CN116055651B - Shared access method, device, equipment and medium for multi-center energy economic data - Google Patents
Shared access method, device, equipment and medium for multi-center energy economic data Download PDFInfo
- Publication number
- CN116055651B CN116055651B CN202310032496.XA CN202310032496A CN116055651B CN 116055651 B CN116055651 B CN 116055651B CN 202310032496 A CN202310032496 A CN 202310032496A CN 116055651 B CN116055651 B CN 116055651B
- Authority
- CN
- China
- Prior art keywords
- image data
- data
- energy economy
- target
- encoder
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 230000006870 function Effects 0.000 claims description 42
- 238000012549 training Methods 0.000 claims description 24
- 238000003860 storage Methods 0.000 claims description 16
- 238000011176 pooling Methods 0.000 claims description 12
- 230000004913 activation Effects 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 10
- 230000007246 mechanism Effects 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 11
- 238000013528 artificial neural network Methods 0.000 description 8
- 238000013135 deep learning Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- 230000006872 improvement Effects 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000013478 data encryption standard Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005489 elastic deformation Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/44—Secrecy systems
- H04N1/4406—Restricting access, e.g. according to user identity
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/44—Secrecy systems
- H04N1/448—Rendering the image unintelligible, e.g. scrambling
- H04N1/4486—Rendering the image unintelligible, e.g. scrambling using digital data encryption
-
- 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
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Storage Device Security (AREA)
Abstract
The application discloses a sharing access method, a device, equipment and a medium of multi-center energy economy data, which are characterized in that the energy economy image data of a plurality of data centers are acquired, the energy economy image data are subjected to data alignment to obtain target image data, and the systematicness and the stability are enhanced; encrypting the target image data by utilizing an encoder in the convolutional self-encoder model which is trained in advance to obtain hidden layer space characteristics of the target image data; and establishing an encryption database of the energy economy image data based on the hidden layer space characteristics, wherein the hidden layer space characteristics in the encryption database are used for decrypting the target decoder corresponding to the authority access method to obtain the target energy economy image data. The complex condition of adapting to the energy economy data is realized, only the authority access party with the access authority can call the decoder, and only the specific energy economy data can be accessed, so that the reliability, the safety and the confidentiality of the energy economy data are enhanced.
Description
Technical Field
The present application relates to the field of data security technologies, and in particular, to a method, an apparatus, a device, and a medium for sharing access to multi-center energy economic data.
Background
The energy economy data is rapidly increased, and the characteristics of more data centers, large data volume and the like are presented. When each organization performs data sharing interaction, the security, confidentiality and effective management and access of the data resources are all required to be absolutely protected.
Currently, the data sharing method mainly comprises identity authentication, information resource storage and shared resource access control. However, the current method is an application of small terminal sharing environment from remote medical treatment, remote physical environment detection, unmanned vehicle driving and the like, and has the characteristics of more devices, small data volume and simple data types. However, the large energy economy data has the complex situations of multiple centers, multiple data common storage and multiple mechanisms sharing read access, and the current method is difficult to cope with large-scale and various energy economy data scenes. Meanwhile, in terms of security protection, the current method generally relies on conventional mathematical encryption algorithms such as DES (data encryption standard) and AES (advanced encryption standard), but cannot accommodate energy economy data containing a large number of images, nor can it meet the high confidentiality requirements of the energy economy data. Therefore, the current method has the problems that the method cannot adapt to the complex condition of the energy economic data and cannot meet the high confidentiality requirement of the energy economic data.
Disclosure of Invention
The application provides a shared access method, a device, equipment and a medium for multi-center energy economic data, which are used for solving the technical problems that the prior method cannot adapt to the complex situation of the energy economic data and cannot meet the high confidentiality requirement of the energy economic data.
In order to solve the above technical problems, in a first aspect, the present application provides a method for sharing access to multi-center energy economy data, including:
acquiring energy economy image data of a plurality of data centers;
carrying out data alignment on the energy economy image data to obtain target image data; the energy economy image data are subjected to matrixing and normalization to obtain intermediate image data; extracting high-dimensional characteristics of the intermediate image data by using a preset multi-layer attention mechanism to finish data alignment of the energy economy image so as to obtain the target image data;
encrypting the target image data by utilizing an encoder in a convolutional self-encoder model which is trained in advance to obtain hidden layer space characteristics of the target image data; the target image data is input to the encoder, and after multiple convolution operations and multiple pooling operations are performed on the target image data by the encoder, the hidden layer spatial feature is obtained, wherein the convolution operations are as follows: h is a k ＝σ(x×w k +b k )；h k Represents hidden layer space characteristics obtained after the kth convolution operation is carried out on target image data, sigma represents an activation function, x represents the target image data, and w k And b k Representing the convolution parameters in the kth convolution operation; performing iterative training on a preset convolution self-encoder model based on a preset energy economy image data sample, and calculating a loss function of each iteration; updating model parameters of the preset convolution self-encoder model based on the loss function by using a back propagation algorithm until the preset convolution self-encoder model reaches a preset convergence condition to obtain the convolution self-encoder model; wherein, the expression of the loss function is:the expression of the back propagation algorithm is: />Wherein L is loss Representing the loss function value, x i Representing the output value of the preset convolution self-encoder model at the ith iteration, y i Representing the expected value; x represents target image data δh k Representing the hidden layer spatial feature increment obtained by the encoder, δy representing the original image data increment obtained by the decoder,/->Representing hidden layer space characteristics obtained by the encoder;
establishing an encryption database of the energy economy image data based on the hidden layer space features, wherein the hidden layer space features in the encryption database are used for encrypting a target decoder corresponding to the authority access method to obtain target energy economy image data; the access request of the target authority access party is acquired; based on the access request, reading hidden layer space characteristics in the encryption database, and calling a target decoder corresponding to the target authority access party; decrypting the hidden layer space features through the target decoder to obtain the target energy economy image data; the convolutional self-encoder model includes a plurality of decoders, one for each authorized party.
