Electronic component noise test system and test method based on data model

Abstract

The invention relates to the technical field of electronic component testing, and discloses an electronic component noise testing system and method based on a data model. The method comprises the following steps: in the working process of the electronic component, acquiring current working environment parameter data of the electronic component, and inputting the current working environment parameter data into a pre-trained noise feature extraction model to obtain noise feature data; performing noise comparison operation according to the noise feature data to obtain model adjustment parameters of the noise prediction model; performing difference degree adjustment operation according to the model adjustment parameters to obtain an adjusted noise prediction model; and performing noise prediction operation according to the adjusted noise prediction model to obtain predicted noise characteristics of the electronic component. According to the invention, the accuracy of noise testing can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic component testing, and in particular to an electronic component noise testing system and a testing method based on a data model. Background Art

[0002] At present, with the rapid development of science and technology, there is a large demand for electronic components, but electronic components will have noise during operation, and the noise changes with the temperature, voltage, current and other environmental conditions of the device, such as photodetectors, transistors, noise amplifiers, etc. The noise of electronic components during operation will interfere with the normal operation of the communication system, especially some precision measurement systems, which have very high requirements on the noise level of electronic components during operation.

[0003] Among the existing noise testing methods, one method can perform noise testing on various types of components, which helps to optimize component design to obtain low-noise, high-performance components. However, it does not consider the impact of environmental conditions such as device temperature, voltage and operating current on the noise level, nor does it consider establishing a noise prediction model for electronic components under working environmental conditions. It is impossible to accurately obtain the noise level of electronic components under certain working environmental conditions during testing. Although the other method provides a test system for a high-power device model, it does not test the noise and cannot provide a reference for noise level prediction.

[0004] Although the existing technology can obtain the noise of electronic components, there is no good solution to accurately test the noise level of electronic components under certain working environment conditions. Summary of the invention

[0005] The present invention provides an electronic component noise testing system and a testing method based on a data model, so as to improve the accuracy of the noise testing.

[0006] In a first aspect, in order to solve the above-mentioned technical problems, the present invention provides an electronic component noise testing method based on a data model, comprising: during the operation of the electronic component, obtaining the current working environment parameter data of the electronic component, and inputting it into a pre-trained noise feature extraction model to obtain noise feature data; performing a noise comparison operation according to the noise feature data to obtain a model adjustment parameter of an initial noise prediction model; performing a difference degree adjustment operation according to the model adjustment parameter to obtain an adjusted noise prediction model; performing a noise prediction operation according to the adjusted noise prediction model to obtain the predicted noise characteristics of the electronic component; wherein the training process of the noise feature extraction model comprises: obtaining historical working environment parameter data, the historical working environment parameter data including temperature, voltage and current; performing data preprocessing and parameter range determination operations according to the historical working environment parameter data to obtain a parameter range; performing a parameter numerical interval conversion operation according to the parameter range to obtain a numerical interval corresponding to the parameter.

[0007] As an optional implementation manner, performing a noise comparison operation according to the noise feature data to obtain a model adjustment parameter of the noise prediction model includes: a calculation formula for the model adjustment parameter is: in, How good is the model's prediction? To predict the number of noise data, For the The original noise characteristic data, For the predicted Noise characteristic data, is the preset learning rate, Adjust the parameters for the preset initial model, Parameters are adjusted for the model.

[0008] As an optional implementation, the difference degree adjustment operation is performed according to the model adjustment parameters to obtain the adjusted noise prediction model, including: the difference degree adjustment operation formula is: in, is the function of the adjusted noise prediction model, is the preset coefficient, is the noise characteristic data, Adjust parameters for the model, is the error term.

[0009] As an optional implementation, the training process of the noise prediction model includes: the training process of the noise prediction model includes: obtaining historical noise feature data; performing a judgment set establishment operation according to the historical noise feature data to obtain an evaluation set; performing a matrix construction operation according to the evaluation set to obtain an evaluation matrix; performing an entropy value calculation operation according to the evaluation matrix to obtain an entropy value; performing a goodness calculation operation according to the entropy value to obtain a goodness index; performing a model construction operation according to the evaluation matrix, the entropy value and the goodness index to obtain a noise prediction model; wherein the evaluation matrix is: in, is the evaluation matrix, No. Line The rating matrix value of the column.

[0010] As an optional implementation, performing an entropy value calculation operation according to the evaluation matrix to obtain an entropy value includes: the entropy value calculation formula is: in, is the entropy value, For the Line The evaluation matrix value of the column, is the total number of matrix values ​​in the evaluation matrix.

