Performance data prediction method of semiconductor ceramic composite material

Abstract

The invention relates to the technical field of data processing, in particular to a method for predicting performance data of a semiconductor ceramic composite material, which comprises the following steps of: performing any round of repeatability test on a target piezoresistor according to a reference parameter, a test parameter and test times to obtain a performance characteristic sequence of any round of repeatability test; obtaining the stability coefficient of the target varistor according to the data change condition in the performance characteristic sequence, obtaining the performance expression index of the target varistor according to the performance characteristic sequence and the stability coefficient, and obtaining the new test parameters and new test times when the next round of repeatability test is carried out on the target varistor; according to the reference parameter, the new test parameter and the new test frequency, the performance expression index of the target varistor in the next round of repetitive test is obtained, the performance expression index of the target varistor under any test parameter is predicted according to all the performance expression indexes, and the accuracy of predicting the key performance of the varistor is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly relates to a method for predicting performance data of a semiconductor ceramic composite material. Background Art

[0002] Semiconductor ceramic composites are a type of functional material that combines semiconductor properties and ceramic structural characteristics, usually composed of transition metal oxides (such as etc.) and various doping or auxiliary phases. Due to their excellent electrical properties and other characteristics, they are widely used in the fields of electronic components, etc. Among them, efficient prediction of the key performance of semiconductor ceramic composites can significantly improve the efficiency and reliability of material research and device design: on the one hand, without the need for a large number of experiments, it can quickly judge the performance of materials under different formulations, processes or application conditions, thereby shortening the development cycle and reducing the R & D cost; on the other hand, accurate performance prediction can also be used for quality control and device performance screening to ensure that the final product has excellent stability, response characteristics and life performance in actual use, especially suitable for application scenarios of electronic components with high requirements for performance consistency such as varistors and thermistors.

[0003] For example, a varistor made of a semiconductor ceramic composite is an electronic component with good non-linear volt-ampere characteristics. Its core function is to sharply reduce the resistance when the voltage exceeds the threshold, thereby realizing the protection of the circuit against overvoltage, and is widely used in scenarios such as circuit protection, surge suppression, and lightning protection. Such devices usually use zinc oxide (ZnO) as the main matrix material, supplemented by a variety of metal oxides (such as etc.) to form a ceramic system, and the required electrical properties are achieved by controlling its microstructure (grains, grain boundaries, second phases, etc.).

[0004] In the existing methods for predicting the key performance of varistors, usually, the varistor is actually measured according to preset test conditions to obtain sample data for modeling analysis, so as to establish its performance prediction model and realize the performance prediction of the varistor under various use environments. However, in the actual test process, the existing methods have a strong dependence on the material system and process conditions, and it is difficult to adapt to the performance differences of varistors under different formulations or manufacturing processes. For example, due to the microstructure differences of the varistor itself or the errors caused by the preparation process, the accuracy and stability of the prediction results are often affected. Therefore, relying on the existing methods for performance prediction has significant limitations in terms of intelligence and generality, and it is difficult to meet the need for accurate prediction of the performance of varistors under complex systems.

[0005] Therefore, how to improve the accuracy of predicting the key performance of varistors has become an urgent problem to be solved. Summary of the Invention

[0006] In view of this, an embodiment of the present invention provides a method for predicting performance data of a semiconductor ceramic composite material to solve the problem of improving the accuracy of predicting the key performance of a varistor.

[0007] An embodiment of the present invention provides a method for predicting performance data of a semiconductor ceramic composite material, the method comprising the following steps: According to the reference parameters, test parameters, and number of tests, perform any round of repetitive tests on the target varistor to obtain the performance characteristic coefficients after each test, and form a performance characteristic sequence with all the performance characteristic coefficients, where the number of tests is greater than or equal to 2; According to the data change situation in the performance characteristic sequence, obtain the stability coefficient of the target varistor during the any round of repetitive tests, and according to the performance characteristic sequence and the stability coefficient, obtain the performance performance index of the target varistor during the any round of repetitive tests, as well as the parameter adjustment coefficient and the new number of tests for the next round of repetitive tests on the target varistor, and adjust the test parameters according to the parameter adjustment coefficient to obtain new test parameters; If the new test parameters do not meet the preset conditions, then according to the reference parameters, the new test parameters, and the new number of tests, perform the next round of repetitive tests on the target varistor to obtain the performance performance index of the target varistor during the next round of repetitive tests, until the new test parameters meet the preset conditions, obtain the performance performance index of the target varistor during each round of repetitive tests, and predict the performance performance index of the target varistor under any test parameters according to all the performance performance indexes.

[0008] Preferably, the reference parameters include a reference voltage and a reference current, and the test parameters include a test voltage and a test current. Then, according to the reference parameters, test parameters, and number of tests, performing any round of repetitive tests on the target varistor to obtain the performance characteristic coefficients after each test includes: For any one test in the any round of repetitive tests, apply the reference voltage and the test voltage to the target varistor respectively to obtain the reference output current and the test output current flowing through the target varistor; Calculate the ratio between the reference output current and the reference current to obtain a reference resistance characteristic index, calculate the ratio between the test output current and the test current to obtain a test resistance characteristic index, and use the ratio between the test resistance characteristic index and the reference resistance characteristic index as the independent variable of a logarithmic function with base 10 to obtain a first function value; Use the ratio between the test voltage and the reference voltage as the independent variable of a logarithmic function with base 10 to obtain a second function value; Calculate the ratio between the first function value and the second function value to obtain the performance characteristic coefficient of the target varistor after any one test.

