An artificial intelligence parasitic parameter analysis and automatic optimization method

Through artificial intelligence methods combined with parasitic capturers and multi-physics simulation software, the circuit layout is automatically adjusted, which solves the problem of low parasitic parameter optimization efficiency in circuit design, and improves circuit performance and design efficiency.

CN118627464BActive Publication Date: 2024-12-06青岛展诚科技有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410864941.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-28
Publication Date
2024-12-06
Estimated Expiration
2044-06-28

AI Technical Summary

Technical Problem

The prior art is difficult to efficiently capture and interpret complex relationships between parasitic parameters in circuit design, resulting in inefficient circuit performance optimization and difficult to automate processing.

Method used

Using artificial intelligence methods combined with parasitic capturers and multi-physics simulation software, the circuit layout is analyzed layer by layer, independent contribution factors are identified, and the multi-physics simulation software is used to simulate the interference of parasitic parameters on electrical performance, and the circuit layout and design are automatically adjusted. Parameters to minimize parasitic effects.

Benefits of technology

Improves the signal integrity and interference resistance of the circuit, automates the design process, and significantly improves the design efficiency and circuit performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118627464B_ABST
    Figure CN118627464B_ABST
Patent Text Reader

Abstract

The present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which belongs to the technical field of circuit simulation. The method comprises: using a parasitic capturer to determine the parasitic parameters of a circuit under different test states, and recording the operation data of the circuit during the capture process, thereby determining the parasitic situation; hierarchically dividing the circuit layout based on the operation data, analyzing the circuit layout layer by layer according to the levels, thereby identifying independent contribution factors; using the operation data to determine the multi-physics field simulation software to simulate the influence law of the independent contribution factors on the parasitic parameters and the interference of the parasitic parameters on the electrical performance; automatically adjusting the circuit layout and circuit-related design parameters according to the interference situation, influence law and parasitic situation, thereby minimizing the parasitic effect, improving the signal integrity and interference resistance of the circuit, and automating the adjustment to improve the design efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of circuit simulation, and in particular to an artificial intelligence parasitic parameter analysis and automatic optimization method. Background Art

[0002] As the complexity of electronic circuit design increases, parasitic parameters (such as parasitic capacitance, parasitic inductance, etc.) have an increasing impact on circuit performance. These parasitic effects may cause problems such as signal distortion, increased delay, increased power consumption, and poor frequency response. Currently, engineers usually need to manually adjust the circuit layout to reduce these parasitic effects. This process often uses circuit simulation software to assist in the design, and optimizes the circuit layout and parameters through experience and trial and error. However, this process is inefficient, and it is not easy to find the optimal solution, and it is difficult to fully capture and explain the complex relationship between various parasitic effects.

[0003] Therefore, the present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method. Summary of the invention

[0004] The present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which accurately captures the parasitic parameters of the circuit through a parasitic catcher, and conducts a detailed analysis of the circuit board layer in combination with the circuit operation data, and uses multi-physics field simulation software to predict the specific impact of these factors and how the parasitic parameters interfere with the circuit performance. The system automatically adjusts the circuit layout and design parameters according to these analysis results, improves the signal integrity and interference resistance of the circuit, automates the adjustment, and improves the design efficiency.

[0005] The present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, comprising:

[0006] Step 1: Use a parasitic capturer to determine the parasitic parameters of the circuit under different test conditions, and record the operating data of the circuit during the capture process to determine the parasitic conditions;

[0007] Step 2: hierarchically divide the circuit layout based on the operation data, analyze the circuit layout layer by layer according to the hierarchy, and then identify independent contribution factors;

[0008] Step 3: Using the operating data to determine the multi-physics field simulation software to simulate the influence of the independent contribution factors on the parasitic parameters and the interference of the parasitic parameters on the electrical performance;

[0009] Step 4: Automatically adjust circuit layout and circuit-related design parameters based on interference conditions, impact rules and parasitic conditions to minimize parasitic effects.

[0010] The present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which uses a parasitic capturer to determine the parasitic parameters of a circuit under different test states, and records the operating data of the circuit during the capture process, thereby determining the parasitic situation, including:

[0011] Determine the function and corresponding principle of the circuit according to the existing circuit diagram related information, and predict the first key node of the circuit according to the function and corresponding principle;

[0012] At the same time, a hot spot analysis is performed on the circuit operation status, and the overheated component is identified as the second critical node;

[0013] Calculating a union of the first key node and the second key node, wherein each node included in the union is regarded as a monitoring node;

[0014] Determine the circuit type of the existing circuit diagram and the type characteristics of the corresponding type according to the function and the corresponding principle, and then determine the state of the circuit to be tested;

[0015] configuring corresponding monitoring parameters for the monitoring node based on different states to be tested, and setting a parasitic catcher based on the monitoring parameters;

[0016] Use a parasitic catcher to collect parasitic parameters of the circuit corresponding to the monitoring nodes in different test states, and determine the operation data of the circuit in different test states;

[0017] The collected parasitic parameters and operating data are preliminarily analyzed to determine the changing trends of the parasitic parameters and the abnormal values ​​of the operating data under different test conditions, and based on the changing trends and the abnormal values, the parasitic conditions of the parasitic parameters on the circuit are obtained.

