CN116718894A

CN116718894A - Circuit stability test method and system for corn lamp

Info

Publication number: CN116718894A
Application number: CN202310724383.6A
Authority: CN
Inventors: 邱财明; 邱吉伊
Original assignee: Shangrao Guangqiang Electronic Technology Co ltd
Current assignee: Shangrao Guangqiang Electronic Technology Co ltd
Priority date: 2023-06-19
Filing date: 2023-06-19
Publication date: 2023-09-08
Anticipated expiration: 2043-06-19
Also published as: CN116718894B

Abstract

A circuit stability test method and system of corn lamp, it obtains voltage value, current value, power value and temperature value of a plurality of predetermined time points in the predetermined time quantum of the corn lamp detected; the intelligent test of the stability of the corn lamp is realized by utilizing the voltage, current, power and temperature data of the corn lamp and combining deep learning and artificial intelligence technology. Therefore, the stability of the corn lamp can be comprehensively evaluated, the objectivity and the accuracy of the test are improved, and the labor cost and the time cost of the test are reduced.

Description

Circuit stability test method and system for corn lamp

Technical Field

The application relates to the technical field of intelligent testing, in particular to a circuit stability testing method and system of a corn lamp.

Background

Corn light is a common energy-saving light widely used for household, commercial and industrial lighting. The LED light source is characterized in that the LED light source is adopted, the shape of the LED light source is like a corn cob, the LED light source can emit light at 360 degrees, the light efficiency is high, the service life is long, and the LED light source is environment-friendly and pollution-free. The circuit stability of corn lamps is an important factor affecting their performance and life, and thus, need to be tested effectively.

At present, the traditional corn lamp circuit stability testing method is mainly based on manual observation and judgment, lacks objectivity and accuracy and is low in efficiency. Thus, a solution is desired.

Disclosure of Invention

The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides a circuit stability testing method and a system thereof for a corn lamp, wherein the circuit stability testing method and the system acquire voltage values, current values, power values and temperature values of a plurality of preset time points in a preset time period of the detected corn lamp; the intelligent test of the stability of the corn lamp is realized by utilizing the voltage, current, power and temperature data of the corn lamp and combining deep learning and artificial intelligence technology. Therefore, the stability of the corn lamp can be comprehensively evaluated, the objectivity and the accuracy of the test are improved, and the labor cost and the time cost of the test are reduced.

In a first aspect, a method for testing circuit stability of a corn lamp is provided, comprising:

acquiring voltage values, current values, power values and temperature values of a plurality of preset time points in a preset time period of a detected corn lamp;

arranging the voltage values, the current values, the power values and the temperature values of a plurality of preset time points in the preset time period into a voltage input vector, a current input vector, a power input vector and a temperature input vector according to the time dimension respectively;

Arranging the voltage input vector, the current input vector, the power input vector and the temperature input vector into a two-dimensional input matrix, and then obtaining a parameter association characteristic matrix through an inter-parameter association characteristic extractor based on a convolutional neural network model;

respectively passing the voltage input vector, the current input vector, the power input vector and the temperature input vector through a time sequence feature extractor based on a one-dimensional convolutional neural network model to obtain a voltage time sequence feature vector, a current time sequence feature vector, a power time sequence feature vector and a temperature time sequence feature vector;

respectively taking the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector as query feature vectors, and calculating a matrix product between the query feature vectors and the parameter association feature matrix to obtain a corrected voltage time sequence feature vector, a corrected current time sequence feature vector, a corrected power time sequence feature vector and a corrected temperature time sequence feature vector;

cascading the corrected voltage time sequence feature vector, the corrected current time sequence feature vector, the corrected power time sequence feature vector and the corrected temperature time sequence feature vector to obtain a classification feature vector; and

And the classification feature vector is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the circuit stability of the detected corn lamp meets the preset requirement.

In the above method for testing circuit stability of a corn lamp, the step of arranging the voltage input vector, the current input vector, the power input vector and the temperature input vector into a two-dimensional input matrix and obtaining a parameter correlation feature matrix through an inter-parameter correlation feature extractor based on a convolutional neural network model comprises the following steps: and respectively carrying out convolution processing, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transfer of layers by using each layer of the inter-parameter correlation feature extractor based on the convolutional neural network model, wherein the output of the last layer of the inter-parameter correlation feature extractor based on the convolutional neural network model is taken as the parameter correlation feature matrix, and the input of the first layer of the inter-parameter correlation feature extractor based on the convolutional neural network model is taken as the two-dimensional input matrix.

In the above method for testing circuit stability of a corn lamp, the step of passing the voltage input vector, the current input vector, the power input vector and the temperature input vector through a time sequence feature extractor based on a one-dimensional convolutional neural network model to obtain a voltage time sequence feature vector, a current time sequence feature vector, a power time sequence feature vector and a temperature time sequence feature vector, includes: and respectively carrying out convolution processing, feature matrix-based averaging pooling processing and nonlinear activation processing on input data in forward transfer of layers by using each layer of the time sequence feature extractor based on the one-dimensional convolutional neural network model, wherein the output of the last layer of the time sequence feature extractor based on the one-dimensional convolutional neural network model is the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector, and the input of the first layer of the time sequence feature extractor based on the one-dimensional convolutional neural network model is the voltage input vector, the current input vector, the power input vector and the temperature input vector.

In the above method for testing circuit stability of a corn lamp, respectively using the voltage timing characteristic vector, the current timing characteristic vector, the power timing characteristic vector and the temperature timing characteristic vector as query characteristic vectors, calculating a matrix product between the query characteristic vector and the parameter correlation characteristic matrix to obtain a corrected voltage timing characteristic vector, a corrected current timing characteristic vector, a corrected power timing characteristic vector and a corrected temperature timing characteristic vector, including: calculating the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector by using the following product formula as a matrix product between the query feature vector and the parameter association feature matrix to obtain a corrected voltage time sequence feature vector, a corrected current time sequence feature vector, a corrected power time sequence feature vector and a corrected temperature time sequence feature vector; wherein, the product formula is:

wherein M is ₁ Representing the parameter association characteristic matrix, V _ai Representing the voltage timing feature vector or the current timing feature vector or the power timing feature vector or the temperature timing feature vector, V _bi Representing the corrected voltage timing feature vector or the corrected current timing feature vector or the corrected power timing feature vector or the corrected temperature timing feature vector, Representing matrix multiplication.

In the above method for testing circuit stability of a corn lamp, cascading the corrected voltage timing characteristic vector, the corrected current timing characteristic vector, the corrected power timing characteristic vector and the corrected temperature timing characteristic vector to obtain a classification characteristic vector includes: cascading the corrected voltage time sequence feature vector, the corrected current time sequence feature vector, the corrected power time sequence feature vector and the corrected temperature time sequence feature vector by using the following cascading formula to obtain a classification feature vector; wherein, the cascade formula is:

V _c ＝Concat[V _b1 ，V _b2 ，V _b3 ，V _b4 ]

wherein V is _b1 ，V _b2 ，V _b3 ，V _b4 Representing the corrected voltage timing characteristic vector, corrected current timing characteristic vector, corrected power timing characteristic vector and corrected temperature timing characteristic vector, concat [. Cndot. ]]Representing a cascade function, V _c Representing the classification feature vector.

