Method and system for performance analysis of semiconductor materials based on intelligent AI

Abstract

The present invention relates to the technical field of semiconductor material performance analysis, and discloses a method for realizing semiconductor material performance analysis based on intelligent AI, including: obtaining material data of semiconductor materials, analyzing the electrical properties of semiconductor materials; calculating the light energy absorption coefficient of semiconductor materials, analyzing the light energy absorption performance of semiconductor materials, detecting the light energy refractive index and luminous efficiency of semiconductor materials, and analyzing the optical properties of semiconductor materials; calculating the thermal conductivity of semiconductor materials, querying the thermal expansion coefficient of semiconductor materials, and analyzing the thermal properties of semiconductor materials; querying the multi-source stable test data of semiconductor materials, constructing an environmental performance analysis model of semiconductor materials, and analyzing the chemical properties of semiconductor materials; analyzing the material properties of semiconductor materials based on electrical properties, optical properties, thermal properties and chemical properties. The present invention can improve the accuracy of semiconductor material performance analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of semiconductor material performance analysis, and in particular to a method and system for realizing semiconductor material performance analysis based on intelligent AI. Background Art

[0002] Semiconductor materials are a type of material with electrical conductivity between conductors (such as metals) and insulators (such as plastics). Common semiconductor materials include silicon (Si), germanium (Ge), gallium arsenide (GaAs), gallium nitride (GaN), etc. At the same time, semiconductor materials have unique electrical properties, and their conductivity can be regulated by external conditions such as doping, light, and temperature. Therefore, semiconductor materials play a vital role in the field of modern science and technology, and their performance analysis is crucial for the development and optimization of electronic devices.

[0003] At present, the performance analysis method of semiconductor materials generally uses traditional experimental tests and theoretical calculations to build a relatively fixed analysis process and model to achieve performance analysis of semiconductor materials. However, through this method, many analysis processes rely on manual settings and operations. Since the performance of semiconductor materials involves many complex parameters and variables, it is easy to make mistakes when processing large amounts of data and complex situations, resulting in incomplete analysis results, which will affect the accuracy of the analysis of semiconductor material performance. Summary of the invention

[0004] The present invention provides a method and system for realizing performance analysis of semiconductor materials based on intelligent AI, the main purpose of which is to improve the accuracy of performance analysis of semiconductor materials.

[0005] To achieve the above objectives, the present invention provides a method for analyzing the performance of semiconductor materials based on intelligent AI, comprising:

[0006] Acquiring material data of a semiconductor material, calculating a carrier concentration of the semiconductor material based on the material data, identifying a material composition, a material structure, and a material process of the semiconductor material, analyzing a resistance property of the semiconductor material based on the material composition, the material structure, and the material process, and analyzing an electrical property of the semiconductor material based on the carrier concentration and the resistance property;

[0007] Collecting the photoelectron spectrum of the semiconductor material, calculating the light energy absorption coefficient of the semiconductor material using the photoelectron spectrum, analyzing the light energy absorption performance of the semiconductor material based on the material structure and the material composition, detecting the light energy refractive index and luminous efficiency of the semiconductor material, and analyzing the optical properties of the semiconductor material according to the light energy absorption coefficient, the light energy absorption performance, the light energy refractive index and the luminous efficiency;

[0008] Querying the thermal diffusion coefficient and specific heat capacity of the semiconductor material, calculating the thermal conductivity of the semiconductor material according to the thermal diffusion coefficient and the specific heat capacity, querying the thermal expansion coefficient of the semiconductor material, and analyzing the thermal properties of the semiconductor material using the thermal conductivity and the thermal expansion coefficient;

[0009] querying multi-source stable test data of the semiconductor material, the multi-source stable test data including thermogravimetric test data, differential thermal test data, immersion test data and corrosion test data, constructing an environmental performance analysis model of the semiconductor material based on the multi-source stable test data, and analyzing the chemical properties of the semiconductor material using the environmental performance analysis model;

[0010] Based on the electrical properties, the optical properties, the thermal properties and the chemical properties, the material properties of the semiconductor material are analyzed using a pre-trained material analysis model.

[0011] Optionally, calculating the carrier concentration of the semiconductor material based on the material data includes:

[0012] Calculate the conduction band effective state density and the valence band effective state density of the semiconductor material based on the material data;

[0013] Based on the conduction band effective state density and the valence band effective state density, the carrier concentration of the semiconductor material is calculated using the following formula:

[0014]

[0015] in, represents the carrier concentration, represents the conduction band effective state density, represents the valence band effective state density, represents the band gap of the semiconductor material, k represents the Boltzmann constant, Indicates the absolute temperature of semiconductor materials.

[0016] Optionally, after collecting the photoelectron spectrum of the semiconductor material, the method further comprises:

[0017] Classifying the photoelectron spectrum into categories of absorption performance;

[0018] Wherein, the classification of the absorption performance of the photoelectron spectrum includes:

[0019] converting the photoelectron spectrum into vector data;

[0020] Based on the vector data, construct sample labels of the photoelectron energy spectrum;

[0021] The light energy absorption performance category of the semiconductor material is classified based on the sample label using the following formula:

[0022]

[0023] in, Indicates the light energy absorption performance category, indicating Symbolic function, M represents the number of light energy absorption performance categories, represents the Lagrange multiplier, represents the i-th label category, represents the kernel function, represents the i-th sample label, represents the bias term, Represents vector data.

[0024] Optionally, analyzing the optical properties of the semiconductor material according to the light energy absorption coefficient, the light energy absorption performance, the light energy refractive index and the luminous efficiency includes:

[0025] After determining the application scenario of the semiconductor material, the light energy absorption coefficient, the light energy absorption performance, the light energy refractive index and the luminous efficiency are used as optical performance evaluation indicators of the semiconductor material;

[0026] Refining the evaluation index to obtain a refined index;

[0027] Performing weight assignment on the refined indexes to obtain index weights;

[0028] Calculating a weight score of the indicator weight;

[0029] Based on the weighted scores, optical properties of the semiconductor material are analyzed.

[0030] Optionally, analyzing the thermal properties of the semiconductor material by using the thermal conductivity and the thermal expansion coefficient includes:

[0031] constructing a three-dimensional model of the semiconductor material;

[0032] Defining properties of the semiconductor material in the three-dimensional model to obtain a first operating three-dimensional model;

[0033] Setting boundaries for the first operating three-dimensional model to obtain a second operating three-dimensional model;

[0034] Based on the thermal conductivity, a thermal conduction analysis is performed on the semiconductor material to obtain a thermal conduction analysis result;

[0035] Performing thermal stress analysis on the semiconductor material based on the thermal expansion coefficient to obtain a thermal stress analysis result;

[0036] Based on the heat conduction analysis results and the thermal stress analysis results, construct a temperature cloud map and a stress cloud map of the semiconductor material;

[0037] Using the temperature cloud map and the stress cloud map, filling the second operating three-dimensional model to obtain a target three-dimensional model;

[0038] The target three-dimensional model is used to analyze the thermal properties of the semiconductor material.