[0011] As an optional implementation, performing a goodness calculation operation according to the entropy value to obtain a goodness index includes: the goodness calculation formula is: in, is the goodness index, is the entropy value.

[0012] As an optional implementation, the model building operation is performed according to the evaluation matrix, the entropy value and the goodness index to obtain a noise prediction model, including: the model building formula is: in, is the noise prediction model, is the entropy matrix under the subjective weight set, is the entropy matrix under the objective weight set, W1 is the subjective weight set, W2 is the objective weight set, is the entropy value, is the preset scale factor, is the goodness index.

[0013] As an optional implementation, performing a parameter value interval conversion operation according to the parameter range to obtain a value interval corresponding to the parameter includes: the parameter value interval conversion formula is: in, is the numerical interval, is the minimum value of the parameter range, is the maximum value of the parameter range, is the first data.

[0014] As an optional implementation, the support vector machine regression operation is performed according to the numerical interval to obtain a noise feature extraction model, including: the support vector machine regression formula is: in, is the noise feature extraction model, is the number of support vectors, and is the default Lagrange multiplier, is the kernel function, is the bias term, is the newly input sample data, is the first data.

[0015] In a second aspect, the present invention provides an electronic component noise testing system based on a data model, comprising: a data acquisition module, used to acquire current working environment parameter data of the electronic component during the working process of the electronic component; a model building module, used to establish a noise feature extraction model and a noise prediction model; a model optimization module, used to optimize the noise prediction model to obtain an adjusted noise prediction model; a noise prediction module, used to perform a noise prediction operation according to the adjusted noise prediction model to obtain the predicted noise characteristics of the electronic component; wherein the training process of the noise feature extraction model comprises: acquiring historical working environment parameter data, the historical working environment parameter data including temperature, voltage and current; performing data preprocessing and parameter range determination operations according to the historical working environment parameter data to obtain a parameter range; performing a parameter value interval conversion operation according to the parameter range to obtain a value interval corresponding to the parameter; performing a support vector machine regression operation according to the value interval to obtain a noise feature extraction model.

[0016] In a third aspect, the present invention further provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the electronic component noise testing method based on the data model as described above is implemented.

[0017] In a fourth aspect, the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the electronic component noise testing method based on the data model as described above.

[0018] Compared with the prior art, the present invention has the following beneficial effects: the present invention provides an electronic component noise testing method based on a data model, comprising: in the working process of the electronic component, obtaining the current working environment parameter data of the electronic component, and inputting it into a pre-trained noise feature extraction model to obtain noise feature data; performing a noise comparison operation according to the noise feature data to obtain the model adjustment parameters of the noise prediction model; performing a difference degree adjustment operation according to the model adjustment parameters to obtain the adjusted noise prediction model; performing a noise prediction operation according to the adjusted noise prediction model to obtain the predicted noise characteristics of the electronic component; wherein the training process of the noise feature extraction model comprises: obtaining historical working environment parameter data, the historical working environment parameter data including temperature, voltage and current; performing data preprocessing and parameter range determination operations according to the historical working environment parameter data to obtain a parameter range; performing a parameter value interval conversion operation according to the parameter range to obtain a value interval corresponding to the parameter; performing a support vector machine regression operation according to the value interval to obtain a noise feature extraction model.

