR semiconductor packaging, and combined with the specific packaging design specifications of the target DDR, the overall packaging structure is decomposed into multiple areas with independent physical boundaries and dedicated functional positioning. Primary target packaging parameters that act on the same packaging function and are closely related in physical location are mapped to the same area, dividing the target packaging areas into regions such as RDL wiring areas, bump interconnect areas, and buried copper layer heat dissipation areas. Each final target packaging area has a clear physical range and corresponds to a dedicated set of primary target packaging parameters.

[0038] S1042: Calculate the comprehensive field response score of each target packaging region relative to the first unbalanced physical field combination; the comprehensive field response score is used to characterize the influence relationship of all packaging parameters in the region on the physical field performance when they work together.

[0039] The comprehensive field response score quantifies the impact and correlation of the synergistic effect of all encapsulation parameters within a single target encapsulation region on the overall physical field performance of the first unbalanced physical field combination. This score characterizes the comprehensive trend of performance changes in each physical field within the first unbalanced physical field combination caused by a unit amplitude change in all encapsulation parameters within the region. This score directly reflects the encapsulation region's potential to regulate the performance of the unbalanced physical field combination. The score value is positively correlated with the strength of the impact of synergistic parameter adjustments on physical field performance; a higher score indicates a more significant improvement or deterioration in the performance of the first unbalanced physical field combination due to synergistic parameter changes, while a lower score indicates a weaker impact.

[0040] The comprehensive field response score is calculated based on coupled simulation across physical fields. For each target encapsulation region, coupled physical field simulation is used to simulate the changes in operating conditions caused by a unit amplitude change in all encapsulation parameters within it. Simultaneously, performance index change data for each physical field in the first unbalanced physical field combination is collected under this change. This performance index change data includes information such as the magnitude and trend of the performance index changes. Subsequently, based on the degree of imbalance and importance weight of each physical field in the first unbalanced physical field combination, corresponding weights are assigned to the performance change indicators of each physical field, with higher weights assigned to physical fields with more severe imbalances. Finally, a weighted summation method is used to calculate the comprehensive field response score for the target encapsulation region.

[0041] Specifically, the formula for calculating the comprehensive field response score is as follows: ; In the formula, The overall score of the field response of the target encapsulation region is represented by m, which represents the number of physical fields contained in the first unbalanced physical field combination. This represents the weight coefficient of the j-th physical field. The more severe the imbalance, the larger the weight coefficient, and the sum of the weight coefficients of all physical fields is 1. This represents the change in the performance index of the j-th physical field after all packaging parameters within the target packaging area undergo a coordinated change of unit magnitude. This represents the baseline value of the performance index of the j-th physics field. This represents the relative rate of change of the performance of the j-th physical field.

[0042] In this way, the comprehensive field response score can identify target encapsulation regions that have a significant impact on physical field performance under the synergistic effect of parameters within the region, while also identifying regions with a weak impact, thereby eliminating regions that are not worth optimizing and improving the accuracy and efficiency of overall parameter optimization.

[0043] S1043: Calculate the field coupling-mediated score of each target packaging region relative to the first unbalanced physical field combination; the field coupling-mediated score is used to characterize the influence relationship of all packaging parameters in the region on the cross-field coupling strength when they act together.

[0044] In conjunction with the comprehensive field response score, the field coupling-mediated score quantifies the influence and control capability of the synergistic effect of all encapsulation parameters within a single target encapsulation region on the interfield coupling strength of the first imbalanced physical field combination. It characterizes the specific trend of change in interfield coupling strength between physical fields in the first imbalanced physical field combination caused by a unit-amplitude synergistic change in all parameters within the region. The score is positively correlated with the strength of the synergistic adjustment of all parameters within the region on the interfield coupling strength; a higher score indicates a more prominent effect of the synergistic change of parameters in that region on improving or suppressing the interfield coupling of the first imbalanced physical field combination, while a lower score indicates a weaker effect. As a complementary indicator to the comprehensive field response score, the field coupling-mediated score is also an important quantitative basis for subsequent screening of core encapsulation adjustment regions and the division of primary and secondary optimization regions.

[0045] The calculation of the field coupling-mediated score is similar to that of the field response comprehensive score, and it also requires simulation based on cross-physical field coupling. The score ranges from [0,1]. The higher the score, the more significant the cross-field coupling strength of the target encapsulation region on the first unbalanced physical field combination. The calculation formula for the field coupling-mediated score is as follows: ; In the formula, The field coupling-mediated score represents the target encapsulation region; n represents the number of inter-field coupling pairs in the first imbalanced physical field combination. This represents the weight coefficient of the t-th inter-field coupling pair. The stronger the coupling, the larger the weight coefficient, and the sum of all weight coefficients is 1. This represents the change in coupling strength of the t-th inter-field coupling pair after all packaging parameters within the target packaging region undergo a unit-amplitude coordinated change. This represents the baseline value of the coupling strength of the t-th inter-field coupling pair. This represents the relative rate of change of the coupling strength of the t-th inter-field coupling pair.

[0046] S1044: Construct the set of encapsulation adjustment regions based on the comprehensive score of the field response of each target encapsulation region relative to the first unbalanced physical field combination and the field coupling-mediated score.