[0018] The present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which finally obtains the parasitic condition of the parasitic parameter on the circuit based on the change trend and the abnormal value, including:

[0019] Performing statistical descriptive analysis on parasitic parameters and operating data to obtain basic characteristics of the parasitic parameters and operating data, using charts to visualize changes in parasitic parameters under different test conditions, and determining the change trend of the parasitic parameters;

[0020] Identify abnormal values ​​in the operating data according to the basic characteristics, perform correlation analysis on the abnormal values ​​and the change trend, determine the correlation, and then determine the causal relationship between the parasitic parameter change and the circuit performance;

[0021] The parasitic condition of the parasitic parameters on the circuit is obtained according to the cause-effect relationship.

[0022] The present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which obtains the parasitic condition of the parasitic parameter on the circuit according to the cause-effect relationship, including:

[0023] Combining the correlation with circuit design-related knowledge, inferring the possible causal relationship between the abnormal value and the change trend, screening the possible causal relationship through setting control variable experiments, and determining the causal relationship;

[0024] According to the established cause-effect relationship, specific parasitic parameters associated with circuit performance are identified, the specific impact of the specific parasitic parameters on circuit performance is determined, and the parasitic situation is obtained.

[0025] The present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which divides the circuit layout into levels based on the operation data, analyzes the circuit layout layer by layer according to the levels, and then identifies independent contribution factors, including:

[0026] Preliminarily plan the levels of the circuit layout according to the operation data, determine the main functions and layout of each level, extract electrical parameters of different levels based on the connection between layers, and perform electrical evaluation to determine the electrical performance of each level;

[0027] Performing a first sensitivity analysis on electrical parameters and electrical performance based on the parasitic conditions to identify parasitic optimization parameters that have the greatest impact on electrical performance;

[0028] A second sensitivity analysis is performed on the parasitic optimization parameters, and an independent contribution factor is determined in combination with the result of the first sensitivity analysis.

[0029] The present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, and a calculation formula for sensitivity analysis, including:

[0030] ;

[0031] ;

[0032] ;

[0033] express The global first sensitivity corresponding to the independent impact; express The interaction between them affects the corresponding global second sensitivity; represents global sensitivity; Represents the total variance of the electrical performance index; represents the i-th electrical parameter; Indicates electrical performance; Indicates that when integrating The joint probability distribution of all other electrical parameters except Indicates that when integrating The joint probability distribution of all other electrical parameters except ; Indicates that when integrating The joint probability distribution of In addition to All other electrical parameters except for are multi-integrated; right Perform multiple integrations; Divide All other electrical parameters except are multiply integrated, where the number of multiply integrations depends on the dimension of the parameter; Indicates that the condition Down The conditional expectation of The variance of In the conditions Down The conditional expectation of The variance of Indicates that the condition Down The conditional expectation of The variance of In addition to the condition Other conditions besides The conditional expectation of The variance of

[0034] If the global first sensitivity and the global sensitivity of the electrical parameter are within a set similar range, the electrical parameter is determined as an independent contribution factor in combination with the global second sensitivity.

[0035] The present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which uses operating data to determine the multi-physics field simulation software to simulate the influence of the independent contribution factors on the parasitic parameters and the interference of the parasitic parameters on the electrical performance, including:

[0036] Determine the multi-physics simulation software to be used according to the operation data, and define the corresponding physical processes in the multi-physics simulation software;

[0037] Implementing parameter scanning, changing the values ​​of independent contribution factors and parasitic parameters, and determining the interference of the independent contribution factors and parasitic parameters on electrical performance;

[0038] Analyze the simulation results to determine the influence of independent contribution factors on parasitic parameters.

[0039] The present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which automatically adjusts the circuit layout and circuit-related design parameters according to the interference situation, influence law and parasitic situation, thereby minimizing the parasitic effect, including:

[0040] Identify the most critical parasitic effects based on the interference conditions, impact rules and parasitic conditions;

[0041] According to the influencing rules, the key design parameters and circuit layout features that affect the electrical performance are extracted, and then the path-finding algorithm is developed. At the same time, possible circuit design improvement points are identified and optimization suggestions are put forward;

[0042] Iteratively applying the pathfinding algorithm and automatically adjusting circuit layout and design parameters in combination with the optimization suggestions;

[0043] If the optimization suggestion is applicable, perform a design sensitivity optimization cycle to adjust and test new design variables;

[0044] After the iteration, the electrical performance is analyzed to determine the changes, and the next optimization direction of the path-finding algorithm is determined according to the changes;

[0045] Compare the simulation results of each iterative step, evaluate the changing trend of the parasitic effect, screen the optimization direction according to the comparison and evaluation results, and repeat the optimization process until the parasitic effect is minimized.

