Intelligent evaluation method and device for multi-modal test data

Abstract

The invention discloses an intelligent evaluation method and device for multi-modal test data. The method comprises the following steps: acquiring a multi-modal test data set in an equipment performance test process; the multi-modal test data set comprises a multi-source data subset of each technical index measured in an equipment performance test process; preprocessing the multi-modal test data set to obtain a preprocessed data set; performing fusion evaluation on the multi-source data subset of each technical index in the preprocessed data set to obtain an evaluation value of each technical index; and performing comprehensive performance evaluation processing on the evaluation values of all the technical indexes to obtain a test evaluation value of the equipment performance. According to the intelligent evaluation method, weight distribution can be flexibly adjusted according to the characteristics of different technical indexes, so that the accuracy and reliability of an evaluation result are remarkably improved, and powerful support is provided for comprehensive evaluation of equipment performance.

Description

Technical Field

[0001] The present invention relates to the fields of industrial data processing and equipment system effectiveness evaluation technology, and in particular to an intelligent evaluation method and device for multimodal test data. Background Art

[0002] In the field of equipment performance testing, with the continuous development of technology and the increase in test complexity, the data generated during the test process has shown multimodal, multi-source and heterogeneous characteristics. Traditional test data evaluation methods mainly rely on the analysis of single-modal data (such as numerical data), ignoring the rich information contained in other modal data such as images and text. This single-modal data evaluation method has many limitations. For example, it is difficult to fully reflect the performance of equipment under complex working conditions by relying solely on numerical data, while image data and text data (such as records and descriptions of test personnel) often contain key fault information and abnormal conditions, but this information is often ignored in traditional evaluation methods. In addition, existing methods lack systematic processing of multi-source heterogeneous data in the data preprocessing stage, resulting in problems such as data noise, inconsistent formats and abnormal data that seriously affect the accuracy of the evaluation results.

[0003] Therefore, how to effectively integrate multimodal test data and achieve a comprehensive and accurate evaluation of equipment performance through intelligent evaluation methods is a technical problem that needs to be urgently solved in the current field of equipment performance testing. Summary of the Invention

[0004] The present invention mainly solves the problem of how to effectively integrate multimodal test data and realize comprehensive and accurate evaluation of equipment performance through intelligent evaluation methods. The present invention discloses an intelligent evaluation method and device for multimodal test data.

[0005] In a first aspect, an embodiment of the present invention discloses an intelligent evaluation method for multimodal test data, comprising:

[0006] S1, collecting a multimodal test data set during an equipment performance test; the multimodal test data set includes a multi-source data subset of each technical indicator measured during the equipment performance test; the multi-source data subset includes an image data sequence, a text data sequence, and a test data sequence;

[0007] S2, preprocessing the multimodal test dataset to obtain a preprocessed dataset;

[0008] S3, performing fusion evaluation on the multi-source data subsets of each technical indicator in the preprocessed data set to obtain an evaluation value of each technical indicator;

[0009] S4, conduct a comprehensive performance evaluation on the evaluation values of all technical indicators to obtain the test evaluation value of the equipment performance.

[0010] The preprocessing of the multi-source heterogeneous data set to obtain a preprocessed data set includes:

[0011] S21, performing data cleaning processing on the multi-source heterogeneous data set to obtain a first data set;

[0012] S22, performing format detection processing on the first data set to obtain a second data set;

[0013] S23: Perform data pattern matching detection processing on the second data set to obtain a preprocessed data set.

[0014] The performing data pattern matching detection processing on the second data set to obtain a preprocessed data set includes:

[0015] S231, for each data attribute of the second data set, perform autoregressive-sliding average modeling with the data collection information of the data as the independent variable and the data value of the data as the dependent variable to obtain a regression model for each data attribute of the class;

[0016] S232, using the regression model, performing calculation processing on the independent variable to obtain a regression data value; determining whether the absolute value of the difference between the regression data value and the corresponding dependent variable value is greater than a set first regression discrimination threshold; if the absolute value is greater than the first regression discrimination threshold, deleting the data from the second data set; if the absolute value is less than or equal to the first regression discrimination threshold, not processing the data;

[0017] S233: Perform fusion processing on all data of the second data set after the data pattern matching detection processing is performed to obtain a preprocessed data set.