[0047] Finally, by selecting encapsulation regions based on the comprehensive field response score and field coupling-mediated score of each target encapsulation region relative to the first imbalanced physical field combination, a set of encapsulation adjustment regions can be constructed. Specifically, the process of constructing the set of encapsulation adjustment regions in this step is achieved through the following three steps: Step 1: Based on the comprehensive score of the field response and the field coupling-mediated score of each target encapsulation region relative to the first unbalanced physical field combination, the scores are ranked to determine the high field response interval and the high field coupling interval; the high field response interval and the high field coupling interval are the intervals whose score ranking is higher than the ranking threshold.

[0048] In constructing the set of package adjustment regions, the overall field response scores of all target package regions are first independently sorted in descending order, and a ranking threshold is set. The intervals corresponding to scores ranking above this threshold are defined as high field response intervals. Package regions falling into this interval indicate that their parameter co-variations have a significant impact on the physical field performance of the first unbalanced physical field combination. Simultaneously, the field coupling-mediated scores of all target package regions are also independently sorted in descending order using the same ranking threshold. Again, the intervals corresponding to scores ranking above this threshold are defined as high field coupling intervals. Package regions within this interval represent those whose parameter co-variations have a prominent impact on cross-field coupling strength. The ranking thresholds for both types of intervals can be shared or adjusted according to the target DDR package specifications and application scenarios.

[0049] Step 2: The target encapsulation region that falls into both the high field response range and the high field coupling range is determined as the primary conflict optimization region; and the target encapsulation region that falls only into either the high field response range or the high field coupling range is determined as the secondary conflict adaptation region; and the target encapsulation region that does not fall into either the high field response range or the high field coupling range is determined as the fine-tuning region. Step 3: Based on the main conflict optimization region, the secondary conflict adaptation region, and the fine-tuning region, construct the encapsulation adjustment region set.

[0050] After determining the high-field response interval and the high-field coupling interval, each target encapsulation region is categorized based on its corresponding comprehensive field response score and field coupling-mediated score to divide the regions. Specifically, it is checked whether the two scores fall into the corresponding high-field response interval and high-field coupling interval, respectively. If both scores meet the interval assignment criteria, the region is directly identified as the primary conflict optimization region, completing the screening of the first type of region. Then, for the remaining target encapsulation regions not classified as primary conflict optimization regions, the score assignment is checked region by region. If only the comprehensive field response score falls into the high-field response interval, or only the field coupling-mediated score falls into the high-field coupling interval, either condition is met, and the region is identified as the secondary conflict adaptation region, completing the division of the second type of region. Finally, regions where both scores fall into the interval are identified as fine-tuning regions. These three types of regions are included in the set, thus completing the construction of the encapsulation adjustment region set.

[0051] In this embodiment, the primary conflict optimization region is the key region for DDR packaging optimization. This type of region has a significant impact on the overall physical field performance of the first unbalanced physical field combination and is the region most directly related to the physical field imbalance problem and cross-field coupling contradiction of DDR packaging. Parameter adjustments in this type of region can fundamentally improve the physical field performance and inter-field coupling state of the package. Therefore, this type of region is classified as the primary conflict optimization region and is the region that needs to be optimized first in the parameter optimization process. The parameter adjustments of the secondary conflict adaptation region need to be based on the parameter adjustments in the primary conflict optimization region. The secondary conflict adaptation region is the adaptation region for packaging optimization. This type of region can only play a significant role in one dimension of physical field performance improvement or cross-field coupling strength regulation, and cannot achieve dual regulation effects simultaneously. Therefore, in the parameter optimization process, large-scale parameter adjustments cannot be made independently. Adaptive adjustments must be made in conjunction with the parameter adjustments in the primary conflict optimization region to avoid causing new inter-field coupling problems due to individual adjustments. The fine-tuning region is a non-core region of packaging optimization. This type of region has no significant impact on the physical field performance and cross-field coupling strength of the first unbalanced physical field combination. Therefore, subsequent parameter optimization only requires fine-tuning of the packaging parameters in this region according to the actual packaging process requirements of DDR.

[0052] S105: Based on the set of encapsulation adjustment regions and the first unbalanced physical field combination, perform hierarchical collaborative optimization on the first target encapsulation parameters to obtain the second target encapsulation parameters.

[0053] Finally, based on the three types of regions categorized in the encapsulation adjustment region set and the clearly defined physical field types within the first imbalance physical field combination, the encapsulation parameters (i.e., the first target encapsulation parameters) within the aforementioned three types of regions are progressively optimized in a hierarchical and collaborative manner. This completes the optimization process of the encapsulation parameters and yields the second target encapsulation parameters. Next, the process of hierarchical and collaborative optimization of the first target encapsulation parameters in S105 will be described in conjunction with the accompanying drawings of a specific process embodiment.

[0054] See Figure 3 The figure is a schematic diagram of a process for hierarchical collaborative optimization of the first target encapsulation parameters provided in an embodiment of this application, specifically including the following steps: S1051: Based on the first unbalanced physical field combination, perform coupled simulation to obtain a cross-physical field coupling model for the first unbalanced physical field combination; the cross-physical field coupling model is used to characterize the influence of the change in packaging parameters within a unit amplitude on the physical field performance.