[0046] Compared with the prior art, the beneficial effects of the present application are as follows: the parasitic parameters of the circuit are accurately captured by the parasitic catcher, and the circuit board layer hierarchy is carefully analyzed in combination with the circuit operation data, and multi-physics field simulation software is used to predict the specific impact of these factors and how the parasitic parameters interfere with the circuit performance. The system automatically adjusts the circuit layout and design parameters according to these analysis results, improves the signal integrity and interference resistance of the circuit, automates the adjustment, and improves the design efficiency.

[0047] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0048] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0050] Figure 1It is a flow chart of an artificial intelligence parasitic parameter analysis and automatic optimization method provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0052] Embodiment 1:

[0053] The embodiment of the present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, such as Figure 1 As shown, including:

[0054] Step 1: Use a parasitic capturer to determine the parasitic parameters of the circuit under different test conditions, and record the operating data of the circuit during the capture process to determine the parasitic conditions;

[0055] Step 2: hierarchically divide the circuit layout based on the operation data, analyze the circuit layout layer by layer according to the hierarchy, and then identify independent contribution factors;

[0056] Step 3: Using the operating data to determine the multi-physics field simulation software to simulate the influence of the independent contribution factors on the parasitic parameters and the interference of the parasitic parameters on the electrical performance;

[0057] Step 4: Automatically adjust circuit layout and circuit-related design parameters based on interference conditions, impact rules and parasitic conditions to minimize parasitic effects.

[0058] In this embodiment, the parasitic condition refers to how the parasitic parameters of each monitoring node change under different test states of the circuit, and how these changes affect the overall performance and stability of the circuit, including the increase of parasitic capacitance, the influence of parasitic inductance or the change of parasitic resistance.

[0059] In this embodiment, the hierarchical planning process is a preliminary planning of the levels of the circuit layout, typically including determining the physical structure of the circuit board, such as a single-layer, double-layer or multi-layer PCB, determining the main circuits, components and connections carried by each level, and considering the wiring of signals, power and ground between layers.

[0060] In this embodiment, the layer-by-layer analysis is to guide the layout level planning through the circuit operation data, extract and evaluate the electrical parameters at each level, and analyze the circuit layout.

[0061] In this embodiment, the independent contribution factor refers to a single factor or parameter that has a significant impact on a specific performance indicator in the performance of the entire circuit or system.

[0062] In this embodiment, the interference situation determination process is to set a parameter scanning scheme, in which the values ​​of independent contribution factors (such as power supply voltage, current, component size, etc.) and parasitic parameters (such as parasitic capacitance, parasitic inductance, etc.) are changed, a series of simulations are performed, and the impact of these variables on electrical performance (such as voltage, current, frequency response, etc.) is observed. For example, in a circuit, increasing the parasitic capacitance value may cause the cutoff frequency of the RC low-pass filter to decrease, which will change the circuit's processing of signals, thereby affecting the performance of the entire system.

[0063] In this embodiment, the influence rule describes the quantitative or qualitative influence of the independent contribution factor on the parasitic parameter. For example, the increase in the size of the resistor will increase the influence of the parasitic inductance, causing the impedance of the circuit to increase faster with the increase of frequency.

[0064] In this embodiment, the process of minimizing parasitic effects is to identify key parasitic effects, continuously adjust the circuit design using a path-finding algorithm combined with optimization suggestions, minimize the impact of parasitic effects on performance through iterative cycle optimization, and accurately identify and weaken major parasitic effects.

[0065] The working principle and beneficial effects of the above technical solution are: the parasitic parameters of the circuit are accurately captured by the parasitic catcher, and the circuit board layer is carefully analyzed in combination with the circuit operation data, and the multi-physics field simulation software is used to predict the specific impact of these factors and how the parasitic parameters interfere with the circuit performance. The system automatically adjusts the circuit layout and design parameters according to these analysis results, improves the signal integrity and interference resistance of the circuit, automates the adjustment, and improves the design efficiency.

[0066] Embodiment 2:

[0067] The embodiment of the present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which uses a parasitic capturer to determine the parasitic parameters of a circuit under different test states, and records the operation data of the circuit during the capture process, thereby determining the parasitic situation, including:

[0068] Determine the function and corresponding principle of the circuit according to the existing circuit diagram related information, and predict the first key node of the circuit according to the function and corresponding principle;

[0069] At the same time, a hot spot analysis is performed on the circuit operation status, and the overheated component is identified as the second critical node;

[0070] Calculating a union of the first key node and the second key node, wherein each node included in the union is regarded as a monitoring node;

[0071] Determine the circuit type of the existing circuit diagram and the type characteristics of the corresponding type according to the function and the corresponding principle, and then determine the state of the circuit to be tested;

[0072] configuring corresponding monitoring parameters for the monitoring node based on different states to be tested, and setting a parasitic catcher based on the monitoring parameters;

[0073] Use a parasitic catcher to collect parasitic parameters of the circuit corresponding to the monitoring nodes in different test states, and determine the operation data of the circuit in different test states;

[0074] The collected parasitic parameters and operating data are preliminarily analyzed to determine the changing trends of the parasitic parameters and the abnormal values ​​of the operating data under different test conditions, and based on the changing trends and the abnormal values, the parasitic conditions of the parasitic parameters on the circuit are obtained.

