Engineering data security management method and system based on real-time monitoring

Abstract

The invention relates to the technical field of data management, and discloses an engineering data security management method and system based on real-time monitoring. According to the method, engineering historical abnormal data are collected, a training set is constructed, the engineering historical abnormal data in the training set are processed, and feature extraction and storage are performed on the processed engineering historical abnormal data through a convolutional neural network. Comparing the obtained characteristics of the real-time collected engineering data with the characteristics of the historical abnormal data of the engineering, filtering the real-time collected engineering data according to a comparison result, and classifying the filtered engineering data through a data clustering mode based on the obtained filtered engineering data, so as to obtain the classified engineering data; and the classified engineering data is managed, so that the safety management accuracy of the engineering data is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to a method and system for engineering data security management based on real-time monitoring. Background Art

[0002] Monitoring configuration software is an important part of industrial software and a tool platform for industrial data. In recent years, some technologies closely related to monitoring configuration software, such as OPC UA, fieldbus, industrial Internet of Things, edge computing, cloud computing, 5G and other technologies, have achieved rapid development. However, the real-time management capability and security energy of monitoring configuration software are insufficient to meet the management and monitoring of engineering manufacturing data.

[0003] The existing Chinese patent CN116957404A, this method parses the collected cost documents, settlement business reports and electronic engineering documents of the power transmission and transformation project, and obtains the engineering information and basic cost indicators of the power transmission and transformation project through parsing; at the same time, based on the engineering type of the power transmission and transformation project and the basic cost indicators, the index data of the power transmission and transformation project is extracted, and then the index data that passes the data verification is verified, and the corresponding index analysis results are obtained, and an analysis report is generated. However, since this method only verifies the index data, it can only ensure that the index data is not missing, and cannot identify whether the index data is valid, which has certain limitations. Summary of the invention

[0004] 1. Technical issues to be resolved In view of the shortcomings of the prior art, the present invention provides a method and system for engineering data security management based on real-time monitoring, which has the advantages of security, real-time and accuracy, and solves the problem that the real-time management capability and security energy of the monitoring configuration software are insufficient, and cannot meet the management and monitoring of engineering manufacturing data.

[0005] In order to solve the technical problem that the above monitoring configuration software has insufficient real-time management capability and safety energy and cannot meet the management and monitoring of engineering manufacturing data, the present invention provides the following technical solutions: The present invention discloses a method for engineering data security management based on real-time monitoring, which specifically comprises the following steps:

[0006] (II) Technical solution

[0007] S1. Collect engineering history abnormal data, construct a training set, and process the engineering history abnormal data in the training set to obtain processed engineering history abnormal data; S2. Extracting features of the processed engineering history abnormal data through a convolutional neural network to obtain and save features of the engineering history abnormal data; S3, collecting engineering data in real time, and inputting the engineering data collected in real time into a convolutional neural network to obtain features of the engineering data collected in real time, comparing the features of the engineering data collected in real time with features of historical abnormal engineering data, filtering the engineering data collected in real time according to the comparison results, and obtaining filtered engineering data; S4. Based on the obtained filtered engineering data, classify the filtered engineering data by data clustering to obtain classified engineering data, wherein the classification results include: engineering data type and engineering data of each classification; S5. Manage the classified engineering data.

[0008] The present invention collects engineering history abnormal data, constructs a training set and processes the engineering history abnormal data in the training set, and extracts and saves features of the processed engineering history abnormal data through a convolutional neural network, and then compares the features of the real-time collected engineering data with the features of the engineering history abnormal data, filters the real-time collected engineering data according to the comparison results, and classifies the filtered engineering data through data clustering based on the obtained filtered engineering data, and manages the classified engineering data, thereby ensuring the accuracy of engineering data security management.

[0009] Preferably, the collecting of engineering history abnormal data, constructing a training set and processing the engineering history abnormal data in the training set to obtain the processed engineering history abnormal data comprises the following steps: Create an array of length n and select A hash function is used to traverse each set of engineering history abnormal data in the training set, and the traversal results are saved in an array; during the traversal process through the hash function, when the traversal results of two sets of data are the same, each data information in the two sets of data is compared; When the comparison results are consistent, the current two sets of data are set to be the same data, and based on the traversal order of each set of engineering history abnormal data, the engineering history abnormal data traversed later is deleted; after the traversal is completed, the engineering history abnormal data stored in the array during the traversal process is summarized to obtain the processed engineering history abnormal data.