[0055] In this step, the cross-physics coupling model serves as the medium for quantifying the packaging parameters and physical field performance within the first imbalanced physical field combination. The construction of this model requires using the first imbalanced physical field combination as the core, combining the target DDR's packaging structure and the first target packaging parameters to build a basic framework, thereby reconstructing the characteristics of each physical field within the combination and the basic coupling relationships between fields. Based on this, a controllable perturbation of unit amplitude is applied to the packaging parameters, and multiple rounds of coupling simulation experiments are conducted. The response data of the physical field performance after parameter changes are collected, and these response data are preprocessed to uncover the correlation between parameter changes and physical field performance. After iterative optimization, the cross-physics coupling model is obtained.

[0056] S1052: Based on the cross-physics coupling model, performance simulation is performed to determine a number of main region encapsulation parameters whose SOBOL exponent is greater than the exponent threshold within the main conflict optimization region; the SOBOL exponent is used to characterize the sensitivity of the encapsulation parameters to the physical field performance; the main region encapsulation parameters are parameters within the first target encapsulation parameters.

[0057] Understandably, in practical applications, although the main conflict optimization region is identified as the area requiring priority parameter adjustment, not all encapsulation parameters need to be adjusted. Therefore, the purpose of this step is to select the encapsulation parameters with the highest impact on physical field performance from the main conflict optimization region and to focus on adjusting these encapsulation parameters, thereby improving the efficiency of parameter optimization.

[0058] In this step, the Sobol (Sobol Sensitivity Analysis) index is used to measure the impact of encapsulation parameters on physical field performance. The determination of the Sobol index for each type of encapsulation parameter is based on simulation data from the cross-physics coupling model and the principles of Sobol analysis. In determining the Sobol index, a reasonable range of values ​​for each type of encapsulation parameter is first defined as the input space for Sobol analysis. Then, uniform random sampling is performed within this input space to generate multiple sets of different parameter combinations. All samples are then substituted into the cross-physics coupling model to conduct performance simulations, obtaining the first unbalanced physical field combination performance index output for each set of samples. Based on the calculation logic of the Sobol index, the total variance of the physical field performance index is decomposed into the variance generated by the individual action of each parameter and the interaction variance generated by the interaction between parameters. The first-order Sobol index and the total-order Sobol index for each parameter are calculated respectively. The first-order index characterizes the sensitivity of a parameter individually to physical field performance, while the total-order index characterizes the combined sensitivity of the parameter itself and its interaction with other parameters. Finally, the calculated SOBO index is compared with the preset index threshold, and parameters with index values ​​greater than the threshold are selected as the final main region encapsulation parameters. S1053: Based on the target constraint conditions corresponding to the first unbalanced physical field combination in the physical field constraint condition set, the parameters of the multiple main region encapsulation parameters are optimized to obtain multiple main region optimization parameters.

[0059] After determining the main region encapsulation parameters that need to be focused on for adjustment within the main conflict optimization area, the target constraint is further used as the benchmark for parameter optimization to optimize the main region encapsulation parameters. Among them, the target constraint is the constraint used to trigger the first unbalanced physical field in the first combination of unbalanced physical fields. This constraint is a constraint that cannot be satisfied in the physical field and is the cause of the physical field imbalance. Therefore, it needs to be used as the benchmark for optimizing the main region encapsulation parameters.

[0060] Specifically, the process of optimizing the main region encapsulation parameters to obtain the main region optimization parameters is achieved through the following four steps: Step 1: Determine the hard constraints and soft constraints in the target constraints.

[0061] First, the constraints are categorized and defined. From the target constraints corresponding to the first unbalanced physical field combination, and considering the DDR package product design specifications, actual application conditions, and performance requirements, each constraint index is divided into hard constraints and soft constraints. Hard constraints are the rigid requirements that the DDR package physical field performance must meet, while soft constraints are flexible optimization goals set to improve the overall package performance on the basis of meeting the hard constraints. The entire process requires defining the specific scope, index thresholds, and judgment criteria for both types of constraints.

[0062] Step 2: Convert the hard constraints into parameter optimization boundaries, and perform function transformation based on the soft constraints to obtain the target optimization function for all encapsulated parameters of the main regions; Step 3: Based on the NSGA-II algorithm, perform multi-objective optimization on multiple main region encapsulation parameters through the parameter optimization boundary and the objective optimization function to obtain the optimal parameter solution set; the NSGA-II algorithm is a multi-objective optimization algorithm.

[0063] After classifying the constraints, the next step is to quantify and transform them into algorithms. First, the defined hard constraints are transformed into parameter optimization boundaries. Based on the performance requirements corresponding to the hard constraints, parameters are encapsulated for each principal region, and their value ranges are defined to ensure that parameter optimization always stays within this range. Next, the soft constraints undergo function transformation. Combining the weight of each soft constraint in the physics performance optimization and the correlation between the encapsulated parameters of each principal region and the soft constraint indicators, the abstract flexible optimization requirements are transformed into a target optimization function that the algorithm can directly recognize, ensuring that the function accurately maps the optimization needs of the soft constraints. Subsequently, based on the NSGA-II (Non-dominated Sorting Genetic Algorithm II) multi-objective optimization algorithm, the parameter optimization boundary is used as the constraint condition for the algorithm's solution, and the objective optimization function is used as the algorithm's optimization objective. Multiple principal region encapsulation parameters are imported into the algorithm, and genetic operations such as encoding, crossover, and mutation are completed according to the algorithm logic to obtain an initial solution group. Then, the solution group is processed in a hierarchical manner through non-dominated sorting, and high-quality solutions with uniform distribution are selected by combining crowding degree calculation. After multiple rounds of iterative evolution, Pareto optimal solutions that take into account the performance optimization requirements of multiple physics fields can finally be obtained within the parameter optimization boundary, forming the optimal parameter solution set.