[0075] In this embodiment, hot spot analysis refers to finding areas of elevated temperature in a circuit through temperature monitoring or thermal simulation analysis during circuit operation. These areas of elevated temperature are usually caused by current concentration or poor heat dissipation of components, and are used to identify components or parts that may affect circuit stability and life due to excessive temperature.

[0076] In this embodiment, the first key node is the most functionally important or most sensitive part of the circuit, such as the output of a power converter, the input of a signal amplifier, or a key node of a logic operation; based on hot spot analysis, the second key node may be a component in the circuit that is prone to overheating, such as a power device (such as a transistor), a high current connection point, or a power management module.

[0077] In this embodiment, the state to be tested includes various working modes of the circuit, such as a standby mode, a full-load working mode, a low-voltage operating mode and other different working conditions.

[0078] In this embodiment, the monitoring parameters include electrical parameters such as current, voltage, power consumption, frequency, etc., as well as thermal parameters such as temperature that may need to be monitored.

[0079] In this embodiment, the parasitic catcher refers to a tool or device for monitoring and recording parasitic parameters (such as parasitic capacitance, parasitic inductance, etc.) of a circuit node under a specific test state. This may be an actual test device or a functional module in a simulation software.

[0080] In this embodiment, the preliminary analysis process is to compare and analyze the collected parasitic parameters and operating data to find the variation pattern of the parasitic parameters under different test conditions and the deviation from the expected operating data, which also includes identifying those data points (outliers) that are beyond the normal range.

[0081] The working principle and beneficial effects of the above technical solution are: comprehensively using circuit functional principles and hot spot analysis to identify key monitoring nodes, and collecting data under different operating conditions through parasitic catchers, thereby revealing the impact of parasitic parameters on circuit performance and improving the accuracy of circuit fault diagnosis. It can reduce maintenance costs and downtime by preventing faults, optimize circuit performance in the design stage, and improve system reliability.

[0082] Embodiment 3:

[0083] The embodiment of the present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which finally obtains the parasitic condition of the parasitic parameter on the circuit based on the change trend and the abnormal value, including:

[0084] Performing statistical descriptive analysis on parasitic parameters and operating data to obtain basic characteristics of the parasitic parameters and operating data, using charts to visualize changes in parasitic parameters under different test conditions, and determining the change trend of the parasitic parameters;

[0085] Identify abnormal values ​​in the operating data according to the basic characteristics, perform correlation analysis on the abnormal values ​​and the change trend, determine the correlation, and then determine the causal relationship between the parasitic parameter change and the circuit performance;

[0086] The parasitic condition of the parasitic parameters on the circuit is obtained according to the cause-effect relationship.

[0087] In this embodiment, the statistical description analysis process is to collect parasitic parameters and operating data, and then use statistical methods such as mean, median, mode, standard deviation and variance to quantitatively describe the data, and draw charts such as line charts, bar charts, scatter plots, etc. to visualize the data and analyze the data distribution and central tendency.

[0088] In this embodiment, the basic characteristics include the average value, median, mode, extreme value, range, dispersion degree (variance or standard deviation), distribution shape, etc. of the parasitic parameters and operating data.

[0089] In this embodiment, the variation trend is manifested as a regular increase or decrease with a certain test state, such as the nonlinear growth of parasitic capacitance with temperature, or the variation of parasitic inductance at different frequencies.

[0090] In this embodiment, the correlation analysis process is to use correlation analysis tools, such as Pearson correlation coefficient, Spearman rank correlation, etc., to quantify the strength and direction of the relationship between the parasitic parameter changes and the abnormal values ​​of the operating data after the abnormal values ​​are collected. It usually involves performing statistical tests on the data to obtain a probability value (P-value) to determine whether there is a statistically significant correlation between the two sets of data.

[0091] In this embodiment, the correlation refers to the relationship between the parasitic parameters and the operating data, which may be expressed as a direct positive correlation, a negative correlation, or no obvious correlation.

[0092] In this embodiment, causality refers to determining through analysis how a change in a parasitic parameter leads to a specific performance in the operating data, for example, an increase in parasitic capacitance causes an increase in circuit delay, or noise generated by parasitic inductance affects the transmission quality of the signal.

[0093] In this embodiment, the specific steps of deriving the parasitic situation from the cause-effect relationship are to classify and process the collected parasitic parameters; analyze the statistical correlation between the parasitic parameters and the operating data of the circuit performance; find the parasitic parameter changes that occur simultaneously with the negative changes in the circuit performance and establish a preliminary connection; compare experiments or simulations, change a certain parasitic parameter through the control variable method, and observe the changes in circuit performance; establish statistical or logical evidence for the cause-effect relationship found through repeated verification; according to the above steps, comprehensively analyze and determine how the changes in the parasitic parameters affect the overall parasitic situation of the circuit.