[0010] The present invention establishes an array and selects a hash function to traverse each group of engineering history abnormal data in a training set, finds repeated engineering history abnormal data in the training set, and deletes the repeated engineering history abnormal data, thereby improving the reliability of the engineering history abnormal data.

[0011] Preferably, the process of extracting features from the processed engineering history abnormal data by using a convolutional neural network to obtain and save features of the engineering history abnormal data comprises the following steps: The convolutional neural network includes: a convolutional layer, a pooling layer and a fully connected layer; S21, setting each group of processed engineering history abnormal data as a processed engineering history abnormal data block; S22, inputting each processed engineering history abnormal data block into a convolutional neural network; S23, after receiving the input processed engineering history abnormal data block, the convolution layer of the convolutional neural network moves on the input processed engineering history abnormal data block according to the set step size through the convolution kernel, and performs convolution calculation on the corresponding area of ​​each step with the characteristic value of the area; S24, setting an activation function, and in the continuous convolution of the neural network, processing the input processed engineering history abnormal data block through the set activation function; S24, setting an activation function, and in the continuous convolution of the neural network, processing the input processed engineering history abnormal data block through the set activation function; S25, optimizing the features of the processed engineering history abnormal data blocks in the process of continuous stacking of convolutions and pooling; S26, passing the optimized processed engineering history abnormal data block features into the fully connected layer; S27, expand and combine the feature data through the fully connected layer to obtain a feature array and save it; The obtained feature array is set as the engineering history abnormal data feature.

[0012] Preferably, the convolution calculation formula is as follows: ; Wherein, I(z) represents the input processed engineering history abnormal data block, z represents the processed engineering history abnormal data value, w represents the weight of the corresponding convolution kernel, and b represents the bias value.

[0013] Preferably, the activation function processing step: Set the input value of the processed engineering history abnormal data block to z i (i=1,2,...,m), each input value z i The corresponding input weight is w i , b is the offset, and the output result obtained after inputting these input values ​​into the neuron is: ; Among them, g is the corresponding activation function, w i is the corresponding input value, y is the output processed by the activation function, and m represents the number of input values.

[0014] The present invention convolves the processed engineering history abnormal data using a convolutional neural network and activates the convolved engineering history abnormal data through an activation function, thereby extracting the features of the processed engineering history abnormal data and improving the accuracy of the features of the engineering history abnormal data.

[0015] Preferably, the step of comparing the characteristics of the obtained real-time collected engineering data with the characteristics of the engineering historical abnormal data, filtering the real-time collected engineering data according to the comparison result, and obtaining the filtered engineering data comprises the following steps: Compare the obtained features of the real-time collected engineering data with the features of the engineering history abnormal data, and calculate the similarity between the features of the real-time collected engineering data and the features of the engineering history abnormal data; The similarity calculation formula is: ; in, A vector representation of the features of engineering data collected in real time, A vector representation of the characteristics of engineering history abnormal data, Indicates the similarity between the features of real-time collected engineering data and the features of historical engineering abnormal data; set the similarity threshold [-1,1], when the similarity between the features of real-time collected engineering data and the features of historical engineering abnormal data is less than or equal to 0 after comparison, it is judged as dissimilar, indicating that the current real-time collected engineering data is abnormal data; When the characteristics of real-time collected engineering data are compared with the characteristics of historical engineering abnormal data and the similarity is greater than 0, it is determined to be similar, indicating that the current real-time collected engineering data is normal data; The engineering data collected in real time after the abnormal data is removed is set as the filtered engineering data. The present invention compares the characteristics of the real-time collected engineering data with the characteristics of the engineering history abnormal data, calculates the similarity between the characteristics of the real-time collected engineering data and the characteristics of the engineering history abnormal data, and completes the filtering of the real-time collected engineering data by setting a similarity threshold, thereby ensuring the security and real-time nature of the real-time collected engineering data.

[0016] Preferably, the filtered engineering data is classified by a data classification method based on the obtained filtered engineering data to obtain the classified engineering data, which comprises the following steps: The filtered engineering data is set to include engineering monitoring data and engineering scheduling data; Divide the filtered engineering data into two parts until the divided engineering data converges; After the division is completed, the variance of the feature data in each category is calculated and the variance threshold is set; ; Among them, ε 2is the variance of the distance between engineering data in the current classification, d j Represents the distance between the jth group of engineering data in the current classification, represents the average distance between engineering data in the current classification, and J represents the number of engineering data in the current classification; when the variance of the distance between engineering data in each classification after division is less than or equal to the set variance threshold, it means that the result of the current classification is the optimal result; When the variance of the distance between engineering data in each classification after division is greater than the set variance threshold, it means that the current classification result is not the optimal result; When the classification result is not the optimal result, the binary division is repeated until the engineering data converges, the iteration stops, and the optimal result after classification is output, and the optimal result is set as the classified engineering data.

