DCS cross-domain data collaborative optimization method based on privacy calculation

Abstract

The invention relates to the field of data collaboration, and discloses a DCS cross-domain data collaborative optimization method based on privacy computing, which is used for realizing efficient collaboration and dynamic prediction of cross-domain data on the premise of ensuring data security. Comprising the following steps: extracting topological features from generator set, power grid dispatching and equipment monitoring data, forming an encryption feature set through algebraic coding and homomorphic encryption, constructing a hierarchical encryption architecture based on secure multi-party calculation and hierarchical storage management, supporting cross-system feature alignment and weighted fusion, generating a dynamically optimized global encryption topological graph, and generating a dynamic optimization global encryption topological graph. High-precision state deduction is realized by combining uncertainty quantization, a feature extraction strategy, an encryption parameter and a prediction model are continuously optimized through a closed-loop feedback mechanism, and a privacy protection audit log is generated. The method breaks through the limitation of a traditional data island, improves the cross-domain cooperation efficiency on the premise of guaranteeing privacy security, and is suitable for real-time decision making and intelligent optimization of high-security demand scenes such as electric power and energy.

Description

Technical Field

[0001] This invention relates to the field of data collaboration, and in particular to a DCS cross-domain data collaboration optimization method based on privacy computing. Background Technology

[0002] With the rapid development of the energy internet, distributed control systems (DCS) are increasingly being used in the power industry, and their cross-domain data collaboration capabilities have become crucial for improving the operational efficiency and security of power systems. However, traditional DCS cross-domain data collaboration faces two core challenges: data privacy protection and cross-domain feature fusion.

[0003] Core data in power systems, such as generator operation data, grid dispatch data, and equipment status monitoring data, are often distributed across different power plant subsystems and energy entities, involving commercially confidential and security-sensitive information. Existing technologies mostly employ centralized data aggregation or plaintext transmission methods, leading to a high risk of data leakage and failing to meet regulatory requirements for privacy protection.

[0004] On the other hand, cross-domain data collaboration requires feature fusion and state prediction of multi-source heterogeneous data. Existing methods mostly rely on centralized modeling, extracting features based on only single-dimensional data and ignoring multi-parameter coupling relationships, resulting in insufficient topological feature representation capabilities; cross-domain data fusion requires decryption and post-processing, which has high computational overhead and poor real-time performance, failing to meet the second-level response requirements of power systems; traditional prediction functions rely on fixed parameters, making it difficult to adapt to dynamic scenarios such as equipment aging and changes in operating conditions, leading to a decrease in prediction accuracy; Therefore, we propose a privacy-preserving computation-based cross-domain data collaborative optimization method for DCS to address the above problems. Summary of the Invention

[0005] This invention provides a privacy-preserving computation-based method for cross-domain data collaboration and optimization in DCS, which enables efficient collaboration and dynamic prediction of cross-domain data while ensuring data security.

[0006] The first aspect of this invention provides a DCS cross-domain data collaborative optimization method based on privacy computing. The method includes: each power plant subsystem extracting topological features from generator unit operation data, grid dispatch data, and equipment status monitoring data to generate a power plant operation topological feature set; encrypting the power plant operation topological feature set using a homomorphic encryption algorithm to generate an encrypted power plant topological feature set; fusing the encrypted power plant topological feature set in the encrypted state to generate a global power plant encrypted topological feature map; constructing a power plant operation prediction function in the encrypted state using the global power plant encrypted topological feature map; and, upon receiving an operation status query request, inputting the encrypted topological features generated from the query operation parameters into the power plant operation prediction function, performing matching calculations with the global power plant encrypted topological feature map in the encrypted state, outputting the encrypted prediction result, and authorizing decryption into an operation status prediction value.

[0007] Optionally, in a first implementation of the first aspect of the present invention, the method includes: constructing an operational data complex based on generator vibration data, temperature gradient data, and pressure fluctuation data to generate an operational data complex structure; performing multi-scale homology analysis on the operational data complex structure to generate an operational data homology group sequence; extracting the birth and death times of topological features from the operational data homology group sequence to generate operational state persistence features; and associating and integrating the operational state persistence features with equipment operating condition parameters to generate a power plant operational topological feature set.