[0018] The fusion evaluation of each multi-source data subset of each technical indicator in the pre-processed data set is performed separately to obtain an evaluation value of each technical indicator, including:

[0019] S31, using the training data set corresponding to each technical indicator, training a preset evaluation network to obtain an evaluation network corresponding to each technical indicator;

[0020] S32, inputting the image data sequence in the multi-source data subset of each technical indicator in the pre-processed data set into the evaluation network corresponding to the technical indicator to obtain the corresponding image evaluation value;

[0021] S33, performing text evaluation processing on the text data sequence in the multi-source data subset of each technical indicator in the pre-processed data set to obtain a corresponding text evaluation value;

[0022] S34, performing numerical evaluation processing on the test data sequence in the multi-source data subset for each technical indicator in the preprocessed data set to obtain a corresponding numerical evaluation value;

[0023] S35 , performing weighted sum processing on the image evaluation value, text evaluation value, and numerical evaluation value corresponding to each technical indicator to obtain the evaluation value of each technical indicator.

[0024] The evaluation network includes a first processing unit, a second processing unit and a fully connected unit;

[0025] The output end of the first processing unit is connected to the input end of the second processing unit; the output end of the second processing unit is connected to the input end of the fully connected layer;

[0026] The input of the first processing unit is a picture data sequence;

[0027] The first processing unit includes a convolutional layer, a pooling layer and a fully connected layer;

[0028] The convolution layer uses a two-dimensional convolution kernel w∈R 3×3 Perform feature extraction on the image data sequence to obtain the feature matrix C n ;

[0029] The pooling layer uses the maximum pooling method to extract features. The calculation expression of the maximum pooling method is:

[0030] p u =Max 2×2 [C n ],

[0031] Among them, u represents the number of pooling, Max 2×2 Represents the maximum pooling operation method of a 2×2 matrix, p u is the extracted features;

[0032] The fully connected layer, for the feature p u Performing dimensionality transformation processing to obtain a variable that matches the input dimension of the second processing unit; the output end of the fully connected layer is connected to the input end of the second processing unit;

[0033] The second processing unit is configured to perform time series attribute extraction processing on the input data to obtain an output value, and input the output value into the fully connected unit;

[0034] The fully connected unit is used to calculate the output value of the second processing unit to obtain a picture evaluation value.

[0035] The comprehensive performance evaluation of all technical indicators is performed to obtain the test evaluation value of equipment performance, including:

[0036] S41, obtaining a standard evaluation value of each technical indicator;

[0037] S42, performing differential fusion calculation processing on the evaluation values of all technical indicators and the corresponding standard evaluation values to obtain the test evaluation value of the equipment performance.

[0038] The expression of the difference fusion calculation process is:

[0039]

[0040] Among them, qe is the test evaluation value of equipment performance, q i is the evaluation value of the i-th technical indicator, e i is the standard evaluation value corresponding to the i-th technical indicator, and N is the number of technical indicators.

[0041] According to a second aspect of the present invention, an intelligent evaluation device for multimodal test data is disclosed, comprising:

[0042] a memory storing executable program code;

[0043] a processor coupled to the memory;

[0044] The processor calls the executable program code stored in the memory to execute the intelligent evaluation method for multimodal test data.

[0045] According to a third aspect of the present invention, a computer-storable medium is disclosed, wherein the computer-storable medium stores computer instructions, and when the computer instructions are called by a computer, the computer-storage medium is used to execute the intelligent evaluation method for multimodal test data.

[0046] According to a fourth aspect of the present invention, an information data processing terminal is disclosed. The information data processing terminal is used to implement the intelligent evaluation method for multimodal test data.

[0047] The beneficial effects of the present invention are:

[0048] The multimodal test data intelligent evaluation method of the present invention collects multimodal data sets during equipment performance testing, including pictures, text and numerical data, and performs systematic preprocessing and fusion evaluation on them, effectively solving the problems of single data modality and insufficient preprocessing in the existing technology.