[0017] The present invention classifies the filtered engineering data by a data classification method, specifically by calculating the variance of the feature data in each classification and setting a variance threshold, thereby ensuring the accuracy of engineering data classification.

[0018] Preferably, the management of the classified engineering data comprises the following steps: for the engineering monitoring data in the classified engineering data, setting an early warning range threshold for each engineering monitoring data, and when the engineering monitoring data collected in real time has engineering monitoring data exceeding the set early warning range threshold, giving an alarm; For the engineering scheduling data in the classified engineering data, the engineering scheduling data is sequentially executed based on the real-time collection sequence of the engineering scheduling data.

[0019] This embodiment also discloses an engineering data safety management system based on real-time monitoring, including: a data acquisition module, a data processing module, a feature extraction module, a comparison and filtering module, a data classification module, an alarm module and an execution module; The data acquisition module is used to collect engineering data in the system in real time and transmit the collected engineering data to the data processing module; The data processing module is used to process the received engineering data and transmit the processed engineering data to the feature extraction module; The feature extraction module is used to extract features from the received processed engineering data to obtain features of the engineering data; The comparison and filtering module is used to compare the engineering data features extracted in real time with the historical engineering data features, and filter the engineering data extracted in real time according to the comparison result to obtain filtered engineering data; The data classification module is used to classify the filtered engineering data to obtain classified engineering data; The alarm module is used to set a threshold value for engineering data, and when engineering data exceeding the threshold value is detected, an alarm is immediately processed; The execution module is used to sequentially execute engineering data.

[0020] (III) Beneficial effects

[0021] Compared with the prior art, the present invention provides a method and system for engineering data security management based on real-time monitoring, which has the following beneficial effects: 1. The invention collects engineering historical abnormal data, constructs a training set and processes the engineering historical abnormal data in the training set, and extracts and saves features of the processed engineering historical abnormal data through a convolutional neural network, and then compares the features of the real-time collected engineering data with the features of the engineering historical abnormal data, and filters the real-time collected engineering data according to the comparison results. At the same time, based on the obtained filtered engineering data, the filtered engineering data is classified by data clustering, and the classified engineering data is managed, thereby ensuring the accuracy of engineering data security management.

[0022] 2. The invention establishes an array and selects a hash function to traverse each group of engineering history abnormal data in the training set, finds repeated engineering history abnormal data in the training set, and deletes the repeated engineering history abnormal data, thereby improving the reliability of the engineering history abnormal data.

[0023] 3. The present invention extracts features of processed engineering history abnormal data by using a convolutional neural network to convolve the processed engineering history abnormal data and activating the convolved engineering history abnormal data through an activation function, thereby improving the accuracy of the features of the engineering history abnormal data.

[0024] 4. The invention compares the characteristics of the real-time collected engineering data with the characteristics of the engineering history abnormal data, calculates the similarity between the characteristics of the real-time collected engineering data and the characteristics of the engineering history abnormal data, and at the same time filters the real-time collected engineering data by setting a similarity threshold, thereby ensuring the security and real-time nature of the real-time collected engineering data.

[0025] 5. The invention classifies the filtered engineering data through data classification, specifically by calculating the variance of the feature data in each classification and setting a variance threshold, thereby ensuring the accuracy of engineering data classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a structural schematic diagram of the engineering data security management process of real-time monitoring of the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] Example 1