[0008] Optionally, in a second implementation of the first aspect of the present invention, the method includes: generating a public key and a private key required for homomorphic encryption based on polynomial ring theory, and generating an encryption key pair; performing algebraic encoding on the topological invariants and persistence patterns in the power plant operating topological feature set, and generating algebraically encoded topological features; using the public key in the encryption key pair to perform homomorphic encryption on the algebraically encoded topological features, and generating a primary encrypted topological feature; performing algebraic operation verification tests on the primary encrypted topological feature, and generating an encrypted power plant topological feature set.

[0009] Optionally, in the third implementation of the first aspect of the present invention, based on the power plant's operational safety level and time cycle factors, homomorphic encryption key pairs are periodically updated to generate key update rules; the encrypted power plant topology feature set is cross-validated through a secure multi-party computation protocol to generate verified secure topology features; based on the power plant's data security level requirements, the verified secure topology features are hierarchically stored and managed to generate a layered encrypted storage architecture; integrity verification information is added to the encrypted topology features in the layered encrypted storage architecture to generate encrypted topology features with integrity protection; and a secure transmission record is generated from the encrypted topology features with integrity protection through a secure channel protocol.

[0010] Optionally, in the fourth implementation of the first aspect of the present invention, the method includes: aligning the feature space of the encrypted power plant topology feature set from the power generation system, transmission and distribution system, and power consumption system to generate aligned encrypted topology features; weighting and fusing the aligned encrypted topology features through homomorphic encryption addition to generate pre-fused encrypted topology features; constructing a graph structure containing node associations and feature similarities based on the pre-fused encrypted topology features to generate a primary global encrypted topology graph; and performing topology consistency verification and redundant feature removal on the primary global encrypted topology graph to generate a global power plant encrypted topology feature graph.

[0011] Optionally, in the fifth implementation of the first aspect of the present invention, based on the final global power plant encrypted topology feature map, the correlation between the topology features of the power generation system, transmission and distribution system, and power consumption system is analyzed to generate cross-system topology association rules; according to the cross-system topology association rules, combined with the importance assessment of the operating status of each subsystem, a dynamic topology feature weight allocation scheme is generated; based on the dynamic topology feature weight allocation scheme, the initially fused encrypted topology features are weighted and optimized in the encrypted state to generate optimized encrypted fusion features; the topology structure changes of the initial global encrypted topology map are continuously monitored to generate a topology evolution trend analysis report; based on the topology evolution trend analysis report, the topology consistency verification standard is dynamically adjusted to generate an adaptively optimized global power plant encrypted topology feature map.

[0012] Optionally, in the sixth implementation of the first aspect of the present invention, the method includes: establishing a parameterized probability distribution model of encrypted topological features based on a global encrypted topological feature map of the power plant, and generating an encrypted statistical manifold structure; defining a metric tensor on the encrypted statistical manifold structure to generate an encrypted Riemann metric field; establishing a shortest path calculation method in the encrypted feature space based on the encrypted Riemann metric field, and generating an encrypted geodesic calculation function; and integrating the encrypted geodesic calculation function with a topological feature matching algorithm to generate a power plant operation prediction function under encrypted conditions.

[0013] Optionally, in the seventh implementation of the first aspect of the present invention, based on the spatiotemporal characteristics of power plant operation data, the encrypted statistical manifold structure is extended to a multi-scale representation to generate a multi-scale encrypted statistical manifold; according to the changing characteristics of power plant operation status, the parameter settings of the encrypted Riemann metric field are dynamically adjusted to generate an adaptive encrypted Riemann metric field; based on the real-time requirements of power plant operation data, the computational efficiency of the encrypted geodesic calculation function is optimized to generate an efficient encrypted geodesic calculation function; based on the power plant operation prediction function, an uncertainty quantification function for the prediction results is added to generate a prediction function with confidence assessment; based on the continuous comparative analysis of actual operation data and prediction results, the parameter configuration of the prediction function with confidence assessment is dynamically optimized to generate a self-learning prediction function.