In some implementations, the decrypting, by the target decoder, the hidden layer spatial feature to obtain the target energy economy image data includes:
and inputting the hidden layer space characteristics to the target decoder, and obtaining the target energy economy image data after carrying out deconvolution operation and anti-pooling operation on the hidden layer space characteristics for a plurality of times by the target decoder.
In a second aspect, the present application also provides a shared access device for multi-center energy economy data, including:
the acquisition module is used for acquiring the energy economy image data of the plurality of data centers;
the alignment module is used for carrying out data alignment on the energy economy image data to obtain target image data; the energy economy image data are subjected to matrixing and normalization to obtain intermediate image data; extracting high-dimensional characteristics of the intermediate image data by using a preset multi-layer attention mechanism to finish data alignment of the energy economy image so as to obtain the target image data;
the encryption module is used for encrypting the target image data by utilizing an encoder in the convolutional self-encoder model which is trained in advance to obtain hidden layer space characteristics of the target image data; wherein the target image data is input toThe encoder performs multiple convolution operations and multiple pooling operations on the target image data through the encoder to obtain the hidden layer spatial feature, wherein the convolution operations are as follows: h is a k ＝σ(x×w k +b k )；h k Represents hidden layer space characteristics obtained after the kth convolution operation is carried out on target image data, sigma represents an activation function, x represents the target image data, and w k And b k Representing the convolution parameters in the kth convolution operation; performing iterative training on a preset convolution self-encoder model based on a preset energy economy image data sample, and calculating a loss function of each iteration; updating model parameters of the preset convolution self-encoder model based on the loss function by using a back propagation algorithm until the preset convolution self-encoder model reaches a preset convergence condition to obtain the convolution self-encoder model; wherein, the expression of the loss function is:the expression of the back propagation algorithm is: />Wherein L is loss Representing the loss function value, x i Representing the output value of the preset convolution self-encoder model at the ith iteration, y i Representing the expected value; x represents target image data δh k Representing the hidden layer spatial feature increment obtained by the encoder, δy representing the original image data increment obtained by the decoder,/->Representing hidden layer space characteristics obtained by the encoder;
the establishing module is used for establishing an encryption database of the energy economic image data based on the hidden layer space characteristics, wherein the hidden layer space characteristics in the encryption database are used for encrypting a target decoder corresponding to the authority access method to obtain target energy economic image data; the access request of the target authority access party is acquired; based on the access request, reading hidden layer space characteristics in the encryption database, and calling a target decoder corresponding to the target authority access party; decrypting the hidden layer space features through the target decoder to obtain the target energy economy image data; the convolutional self-encoder model includes a plurality of decoders, one for each authorized party.
In a third aspect, the present application also provides a computer device comprising a processor and a memory for storing a computer program which, when executed by the processor, implements the method of shared access to multi-central energy economy data according to the first aspect.
In a fourth aspect, the present application also provides a computer readable storage medium storing a computer program which, when executed by a processor, implements the method for sharing access of multi-central energy economy data according to the first aspect.
Compared with the prior art, the application has at least the following beneficial effects:
the method comprises the steps of obtaining energy economy image data of a plurality of data centers, carrying out data alignment on the energy economy image data to obtain target image data, and realizing unified processing on multi-center data, so that the complex condition of the energy economy data is adapted, and the systematicness and the stability are enhanced; encrypting the target image data by using an encoder in a convolutional self-encoder model which is trained in advance to obtain hidden layer space characteristics of the target image data, encrypting the data by using the black box characteristics of a neural network and performing authority control by using a decoder, and improving the data security and confidentiality of energy economy data in the storage and transmission processes; and finally, establishing an encryption database of the energy economy image data based on the hidden layer space features, wherein the hidden layer space features in the encryption database are used for encrypting target decoders corresponding to the access method of the authorities to obtain the target energy economy image data, the convolution self-encoder model comprises a plurality of decoders, each authority access party corresponds to one decoder, encryption sharing access of multi-center data is realized, only the authority access party with the access authorities can call the decoders, and the corresponding decoders can only decrypt the corresponding hidden layer space features, namely only access to specific energy economy data, so that the security and confidentiality improvement from two dimensions of the access party and the access data is realized, and the reliability, the security and the confidentiality of the energy economy data are enhanced.