[0064] Specifically, the process of determining the optimal parameter solution set based on the NSGA-II algorithm is implemented through the following formula: (1) In the formula, X represents the main region encapsulation parameter vector, m is the total number of main region encapsulation parameters, and F(X) is the objective optimization function set, generated by transforming soft constraints. ... Let be the objective optimization function for each physical field. The parameter optimization boundary for encapsulating the parameters of the i-th principal region is generated by transforming hard constraints. The main region encapsulates the value space of the parameter vector, where k is the number of objective optimization functions.

[0065] Set the optimal parameter solution set as , Let be the Pareto optimal solution set of formula (1).

[0066] Step 4: Determine the robust solution in the optimal parameter solution set, and determine the encapsulation parameter values ​​in the robust solution as the main region optimization parameters.

[0067] After obtaining the optimal parameter solution set, robustness analysis is performed on all Pareto optimal solutions in the set. Robust solutions are selected that maintain stable performance of the first unbalanced physics combination within a certain range of process deviations and environmental changes. Finally, the specific values ​​of the encapsulation parameters for each main region in the robust solution are directly determined as the final main region optimization parameters, completing the entire optimization process for the main region encapsulation parameters. In the optimization process for the main region encapsulation parameters, the above four steps are based on specific constraints. This ensures that the optimization of the main region encapsulation parameters strictly follows the target constraints, achieves collaborative optimization of multi-physics performance through a multi-objective algorithm, and, combined with the selection of robust solutions, adapts the optimization results to the engineering requirements of actual production, avoiding the problem of disconnect between simulation optimization and practical application, and effectively improving the accuracy of parameter optimization.

[0068] S1054: Perform parameter-assisted optimization on the secondary region encapsulation parameters in the secondary conflict adaptation region based on multiple main region optimization parameters to obtain secondary region optimization parameters, and perform fine-tuning parameter optimization on the fine-tuning region encapsulation parameters in the fine-tuning region based on the field response comprehensive score to obtain fine-tuning region optimization parameters. S1055: The main region optimization parameters, the auxiliary region optimization parameters, and the fine-tuning region optimization parameters are determined as the second target encapsulation parameters.

[0069] After optimizing the main region encapsulation parameters to obtain multiple main region optimization parameters, the secondary region encapsulation parameters are further optimized based on the optimization results of the main region optimization parameters to obtain secondary region optimization parameters. Specifically, the process of performing auxiliary parameter optimization on the secondary region encapsulation parameters to obtain secondary region optimization parameters is achieved through the following four steps: Step 1: Calculate the robust solution of the main region corresponding to the multiple main region optimization parameters.

[0070] As discussed earlier regarding the optimization logic of packaging parameters, optimizing the packaging parameters in the auxiliary region requires consideration of the optimization of parameters within the main region. In this step, the calculation of the robust solution for the main region essentially verifies the stability of the optimization results for the parameters in the main conflict optimization region. After completing the multi-objective optimization of the main region's packaging parameters, robust solutions with strong resistance to uncertainties such as process fluctuations need to be selected from the set of optimal parameter solutions. Robust solutions can be calculated using Monte Carlo simulation or sensitivity analysis to quantify the performance fluctuation amplitude of each parameter solution within a preset disturbance range. For example, the stability of thermomechanical reliability can be assessed by calculating the standard deviation of solder joint stress values, or the robustness of electromagnetic compatibility can be measured by the coefficient of variation of the signal eye diagram opening. The final selected robust solution for the main region must satisfy physical field constraints.

[0071] Step 2: Based on the robust solution of the main region and the encapsulation parameters of the auxiliary region, and according to the field coupling-mediated score corresponding to each of them, construct a nonlinear correlation model between the main conflict optimization region and the auxiliary conflict adaptation region; the nonlinear correlation model is used to characterize the interaction relationship between each of the main region optimization parameters and each of the auxiliary region encapsulation parameters.

[0072] Because physical field couplings (such as thermal-electromagnetic coupling and structure-signal coupling) in DDR packaging often exhibit nonlinear characteristics, traditional linear regression models struggle to accurately describe parameter interaction effects. Therefore, machine learning algorithms or physics-inspired nonlinear regression models are necessary. In the modeling process of nonlinear correlation models, the field coupling mediation score serves as a key input indicator, quantifying the contribution of parameters in each region to the cross-field coupling strength. For example, a high field coupling mediation score for a parameter in the auxiliary region indicates a significant impact on the electromagnetic-thermal coupling path, requiring a higher interaction weight in the model. By using the parameter values ​​of the robust solution in the main region and the initial configuration of the auxiliary region parameters as training samples, the model can learn the nonlinear relationship between the optimized parameters of the main region and the packaging parameters of the auxiliary region (e.g., the product effect between the increase in the thickness of the heat dissipation copper layer in the main region and the reduction in the spacing between the signal layers in the auxiliary region, leading to an exponential increase in coupling noise with changes in both). This provides an accurate correlation mapping for subsequent constraint analysis.