[0049] For each multi-source data subset of each technical indicator, the present invention uses evaluation networks, text evaluation, and numerical evaluation methods for processing, and obtains a comprehensive evaluation value through weighted summation, thus achieving deep fusion and complementarity of multimodal data. This fusion evaluation method not only fully utilizes the intuitiveness of image data, the descriptiveness of text data, and the accuracy of numerical data, but also can flexibly adjust the weight distribution according to the characteristics of different technical indicators, thereby significantly improving the accuracy and reliability of the evaluation results. Finally, by comprehensively processing the evaluation values of all technical indicators, the overall test evaluation value of the equipment performance is obtained, which provides strong support for the comprehensive evaluation of equipment performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 4 is an implementation flow chart of the method of the present invention. DETAILED DESCRIPTION

[0051] In order to better understand the content of the present invention, an embodiment is given here.

[0052] Figure 1 4 is an implementation flow chart of the method of the present invention.

[0053] In a first aspect, an embodiment of the present invention discloses an intelligent evaluation method for multimodal test data, comprising:

[0054] S1, collecting a multimodal test data set during an equipment performance test; the multimodal test data set includes a multi-source data subset of each technical indicator measured during the equipment performance test; the multi-source data subset includes an image data sequence, a text data sequence, and a test data sequence;

[0055] S2, preprocessing the multimodal test dataset to obtain a preprocessed dataset;

[0056] S3, performing fusion evaluation on the multi-source data subsets of each technical indicator in the preprocessed data set to obtain an evaluation value of each technical indicator;

[0057] S4, conduct a comprehensive performance evaluation on the evaluation values of all technical indicators to obtain the test evaluation value of the equipment performance.

[0058] The preprocessing of the multi-source heterogeneous data set to obtain a preprocessed data set includes:

[0059] Performing data cleaning on the multi-source heterogeneous data set to obtain a first data set;

[0060] Performing format detection processing on the first data set to obtain a second data set;

[0061] performing data pattern matching detection processing on the second data set to obtain a preprocessed data set;

[0062] The performing data pattern matching detection processing on the second data set to obtain a preprocessed data set includes:

[0063] For each type of data attribute of the second data set, autoregressive-sliding average modeling is performed with the data collection information of the data as the independent variable and the data value of the data as the dependent variable to obtain a regression model for each type of data attribute;

[0064] Using the regression model, calculating and processing the independent variable to obtain a regression data value; determining whether the absolute value of the difference between the regression data value and the corresponding dependent variable value is greater than a set first regression discrimination threshold; if the absolute value is greater than the first regression discrimination threshold, deleting the data from the second data set; if the absolute value is less than or equal to the first regression discrimination threshold, not processing the data;

[0065] performing fusion processing on all data of the second data set after the data pattern matching detection processing is performed to obtain a preprocessed data set;

[0066] The present invention can effectively remove noise data and abnormal data through multi-step preprocessing of data cleaning, format detection and data pattern matching detection, thereby ensuring the quality and consistency of input data.

[0067] The data cleaning process includes filling missing values, smoothing noise data, and smoothing or deleting outliers. Smoothing noise data involves first identifying noise data and then smoothing it based on the preceding and following data. Noise data is defined as values that are less than the sensor's sensitivity or greater than the sensor's upper limit. Kalman filtering can be used to identify outliers. The value to fill missing values can be determined by averaging the measured values within a certain sampling interval before and after the missing value.

[0068] The format detection process includes: detecting whether the data format of each data attribute in the first data set is consistent with a preset format, and deleting data with inconsistent formats from the first data set.