[0039] Optionally, constructing the environmental performance analysis model of the semiconductor material based on the multi-source stable test data includes:

[0040] Performing data preprocessing on the multi-source stable test data to obtain preprocessed data;

[0041] constructing an initial environmental analysis model of the semiconductor material using the preprocessed data;

[0042] Performing model training on the initial environment analysis model to obtain a training model;

[0043] Performing performance evaluation on the training model to obtain a performance evaluation result;

[0044] When the performance evaluation result meets the preset result, obtaining the target model;

[0045] The target model is optimized to obtain an environmental performance analysis model.

[0046] Optionally, performing performance evaluation on the training model to obtain a performance evaluation result includes:

[0047] Constructing a test sample of the training model;

[0048] Based on the test samples, the mean absolute error of the training model is calculated using the following formula:

[0049]

[0050] in, represents the mean absolute error, n represents the number of test samples, represents the true value of the jth sample of the test sample, Represents the predicted value of the jth sample of the test sample;

[0051] The performance of the training model is evaluated according to the mean absolute error to obtain a performance evaluation result.

[0052] Optionally, analyzing the material properties of the semiconductor material based on the electrical properties, the optical properties, the thermal properties and the chemical properties using a pre-trained material analysis model includes:

[0053] querying the performance parameters of the electrical properties, the optical properties, the thermal properties and the chemical properties;

[0054] Performing standardization processing on the performance parameters to obtain standardized parameters;

[0055] After the standardized parameters are input as test data into the pre-trained material analysis model, the model fit of the pre-trained material analysis model is calculated using the test data in combination with the following formula:

[0056]

[0057] in, represents the model fit, N represents the amount of test data, Represents the predicted value of the u-th data of the test data, Represents the true value of the u-th data in the test data, Represents the mean of the true value of the u-th data in the test data;

[0058] When the degree of fit meets a preset degree of fit, taking the pre-trained material analysis model as a target material analysis model;

[0059] Calculating a material application score of the semiconductor material under preset material application conditions using the target material analysis model;

[0060] Based on the material application score, material properties of the semiconductor material are analyzed.

[0061] Optionally, the using the target material analysis model to calculate the material application score of the semiconductor material under a preset material application condition includes:

[0062] Using the target material analysis model, identifying the scene adaptation coefficient and reliability coefficient of the semiconductor material;

[0063] Based on the scenario adaptation coefficient and the reliability coefficient, the material application score of the semiconductor material is calculated using the following formula:

[0064]

[0065] in, Indicates the material application score, Indicates the electrical performance score, represents the importance coefficient of electrical performance, Indicates the optical performance score, represents the optical performance importance coefficient, represents the thermal performance score, represents the thermal performance importance coefficient, Indicates the chemical performance score, represents the chemical property importance coefficient, represents the scene adaptation coefficient, Represents the reliability coefficient.

[0066] In order to solve the above problems, the present invention also provides a performance analysis system for semiconductor materials based on intelligent AI, the system comprising:

[0067] An electrical performance analysis module, used to obtain material data of a semiconductor material, calculate a carrier concentration of the semiconductor material based on the material data, identify a material composition, a material structure and a material process of the semiconductor material, analyze a resistance property of the semiconductor material based on the material composition, the material structure and the material process, and analyze an electrical property of the semiconductor material based on the carrier concentration and the resistance property;

[0068] an optical performance analysis module, for collecting the photoelectron energy spectrum of the semiconductor material, calculating the light energy absorption coefficient of the semiconductor material using the photoelectron energy spectrum, analyzing the light energy absorption performance of the semiconductor material based on the material structure and the material composition, detecting the light energy refractive index and luminous efficiency of the semiconductor material, and analyzing the optical performance of the semiconductor material according to the light energy absorption coefficient, the light energy absorption performance, the light energy refractive index and the luminous efficiency;

[0069] A thermal performance analysis module, used to query the thermal diffusion coefficient and specific heat capacity of the semiconductor material, calculate the thermal conductivity of the semiconductor material according to the thermal diffusion coefficient and the specific heat capacity, query the thermal expansion coefficient of the semiconductor material, and analyze the thermal performance of the semiconductor material using the thermal conductivity and the thermal expansion coefficient;

[0070] A chemical performance analysis module, used to query multi-source stable test data of the semiconductor material, the multi-source stable test data including thermogravimetric test data, differential thermal test data, immersion test data and corrosion test data, and to construct an environmental performance analysis model of the semiconductor material based on the multi-source stable test data, and to analyze the chemical properties of the semiconductor material using the environmental performance analysis model;

[0071] The material property analysis module is used to analyze the material properties of the semiconductor material based on the electrical property, the optical property, the thermal property and the chemical property using a pre-trained material analysis model.

[0072] The present invention first analyzes the electrical properties, chemical properties, thermal properties and optical properties of the semiconductor material, so as to understand the material properties of the semiconductor material from multiple aspects and ensure the comprehensiveness of the analysis; secondly, when performing electrical and chemical property analysis, the present invention mainly analyzes the material structure and composition of the semiconductor material to ensure the deep division of data in the data processing process and the reliability of data processing, and combines the carrier concentration and the resistance performance of the semiconductor material in the electrical property analysis to achieve the reliability of the electrical property analysis in a complex environment, thereby ensuring the accuracy of the subsequent performance analysis of the semiconductor material, and combines the light energy absorption performance, light energy refractive index and luminous efficiency to achieve the reliability of the electrical property analysis in a complex environment. Analyze the comprehensive performance of semiconductor materials in optical applications; further, the embodiments of the present invention analyze the thermal properties of semiconductor materials by using thermal conductivity and thermal expansion coefficient to achieve a comprehensive evaluation of the thermal properties of semiconductor materials, thereby providing important thermal performance basis for the design and application of semiconductor devices; further, the embodiments of the present invention analyze the chemical property changes of semiconductor materials under various environmental conditions by using environmental performance analysis models, such as the possibility of reaction between materials and chemical substances in the environment, reaction rate, product generation, etc., to understand the chemical stability, oxidation resistance, corrosion resistance and other chemical performance indicators of semiconductor materials, so as to ensure accurate analysis of semiconductor materials used in various complex environments. Therefore, the present invention can improve the accuracy of semiconductor material performance analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 A schematic diagram of a process for realizing performance analysis of semiconductor materials based on intelligent AI provided by one embodiment of the present invention;

[0074] Figure 2 A functional module diagram of a semiconductor material performance analysis system based on intelligent AI provided by one embodiment of the present invention;

[0075] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0076] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0077] The embodiment of the present application provides a method for performance analysis of semiconductor materials based on intelligent AI. The execution subject of the method for performance analysis of semiconductor materials based on intelligent AI includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided in the embodiment of the present application. In other words, the method for performance analysis of semiconductor materials based on intelligent AI can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.