[0094] The working principle and beneficial effects of the above technical solution are: to describe the characteristics of parasitic parameters and operating data through statistical methods, and to determine the parameter change trend through graphical analysis, and then analyze the correlation between outliers and trends, determine the causal relationship between parasitic effects and circuit performance, and clarify the direct impact relationship between parasitic parameters and circuit performance. Through the identification of outliers, circuit failures can be prevented in advance, and the understanding of complex circuit behavior is simplified, which helps to achieve more accurate circuit debugging and performance improvement.

[0095] Embodiment 4:

[0096] The embodiment of the present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which obtains the parasitic condition of the parasitic parameter on the circuit according to the cause-effect relationship, including:

[0097] Combining the correlation with circuit design-related knowledge, inferring the possible causal relationship between the abnormal value and the change trend, screening the possible causal relationship through setting control variable experiments, and determining the causal relationship;

[0098] According to the established cause-effect relationship, specific parasitic parameters associated with circuit performance are identified, the specific impact of the specific parasitic parameters on circuit performance is determined, and the parasitic situation is obtained.

[0099] In this embodiment, the inference of possible causal relationships is that when certain abnormal values ​​or performance change trends are observed, it is necessary to analyze possible factors including design defects, material properties, differences in the manufacturing process, etc. Through circuit-related knowledge and experience, the possible causal relationships behind these abnormal values ​​can be inferred.

[0100] In this embodiment, the screening process is to screen possible causal relationships through experimental design of controlled variables, that is, when conducting experiments, all variables except the variables being tested remain unchanged. In this way, interference from other factors can be eliminated, thereby verifying and establishing the true causal relationship. For example, if it is suspected that parasitic capacitance causes the amplifier frequency response to drop, only the value of the parasitic capacitance can be changed while keeping all other design parameters unchanged, and observing how the frequency response changes.

[0101] In this embodiment, the specific parasitic parameters refer to parameters that are verified to have a significant impact on circuit performance through screening and verification among a series of possible parasitic factors, such as the parasitic capacitance, inductance and resistance mentioned above.

[0102] In this embodiment, the specific impact is that after confirming the specific impact of specific parasitic parameters on circuit performance, a clear understanding of the source of these parameters, extracting their values, and why they affect circuit performance can be obtained, including signal distortion, gain change, corresponding frequency change, noise increase, etc. Taking parasitic capacitance as an example, it may cause phase shift of the signal, reduce the transmission efficiency of high-frequency signals, and affect signal integrity. After determining these specific parasitic parameters, engineers can take measures to optimize them, such as re-layout to reduce the capacitance effect, or add corresponding compensation mechanisms in the simulation model.

[0103] The working principle and beneficial effects of the above technical solution are: through correlation analysis and controlled variable experiments, the deterministic causal relationship between parasitic parameters and circuit performance is comprehensively determined, and then the parasitic situation is determined, the accuracy of circuit design is enhanced, and the risks brought by unknown factors are reduced. The results of theoretical analysis are experimentally verified, the effectiveness of circuit adjustment is improved, the circuit stability and reliability are improved, and the product life is extended.

[0104] Embodiment 5:

[0105] The embodiment of the present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which divides the circuit layout into levels based on the operation data, analyzes the circuit layout layer by layer according to the levels, and then identifies independent contribution factors, including:

[0106] Preliminarily plan the levels of the circuit layout according to the operation data, determine the main functions and layout of each level, extract electrical parameters of different levels based on the connection between layers, and perform electrical evaluation to determine the electrical performance of each level;

[0107] Performing a first sensitivity analysis on electrical parameters and electrical performance based on the parasitic conditions to identify parasitic optimization parameters that have the greatest impact on electrical performance;

[0108] A second sensitivity analysis is performed on the parasitic optimization parameters, and an independent contribution factor is determined in combination with the result of the first sensitivity analysis.

[0109] In this embodiment, the main functions and layout of each layer include the top layer: used to place most of the main components and microstrip lines; the inner layer: can be used for the power plane and ground plane to provide stable power supply and grounding for the circuit; the bottom layer: can be used to place secondary components or auxiliary routing.

[0110] In this embodiment, the electrical parameter extraction results include the board resistance, capacitance, inductance, etc. of each layer, which are key electrical characteristics that need to be considered during the circuit simulation process.

[0111] In this embodiment, the result of the electrical evaluation is to analyze the impact of the electrical parameters of each level on the performance of the entire circuit, which may include evaluations of signal integrity (SI), power integrity (PI), and electromagnetic compatibility (EMC).

[0112] In this embodiment, electrical performance refers to operating characteristics of the circuit, such as operating frequency, noise level, power consumption, sensitivity, stability and other indicators.

[0113] In this embodiment, the first sensitivity analysis refers to analyzing the impact of each parameter in the circuit model on the circuit performance when it changes individually, helping to identify sensitive parameters that will significantly change the circuit performance when the parameters change slightly, namely parasitic optimization parameters.

[0114] In this embodiment, parasitic optimization parameters refer to parasitic electrical parameters that have a significant impact on circuit performance, such as parasitic capacitance, parasitic inductance, parasitic resistance, etc. These parameters are usually not directly set in the design, but are non-ideal characteristics that appear in the circuit due to several factors.