[0019] In the present invention, the noise test method of electronic components based on the data model has the following advantages: First, by obtaining the noise characteristic data of the electronic components under different working environment parameters (such as temperature, current, voltage, etc.), and applying these data to the construction of the noise prediction model, the accuracy of the noise test can be effectively improved. This method not only reduces the common errors in traditional noise testing, but also greatly reduces the difficulty of testing, because the data model can provide accurate noise characteristic prediction under various environmental conditions, thereby avoiding a large number of cumbersome manual tests and experimental processes. Second, the method can realize real-time prediction of the working environment parameters of the electronic components to be tested, thereby providing a more accurate basis for environmental conditions for noise testing. This prediction capability can significantly improve the efficiency of noise testing and avoid the cumbersome steps of adjusting environmental parameters one by one in traditional testing. Third, by optimizing simulation parameters, the test process can also be made more efficient, unnecessary repeated tests can be reduced, and time and resources can be saved. According to the above method, the noise test method of electronic components is implemented, relying on the established noise feature extraction model and noise prediction model, and combining these two models to accurately predict the noise. Through the continuous optimization of the noise prediction model, the accuracy of the test is further improved, making the noise test results more reliable. At the same time, by predicting and adjusting the working environment parameters, not only can the accuracy of noise testing be improved, but the testing process can also be completed in a shorter time, thereby greatly improving the overall testing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of a method for testing noise of electronic components based on a data model provided by an embodiment of the present invention; Figure 2 is a training schematic diagram of a noise feature extraction model provided by an embodiment of the present invention; Figure 3 is a training schematic diagram of a noise prediction model provided by an embodiment of the present invention; Figure 4 It is a structural schematic diagram of an electronic component noise testing system based on a data model provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] It should be noted that, at present, with the rapid development of science and technology, there is a large demand for electronic components, but electronic components will have noise during operation, and the noise changes with the changes in environmental conditions such as temperature, voltage, current, etc. of the device, such as photodetectors, triodes, noise amplifiers, etc. The noise of electronic components during operation will interfere with the normal operation of the communication system, especially some precision measurement systems, which have very high requirements on the noise level of electronic components during operation. Among the existing noise test methods, one of them can perform noise tests on various types of components, which helps to optimize component design to obtain low-noise, high-performance components, but does not consider the impact of environmental conditions such as device temperature, voltage and working current on the noise level, nor does it consider establishing a noise prediction model for electronic components under working environmental conditions, and cannot accurately obtain the noise level of electronic components under certain working environmental conditions during testing; the other provides a test system for a high-power device model, but does not test the noise, and cannot provide a reference for noise level prediction. Although the existing technology can achieve the acquisition of the noise of electronic components, there is no good solution to accurately test the noise level of electronic components under certain working environmental conditions.

[0023] For the above problems, refer to Figure 1 The first embodiment of the present invention provides a method for testing noise of electronic components based on a data model, comprising the following steps: S1, during the working process of the electronic component, obtaining the current working environment parameter data of the electronic component, and inputting it into a pre-trained noise feature extraction model to obtain noise feature data; S2, performing a noise comparison operation according to the noise characteristic data to obtain a model adjustment parameter of an initial noise prediction model; S3, adjusting the parameters according to the model, performing a difference degree adjustment operation, and obtaining an adjusted noise prediction model; S4, performing a noise prediction operation according to the adjusted noise prediction model to obtain a predicted noise feature of the electronic component.

[0024] In step S1, during the operation of the electronic component, current operating environment parameter data of the electronic component is obtained and input into a pre-trained noise feature extraction model to obtain noise feature data.

[0025] In the above embodiment, the test system obtains the current working environment parameter data (such as temperature, voltage, current, etc.) of the electronic components and inputs these data into the pre-trained noise feature extraction model. The model predicts and extracts the corresponding noise feature data according to the environmental changes, so that the noise feature data can be accurately extracted. These noise feature data can be used to monitor the working status of electronic components under different environmental conditions, discover potential faults or performance problems in advance, optimize the data distortion problem during the component testing process, and thus improve the accuracy of the noise test. At the same time, the noise test efficiency can be improved and the simulation parameters can be optimized by predicting the working environment parameters of the environment.

[0026] like Figure 2 As shown, in step S1, during the working process of the electronic component, the current working environment parameter data of the electronic component is obtained and input into a pre-trained noise feature extraction model to obtain noise feature data. The noise feature extraction model includes: S11, acquiring historical working environment parameter data, wherein the historical working environment parameter data includes temperature, voltage and current; S12, performing data preprocessing and parameter range determination operations according to the historical working environment parameter data to obtain a parameter range; S13, performing a parameter value interval conversion operation according to the parameter range to obtain a value interval corresponding to the parameter; S14, performing a support vector machine regression operation according to the numerical interval to obtain a noise feature extraction model.

[0027] In step S12, data preprocessing and parameter range determination operations are performed based on the historical working environment parameter data to obtain a parameter range.

[0028] It should be specifically noted that the data preprocessing is crucial, and the specific steps are as follows: First, data cleaning is performed. For missing values, the corresponding records can be deleted if there are few missing values, or the mean, median, or model prediction values ​​can be used as substitutes; then, outliers are identified through box plots based on the upper and lower quartiles and interquartile ranges. False positives can be directly eliminated, while special extreme values ​​need to be retained based on actual conditions. Finally, standardization operations are performed to convert the data into a distribution with a mean of 0 and a standard deviation of 1 to eliminate unit magnitude differences. After this processing, the data is more suitable for training models, the accuracy of noise data acquisition is improved, and an effective data basis is provided for subsequent model training. The formula for standardization operations is: The meanings of the parameters are represents the standardized data, represents the data before normalization, represents the mathematical expectation of the data before standardization, Represents the variance of the data before standardization.