In the above method for testing the circuit stability of a corn lamp, the classifying feature vector is passed through a classifier to obtain a classifying result, where the classifying result is used to indicate whether the circuit stability of the detected corn lamp meets a predetermined requirement, and the method includes: performing full-connection coding on the classification feature vectors by using a plurality of full-connection layers of the classifier to obtain coded classification feature vectors; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.

The circuit stability testing method of the corn lamp further comprises the following training steps: training the inter-parameter correlation feature extractor based on the convolutional neural network model, the time sequence feature extractor based on the one-dimensional convolutional neural network model and the classifier; wherein the training step comprises: acquiring training data, wherein the training data comprises training voltage values, training current values, training power values and training temperature values of a plurality of preset time points in a preset time period of a detected corn lamp, and whether the circuit stability of the detected corn lamp reaches a true value of a preset requirement; respectively arranging the training voltage values, the training current values, the training power values and the training temperature values of a plurality of preset time points in the preset time period into training voltage input vectors, training current input vectors, training power input vectors and training temperature input vectors according to time dimensions; arranging the training voltage input vector, the training current input vector, the training power input vector and the training temperature input vector into a two-dimensional input matrix, and then obtaining a training parameter association feature matrix through the inter-parameter association feature extractor based on the convolutional neural network model; respectively passing the training voltage input vector, the training current input vector, the training power input vector and the training temperature input vector through the time sequence feature extractor based on the one-dimensional convolutional neural network model to obtain a training voltage time sequence feature vector, a training current time sequence feature vector, a training power time sequence feature vector and a training temperature time sequence feature vector; respectively taking the training voltage time sequence feature vector, the training current time sequence feature vector, the training power time sequence feature vector and the training temperature time sequence feature vector as query feature vectors, and calculating a matrix product between the query feature vectors and the training parameter association feature matrix to obtain a training corrected voltage time sequence feature vector, a training corrected current time sequence feature vector, a training corrected power time sequence feature vector and a training corrected temperature time sequence feature vector; cascading the training corrected voltage time sequence feature vector, the training corrected current time sequence feature vector, the training corrected power time sequence feature vector and the training corrected temperature time sequence feature vector to obtain a training classification feature vector; the training classification feature vector passes through a classifier to obtain a classification loss function value; calculating pseudo-cycle difference penalty factors of the training voltage time sequence feature vector, the training current time sequence feature vector, the training power time sequence feature vector, the training temperature time sequence feature vector, the training corrected voltage time sequence feature vector, the training corrected current time sequence feature vector, the training corrected power time sequence feature vector and the training corrected temperature time sequence feature vector; and training the inter-parameter correlation feature extractor based on the convolutional neural network model, the timing feature extractor based on the one-dimensional convolutional neural network model and the classifier by taking a weighted sum of the classification loss function value and the pseudo-cyclic difference penalty factor as a loss function value and propagating in a gradient descent direction.

In the above method for testing circuit stability of a corn lamp, calculating pseudo-cyclic difference penalty factors of the training voltage timing feature vector, the training current timing feature vector, the training power timing feature vector, the training temperature timing feature vector, the training corrected voltage timing feature vector, the training corrected current timing feature vector, the training corrected power timing feature vector, and the training corrected temperature timing feature vector, the method comprises: calculating pseudo-cycle difference penalty factors of the training voltage time sequence feature vector, the training current time sequence feature vector, the training power time sequence feature vector, the training temperature time sequence feature vector, the training corrected voltage time sequence feature vector, the training corrected current time sequence feature vector, the training corrected power time sequence feature vector and the training corrected temperature time sequence feature vector according to the following optimization formula; wherein, the optimization formula is:

wherein V is _1i Representing one of the training voltage timing feature vector, the training current timing feature vector, the training power timing feature vector, and the training temperature timing feature vector, V _1i ' represents one of the training corrected voltage timing feature vector, the training corrected current timing feature vector, the training corrected power timing feature vector, and the training corrected temperature timing feature vector, D (V _1i ，V _1i ')) is a feature vector V _1i And V _1i Distance matrix between' i|·|| _F The Frobenius norm of the matrix, L is the length of the eigenvector, d (V _1i ，V _1i ' is the feature vector V _1i And V _1i ' Euclidean distance between |·|| ₂ Is the two norms of the vector, log represents the base 2 logarithm, and alpha and beta are weighted hyper-parameters,representing the pseudo-cyclic difference penalty factor.

In a second aspect, a circuit stability testing system for a corn light is provided, comprising:

the data acquisition module is used for acquiring voltage values, current values, power values and temperature values of a plurality of preset time points in a preset time period of the detected corn lamp;

the vector arrangement module is used for respectively arranging the voltage values, the current values, the power values and the temperature values of a plurality of preset time points in the preset time period into a voltage input vector, a current input vector, a power input vector and a temperature input vector according to the time dimension;

the parameter correlation feature extraction module is used for arranging the voltage input vector, the current input vector, the power input vector and the temperature input vector into a two-dimensional input matrix and then obtaining a parameter correlation feature matrix through a parameter correlation feature extractor based on a convolutional neural network model;

The time sequence feature extraction module is used for respectively enabling the voltage input vector, the current input vector, the power input vector and the temperature input vector to pass through a time sequence feature extractor based on a one-dimensional convolutional neural network model so as to obtain a voltage time sequence feature vector, a current time sequence feature vector, a power time sequence feature vector and a temperature time sequence feature vector;

the correction module is used for respectively taking the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector as query feature vectors, and calculating a matrix product between the query feature vectors and the parameter association feature matrix to obtain corrected voltage time sequence feature vectors, corrected current time sequence feature vectors, corrected power time sequence feature vectors and corrected temperature time sequence feature vectors;

the cascade module is used for cascading the corrected voltage time sequence feature vector, the corrected current time sequence feature vector, the corrected power time sequence feature vector and the corrected temperature time sequence feature vector to obtain a classification feature vector; and

and the circuit stability result generation module is used for passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the circuit stability of the detected corn lamp meets the preset requirement.

In the above circuit stability test system for a corn lamp, the inter-parameter correlation feature extraction module is configured to: and respectively carrying out convolution processing, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transfer of layers by using each layer of the inter-parameter correlation feature extractor based on the convolutional neural network model, wherein the output of the last layer of the inter-parameter correlation feature extractor based on the convolutional neural network model is taken as the parameter correlation feature matrix, and the input of the first layer of the inter-parameter correlation feature extractor based on the convolutional neural network model is taken as the two-dimensional input matrix.

Compared with the prior art, the circuit stability testing method and the system thereof of the corn lamp provided by the application acquire voltage values, current values, power values and temperature values of a plurality of preset time points in a preset time period of the detected corn lamp; the intelligent test of the stability of the corn lamp is realized by utilizing the voltage, current, power and temperature data of the corn lamp and combining deep learning and artificial intelligence technology. Therefore, the stability of the corn lamp can be comprehensively evaluated, the objectivity and the accuracy of the test are improved, and the labor cost and the time cost of the test are reduced.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic diagram of a circuit stability test method of a corn lamp according to an embodiment of the application.