[0078] Reference Figure 1 FIG. 1 is a flow chart of a method for realizing performance analysis of semiconductor materials based on intelligent AI according to an embodiment of the present invention. In this embodiment, the method for realizing performance analysis of semiconductor materials based on intelligent AI includes:

[0079] S1. Acquire material data of a semiconductor material, calculate the carrier concentration of the semiconductor material based on the material data, identify the material composition, material structure and material process of the semiconductor material, analyze the resistance performance of the semiconductor material based on the material composition, the material structure and the material process, and analyze the electrical properties of the semiconductor material based on the carrier concentration and the resistance performance.

[0080] The embodiment of the present invention can understand the performance characteristics of semiconductor materials under different compositions, structures and process conditions by acquiring material data of semiconductor materials. The semiconductor material is a type of material with conductivity between that of a conductor and an insulator. Common semiconductor materials include silicon, germanium, gallium arsenide, etc. The material data refers to the relevant data of the semiconductor material, such as material category, composition, and related material measurement data.

[0081] The embodiment of the present invention can help users evaluate the quality of semiconductor materials by calculating the carrier concentration of the semiconductor material based on the material data.

[0082] As an embodiment of the present invention, the calculating the carrier concentration of the semiconductor material based on the material data includes: calculating the conduction band effective state density and the valence band effective state density of the semiconductor material based on the material data, and calculating the carrier concentration of the semiconductor material based on the conduction band effective state density and the valence band effective state density using the following formula:

[0083]

[0084] in, represents the carrier concentration, represents the conduction band effective state density, represents the valence band effective state density, represents the band gap of the semiconductor material, k represents the Boltzmann constant, Indicates the absolute temperature of semiconductor materials.

[0085] Among them, the conduction band effective state density refers to the number of effective quantum states that can be occupied by electrons in the conduction band. It reflects the ability of semiconductor materials to accommodate electrons in the conduction band at a certain temperature. It can be understood as the number of "positions" available for electrons to occupy in the conduction band, and the valence band effective state density refers to the number of effective quantum states that can be occupied by holes in the valence band. It reflects the ability of semiconductor materials to accommodate holes in the valence band at a certain temperature, that is, the number of "positions" available for holes to occupy in the valence band.

[0086] Optionally, the calculation of the conduction band effective state density and the valence band effective state density of the semiconductor material based on the material data can be obtained by identifying the crystal structure, energy band structure and temperature data of the semiconductor material using the material data.

[0087] Furthermore, the embodiments of the present invention can understand the basic properties of semiconductor materials by identifying the material composition, material structure and material process of the semiconductor materials, which can be obtained by querying the material data.

[0088] The embodiments of the present invention can help users evaluate the performance of the material in terms of power consumption, signal transmission, etc. in the circuit by analyzing the resistance performance of the semiconductor material based on the material composition, the material structure, and the material process.

[0089] Optionally, the analysis of the resistance performance of the semiconductor material based on the material composition, the material structure and the material process may calculate the resistivity of the semiconductor material through the material composition, the material structure and the material process, so as to analyze the resistance performance of the semiconductor material based on the resistivity.

[0090] The embodiment of the present invention can more comprehensively evaluate the electrical properties of the semiconductor material by analyzing the electrical properties of the semiconductor material according to the carrier concentration and the resistance performance.

[0091] Optionally, the electrical properties of the semiconductor material are analyzed according to the carrier concentration and the resistance performance by inputting the corresponding parameters of the carrier concentration and the resistance performance into a preconfigured deep learning model, and using the deep learning model to analyze the electrical properties of the semiconductor material according to historical semiconductor material analysis results. The preconfigured deep learning model is a machine learning method based on artificial neural networks, which automatically learns features and patterns from large amounts of data by constructing a neural network structure with multiple levels to achieve intelligent processing of various tasks.

[0092] S2. Collect the photoelectron energy spectrum of the semiconductor material, calculate the light energy absorption coefficient of the semiconductor material using the photoelectron energy spectrum, analyze the light energy absorption performance of the semiconductor material based on the material structure and the material composition, detect the light energy refractive index and luminous efficiency of the semiconductor material, and analyze the optical properties of the semiconductor material according to the light energy absorption coefficient, the light energy absorption performance, the light energy refractive index and the luminous efficiency.

[0093] The embodiment of the present invention can obtain electronic structure information about the semiconductor material by collecting the photoelectron energy spectrum of the semiconductor material, including the valence band, conduction band and surface electronic state, etc., and the photoelectron energy spectrum can be collected by an X-ray photoelectron spectrometer.

[0094] The photoelectron spectroscopy is an analytical technique used to study the electronic structure of a material surface.

[0095] Furthermore, in the present invention, after collecting the photoelectron energy spectrum of the semiconductor material, it also includes classifying the photoelectron energy spectrum into categories of absorption performance, so that when the light energy absorption coefficient of the semiconductor material is subsequently calculated through the photoelectron energy spectrum, the absorption capacity of the semiconductor material can be understood, thereby ensuring the accuracy of the subsequent optical performance analysis of the semiconductor material.

[0096] Specifically, the classification of the absorption performance of the photoelectron energy spectrum includes: converting the photoelectron energy spectrum into vector data, constructing a sample label of the photoelectron energy spectrum based on the vector data, and classifying the light energy absorption performance category of the semiconductor material based on the sample label using the following formula:

[0097]

[0098] in, Indicates the light energy absorption performance category, represents the sign function, M represents the number of light energy absorption performance categories, represents the Lagrange multiplier, represents the i-th label category, represents the kernel function, represents the i-th sample label, represents the bias term, Represents vector data.

[0099] It should be noted that the It can be the light energy absorption performance category corresponding to different photoelectron energy spectrum characteristics, It refers to a label constructed based on the photoelectron energy spectrum. Furthermore, since the photoelectron energy spectrum contains rich information about the electronic structure of semiconductor materials, these complex data can be represented in a more convenient form through vector conversion. The conversion of the photoelectron energy spectrum into vector data can be performed through a vector conversion matrix. The sample label of the photoelectron energy spectrum based on the vector data can be constructed through a script generated by java.