Drawings
FIG. 1 is a flow chart of a method for sharing access to multi-center energy economy data according to an embodiment of the present application;
FIG. 2 is a schematic diagram of an implementation flow of a shared access method according to an embodiment of the present application;
FIG. 3 is a schematic diagram of an encoder structure according to an embodiment of the present application;
FIG. 4 is a schematic diagram of a decoder according to an embodiment of the present application;
FIG. 5 is a schematic diagram of a convolutional self-encoder model shown in an embodiment of the present application;
fig. 6 is a schematic structural diagram of a multi-center energy economy data sharing access device according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of a computer device according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Referring to fig. 1, fig. 1 is a flow chart of a method for sharing access of multi-center energy economic data according to an embodiment of the present application. The multi-center energy economy data sharing access method can be applied to computer equipment, wherein the computer equipment comprises, but is not limited to, intelligent mobile phones, notebook computers, tablet computers, desktop computers, physical servers, cloud servers and the like. As shown in fig. 1, the sharing access method of multi-center energy economy data of the present embodiment includes steps S101 to S104, which are described in detail as follows:
step S101, energy economy image data of a plurality of data centers are acquired.
And step S102, carrying out data alignment on the energy economy image data to obtain target image data.
Step S103, encrypting the target image data by utilizing an encoder in the convolutional self-encoder model which is trained in advance, and obtaining hidden layer space characteristics of the target image data.
Step S104, based on the hidden layer space characteristics, an encryption database of the energy economy image data is established, the hidden layer space characteristics in the encryption database are used for encrypting a target decoder corresponding to the authority access method to obtain the target energy economy image data, the convolution self-encoder model comprises a plurality of decoders, and each authority access party corresponds to one decoder.
In this embodiment, as shown in the implementation flow diagram of fig. 2, in this embodiment, by acquiring energy economic image data of a plurality of data centers, data alignment is performed on the energy economic image data to obtain target image data, so as to implement unified processing on multi-center data, thereby adapting to the complex situation of the energy economic data and enhancing systemicity and stability; encrypting the target image data by using an encoder in a convolutional self-encoder model which is trained in advance to obtain hidden layer space characteristics of the target image data, encrypting the data by using the black box characteristics of a neural network and performing authority control by using a decoder, and improving the data security and confidentiality of energy economy data in the storage and transmission processes; and finally, establishing an encryption database of the energy economy image data based on the hidden layer space features, wherein the hidden layer space features in the encryption database are used for encrypting target decoders corresponding to the access method of the authorities to obtain the target energy economy image data, the convolution self-encoder model comprises a plurality of decoders, each authority access party corresponds to one decoder, encryption sharing access of multi-center data is realized, only the authority access party with the access authorities can call the decoders, and the corresponding decoders can only decrypt the corresponding hidden layer space features, namely only access to specific energy economy data, so that the security and confidentiality improvement from two dimensions of the access party and the access data is realized, and the reliability, the security and the confidentiality of the energy economy data are enhanced.
In some embodiments, the step S102 includes:
matrixing and normalizing the energy economy image data to obtain intermediate image data;
and extracting high-dimensional characteristics of the intermediate image data by using a preset multi-layer attention mechanism to finish data alignment of the energy economy image so as to obtain the target image data.
In this embodiment, the original multi-center energy economy image data is matrixed: cutting images with different sizes into a plurality of image blocks with specified sizes so as to meet the input and output of the fixed size of the neural network, carrying out standardized data enhancement means such as zooming, rotation, overturning, stretching, contrast enhancement, gamma enhancement, gaussian noise addition, elastic deformation and the like on the image blocks, and strengthening the data expression capability. Normalizing the matrixed image data: and normalizing all the image blocks to the [0-1] interval to obtain intermediate image data.
Further, since the multi-center image data originates from different data centers and has data characteristics and data distribution unique to each other, the present embodiment constructs a data alignment module based on an attention mechanism, which is composed of multiple attention layers, in order to align the multi-center data in a high-dimensional space, so that an encoder and a decoder do not need to pay additional attention to data differences. And the intermediate image data is subjected to data alignment to obtain an image characteristic image after data alignment, namely target image data.
In some embodiments, the step S103 includes:
inputting the target image data to the encoder, and performing multiple convolution operations and multiple pooling operations on the target image data by the encoder to obtain the hidden layer spatial feature, wherein the convolution operations are as follows:
h k ＝σ(x×w k +b k )；
h k represents hidden layer space characteristics obtained after the kth convolution operation is carried out on target image data, sigma represents an activation function, x represents the target image data, and w k And b k A convolution parameter representing the kth convolution kernel.