[0009] Preferably, obtaining the stability coefficient of the target varistor during any round of repeatability test according to the data change situation in the performance characteristic sequence includes: Obtain the first-order difference sequence of the performance characteristic sequence, calculate the standard deviation of the first-order difference sequence, take the standard deviation as the independent variable of the exponential function with the natural constant as the base, and obtain the data dispersion degree of the performance characteristic sequence; Denote the first data in the performance characteristic sequence as the initial data, calculate the absolute value of the difference between each data other than the initial data in the performance characteristic sequence and the initial data respectively, and take the mean value of all absolute values of differences as the data change degree of the performance characteristic sequence; Calculate the product between the data dispersion degree and the data change degree to obtain the overall fluctuation degree of the performance characteristic sequence, calculate the ratio between the overall fluctuation degree and the number of all elements in the performance characteristic sequence, calculate the reciprocal of the sum of the preset constant and the ratio, and obtain the stability coefficient of the target varistor during any round of repeatability test.

[0010] Preferably, obtaining the performance performance index of the target varistor during any round of repeatability test according to the performance characteristic sequence and the stability coefficient includes: Calculate the product between the average value of all data in the performance characteristic sequence and the stability coefficient to obtain the performance performance index of the target varistor during any round of repeatability test.

[0011] Preferably, the method for obtaining the parameter regulation coefficient during the next round of repeatability test of the target varistor includes: Obtain the maximum withstand voltage value of the target varistor, calculate the proportion of the test voltage in the test parameters in the maximum withstand voltage value, and take the difference between the constant 1 and the proportion as the first characteristic value; Take the stability coefficient as the independent variable of the exponential function with the natural constant as the base to obtain the second characteristic value; Calculate the average value of all data in the performance characteristic sequence to obtain the average characteristic coefficient. If any round of repeatability test is the first round of repeatability test, calculate the reciprocal of the average characteristic coefficient to obtain the third characteristic value; If any round of the repetitive tests is not the first round of repetitive tests, calculate the absolute value of the difference between the average characteristic coefficient corresponding to any round of the repetitive tests and the average characteristic coefficient corresponding to the previous round of the repetitive tests to obtain the characteristic fluctuation amplitude, and calculate the reciprocal of the sum of the preset constant and the characteristic fluctuation amplitude to obtain the third characteristic value; Perform linear normalization on the product result among the first characteristic value, the second characteristic value, and the third characteristic value to obtain the voltage parameter regulation coefficient for the next round of repetitive tests on the target varistor; Based on the maximum allowable surge current value of the target varistor and the test current in the test parameters, obtain the current parameter regulation coefficient for the next round of repetitive tests on the target varistor.

[0012] Preferably, the method for obtaining the new number of tests includes: Calculate the sum of the constant 1 and the stability coefficient to obtain the number regulation ratio, and calculate the product of the number of tests and the number regulation ratio to obtain the new number of tests.

[0013] Preferably, the new test parameters include a new test current and a new test voltage, and the test parameters include a test voltage and a test current. Then, adjusting the test parameters according to the parameter regulation coefficient to obtain new test parameters includes: Calculate the sum of the constant 1 and the voltage parameter regulation coefficient to obtain the voltage regulation ratio, and calculate the product of the test voltage in the test parameters and the voltage regulation ratio to obtain the new test voltage; Calculate the sum of the constant 1 and the current parameter regulation coefficient to obtain the current regulation ratio, and calculate the product of the test current in the test parameters and the current regulation ratio to obtain the new test current.

[0014] Preferably, predicting the performance index of the target varistor under any test parameters according to all performance index includes: Use the reference parameters and test parameters during each round of repetitive tests as the input set of the performance prediction model, and use the performance index during each round of repetitive tests as the output set of the performance prediction model. Train the performance prediction model according to the input set and output set to obtain a trained performance prediction model, and use the trained performance prediction model to predict the performance index of the target varistor under any test parameters.

[0015] Preferably, the preset conditions include: The new test voltage in the new test parameters is greater than the maximum withstand voltage value of the target varistor, or the new test current in the new test parameters is greater than the maximum allowable surge current value of the target varistor.

[0016] Preferably, after the new test parameters meet the preset conditions, the following steps are further included: Based on the reference parameters, the new number of test times, as well as the maximum withstand voltage value and the maximum allowable surge current value of the target varistor, conduct the last round of repetitive tests on the target varistor, and predict the performance index of the target varistor under any test parameters according to the performance index during each round of repetitive tests.