[0069] The fusion evaluation of each multi-source data subset of each technical indicator in the pre-processed data set is performed separately to obtain an evaluation value of each technical indicator, including:

[0070] S31, using the training data set corresponding to each technical indicator, training a preset evaluation network to obtain an evaluation network corresponding to each technical indicator;

[0071] S32, inputting the image data sequence in the multi-source data subset of each technical indicator in the pre-processed data set into the evaluation network corresponding to the technical indicator to obtain the corresponding image evaluation value;

[0072] S33, performing text evaluation processing on the text data sequence in the multi-source data subset of each technical indicator in the pre-processed data set to obtain a corresponding text evaluation value;

[0073] S34, performing numerical evaluation processing on the test data sequence in the multi-source data subset for each technical indicator in the preprocessed data set to obtain a corresponding numerical evaluation value;

[0074] S35, performing weighted sum processing on the image evaluation value, text evaluation value, and numerical evaluation value corresponding to each technical indicator to obtain an evaluation value of each technical indicator;

[0075] The evaluation network includes a first processing unit, a second processing unit and a fully connected unit;

[0076] The output end of the first processing unit is connected to the input end of the second processing unit; the output end of the second processing unit is connected to the input end of the fully connected layer;

[0077] The input of the first processing unit is a picture data sequence;

[0078] The first processing unit includes a convolutional layer, a pooling layer and a fully connected layer;

[0079] The convolution layer uses a two-dimensional convolution kernel w∈R 3×3 Perform feature extraction on the image data sequence to obtain the feature matrix C n ;

[0080] The pooling layer uses the maximum pooling method to extract features. The calculation expression of the maximum pooling method is:

[0081] p u =Max 2×2 [C n ],

[0082] Among them, u represents the number of pooling, Max 2×2 Represents the maximum pooling operation method of a 2×2 matrix, p u is the extracted features;

[0083] The fully connected layer, for the feature p u Performing dimensionality transformation processing to obtain a variable that matches the input dimension of the second processing unit; the output end of the fully connected layer is connected to the input end of the second processing unit;

[0084] The second processing unit is configured to perform time series attribute extraction processing on the input data to obtain an output value, and input the output value into the fully connected unit;

[0085] The fully connected unit is configured to calculate the output value of the second processing unit to obtain an image evaluation value;

[0086] The convolution layer is a cascaded structure of 13 convolution modules; the maximum pooling layer is a cascaded structure of 5 maximum pooling modules; the fully connected layer is a cascaded structure of 3 fully connected modules; the convolution layer is connected to the pooling layer, and the pooling layer is connected to the fully connected layer;

[0087] The calculation expression of the convolutional layer is:

[0088]

[0089] Among them, n represents the number of convolution operations, m represents the number of convolution kernels, and p i represents the acquired i-th feature matrix, f represents the nonlinear activation function, · represents the corresponding operation of the shared weight of the convolution kernel and the feature matrix, w represents the weight of the convolution kernel, b represents the bias value, R 3×3 Represents a 3×3 real matrix;

[0090] The dimensionality transformation process can be implemented using the Reshape function.

[0091] The fully connected unit can be implemented using an FC layer.

[0092] The processing process of the second processing unit is expressed as:

[0093] o t =g(V st +V′ sT′ ),

[0094]

[0095] Among them, s t represents the output of the second processing unit at time t, s′ t It represents the output of the reverse sequential input of the second processing unit at time t, U Xt Represents the initial input of the second processing unit in the forward direction, U′ Xt represents the initial input quantity of the reverse sequence of the second processing unit, Indicates the positive time series input quantity of the second processing unit at the previous moment, is the reverse time series input quantity at the next moment, o trepresents the standard output of the second processing unit at time t, g and f are the activation function and sigmoid function respectively.

[0096] The output of the second processing unit includes s t , s′ t 、o t .

[0097] The input of the forward time series is the extracted feature p u ;

[0098] The input of the reverse time series is the extracted feature p u The flashback arrangement;

[0099] The second processing unit can be implemented by using a bidirectional LSTM network.

[0100] The comprehensive performance evaluation of all technical indicators is performed to obtain the test evaluation value of equipment performance, including:

[0101] S41, obtaining a standard evaluation value of each technical indicator;

[0102] S42, performing differential fusion calculation processing on the evaluation values of all technical indicators and the corresponding standard evaluation values to obtain the test evaluation value of the equipment performance.

