CN114386067A - Equipment production data safe transmission method and system based on artificial intelligence - Google Patents

Equipment production data safe transmission method and system based on artificial intelligence Download PDF

Info

Publication number
CN114386067A
CN114386067A CN202210007885.2A CN202210007885A CN114386067A CN 114386067 A CN114386067 A CN 114386067A CN 202210007885 A CN202210007885 A CN 202210007885A CN 114386067 A CN114386067 A CN 114386067A
Authority
CN
China
Prior art keywords
data
obtaining
data type
encrypted
network loss
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202210007885.2A
Other languages
Chinese (zh)
Other versions
CN114386067B (en
Inventor
曲佳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chengde Qingyun Information Technology Co.,Ltd.
Original Assignee
Chengde Petroleum College
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chengde Petroleum College filed Critical Chengde Petroleum College
Priority to CN202210007885.2A priority Critical patent/CN114386067B/en
Publication of CN114386067A publication Critical patent/CN114386067A/en
Application granted granted Critical
Publication of CN114386067B publication Critical patent/CN114386067B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G06F21/602—Providing cryptographic facilities or services
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/901—Indexing; Data structures therefor; Storage structures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Molecular Biology (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Evolutionary Computation (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Bioethics (AREA)
  • Computer Hardware Design (AREA)
  • Computer Security & Cryptography (AREA)
  • Compression, Expansion, Code Conversion, And Decoders (AREA)

Abstract

The invention relates to the technical field of data transmission, in particular to a method and a system for safely transmitting equipment production data based on artificial intelligence. The method sets a self-encoding neural network comprising an encoder and a plurality of decoders, trains through historical production data of a plurality of data types, one data type corresponds to one decoder, and controls network loss through required precision of the historical production data of each data type. And randomly arranging the initial target data according to the data type to obtain first target production data and a data type index sequence. And encrypting the data type index sequence to obtain first encrypted data, sending the first target production data into an encoder to obtain second encrypted data, obtaining a digital abstract of the data type index sequence, and transmitting the first encrypted data, the second encrypted data and the digital abstract as transmission data. The invention realizes the encrypted transmission of the target production data by improving the self-coding neural network.