[0017] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows: According to the reference parameters, test parameters and number of test times, the present invention conducts any round of repetitive tests on the target varistor to obtain the performance characteristic coefficient after each test, and forms a performance characteristic sequence with all the performance characteristic coefficients, where the number of test times is greater than or equal to 2; according to the data change situation in the performance characteristic sequence, obtain the stability coefficient of the target varistor during the any round of repetitive tests, and according to the performance characteristic sequence and the stability coefficient, obtain the performance index of the target varistor during the any round of repetitive tests, as well as the parameter adjustment coefficient and the new number of test times for the next round of repetitive tests on the target varistor, and adjust the test parameters according to the parameter adjustment coefficient to obtain new test parameters; if the new test parameters do not meet the preset conditions, then according to the reference parameters, the new test parameters and the new number of test times, conduct the next round of repetitive tests on the target varistor to obtain the performance index of the target varistor during the next round of repetitive tests until the new test parameters meet the preset conditions, obtain the performance index of the target varistor during each round of repetitive tests, and predict the performance index of the target varistor under any test parameters according to all the performance indexes. Among them, the non-linear parameters of the target varistor are improved, and the performance characteristic coefficient of the target varistor between any two voltages and currents (reference parameters and test parameters) is constructed. By the stability (stability coefficient) of the target varistor during each round of repetitive tests and the performance characteristic coefficient of each test, the performance index of the target varistor is obtained. At the same time, according to the change relationship between the performance characteristic coefficient and the test parameters, the test parameters for the next round of repetitive tests on the target varistor are adaptively adjusted, improving the test efficiency and the accuracy of the performance prediction of the target varistor. Description of the Drawings

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 It is a flowchart of a method for predicting the performance data of a semiconductor ceramic composite material provided in the first embodiment of the present invention. Detailed implementation manners

[0020] The embodiments of the present disclosure will be described in detail below. The examples of the embodiments are shown in the accompanying drawings. The embodiments described below by referring to the accompanying drawings are exemplary and are intended to explain the present disclosure, but should not be construed as a limitation to the present disclosure.

[0021] It should be noted that the terms "first", "second", etc. in the specification of the present disclosure and the above accompanying drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data used in appropriate cases can be interchanged so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present disclosure. On the contrary, they are only examples of devices and methods consistent with some aspects of the present disclosure.

[0022] In order to illustrate the technical solution of the present invention, specific embodiments will be used for illustration below.

[0023] See Figure 1 , which is a flowchart of a method for predicting the performance data of a semiconductor ceramic composite material provided in the first embodiment of the present invention. As Figure 1 shown, the method may include: Step S101, perform any round of repetitive tests on the target varistor according to the reference parameters, test parameters, and number of tests, obtain the performance characteristic coefficients after each test, and form a performance characteristic sequence with all the performance characteristic coefficients. The number of tests is greater than or equal to 2.

[0024] The nonlinear coefficient of a varistor refers to a parameter calculated according to the volt-ampere characteristic curve within a specific current and voltage range, that is , where represents the current value flowing through the varistor when the voltage value applied to the varistor is , represents the current value flowing through the varistor when the voltage value applied to the varistor is , The larger

[0025] is, the more sensitive the varistor is to voltage changes, that is, a small voltage change will cause a sharp change in current, and its performance is better. and voltage Perform actual measurement on the varistor to obtain the corresponding non-linear coefficient, thereby acquiring the sample data for modeling analysis, and using it to establish a performance prediction model to achieve performance prediction of the varistor under various usage environments.

[0026] However, during the actual testing process, the existing methods are highly dependent on the material system and process conditions, and it is difficult to adapt to the performance differences of varistors under different materials or manufacturing processes. For example, due to the microscopic structure differences of the varistor itself or the errors caused by the manufacturing process, the accuracy and stability of the prediction results are often affected. Therefore, relying on the existing methods for performance prediction has significant limitations in terms of intelligence and generality, and it is difficult to meet the requirement of accurately predicting the performance of varistors under complex systems. At the same time, in practical applications, the varistor may be subjected to long-term surge impacts. After long-term surge impacts, the interface aging inside the material of the varistor and the changes in the microstructure may cause its performance to change, thereby affecting the performance stability and service life of the varistor.

[0027] Therefore, in the embodiments of the present invention, any varistor with a performance to be predicted is denoted as the target varistor, the test conditions are adjusted according to the characteristics of the target varistor, and then the target varistor is repeatedly tested according to each test condition, so as to establish a performance prediction model based on the non-linear coefficient after each test under all test conditions, and improve the accuracy of performance prediction of the varistor under various usage environments.

[0028] First, obtain the maximum withstand voltage value and the maximum allowable surge current value of the target varistor, and set the test conditions for the first round of repeated testing of the target varistor: respectively use 0.005 times the maximum withstand voltage value and the maximum allowable surge current value as the reference parameters; respectively use 0.01 times the maximum withstand voltage value and the maximum allowable surge current value as the test parameters for the first round of repeated testing of the target varistor; the duration of each test during the first round of repeated testing is , the number of tests is 100 times, which is not limited here, and the implementer can set it according to the specific scenario, where the number of tests must be greater than or equal to 2. At the same time, denote 0.005 times the maximum withstand voltage value as the reference voltage, 0.005 times the maximum allowable surge current value as the reference current, 0.01 times the maximum withstand voltage value as the test voltage, and 0.01 times the maximum allowable surge current value as the test current.