Fig. 2 is a flow chart of a method for testing circuit stability of a corn lamp according to an embodiment of the application.

Fig. 3 is a schematic diagram of a circuit stability testing method of a corn lamp according to an embodiment of the application.

Fig. 4 is a flowchart of the sub-steps of step 170 in a method for testing the circuit stability of a corn lamp according to an embodiment of the application.

Fig. 5 is a flowchart of the sub-steps of step 180 in the circuit stability test method of the corn lamp according to an embodiment of the application.

Fig. 6 is a block diagram of a circuit stability testing system for a corn light in accordance with an embodiment of the present application.

Detailed Description

The following description of the technical solutions according to the embodiments of the present application will be given with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

Unless defined otherwise, all technical and scientific terms used in the embodiments of the application have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application.

In describing embodiments of the present application, unless otherwise indicated and limited thereto, the term "connected" should be construed broadly, for example, it may be an electrical connection, or may be a communication between two elements, or may be a direct connection, or may be an indirect connection via an intermediate medium, and it will be understood by those skilled in the art that the specific meaning of the term may be interpreted according to circumstances.

It should be noted that, the term "first\second\third" related to the embodiment of the present application is merely to distinguish similar objects, and does not represent a specific order for the objects, it is to be understood that "first\second\third" may interchange a specific order or sequence where allowed. It is to be understood that the "first\second\third" distinguishing objects may be interchanged where appropriate such that embodiments of the application described herein may be practiced in sequences other than those illustrated or described herein.

Aiming at the technical problems, the technical conception of the application comprehensively utilizes the voltage, current, power and temperature data of the corn lamp, and combines deep learning and artificial intelligence technology to realize intelligent test on the stability of the corn lamp.

Specifically, in the technical scheme of the application, firstly, voltage values, current values, power values and temperature values of a plurality of preset time points in a preset time period of a corn lamp to be detected are obtained. The circuit stability of the corn lamp at different time points can be comprehensively known from the data, and the circuit stability is an important data base for subsequent data processing and analysis.

Then, the voltage value, the current value, the power value and the temperature value at a plurality of preset time points in the preset time period are respectively arranged into a voltage input vector, a current input vector, a power input vector and a temperature input vector according to the time dimension. That is, time-series discrete distributions of voltage values, current values, power values, and temperature values are respectively constructed as the structured voltage input vector, current input vector, power input vector, and temperature input vector.

And the correlation relationship among parameters of the extracted corn lamps, namely the mutual influence and the dependence relationship among the parameters are considered. In the technical scheme of the application, the voltage input vector, the current input vector, the power input vector and the temperature input vector are arranged into a two-dimensional input matrix, and then the parameter correlation characteristic matrix is obtained through a parameter correlation characteristic extractor based on a convolutional neural network model. That is, by arranging the input vectors of the respective parameters into a two-dimensional input matrix, local feature sensing and downsampling operations can be performed on the input matrix by using the convolutional layer and the pooling layer of the convolutional neural network model, thereby extracting high-level and abstract features between the parameters. These features can reflect the law of variation of various parameters of the corn light circuit.

And simultaneously, respectively passing the voltage input vector, the current input vector, the power input vector and the temperature input vector through a time sequence feature extractor based on a one-dimensional convolutional neural network model to obtain a voltage time sequence feature vector, a current time sequence feature vector, a power time sequence feature vector and a temperature time sequence feature vector. It should be appreciated that in circuit stability testing, not only the correlation between different parameters, but also the variation of the same parameter at different points in time need to be considered. That is, the voltage value, the current value, the power value, and the temperature value have dynamic variability in the time dimension. Here, the feature extraction and analysis are performed by using a one-dimensional convolutional neural network model, so that the change condition of the circuit parameters in the time dimension can be obtained. In particular, the network structure of a one-dimensional convolutional neural network model generally includes an input layer, a convolutional layer, and a pooling layer. The input layer receives one-dimensional sequence data. The convolution layer performs convolution operations on the input sequence using a set of learnable filters to extract local features. The pooling layer downsamples the output of the convolutional layer to reduce the number of parameters and computation while preserving important feature information.

And then, respectively taking the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector as query feature vectors, and calculating a matrix product between the query feature vectors and the parameter association feature matrix to obtain a corrected voltage time sequence feature vector, a corrected current time sequence feature vector, a corrected power time sequence feature vector and a corrected temperature time sequence feature vector. As described above, the inter-parameter correlation feature matrix is extracted from the two-dimensional input matrix by the convolutional neural network model, and reflects the interaction and change rule between different parameters. By taking the time sequence feature vector as the query feature vector and performing matrix multiplication with the inter-parameter association feature matrix, the feature vector can more accurately describe the circuit stability of the corn lamp, interference and noise among different parameters are eliminated, and the feature expression capability is improved.

Since in circuit stability testing, the voltage, current, power and temperature parameters are interrelated, neither type of parameter exists independently, and the trend of variation between them can reflect the stability of the circuit. Therefore, in the technical scheme of the application, the corrected voltage time sequence feature vector, the corrected current time sequence feature vector, the corrected power time sequence feature vector and the corrected temperature time sequence feature vector are cascaded to more comprehensive and accurate feature vectors, so that the classified feature vectors are obtained.

Further, the classification feature vector is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the circuit stability of the detected corn lamp meets a preset requirement. Wherein the classifier is capable of mapping the input data into predefined classification labels. Specifically, in the circuit stability test, the classification feature vector is mapped to one of two categories of "the circuit stability of the detected corn light reaches a predetermined requirement" or "the circuit stability of the detected corn light does not reach the predetermined requirement" through a soft maximum function. In this way, the circuit stability of the corn lamp can be accurately assessed, providing a reference for subsequent production and quality control.

In the technical scheme of the application, when the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector are respectively used as query feature vectors, and the matrix product between the query feature vectors and the parameter correlation feature matrix is calculated to obtain corrected voltage time sequence feature vectors, corrected current time sequence feature vectors, corrected power time sequence feature vectors and corrected temperature time sequence feature vectors, the time sequence-sample cross dimension local correlation features of the parameters expressed by the parameter correlation feature matrix are essentially mapped into the time sequence distribution feature space of each individual feature, so that the corrected feature vectors have unbalance between overall feature distribution relative to the feature vectors before correction, thereby influencing the expression effect of the corrected feature vectors on the feature time sequence distribution, and also influencing the expression effect of the classification feature vectors obtained by cascading the corrected feature vectors.

Based on this, the applicant of the present application furtherIntroducing feature vectors before correction, e.g. denoted V, for each _1i And corrected feature vectors, e.g. denoted V _1i The 'pseudo-cyclic difference penalty factor' is expressed as a loss function:

D(v _1i ,V _1i ')) is a feature vector V _1i And V _1i The eigenvalues of the distance matrix between', i.e. the (j, k) th position of said distance matrix, are eigenvectors V _1i The j-th eigenvalue v _1ij And feature vector V _ii ' kth eigenvalue v _1ik The distance between the's points, I.I _F The Frobenius norm of the matrix, L is the length of the eigenvector, d (V _1i ，V _1i ' is the feature vector V _1i And V _1i Distance between' e.g. Euclidean distance, ||. | ₂ Is the two norms of the vector, log represents the base 2 logarithm, and α and β are weighted hyper-parameters.