Description

Equipment production data safe transmission method and system based on artificial intelligence
Technical Field
The invention relates to the technical field of data transmission, in particular to a method and a system for safely transmitting equipment production data based on artificial intelligence.
Background
The manufacturing workshop equipment is diversified, various state data of the equipment in the workshop, such as equipment rotating speed, temperature, pressure and the like, need to be monitored in real time for an intelligent manufacturing scene, and various sensors need to be arranged for acquiring various types of data. The accuracy requirements and data ranges of different types of sensor data are different, and the communication protocols of different brands of equipment are also different, so that the sensor data are difficult to transmit uniformly. In addition, in the data transmission process, because the data volume is large, the data needs to be transmitted after being compressed, and if the data is tampered due to malicious attack received in the transmission process, wrong data can be transmitted to a wrong instruction of the control center, so that production loss is caused.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a method and a system for safely transmitting equipment production data based on artificial intelligence, and the adopted technical scheme is as follows:
the invention provides an equipment production data safe transmission method based on artificial intelligence, which comprises the following steps:
acquiring historical production data of multiple data types; obtaining the required precision corresponding to the data type according to the difference between the historical production data of the same data type; training a self-coding neural network according to the historical production data; the self-coding neural network comprises an encoder and a plurality of decoders, and each data type corresponds to one decoder; obtaining a network loss weight according to the required precision, and adjusting the network loss of the self-coding neural network according to the network loss weight;
acquiring initial target production data of a plurality of data types; randomly arranging the initial target production data according to the data type to obtain first target production data and a data type index sequence; encrypting the data type index sequence to obtain first encrypted data; sending the first target production data into the self-coding neural network, and obtaining second encrypted data according to the encoder; obtaining a digital summary of the data type index sequence;
and transmitting the first encrypted data, the second encrypted data and the digital abstract as transmission data.
Further, the obtaining the required precision corresponding to the data type according to the difference between the historical production data of the same data type includes:
taking an average difference between adjacent historical data of the same data type as the required precision.
Further, the obtaining a network loss weight according to the required precision, and the adjusting the network loss of the self-coding neural network according to the network loss weight comprises:
the required precision is in inverse proportion to the network loss weight; the initial network loss function of the self-coding neural network adopts a mean square error loss function, and the initial network loss value of each data type is obtained by multiplying the initial network loss function by the network loss weight; and taking the average value of the initial network loss values as the network loss.
Further, the inversely proportional relationship between the required accuracy and the network loss weight comprises:
Figure BDA0003457691850000021
wherein, JDnFor the required precision corresponding to the nth data type, max (jd) is the maximum required precision among all the data types, and min (jd) is the minimum required precision among all the data types.
Further, the encrypting the data type index sequence to obtain the first encrypted data includes:
and encrypting the data type index sequence by adopting a symmetric encryption method to obtain first encrypted data.
Further, the method further comprises a decryption process, the decryption process comprising:
splitting the transmission data into the first encrypted data, the second encrypted data and the digital digest;
decrypting the first encrypted data to obtain a decrypted data type index sequence; obtaining the encoder of the corresponding self-encoding neural network according to the decrypted data type index sequence; and processing the second encrypted data by the corresponding encoders in sequence to obtain decrypted data.
Further, the decryption process further includes a security verification process, the security verification process including:
obtaining a decrypted digital digest of the decrypted data type index sequence; obtaining a correspondence between the decrypted digital digest and the digital digest;
obtaining a data range and a range radius of each data type according to the historical production data; acquiring central data of the data range; obtaining a differential distance between the decrypted data of each of the data types and the corresponding central data; taking the square of the difference between the difference distance and the corresponding range radius as a data error distance; taking the average data error distance of all the data types as the abnormal degree;
obtaining the data attack probability according to the consistency and the abnormal degree; and judging whether the transmission data is safe or not according to the attack probability of the data.
Further, the obtaining of the data attack probability according to the consistency and the degree of abnormality comprises:
obtaining the data attack probability according to a data attack probability formula; the data attack probability formula comprises:
ZG＝1-YS*(1-YC)
wherein ZG is the data attack probability, YS is the consistency, and YC is the abnormal degree.
The invention also provides an equipment production data safe transmission system based on artificial intelligence, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein the processor realizes any step of the equipment production data safe transmission method based on artificial intelligence when executing the computer program.
The invention has the following beneficial effects:
1. the self-coding neural network provided by the embodiment of the invention comprises an encoder and decoders corresponding to various data types one by one, so that the unified transmission of data of various data types is realized, namely, the data of all data types are encrypted by the encoder in the self-coding neural network, and the decryption process is realized according to the respective corresponding decoders in the subsequent decryption process.
2. According to the embodiment of the invention, the initial target production data are randomly sequenced according to the data type, and the first target production data and the data type index sequence are obtained. The first target production data and the data type index sequence are respectively encrypted, the transmission data are obtained by combining the digital abstract of the data type index sequence, the safety of the transmission data is guaranteed, the data are protected through the first encrypted data, the second encrypted data and the digital abstract, and interception and tampering in the transmission process are avoided.