[0115] In this embodiment, the second sensitivity analysis is a further analysis based on the first sensitivity analysis, which considers the effect of the change in one parameter on the circuit performance while keeping other parameters unchanged, and helps identify the main independent parameters that affect the circuit performance in a complex circuit system. For example, when designing an amplifier circuit, the first sensitivity analysis may find that the parasitic capacitance has a significant effect on the bandwidth of the signal amplification. Through the second sensitivity analysis, it may be found that increasing the bias current of the transistor can reduce the effect of the parasitic capacitance, thereby improving the performance of the circuit. At this point, the bias current of the transistor is considered to be an independent contribution factor because it can optimize and adjust the effect of the parasitic parameters.

[0116] In this embodiment, the process of determining the independent contribution factors is to use a statistical analysis method of parameters, such as regression analysis, variance analysis, etc., to identify and quantify the independent factors that affect the electrical performance.

[0117] The working principle and beneficial effects of the above technical solution are: guiding the layout hierarchical planning through circuit operation data, extracting and evaluating electrical parameters at each level, and determining the parasitic parameters that most affect the electrical performance through sensitivity analysis, so as to find independent contribution factors, optimize the circuit layout, reduce the adverse effects caused by parasitic effects, and ensure the robustness and reliability of the circuit design through sensitivity analysis.

[0118] Embodiment 6:

[0119] The embodiment of the present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, and the calculation formula of sensitivity analysis includes:

[0120] ;

[0121] ;

[0122] ;

[0123] express The global first sensitivity corresponding to the independent impact; express The interaction between them affects the corresponding global second sensitivity; represents global sensitivity; Represents the total variance of the electrical performance index; represents the i-th electrical parameter; Indicates electrical performance; Indicates that when integrating The joint probability distribution of all other electrical parameters except Indicates that when integrating The joint probability distribution of all other electrical parameters except ; Indicates that when integrating The joint probability distribution of In addition to All other electrical parameters except for are multi-integrated; right Perform multiple integrations; Divide All other electrical parameters except are multiply integrated, where the number of multiply integrations depends on the dimension of the parameter; Indicates that the condition Down The conditional expectation of The variance of In the conditions Down The conditional expectation of The variance of Indicates that the condition Down The conditional expectation of The variance of In addition to the condition Other conditions besides The conditional expectation of The variance of

[0124] If the global first sensitivity and the global sensitivity of the electrical parameter are within a set similar range, the electrical parameter is determined as an independent contribution factor in combination with the global second sensitivity.

[0125] In this embodiment, the global first sensitivity evaluates the basis of the independent impact of each parameter, which can help determine which parameters have a significant individual impact on electrical performance. These parameters are considered to be key parameters in circuit design; the global second sensitivity helps understand the importance of the interaction between different parameters and reveals the impact of multiple parameter combinations.

[0126] In this embodiment, the global first sensitivity is used in the first sensitivity analysis, which is mainly used to preliminarily identify individual parameters with significant impacts, while the second sensitivity analysis uses the global second sensitivity and the global sensitivity to more deeply analyze the interactions between parameters and the overall impact of the parameters, and determine the independent contribution factors that ultimately affect the output.

[0127] The working principle and beneficial effect of the above technical solution are: determining independent contribution factors by calculating the global first sensitivity, the global second sensitivity and the global sensitivity helps to optimize the circuit design more accurately and ensure the robustness and reliability of the circuit design.

[0128] Embodiment 7:

[0129] The embodiment of the present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which uses operation data to determine that multi-physics field simulation software simulates the influence of the independent contribution factors on the parasitic parameters and the interference of the parasitic parameters on the electrical performance, including:

[0130] Determine the multi-physics simulation software to be used according to the operation data, and define the corresponding physical processes in the multi-physics simulation software;

[0131] Implementing parameter scanning, changing the values ​​of independent contribution factors and parasitic parameters, and determining the interference of the independent contribution factors and parasitic parameters on electrical performance;

[0132] Analyze the simulation results to determine the influence of independent contribution factors on parasitic parameters.

[0133] In this embodiment, the multi-physics simulation software includes ANSYS Multiphysics: providing multi-field coupling analysis such as structure, fluid, and electromagnetic fields; COMSOL Multiphysics: users can customize physical fields and the interactions between them; Autodesk Simulation: suitable for a variety of engineering simulations and supports multi-physics field analysis; Altair HyperWorks: covers multi-physics field simulations such as structure, heat, and electromagnetics.

[0134] In this embodiment, the physical processes include heat conduction, fluid dynamics, electromagnetism, structural mechanics, etc. Each process describes the changes and interactions of different physical quantities, such as temperature changes, fluid flow, electric and magnetic field distributions, stress and strain, etc.

[0135] The working principle and beneficial effects of the above technical solution are: using multi-physics simulation software to simulate actual physical processes, changing key factors through parameter scanning technology, analyzing the specific impact of these factors on electrical performance, and revealing their mechanism of action, analyzing the interaction between independent contribution factors and parasitic parameters, providing a scientific basis for optimization, promoting the refinement of circuit design, and improving the performance and reliability of electronic equipment.