[0029] Then, a first-round repeatability test is performed on the target varistor according to the reference parameters, test parameters, and number of tests. Since in actual work, the varistor may be in a scenario where the voltage and current change in real time, in the embodiments of the present invention, according to the function and characteristics of the non-linear coefficient, the calculation formula of the non-linear coefficient is optimized to obtain the performance characteristic coefficient of the target varistor after each test, which is used to characterize the performance of the target varistor after each test. The specific method for obtaining the performance characteristic coefficient of the target varistor after each test is as follows: For any one test in the first-round repeatability test, the reference voltage and the test voltage are respectively applied to the target varistor, and the reference output current and the test output current flowing through the target varistor are obtained through a current sensor; Calculate the ratio between the reference output current and the reference current to obtain the reference resistance characteristic index, calculate the ratio between the test output current and the test current to obtain the test resistance characteristic index, and use the ratio between the test resistance characteristic index and the reference resistance characteristic index as the independent variable of a logarithmic function with base 10 to obtain the first function value; Use the ratio between the test voltage and the reference voltage as the independent variable of a logarithmic function with base 10 to obtain the second function value; Calculate the ratio between the first function value and the second function value to obtain the performance characteristic coefficient of the target varistor after the arbitrary one test.

[0030] In one embodiment, taking the i-th test as an example, the calculation formula for the performance characteristic coefficient of the target varistor after the i-th test is:

[0031] Among them, represents the performance characteristic coefficient of the target varistor after the i-th test in the j-th round of repeatability test (at this time j = 1, that is, the first-round repeatability test), represents the test output current at the i-th test in the j-th round of repeatability test (at this time j = 1, that is, the first-round repeatability test), represents the test current (0.01 times the maximum allowable surge current value) in the j-th round of repeatability test (at this time j = 1, that is, the first-round repeatability test), represents the reference output current at the i-th test, represents the reference current (0.005 times the maximum allowable surge current value), represents the test voltage (0.01 times the maximum withstand voltage value) in the first-round repeatability test (at this time j = 1, that is, the first-round repeatability test), represents the reference voltage (0.005 times the maximum withstand voltage value). represents the logarithmic function with base 10.

[0032] It should be noted that the larger it is, the more sensitive the varistor is to voltage changes, that is, a small voltage change will cause a sharp change in current, and the better the performance of the target varistor in the i-th test.

[0033] Similarly, obtain the performance characteristic coefficients after each test of the target varistor in the first round of repeatability tests, and form a performance characteristic sequence with all the performance characteristic coefficients. Analyze the performance characteristic sequence to obtain the test conditions for the target varistor when performing the second round of repeatability tests.

[0034] Step S102, according to the data change situation in the performance characteristic sequence, obtain the stability coefficient of the target varistor in any round of repeatability tests. According to the performance characteristic sequence and the stability coefficient, obtain the performance performance index of the target varistor in any round of repeatability tests, as well as the parameter adjustment coefficient and new test times for the target varistor when performing the next round of repeatability tests. Adjust the test parameters according to the parameter adjustment coefficient to obtain new test parameters.

[0035] If the data in the performance characteristic sequence fluctuates more, it means that in the first round of repeatability tests, the stability of the performance of the target varistor is worse, its life may decay faster, and thus the performance of the target varistor is worse. Therefore, in the embodiments of the present invention, according to the data change situation in the performance characteristic sequence corresponding to the target varistor in the first round of repeatability tests, obtain the stability coefficient of the target varistor in the first round of repeatability tests, which is used to characterize the stability of the target varistor in the first round of repeatability tests. Specifically: Obtain the first-order difference sequence of the performance characteristic sequence, calculate the standard deviation of the first-order difference sequence, take the standard deviation as the independent variable of the exponential function with the natural constant as the base, and obtain the data dispersion degree of the performance characteristic sequence; Record the first data in the performance characteristic sequence as the initial data, and calculate the absolute value of the difference between each other data in the performance characteristic sequence except the initial data and the initial data respectively. Take the mean value of all the absolute values of the differences as the data change degree of the performance characteristic sequence; Calculate the product between the data dispersion degree and the data change degree to obtain the overall fluctuation degree of the performance characteristic sequence. Calculate the ratio between the overall fluctuation degree and the number of all elements in the performance characteristic sequence, and calculate the reciprocal of the sum of the preset constant and the ratio to obtain the stability coefficient of the target varistor in the first round of repeatability tests.

[0036] In one embodiment, the calculation formula for the stability coefficient of the target varistor in the first round of repeatability tests is as follows:

[0037] Wherein, represents the stability coefficient of the target varistor in the j-th round of repeatability tests (at this time, j = 1, that is, the first round of repeatability tests), represents the standard deviation of the first-order difference sequence of the performance characteristic sequence of the target varistor in the j-th round of repeatability tests (at this time, j = 1, that is, the first round of repeatability tests), represents the number of all elements in the performance characteristic sequence of the target varistor in the j-th round of repeatability tests (at this time, j = 1, that is, the first round of repeatability tests), that is, the number of tests of the target varistor in the j-th round of repeatability tests, represents the first performance characteristic coefficient (initial data) in the performance characteristic sequence, represents the i-th performance characteristic coefficient in the performance characteristic sequence except the initial data, that is, the performance characteristic coefficient at the i-th test, C represents a preset constant, represents the absolute value symbol, and e represents the natural constant.