3. The embodiment of the invention controls the network loss of the self-coding neural network according to the precision of each data type, and improves the performance of network encryption and decryption.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a flow chart of a method for securely transmitting production data of an artificial intelligence-based device according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a self-coding neural network according to an embodiment of the present invention.
Detailed Description
In order to further explain the technical means and effects of the present invention adopted to achieve the predetermined objects, the following detailed description, the structure, the features and the effects of the method and the system for securely transmitting the production data of the equipment based on artificial intelligence according to the present invention are provided with the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following describes a specific scheme of the equipment production data secure transmission method and system based on artificial intelligence in detail with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for securely transmitting production data of an apparatus based on artificial intelligence according to an embodiment of the present invention is shown, where the method includes:
step S1: acquiring historical production data of multiple data types; obtaining the required precision corresponding to the data type according to the difference between historical production data of the same data type; training a self-coding neural network according to historical production data; the self-coding neural network comprises an encoder and a plurality of decoders, wherein each data type corresponds to one decoder; and obtaining the network loss weight according to the required precision, and adjusting the network loss of the self-coding neural network according to the network loss weight.
During the production process, various data can be acquired through sensors in a production workshop, for example, a code disc sensor is used for acquiring rotation angular velocity data, a piezoelectric ceramic sensor is used for acquiring pressure data, a temperature sensor is used for acquiring temperature data and the like. Wherein the transmission types of different data are different, such as binary transmission, text transmission, ethernet mode transmission based on TCP/IP protocol, etc., and the data type of each data can be obtained according to the transmission type and the sensor data type, such as pressure data, etc., in which the type of data a is binary. And numbering the data types and corresponding the data types to the acquired data one by one to obtain historical production data of various data types.
The accuracy and range of data obtained by different sensors are different, and the accuracy requirement of subsequently generated data is also different. The size of the data collected by the same sensor under the same working condition is fluctuated within a range, the fluctuation size is the precision of the data, so that the precision required by the data type corresponding to the data type can be obtained according to the difference between historical production data of the same data type, and the method specifically comprises the following steps:
the data of the same data type can be regarded as one-dimensional data, each element in the one-dimensional data corresponds to the data acquired by the sensor once, and the average difference between the adjacent historical data in the one-dimensional data of the same data type is used as the required precision.
In the industrial data transmission scene of equipment production, because the data volume is large and the data needs to be compressed and transmitted, the compression and the encryption are often realized together, so the compression and the encryption of the data can be realized through a self-coding neural network. In a conventional self-coding neural network, only one encoder and one decoder are included, and if data of all data types are processed by one coding and decoding structure, ambiguity of data is easily caused, namely high-dimensional data is coded into low-dimensional data, and the same low-dimensional data can be generated by coding different high-dimensional data.
Therefore, the decoder in the self-coding neural network is expanded to realize a plug-and-play self-coding neural network. Referring to fig. 2, a schematic diagram of a self-coding neural network according to an embodiment of the present invention is shown. The network comprises an encoder A and a plurality of decoders B, each data type corresponds to one decoder, and N data types including N decoders B can be collected in the production process. The pressure data and the temperature data can be combined and used according to different data types in the using process, for example, the pressure data and the temperature data are jointly coded by a coder to obtain the same coded data, the coded data are respectively input into a pressure decoder and a temperature decoder, and the decoded pressure data and the decoded temperature data are output.
The self-coding neural network is a self-supervision learning network with self as a label, namely, the network does not need artificial marking, and the network trains network parameters according to the difference relation between input data and output data, namely, the input data volume of an encoder is consistent with the output data volume of a decoder, and a mean square error loss function is generally adopted as a network initial loss function.
Taking historical production data as training data of the self-coding neural network, considering that required accuracies of the historical production data of different data types are different, obtaining a network loss weight according to the required accuracy corresponding to the training data in network training, and adjusting the network loss of the self-coding neural network according to the network loss weight, specifically comprising the following steps:
and obtaining the network loss weight according to the required precision, wherein the network loss weight and the required precision are in an inverse proportional relation. The initial network loss function adopts a mean square error loss function EM, namely:
Figure BDA0003457691850000051
wherein EM is an initial network loss function, T is the batch size of network input training data, Y is the T-th decoding output data, StIs the t-th historical production data.
And the EM represents the initial network loss of one data type in a training process, and the network loss weight is multiplied by the initial network loss function to obtain the initial network loss value of each data type. Since the self-coding neural network includes a plurality of decoders, each decoder corresponding to one data type, the average value of the initial network loss values is taken as the network loss, that is:
Figure BDA0003457691850000052
where ZL is network loss, N is number of data type types, fnFor the network loss weight corresponding to the nth data type, EMnAnd the function is the initial network loss function corresponding to the nth data type.
By introducing the required precision, the network is made more concerned about the data type with less required precision, i.e. the less required precision, the more the network loses weight. And when the network loss reaches the minimum, finishing the training and updating of the whole network.
Network loss weight fnThe method specifically comprises the following steps:
Figure BDA0003457691850000053
wherein, JDnAnd max (JD) is the maximum required precision in all the data types, and min (JD) is the minimum required precision in all the data types.