[0029] See also Figure 1 The present invention discloses a method for engineering data security management based on real-time monitoring, which specifically includes the following steps: S1. Collect engineering history abnormal data, construct a training set, and process the engineering history abnormal data in the training set to obtain processed engineering history abnormal data; Collecting engineering history abnormal data, building a training set and processing the engineering history abnormal data in the training set to obtain the processed engineering history abnormal data includes the following steps: Create an array of length n and select A hash function is used to traverse each set of engineering history abnormal data in the training set, and the traversal results are saved in an array; during the traversal process through the hash function, when the traversal results of two sets of data are the same, each data information in the two sets of data is compared; When the comparison results are consistent, the current two sets of data are set to be the same data, and based on the traversal order of each set of engineering history abnormal data, the engineering history abnormal data traversed later is deleted; After the traversal is completed, the engineering history abnormal data stored in the array during the traversal process is summarized to obtain the processed engineering history abnormal data; S2. Extracting features of the processed engineering history abnormal data through a convolutional neural network to obtain and save features of the engineering history abnormal data; The process of extracting features from the processed engineering history abnormal data through a convolutional neural network, obtaining the features of the engineering history abnormal data and saving them includes the following steps: The convolutional neural network includes: a convolutional layer, a pooling layer and a fully connected layer; S21, setting each group of processed engineering history abnormal data as a processed engineering history abnormal data block; S22, inputting each processed engineering history abnormal data block into a convolutional neural network; S23, after receiving the input processed engineering history abnormal data block, the convolution layer of the convolutional neural network moves on the input processed engineering history abnormal data block according to the set step size through the convolution kernel, and performs convolution calculation on the corresponding area of ​​each step with the characteristic value of the area; The convolution calculation formula is as follows: ; Among them, I(z) represents the processed engineering history abnormal data block input, z represents the processed engineering history abnormal data value, w represents the weight of the corresponding convolution kernel, and b represents the bias value; S24, setting an activation function, and processing the input processed engineering history abnormal data block through the set activation function in the continuous convolution of the neural network; Activation function processing steps: Set the input value of the processed engineering history abnormal data block to z i (i=1,2,...,m), each input value z i The corresponding input weight is w i , b is the offset, and the output result obtained after inputting these input values ​​into the neuron is: ; Among them, g is the corresponding activation function, w i is the corresponding input value, y is the output processed by the activation function, and m represents the number of input values; S25, optimizing the features of the processed engineering history abnormal data blocks in the process of continuous stacking of convolutions and pooling; S26, passing the optimized processed engineering history abnormal data block features into the fully connected layer; S27, expand and combine the feature data through the fully connected layer to obtain a feature array and save it; The obtained feature array is set as the engineering history abnormal data feature; S3, collecting engineering data in real time, and inputting the engineering data collected in real time into a convolutional neural network to obtain features of the engineering data collected in real time, comparing the features of the engineering data collected in real time with features of historical abnormal engineering data, filtering the engineering data collected in real time according to the comparison results, and obtaining filtered engineering data; Comparing the characteristics of the real-time collected engineering data with the characteristics of the engineering historical abnormal data, filtering the real-time collected engineering data according to the comparison result, and obtaining the filtered engineering data includes the following steps: Compare the obtained features of the real-time collected engineering data with the features of the engineering history abnormal data, and calculate the similarity between the features of the real-time collected engineering data and the features of the engineering history abnormal data; The similarity calculation formula is: ; in, A vector representation of the features of engineering data collected in real time, A vector representation of the characteristics of engineering history abnormal data, Indicates the similarity between the features of real-time collected engineering data and the features of historical abnormal engineering data; Set the similarity threshold [-1, 1]. When the similarity between the features of the real-time collected engineering data and the features of the historical engineering abnormal data is less than or equal to 0, it is judged as dissimilar, indicating that the current real-time collected engineering data is abnormal data; When the similarity between the features of the real-time collected engineering data and the features of the historical engineering abnormal data is greater than 0, it is determined to be similar, indicating that the current real-time collected engineering data is normal data; further, the real-time collected engineering data after removing the abnormal data is set as the filtered engineering data; S4. Based on the obtained filtered engineering data, classify the filtered engineering data by data clustering to obtain classified engineering data, wherein the classification results include: engineering data type and engineering data of each classification; Based on the obtained filtered engineering data, the filtered engineering data is classified by a data classification method, and obtaining the classified engineering data includes the following steps: The filtered engineering data is set to include engineering monitoring data and engineering scheduling data; Divide the filtered engineering data into two parts until the divided engineering data converges; After the division is completed, the variance of the feature data in each category is calculated and the variance threshold is set; ; Among them, ε 2 is the variance of the distance between engineering data in the current classification, d j Represents the distance between the jth group of engineering data in the current classification, represents the average distance between engineering data in the current classification, and J represents the number of engineering data in the current classification; When the variance of the distance between engineering data in each classification after division is less than or equal to the set variance threshold, it means that the result of the current classification is the optimal result; When the variance of the distance between engineering data in each classification after division is greater than the set variance threshold, it means that the current classification result is not the optimal result; When the classification result is not the optimal result, the binary division is repeated until the engineering data converges, the iteration stops, and the optimal result after classification is output, and the optimal result is set as the classified engineering data; S5. Manage the classified engineering data; Managing classified engineering data includes the following steps: For the engineering monitoring data in the classified engineering data, set the warning range threshold of each engineering monitoring data, and when the engineering monitoring data collected in real time exceeds the set warning range threshold, an alarm is issued; For the engineering scheduling data in the classified engineering data, the engineering scheduling data is sequentially executed based on the real-time collection order of the engineering scheduling data; Example 2