[0073] Step 3: Perform parameter constraint analysis based on the nonlinear correlation model to obtain the collaborative constraint factor; the collaborative constraint factor is used to calibrate the parameter adjustment range of the auxiliary region encapsulation parameters.

[0074] Based on the established nonlinear correlation model, the potential impact paths of auxiliary region parameter adjustments on the main region performance can be identified through sensitivity analysis or constraint propagation algorithms. For example, if the model shows that adjusting the RDL linewidth in the auxiliary region will have a feedback effect on the signal integrity index of the main region through the electromagnetic coupling path, then the maximum adjustment range of the RDL linewidth needs to be derived based on the performance tolerance of the robust solution in the main region (such as the maximum allowable compression of the eye diagram opening). The co-constraint factor, as the core parameter for quantifying this cross-regional constraint relationship, can be presented in the form of a proportional coefficient or an interval threshold. For example, the adjustment range of a parameter in the auxiliary region must not exceed ±K% of the corresponding robust solution parameter value in the main region, where K is determined by the output sensitivity of the correlation model and the process error tolerance. This constraint mechanism allows the auxiliary region parameters to be optimized within a reasonable range while preventing the optimized physical field performance in the main region from rebounding or becoming unbalanced due to over-adjustment.

[0075] Step 4: Based on the high-field response interval or high-field coupling interval into which the auxiliary conflict adaptation region falls, optimize the encapsulation parameters of the auxiliary region according to the nonlinear correlation model and the cooperative constraint factor to obtain the optimized parameters of the auxiliary region.

[0076] Finally, the optimization method for parameters is determined based on the type of interval into which the secondary conflict adaptation region falls. If the secondary region falls into the high-field response interval, it indicates that its parameters have a significant impact on the performance of a single physical field but a weak cross-field coupling effect. In this case, the optimization focus is on using a nonlinear correlation model to predict the response trend of the robust solution of the main region to changes in the parameters of the secondary region, and using a local search algorithm to find the optimal solution for the physical field performance within the range limited by the cooperative constraint factor. For example, under the premise of ensuring the thermal performance stability of the main region, the heat dissipation hole density of the secondary region can be optimized separately to further reduce the local hot spot temperature. For the secondary region falling into the high-field coupling interval, parameter adjustment may trigger complex cross-field interaction effects. Therefore, a multi-objective optimization algorithm is needed to simultaneously optimize multiple correlated physical field indices. For example, when adjusting the power supply and ground plane spacing, the optimized signal return path requirements of the main region and the newly added electromagnetic coupling suppression constraints must be satisfied simultaneously. In this process, the cooperative constraint factor, as a dynamic boundary condition, will adaptively adjust according to the parameter characteristics of the robust solution of the main region. For example, when the main region adopts a high-density bump layout, the bump spacing adjustment constraint of the secondary region will be more stringent to avoid exacerbating thermal stress concentration.

[0077] In step S1054, while determining the auxiliary region optimization parameters, it is also necessary to optimize the fine-tuning region encapsulation parameters within the fine-tuning region based on the comprehensive field response score. Specifically, the fine-tuning parameter optimization process is implemented through the following three steps: Step 1: Extract the field response score of the encapsulation parameter for each fine-tuning region; the field response score is used to characterize the influence of a unit amplitude change of a single encapsulation parameter within the region on the physical field performance.

[0078] The first step in optimizing the encapsulation parameters of the fine-tuning region is to extract the field response score corresponding to each encapsulation parameter of the fine-tuning region. This score is different from the comprehensive field response score that characterizes the synergistic effect of all parameters in the region. It is specifically used to quantify the independent influence of a single encapsulation parameter in the first unbalanced physical field composition on the physical field performance when a unit amplitude change occurs. It is a quantitative indicator for measuring the performance influence of a single fine-tuning parameter.

[0079] During the extraction process, data retrieval and calculation are carried out one by one for each encapsulation parameter in the fine-tuning area. Based on the previously constructed cross-physics coupling model and physics performance simulation data, combined with the unit amplitude change of a single parameter and the corresponding physics performance change data, a field response score for each parameter is obtained through standardized calculation methods to ensure that each score can accurately match the corresponding single encapsulation parameter.

[0080] Step 2: Determine the target fine-tuning parameter based on the field response score of the encapsulation parameter of each fine-tuning region; the target fine-tuning parameter is the parameter whose field response score is less than the fine-tuning score threshold.

[0081] After extracting the field response scores, the next step is to determine the target fine-tuning parameters. The core of this step is to use quantitative thresholds to screen out encapsulated parameters that are necessary and feasible for fine-tuning, thus defining specific targets for subsequent targeted optimization. The fine-tuning score threshold is the dividing line for judging the degree of influence of a single fine-tuning parameter on the physical field performance. Fine-tuning parameters with scores higher than this threshold mean that they still have some impact on the physical field performance, and to avoid performance fluctuations caused by adjustments, these parameters are excluded from the fine-tuning range. Subsequently, the field response scores of each extracted encapsulated parameter in the fine-tuning region are compared one by one with the preset fine-tuning score threshold. Encapsulated parameters with field response scores lower than this threshold are selected and determined as target fine-tuning parameters. These parameters have extremely weak influence on the physical field performance; even small adjustments will not have a significant adverse impact on the performance of the first unbalanced physical field combination, making them the core targets for conducting process-specific fine-tuning.