It should be noted that, when a new sensor is added in the equipment production plant, that is, the sensor has a new data type, a new decoder may be added on the basis of the invariance of the current self-coding neural network parameters, and training may be performed on the new decoder.
Step S2: acquiring initial target production data of multiple data types; randomly arranging the initial target production data according to the data type to obtain first target production data and a data type index sequence; encrypting the data type index sequence to obtain first encrypted data; sending the first target production data into a self-coding neural network, and obtaining second encrypted data according to an encoder; a digital digest of the data type index sequence is obtained.
Initial target production data of a plurality of data types to be encrypted is obtained. And the number of the data type in the initial target production data is consistent with that of the historical production data, and the initial target production data is randomly arranged according to the data type to obtain first target production data and a data type index sequence. The random arrangement can be regarded as a method for protecting data from leakage of a decryption method due to a fixed arrangement order, for example, the data type index sequence LX of the conventional initial target production data is: LX ═ 1,2,3,4,5,6,7,8,9,10], after random arrangement: and LX is [9,5,8,3,2,10,1,6,4,7], wherein each index corresponds to a decoder of a self-coding neural network, and the random arrangement ensures that the decoding sequence is different for different data types each time, thereby ensuring the data security.
For further protection of the data type index sequence, the data type index sequence is encrypted to obtain first encrypted data.
Preferably, the data type index sequence is encrypted by a symmetric encryption method to obtain first encrypted data.
And sending the first target production data into a self-coding neural network, and processing the first target production data through an encoder to take the encoding information as second encryption data. It should be noted that the second encrypted data is a set of encoded information obtained by respectively processing different data types by an encoder.
And processing the data type index sequence by a Hash algorithm to obtain a digital abstract of the data type index sequence. The digital abstract is a short message with a fixed length and can be used for a subsequent security evaluation process.
Step S3: and transmitting the first encrypted data, the second encrypted data and the digital abstract as transmission data.
And setting separators among the first encrypted data, the second encrypted data and the digital abstract, and integrating the three kinds of data into one transmission data. And carrying out network transmission on the transmission data, and carrying out decryption operation at a receiving end to realize safe transmission of the equipment production data.
The decryption process of the data receiving end specifically comprises the following steps:
the transmission data is divided into first encryption data, second encryption data and a digital summary according to separators in the transmission data.
And decrypting the first encrypted data according to a decryption method corresponding to the encryption method of the first encrypted data to obtain a decrypted data type index sequence. And obtaining the corresponding encoder of the self-encoding neural network according to the decrypted data type index sequence. And processing the second encrypted data by the corresponding encoder in sequence to obtain decrypted data.
After the decrypted data is obtained, the transmission security can be evaluated according to the decrypted data, and the method specifically comprises the following steps:
a decrypted digital digest of the decrypted data type index sequence is obtained. Consistency of the decrypted digital digest and the digital digest is obtained. In the embodiment of the present invention, considering that the digital digest is a short message of a fixed length, the data length and the data size are small, so if the decrypted digital digest and the digital digest are completely identical, the consistency is 1, otherwise, the consistency is 0.
The data range and range radius for each data type is obtained from historical production data. Central data of the data range is acquired. The difference distance of the decrypted data of each data type and the corresponding center data is obtained. The data error distance is taken as the square of the difference between the difference distance and the corresponding range radius. The average data error distance of all data types is taken as the degree of abnormality YC. Is formulated as:
Figure BDA0003457691850000071
where YC is the degree of abnormality, N is the number of data type categories, D () is the difference distance calculation function, YnFor decrypted data corresponding to the nth data type, CnFor decrypting the central data corresponding to the data, RnThe range radius corresponds to the nth data type.
Wherein the data range corresponds to the maximum value max (X) of the data typenAnd minimum value min (X)nThe interval in between. Radius of range of
Figure BDA0003457691850000072
The central data is min (X)n+Rn。
The degree of abnormality indicates that the more each decrypted data is out of the range of the data of the corresponding data type, the more abnormal.
Obtaining the data attack probability according to the consistency and the abnormal degree, specifically comprising:
obtaining the data attack probability according to a data attack probability formula; the data attack probability formula comprises:
ZG＝1-YS*(1-YC)
wherein ZG is the data attack probability, YS is consistency, and YC is the abnormal degree.
And judging whether the transmission data is safe or not according to the attack probability of the data, and when the attack probability of the data is greater than a preset probability threshold, indicating that the current data is abnormally sent and tampered, the data is unsafe and the content is not credible. In the present embodiment, the probability threshold is set to 0.7.
In summary, the embodiment of the present invention sets a self-encoding neural network including one encoder and a plurality of decoders, and trains through historical production data of a plurality of data types, one data type corresponds to one decoder, and network loss is controlled through required accuracy of the historical production data of each data type. And randomly arranging the initial target data according to the data type to obtain first target production data and a data type index sequence. And encrypting the data type index sequence to obtain first encrypted data, sending the first target production data into an encoder to obtain second encrypted data, obtaining a digital abstract of the data type index sequence, and transmitting the first encrypted data, the second encrypted data and the digital abstract as transmission data. The embodiment of the invention realizes the encrypted transmission of the target production data by improving the self-coding neural network.
The invention also provides an equipment production data safe transmission system based on artificial intelligence, which comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein when the processor executes the computer program, any step of the equipment production data safe transmission method based on artificial intelligence is realized.
It should be noted that: the precedence order of the above embodiments of the present invention is only for description, and does not represent the merits of the embodiments. And specific embodiments thereof have been described above. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims (9)