[0038] It should be noted that, The larger the , the more discrete the data distribution in the first-order difference sequence of the performance characteristic sequence, the greater the data fluctuation in the performance characteristic sequence, and thus The larger the , the smaller the stability of the target varistor in the j-th round of repeatability tests; The larger the

[0039] , it indicates that as the number of tests increases, the change in the performance characteristic coefficient of the target varistor is greater, and thus The smaller the

[0040] , the smaller the stability of the target varistor in the j-th round of repeatability tests. C is used to prevent the denominator from being 0. In the embodiments of the present invention, C = 1 is set, and there is no limitation here. Implementers can set it according to specific scenarios.

[0041] Among them, represents the performance index of the target varistor in the j-th round of repeatability test (at this time j = 1, that is, the first round of repeatability test); represents the stability coefficient of the target varistor in the j-th round of repeatability test (at this time j = 1, that is, the first round of repeatability test); represents the average value of all data in the performance characteristic sequence of the target varistor in the j-th round of repeatability test (at this time j = 1, that is, the first round of repeatability test).

[0042] It should be noted that the larger it is, the better the stability of the target varistor, and thus the larger it is, the better the performance of the target varistor under the test conditions of the j-th round of repeatability test; the larger it is, the more sensitive the target varistor is to the change of the test voltage in the j-th round of repeatability test, and thus the larger it is, the better the performance of the target varistor under the test conditions of the j-th round of repeatability test.

[0043] So far, the first round of repeatability test of the target varistor is completed. Further, the second round of repeatability test is carried out on the target varistor.

[0044] In the embodiment of the present invention, by analyzing the performance characteristic sequence and the stability coefficient of the target varistor in the first round of repeatability test, the parameter adjustment coefficient of the target varistor in the second round of repeatability test is obtained, so as to adjust the test conditions of the target varistor in the second round of repeatability test according to the self-characteristics of the target varistor: if the stability coefficient of the target varistor in the first round of repeatability test is large, it means that the performance of the target varistor is good. When performing the second round of repeatability test, the number of tests and the step sizes of the increased current and voltage should be appropriately increased, so as to quickly obtain the performance change of the target varistor; on the contrary, the step sizes of the increased current and voltage applied need to be appropriately reduced, so as to obtain more subtle performance changes of the target varistor and improve the accuracy of the performance prediction of the target varistor. Specifically: Obtain the maximum withstand voltage value of the target varistor, calculate the ratio of the test voltage in the test parameters to the maximum withstand voltage value, and take the difference between the constant 1 and the ratio as the first characteristic value; Take the stability coefficient as the independent variable of the exponential function with the natural constant as the base to obtain the second characteristic value; Calculate the average value of all data in the performance characteristic sequence to obtain the average characteristic coefficient, and calculate the reciprocal of the average characteristic coefficient to obtain the third characteristic value; Perform linear normalization on the product result among the first eigenvalue, the second eigenvalue, and the third eigenvalue to obtain the voltage parameter regulation coefficient for the second-round repeatability test of the target varistor. Here, linear normalization is a prior art and will not be elaborated herein.

[0045] In one embodiment, the calculation formula for the voltage parameter regulation coefficient for the second-round repeatability test of the target varistor is:

[0046] where, represents the voltage parameter regulation coefficient for the (j + 1)-th round repeatability test of the target varistor (at this time, j = 1, that is, the second-round repeatability test), represents the test voltage for the j-th round repeatability test (at this time, j = 1, that is, the first-round repeatability test), represents the maximum withstand voltage value of the target varistor, represents the average value of all data in the performance characteristic sequence for the j-th round repeatability test (at this time, j = 1, that is, the first-round repeatability test), represents the average value of all data in the performance characteristic sequence for the (j - 1)-th round repeatability test, represents the stability coefficient for the j-th round repeatability test of the target varistor (at this time, j = 1, that is, the first-round repeatability test), represents the linear normalization function, and C represents a preset constant.

[0047] It should be noted that, is the first eigenvalue, is the second eigenvalue, is the third eigenvalue. Since j = 1 at this time, that is, the first-round repeatability test, it is the first-round repeatability test. Therefore, the third eigenvalue at this time is , that is, when j = 1, . The smaller the , the larger the adjustable space of the test voltage in the (j + 1)-th round repeatability test. Furthermore, the larger the , and when the larger the , the larger the step size of the increase in the voltage applied to the target varistor in the (j + 1)-th round repeatability test. Furthermore, the larger the ; The larger the , the smaller the performance difference between the target varistor in this round (the j-th round) and the previous round repeatability test under the test conditions. The more stable the target varistor and the better its performance, the larger the step size of the increase in the voltage applied to the target varistor in the (j + 1)-th round repeatability test to quickly obtain the performance change of the target varistor. Furthermore, The larger it is; C is used to prevent the denominator from being 0. In the present invention, C is set to 1, which is not limited here, and the implementer can set it according to the specific scenario.