[0029] In step S13, a parameter value interval conversion operation is performed according to the parameter range to obtain a value interval corresponding to the parameter.

[0030] In one implementation, performing a parameter value interval conversion operation according to the parameter range to obtain a value interval corresponding to the parameter includes: the parameter value interval conversion formula is: in, is the numerical interval, is the minimum value of the parameter range, is the maximum value of the parameter range, is the first data.

[0031] It should be specifically noted that the parameter range is , the parameters include current, voltage and temperature.

[0032] In step S2, a noise comparison operation is performed according to the noise characteristic data to obtain model adjustment parameters of the noise prediction model.

[0033] In one implementation, performing a noise comparison operation according to the noise feature data to obtain a model adjustment parameter of the noise prediction model includes: a calculation formula for the model adjustment parameter is: in, How good is the model's prediction? To predict the number of noise data, For the The original noise characteristic data, For the predicted Noise characteristic data, is the preset learning rate, Adjust the parameters for the preset initial model, Parameters are adjusted for the model.

[0034] In the above embodiment, it is first necessary to collect noise feature data, and then use regression analysis method on the selection of noise prediction model. The noise prediction model can capture the relationship between noise source and its influence, so as to provide accurate noise prediction. The adjustment of noise prediction model is very important, and parameters such as learning rate, model complexity, etc. will directly affect the prediction accuracy. After performing multiple noise comparison operations, the optimal model adjustment parameters are obtained, which provides the basis for adjusting parameters for optimizing the noise prediction model in the next step, and finally improves the accuracy of test noise.

[0035] It should be noted that the learning rate is a very critical hyperparameter in the process of machine learning model training. It determines the step size of the model parameter update in each iteration (for example, each iteration of the gradient descent algorithm). Its function is to adjust the parameters according to the gradient of the loss function with respect to the parameters, so that the value of the loss function continues to decrease. If the learning rate is too large, it may cause the model to "skip" the optimal solution during training, and it may not converge or oscillate back and forth near the optimal solution; if the learning rate is too small, the model training process will become very slow, and more iterations are required to achieve better results. In this noise contrast formula, the preset learning rate is combined with the model adjustment parameters to adjust the model. It controls the speed at which the model adjusts the model according to the difference between the predicted value and the actual value. When calculating the quality of the model, the learning rate affects the update amplitude of the model adjustment parameters. If the preset learning rate is large, then each time the model adjustment parameters are adjusted based on the prediction error, the change in the model adjustment parameters will be large, and the model can adapt to the data faster, but there are certain risks, such as over-adjustment, which may lead to model instability; on the contrary, if the preset learning rate is small, the change in the model adjustment parameters will be small, and the model will adjust more slowly, but it can converge to a better state more stably. Balancing the speed and stability of model adjustment during model training affects the model's prediction effect on noisy data, and further affects the final calculation result of the model's quality.

[0036] It should also be specified that regression analysis is a statistical analysis method used to study the relationship between variables. It mainly describes the dependency between a dependent variable and one or more independent variables by establishing a mathematical model. Regression analysis plays an important role in the selection of noise prediction models. It can help find the quantitative relationship between factors (such as independent variables such as operating current, voltage, and temperature) related to the noise characteristics (dependent variables) of electronic components. The model established by regression analysis can accurately predict the size, frequency and other characteristics of noise based on the specific values ​​of these factors. This method can also evaluate the degree of influence of each independent variable on noise and help determine the key factors. This is conducive to focusing on these key factors when building and optimizing the model, so as to more accurately predict noise and provide a reliable basis for controlling and reducing noise.

[0037] like Figure 2 As shown, in step S1, a noise comparison operation is performed according to the noise characteristic data to obtain a model adjustment parameter of the noise prediction model, and the noise prediction model includes: S21, obtaining historical noise characteristic data; S22, performing a judgment set establishment operation according to the historical noise characteristic data to obtain an evaluation set; S23, performing a matrix construction operation according to the evaluation set to obtain an evaluation matrix; S24, performing an entropy value calculation operation according to the evaluation matrix to obtain an entropy value; S25, performing a goodness calculation operation according to the entropy value to obtain a goodness index; S26, performing a model building operation according to the evaluation matrix, the entropy value and the goodness index to obtain a noise prediction model.

[0038] In step S21, historical noise characteristic data is obtained.