Here, consider each pre-correction feature vector V _1i And each corrected feature vector V _1i The imbalance distribution between' causes gradient propagation anomalies during the back-propagation model training process based on gradient descent, thus forming pseudo-loops of model parameter updates, which are treated as true loops during the model training process of minimizing the loss function by introducing penalty factors for expressing both spatial and numerical relationships of closely related numerical pairs of eigenvalues, to achieve each pre-correction eigenvector V by simulated activation of gradient propagation _1i And each corrected feature vector V _1i ' progressive coupling of the respective feature distributions, thereby improving the expression effect of the corrected feature vectors on the feature timing distribution, to improve the expression effect of the classification feature vectors obtained by the corrected feature vector concatenation.

The application has the following technical effects: 1. an intelligent circuit stability test scheme for a corn lamp is provided. 2. According to the scheme, the voltage, current, power and temperature parameters of the corn lamp can be tested, the stability of the corn lamp can be comprehensively evaluated, the objectivity and accuracy of the test are improved, and the labor cost and the time cost of the test are reduced.

Fig. 1 is a schematic diagram of a circuit stability test method of a corn lamp according to an embodiment of the application. As shown in fig. 1, in this application scenario, first, voltage values (e.g., C1 as illustrated in fig. 1), current values (e.g., C2 as illustrated in fig. 1), power values (e.g., C3 as illustrated in fig. 1), and temperature values (e.g., C4 as illustrated in fig. 1) at a plurality of predetermined time points within a predetermined period of time of a detected corn lamp (e.g., M as illustrated in fig. 1) are acquired; the obtained voltage value, current value, power value and temperature value are then input into a server (e.g., S as illustrated in fig. 1) where a circuit stability test algorithm for the corn light is deployed, wherein the server is capable of processing the voltage value, the current value, the power value and the temperature value based on the circuit stability test algorithm for the corn light to generate a classification result indicating whether the circuit stability of the detected corn light meets a predetermined requirement.

Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.

In one embodiment of the present application, fig. 2 is a flow chart of a method for testing the circuit stability of a corn lamp according to an embodiment of the present application. As shown in fig. 2, a circuit stability testing method 100 of a corn lamp according to an embodiment of the present application includes: 110, obtaining voltage values, current values, power values and temperature values of a plurality of preset time points in a preset time period of the detected corn lamp; 120, arranging the voltage values, the current values, the power values and the temperature values of a plurality of preset time points in the preset time period into a voltage input vector, a current input vector, a power input vector and a temperature input vector according to the time dimension respectively; 130, arranging the voltage input vector, the current input vector, the power input vector and the temperature input vector into a two-dimensional input matrix, and then obtaining a parameter association feature matrix through an inter-parameter association feature extractor based on a convolutional neural network model; 140, passing the voltage input vector, the current input vector, the power input vector and the temperature input vector through a time sequence feature extractor based on a one-dimensional convolutional neural network model to obtain a voltage time sequence feature vector, a current time sequence feature vector, a power time sequence feature vector and a temperature time sequence feature vector; 150, respectively taking the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector as query feature vectors, and calculating a matrix product between the query feature vectors and the parameter association feature matrix to obtain a corrected voltage time sequence feature vector, a corrected current time sequence feature vector, a corrected power time sequence feature vector and a corrected temperature time sequence feature vector; 160, concatenating the corrected voltage timing feature vector, corrected current timing feature vector, corrected power timing feature vector, and corrected temperature timing feature vector to obtain a classification feature vector; and 170, passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the circuit stability of the detected corn lamp meets the preset requirement.

Fig. 3 is a schematic diagram of a circuit stability testing method of a corn lamp according to an embodiment of the application. As shown in fig. 3, in the network architecture, first, voltage values, current values, power values and temperature values at a plurality of predetermined time points within a predetermined period of time of a detected corn lamp are obtained; then, arranging the voltage value, the current value, the power value and the temperature value of a plurality of preset time points in the preset time period into a voltage input vector, a current input vector, a power input vector and a temperature input vector according to the time dimension respectively; then, arranging the voltage input vector, the current input vector, the power input vector and the temperature input vector into a two-dimensional input matrix, and then obtaining a parameter association feature matrix through an inter-parameter association feature extractor based on a convolutional neural network model; then, the voltage input vector, the current input vector, the power input vector and the temperature input vector are respectively passed through a time sequence feature extractor based on a one-dimensional convolutional neural network model to obtain a voltage time sequence feature vector, a current time sequence feature vector, a power time sequence feature vector and a temperature time sequence feature vector; then, respectively taking the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector as query feature vectors, and calculating a matrix product between the query feature vectors and the parameter association feature matrix to obtain a corrected voltage time sequence feature vector, a corrected current time sequence feature vector, a corrected power time sequence feature vector and a corrected temperature time sequence feature vector; then, cascading the corrected voltage time sequence feature vector, the corrected current time sequence feature vector, the corrected power time sequence feature vector and the corrected temperature time sequence feature vector to obtain a classification feature vector; and finally, the classification feature vector is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the circuit stability of the detected corn lamp meets the preset requirement.

Specifically, in step 110, voltage values, current values, power values, and temperature values at a plurality of predetermined time points within a predetermined period of time of the detected corn light are obtained. Aiming at the technical problems, the technical conception of the application comprehensively utilizes the voltage, current, power and temperature data of the corn lamp, and combines deep learning and artificial intelligence technology to realize intelligent test on the stability of the corn lamp.

Specifically, in step 120, the voltage values, the current values, the power values, and the temperature values at a plurality of predetermined time points within the predetermined time period are respectively arranged into a voltage input vector, a current input vector, a power input vector, and a temperature input vector according to a time dimension. Then, the voltage value, the current value, the power value and the temperature value at a plurality of preset time points in the preset time period are respectively arranged into a voltage input vector, a current input vector, a power input vector and a temperature input vector according to the time dimension. That is, time-series discrete distributions of voltage values, current values, power values, and temperature values are respectively constructed as the structured voltage input vector, current input vector, power input vector, and temperature input vector.

Specifically, in step 130, the voltage input vector, the current input vector, the power input vector and the temperature input vector are arranged into a two-dimensional input matrix, and then the two-dimensional input matrix is passed through an inter-parameter correlation feature extractor based on a convolutional neural network model to obtain a parameter correlation feature matrix. And the correlation relationship among parameters of the extracted corn lamps, namely the mutual influence and the dependence relationship among the parameters are considered. In the technical scheme of the application, the voltage input vector, the current input vector, the power input vector and the temperature input vector are arranged into a two-dimensional input matrix, and then the parameter correlation characteristic matrix is obtained through a parameter correlation characteristic extractor based on a convolutional neural network model. That is, by arranging the input vectors of the respective parameters into a two-dimensional input matrix, local feature sensing and downsampling operations can be performed on the input matrix by using the convolutional layer and the pooling layer of the convolutional neural network model, thereby extracting high-level and abstract features between the parameters. These features can reflect the law of variation of various parameters of the corn light circuit.