[0103] The expression of the difference fusion calculation process is:

[0104]

[0105] Among them, qe is the test evaluation value of equipment performance, q i is the evaluation value of the i-th technical indicator, e i is the standard evaluation value corresponding to the i-th technical indicator, and N is the number of technical indicators.

[0106] The difference fusion calculation process described above can comprehensively and accurately reflect the overall performance of the equipment by performing a differential fusion calculation on the evaluation values of all technical indicators and the standard evaluation values. This expression quantitatively weights and integrates the differences between each technical indicator and the standard value, avoiding the one-sidedness of single indicator evaluation and fully considering the multi-dimensional characteristics of equipment performance. At the same time, the calculation process based on this expression is simple and efficient, and can quickly process large amounts of technical indicator data. After the equipment performance test is completed, accurate test evaluation values can be generated in a timely manner, providing a reliable basis for equipment optimization and improvement, quality inspection, and acceptance decision-making, greatly improving the scientific nature and practicality of equipment performance evaluation.

[0107] The text evaluation process includes:

[0108] Get standard text data sequence;

[0109] For the text data sequence and the standard text data sequence, autoregressive-sliding average modeling is performed respectively to obtain the first regression model and the second regression model;

[0110] Extract the coefficient vectors of the two regression models, calculate their cross-correlation matrix, and calculate the maximum eigenvalue of the cross-correlation matrix;

[0111] Performing evaluation and calculation processing on the maximum eigenvalue to obtain a text evaluation value;

[0112] The expression for the evaluation calculation process is:

[0113]

[0114] Among them, pg is the text evaluation value, M is the length of the text data sequence, and f max is the maximum eigenvalue, a i is the i-th item of the text data sequence, la i is the i-th item in the standard text data sequence.

[0115] This expression extracts the features of text data sequences and standard text data sequences through autoregressive-sliding average modeling, and constructs evaluation calculation logic based on feature coefficients. It not only captures the differences between text data sequences and standard sequences in various aspects of content in detail, but also calculates the relative degree of difference. It also combines the maximum eigenvalue of the regression model features and uses an exponential function to adjust the weights of the differences, highlighting the impact of important differences, thereby achieving an accurate evaluation of the quality and accuracy of the text data sequence. In the evaluation scenario of text data involved in equipment performance testing (such as test reports, operating instructions, etc.), this expression can effectively determine whether the text data meets the standard specifications, providing a scientific and effective quantitative means for the reliability evaluation of text information during equipment testing, and improving the accuracy and comprehensiveness of text evaluation in multimodal test data.

[0116] The numerical evaluation process comprises:

[0117] Obtain the standard measurement values corresponding to technical indicators;

[0118] Performing calculations on the test data sequence and the standard measurement value to obtain a numerical evaluation value;

[0119] The calculation expression of the numerical evaluation value is:

[0120]

[0121] Wherein, P is the length of the test data sequence, R is the numerical evaluation value, and p iis the i-th element of the test data sequence, and p0 is the standard measurement value.

[0122] The training data set corresponding to each technical indicator can be constructed based on historical image data and indicator evaluation values during the equipment test process.

[0123] The weight values of the weighted summation in S35 may be 0.2, 0.4, and 0.3 for the image evaluation value, the text evaluation value, and the numerical evaluation value, respectively.

[0124] The data attributes include images, text, and test data;

[0125] The data collection information is collection time information.

[0126] The text data sequence is obtained by converting the collected text into a data vector, which is achieved by using the word2vec function.

[0127] The method of using the training data set corresponding to each technical indicator to train the preset evaluation network to obtain the evaluation network corresponding to each technical indicator includes:

[0128] Obtaining a training data set corresponding to the technical indicators; the training data set includes training data and corresponding label information; the label information is an image evaluation value; the training data is a picture data sequence;

[0129] Initialize the number of training iterations;

[0130] Input the training data in the training data set as input data into the evaluation network;

[0131] Using the evaluation network, processing the input data to obtain a predicted value;

[0132] Performing a difference calculation process on the obtained predicted value and the label information corresponding to the input data to obtain a difference value;

[0133] Determine whether the difference value meets the convergence condition, and obtain a first determination result;