1. An artificial intelligence-based equipment production data secure transmission method is characterized by comprising the following steps:
acquiring historical production data of multiple data types; obtaining the required precision corresponding to the data type according to the difference between the historical production data of the same data type; training a self-coding neural network according to the historical production data; the self-coding neural network comprises an encoder and a plurality of decoders, and each data type corresponds to one decoder; obtaining a network loss weight according to the required precision, and adjusting the network loss of the self-coding neural network according to the network loss weight;
acquiring initial target production data of a plurality of data types; randomly arranging the initial target production data according to the data type to obtain first target production data and a data type index sequence; encrypting the data type index sequence to obtain first encrypted data; sending the first target production data into the self-coding neural network, and obtaining second encrypted data according to the encoder; obtaining a digital summary of the data type index sequence;
and transmitting the first encrypted data, the second encrypted data and the digital abstract as transmission data.
2. The method for securely transmitting the production data of the equipment based on the artificial intelligence as claimed in claim 1, wherein the obtaining the required accuracy corresponding to the data type according to the difference between the historical production data of the same data type comprises:
taking an average difference between adjacent historical data of the same data type as the required precision.
3. The method of claim 1, wherein the obtaining of the network loss weight according to the required accuracy, and the adjusting of the network loss of the self-coding neural network according to the network loss weight comprises:
the required precision is in inverse proportion to the network loss weight; the initial network loss function of the self-coding neural network adopts a mean square error loss function, and the initial network loss value of each data type is obtained by multiplying the initial network loss function by the network loss weight; and taking the average value of the initial network loss values as the network loss.
4. The method of claim 3, wherein the inverse scaling of the required accuracy to the network loss weight comprises:
Figure FDA0003457691840000011
wherein, JDnFor the required precision corresponding to the nth data type, max (jd) is the maximum required precision among all the data types, and min (jd) is the minimum required precision among all the data types.
5. The method for securely transmitting the equipment production data based on the artificial intelligence as claimed in claim 1, wherein the encrypting the data type index sequence to obtain the first encrypted data comprises:
and encrypting the data type index sequence by adopting a symmetric encryption method to obtain first encrypted data.
6. The method for securely transmitting artificial intelligence based equipment production data according to claim 1, wherein the method further comprises a decryption process, the decryption process comprising:
splitting the transmission data into the first encrypted data, the second encrypted data and the digital digest;
decrypting the first encrypted data to obtain a decrypted data type index sequence; obtaining the encoder of the corresponding self-encoding neural network according to the decrypted data type index sequence; and processing the second encrypted data by the corresponding encoders in sequence to obtain decrypted data.
7. The method for securely transmitting the production data of the artificial intelligence-based equipment according to claim 6, wherein the decryption process further comprises a security verification process, the security verification process comprising:
obtaining a decrypted digital digest of the decrypted data type index sequence; obtaining a correspondence between the decrypted digital digest and the digital digest;
obtaining a data range and a range radius of each data type according to the historical production data; acquiring central data of the data range; obtaining a differential distance between the decrypted data of each of the data types and the corresponding central data; taking the square of the difference between the difference distance and the corresponding range radius as a data error distance; taking the average data error distance of all the data types as the abnormal degree;
obtaining the data attack probability according to the consistency and the abnormal degree; and judging whether the transmission data is safe or not according to the attack probability of the data.
8. The method for securely transmitting the production data of the equipment based on the artificial intelligence as claimed in claim 7, wherein the obtaining the attack probability of the data according to the consistency and the degree of the anomaly comprises:
obtaining the data attack probability according to a data attack probability formula; the data attack probability formula comprises:
ZG＝1-YS*(1-YC)
wherein ZG is the data attack probability, YS is the consistency, and YC is the abnormal degree.
9. An artificial intelligence based system for secure transmission of production data of a device, comprising a memory, a processor and a computer program stored in said memory and executable on said processor, wherein said processor when executing said computer program implements the steps of the method according to any one of claims 1 to 8.
CN202210007885.2A 2022-01-06 2022-01-06 Equipment production data safe transmission method and system based on artificial intelligence Active CN114386067B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210007885.2A CN114386067B (en) 2022-01-06 2022-01-06 Equipment production data safe transmission method and system based on artificial intelligence