[0100] Optionally, the light energy absorption coefficient of the semiconductor material calculated by using the photoelectron energy spectrum can be obtained by combining the photoelectron energy spectrum with a light energy absorption coefficient calculation formula.

[0101] The embodiments of the present invention analyze the light energy absorption performance of the semiconductor material based on the material structure and the material composition, and conduct in-depth research on the material's absorption characteristics of light energy, including the absorption wavelength range, absorption intensity, etc. For example, by analyzing the light energy absorption performance, the applicability of the material in a specific application scenario can be clarified.

[0102] Among them, material structure refers to the organization and arrangement of materials at the microscopic and macroscopic scales, which determines many physical and chemical properties of the material, and the material composition refers to the chemical elements that constitute the material and their relative contents.

[0103] Optionally, the light energy absorption performance of the semiconductor material based on the material structure and the material composition can be analyzed by analyzing the influence of the material structure through the photoelectron energy spectrum and light energy absorption coefficient of semiconductor materials with different structures, and the absorption performance of the material composition can be analyzed through the photoelectron energy spectrum and light energy absorption coefficient of semiconductor materials with different material compositions.

[0104] The embodiment of the present invention can determine the propagation characteristics of light in the material by detecting the light energy refractive index and luminous efficiency of the semiconductor material, and the luminous efficiency reflects the ability of the material to emit light when excited, thereby understanding the optical characteristics of the semiconductor material. The light energy refractive index of the semiconductor material can be measured by an ellipsometer, and the luminous efficiency can be measured by a photoluminescence spectrometer.

[0105] The light energy refractive index refers to a measure of the semiconductor material's ability to refract light, and the luminous efficiency refers to an indicator used to measure the luminous performance of a semiconductor material.

[0106] Furthermore, the embodiments of the present invention can understand the comprehensive performance of semiconductor materials in optical applications by analyzing the optical properties of the semiconductor material according to the light energy absorption performance, the light energy refractive index and the luminous efficiency. Different optical applications have different performance requirements for materials. For example, high light energy absorption performance and moderate refractive index are required in solar cells, while luminous efficiency is more important in light-emitting devices. By comprehensively analyzing the optical properties, guidance can be provided for the application of materials in different fields, and goals and directions can be provided for further optimizing material performance.

[0107] As an embodiment of the present invention, the optical properties of the semiconductor material are analyzed based on the light energy absorption coefficient, the light energy absorption performance, the light energy refractive index and the luminous efficiency, including: after determining the application scenario of the semiconductor material, using the light energy absorption coefficient, the light energy absorption performance, the light energy refractive index and the luminous efficiency as optical performance evaluation indicators of the semiconductor material, and refining the evaluation indicators to obtain refined indicators, assigning weights to the refined indicators to obtain indicator weights, calculating weight scores of the indicator weights, and analyzing the optical properties of the semiconductor material based on the weight scores.

[0108] Among them, the application scenario of the semiconductor material can be set to provide an efficient and stable lighting source scenario or an application scenario of efficiently absorbing sunlight and converting it into electrical energy. The light energy absorption performance, the light energy refractive index and the luminous efficiency as the optical performance evaluation index of the semiconductor material can be set accordingly through the application scenario. For example, in solar cell applications, light energy absorption performance and light energy refractive index are crucial to improving light capture and transmission, while in LED lighting, luminous efficiency is one of the core indicators. The evaluation index is refined to obtain the refined index by the following method. For example, for light energy absorption performance, it can be refined into indicators such as absorption coefficient and absorption bandwidth at different wavelengths. For example, for solar cell materials, their absorption coefficients in the visible light and near-infrared regions can be measured respectively to evaluate the absorption capacity of sunlight. The light energy refractive index can be refined into indicators such as refractive index at different wavelengths and changes in refractive index with temperature. For example, in the field of optical communications, it is necessary to consider the stability of the refractive index of the material at different temperatures to ensure the transmission quality of the optical signal. The luminous efficiency can be refined into indicators such as internal quantum efficiency, external quantum efficiency, and luminous intensity. For example, in LED lighting, internal quantum efficiency and external quantum efficiency reflect the luminous efficiency inside and outside the material respectively, and luminous intensity can intuitively reflect the luminous ability of the material. The weights of the detailed indicators can be obtained by assigning weights to the indicator weights to determine the importance of each detailed indicator according to the application scenario. For example, in a solar cell, the importance of light energy absorption performance may be relatively high, so the weight of its refined index may be relatively large. Then, the hierarchical analysis method may be used as an example, with optical performance as the target layer, light energy absorption performance, light energy refractive index and luminous efficiency as the criterion layer, and refined index as the solution layer. By constructing a judgment matrix, the relative importance of each index is calculated to obtain the index weight. The weight score of the index weight may be determined by determining a score for each refined index according to its actual measurement value or evaluation result. For example, the measurement value of the refined index may be compared with the preset standard value, and different scores may be given according to the degree of deviation. Then, the score of each refined index may be multiplied by its corresponding index weight to obtain a weighted score. Finally, the weighted scores of all refined indexes may be added to obtain a weighted score. The optical performance of the semiconductor material may be analyzed based on the weighted score to judge the optical performance of the semiconductor material according to the weighted score. For example, if the weighted score is 80 points and the preset weighted score is 75, it means that the optical performance of the semiconductor material is better.

[0109] S3. Query the thermal diffusion coefficient and specific heat capacity of the semiconductor material, calculate the thermal conductivity of the semiconductor material based on the thermal diffusion coefficient and the specific heat capacity, query the thermal expansion coefficient of the semiconductor material, and use the thermal conductivity and the thermal expansion coefficient to analyze the thermal properties of the semiconductor material.

[0110] The embodiment of the present invention can help users understand the heat dissipation performance of the semiconductor material during application by querying the thermal diffusion coefficient and specific heat capacity of the semiconductor material.

[0111] Optionally, the thermal diffusion coefficient can be measured by a laser flash method, and the specific heat capacity can be measured by a differential scanning calorimetry method.

[0112] Furthermore, the embodiments of the present invention can fully understand the heat transfer performance of semiconductor materials by calculating the thermal conductivity of the semiconductor material according to the thermal diffusion coefficient and the specific heat capacity, because in the design of electronic devices, accurate thermal conductivity data is crucial for optimizing the heat dissipation structure, selecting appropriate packaging materials, and improving the performance and stability of the device. The thermal conductivity can be calculated by querying the thermal diffusion coefficient, material density and specific heat capacity of the semiconductor material, and then according to the thermal conductivity calculation formula.