The method for obtaining the parameter correlation feature matrix through the inter-parameter correlation feature extractor based on the convolutional neural network model after arranging the voltage input vector, the current input vector, the power input vector and the temperature input vector into a two-dimensional input matrix comprises the following steps: and respectively carrying out convolution processing, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transfer of layers by using each layer of the inter-parameter correlation feature extractor based on the convolutional neural network model, wherein the output of the last layer of the inter-parameter correlation feature extractor based on the convolutional neural network model is taken as the parameter correlation feature matrix, and the input of the first layer of the inter-parameter correlation feature extractor based on the convolutional neural network model is taken as the two-dimensional input matrix.

The convolutional neural network (Convolutional Neural Network, CNN) is an artificial neural network and has wide application in the fields of image recognition and the like. The convolutional neural network may include an input layer, a hidden layer, and an output layer, where the hidden layer may include a convolutional layer, a pooling layer, an activation layer, a full connection layer, etc., where the previous layer performs a corresponding operation according to input data, outputs an operation result to the next layer, and obtains a final result after the input initial data is subjected to a multi-layer operation.

The convolutional neural network model has excellent performance in the aspect of image local feature extraction by taking a convolutional kernel as a feature filtering factor, and has stronger feature extraction generalization capability and fitting capability compared with the traditional image feature extraction algorithm based on statistics or feature engineering.

Specifically, in step 140, the voltage input vector, the current input vector, the power input vector, and the temperature input vector are respectively passed through a time series feature extractor based on a one-dimensional convolutional neural network model to obtain a voltage time series feature vector, a current time series feature vector, a power time series feature vector, and a temperature time series feature vector. And simultaneously, respectively passing the voltage input vector, the current input vector, the power input vector and the temperature input vector through a time sequence feature extractor based on a one-dimensional convolutional neural network model to obtain a voltage time sequence feature vector, a current time sequence feature vector, a power time sequence feature vector and a temperature time sequence feature vector.

It should be appreciated that in circuit stability testing, not only the correlation between different parameters, but also the variation of the same parameter at different points in time need to be considered. That is, the voltage value, the current value, the power value, and the temperature value have dynamic variability in the time dimension. Here, the feature extraction and analysis are performed by using a one-dimensional convolutional neural network model, so that the change condition of the circuit parameters in the time dimension can be obtained.

In particular, the network structure of a one-dimensional convolutional neural network model generally includes an input layer, a convolutional layer, and a pooling layer. The input layer receives one-dimensional sequence data. The convolution layer performs convolution operations on the input sequence using a set of learnable filters to extract local features. The pooling layer downsamples the output of the convolutional layer to reduce the number of parameters and computation while preserving important feature information.

The step of passing the voltage input vector, the current input vector, the power input vector and the temperature input vector through a time sequence feature extractor based on a one-dimensional convolutional neural network model to obtain a voltage time sequence feature vector, a current time sequence feature vector, a power time sequence feature vector and a temperature time sequence feature vector, comprises the following steps: and respectively carrying out convolution processing, feature matrix-based averaging pooling processing and nonlinear activation processing on input data in forward transfer of layers by using each layer of the time sequence feature extractor based on the one-dimensional convolutional neural network model, wherein the output of the last layer of the time sequence feature extractor based on the one-dimensional convolutional neural network model is the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector, and the input of the first layer of the time sequence feature extractor based on the one-dimensional convolutional neural network model is the voltage input vector, the current input vector, the power input vector and the temperature input vector.

Specifically, in step 150, the voltage timing feature vector, the current timing feature vector, the power timing feature vector, and the temperature timing feature vector are used as query feature vectors, respectively, and the matrix product between the query feature vectors and the parameter correlation feature matrix is calculated to obtain a corrected voltage timing feature vector, a corrected current timing feature vector, a corrected power timing feature vector, and a corrected temperature timing feature vector.

The method for calculating the matrix product between the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector serving as query feature vectors respectively and the parameter association feature matrix to obtain a corrected voltage time sequence feature vector, a corrected current time sequence feature vector, a corrected power time sequence feature vector and a corrected temperature time sequence feature vector comprises the following steps: calculating the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector by using the following product formula as a matrix product between the query feature vector and the parameter association feature matrix to obtain a corrected voltage time sequence feature vector, a corrected current time sequence feature vector, a corrected power time sequence feature vector and a corrected temperature time sequence feature vector; wherein, the product formula is:

Specifically, in step 160, the corrected voltage timing feature vector, the corrected current timing feature vector, the corrected power timing feature vector, and the corrected temperature timing feature vector are concatenated to obtain a classification feature vector. Since in circuit stability testing, the voltage, current, power and temperature parameters are interrelated, neither type of parameter exists independently, and the trend of variation between them can reflect the stability of the circuit. Therefore, in the technical scheme of the application, the corrected voltage time sequence feature vector, the corrected current time sequence feature vector, the corrected power time sequence feature vector and the corrected temperature time sequence feature vector are cascaded to more comprehensive and accurate feature vectors, so that the classified feature vectors are obtained.

The step of cascading the corrected voltage time sequence feature vector, the corrected current time sequence feature vector, the corrected power time sequence feature vector and the corrected temperature time sequence feature vector to obtain a classification feature vector comprises the following steps: cascading the corrected voltage time sequence feature vector, the corrected current time sequence feature vector, the corrected power time sequence feature vector and the corrected temperature time sequence feature vector by using the following cascading formula to obtain a classification feature vector; wherein, the cascade formula is:

V _c ＝Concat[V _b1 ，V _b2 ，V _b3 ，V _b4 ]

Specifically, in step 170, the classification feature vector is passed through a classifier to obtain a classification result, where the classification result is used to indicate whether the circuit stability of the detected corn lamp meets a predetermined requirement. Further, the classification feature vector is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether the circuit stability of the detected corn lamp meets a preset requirement. Wherein the classifier is capable of mapping the input data into predefined classification labels. Specifically, in the circuit stability test, the classification feature vector is mapped to one of two categories of "the circuit stability of the detected corn light reaches a predetermined requirement" or "the circuit stability of the detected corn light does not reach the predetermined requirement" through a soft maximum function. In this way, the circuit stability of the corn lamp can be accurately assessed, providing a reference for subsequent production and quality control.

Fig. 4 is a flowchart of a sub-step of step 170 in a method for testing the circuit stability of a corn lamp according to an embodiment of the present application, as shown in fig. 4, the classification feature vector is passed through a classifier to obtain a classification result, where the classification result is used to indicate whether the circuit stability of the tested corn lamp meets a predetermined requirement, and the method includes: 171, performing full-connection encoding on the classification feature vector by using a plurality of full-connection layers of the classifier to obtain an encoded classification feature vector; and, 172, passing the encoded classification feature vector through a Softmax classification function of the classifier to obtain the classification result.