[0039] It should be noted that the historical noise characteristic data is obtained by simulation to obtain the noise characteristic data of electronic components. The first step is to build an accurate simulation environment that conforms to actual working conditions. Based on the physical properties, material parameters, electrical specifications and other detailed information of electronic components, after a large number of virtual time step iterative operations, the dynamic changes of multiple physical fields such as electromagnetic fields, stress fields, and thermal fields inside the components are tracked, and the sound wave fluctuations caused by electromagnetic vibrations, thermal expansion and contraction are captured. From microscopic particle vibrations to macroscopic sound wave radiation, all-round analysis is performed, and finally the historical noise characteristic data that conforms to reality is calculated and output.

[0040] In step S22, a judgment set establishment operation is performed based on the historical noise characteristic data to obtain an evaluation set.

[0041] In the above embodiment, the historical noise feature data is obtained by collecting, recording and analyzing the noise data within a certain period of time in the past. These data include information such as noise levels, frequency characteristics, and time domain characteristics at different times, different locations, and different environmental conditions. Through in-depth mining and statistical analysis of these noise feature data, the relationship between noise characteristics and noise sources and environmental factors can be established. When establishing a judgment set, these historical noise feature data will be used to mark and classify samples of different noise types, and group them according to different noise conditions, so as to construct a multi-dimensional and multi-level evaluation set. This evaluation set will serve as the basic data for training the noise prediction model to help the model learn the mapping relationship between noise patterns and noise sources. By using this evaluation set, the trained noise prediction model can more accurately predict new noise data, thereby improving the test system's test of electronic component noise.

[0042] In step S23, a matrix construction operation is performed according to the evaluation set to obtain an evaluation matrix.

[0043] In one embodiment, the evaluation matrix is: in, is the evaluation matrix, No. Line The rating matrix value of the column.

[0044] It should be specifically noted that the evaluation set is established based on obtaining noise values ​​of electronic components under different working environment parameter conditions (including various environmental parameters such as different temperatures, voltages and currents).

[0045] In step S24, an entropy value calculation operation is performed according to the evaluation matrix to obtain an entropy value.

[0046] In one implementation, performing an entropy value calculation operation according to the evaluation matrix to obtain an entropy value includes: the entropy value calculation formula is: in, is the entropy value, For the Line The evaluation matrix value of the column, is the total number of matrix values ​​in the evaluation matrix.

[0047] In step S25, a goodness calculation operation is performed according to the entropy value to obtain a goodness index.

[0048] In one implementation, performing a goodness calculation operation according to the entropy value to obtain a goodness index includes: the goodness calculation formula is: in, is the goodness index, is the entropy value.

[0049] In step S26, a model building operation is performed according to the evaluation matrix, the entropy value and the goodness index to obtain a noise prediction model.

[0050] In one implementation, the model building operation is performed according to the evaluation matrix, the entropy value and the goodness index to obtain a noise prediction model, including: the model building formula is: in, is the noise prediction model, is the entropy matrix under the subjective weight set, is the entropy matrix under the objective weight set, W1 is the subjective weight set, W2 is the objective weight set, is the entropy value, is the preset scale factor, is the goodness index.

[0051] In step S3, a difference degree adjustment operation is performed according to the model adjustment parameters to obtain an adjusted noise prediction model.

[0052] In one implementation, the difference degree adjustment operation is performed according to the model adjustment parameters to obtain the adjusted noise prediction model, including: the difference degree adjustment operation formula is: in, is the function of the adjusted noise prediction model, is the preset coefficient, is the noise characteristic data, Adjust parameters for the model, is the error term.

[0053] In the above embodiment, a large amount of noise data is first collected, and the data is preprocessed to remove outliers and noise points to ensure the accuracy and representativeness of the data. These data can be obtained through sensors and monitoring equipment, wherein the noise data includes key features such as the spectrum, intensity, and volatility of the noise. When performing the difference degree adjustment operation, the data from different sources will first be compared and analyzed to find out the differences between different data, and the parameters of the model will be optimized and adjusted using the difference information. The weights and hyperparameters in the model are adjusted, and new features are introduced to allow the model to more accurately predict the noise performance under different circumstances. In this way, the model can better adapt to a variety of noise environments and improve the accuracy of prediction. Through the optimization operation of this noise prediction model, the accuracy of electronic device noise testing can be further improved.