[0014] Optionally, in the eighth implementation of the first aspect of the present invention, the method includes: extracting topological features based on real-time collected query operating parameters using algebraic topology theory to generate a query topological feature set; encrypting the query topological feature set using a homomorphic encryption algorithm to generate encrypted query topological features; inputting the encrypted query topological features into a power plant operation prediction function, performing similarity matching with the global power plant encrypted topological feature map in the encrypted state to generate an encrypted matching result; using the encrypted matching result, performing state deduction in the encrypted state using manifold interpolation and geodesic distance calculation methods to generate an encrypted prediction result; and performing authorized decryption processing on the encrypted prediction result to generate a final operating state prediction value.

[0015] Optionally, in the ninth implementation of the first aspect of the present invention, the method further includes: constructing an anomaly detection model based on the deviation analysis between the predicted operating status value and the actual operating parameters, and generating an anomaly detection report; dynamically adjusting the topology feature extraction strategy based on the equipment status change information in the anomaly detection report, and generating updated topology feature extraction rules; dynamically optimizing the parameter configuration of the homomorphic encryption algorithm based on the encryption processing efficiency and security evaluation results, and generating an optimized encryption parameter set; using the comparative analysis of the decryption prediction results and the actual operating data, adjusting the parameter settings of the power plant operation prediction function, and generating an optimized prediction function version; and recording the entire process of encrypted calculation operations from topology feature extraction to prediction result output, and generating a complete privacy protection audit log.

[0016] Beneficial effects: From key generation (polynomial ring theory) to algebraic coding topological feature encryption, and then to encryption domain computation (addition / multiplication homomorphism), data is made "usable but invisible", avoiding the risk of leakage caused by decryption in traditional methods; By combining security level with regular key updates over time, and with secure multi-party computation cross-validation, multi-layered dynamic protection is constructed, significantly improving the ability to resist attacks; Based on the complex structure and homology analysis of running data, this method extracts persistent information such as "birth-death time" of topological features, captures multi-parameter coupling relationships, and solves the problem of feature one-sidedness in traditional methods. Homomorphic encryption addition is used to encrypt, align, and weight the features of power generation, transmission, distribution, and consumption systems, generating a global topology map that preserves the original semantics of the data while avoiding decryption overhead. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of an embodiment of the DCS cross-domain data collaborative optimization method based on privacy computing in this invention. Figure 2 This is a schematic diagram of an embodiment of the DCS cross-domain data collaborative optimization device based on privacy computing in this invention. Detailed Implementation

[0018] This invention provides a privacy-preserving computation-based DCS cross-domain data collaborative optimization method, which enables efficient cross-domain data collaboration and dynamic prediction while ensuring data security. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0019] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the DCS cross-domain data collaborative optimization method based on privacy computing in this invention includes: 101. Each power plant subsystem extracts topological features from generator unit operation data, grid dispatch data, and equipment status monitoring data to generate a power plant operation topological feature set; the power plant operation topological feature set includes topological invariants and persistent patterns identified from the operating parameter sequence; It is understood that the executing entity of this invention can be a privacy-based computing-based DCS cross-domain data collaborative optimization device, or it can be a terminal or a server; the specific implementation is not limited here. This embodiment of the invention will be described using a server as an example.

[0020] Specifically, an operational data complex is constructed based on generator vibration data, temperature gradient data, and pressure fluctuation data to generate an operational data complex structure; the operational data complex structure establishes the topological connection relationship between data points through parameterized thresholds; Multi-scale cohomology analysis was performed on the complex structure of the running data to generate a cohomology group sequence of the running data; the cohomology group sequence of the running data records the evolution process of topological holes and connected components at different scales; The birth and death times of topological features are extracted from the homology group sequence of the running data to generate running state persistence features; the running state persistence features represent the stability features of the running state in the form of topological barcodes. The continuous characteristics of the operating status are correlated and integrated with the operating parameters of the equipment to generate a set of power plant operating topology features; the set of power plant operating topology features is used as the input features for subsequent encryption processing.