[0048] Similarly, according to the maximum allowable surge current value of the target varistor and the test current in the test parameters, the current parameter regulation coefficient for the second-round repetitive test of the target varistor is obtained. That is, the calculation formula for the current parameter regulation coefficient for the second-round repetitive test of the target varistor is: , where represents the test current of the j-th round of repetitive test (the previous round of repetitive test of the (j + 1)-th round of repetitive test, where j = 1 at this time, that is, the first round of repetitive test), represents the maximum allowable surge current value of the target varistor.

[0049] Furthermore, according to the voltage parameter regulation coefficient and the current parameter regulation coefficient for the second-round repetitive test of the target varistor, the new test parameters for the second-round repetitive test of the target varistor are obtained. The new test parameters include the new test voltage and the new test current. Among them, the specific method for obtaining the new test voltage is: Calculate the sum between the constant 1 and the voltage parameter regulation coefficient to obtain the voltage regulation ratio, and calculate the product between the test voltage in the test parameters and the voltage regulation ratio to obtain the new test voltage.

[0050] In an embodiment, the calculation formula for the new test voltage for the second-round repetitive test of the target varistor is:

[0051] where represents the new test voltage for the (j + 1)-th round of repetitive test of the target varistor (where j = 1 at this time, that is, the second round of repetitive test), represents the test voltage of the j-th round of repetitive test (where j = 1 at this time, that is, the first round of repetitive test), represents the voltage parameter regulation coefficient for the (j + 1)-th round of repetitive test of the target varistor (where j = 1 at this time, that is, the second round of repetitive test).

[0052] It should be noted that the larger the voltage parameter regulation coefficient, the larger the new test voltage for the second-round repetitive test of the target varistor.

[0053] Among them, the specific method for obtaining the new test current is: Calculate the sum between the constant 1 and the current parameter regulation coefficient to obtain the current regulation ratio, and calculate the product between the test current in the test parameters and the current regulation ratio to obtain the new test current.

[0054] In one embodiment, the calculation formula for the new test current during the second-round repeatability test of the target varistor is:

[0055] Wherein, represents the test current during the (j + 1)-th round repeatability test of the target varistor (at this time, j = 1, that is, the second-round repeatability test); represents the test current during the j-th round repeatability test (at this time, j = 1, that is, the first-round repeatability test); represents the current parameter regulation coefficient during the (j + 1)-th round repeatability test of the target varistor (at this time, j = 1, that is, the second-round repeatability test).

[0056] It should be noted that the larger the current parameter regulation coefficient, the larger the new test current during the second-round repeatability test of the target varistor.

[0057] Furthermore, according to the stability coefficient during the previous-round repeatability test (the first-round repeatability test) of the target varistor in the second-round repeatability test, the number of test times during the second-round repeatability test of the target varistor is obtained, denoted as the new test times. Specifically: Calculate the sum between the constant 1 and the stability coefficient to obtain the number regulation ratio, and calculate the product of the number of test times and the number regulation ratio and round up to obtain the new test times.

[0058] In one embodiment, the calculation formula for the new test times during the second-round repeatability test of the target varistor is:

[0059] Wherein, represents the new test times during the (j + 1)-th round repeatability test of the target varistor (at this time, j = 1, that is, the second-round repeatability test); represents the number of test times during the j-th round repeatability test of the target varistor (at this time, j = 1, that is, the first-round repeatability test); represents the stability coefficient during the j-th round repeatability test of the target varistor (at this time, j = 1, that is, the first-round repeatability test), 1 represents a constant, represents the rounding-up symbol.

[0060] It should be noted that the larger the stability coefficient during the j-th round repeatability test, the better the stability of the target varistor, and the more test times during the (j + 1)-th round repeatability test, which is convenient for more sensitively capturing the attenuation of performance such as the lifespan of the target varistor.

[0061] So far, the acquisition of new test parameters and new test times for the second round of repeatability testing of the target varistor has been completed.

[0062] Step S103: If the new test parameters do not meet the preset conditions, then according to the reference parameters, the new test parameters, and the new test times, conduct the next round of repeatability testing on the target varistor to obtain the performance index of the target varistor in the next round of repeatability testing until the new test parameters meet the preset conditions, obtain the performance index of the target varistor in each round of repeatability testing, and predict the performance index of the target varistor under any test parameters based on all performance indexes.

[0063] To prevent the new test voltage in the second round of repeatability testing from being greater than the maximum withstand voltage of the target varistor, or the new test current from being greater than the maximum allowable surge current value of the target varistor, which may cause damage to the target varistor. Therefore, after obtaining the new test voltage and new test current in the second round of repeatability testing through step S102, it is necessary to make a judgment first: if the new test voltage is less than or equal to the maximum withstand voltage of the target varistor, and the new test current is less than or equal to the maximum allowable surge current value of the target varistor, then according to the reference parameters (0.005 times the maximum withstand voltage and the maximum allowable surge current value), the new test parameters (new test voltage and new test current), and the new test times, conduct the second round of repeatability testing on the target varistor to obtain the performance index of the target varistor in the second round of repeatability testing, that is , and the new test parameters and new test times for the next round of repeatability testing, where the duration of each test is still , which is the same as the duration of each test in the first round of repeatability testing.