[0054] It should be specifically noted that the error of the electronic component noise prediction model initially calculated by this method is 10%, which meets the engineering measurement requirements, which means that in practical applications, the prediction error is sufficient to ensure the accuracy of noise monitoring and control and can meet most engineering requirements. However, in order to further improve the accuracy of the prediction model and provide more accurate noise prediction results in more stringent application scenarios, the optimization coefficient can be adjusted by the method in S3 to reduce the prediction error of the noise spectrum density of electronic components, thereby improving the accuracy and efficiency of the entire noise control system.

[0055] S4, performing a noise prediction operation according to the adjusted noise prediction model to obtain a predicted noise feature of the electronic component.

[0056] In the above embodiment, first, the model is trained and optimized by collecting the operating data of electronic components under different working conditions, including multiple parameters such as current, voltage, temperature, etc., and combining with historical noise test data. The adjusted noise prediction model can quickly predict the noise characteristics of electronic components under specific conditions based on real-time data input. Through the optimized noise prediction model, noise can be tested more accurately and test efficiency can be improved.

[0057] The following describes the working process of the present invention using a common scenario as an example. Figure 1 , Figure 1 The invention is a flowchart of a noise testing method for electronic components based on a data model.

[0058] Step 1: When an electronic component enters the noise test system, the system starts to obtain the current working environment parameter data of the electronic component and inputs it into the pre-trained noise feature extraction model to obtain noise feature data; Step 2: The system inputs the noise characteristic data obtained in step 1 under different working environments into the noise prediction model for testing, and the noise prediction model starts to perform optimization operations to obtain an optimized noise prediction model; Step 3: The system uses the optimized noise prediction model as the final prediction model, starts noise detection, and transmits the obtained predicted noise characteristics to the electronic device; Step 4: The electronic device obtains the predicted noise characteristics obtained from the noise test in the electronic component system and saves it to the cloud server for user reference.

[0059] In summary, the present invention provides an electronic component noise testing method based on a data model, comprising: during the working process of the electronic component, obtaining the current working environment parameter data of the electronic component, and inputting it into a pre-trained noise feature extraction model to obtain noise feature data; performing a noise comparison operation according to the noise feature data to obtain the model adjustment parameters of the noise prediction model; performing a difference degree adjustment operation according to the model adjustment parameters to obtain the adjusted noise prediction model; performing a noise prediction operation according to the adjusted noise prediction model to obtain the predicted noise characteristics of the electronic component; wherein the training process of the noise feature extraction model comprises: obtaining historical working environment parameter data, the historical working environment parameter data including temperature, voltage and current; performing data preprocessing and parameter range determination operations according to the historical working environment parameter data to obtain a parameter range; performing a parameter value interval conversion operation according to the parameter range to obtain a value interval corresponding to the parameter; performing a support vector machine regression operation according to the value interval to obtain a noise feature extraction model.

[0060] The embodiment of the present invention also provides an electronic component noise testing system based on a data model, such as Figure 4As shown, it is a structural block diagram of an electronic component noise test system based on a data model provided by an embodiment of the present invention, comprising: a data acquisition module 1, used to obtain the current working environment parameter data of the electronic component during the working process of the electronic component; a model establishment module 2, used to establish a noise feature extraction model and a noise prediction model; a model optimization module 3, used to optimize the noise prediction model to obtain an adjusted noise prediction model; a noise prediction module 4, used to perform a noise prediction operation according to the adjusted noise prediction model to obtain the predicted noise characteristics of the electronic component.

[0061] In an optional embodiment, the model building module 2 is also used to: obtain historical working environment parameter data, the historical working environment parameter data including temperature, voltage and current; perform data preprocessing and parameter range determination operations based on the historical working environment parameter data to obtain a parameter range; perform parameter value interval conversion operations based on the parameter range to obtain a value interval corresponding to the parameter.

[0062] In an optional implementation, the model building module 2 is also used for: the training process of the noise prediction model includes: obtaining historical noise feature data; performing a judgment set building operation based on the historical noise feature data to obtain an evaluation set; performing a matrix construction operation based on the evaluation set to obtain an evaluation matrix; performing an entropy value calculation operation based on the evaluation matrix to obtain an entropy value; performing a goodness calculation operation based on the entropy value to obtain a goodness index; performing a model construction operation based on the evaluation matrix, the entropy value and the goodness index to obtain a noise prediction model; wherein the evaluation matrix is: in, is the evaluation matrix, No. Line The rating matrix value of the column.