In this embodiment, as shown in the schematic encoder structure of fig. 3, the encoder includes a plurality of convolution layers and a pooling layer, and the image data is input into the encoder and then subjected to convolution and then to pooling by the multi-layer convolutional neural network. Let k convolution kernels, x representing the input, each convolution kernel being defined by the parameter w k And b k Composition, sigma is the activation function, using h k The output feature map is shown. And carrying out pooling operation on the generated characteristic map, and obtaining hidden layer spatial characteristics, namely encrypted image data, through a multi-layer convolutional neural network.
In some embodiments, after the step S104, further includes:
acquiring an access request of the target authority access party;
based on the access request, reading hidden layer space characteristics in the encryption database, and calling a target decoder corresponding to the target authority access party;
and decrypting the hidden layer space features through the target decoder to obtain the target energy economy image data.
In this embodiment, the first half of the model, namely the data preprocessing stage, the data alignment module and its parameters, and the encoder and its parameters of the deep learning convolution self-encoder model are assigned to the total data center, while the decoder module and its parameters are assigned to the various institutions (i.e. rights accessors) with shared access rights. Illustratively, when all data centers collect brand new energy economy image data, the image data only need to be sent into a model, collected into a total data center, and then hidden layer spatial features, namely an encryption database, are obtained through an encoder. Then, each data processing node decrypts the hidden layer space features through a decoder thereof to obtain an energy economy image with access rights of the corresponding node, and the image is very similar to the original image data. And any noise image will be obtained for the energy economy image data without access. The process can protect the integrity of the original data, ensure the safety and achieve the aim of limited shared access of the data.
In some embodiments, the decrypting, by the target decoder, the hidden layer spatial feature to obtain the target energy economy image data includes:
inputting the hidden layer space characteristics to the target decoder, and performing deconvolution operation and antipollution operation on the hidden layer space characteristics for a plurality of times by the target decoder to obtain target energy economy image data, wherein the deconvolution operation is as follows:
y represents target energy economy image data, h k Represents hidden layer space characteristics obtained after the kth convolution operation is performed on the target image data, sigma represents an activation function,the convolution parameter after the kth convolution kernel transpose is represented, and c represents the offset parameter.
In this embodiment, as shown in the schematic decoder structure of fig. 4, the decoder includes a plurality of deconvolution layers and a deconvolution layer, wherein the convolution kernel in the deconvolution layer is a transpose of the convolution kernel in the encoder convolution layer. And taking the output result of the encoder as the input of the decoder, and carrying out deconvolution and anti-pooling on the image by using the multi-layer deconvolution neural network. Each feature map h k The transpose of the convolution kernel corresponding thereto performs the convolution operation and sums the results, adding the offset c, the activation function remains σ. Each different decoder restores the encrypted data into the original image required by the corresponding parameterData enabling individual institutions to achieve data sharing access.
In some embodiments, before the step S103, the method further includes:
performing iterative training on a preset convolution self-encoder model based on a preset energy economy image data sample, and calculating a loss function of each iteration;
and updating model parameters of the preset convolution self-encoder model based on the loss function by using a back propagation algorithm until the preset convolution self-encoder model reaches a preset convergence condition to obtain the convolution self-encoder model.
In this embodiment, the deep neural network approach requires a certain scale of data for training. Thus, it is first necessary to prepare the corresponding image data for training of the convolutional self-encoder model for deep learning. The convolution self-encoder model adopted by the application can automatically and efficiently learn the high-dimensional expression of the data from the data without the label. The training data used are derived from the energy economy image data of different data centers and are all aligned by the data. The method for training the model is based on self-supervision learning, and does not need data labels, so that the method is favorable for massive acquisition of data and rapid training of the model. The model training process based on the deep learning is used for mutually transmitting weight parameters through an encoder and a decoder of the convolutional self-encoder model so as to finish the iterative updating of the model.
As shown in fig. 5, which is a schematic diagram of a convolutional self-encoder model, the network structure of the deep learning convolutional self-encoder model used in the present application is composed of two parts: the first part is an encoder that encrypts multi-center image data; the second part is a plurality of decoders, and the data sharing access is realized by respectively carrying out corresponding decryption operation on the generated encrypted data. For an input image, an encoder extracts the code, maps features to a hidden space, the features of the hidden space represent encrypted data, and then each decoder decodes the encrypted data to obtain an image very close to the original data so as to realize multi-center data sharing access.
In some embodiments, the training purpose of the network is to have each decoder get as similar an image as possible to the input of the encoder for each energy efficient image data that has access to it, while getting any noise output for energy efficient image data that has no access to it. Therefore, the minimum mean square error function can be adopted as a loss function, the training loss of the weighted energy economy image data is continuously reduced through a back propagation algorithm, and the back propagation of the training loss of the weighted energy economy image data is cut off, so that the training purpose is achieved. The loss function formula is as follows, noting that the unweighted data will not be back-propagated:
wherein L is loss Representing the loss function value, x i Representing the output value of the preset convolution self-encoder model at the ith iteration, y i Representing the expected value.