[0064] Similarly, obtain the test parameters of the target varistor in each round of repeatability testing until the calculated new test voltage in the next round of repeatability testing is greater than the maximum withstand voltage of the target varistor, or the new test current is greater than the maximum allowable surge current value of the target varistor. At this time, according to the maximum withstand voltage, the maximum allowable surge current value, the reference parameters, and the new test times in the next round of repeatability testing, conduct the last round of repeatability testing on the target varistor to obtain the performance index of the target varistor in the last round of repeatability testing.

[0065] So far, the performance indexes of the target varistor in each round of repeatability testing have been obtained.

[0066] In an embodiment of the present invention, a performance prediction model for predicting the performance of a varistor is obtained by training an SVR model. Therefore, after obtaining all performance index, the reference parameters and test parameters during each round of repetitive testing are used as the input set of the SVR model, and the performance index during each round of repetitive testing is used as the output set of the SVR model. The SVR model is trained according to the input set and the output set to obtain the performance prediction model. According to the performance prediction model, the performance index of all varistors with the same material system and process conditions as the target varistor under any test parameters is predicted. Among them, the training of the SVR model is a prior art and will not be elaborated here.

[0067] In summary, according to the reference parameters, test parameters and number of tests, any round of repetitive testing is performed on the target varistor to obtain the performance characteristic coefficient after each test, and all performance characteristic coefficients are combined into a performance characteristic sequence, where the number of tests is greater than or equal to 2; according to the data change situation in the performance characteristic sequence, the stability coefficient of the target varistor during any round of repetitive testing is obtained, and according to the performance characteristic sequence and the stability coefficient, the performance index of the target varistor during any round of repetitive testing, as well as the parameter adjustment coefficient and the new number of tests for the next round of repetitive testing of the target varistor are obtained. The test parameters are adjusted according to the parameter adjustment coefficient to obtain new test parameters; if the new test parameters do not meet the preset conditions, then according to the reference parameters, the new test parameters and the new number of tests, the next round of repetitive testing is performed on the target varistor to obtain the performance index of the target varistor during the next round of repetitive testing, until the new test parameters meet the preset conditions, the performance index of the target varistor during each round of repetitive testing is obtained, and according to all performance indexes, the performance index of the target varistor under any test parameters is predicted. Among them, the non-linear parameters of the target varistor are improved, and the performance characteristic coefficient of the target varistor between any two voltages and currents (reference parameters and test parameters) is constructed. By the stability (stability coefficient) of the target varistor during each round of repetitive testing and the performance characteristic coefficient of each test, the performance index of the target varistor is obtained. At the same time, according to the change relationship between the performance characteristic coefficient and the test parameters, the test parameters of the target varistor during the next round of repetitive testing are adaptively adjusted to improve the test efficiency and the accuracy of the performance prediction of the target varistor.

[0068] 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 foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for predicting the performance data of a semiconductor ceramic composite material, characterized in that, The performance data prediction method for the semiconductor ceramic composite material includes: Performing any round of repetitive tests on the target varistor according to the reference parameters, test parameters, and number of tests, obtaining the performance characteristic coefficients after each test, and forming a performance characteristic sequence with all the performance characteristic coefficients, where the number of tests is greater than or equal to 2; According to the data change situation in the performance characteristic sequence, obtaining the stability coefficient of the target varistor during the any round of repetitive tests, obtaining the performance performance index of the target varistor during the any round of repetitive tests, as well as the parameter adjustment coefficient and new number of tests for the next round of repetitive tests on the target varistor according to the performance characteristic sequence and the stability coefficient, and adjusting the test parameters according to the parameter adjustment coefficient to obtain new test parameters; If the new test parameters do not meet the preset conditions, then perform the next round of repetitive tests on the target varistor according to the reference parameters, the new test parameters, and the new number of tests, obtaining the performance performance index of the target varistor during the next round of repetitive tests, until the new test parameters meet the preset conditions, obtaining the performance performance index of the target varistor during each round of repetitive tests, and predicting the performance performance index of the target varistor under any test parameters according to all the performance performance indexes.

2. The performance data prediction method of a semiconductor ceramic composite material according to claim 1, characterized in that The reference parameters include the reference voltage and the reference current, and the test parameters include the test voltage and the test current. Then, performing any round of repetitive tests on the target varistor according to the reference parameters, test parameters, and number of tests, and obtaining the performance characteristic coefficients after each test includes: For any one test in the any round of repetitive tests, applying the reference voltage and the test voltage to the target varistor respectively, and obtaining the reference output current and the test output current flowing through the target varistor; Calculating the ratio between the reference output current and the reference current to obtain the reference resistance characteristic index, calculating the ratio between the test output current and the test current to obtain the test resistance characteristic index, and taking the ratio between the test resistance characteristic index and the reference resistance characteristic index as the independent variable of the logarithmic function with base 10 to obtain the first function value; Taking the ratio between the test voltage and the reference voltage as the independent variable of the logarithmic function with base 10 to obtain the second function value; Calculating the ratio between the first function value and the second function value to obtain the performance characteristic coefficient of the target varistor after the any one test.