[0134] When the first judgment result is no, determining whether the number of training iterations is equal to a training number threshold, and obtaining a second judgment result;

[0135] When the second judgment result is no, determining that the model training state does not meet the training termination condition;

[0136] When the second judgment result is yes, determining that the model training state satisfies the training termination condition;

[0137] When the first judgment result is yes, determining that the model training state satisfies the training termination condition;

[0138] When the model training status does not meet the training termination condition, the evaluation network parameters are updated using the parameter update model, the number of training iterations is increased by 1, and the training data in the training data set is triggered to be input into the evaluation network as input data;

[0139] When the model training state satisfies the training termination condition, the training process of the evaluation network is completed to obtain a trained evaluation network.

[0140] The difference value satisfies the convergence condition, which means that the difference value is less than a preset convergence threshold; the difference value does not satisfy the convergence condition, which means that the difference value is not less than the preset convergence threshold.

[0141] The difference calculation process can be implemented using a loss function.

[0142] The loss function may be a cross entropy loss function.

[0143] The parameter update model is:

[0144]

[0145] θ←θ+v;

[0146] Where, is the difference value calculated for the i-th training data in the training data set, v is the parameter update value, θ is the parameter of the evaluation network, η is the initial parameter learning rate, α is the momentum angle parameter, 0≤α≤π / 4, Indicates partial derivative with respect to variable θ;

[0147] The comprehensive performance evaluation of all technical indicators is performed to obtain the test evaluation value of equipment performance, including:

[0148] Get the standard evaluation value of each technical indicator;

[0149] Perform comprehensive performance evaluation calculation on the evaluation values and standard evaluation values of all technical indicators to obtain the test evaluation value of equipment performance;

[0150] The expression for the comprehensive performance evaluation calculation is:

[0151]

[0152] Among them, zi is the evaluation value of the i-th technical indicator, bzi is the standard evaluation value of the i-th technical indicator, K is the number of technical indicators, and S is the test evaluation value of equipment performance.

[0153] According to a second aspect of the present invention, an intelligent evaluation device for multimodal test data is disclosed, comprising:

[0154] a memory storing executable program code;

[0155] a processor coupled to the memory;

[0156] The processor calls the executable program code stored in the memory to execute the intelligent evaluation method for multimodal test data.

[0157] According to a third aspect of the present invention, a computer-storable medium is disclosed, wherein the computer-storable medium stores computer instructions, and when the computer instructions are called by a computer, the computer-storage medium is used to execute the intelligent evaluation method for multimodal test data.

[0158] According to a fourth aspect of the present invention, an information data processing terminal is disclosed. The information data processing terminal is used to implement the intelligent evaluation method for multimodal test data.

[0159] The foregoing is merely an embodiment of the present invention and is not intended to limit the present invention. It will be apparent to those skilled in the art that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention are intended to be included within the scope of the claims of the present invention.

Claims

1. An intelligent evaluation method for multimodal test data, characterized in that: include: S1, collects the multimodal test data set during the equipment performance test; The multimodal test data set includes a multi-source data subset of each technical indicator measured during the equipment performance test; the multi-source data subset includes an image data sequence, a text data sequence, and a test data sequence; S2, preprocessing the multimodal test dataset to obtain a preprocessed dataset; S3, performing fusion evaluation on the multi-source data subsets of each technical indicator in the preprocessed data set to obtain an evaluation value of each technical indicator; S4, conduct a comprehensive performance evaluation on the evaluation values of all technical indicators to obtain the test evaluation value of the equipment performance.

2. The intelligent evaluation method for multimodal test data according to claim 1, wherein: The preprocessing of the multi-source heterogeneous data set to obtain a preprocessed data set includes: S21, performing data cleaning processing on the multi-source heterogeneous data set to obtain a first data set; S22, performing format detection processing on the first data set to obtain a second data set; S23: Perform data pattern matching detection processing on the second data set to obtain a preprocessed data set.