[0082] Step 3: Optimize the target fine-tuning parameters based on the field response score to obtain the optimized parameters for the fine-tuning region; the preset parameter fine-tuning range is determined based on the process calibration parameters of the target DDR.

[0083] When optimizing target fine-tuning parameters, the first step is to obtain the packaging process parameters of the target DDR. These parameters are process parameters from the actual DDR manufacturing process, encompassing the precision indicators of packaging equipment, process tolerances for each step, material processing adaptability, and other data relevant to the actual process implementation. Subsequently, based on the obtained packaging process parameters, process deviation analysis is conducted. Considering unavoidable equipment errors and process fluctuations in production, an adjustment range that aligns with actual production capabilities is defined for each target fine-tuning parameter. Simultaneously, the corresponding fine-tuning step size is determined based on the field response score. These two factors are negatively correlated: a lower field response score indicates a weaker impact of the parameter on physical field performance, allowing for a larger fine-tuning step size. Conversely, a score closer to the fine-tuning score threshold indicates a potentially slight impact on physical field performance, necessitating a smaller fine-tuning step size. This ensures the safety of parameter adjustments and avoids fluctuations in physical field performance caused by excessive adjustments.

[0084] After determining the parameter adjustment range and fine-tuning step size, specific fine-tuning optimization operations are carried out on the target fine-tuning parameters based on these two indicators. Within the defined adjustment range, the target fine-tuning parameters are gradually adjusted according to the determined fine-tuning step size. Each adjustment can be verified by simple physical field performance simulation based on the cross-physics coupling model to confirm that the adjusted parameters will not adversely affect the overall performance of the first unbalanced physical field combination, nor will they interfere with the optimization results of the main conflict optimization region and the auxiliary conflict adaptation region. After completing the adjustment and verification of a single parameter, the remaining target fine-tuning parameters are optimized sequentially according to the same rules until all target fine-tuning parameters have completed the adaptation adjustment. Finally, the specific values ​​of all optimized target fine-tuning parameters are determined as the fine-tuning region optimization parameters. These parameters not only meet the actual process production requirements of DDR packaging, but also form a good fit with the main region optimization parameters and the auxiliary region optimization parameters.

[0085] This application provides a method for optimizing DDR packaging structure. In this method, physical field performance simulation is performed on the first target packaging parameters of the target DDR to quantify the influence of parameters on physical field performance. Then, it is determined whether the simulation data satisfies the set of physical field constraints to locate the first unbalanced physical field that does not meet the constraints. Inter-field coupling analysis is then performed on the first unbalanced physical field to obtain a first unbalanced physical field combination, thus identifying the coupling relationships between physical fields. This method is not limited to a single physical field and optimizes packaging parameters based on the inter-field coupling relationships. Furthermore, a set of packaging adjustment regions is determined by combining the first unbalanced physical field combination and the first target packaging parameter coupling analysis. This allows the region division for packaging optimization to be associated with the coupling between multiple physical fields, rather than optimizing a single physical field or packaging region in isolation. Finally, hierarchical collaborative optimization is performed on the first target packaging parameters based on the set of packaging adjustment regions and the first unbalanced physical field combination. This achieves collaborative optimization and adjustment of multiple physical fields and multiple packaging regions, avoiding cross-field coupling conflicts and reduced structural reliability caused by optimizing a single physical field or a single region, effectively improving the overall physical field performance of the DDR packaging structure.

[0086] The following describes a DDR packaging structure optimization system provided in the embodiments of this application. The DDR packaging structure optimization system described below and the DDR packaging structure optimization method described above can be referred to and correspond to each other.

[0087] See Figure 4 The figure is a schematic diagram of a DDR packaging structure optimization system provided in an embodiment of this application, which specifically includes the following modules: The performance simulation module 100 is used to perform physical field performance simulation based on the first target package parameters of the target DDR, and obtain multiple sets of first physical field performance simulation data; the first physical field performance simulation data is used to quantify the influence relationship of the first target package parameters on the physical field performance. The determination module 200 is used to determine whether the first physical field performance simulation data of each group meets the preset set of physical field constraint conditions; The first coupling analysis module 300 is used to perform inter-field coupling analysis on a first unbalanced physical field that does not meet the set of physical field constraints when it is determined that any of the first physical field performance simulation data cannot meet the set of physical field constraints, and obtain a first unbalanced physical field combination; the first unbalanced physical field combination is used to characterize physical fields whose coupling strength with the first unbalanced physical field is greater than the coupling strength threshold. The second coupling analysis module 400 is used to perform coupling analysis based on the first imbalance physical field combination and the first target packaging parameters to determine the set of packaging adjustment regions for the target DDR. The parameter optimization module 500 is used to perform hierarchical collaborative optimization of the first target packaging parameters based on the packaging adjustment region set and the first imbalance physical field combination to obtain the second target packaging parameters.