The circuit stability testing method of the corn lamp further comprises the training steps of: training the inter-parameter correlation feature extractor based on the convolutional neural network model, the time sequence feature extractor based on the one-dimensional convolutional neural network model and the classifier; fig. 5 is a flowchart showing the sub-steps of step 180 in the circuit stability test method of the corn lamp according to the embodiment of the application, and as shown in fig. 5, the training step 180 includes: 181, obtaining training data, wherein the training data comprises training voltage values, training current values, training power values and training temperature values of a plurality of preset time points in a preset time period of the detected corn lamp, and whether the circuit stability of the detected corn lamp reaches a true value of a preset requirement; 182, arranging the training voltage values, the training current values, the training power values and the training temperature values of a plurality of preset time points in the preset time period into a training voltage input vector, a training current input vector, a training power input vector and a training temperature input vector according to the time dimension respectively; 183, arranging the training voltage input vector, the training current input vector, the training power input vector and the training temperature input vector into a two-dimensional input matrix, and then obtaining a training parameter association feature matrix through the inter-parameter association feature extractor based on the convolutional neural network model; 184, passing the training voltage input vector, the training current input vector, the training power input vector and the training temperature input vector through the one-dimensional convolutional neural network model-based time sequence feature extractor to obtain a training voltage time sequence feature vector, a training current time sequence feature vector, a training power time sequence feature vector and a training temperature time sequence feature vector; 185, respectively taking the training voltage time sequence feature vector, the training current time sequence feature vector, the training power time sequence feature vector and the training temperature time sequence feature vector as query feature vectors, and calculating a matrix product between the query feature vectors and the training parameter association feature matrix to obtain a training corrected voltage time sequence feature vector, a training corrected current time sequence feature vector, a training corrected power time sequence feature vector and a training corrected temperature time sequence feature vector; 186, cascading the training corrected voltage time sequence feature vector, the training corrected current time sequence feature vector, the training corrected power time sequence feature vector and the training corrected temperature time sequence feature vector to obtain a training classification feature vector; 187, passing the training classification feature vector through a classifier to obtain a classification loss function value; 188, calculating pseudo-cycle difference penalty factors of the training voltage time sequence feature vector, the training current time sequence feature vector, the training power time sequence feature vector, the training temperature time sequence feature vector, the training corrected voltage time sequence feature vector, the training corrected current time sequence feature vector, the training corrected power time sequence feature vector and the training corrected temperature time sequence feature vector; and 189 training the inter-parameter correlation feature extractor based on the convolutional neural network model, the timing feature extractor based on the one-dimensional convolutional neural network model, and the classifier with a weighted sum of the classification loss function value and the pseudo-cyclic difference penalty factor as a loss function value, and traveling in a direction of gradient descent.

Based on this, the applicant of the present application further introduced for each pre-correction feature vector, e.g. denoted V, in addition to the classification loss function for the classification feature vector _1i And corrected feature vectors, e.g. denoted V _1i The 'pseudo-cyclic difference penalty factor' is expressed as a loss function: calculating pseudo-cycle difference penalty factors of the training voltage time sequence feature vector, the training current time sequence feature vector, the training power time sequence feature vector, the training temperature time sequence feature vector, the training corrected voltage time sequence feature vector, the training corrected current time sequence feature vector, the training corrected power time sequence feature vector and the training corrected temperature time sequence feature vector according to the following optimization formula; wherein, the optimization formula is:

In summary, a method 100 for testing the circuit stability of a corn lamp according to an embodiment of the present application is illustrated, which obtains voltage values, current values, power values, and temperature values at a plurality of predetermined time points within a predetermined time period of the corn lamp to be tested; the intelligent test of the stability of the corn lamp is realized by utilizing the voltage, current, power and temperature data of the corn lamp and combining deep learning and artificial intelligence technology. Therefore, the stability of the corn lamp can be comprehensively evaluated, the objectivity and the accuracy of the test are improved, and the labor cost and the time cost of the test are reduced.

In one embodiment of the application, fig. 6 is a block diagram of a circuit stability testing system for a corn lamp in accordance with an embodiment of the application. As shown in fig. 6, a circuit stability test system 200 of a corn lamp according to an embodiment of the present application includes: a data acquisition module 210, configured to acquire voltage values, current values, power values and temperature values at a plurality of predetermined time points within a predetermined time period of the detected corn light; a vector arrangement module 220, configured to arrange the voltage values, the current values, the power values, and the temperature values at a plurality of predetermined time points within the predetermined time period into a voltage input vector, a current input vector, a power input vector, and a temperature input vector according to a time dimension, respectively; the inter-parameter correlation feature extraction module 230 is configured to arrange the voltage input vector, the current input vector, the power input vector, and the temperature input vector into a two-dimensional input matrix, and then obtain a parameter correlation feature matrix through an inter-parameter correlation feature extractor based on a convolutional neural network model; the time sequence feature extraction module 240 is configured to pass the voltage input vector, the current input vector, the power input vector and the temperature input vector through a time sequence feature extractor based on a one-dimensional convolutional neural network model to obtain a voltage time sequence feature vector, a current time sequence feature vector, a power time sequence feature vector and a temperature time sequence feature vector; the correction module 250 is configured to calculate a matrix product between the voltage timing feature vector, the current timing feature vector, the power timing feature vector, and the temperature timing feature vector, which are used as query feature vectors, respectively, and the parameter correlation feature matrix to obtain a corrected voltage timing feature vector, a corrected current timing feature vector, a corrected power timing feature vector, and a corrected temperature timing feature vector; a cascade module 260, configured to cascade the corrected voltage timing feature vector, the corrected current timing feature vector, the corrected power timing feature vector, and the corrected temperature timing feature vector to obtain a classification feature vector; and a circuit stability result generating module 270, configured to pass the classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether the circuit stability of the detected corn lamp meets a predetermined requirement.

In a specific example, in the circuit stability test system of the corn light, the inter-parameter correlation feature extraction module is configured to: and respectively carrying out convolution processing, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transfer of layers by using each layer of the inter-parameter correlation feature extractor based on the convolutional neural network model, wherein the output of the last layer of the inter-parameter correlation feature extractor based on the convolutional neural network model is taken as the parameter correlation feature matrix, and the input of the first layer of the inter-parameter correlation feature extractor based on the convolutional neural network model is taken as the two-dimensional input matrix.

In a specific example, in the circuit stability test system of the corn light, the timing characteristic extraction module is configured to: and respectively carrying out convolution processing, feature matrix-based averaging pooling processing and nonlinear activation processing on input data in forward transfer of layers by using each layer of the time sequence feature extractor based on the one-dimensional convolutional neural network model, wherein the output of the last layer of the time sequence feature extractor based on the one-dimensional convolutional neural network model is the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector, and the input of the first layer of the time sequence feature extractor based on the one-dimensional convolutional neural network model is the voltage input vector, the current input vector, the power input vector and the temperature input vector.