[0113] Furthermore, the embodiments of the present invention can comprehensively evaluate the thermal performance of semiconductor materials by querying the thermal expansion coefficient of the semiconductor material and analyzing the thermal performance of the semiconductor material using the thermal conductivity and the thermal expansion coefficient, thereby providing important thermal performance basis for the design and application of semiconductor devices. The thermal expansion coefficient refers to the rate of dimensional change of a material when the temperature changes, which can be measured by a thermomechanical analysis method.

[0114] As an embodiment of the present invention, the use of the thermal conductivity and the thermal expansion coefficient to analyze the thermal properties of the semiconductor material includes: constructing a three-dimensional model of the semiconductor material, defining properties of the semiconductor material in the three-dimensional model to obtain a first operating three-dimensional model, setting boundaries for the first operating three-dimensional model to obtain a second operating three-dimensional model, performing a thermal conduction analysis on the semiconductor material based on the thermal conductivity to obtain a thermal conduction analysis result, performing a thermal stress analysis on the semiconductor material based on the thermal expansion coefficient to obtain a thermal stress analysis result, constructing a temperature cloud map and a stress cloud map of the semiconductor material based on the thermal conduction analysis result and the thermal stress analysis result, using the temperature cloud map and the stress cloud map to fill the second operating three-dimensional model to obtain a target three-dimensional model, and using the target three-dimensional model to analyze the thermal properties of the semiconductor material.

[0115] Among them, the temperature cloud map refers to an image representation method that uses color coding to intuitively display the surface or internal temperature distribution of an object, and the stress cloud map refers to an image representation method that uses color to represent the internal or surface stress distribution of an object.

[0116] Optionally, the three-dimensional model can be constructed by CATIA tools, and the property definition of the three-dimensional model can be achieved by inputting thermal performance parameters such as thermal conductivity, thermal expansion coefficient and mechanical performance parameters such as elastic modulus of the semiconductor material into the three-dimensional model. The boundary setting of the first operating three-dimensional model can be achieved by setting the temperature value, heat flux density value and displacement boundary condition on the boundary of the three-dimensional model. The thermal conduction analysis of the semiconductor material can be calculated by the heat conduction formula, and the temperature cloud map and stress cloud map can be constructed by Microsoft Excel tools. The use of the target three-dimensional model to analyze the thermal properties of the semiconductor material can be understood through intuitive observation of the model.

[0117] It should be noted that the temperature boundary condition indicates the temperature value on the boundary of the specified model. For example, the temperature of certain boundaries can be set to a constant value, or the heat exchange conditions with the surrounding environment (such as convection heat exchange, radiation heat exchange, etc.) can be set according to actual conditions; the heat flow boundary condition indicates the heat flux density value on the boundary of the specified model. For example, certain boundaries can be set to have specific heat flow input or output.

[0118] The displacement boundary condition setting means that in order to determine the deformation of the material, the displacement boundary condition needs to be set. For example, the displacement on certain boundaries can be limited to zero, or the connection conditions with the surrounding structure can be set according to the actual situation.

[0119] S4. Query the multi-source stable test data of the semiconductor material, wherein the multi-source stable test data includes thermogravimetric test data, differential thermal test data, immersion test data and corrosion test data; construct an environmental performance analysis model of the semiconductor material based on the multi-source stable test data; and analyze the chemical properties of the semiconductor material using the environmental performance analysis model.

[0120] The embodiment of the present invention can understand the quality performance of the semiconductor material under different environments by querying the multi-source stable test data of the semiconductor material, wherein the multi-source stable test data includes thermogravimetric test data, differential thermal test data, immersion test data and corrosion test data, wherein the thermogravimetric test data refers to the mass change data of the semiconductor material under different temperature experiments, and the differential thermal test data refers to the performance change test data of the semiconductor material after being immersed in a specific liquid environment (such as different solvents, aqueous solutions, etc.) for a period of time, including data such as mass change, size change, appearance change, and possible component dissolution or precipitation, etc. The immersion test data refers to the performance change test data of the semiconductor material after being immersed in a specific liquid environment (such as different solvents, aqueous solutions, etc.) for a period of time, and the corrosion test data refers to the corrosion resistance test data of the semiconductor material in a specific corrosion environment (such as acid, alkali, salt solution, etc.).

[0121] Furthermore, the embodiments of the present invention can integrate and correlate various types of test data by constructing an environmental performance analysis model of the semiconductor material based on the multi-source stable test data, and can comprehensively and systematically describe the performance and change patterns of semiconductor materials under different environmental conditions by establishing a mathematical model or logical framework.

[0122] As an embodiment of the present invention, the environmental performance analysis model of the semiconductor material is constructed based on the multi-source stable test data, including: performing data preprocessing on the multi-source stable test data to obtain preprocessed data, using the preprocessed data to construct an initial environmental analysis model of the semiconductor material, performing model training on the initial environmental analysis model to obtain a training model, performing performance evaluation on the training model to obtain a performance evaluation result, obtaining a target model when the performance evaluation result meets a preset result, and performing model optimization on the target model to obtain an environmental performance analysis model.

[0123] In the present invention, the data preprocessing includes processing missing values, outliers, duplicate values ​​and performing feature engineering and other operations. The initial environmental analysis model of the semiconductor material can be constructed by utilizing the preprocessed data in combination with a multivariate regression function. The model training of the initial environmental analysis model can be performed by using the test results of the corresponding multi-source stable test data in the preprocessed data as the model output value and the multi-source stable test data as the test value as the input value for model training. When the performance evaluation result meets the preset result, obtaining the target model means that the corresponding evaluation value of the performance evaluation is not less than 0.9, which can be specifically set in combination with the actual application scenario. The model optimization of the target model to obtain the environmental performance analysis model means adjusting the parameters of the model to improve the accuracy of the model analysis, which can be achieved by adjusting the model hyperparameters through feature engineering.

[0124] Further, as another optional embodiment of the present invention, the performance evaluation of the training model to obtain the performance evaluation result includes: constructing a test sample of the training model, and based on the test sample, calculating the mean absolute error of the training model using the following formula:

[0125]

[0126] in, represents the mean absolute error, n represents the number of test samples, represents the true value of the jth sample of the test sample, Represents the predicted value of the jth sample of the test sample;

[0127] The performance of the training model is evaluated according to the mean absolute error to obtain a performance evaluation result.