[0136] Embodiment 8:

[0137] The embodiment of the present invention provides an artificial intelligence parasitic parameter analysis and automatic optimization method, which automatically adjusts the circuit layout and circuit-related design parameters according to the interference situation, influence law and parasitic situation, thereby minimizing the parasitic effect, including:

[0138] Identify the most critical parasitic effects based on the interference conditions, impact rules and parasitic conditions;

[0139] According to the influencing rules, the key design parameters and circuit layout features that affect the electrical performance are extracted, and then the path-finding algorithm is developed. At the same time, possible circuit design improvement points are identified and optimization suggestions are put forward;

[0140] Iteratively applying the pathfinding algorithm and automatically adjusting circuit layout and design parameters in combination with the optimization suggestions;

[0141] If the optimization suggestion is applicable, perform a design sensitivity optimization cycle to adjust and test new design variables;

[0142] After the iteration, the electrical performance is analyzed to determine the changes, and the next optimization direction of the path-finding algorithm is determined according to the changes;

[0143] Compare the simulation results of each iterative step, evaluate the changing trend of the parasitic effect, screen the optimization direction according to the comparison and evaluation results, and repeat the optimization process until the parasitic effect is minimized.

[0144] In this embodiment, the process of determining the parasitic effect is to analyze the interaction between the various components of the electrical system. First, the electrical performance of the entire electrical system is simulated using simulation software to identify various interference and parasitic conditions in the system. Then, the simulation results are analyzed in detail to identify the key factors affecting the system performance, from which the most critical parasitic effects are determined.

[0145] In this embodiment, the parasitic effects include parasitic capacitance, parasitic inductance, parasitic resistance, etc. These effects may be caused by circuit layout, manufacturing process or other non-ideal characteristics.

[0146] In this embodiment, the process of developing a pathfinding algorithm is to guide the algorithm design through the information revealed by the simulation results. The algorithm needs to be able to identify and adjust the circuit layout and design parameters to optimize circuit performance and minimize parasitic effects. This may involve heuristic search, optimization theory, and algorithm design techniques.

[0147] In this embodiment, the optimization suggestion includes modifying circuit layout features, adjusting component placement, improving conductive paths, selecting more suitable materials or components, etc.

[0148] In this embodiment, the next optimization direction may be based on simulation results and design sensitivity analysis. Possible optimization directions include improved circuit layout, enhanced or reduced use of specific components, or fine-tuning of the most critical design parameters.

[0149] In this embodiment, the process of comparing simulation results is to collect and analyze data after each simulation iteration, and compare the change of electrical performance and the reduction trend of parasitic effects. By drawing the performance index curve, the effects of various design modifications can be intuitively seen.

[0150] In this embodiment, the process of repeated optimization is iterative, which means that after each modification of the circuit design, the simulation test will be re-performed. By looping this process, the new and old simulation results are constantly compared, and the design is adjusted until the goal of minimizing parasitic effects is achieved. Each iteration will be analyzed based on the results of the previous simulation to determine the new optimization direction and the design parameters that need to be adjusted.

[0151] The working principle and beneficial effects of the above technical solution are: by identifying key parasitic effects, using path-finding algorithms combined with optimization suggestions to continuously adjust the circuit design, minimizing the impact of parasitic effects on performance through iterative cycle optimization, accurately identifying and weakening major parasitic effects, improving circuit performance, realizing automatic optimization of circuit design, and improving design efficiency.

[0152] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An artificial intelligence parasitic parameter analysis and automatic optimization method, characterized in that: include: Step 1: Use a parasitic capturer to determine the parasitic parameters of the circuit under different test conditions, and record the operating data of the circuit during the capture process to determine the parasitic conditions; Step 2: hierarchically divide the circuit layout based on the operation data, analyze the circuit layout layer by layer according to the hierarchy, and then identify independent contribution factors; Step 3: Using the operating data to determine the multi-physics field simulation software to simulate the influence of the independent contribution factors on the parasitic parameters and the interference of the parasitic parameters on the electrical performance; Step 4: Automatically adjust the circuit layout and circuit-related design parameters according to the interference situation, influence rules and parasitic conditions to minimize the parasitic effects; Wherein, step 2 includes: Preliminarily plan the levels of the circuit layout according to the operation data, determine the main functions and layout of each level, extract electrical parameters of different levels based on the connection between layers, and perform electrical evaluation to determine the electrical performance of each level; Performing a first sensitivity analysis on electrical parameters and electrical performance based on the parasitic conditions to identify parasitic optimization parameters that have the greatest impact on electrical performance; Performing a second sensitivity analysis on the parasitic optimization parameters, and determining an independent contribution factor in combination with the result of the first sensitivity analysis; The calculation formula for sensitivity analysis includes: ; ; ; express The global first sensitivity corresponding to the independent impact; express The interaction between them affects the corresponding global second sensitivity; represents global sensitivity; Represents the total variance of the electrical performance index; represents the i-th electrical parameter; Indicates electrical performance; Indicates that when integrating The joint probability distribution of all other electrical parameters except Indicates that when integrating The joint probability distribution of all other electrical parameters except ; Indicates that when integrating The joint probability distribution of In addition to All other electrical parameters except for are multi-integrated; right Perform multiple integrations; Divide All other electrical parameters except are multiply integrated, where the number of multiply integrations depends on the dimension of the parameter; Indicates that the condition Down The conditional expectation of The variance of In the conditions Down The conditional expectation of The variance of Indicates that the condition Down The conditional expectation of The variance of In addition to the condition Other conditions besides The conditional expectation of The variance of If the global first sensitivity of the electrical parameter is within a set similar range to the global sensitivity value, the electrical parameter is determined as an independent contribution factor in combination with the global second sensitivity.