In a specific example, in the circuit stability test system of the corn light, the correction module is configured to: calculating the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector by using the following product formula as a matrix product between the query feature vector and the parameter association feature matrix to obtain a corrected voltage time sequence feature vector, a corrected current time sequence feature vector, a corrected power time sequence feature vector and a corrected temperature time sequence feature vector; wherein, the product formula is:

wherein M is ₁ Representing the parameter association characteristic matrix, V _ai Representing the voltage timing feature vector or the current timing feature vector or the power timing feature vector or the temperature timing feature vector, V _bi Representing the corrected voltage timing feature vector or the corrected current timing feature vector or the corrected power timing feature vector or the corrected temperature timing feature vector,representing matrix multiplication.

In a specific example, in the circuit stability test system of the corn light, the cascade module is configured to: cascading the corrected voltage time sequence feature vector, the corrected current time sequence feature vector, the corrected power time sequence feature vector and the corrected temperature time sequence feature vector by using the following cascading formula to obtain a classification feature vector; wherein, the cascade formula is:

V _c ＝Concat[V _b1 ，V _b2 ，V _b3 ，V _b4 ]

In a specific example, in the circuit stability test system of the corn lamp, the circuit stability result generating module includes: the coding unit is used for carrying out full-connection coding on the classification characteristic vectors by using a plurality of full-connection layers of the classifier so as to obtain coded classification characteristic vectors; and a classification result unit, configured to pass the encoded classification feature vector through a Softmax classification function of the classifier to obtain the classification result.

In a specific example, in the circuit stability test system of the corn light, the circuit stability test system further comprises a training module for training the inter-parameter correlation feature extractor based on the convolutional neural network model, the time sequence feature extractor based on the one-dimensional convolutional neural network model and the classifier; wherein, training module includes: the system comprises a training data acquisition unit, a detection unit and a control unit, wherein the training data acquisition unit is used for acquiring training data, the training data comprises training voltage values, training current values, training power values and training temperature values of a plurality of preset time points in a preset time period of a detected corn lamp, and whether the circuit stability of the detected corn lamp reaches a true value of a preset requirement or not; the training vector arrangement unit is used for respectively arranging the training voltage values, the training current values, the training power values and the training temperature values of a plurality of preset time points in the preset time period into training voltage input vectors, training current input vectors, training power input vectors and training temperature input vectors according to the time dimension; the training inter-parameter correlation feature extraction unit is used for arranging the training voltage input vector, the training current input vector, the training power input vector and the training temperature input vector into a two-dimensional input matrix and then obtaining a training parameter correlation feature matrix through the inter-parameter correlation feature extractor based on the convolutional neural network model; the training time sequence feature extraction unit is used for enabling the training voltage input vector, the training current input vector, the training power input vector and the training temperature input vector to respectively pass through the time sequence feature extractor based on the one-dimensional convolutional neural network model so as to obtain a training voltage time sequence feature vector, a training current time sequence feature vector, a training power time sequence feature vector and a training temperature time sequence feature vector; the training correction unit is used for respectively taking the training voltage time sequence feature vector, the training current time sequence feature vector, the training power time sequence feature vector and the training temperature time sequence feature vector as query feature vectors, and calculating a matrix product between the query feature vector and the training parameter association feature matrix to obtain a training corrected voltage time sequence feature vector, a training corrected current time sequence feature vector, a training corrected power time sequence feature vector and a training corrected temperature time sequence feature vector; the training cascade unit is used for cascading the training corrected voltage time sequence feature vector, the training corrected current time sequence feature vector, the training corrected power time sequence feature vector and the training corrected temperature time sequence feature vector to obtain a training classification feature vector; the classification loss unit is used for passing the training classification feature vector through a classifier to obtain a classification loss function value; the penalty factor calculation unit is used for calculating pseudo-cycle difference penalty factors of the training voltage time sequence feature vector, the training current time sequence feature vector, the training power time sequence feature vector, the training temperature time sequence feature vector, the training corrected voltage time sequence feature vector, the training corrected current time sequence feature vector, the training corrected power time sequence feature vector and the training corrected temperature time sequence feature vector; and a training unit for training the inter-parameter correlation feature extractor based on the convolutional neural network model, the one-dimensional convolutional neural network model-based timing feature extractor, and the classifier with a weighted sum of the classification loss function value and the pseudo-cyclic difference penalty factor as a loss function value, and traveling in a gradient descent direction.

In a specific example, in the circuit stability test system of the corn light, the penalty factor calculating unit is configured to: calculating pseudo-cycle difference penalty factors of the training voltage time sequence feature vector, the training current time sequence feature vector, the training power time sequence feature vector, the training temperature time sequence feature vector, the training corrected voltage time sequence feature vector, the training corrected current time sequence feature vector, the training corrected power time sequence feature vector and the training corrected temperature time sequence feature vector according to the following optimization formula; wherein, the optimization formula is:

Here, it will be understood by those skilled in the art that the specific functions and operations of the respective units and modules in the above-described circuit stability test system of the corn lamp have been described in detail in the above description of the circuit stability test method of the corn lamp with reference to fig. 1 to 5, and thus, repetitive descriptions thereof will be omitted.

As described above, the circuit stability test system 200 for a corn lamp according to an embodiment of the present application may be implemented in various terminal devices, such as a server for circuit stability test of a corn lamp, and the like. In one example, the circuit stability test system 200 of the corn light according to an embodiment of the present application may be integrated into the terminal device as a software module and/or hardware module. For example, the circuit stability test system 200 of the corn light may be a software module in the operating system of the terminal device or may be an application developed for the terminal device; of course, the corn light circuit stability test system 200 could equally be one of many hardware modules of the terminal device.

Alternatively, in another example, the corn light circuit stability test system 200 and the terminal device may be separate devices, and the corn light circuit stability test system 200 may be connected to the terminal device through a wired and/or wireless network and transmit the interactive information in a agreed data format.

The present application also provides a computer program product comprising instructions which, when executed, cause an apparatus to perform operations corresponding to the above-described method.

In one embodiment of the present application, there is also provided a computer-readable storage medium storing a computer program for executing the above-described method.

It should be appreciated that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the forms of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects may be utilized. Furthermore, the computer program product may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

Methods, systems, and computer program products of embodiments of the present application are described in the flow diagrams and/or block diagrams. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not intended to be limiting, and these advantages, benefits, effects, etc. are not to be considered as essential to the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not necessarily limited to practice with the above described specific details.

The block diagrams of the devices, apparatuses, devices, systems referred to in the present application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.

It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent aspects of the present application.

The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.

The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.

Claims

1. A method for testing the circuit stability of a corn lamp, comprising:

2. The method of claim 1, wherein the step of arranging the voltage input vector, the current input vector, the power input vector, and the temperature input vector into a two-dimensional input matrix and then obtaining a parameter correlation feature matrix by a parameter correlation feature extractor based on a convolutional neural network model comprises the steps of: and respectively carrying out convolution processing, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transfer of layers by using each layer of the inter-parameter correlation feature extractor based on the convolutional neural network model, wherein the output of the last layer of the inter-parameter correlation feature extractor based on the convolutional neural network model is taken as the parameter correlation feature matrix, and the input of the first layer of the inter-parameter correlation feature extractor based on the convolutional neural network model is taken as the two-dimensional input matrix.