Through a back propagation algorithm, the loss function derives w to update the weight w, and the formula is as follows:
x represents target image data δh k Representing the hidden layer space feature increment obtained by the encoder, delta y represents the original image data increment obtained by the decoder, w k Model parameters representing a preset convolutional self-encoder model,representing hidden layer spatial features obtained by the encoder.
Optionally, the training optimization algorithm adopts Adam optimization algorithm, the initial learning rate is set to 0.001, and the weight attenuation parameter is set to 5×10 -4 . If the training of 20 cases of data is not reduced after the single case error continuously passes, the learning rate is multiplied by the attenuation coefficient of 0.8. Since the scale of the input data is different in each case, the input data is processed onceThe training batch was set to 1 and the number of learning iterations was 100. The network training employs a BP feedback propagation algorithm to update the parameters of both the encoder and decoder simultaneously using the loss function. The web learning updates the parameters once for each batch. After each stage is subjected to iterative learning, the model judges the total error of each stage, if the current error is smaller than the error of the previous iteration, the current model of the current stage is saved, and then training is continued. If the training reaches the maximum iteration number or the total error does not drop continuously through 10 iterations, the training is stopped.
Compared with the current multi-center energy economy big data sharing access framework, the framework abandons the traditional encryption algorithm, and performs data encryption by deep learning on a model, so that the framework has good safety, and the problem of data leakage is not required in the storage and transmission processes.
Compared with the current multi-center energy economy big data sharing access method, the method adopts the encryption model based on the deep neural network, encrypts the data and reduces the dimension by utilizing the black box characteristic of the neural network, and ensures the robustness and the stability of a sharing access system.
Compared with the current encryption method of the multi-center energy economy big data, the method has the advantages that the black box characteristic of the neural network is utilized to control the authority of the energy economy data, and the data encrypted by the encoder can only be decrypted and shared to be accessed by the corresponding decoder, so that the safety and confidentiality of data information in storage and transmission are ensured.
In order to execute the multi-center energy economy data sharing access method corresponding to the method embodiment, corresponding functions and technical effects are achieved. Referring to fig. 6, fig. 6 shows a block diagram of a multi-center energy economy data sharing access device according to an embodiment of the present application. For convenience of explanation, only the parts related to this embodiment are shown, and the sharing access device for multi-center energy economy data provided in this embodiment of the present application includes:
an acquisition module 601, configured to acquire energy economy image data of a plurality of data centers;
an alignment module 602, configured to perform data alignment on the energy economy image data to obtain target image data;
an encryption module 603, configured to encrypt the target image data by using an encoder in the convolutional self-encoder model that is trained in advance, to obtain hidden layer spatial features of the target image data;
the establishing module 604 is configured to establish an encryption database of the energy economic image data based on the hidden layer spatial feature, where the hidden layer spatial feature in the encryption database is used for encrypting a target decoder corresponding to the rights access method to obtain the target energy economic image data, and the convolutional self-encoder model includes a plurality of decoders, and each rights access party corresponds to one decoder.
In some embodiments, the alignment module 602 is specifically configured to:
matrixing and normalizing the energy economy image data to obtain intermediate image data;
and extracting high-dimensional characteristics of the intermediate image data by using a preset multi-layer attention mechanism to finish data alignment of the energy economy image so as to obtain the target image data.
In some embodiments, the encryption module 603 is specifically configured to:
inputting the target image data to the encoder, and performing multiple convolution operations and multiple pooling operations on the target image data by the encoder to obtain the hidden layer spatial feature, wherein the convolution operations are as follows:
h k ＝σ(x×w k +b k )；
h k represents hidden layer space characteristics obtained after the kth convolution operation is carried out on target image data, sigma represents an activation function, x represents the target image data, and w k And b k A convolution parameter representing the kth convolution kernel.
In some embodiments, the shared access apparatus further comprises:
the second acquisition module is used for acquiring the access request of the target authority access party;
the reading module is used for reading hidden layer space characteristics in the encryption database based on the access request and calling a target decoder corresponding to the target authority access party;
and the decryption module is used for decrypting the hidden layer space features through the target decoder to obtain the target energy economy image data.
In some embodiments, the decryption module is specifically configured to:
inputting the hidden layer space characteristics to the target decoder, and performing deconvolution operation and antipollution operation on the hidden layer space characteristics for a plurality of times by the target decoder to obtain target energy economy image data, wherein the deconvolution operation is as follows:
y represents target energy economy image data, h k Represents hidden layer space characteristics obtained after the kth convolution operation is performed on the target image data, sigma represents an activation function,the convolution parameter after the kth convolution kernel transpose is represented, and c represents the offset parameter.