[0128] Optionally, the test samples for constructing the training model can be constructed by using multi-source stable test data, and the performance of the training model is evaluated based on the mean absolute error. The performance evaluation result is evaluated based on the calculation result of the mean absolute error. If the mean absolute error is not greater than 0.1, it indicates that the performance of the training model is excellent.

[0129] Furthermore, the embodiment of the present invention can analyze the chemical properties of the semiconductor material by using the environmental performance analysis model. Based on the constructed environmental performance analysis model, the semiconductor material can focus on studying the chemical property changes of the semiconductor material under various environmental conditions, such as the possibility of the material reacting with the chemical substances in the environment, the reaction rate, the product generation, etc. The chemical performance indicators such as the chemical stability, oxidation resistance, and corrosion resistance of the material can be analyzed.

[0130] S5. Based on the electrical properties, the optical properties, the thermal properties and the chemical properties, use a pre-trained material analysis model to analyze the material properties of the semiconductor material.

[0131] The embodiment of the present invention analyzes the material properties of the semiconductor material based on the electrical properties, the optical properties, the thermal properties and the chemical properties using a pre-trained material analysis model, which can comprehensively consider the electrical, optical, thermal and chemical properties to more comprehensively understand the overall characteristics of the semiconductor material and provide accurate evaluation for its application in different fields. Among them, the pre-trained material analysis model refers to an intelligent AI model for realizing material performance analysis, which can be obtained through data training by combining a deep learning model with the historical material data of the semiconductor material.

[0132] As an embodiment of the present invention, the material properties of the semiconductor material are analyzed based on the electrical properties, the optical properties, the thermal properties and the chemical properties using a pre-trained material analysis model, including: querying the performance parameters of the electrical properties, the optical properties, the thermal properties and the chemical properties, standardizing the performance parameters to obtain standardized parameters, inputting the standardized parameters as test data into the pre-trained material analysis model, and calculating the model fit of the pre-trained material analysis model using the test data in combination with the following formula:

[0133]

[0134] in, represents the model fit, N represents the amount of test data, Represents the predicted value of the u-th data of the test data, Represents the true value of the u-th data in the test data, Represents the mean of the true value of the u-th data in the test data;

[0135] When the degree of fit meets the preset degree of fit, the pre-trained material analysis model is used as the target material analysis model, and the target material analysis model is used to calculate the material application score of the semiconductor material under preset material application conditions, and based on the material application score, the material properties of the semiconductor material are analyzed.

[0136] The material application conditions refer to the material application scenarios of the model, such as electrical application scenarios, optical application scenarios, and thermal application scenarios.

[0137] It should be noted that when the degree of fit satisfies a preset degree of fit, it means that the degree of fit is not less than 0.9. The preset degree of fit is a preset value, which is 0.9, and can also be set according to actual application scenarios.

[0138] Optionally, the standardization of the performance parameters to obtain standardized parameters refers to preprocessing of the data, which can be processed by a data normalization method. Based on the material application score, the material performance of the semiconductor material can be analyzed by comparing the material application score with a preset score. For example, if the material application score is below 0.5, it means that the material performance is average; if the material application score is between 0.5 and 0.8, it means that the material performance is better; and if the material application score is above 0.8, it means that the material performance is excellent.

[0139] As an optional embodiment of the present invention, the use of the target material analysis model to calculate the material application score of the semiconductor material under preset material application conditions includes: using the target material analysis model to identify the scene adaptation coefficient and reliability coefficient of the semiconductor material, and based on the scene adaptation coefficient and the reliability coefficient, using the following formula to calculate the material application score of the semiconductor material:

[0140]

[0141] in, Indicates the material application score, Indicates the electrical performance score, represents the importance coefficient of electrical performance, Indicates the optical performance score, represents the optical performance importance coefficient, represents the thermal performance score, represents the thermal performance importance coefficient, Indicates the chemical performance score, represents the chemical property importance coefficient, represents the scene adaptation coefficient, Represents the reliability coefficient.

[0142] Among them, the scenario adaptation coefficient refers to the adaptation degree index of the material in a specific application scenario, and the reliability coefficient refers to a parameter for measuring the long-term reliability and stability of the material.

[0143] Optionally, the scene adaptation coefficient and reliability coefficient of the semiconductor material identified by using the target material analysis model can determine the importance of each performance and set parameters for material performance through different application scenarios, such as in the field of semiconductor chip manufacturing: assuming that the conductivity score in the electrical performance is 80 points, the conductivity weight is 0.3; the resistivity score is 70 points, and the resistivity weight is 0.2. The transmittance score in the optical performance is 60 points, the transmittance weight is 0.1; the refractive index score is 75 points, and the refractive index weight is 0.1. The thermal conductivity score in the thermal performance is 70 points, the thermal conductivity weight is 0.15; the thermal expansion coefficient score is 65 points, and the thermal expansion coefficient weight is 0.1. The corrosion resistance score in the chemical performance is 85 points, the corrosion resistance weight is 0.05; the stability score is 75 points, and the stability weight is 0.05. The importance coefficient is set to 0.4 for electrical performance, 0.2 for optical performance, 0.3 for thermal performance, and 0.1 for chemical performance according to the characteristics of the field. The application scenario adaptation coefficient is assumed to be 0.8 (taking into account factors such as high temperature and complex environment during chip manufacturing). The reliability coefficient is determined to be 0.9 through reliability testing.

[0144] like Figure 2 As shown, it is a functional module diagram of a semiconductor material performance analysis system based on intelligent AI provided by one embodiment of the present invention.

[0145] The performance analysis system 200 for semiconductor materials based on intelligent AI described in the present invention can be installed in an electronic device. According to the functions implemented, the performance analysis system 200 for semiconductor materials based on intelligent AI can include an electrical performance analysis module 201, an optical performance analysis module 202, a thermal performance analysis module 203, a chemical performance analysis module 204, and a material performance analysis module 205. The module described in the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0146] In this embodiment, the functions of each module / unit are as follows:

[0147] The electrical performance analysis module 201 is used to obtain material data of semiconductor materials, calculate the carrier concentration of the semiconductor material based on the material data, identify the material composition, material structure and material process of the semiconductor material, analyze the resistance performance of the semiconductor material based on the material composition, the material structure and the material process, and analyze the electrical performance of the semiconductor material according to the carrier concentration and the resistance performance;

[0148] The optical performance analysis module 202 is used to collect the photoelectron spectrum of the semiconductor material, calculate the light energy absorption coefficient of the semiconductor material using the photoelectron spectrum, analyze the light energy absorption performance of the semiconductor material based on the material structure and the material composition, detect the light energy refractive index and luminous efficiency of the semiconductor material, and analyze the optical performance of the semiconductor material according to the light energy absorption coefficient, the light energy absorption performance, the light energy refractive index and the luminous efficiency;