[0021] 102. The power plant operating topology feature set is encrypted using a homomorphic encryption algorithm to generate an encrypted power plant topology feature set; the encrypted power plant topology feature set maintains the algebraic structure of the operating features and supports computational operations in the encrypted state; Specifically, based on polynomial ring theory, the public and private keys required for homomorphic encryption are generated to create an encryption key pair; the public key in the encryption key pair is used for encryption operations, and the private key is used for authorized decryption operations. The topological invariants and persistence patterns in the power plant's operational topological feature set are algebraically encoded to generate algebraically encoded topological features; the algebraically encoded topological features are then converted into an algebraic representation that supports homomorphic encryption operations. Using the public key in the encryption key pair, homomorphic encryption is performed on the algebraically encoded topological features to generate primary encrypted topological features; the primary encrypted topological features retain the algebraic operation properties of the original topological features. Algebraic operations are performed on the primary encrypted topological features to verify and test them, generating a set of encrypted power plant topological features. The set of encrypted power plant topological features supports topological matching calculations in the encrypted state while maintaining the integrity of the algebraic structure.

[0022] It should be noted that, based on the power plant's operational safety level and time cycle factors, homomorphic encryption key pairs are updated periodically to generate key update rules; the key update rules ensure the forward and backward security of the encryption system. The encrypted power plant topology feature set is cross-validated using a secure multi-party computation protocol to generate a verified secure topology feature; the secure topology feature is then verified for authenticity while remaining encrypted. Based on the data security level requirements of power plants, the verified security topology features are managed in a hierarchical storage manner to generate a layered encrypted storage architecture; the layered encrypted storage architecture is assigned different encryption protection strengths according to the data sensitivity. Integrity verification information is added to the encrypted topology features in the hierarchical encrypted storage architecture to generate encrypted topology features with integrity protection; encrypted topology features with integrity protection can detect data tampering and transmission errors. Secure transmission records are generated by using a secure channel protocol to capture encrypted topology features with integrity protection; the secure transmission records contain the entire transmission trajectory of the encrypted features.

[0023] 103. The encrypted power plant topology feature set is fused under encrypted conditions to generate a global encrypted power plant topology feature map; the global encrypted power plant topology feature map integrates the topological structure information of the operating characteristics of each subsystem; Specifically, the encrypted power plant topology feature sets from the power generation system, transmission and distribution system, and power consumption system are aligned in feature space to generate aligned encrypted topology features; the aligned encrypted topology features ensure the comparability of topology features of different subsystems in a unified metric space; The aligned encrypted topological features are weighted and fused using homomorphic encryption addition to generate a preliminary fused encrypted topological feature; the preliminary fused encrypted topological feature integrates key information about the topological structure of multiple systems. Based on the initially fused encrypted topological features, a graph structure containing node associations and feature similarities is constructed to generate a primary global encrypted topological graph; the primary global encrypted topological graph represents the cross-system topological relationship network in encrypted form; The initial global encrypted topology map is subjected to topology consistency verification and redundant feature removal to generate the final global power plant encrypted topology feature map.

[0024] It should be noted that, based on the final global encrypted topology feature map of the power plant, the correlation between the topology features of the power generation system, transmission and distribution system, and power consumption system is analyzed to generate cross-system topology association rules; the cross-system topology association rules describe the mutual influence patterns between the topology features of different systems. Based on cross-system topology association rules and the importance assessment of the operating status of each subsystem, a dynamic topology feature weight allocation scheme is generated; the dynamic topology feature weight allocation scheme assigns differentiated fusion weights to the topology features of different systems. Based on a dynamic topology feature weighting scheme, the initially fused encrypted topology features are weighted and optimized in the encrypted state to generate optimized encrypted fusion features; the optimized encrypted fusion features reflect the actual contribution of each system's topology features. Continuously monitor the topological changes of the primary global encrypted topology map and generate a topology evolution trend analysis report; the topology evolution trend analysis report records the dynamic changes in the topological characteristics of the power plant system; Based on the topology evolution trend analysis report, the topology consistency verification standard is dynamically adjusted to generate an adaptively optimized global power plant encrypted topology feature map; the adaptively optimized global power plant encrypted topology feature map can reflect the latest state of the system topology.