2. The artificial intelligence parasitic parameter analysis and automatic optimization method according to claim 1, characterized in that: Use the parasitic capturer to determine the parasitic parameters of the circuit under different test conditions, and record the operating data of the circuit during the capture process to determine the parasitic conditions, including: Determine the function and corresponding principle of the circuit according to the existing circuit diagram related information, and predict the first key node of the circuit according to the function and corresponding principle; At the same time, hot spot analysis is performed on the circuit operation status, and the overheated component is identified as the second key node; Calculating a union of the first key node and the second key node, wherein each node included in the union is regarded as a monitoring node; Determine the circuit type of the existing circuit diagram and the type characteristics of the corresponding type according to the function and the corresponding principle, and then determine the state of the circuit to be tested; configuring corresponding monitoring parameters for the monitoring node based on different states to be tested, and setting a parasitic catcher based on the monitoring parameters; Use a parasitic catcher to collect parasitic parameters of the circuit corresponding to the monitoring nodes in different test states, and determine the operation data of the circuit in different test states; The collected parasitic parameters and operating data are preliminarily analyzed to determine the changing trends of the parasitic parameters and the abnormal values ​​of the operating data under different test conditions, and based on the changing trends and the abnormal values, the parasitic conditions of the parasitic parameters on the circuit are obtained.

3. The artificial intelligence parasitic parameter analysis and automatic optimization method according to claim 2, characterized in that: Based on the change trend and abnormal values, the parasitic conditions of the parasitic parameters on the circuit are finally obtained, including: Performing statistical descriptive analysis on parasitic parameters and operating data to obtain basic characteristics of the parasitic parameters and operating data, using charts to visualize changes in parasitic parameters under different test conditions, and determining the change trend of the parasitic parameters; Identify abnormal values ​​in the operating data according to the basic characteristics, perform correlation analysis on the abnormal values ​​and the change trend, determine the correlation, and then determine the causal relationship between the parasitic parameter change and the circuit performance; The parasitic condition of the parasitic parameters on the circuit is obtained according to the cause-effect relationship.

4. The artificial intelligence parasitic parameter analysis and automatic optimization method according to claim 3, characterized in that: According to the cause-effect relationship, the parasitic condition of the parasitic parameters on the circuit is obtained, including: Combining the correlation with circuit design-related knowledge, inferring the possible causal relationship between the abnormal value and the change trend, screening the possible causal relationship through setting control variable experiments, and determining the causal relationship; According to the established cause-effect relationship, identifying specific parasitic parameters associated with circuit performance, determining the specific impact of the specific parasitic parameters on circuit performance, and obtaining a parasitic condition; Among them, specific parasitic parameters refer to parameters that are proven to have a significant impact on circuit performance through screening and verification among a series of possible parasitic factors.

5. The artificial intelligence parasitic parameter analysis and automatic optimization method according to claim 1, characterized in that: The multi-physics field simulation software is used to determine the influence of the independent contribution factors on the parasitic parameters and the interference of the parasitic parameters on the electrical performance by using the operating data, including: Determine the multi-physics simulation software to be used according to the operation data, and define the corresponding physical processes in the multi-physics simulation software; Implementing parameter scanning, changing the values ​​of independent contribution factors and parasitic parameters, and determining the interference of the independent contribution factors and parasitic parameters on electrical performance; Analyze the simulation results to determine the influence of independent contribution factors on parasitic parameters.

6. The artificial intelligence parasitic parameter analysis and automatic optimization method according to claim 1, characterized in that: Automatically adjust circuit layout and circuit-related design parameters based on interference conditions, influence rules and parasitic conditions to minimize parasitic effects, including: Identify the most critical parasitic effects based on interference conditions, impact patterns and parasitic conditions; According to the influencing rules, the key design parameters and circuit layout features that affect the electrical performance are extracted, and then the path-finding algorithm is developed. At the same time, possible circuit design improvement points are identified and optimization suggestions are put forward; Iteratively applying the pathfinding algorithm and automatically adjusting circuit layout and design parameters in combination with the optimization suggestions; If the optimization suggestion is applicable, perform a design sensitivity optimization cycle to adjust and test new design variables; After the iteration, the electrical performance is analyzed to determine the changes, and the next optimization direction of the path-finding algorithm is determined according to the changes; Compare the simulation results of each iterative step, evaluate the changing trend of the parasitic effect, screen the optimization direction according to the comparison and evaluation results, and repeat the optimization process until the parasitic effect is minimized.