[0063] In an optional implementation, the model building module 2 is further used for: performing an entropy value calculation operation according to the evaluation matrix to obtain an entropy value, including: the entropy value calculation formula is: in, is the entropy value, For the Line The evaluation matrix value of the column, is the total number of matrix values ​​in the evaluation matrix.

[0064] In an optional implementation, the model building module 2 is further used to: perform a goodness calculation operation according to the entropy value to obtain a goodness index, including: the goodness calculation formula is in, is the goodness index, is the entropy value.

[0065] In an optional implementation, the model building module 2 is further used to: perform a model building operation according to the evaluation matrix, the entropy value and the goodness index to obtain a noise prediction model, including: the model building formula is: in, is the noise prediction model, is the entropy matrix under the subjective weight set, is the entropy matrix under the objective weight set, W1 is the subjective weight set, W2 is the objective weight set, is the entropy value, is the preset scale factor, is the goodness index.

[0066] In an optional implementation, the model building module 2 is further used to: perform a parameter value interval conversion operation according to the parameter range to obtain a value interval corresponding to the parameter, including: the parameter value interval conversion formula is: in, is the numerical interval, is the minimum value of the parameter range, is the maximum value of the parameter range, is the first data.

[0067] In an optional implementation, the model building module 2 is further used to: perform a support vector machine regression operation according to the numerical interval to obtain a noise feature extraction model, including: the support vector machine regression formula is: in, is the noise feature extraction model, is the number of support vectors, and is the default Lagrange multiplier, is the kernel function, is the bias term, is the newly input sample data, is the first data.

[0068] In an optional implementation, the model optimization module 3 is further used to: perform a noise comparison operation according to the noise feature data to obtain a model adjustment parameter of the noise prediction model, including: a calculation formula for the model adjustment parameter is: in, How good is the model's prediction? To predict the number of noise data, for Noise data, For the The original noise characteristic data, For the predicted Noise characteristic data, is the preset learning rate, Adjust the parameters for the preset initial model, Parameters are adjusted for the model.

[0069] In an optional implementation, the model optimization module 3 is further used to: perform a difference degree adjustment operation according to the model adjustment parameters to obtain an adjusted noise prediction model, including: the difference degree adjustment operation formula is: in, is the function of the adjusted noise prediction model, is the preset coefficient, is the noise characteristic data, Adjust parameters for the model, is the error term.

[0070] An embodiment of the present invention further provides an electronic device, which includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the electronic component noise testing method based on the data model as described in the above embodiment when executing the computer program.

[0071] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0072] The electronic device may be a computing device such as a desktop computer, a notebook, a PDA, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. The electronic device may include more or fewer components than the above components, or may combine certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0073] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0074] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0075] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0076] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0077] In the present invention, the noise test method of electronic components based on the data model has the following advantages: First, by obtaining the noise characteristic data of the electronic components under different working environment parameters (such as temperature, current, voltage, etc.), and applying these data to the construction of the noise prediction model, the accuracy of the noise test can be effectively improved. This method not only reduces the common errors in traditional noise testing, but also greatly reduces the difficulty of testing, because the data model can provide accurate noise characteristic prediction under various environmental conditions, thereby avoiding a large number of cumbersome manual tests and experimental processes. Second, the method can realize real-time prediction of the working environment parameters of the electronic components to be tested, thereby providing a more accurate basis for environmental conditions for noise testing. This prediction capability can significantly improve the efficiency of noise testing and avoid the cumbersome steps of adjusting environmental parameters one by one in traditional testing. Third, by optimizing simulation parameters, the test process can also be made more efficient, unnecessary repeated tests can be reduced, and time and resources can be saved. According to the above method, the noise test method of electronic components is implemented, relying on the established noise feature extraction model and noise prediction model, and combining these two models to accurately predict the noise. Through the continuous optimization of the noise prediction model, the accuracy of the test is further improved, making the noise test results more reliable. At the same time, by predicting and adjusting the working environment parameters, not only can the accuracy of noise testing be improved, but the testing process can also be completed in a shorter time, thereby greatly improving the overall testing efficiency.

[0078] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A noise testing method for electronic components based on a data model, characterized in that: include: During the operation of the electronic component, current operating environment parameter data of the electronic component is obtained and input into a pre-trained noise feature extraction model to obtain noise feature data; According to the noise characteristic data, a noise comparison operation is performed to obtain a model adjustment parameter of an initial noise prediction model; according to the model adjustment parameter, a difference degree adjustment operation is performed to obtain an adjusted noise prediction model; According to the adjusted noise prediction model, a noise prediction operation is performed to obtain the predicted noise characteristics of the electronic components; wherein the training process of the noise characteristic extraction model includes: obtaining historical working environment parameter data, the historical working environment parameter data including temperature, voltage and current; according to the historical working environment parameter data, data preprocessing and parameter range determination operations are performed to obtain the parameter range; according to the parameter range, a parameter numerical interval conversion operation is performed to obtain the numerical interval corresponding to the parameter; according to the numerical interval, a support vector machine regression operation is performed to obtain the noise characteristic extraction model.