[0030] This embodiment also discloses an engineering data safety management system based on real-time monitoring, including: a data acquisition module, a data processing module, a feature extraction module, a comparison and filtering module, a data classification module, an alarm module and an execution module; The data acquisition module is used to collect engineering data in the system in real time and transmit the collected engineering data to the data processing module; The data processing module is used to process the received engineering data and transmit the processed engineering data to the feature extraction module; The feature extraction module is used to extract features from the received processed engineering data to obtain features of the engineering data; The comparison and filtering module is used to compare the engineering data features extracted in real time with the historical engineering data features, and filter the engineering data extracted in real time according to the comparison result to obtain filtered engineering data; The data classification module is used to classify the filtered engineering data to obtain classified engineering data; The alarm module is used to set a threshold value for engineering data, and when engineering data exceeding the threshold value is detected, an alarm is immediately processed; The execution module is used to sequentially execute engineering data.

[0031] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for engineering data security management based on real-time monitoring, characterized in that: The following steps are involved: S1. Collect engineering history abnormal data, construct a training set, and process the engineering history abnormal data in the training set to obtain processed engineering history abnormal data; S2. Extracting features of the processed engineering history abnormal data through a convolutional neural network to obtain and save features of the engineering history abnormal data; S3, collecting engineering data in real time, and inputting the engineering data collected in real time into a convolutional neural network to obtain features of the engineering data collected in real time, comparing the features of the engineering data collected in real time with features of historical abnormal engineering data, filtering the engineering data collected in real time according to the comparison results, and obtaining filtered engineering data; S4. Based on the obtained filtered engineering data, classify the filtered engineering data by data clustering to obtain classified engineering data, wherein the classification results include: engineering data type and engineering data of each classification; S5. Manage the classified engineering data.

2. The engineering data security management method based on real-time monitoring according to claim 1 is characterized in that: The collecting of engineering history abnormal data, constructing a training set and processing the engineering history abnormal data in the training set to obtain the processed engineering history abnormal data comprises the following steps: Create an array of length n and select A hash function traverses each set of engineering history abnormal data in the training set and saves the traversal results in an array; In the process of traversing through the hash function, when there are two sets of data with the same traversal results, each piece of data information in the two sets of data is compared; When the comparison results are consistent, the current two sets of data are set to be the same data, and based on the traversal order of each set of engineering history abnormal data, the engineering history abnormal data traversed later is deleted; After the traversal is completed, the engineering history exception data stored in the array during the traversal process is summarized to obtain the processed engineering history exception data.

3. The engineering data security management method based on real-time monitoring according to claim 1 is characterized in that: The process of extracting features from the processed engineering history abnormal data by using a convolutional neural network to obtain and save features of the engineering history abnormal data includes the following steps: The convolutional neural network includes: a convolutional layer, a pooling layer and a fully connected layer; S21, setting each group of processed engineering history abnormal data as a processed engineering history abnormal data block; S22, inputting each processed engineering history abnormal data block into a convolutional neural network; S23, after receiving the input processed engineering history abnormal data block, the convolution layer of the convolutional neural network moves on the input processed engineering history abnormal data block according to the set step size through the convolution kernel, and performs convolution calculation on the corresponding area of ​​each step with the characteristic value of the area; S24, setting an activation function, and in the continuous convolution of the neural network, processing the input processed engineering history abnormal data block through the set activation function; S24, setting an activation function, and in the continuous convolution of the neural network, processing the input processed engineering history abnormal data block through the set activation function; S25, optimizing the features of the processed engineering history abnormal data blocks in the process of continuous stacking of convolutions and pooling; S26, passing the optimized processed engineering history abnormal data block features into the fully connected layer; S27, expand and combine the feature data through the fully connected layer to obtain a feature array and save it; The obtained feature array is set as the engineering history abnormal data feature.