3. The intelligent evaluation method for multimodal test data according to claim 2, wherein: The performing data pattern matching detection processing on the second data set to obtain a preprocessed data set includes: S231, for each data attribute of the second data set, perform autoregressive-sliding average modeling with the data collection information of the data as the independent variable and the data value of the data as the dependent variable to obtain a regression model for each data attribute of the class; S232, using the regression model, performing calculation processing on the independent variable to obtain a regression data value; determining whether the absolute value of the difference between the regression data value and the corresponding dependent variable value is greater than a set first regression discrimination threshold; if the absolute value is greater than the first regression discrimination threshold, deleting the data from the second data set; if the absolute value is less than or equal to the first regression discrimination threshold, not processing the data; S233: Perform fusion processing on all data of the second data set after the data pattern matching detection processing is performed to obtain a preprocessed data set.

4. The intelligent evaluation method for multimodal test data according to claim 1, wherein: The fusion evaluation of each multi-source data subset of each technical indicator in the pre-processed data set is performed separately to obtain an evaluation value of each technical indicator, including: S31, using the training data set corresponding to each technical indicator, training a preset evaluation network to obtain an evaluation network corresponding to each technical indicator; S32, inputting the image data sequence in the multi-source data subset of each technical indicator in the pre-processed data set into the evaluation network corresponding to the technical indicator to obtain the corresponding image evaluation value; S33, performing text evaluation processing on the text data sequence in the multi-source data subset of each technical indicator in the pre-processed data set to obtain a corresponding text evaluation value; S34, performing numerical evaluation processing on the test data sequence in the multi-source data subset for each technical indicator in the preprocessed data set to obtain a corresponding numerical evaluation value; S35 , performing weighted sum processing on the image evaluation value, text evaluation value, and numerical evaluation value corresponding to each technical indicator to obtain the evaluation value of each technical indicator.

5. The intelligent evaluation method for multimodal test data according to claim 4, characterized in that: The evaluation network includes a first processing unit, a second processing unit and a fully connected unit; The output end of the first processing unit is connected to the input end of the second processing unit; the output end of the second processing unit is connected to the input end of the fully connected layer; The input of the first processing unit is a picture data sequence; The first processing unit includes a convolutional layer, a pooling layer and a fully connected layer; The convolution layer uses a two-dimensional convolution kernel w∈R 3×3 Perform feature extraction on the image data sequence to obtain the feature matrix C n ; The pooling layer uses the maximum pooling method to extract features. The calculation expression of the maximum pooling method is: p u =Max 2×2 [C n ], Among them, u represents the number of pooling, Max 2×2 Represents the maximum pooling operation method of a 2×2 matrix, p u is the extracted features; The fully connected layer, for the feature p u Performing dimensionality transformation processing to obtain a variable that matches the input dimension of the second processing unit; the output end of the fully connected layer is connected to the input end of the second processing unit; The second processing unit is configured to perform time series attribute extraction processing on the input data to obtain an output value, and input the output value into the fully connected unit; The fully connected unit is used to calculate the output value of the second processing unit to obtain a picture evaluation value.

6. The intelligent evaluation method for multimodal test data according to claim 4, wherein: The comprehensive performance evaluation of all technical indicators is performed to obtain the test evaluation value of equipment performance, including: S41, obtaining a standard evaluation value of each technical indicator; S42, performing differential fusion calculation processing on the evaluation values of all technical indicators and the corresponding standard evaluation values to obtain the test evaluation value of the equipment performance.

7. The intelligent evaluation method for multimodal test data according to claim 6, characterized in that: The expression of the difference fusion calculation process is: Among them, qe is the test evaluation value of equipment performance, q i is the evaluation value of the i-th technical indicator, e i is the standard evaluation value corresponding to the i-th technical indicator, and N is the number of technical indicators.

8. An intelligent evaluation device for multimodal test data, characterized in that: The device comprises: a memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the intelligent evaluation method for multimodal test data according to any one of claims 1 to 7.

9. A computer storable medium, characterized in that The computer storable medium stores computer instructions, and when the computer instructions are called by a computer, they are used to execute the intelligent evaluation method for multimodal test data according to any one of claims 1 to 7.

10. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the intelligent evaluation method for multimodal test data according to any one of claims 1 to 7.

Images