2. The electronic component noise testing method based on data model according to claim 1, characterized in that: The step of performing a noise comparison operation according to the noise characteristic data to obtain a model adjustment parameter of the noise prediction model includes: a calculation formula of the model adjustment parameter is: in, How good is the model's prediction? To predict the number of noise data, For the The original noise characteristic data, For the predicted Noise characteristic data, is the preset learning rate, Adjust parameters for the preset initial model, Parameters are adjusted for the model.

3. The electronic component noise testing method based on data model according to claim 1, characterized in that: The step of adjusting the parameters according to the model and performing a difference degree adjustment operation to obtain an adjusted noise prediction model includes: the difference degree adjustment operation formula is: in, is the function of the adjusted noise prediction model, is the preset coefficient, is the noise characteristic data, Adjust parameters for the model, is the error term.

4. The electronic component noise testing method based on data model according to claim 1, characterized in that: The training process of the noise prediction model includes: obtaining historical noise feature data; performing a judgment set establishment operation according to the historical noise feature data to obtain an evaluation set; performing a matrix construction operation according to the evaluation set to obtain an evaluation matrix; performing an entropy value calculation operation according to the evaluation matrix to obtain an entropy value; performing a goodness calculation operation according to the entropy value to obtain a goodness index; performing a model construction operation according to the evaluation matrix, the entropy value and the goodness index to obtain a noise prediction model; wherein the evaluation matrix is: in, is the evaluation matrix, No. Line The rating matrix value of the column.

5. The electronic component noise testing method based on data model according to claim 4, characterized in that: The entropy value calculation operation is performed according to the evaluation matrix to obtain the entropy value, including: the entropy value calculation formula is: in, is the entropy value, For the Line The evaluation matrix value of the column, is the total number of matrix values ​​in the evaluation matrix.

6. The electronic component noise testing method based on data model according to claim 4, characterized in that: The goodness calculation operation is performed according to the entropy value to obtain the goodness index, including: the goodness calculation formula is: in, is the goodness index, is the entropy value.

7. The electronic component noise testing method based on data model according to claim 4, characterized in that: The model building operation is performed according to the evaluation matrix, the entropy value and the goodness index to obtain a noise prediction model, including: the model building formula is: in, is the noise prediction model, is the entropy matrix under the subjective weight set, is the entropy matrix under the objective weight set, W1 is the subjective weight set, W2 is the objective weight set, is the entropy value, is the preset scale factor, is the goodness index.

8. The electronic component noise testing method based on data model according to claim 1, characterized in that: The parameter value interval conversion operation is performed according to the parameter range to obtain the value interval corresponding to the parameter, including: the parameter value interval conversion formula is: in, is the numerical interval, is the minimum value of the parameter range, is the maximum value of the parameter range, is the first data.

9. The electronic component noise testing method based on data model according to claim 1, characterized in that: The support vector machine regression operation is performed according to the numerical interval to obtain a noise feature extraction model, including: the support vector machine regression formula is: in, is the noise feature extraction model, is the number of support vectors, and is the default Lagrange multiplier, is the kernel function, is the bias term, is the newly input sample data, is the first data.

10. An electronic component noise test system based on a data model, characterized in that: include: A data acquisition module, used to acquire current working environment parameter data of the electronic component during its working process; Model building module, used to build noise feature extraction model and noise prediction model; A model optimization module is used to optimize the noise prediction model to obtain an adjusted noise prediction model; a noise prediction module is used to perform a noise prediction operation according to the adjusted noise prediction model to obtain the predicted noise characteristics of the electronic components; wherein the training process of the noise characteristic extraction model includes: obtaining historical working environment parameter data, the historical working environment parameter data including temperature, voltage and current; performing data preprocessing and parameter range determination operations according to the historical working environment parameter data to obtain a parameter range; performing a parameter value interval conversion operation according to the parameter range to obtain a value interval corresponding to the parameter; According to the numerical interval, a support vector machine regression operation is performed to obtain a noise feature extraction model.

Images