4. The engineering data security management method based on real-time monitoring according to claim 3 is characterized in that: The convolution calculation formula is as follows: ; Wherein, I(z) represents the input processed engineering history abnormal data block, z represents the processed engineering history abnormal data value, w represents the weight of the corresponding convolution kernel, and b represents the bias value.

5. The engineering data security management method based on real-time monitoring according to claim 3 is characterized in that: The activation function processing steps are: Set the input value of the processed engineering history abnormal data block to z i (i=1,2,...,m), each input value z i The corresponding input weight is w i , b is the offset, and the output result obtained after inputting these input values ​​into the neuron is: ; Among them, g is the corresponding activation function, w i is the corresponding input value, y is the output processed by the activation function, and m represents the number of input values.

6. The engineering data security management method based on real-time monitoring according to claim 1 is characterized in that: The step of comparing the characteristics of the real-time collected engineering data with the characteristics of the historical abnormal engineering data, filtering the real-time collected engineering data according to the comparison result, and obtaining the filtered engineering data comprises the following steps: Compare the obtained features of the real-time collected engineering data with the features of the engineering history abnormal data, and calculate the similarity between the features of the real-time collected engineering data and the features of the engineering history abnormal data; The similarity calculation formula is: ;in, A vector representation of the features of engineering data collected in real time, A vector representation of the characteristics of engineering history abnormal data, Indicates the similarity between the features of real-time collected engineering data and the features of historical abnormal engineering data; Set the similarity threshold [-1,1]. When the similarity between the features of the real-time collected engineering data and the features of the engineering history abnormal data is less than or equal to 0, it is determined to be dissimilar, indicating that the current real-time collected engineering data is abnormal data; when the similarity between the features of the real-time collected engineering data and the features of the engineering history abnormal data is greater than 0, it is determined to be similar, indicating that the current real-time collected engineering data is normal data; The engineering data collected in real time after removing abnormal data is set as filtered engineering data.

7. The engineering data security management method based on real-time monitoring according to claim 1 is characterized in that: Based on the obtained filtered engineering data, classifying the filtered engineering data by a data classification method to obtain the classified engineering data includes the following steps: The filtered engineering data is set to include engineering monitoring data and engineering scheduling data; Divide the filtered engineering data into two parts until the divided engineering data converges; After the division is completed, the variance of the feature data in each category is calculated and the variance threshold is set; ; where ε 2 is the variance of the distance between engineering data in the current classification, d j Represents the distance between the jth group of engineering data in the current classification, represents the average distance between engineering data in the current classification, and J represents the number of engineering data in the current classification; When the variance of the distance between engineering data in each classification after division is less than or equal to the set variance threshold, it means that the result of the current classification is the optimal result; When the variance of the distance between engineering data in each classification after division is greater than the set variance threshold, it means that the current classification result is not the optimal result; When the classification result is not the optimal result, the binary division is repeated until the engineering data converges, the iteration stops, and the optimal result after classification is output, and the optimal result is set as the classified engineering data.

8. The engineering data security management method based on real-time monitoring according to claim 1 is characterized in that: The management of the classified engineering data comprises the following steps: For the engineering monitoring data in the classified engineering data, set the warning range threshold of each engineering monitoring data, and when the engineering monitoring data collected in real time exceeds the set warning range threshold, an alarm is issued; For the engineering scheduling data in the classified engineering data, the engineering scheduling data is sequentially executed based on the real-time collection sequence of the engineering scheduling data.

9. A system for implementing the engineering data security management method based on real-time monitoring according to any one of claims 1 to 8, characterized in that: include: Data acquisition module, data processing module, feature extraction module, comparison and filtering module, data classification module, alarm module and execution module; The data acquisition module is used to collect engineering data in the system in real time and transmit the collected engineering data to the data processing module; the data processing module is used to process the received engineering data and transmit the processed engineering data to the feature extraction module; The feature extraction module is used to extract features from the received processed engineering data to obtain features of the engineering data; The comparison and filtering module is used to compare the engineering data features extracted in real time with the historical engineering data features, and filter the engineering data extracted in real time according to the comparison result to obtain filtered engineering data; The data classification module is used to classify the filtered engineering data to obtain classified engineering data; The alarm module is used to set a threshold value for engineering data, and when engineering data exceeding the threshold value is detected, an alarm is immediately processed; The execution module is used to sequentially execute engineering data.

Images