In some embodiments, the shared access apparatus further comprises:
the training module is used for carrying out iterative training on a preset convolution self-encoder model based on a preset energy economy image data sample and calculating a loss function of each iteration;
and the updating module is used for updating the model parameters of the preset convolution self-encoder model based on the loss function by utilizing a back propagation algorithm until the preset convolution self-encoder model reaches a preset convergence condition to obtain the convolution self-encoder model.
In some embodiments, the loss function is expressed as:
the expression of the back propagation algorithm is:
wherein L is loss Representing the loss function value, x i Representing the output value of the preset convolution self-encoder model at the ith iteration, y i Representing the expected value; x represents target image data δh k Representing the hidden layer space feature increment obtained by the encoder, delta y represents the original image data increment obtained by the decoder, w k Model parameters representing a preset convolutional self-encoder model,representing hidden layer spatial features obtained by the encoder.
The sharing access device for multi-center energy economy data can implement the sharing access method for multi-center energy economy data in the method embodiment. The options in the method embodiments described above are also applicable to this embodiment and will not be described in detail here. The rest of the embodiments of the present application may refer to the content of the above method embodiments, and in this embodiment, no further description is given.
Fig. 7 is a schematic structural diagram of a computer device according to an embodiment of the present application. As shown in fig. 7, the computer device 7 of this embodiment includes: at least one processor 70 (only one is shown in fig. 7), a memory 71 and a computer program 72 stored in the memory 71 and executable on the at least one processor 70, the processor 70 implementing the steps in any of the method embodiments described above when executing the computer program 72.
The computer device 7 may be a smart phone, a tablet computer, a desktop computer, a cloud server, or the like. The computer device may include, but is not limited to, a processor 70, a memory 71. It will be appreciated by those skilled in the art that fig. 7 is merely an example of the computer device 7 and is not limiting of the computer device 7, and may include more or fewer components than shown, or may combine certain components, or different components, such as may also include input-output devices, network access devices, etc.
The processor 70 may be a central processing unit (Central Processing Unit, CPU) and the processor 70 may be other general purpose processors, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), an off-the-shelf programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 71 may in some embodiments be an internal storage unit of the computer device 7, such as a hard disk or a memory of the computer device 7. The memory 71 may in other embodiments also be an external storage device of the computer device 7, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card) or the like, which are provided on the computer device 7. Further, the memory 71 may also include both an internal storage unit and an external storage device of the computer device 7. The memory 71 is used for storing an operating system, application programs, boot loader (BootLoader), data, other programs, etc., such as program codes of the computer program. The memory 71 may also be used for temporarily storing data that has been output or is to be output.
In addition, the embodiment of the present application further provides a computer readable storage medium, where a computer program is stored, where the computer program is executed by a processor to implement the steps in any of the above-mentioned method embodiments.
Embodiments of the present application provide a computer program product which, when run on a computer device, causes the computer device to perform the steps of the method embodiments described above.
In several embodiments provided by the present application, it will be understood that each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing embodiments have been provided for the purpose of illustrating the general principles of the present application, and are not to be construed as limiting the scope of the application. It should be noted that any modifications, equivalent substitutions, improvements, etc. made by those skilled in the art without departing from the spirit and principles of the present application are intended to be included in the scope of the present application.
Claims (5)
1. A method for shared access to multi-center energy economy data, comprising:
acquiring energy economy image data of a plurality of data centers;
carrying out data alignment on the energy economy image data to obtain target image data; the energy economy image data are subjected to matrixing and normalization to obtain intermediate image data; extracting high-dimensional characteristics of the intermediate image data by using a preset multi-layer attention mechanism to finish data alignment of the energy economy image so as to obtain the target image data;
encrypting the target image data by utilizing an encoder in a convolutional self-encoder model which is trained in advance to obtain hidden layer space characteristics of the target image data; the target image data is input to the encoder, and after multiple convolution operations and multiple pooling operations are performed on the target image data by the encoder, the hidden layer spatial feature is obtained, wherein the convolution operations are as follows: h is a k ＝σ(x×w k +b k )；h k Represents hidden layer space characteristics obtained after the kth convolution operation is carried out on target image data, sigma represents an activation function, x represents the target image data, and w k And b k Representing the convolution parameters in the kth convolution operation; performing iterative training on a preset convolution self-encoder model based on a preset energy economy image data sample, and calculating a loss function of each iteration; updating model parameters of the preset convolution self-encoder model based on the loss function by using a back propagation algorithm until the preset convolution self-encoder model reaches a preset convergence condition to obtain the convolution self-encoder model; wherein, the expression of the loss function is:the expression of the back propagation algorithm is: />Wherein L is loss Representing the loss function value, x i Representing a predetermined convolutional self-encoder modelOutput value of the ith iteration, y i Representing the expected value; x represents target image data δh k Representing the hidden layer spatial feature increment obtained by the encoder, δy representing the original image data increment obtained by the decoder,/->Representing hidden layer space characteristics obtained by the encoder;
establishing an encryption database of the energy economy image data based on the hidden layer space features, wherein the hidden layer space features in the encryption database are used for decrypting a target decoder corresponding to an authority access party to obtain target energy economy image data; the method comprises the steps of obtaining an access request of an authority access party; based on the access request, reading hidden layer space characteristics in the encryption database, and calling a target decoder corresponding to the authority access party; decrypting the hidden layer space features through the target decoder to obtain the target energy economy image data; the convolutional self-encoder model includes a plurality of decoders, one for each authorized party.