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210007885.2A CN114386067B (en) 2022-01-06 2022-01-06 Equipment production data safe transmission method and system based on artificial intelligence

Publications (2)

Publication Number Publication Date
CN114386067A true CN114386067A (en) 2022-04-22
CN114386067B CN114386067B (en) 2022-08-23

Family

ID=81199378

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210007885.2A Active CN114386067B (en) 2022-01-06 2022-01-06 Equipment production data safe transmission method and system based on artificial intelligence

Country Status (1)

Country Link
CN (1) CN114386067B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116957991A (en) * 2023-09-19 2023-10-27 北京渲光科技有限公司 Three-dimensional model complement method and three-dimensional model complement model generation method

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190273509A1 (en) * 2018-03-01 2019-09-05 Crowdstrike, Inc. Classification of source data by neural network processing
WO2019185987A1 (en) * 2018-03-29 2019-10-03 Nokia Technologies Oy Entropy-friendly neural network representations and their use in training and using neural networks such as autoencoders
CN111709491A (en) * 2020-06-30 2020-09-25 平安科技（深圳）有限公司 Anomaly detection method, device and equipment based on self-encoder and storage medium
CN111967502A (en) * 2020-07-23 2020-11-20 电子科技大学 Network intrusion detection method based on conditional variation self-encoder
CN112115443A (en) * 2020-11-19 2020-12-22 索信达(北京)数据技术有限公司 Terminal user authentication method and system
CN112789625A (en) * 2018-09-27 2021-05-11 渊慧科技有限公司 Committed information rate variational self-encoder
WO2021137745A1 (en) * 2019-12-30 2021-07-08 Unibap Ab A method for detection of imperfections in products
WO2021165887A1 (en) * 2020-02-19 2021-08-26 Insilico Medicine Ip Limited Adversarial autoencoder architecture for methods of graph to sequence models
WO2021205066A1 (en) * 2020-04-09 2021-10-14 Nokia Technologies Oy Training a data coding system for use with machines
CN113691513A (en) * 2021-08-13 2021-11-23 郑州铁路职业技术学院 Network security encryption method and system based on artificial intelligence