[0088] In one possible implementation, the second coupling analysis module 400 is specifically used for: Based on the first target packaging parameters, a packaging region analysis is performed to determine multiple target packaging regions for the target DDR. Calculate the comprehensive field response score of each target packaging region relative to the first unbalanced physical field combination; the comprehensive field response score is used to characterize the influence relationship of all packaging parameters in the region on the physical field performance when they act together; as well as, Calculate the field coupling-mediated score of each target packaging region relative to the first unbalanced physical field combination; the field coupling-mediated score is used to characterize the influence relationship of all packaging parameters in the region on the cross-field coupling strength when they act together; The set of encapsulation adjustment regions is constructed based on the comprehensive score of the field response and the field coupling-mediated score of each target encapsulation region relative to the first unbalanced physical field combination.

[0089] It should be noted that the various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for methods and systems, since they are basically similar to the method embodiments, the description is relatively simple, and relevant parts can be referred to the description of the method embodiments. The methods and systems described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components indicated as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of the solution in this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0090] The above description is merely one specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for optimizing a DDR package structure, characterized in that, The method includes: Physical field performance simulation is performed based on the first target package parameters of the target DDR to obtain multiple sets of first physical field performance simulation data; the first physical field performance simulation data is used to quantify the influence relationship of the first target package parameters on physical field performance; Determine whether the simulation data of the first physical field performance of each group meets the pre-set set of physical field constraints; If it is determined that any of the first physical field performance simulation data cannot satisfy the set of physical field constraints, an inter-field coupling analysis is performed based on the first unbalanced physical field that does not satisfy the set of physical field constraints to obtain a first unbalanced physical field combination; the first unbalanced physical field combination is used to characterize physical fields whose coupling strength with the first unbalanced physical field is greater than the coupling strength threshold. Based on the first imbalanced physical field combination and the first target packaging parameters, a set of packaging adjustment regions for the target DDR is determined through coupling analysis. Based on the set of encapsulation adjustment regions and the first combination of unbalanced physical fields, the first target encapsulation parameters are optimized in a hierarchical and collaborative manner to obtain the second target encapsulation parameters.

2. The method of claim 1, wherein, The coupling analysis based on the first imbalanced physical field combination and the first target packaging parameters to determine the set of packaging adjustment regions for the target DDR includes: Based on the first target packaging parameters, a packaging region analysis is performed to determine multiple target packaging regions for the target DDR. Calculate the comprehensive field response score of each target packaging region relative to the first unbalanced physical field combination; the comprehensive field response score is used to characterize the influence relationship of all packaging parameters in the region on the physical field performance when they act together; as well as, Calculate the field coupling-mediated score of each target packaging region relative to the first unbalanced physical field combination; the field coupling-mediated score is used to characterize the influence relationship of all packaging parameters in the region on the cross-field coupling strength when they act together; The set of encapsulation adjustment regions is constructed based on the comprehensive score of the field response and the field coupling-mediated score of each target encapsulation region relative to the first unbalanced physical field combination.

3. The method of claim 2, wherein, The set of encapsulation adjustment regions is constructed based on the comprehensive score of the field response of each target encapsulation region relative to the first unbalanced physical field combination and the field coupling-mediated score, including: The high field response interval and the high field coupling interval are determined by ranking the field response comprehensive score and the field coupling-mediated score of each target encapsulation region relative to the first unbalanced physical field combination; the high field response interval and the high field coupling interval are the intervals whose score ranking is higher than the ranking threshold. The target encapsulation region that falls simultaneously into the high-field response range and the high-field coupling range is determined as the main conflict optimization region. as well as, The target encapsulation region that falls only into the high field response range or the high field coupling range is determined as the secondary conflict adaptation region; as well as, The target encapsulation region that does not fall within the high field response range and the high field coupling range is determined as the fine-tuning region; Based on the main conflict optimization region, the secondary conflict adaptation region, and the fine-tuning region, the encapsulation adjustment region set is constructed.

4. The method of claim 2, wherein, The step of performing hierarchical collaborative optimization of the first target encapsulation parameters based on the encapsulation adjustment region set and the first imbalance physical field combination to obtain the second target encapsulation parameters includes: Based on the first unbalanced physical field combination, a coupled simulation is performed to obtain a cross-physical field coupling model for the first unbalanced physical field combination; the cross-physical field coupling model is used to characterize the influence of the change in the packaging parameters within a unit amplitude on the physical field performance; Performance simulations were performed based on the cross-physics coupling model to determine several main region encapsulation parameters whose SOBOL exponents are greater than an exponent threshold within the main conflict optimization region. The SOBOL exponent is used to characterize the sensitivity of the encapsulation parameters to the physical field performance. The main region encapsulation parameters are parameters within the first target encapsulation parameters. Based on the target constraint conditions corresponding to the first unbalanced physical field combination in the physical field constraint condition set, the parameters of multiple main region encapsulation parameters are optimized to obtain multiple main region optimization parameters. Based on multiple main region optimization parameters, the auxiliary region encapsulation parameters in the auxiliary conflict adaptation region are optimized to obtain auxiliary region optimization parameters. Based on the field response comprehensive score, the fine-tuning region encapsulation parameters in the fine-tuning region are optimized to obtain fine-tuning region optimization parameters. The main region optimization parameters, the auxiliary region optimization parameters, and the fine-tuning region optimization parameters are determined as the second target encapsulation parameters.