2. The method for shared access to multi-center energy economy data according to claim 1, wherein the decrypting, by the target decoder, the hidden layer spatial feature to obtain the target energy economy image data includes:
and inputting the hidden layer space characteristics to the target decoder, and obtaining the target energy economy image data after carrying out deconvolution operation and anti-pooling operation on the hidden layer space characteristics for a plurality of times by the target decoder.
3. A multi-center energy economy data sharing access device, comprising:
the acquisition module is used for acquiring the energy economy image data of the plurality of data centers;
the alignment module is used for carrying out data alignment on the energy economy image data to obtain target image data; the energy economy image data are subjected to matrixing and normalization to obtain intermediate image data; extracting high-dimensional characteristics of the intermediate image data by using a preset multi-layer attention mechanism to finish data alignment of the energy economy image so as to obtain the target image data;
the encryption module is used for encrypting the target image data by utilizing an encoder in the convolutional self-encoder model which is trained in advance to obtain hidden layer space characteristics of the target image data; the target image data is input to the encoder, and after multiple convolution operations and multiple pooling operations are performed on the target image data by the encoder, the hidden layer spatial feature is obtained, wherein the convolution operations are as follows: h is a k ＝σ(x×w k +b k )；h k Represents hidden layer space characteristics obtained after the kth convolution operation is carried out on target image data, sigma represents an activation function, x represents the target image data, and w k And b k Representing the convolution parameters in the kth convolution operation; performing iterative training on a preset convolution self-encoder model based on a preset energy economy image data sample, and calculating a loss function of each iteration; updating model parameters of the preset convolution self-encoder model based on the loss function by using a back propagation algorithm until the preset convolution self-encoder model reaches a preset convergence condition to obtain the convolution self-encoder model; wherein, the expression of the loss function is:the expression of the back propagation algorithm is: />Wherein y is loss Representing the loss function value, x i Representing the output value of the preset convolution self-encoder model at the ith iteration, y i Representing the expected value; x represents target image data δh k Representing the hidden layer spatial feature increment obtained by the encoder, δy representing the original image data increment obtained by the decoder,/->Representing hidden layer space characteristics obtained by the encoder;
the establishing module is used for establishing an encryption database of the energy economic image data based on the hidden layer space characteristics, wherein the hidden layer space characteristics in the encryption database are used for decrypting a target decoder corresponding to the authority access party to obtain target energy economic image data; wherein the convolutional self-encoder model comprises a plurality of decoders, one for each authorized party.
4. A computer device comprising a processor and a memory for storing a computer program which when executed by the processor implements the method of shared access to multi-central energy economy data according to any one of claims 1 to 2.
5. A computer-readable storage medium, characterized in that it stores a computer program which, when executed by a processor, implements the multi-center energy economy data sharing access method according to any one of claims 1 to 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310032496.XA CN116055651B (en) | 2023-01-06 | 2023-01-06 | Shared access method, device, equipment and medium for multi-center energy economic data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310032496.XA CN116055651B (en) | 2023-01-06 | 2023-01-06 | Shared access method, device, equipment and medium for multi-center energy economic data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116055651A CN116055651A (en) | 2023-05-02 |
| CN116055651B true CN116055651B (en) | 2023-11-10 |
Family
ID=86123318
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310032496.XA Active CN116055651B (en) | 2023-01-06 | 2023-01-06 | Shared access method, device, equipment and medium for multi-center energy economic data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116055651B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08297638A (en) * | 1995-04-26 | 1996-11-12 | Nippon Telegr & Teleph Corp <Ntt> | User authentication system |
| CN1810030A (en) * | 2003-06-20 | 2006-07-26 | 纳格拉影像股份有限公司 | Decoder system for processing pay-tv data and method for managing at least two decoders |
| KR20190011180A (en) * | 2017-07-24 | 2019-02-01 | 삼성전자주식회사 | Electronic device and Method for controlling the electronic device |