3. The performance data prediction method of a semiconductor ceramic composite material according to claim 1, characterized in that The obtaining the stability coefficient of the target varistor during the any round of repetitive tests according to the data change situation in the performance characteristic sequence includes: Obtaining the first-order difference sequence of the performance characteristic sequence, calculating the standard deviation of the first-order difference sequence, and taking the standard deviation as the independent variable of the exponential function with the natural constant as the base to obtain the data dispersion degree of the performance characteristic sequence; Denote the first data in the performance characteristic sequence as the initial data, and calculate the absolute value of the difference between each data in the performance characteristic sequence except the initial data and the initial data. Take the mean of all the absolute values of the differences as the degree of data change of the performance characteristic sequence; Calculate the product of the degree of data dispersion and the degree of data change to obtain the overall fluctuation degree of the performance characteristic sequence. Calculate the ratio of the overall fluctuation degree to the number of all elements in the performance characteristic sequence, and calculate the reciprocal of the sum of the preset constant and the ratio to obtain the stability coefficient of the target varistor during any round of repeatability test.

4. A method for predicting the performance data of a semiconductor ceramic composite material according to claim 1, characterized in that Based on the performance characteristic sequence and the stability coefficient, obtain the performance performance index of the target varistor during any round of repeatability test, including: Calculate the product of the mean of all data in the performance characteristic sequence and the stability coefficient to obtain the performance performance index of the target varistor during any round of repeatability test.

5. The performance data prediction method of a semiconductor ceramic composite material according to claim 2, wherein The method for obtaining the parameter regulation coefficient during the next round of repeatability test of the target varistor includes: Obtain the maximum withstand voltage value of the target varistor, calculate the proportion of the test voltage in the test parameters in the maximum withstand voltage value, and take the difference between the constant 1 and the proportion as the first characteristic value; Take the stability coefficient as the independent variable of the exponential function with the natural constant as the base to obtain the second characteristic value; Calculate the mean of all data in the performance characteristic sequence to obtain the average characteristic coefficient. If any round of repeatability test is the first round of repeatability test, calculate the reciprocal of the average characteristic coefficient to obtain the third characteristic value; If any round of repeatability test is not the first round of repeatability test, calculate the absolute value of the difference between the average characteristic coefficient corresponding to any round of repeatability test and the average characteristic coefficient corresponding to the previous round of repeatability test to obtain the characteristic fluctuation amplitude, and calculate the reciprocal of the sum of the preset constant and the characteristic fluctuation amplitude to obtain the third characteristic value; Perform linear normalization on the product result of the first characteristic value, the second characteristic value and the third characteristic value to obtain the voltage parameter regulation coefficient for the next round of repeatability test of the target varistor; Based on the maximum allowable surge current value of the target varistor and the test current in the test parameters, obtain the current parameter regulation coefficient for the next round of repeatability test of the target varistor.

6. The performance data prediction method of a semiconductor ceramic composite material according to claim 1, characterized in that The method for obtaining the new number of tests includes: Calculate the sum of the constant 1 and the stability coefficient to obtain the number regulation ratio, and round up the product of the number of tests and the number regulation ratio to obtain the new number of tests.

7. A method for predicting performance data of a semiconductor ceramic composite material according to claim 5, characterized in that The new test parameters include a new test current and a new test voltage, and the test parameters include a test voltage and a test current. Then, adjusting the test parameters according to the parameter regulation coefficient to obtain the new test parameters includes: Calculate the sum of the constant 1 and the voltage parameter regulation coefficient to obtain the voltage regulation ratio, and calculate the product of the test voltage in the test parameters and the voltage regulation ratio to obtain the new test voltage; Calculate the sum between the calculation constant 1 and the current parameter regulation coefficient to obtain the current regulation ratio, and calculate the product between the test current in the test parameters and the current regulation ratio to obtain the new test current.

8. A method for predicting performance data of a semiconductor ceramic composite material according to claim 1, characterized in that Predict the performance index of the target varistor under any test parameters according to all performance indexes, including: Use the reference parameters and test parameters in each round of repeated tests as the input set of the performance prediction model, and use the performance indexes in each round of repeated tests as the output set of the performance prediction model. Train the performance prediction model according to the input set and output set to obtain a trained performance prediction model, and use the trained performance prediction model to predict the performance indexes of the target varistor under any test parameters.

9. The performance data prediction method of a semiconductor ceramic composite material according to claim 1, characterized in that The preset conditions include: The new test voltage in the new test parameters is greater than the maximum withstand voltage value of the target varistor, or the new test current in the new test parameters is greater than the maximum allowable surge current value of the target varistor.

10. A method for predicting the performance data of a semiconductor ceramic composite material according to claim 1, characterized in that, After the new test parameters meet the preset conditions, it further includes: Conduct the last round of repeated tests on the target varistor according to the reference parameters, the new number of test times, and the maximum withstand voltage value and the maximum allowable surge current value of the target varistor, and predict the performance indexes of the target varistor under any test parameters according to the performance indexes in each round of repeated tests.

Images