[0149] The thermal performance analysis module 203 is used to query the thermal diffusion coefficient and specific heat capacity of the semiconductor material, calculate the thermal conductivity of the semiconductor material according to the thermal diffusion coefficient and the specific heat capacity, query the thermal expansion coefficient of the semiconductor material, and analyze the thermal performance of the semiconductor material using the thermal conductivity and the thermal expansion coefficient;

[0150] The chemical performance analysis module 204 is used to query the multi-source stable test data of the semiconductor material, the multi-source stable test data includes thermogravimetric test data, differential thermal test data, immersion test data and corrosion test data, and to construct an environmental performance analysis model of the semiconductor material based on the multi-source stable test data, and to analyze the chemical properties of the semiconductor material using the environmental performance analysis model;

[0151] The material property analysis module 205 is used to analyze the material properties of the semiconductor material based on the electrical properties, the optical properties, the thermal properties and the chemical properties using a pre-trained material analysis model.

[0152] In detail, the modules described in the semiconductor material performance analysis system 200 based on intelligent AI described in the embodiment of the present invention adopt the same technical means as the semiconductor material performance analysis method based on intelligent AI described in the accompanying drawings when used, and can produce the same technical effects, which will not be repeated here.

[0153] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0154] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0155] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0156] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0157] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0158] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0159] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or systems stated in a system claim can also be implemented by one unit or system through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0160] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for analyzing the performance of semiconductor materials based on intelligent AI, characterized in that: The method comprises: Acquiring material data of a semiconductor material, calculating a carrier concentration of the semiconductor material based on the material data, identifying a material composition, a material structure, and a material process of the semiconductor material, analyzing a resistance property of the semiconductor material based on the material composition, the material structure, and the material process, and analyzing an electrical property of the semiconductor material based on the carrier concentration and the resistance property; Collecting the photoelectron spectrum of the semiconductor material, calculating the light energy absorption coefficient of the semiconductor material using the photoelectron spectrum, analyzing the light energy absorption performance of the semiconductor material based on the material structure and the material composition, detecting the light energy refractive index and luminous efficiency of the semiconductor material, and analyzing the optical properties of the semiconductor material according to the light energy absorption coefficient, the light energy absorption performance, the light energy refractive index and the luminous efficiency; Querying the thermal diffusion coefficient and specific heat capacity of the semiconductor material, calculating the thermal conductivity of the semiconductor material according to the thermal diffusion coefficient and the specific heat capacity, querying the thermal expansion coefficient of the semiconductor material, and analyzing the thermal properties of the semiconductor material using the thermal conductivity and the thermal expansion coefficient; querying multi-source stable test data of the semiconductor material, the multi-source stable test data including thermogravimetric test data, differential thermal test data, immersion test data and corrosion test data, constructing an environmental performance analysis model of the semiconductor material based on the multi-source stable test data, and analyzing the chemical properties of the semiconductor material using the environmental performance analysis model; Wherein, constructing the environmental performance analysis model of the semiconductor material based on the multi-source stable test data includes: performing data preprocessing on the multi-source stable test data to obtain preprocessed data, using the preprocessed data to construct an initial environmental analysis model of the semiconductor material, performing model training on the initial environmental analysis model to obtain a training model, performing performance evaluation on the training model to obtain a performance evaluation result, obtaining a target model when the performance evaluation result meets a preset result, and performing model optimization on the target model to obtain an environmental performance analysis model; Furthermore, the performance evaluation of the training model to obtain the performance evaluation result includes: constructing a test sample of the training model, and based on the test sample, calculating the mean absolute error of the training model using the following formula: in, represents the mean absolute error, n represents the number of test samples, represents the true value of the jth sample of the test sample, represents the predicted value of the jth sample of the test sample, and the performance of the training model is evaluated according to the mean absolute error to obtain a performance evaluation result; Analyzing the material properties of the semiconductor material using a pre-trained material analysis model based on the electrical properties, the optical properties, the thermal properties and the chemical properties; Wherein, the material properties of the semiconductor material are analyzed based on the electrical properties, the optical properties, the thermal properties and the chemical properties using a pre-trained material analysis model, including: querying the performance parameters of the electrical properties, the optical properties, the thermal properties and the chemical properties, standardizing the performance parameters to obtain standardized parameters, inputting the standardized parameters as test data into the pre-trained material analysis model, and calculating the model fit of the pre-trained material analysis model using the test data in combination with the following formula: in, represents the model fit, N represents the amount of test data, Represents the predicted value of the u-th data of the test data, Represents the true value of the u-th data in the test data, represents the mean of the true values ​​of the u-th data of the test data, and when the degree of fit satisfies a preset degree of fit, the pre-trained material analysis model is used as a target material analysis model, and the target material analysis model is used to calculate the material application score of the semiconductor material under the preset material application conditions, and based on the material application score, the material properties of the semiconductor material are analyzed; Further, the use of the target material analysis model to calculate the material application score of the semiconductor material under preset material application conditions includes: using the target material analysis model to identify the scene adaptation coefficient and the reliability coefficient of the semiconductor material, and based on the scene adaptation coefficient and the reliability coefficient, using the following formula to calculate the material application score of the semiconductor material: in, Indicates the material application score, Indicates the electrical performance score, represents the importance coefficient of electrical performance, Indicates the optical performance score, represents the optical performance importance coefficient, represents the thermal performance score, represents the thermal performance importance coefficient, Indicates the chemical performance score, represents the chemical property importance coefficient, represents the scene adaptation coefficient, Represents the reliability coefficient.

2. A method for realizing performance analysis of semiconductor materials based on intelligent AI as claimed in claim 1, characterized in that: The calculating the carrier concentration of the semiconductor material based on the material data comprises: Calculate the conduction band effective state density and the valence band effective state density of the semiconductor material based on the material data; Based on the conduction band effective state density and the valence band effective state density, the carrier concentration of the semiconductor material is calculated using the following formula: in, represents the carrier concentration, represents the conduction band effective state density, represents the valence band effective state density, represents the band gap of the semiconductor material, k represents the Boltzmann constant, Indicates the absolute temperature of semiconductor materials.