| CN111310734A (en) * | 2020-03-19 | 2020-06-19 | 支付宝(杭州)信息技术有限公司 | Face recognition method and device for protecting user privacy |
| CN111507386A (en) * | 2020-04-09 | 2020-08-07 | 中国科学院声学研究所南海研究站 | Method and system for detecting encrypted communication of storage file and network data stream |
| CN111553320A (en) * | 2020-05-14 | 2020-08-18 | 支付宝(杭州)信息技术有限公司 | Feature extraction method for protecting personal data privacy, model training method and hardware |
| CN111951781A (en) * | 2020-08-20 | 2020-11-17 | 天津大学 | Chinese prosody boundary prediction method based on graph-to-sequence |
| CN113038089A (en) * | 2021-05-21 | 2021-06-25 | 浙江宇视科技有限公司 | Intelligent identification dynamic self-decoding processing method and device, electronic equipment and storage medium |
| CN114764602A (en) * | 2022-05-07 | 2022-07-19 | 东南大学 | Short-term rainfall prediction method based on space-time attention and data fusion |
-
2023
- 2023-01-06 CN CN202310032496.XA patent/CN116055651B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08297638A (en) * | 1995-04-26 | 1996-11-12 | Nippon Telegr & Teleph Corp <Ntt> | User authentication system |
| CN1810030A (en) * | 2003-06-20 | 2006-07-26 | 纳格拉影像股份有限公司 | Decoder system for processing pay-tv data and method for managing at least two decoders |
| KR20190011180A (en) * | 2017-07-24 | 2019-02-01 | 삼성전자주식회사 | Electronic device and Method for controlling the electronic device |
| CN111310734A (en) * | 2020-03-19 | 2020-06-19 | 支付宝(杭州)信息技术有限公司 | Face recognition method and device for protecting user privacy |
| CN111507386A (en) * | 2020-04-09 | 2020-08-07 | 中国科学院声学研究所南海研究站 | Method and system for detecting encrypted communication of storage file and network data stream |
| CN111553320A (en) * | 2020-05-14 | 2020-08-18 | 支付宝(杭州)信息技术有限公司 | Feature extraction method for protecting personal data privacy, model training method and hardware |
| CN111951781A (en) * | 2020-08-20 | 2020-11-17 | 天津大学 | Chinese prosody boundary prediction method based on graph-to-sequence |
| CN113038089A (en) * | 2021-05-21 | 2021-06-25 | 浙江宇视科技有限公司 | Intelligent identification dynamic self-decoding processing method and device, electronic equipment and storage medium |
| CN114764602A (en) * | 2022-05-07 | 2022-07-19 | 东南大学 | Short-term rainfall prediction method based on space-time attention and data fusion |
Non-Patent Citations (1)
| Title |
|---|
| Stacked Convolutional Auto-Encoders for Hierarchical Feature Extraction;Jonathan Masci;Ueli Meier;Dan Cireşan & Jürgen Schmidhuber;《Artificial Neural Networks and Machine Learning – ICANN 2011》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116055651A (en) | 2023-05-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11902413B2 (en) | Secure machine learning analytics using homomorphic encryption | |
| US11222138B2 (en) | Privacy-preserving machine learning in the three-server model | |
| Mohassel et al. | ABY3: A mixed protocol framework for machine learning | |
| US20220092216A1 (en) | Privacy-preserving machine learning in the three-server model | |
| WO2021228149A1 (en) | Private data protection method, system, and device | |
| CN108595927B (en) | Identity authentication, unlocking and payment method and device, storage medium, product and equipment | |
| US20230106584A1 (en) | Securing User-Entered Text In-Transit | |
| CN106997439A (en) | TrustZone-based data encryption and decryption method and device and terminal equipment | |
| US10891391B2 (en) | Remote file storage with multiple access levels | |
| CN113239391B (en) | Third-party-free logistic regression federal learning model training system and method | |
| CN108881230B (en) | Secure transmission method and device for government affair big data | |
| CN108170753B (en) | Key-Value database encryption and security query method in common cloud | |
| Mohanty et al. | PANDORA: Preserving privacy in PRNU-based source camera attribution | |
| CN113055153B (en) | Data encryption method, system and medium based on fully homomorphic encryption algorithm | |
| CN110069907A (en) | Big data source tracing method and system based on digital watermarking | |
| CN116055651B (en) | Shared access method, device, equipment and medium for multi-center energy economic data | |
| CN115952529B (en) | User data processing method, computing device and storage medium | |
| CN117349685A (en) | Clustering method, system, terminal and medium for communication data | |
| CN116094686B (en) | Homomorphic encryption method, homomorphic encryption system, homomorphic encryption equipment and homomorphic encryption terminal for quantum convolution calculation | |
| CN111104693A (en) | Android platform software data cracking method, terminal device and storage medium | |
| US11455404B2 (en) | Deduplication in a trusted execution environment | |
| Lemieux et al. | Protecting Privacy in Digital Records: The Potential of Privacy-Enhancing Technologies | |
| Pham et al. | Edges of Interpolating Tetrahedron Based Encryption Algorithm for 3D Printing Model | |
| CN115396222B (en) | Device instruction execution method, system, electronic device and readable storage medium | |
| US11424909B1 (en) | System and method for protecting data that is exported to an external entity |
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 |