3. The method of testing circuit stability of a corn lamp of claim 2, wherein passing the voltage input vector, the current input vector, the power input vector, and the temperature input vector through a one-dimensional convolutional neural network model-based time series feature extractor to obtain a voltage time series feature vector, a current time series feature vector, a power time series feature vector, and a temperature time series feature vector, respectively, comprises: and respectively carrying out convolution processing, feature matrix-based averaging pooling processing and nonlinear activation processing on input data in forward transfer of layers by using each layer of the time sequence feature extractor based on the one-dimensional convolutional neural network model, wherein the output of the last layer of the time sequence feature extractor based on the one-dimensional convolutional neural network model is the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector, and the input of the first layer of the time sequence feature extractor based on the one-dimensional convolutional neural network model is the voltage input vector, the current input vector, the power input vector and the temperature input vector.

4. The method of testing circuit stability of a corn lamp according to claim 3, wherein calculating a matrix product between the voltage timing eigenvector, the current timing eigenvector, the power timing eigenvector, and the temperature timing eigenvector as query eigenvectors to obtain a corrected voltage timing eigenvector, a corrected current timing eigenvector, a corrected power timing eigenvector, and a corrected temperature timing eigenvector, respectively, comprises: calculating the voltage time sequence feature vector, the current time sequence feature vector, the power time sequence feature vector and the temperature time sequence feature vector by using the following product formula as a matrix product between the query feature vector and the parameter association feature matrix to obtain a corrected voltage time sequence feature vector, a corrected current time sequence feature vector, a corrected power time sequence feature vector and a corrected temperature time sequence feature vector;

wherein, the product formula is:

5. The method of claim 4, wherein concatenating the corrected voltage timing feature vector, corrected current timing feature vector, corrected power timing feature vector, and corrected temperature timing feature vector to obtain a classification feature vector, comprises: cascading the corrected voltage time sequence feature vector, the corrected current time sequence feature vector, the corrected power time sequence feature vector and the corrected temperature time sequence feature vector by using the following cascading formula to obtain a classification feature vector;

wherein, the cascade formula is:

V _c ＝Concat[V _b1 ，V _b2 ，V _b3 ，V _b4 ]

6. The method of claim 5, wherein passing the classification feature vector through a classifier to obtain a classification result, the classification result being used to indicate whether the circuit stability of the detected corn lamp meets a predetermined requirement, comprising:

Performing full-connection coding on the classification feature vectors by using a plurality of full-connection layers of the classifier to obtain coded classification feature vectors; and

and the coding classification feature vector is passed through a Softmax classification function of the classifier to obtain the classification result.

7. The method of testing the circuit stability of a corn lamp of claim 6, further comprising the training step of: training the inter-parameter correlation feature extractor based on the convolutional neural network model, the time sequence feature extractor based on the one-dimensional convolutional neural network model and the classifier;

wherein the training step comprises:

acquiring training data, wherein the training data comprises training voltage values, training current values, training power values and training temperature values of a plurality of preset time points in a preset time period of a detected corn lamp, and whether the circuit stability of the detected corn lamp reaches a true value of a preset requirement;

respectively arranging the training voltage values, the training current values, the training power values and the training temperature values of a plurality of preset time points in the preset time period into training voltage input vectors, training current input vectors, training power input vectors and training temperature input vectors according to time dimensions;

Arranging the training voltage input vector, the training current input vector, the training power input vector and the training temperature input vector into a two-dimensional input matrix, and then obtaining a training parameter association feature matrix through the inter-parameter association feature extractor based on the convolutional neural network model;

respectively passing the training voltage input vector, the training current input vector, the training power input vector and the training temperature input vector through the time sequence feature extractor based on the one-dimensional convolutional neural network model to obtain a training voltage time sequence feature vector, a training current time sequence feature vector, a training power time sequence feature vector and a training temperature time sequence feature vector;

respectively taking the training voltage time sequence feature vector, the training current time sequence feature vector, the training power time sequence feature vector and the training temperature time sequence feature vector as query feature vectors, and calculating a matrix product between the query feature vectors and the training parameter association feature matrix to obtain a training corrected voltage time sequence feature vector, a training corrected current time sequence feature vector, a training corrected power time sequence feature vector and a training corrected temperature time sequence feature vector;

Cascading the training corrected voltage time sequence feature vector, the training corrected current time sequence feature vector, the training corrected power time sequence feature vector and the training corrected temperature time sequence feature vector to obtain a training classification feature vector;

the training classification feature vector passes through a classifier to obtain a classification loss function value;

calculating pseudo-cycle difference penalty factors of the training voltage time sequence feature vector, the training current time sequence feature vector, the training power time sequence feature vector, the training temperature time sequence feature vector, the training corrected voltage time sequence feature vector, the training corrected current time sequence feature vector, the training corrected power time sequence feature vector and the training corrected temperature time sequence feature vector; and

and training the inter-parameter correlation feature extractor based on the convolutional neural network model, the time sequence feature extractor based on the one-dimensional convolutional neural network model and the classifier by taking the weighted sum of the classification loss function value and the pseudo-cyclic difference penalty factor as the loss function value and transmitting the weighted sum in the gradient descending direction.

8. The method of testing circuit stability of a corn lamp of claim 7, wherein calculating a pseudo-cyclic difference penalty factor for the training voltage timing feature vector, the training current timing feature vector, the training power timing feature vector, and the training temperature timing feature vector with the training corrected voltage timing feature vector, the training corrected current timing feature vector, the training corrected power timing feature vector, and the training corrected temperature timing feature vector comprises: calculating pseudo-cycle difference penalty factors of the training voltage time sequence feature vector, the training current time sequence feature vector, the training power time sequence feature vector, the training temperature time sequence feature vector, the training corrected voltage time sequence feature vector, the training corrected current time sequence feature vector, the training corrected power time sequence feature vector and the training corrected temperature time sequence feature vector according to the following optimization formula;

Wherein, the optimization formula is:

wherein V is _1i Representing the training voltage timing feature vector, the training current timing feature vector, the training power timing feature vector, andone of the training temperature time sequence characteristic vectors, V _1i ' represents one of the training corrected voltage timing feature vector, the training corrected current timing feature vector, the training corrected power timing feature vector, and the training corrected temperature timing feature vector, D (V _1i ,V _1i ')) is a feature vector V _1i And V _1i Distance matrix between' i|·|| _F The Frobenius norm of the matrix, L is the length of the eigenvector, d (V _1i ,V _1i ' is the feature vector V _1i And V _1i ' Euclidean distance between |·|| ₂ Is the two norms of the vector, log represents the base 2 logarithm, and alpha and beta are weighted hyper-parameters,representing the pseudo-cyclic difference penalty factor.

9. A circuit stability testing system for a corn lamp, comprising:

10. The circuit stability testing system of claim 9, wherein the inter-parameter correlation feature extraction module is configured to: and respectively carrying out convolution processing, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transfer of layers by using each layer of the inter-parameter correlation feature extractor based on the convolutional neural network model, wherein the output of the last layer of the inter-parameter correlation feature extractor based on the convolutional neural network model is taken as the parameter correlation feature matrix, and the input of the first layer of the inter-parameter correlation feature extractor based on the convolutional neural network model is taken as the two-dimensional input matrix.