3. A method for realizing performance analysis of semiconductor materials based on intelligent AI as claimed in claim 1, characterized in that: After collecting the photoelectron spectrum of the semiconductor material, the method further includes: classifying the photoelectron spectrum into categories of absorption performance; Wherein, the classification of the absorption performance of the photoelectron spectrum includes: converting the photoelectron spectrum into vector data; Based on the vector data, construct sample labels of the photoelectron energy spectrum; The light energy absorption performance category of the semiconductor material is classified based on the sample label using the following formula: in, Indicates the light energy absorption performance category, indicating Symbolic function, M represents the number of light energy absorption performance categories, represents the Lagrange multiplier, represents the i-th label category, represents the kernel function, represents the i-th sample label, represents the bias term, Represents vector data.

4. The method for realizing performance analysis of semiconductor materials based on intelligent AI according to claim 1, characterized in that: The analyzing the optical properties of the semiconductor material according to the light energy absorption coefficient, the light energy absorption performance, the light energy refractive index and the luminous efficiency includes: After determining the application scenario of the semiconductor material, the light energy absorption coefficient, the light energy absorption performance, the light energy refractive index and the luminous efficiency are used as optical performance evaluation indicators of the semiconductor material; Refining the evaluation index to obtain a refined index; Performing weight assignment on the refined indexes to obtain index weights; Calculating a weight score of the indicator weight; Based on the weighted scores, optical properties of the semiconductor material are analyzed.

5. The method for realizing performance analysis of semiconductor materials based on intelligent AI according to claim 1, characterized in that: The step of analyzing the thermal properties of the semiconductor material by using the thermal conductivity and the thermal expansion coefficient comprises: constructing a three-dimensional model of the semiconductor material; Defining properties of the semiconductor material in the three-dimensional model to obtain a first operating three-dimensional model; Setting boundaries for the first operating three-dimensional model to obtain a second operating three-dimensional model; Based on the thermal conductivity, a thermal conduction analysis is performed on the semiconductor material to obtain a thermal conduction analysis result; Performing thermal stress analysis on the semiconductor material based on the thermal expansion coefficient to obtain a thermal stress analysis result; Based on the heat conduction analysis results and the thermal stress analysis results, construct a temperature cloud map and a stress cloud map of the semiconductor material; Using the temperature cloud map and the stress cloud map, filling the second operating three-dimensional model to obtain a target three-dimensional model; The target three-dimensional model is used to analyze the thermal properties of the semiconductor material.

6. A performance analysis system for semiconductor materials based on intelligent AI, characterized in that: The system is used to perform a method for analyzing semiconductor material performance based on intelligent AI as described in any one of claims 1 to 5, the system comprising: An electrical performance analysis module, used to obtain material data of a semiconductor material, calculate a carrier concentration of the semiconductor material based on the material data, identify a material composition, a material structure and a material process of the semiconductor material, analyze a resistance property of the semiconductor material based on the material composition, the material structure and the material process, and analyze an electrical property of the semiconductor material based on the carrier concentration and the resistance property; an optical performance analysis module, for collecting the photoelectron energy spectrum of the semiconductor material, calculating the light energy absorption coefficient of the semiconductor material using the photoelectron energy spectrum, analyzing the light energy absorption performance of the semiconductor material based on the material structure and the material composition, detecting the light energy refractive index and luminous efficiency of the semiconductor material, and analyzing the optical performance of the semiconductor material according to the light energy absorption coefficient, the light energy absorption performance, the light energy refractive index and the luminous efficiency; A thermal performance analysis module, used to query the thermal diffusion coefficient and specific heat capacity of the semiconductor material, calculate the thermal conductivity of the semiconductor material according to the thermal diffusion coefficient and the specific heat capacity, query the thermal expansion coefficient of the semiconductor material, and analyze the thermal performance of the semiconductor material using the thermal conductivity and the thermal expansion coefficient; A chemical performance analysis module, used to query multi-source stable test data of the semiconductor material, the multi-source stable test data including thermogravimetric test data, differential thermal test data, immersion test data and corrosion test data, and to construct an environmental performance analysis model of the semiconductor material based on the multi-source stable test data, and to analyze the chemical properties of the semiconductor material using the environmental performance analysis model; Wherein, constructing the environmental performance analysis model of the semiconductor material based on the multi-source stable test data includes: performing data preprocessing on the multi-source stable test data to obtain preprocessed data, using the preprocessed data to construct an initial environmental analysis model of the semiconductor material, performing model training on the initial environmental analysis model to obtain a training model, performing performance evaluation on the training model to obtain a performance evaluation result, obtaining a target model when the performance evaluation result meets a preset result, and performing model optimization on the target model to obtain an environmental performance analysis model; Furthermore, the performance evaluation of the training model to obtain the performance evaluation result includes: constructing a test sample of the training model, and based on the test sample, calculating the mean absolute error of the training model using the following formula: in, represents the mean absolute error, n represents the number of test samples, represents the true value of the jth sample of the test sample, represents the predicted value of the jth sample of the test sample, and the performance of the training model is evaluated according to the mean absolute error to obtain a performance evaluation result; A material property analysis module, used for analyzing the material properties of the semiconductor material using a pre-trained material analysis model based on the electrical property, the optical property, the thermal property and the chemical property; Wherein, the material properties of the semiconductor material are analyzed based on the electrical properties, the optical properties, the thermal properties and the chemical properties using a pre-trained material analysis model, including: querying the performance parameters of the electrical properties, the optical properties, the thermal properties and the chemical properties, standardizing the performance parameters to obtain standardized parameters, inputting the standardized parameters as test data into the pre-trained material analysis model, and calculating the model fit of the pre-trained material analysis model using the test data in combination with the following formula: in, represents the model fit, N represents the amount of test data, Represents the predicted value of the u-th data of the test data, Represents the true value of the u-th data in the test data, represents the mean of the true values ​​of the u-th data of the test data, and when the degree of fit satisfies a preset degree of fit, the pre-trained material analysis model is used as a target material analysis model, and the target material analysis model is used to calculate the material application score of the semiconductor material under the preset material application conditions, and based on the material application score, the material properties of the semiconductor material are analyzed; Further, the use of the target material analysis model to calculate the material application score of the semiconductor material under preset material application conditions includes: using the target material analysis model to identify the scene adaptation coefficient and the reliability coefficient of the semiconductor material, and based on the scene adaptation coefficient and the reliability coefficient, using the following formula to calculate the material application score of the semiconductor material: in, Indicates the material application score, Indicates the electrical performance score, represents the importance coefficient of electrical performance, Indicates the optical performance score, represents the optical performance importance coefficient, represents the thermal performance score, represents the thermal performance importance coefficient, Indicates the chemical performance score, represents the chemical property importance coefficient, represents the scene adaptation coefficient, Represents the reliability coefficient.

Images