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190273509A1 (en) * 2018-03-01 2019-09-05 Crowdstrike, Inc. Classification of source data by neural network processing
WO2019185987A1 (en) * 2018-03-29 2019-10-03 Nokia Technologies Oy Entropy-friendly neural network representations and their use in training and using neural networks such as autoencoders
CN112789625A (en) * 2018-09-27 2021-05-11 渊慧科技有限公司 Committed information rate variational self-encoder
WO2021137745A1 (en) * 2019-12-30 2021-07-08 Unibap Ab A method for detection of imperfections in products
WO2021165887A1 (en) * 2020-02-19 2021-08-26 Insilico Medicine Ip Limited Adversarial autoencoder architecture for methods of graph to sequence models
WO2021205066A1 (en) * 2020-04-09 2021-10-14 Nokia Technologies Oy Training a data coding system for use with machines
CN111709491A (en) * 2020-06-30 2020-09-25 平安科技（深圳）有限公司 Anomaly detection method, device and equipment based on self-encoder and storage medium
CN111967502A (en) * 2020-07-23 2020-11-20 电子科技大学 Network intrusion detection method based on conditional variation self-encoder
CN112115443A (en) * 2020-11-19 2020-12-22 索信达(北京)数据技术有限公司 Terminal user authentication method and system
CN113691513A (en) * 2021-08-13 2021-11-23 郑州铁路职业技术学院 Network security encryption method and system based on artificial intelligence

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
任伟: "基于稀疏自编码深度神经网络的入侵检测方法", 《移动通信》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116957991A (en) * 2023-09-19 2023-10-27 北京渲光科技有限公司 Three-dimensional model complement method and three-dimensional model complement model generation method
CN116957991B (en) * 2023-09-19 2023-12-15 北京渲光科技有限公司 Three-dimensional model completion method

Also Published As

Publication number Publication date
CN114386067B (en) 2022-08-23

Similar Documents

Publication Publication Date Title
CN111201749B (en) Method and system for secure data communication
EP2775660B1 (en) Message authentication method in communication system and communication system
CN114386067B (en) Equipment production data safe transmission method and system based on artificial intelligence
US9521120B2 (en) Method for securely transmitting control data from a secure network
CN1455341A (en) Method for long-distance changing of communication cipher code
CN116074123B (en) Method for safely transmitting digital information of Internet of things
CN114205133B (en) Information security enhancement method for vehicle-mounted CAN network and electronic equipment
US11121855B2 (en) System and method for secure exchange
CN112269989A (en) Computer data safety system
CN111641503A (en) Trusted data transmission method for multiple unmanned platforms
CN108173854A (en) A kind of safety monitoring method of electric power proprietary protocol
Kim et al. Unknown payload anomaly detection based on format and field semantics inference in cyber-physical infrastructure systems
CN114124549A (en) Method, system and device for safely accessing mails based on visible light system
CN106549962B (en) Method for realizing communication protocol of universal intelligent control platform
CN116049905B (en) Tamper-proof system based on detecting system file change
US11005764B2 (en) Methods and systems for transmission control protocol (TCP) communications
CN110278068B (en) LoRa communication encryption system based on chaos sequence
CN116112408B (en) Industrial Internet transmission data safety supervision method and system
CN108462690A (en) A kind of numerically-controlled machine tool device data remote communication method
CN115208881B (en) Block chain consensus method, equipment and storage medium
Jaoudi et al. Conversion of an unsupervised anomaly detection system to spiking neural network for car hacking identification
Zha et al. Outlier‐resistant quantized control for T‐S fuzzy systems under multi‐channel‐enabled round‐robin protocol and deception attacks
Liu et al. Resisting the Bergen-Hogan attack on adaptive arithmetic coding
Koh et al. Secret Key and Tag Generation for IIoT Systems Based on Edge Computing
CN112351041B (en) Network request tamper-proof method applied to logistics network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20221122

Address after: 067000 Electrical and Mechanical Laboratory Building 107-2, Chengde Petroleum College, Chengde Development Zone, Hebei Province

Patentee after: Chengde Qingyun Information Technology Co.,Ltd.

Address before: 067000 No. 2 College Road, Chengde High-tech Industrial Development Zone, Hebei Province

Patentee before: Hebei Petroleum University of Technology