5. The method according to claim 4, characterized in that, The step of performing parameter-assisted optimization on the secondary region encapsulation parameters within the secondary conflict adaptation region based on multiple primary region optimization parameters to obtain secondary region optimization parameters includes: Calculate the robust solution of the main region corresponding to multiple main region optimization parameters; Based on the robust solution of the main region and the encapsulation parameters of the auxiliary region, a nonlinear correlation model between the main conflict optimization region and the auxiliary conflict adaptation region is constructed according to the field coupling-mediated score corresponding to each of the two. The nonlinear correlation model is used to characterize the interaction relationship between each of the main region optimization parameters and each of the auxiliary region encapsulation parameters. Based on the aforementioned nonlinear correlation model, parameter constraint analysis is performed to obtain a collaborative constraint factor; the collaborative constraint factor is used to calibrate the parameter adjustment range of the auxiliary region encapsulation parameters. Based on the high-field response interval or high-field coupling interval into which the secondary conflict adaptation region falls, the secondary region encapsulation parameters are optimized according to the nonlinear correlation model and the cooperative constraint factor to obtain the secondary region optimization parameters.

6. The method according to claim 4, characterized in that, The first imbalanced physical field combination is used to optimize multiple main region encapsulation parameters according to the target constraint conditions corresponding to the physical field constraint condition set, resulting in multiple main region optimization parameters, including: Determine the hard and soft constraints in the target constraints; The hard constraints are converted into parameter optimization boundaries, and the soft constraints are transformed into functions to obtain the target optimization function for all the encapsulated parameters of the main regions. Based on the NSGA-II algorithm, the optimal parameter solution set is obtained by multi-objective optimization of multiple main region encapsulation parameters through the parameter optimization boundary and the objective optimization function; the NSGA-II algorithm is a multi-objective optimization algorithm. Determine the robust solution in the optimal parameter solution set, and determine the encapsulation parameter values ​​in the robust solution as the main region optimization parameters.

7. The method according to claim 4, characterized in that, The step of optimizing the fine-tuning region encapsulation parameters within the fine-tuning region based on the comprehensive field response score to obtain optimized fine-tuning region parameters includes: Extract the field response score of the encapsulation parameter for each of the fine-tuning regions; the field response score is used to characterize the influence of a unit amplitude change of a single encapsulation parameter within the region on the physical field performance; Based on the field response score of the encapsulation parameter of each fine-tuning region, a target fine-tuning parameter is determined; the target fine-tuning parameter is the parameter whose field response score is less than the fine-tuning score threshold. The target fine-tuning parameters are optimized based on the field response score to obtain the optimized parameters for the fine-tuning region; the preset parameter fine-tuning range is determined based on the process calibration parameters of the target DDR.

8. The method according to claim 7, characterized in that, The step of optimizing the target fine-tuning parameters based on the field response score to obtain the optimized parameters for the fine-tuning region includes: Obtain the packaging process parameters of the target DDR; Based on the packaging process parameters of the target DDR, a process deviation analysis is performed to determine the parameter adjustment range for the target fine-tuning parameters, and the fine-tuning step size for the target fine-tuning parameters is determined based on the field response score; the field response score is negatively correlated with the fine-tuning step size; Based on the parameter adjustment range and the fine-tuning step size, the target fine-tuning parameters are optimized to obtain the optimized parameters for the fine-tuning region.

9. A DDR packaging structure optimization system, characterized in that, The system includes: The performance simulation module is used to perform physical field performance simulation based on the first target package parameters of the target DDR, and obtain multiple sets of first physical field performance simulation data; the first physical field performance simulation data is used to quantify the influence relationship of the first target package parameters on the physical field performance; The determination module is used to determine whether the first physical field performance simulation data of each group meets the preset set of physical field constraints. The first coupling analysis module is used to perform inter-field coupling analysis on a first unbalanced physical field that does not meet the set of physical field constraints when it is determined that any of the first physical field performance simulation data cannot meet the set of physical field constraints, and obtain a first unbalanced physical field combination; the first unbalanced physical field combination is used to characterize physical fields whose coupling strength with the first unbalanced physical field is greater than the coupling strength threshold. The second coupling analysis module is used to perform coupling analysis based on the first imbalance physical field combination and the first target packaging parameters to determine the set of packaging adjustment regions for the target DDR. The parameter optimization module is used to perform hierarchical collaborative optimization of the first target encapsulation parameters based on the encapsulation adjustment region set and the first imbalance physical field combination to obtain the second target encapsulation parameters.

10. The system according to claim 9, characterized in that, The second coupling analysis module is specifically used for: Based on the first target packaging parameters, a packaging region analysis is performed to determine multiple target packaging regions for the target DDR. Calculate the comprehensive score of the field response of each target encapsulation region relative to the first unbalanced physical field combination; The comprehensive field response score is used to characterize the influence of all encapsulation parameters within the region on the physical field performance when they work together. as well as, Calculate the field coupling-mediated score of each target packaging region relative to the first unbalanced physical field combination; the field coupling-mediated score is used to characterize the influence relationship of all packaging parameters in the region on the cross-field coupling strength when they act together; The set of encapsulation adjustment regions is constructed based on the comprehensive score of the field response and the field coupling-mediated score of each target encapsulation region relative to the first unbalanced physical field combination.

Images