[0025] 104. Utilizing the global encrypted topological feature map of a power plant, a power plant operation prediction function under encrypted conditions is constructed based on information geometry theory; the power plant operation prediction function can perform state deduction and operation trend prediction in the encrypted feature space; Specifically, based on the global power plant encrypted topology feature map, a parameterized probability distribution model of encrypted topology features is established to generate an encrypted statistical manifold structure; the encrypted statistical manifold structure maps the topology features into a curved geometric space; Define a metric tensor on the cryptographic statistical manifold structure to generate a cryptographic Riemannian metric field; the cryptographic Riemannian metric field characterizes the similarity relationship between topological features in the cryptographic state; Based on the encrypted Riemannian metric field, a method for calculating the shortest path in the encrypted feature space is established, and an encrypted geodesic calculation function is generated. The encrypted geodesic calculation function can calculate the geometric distance between feature points without decryption. By integrating the encrypted geodesic calculation function with the topological feature matching algorithm, a power plant operation prediction function under encrypted conditions is generated; the power plant operation prediction function realizes operation state deduction and trend prediction through manifold interpolation.

[0026] It should be noted that, based on the spatiotemporal characteristics of power plant operation data, the encrypted statistical manifold structure is extended to a multi-scale representation, generating a multi-scale encrypted statistical manifold; the multi-scale encrypted statistical manifold can simultaneously capture the macroscopic trends and microscopic fluctuation characteristics of power plant operation. Based on the changing characteristics of the power plant's operating status, the parameter settings of the encrypted Riemann metric field are dynamically adjusted to generate an adaptive encrypted Riemann metric field; the adaptive encrypted Riemann metric field can automatically optimize feature similarity measurement as the operating conditions change. Based on the real-time requirements of power plant operation data, the computational efficiency of the densified geodesic calculation function is optimized to generate a high-efficiency densified geodesic calculation function; the high-efficiency densified geodesic calculation function significantly improves the calculation speed while maintaining accuracy; Based on the power plant operation prediction function, an uncertainty quantification function for the prediction results is added to generate a prediction function with confidence assessment; the prediction function with confidence assessment can output a reliability index for each prediction value. Based on continuous comparative analysis of actual operating data and prediction results, the parameter configuration of the prediction function with confidence assessment is dynamically optimized to generate a self-learning prediction function; the self-learning prediction function can continuously improve prediction accuracy based on operating feedback.

[0027] 105. When a request for querying the operating status is received, the encrypted topology features generated by querying the operating parameters are input into the power plant operation prediction function. Under encrypted conditions, the function performs matching calculations with the global encrypted topology feature map of the power plant, outputs the encrypted prediction results, and authorizes decryption into the predicted operating status value. Specifically, based on the real-time collected query execution parameters, algebraic topology theory is used to extract topological features and generate a query topological feature set; the query topological feature set includes topological invariants and persistent patterns identified from the query execution parameter sequence; The query topology feature set is encrypted using a homomorphic encryption algorithm to generate encrypted query topology features; the encrypted query topology features have the same algebraic structure and encryption properties as the encrypted power plant topology feature set. The encrypted query topology features are input into the power plant operation prediction function, and similarity matching is performed with the global power plant encrypted topology feature map in the encrypted state to generate encrypted matching results. Using the encrypted matching results, state deduction is performed under the encrypted state through manifold interpolation and geodesic distance calculation methods to generate encrypted prediction results; The encrypted prediction results are authorized and decrypted to generate the final running state prediction value.

[0028] 106. Based on the deviation analysis between the predicted operating status values ​​and the actual operating parameters, construct an anomaly detection model for operating status and generate an anomaly detection report; the anomaly detection report includes equipment operating status assessment and fault warning information; Based on the equipment status change information in the anomaly detection report, the topology feature extraction strategy is dynamically adjusted to generate updated topology feature extraction rules; the updated topology feature extraction rules are used to optimize the accuracy and timeliness of subsequent topology feature extraction. Based on the evaluation results of encryption processing efficiency and security, the parameter configuration of the homomorphic encryption algorithm is dynamically optimized to generate an optimized encryption parameter set; the optimized encryption parameter set balances the requirements of encryption strength and computational efficiency. By comparing and analyzing the decrypted prediction results with actual operating data, the parameter settings of the power plant operation prediction function are adjusted to generate an optimized prediction function version; the optimized prediction function version improves the accuracy and reliability of subsequent predictions. Record the entire encrypted computation process from topological feature extraction to prediction result output, and generate a complete privacy protection audit log; the privacy protection audit log is used to verify the privacy protection compliance and data security integrity of each step.

[0029] In this embodiment of the invention, a homomorphic encryption algorithm is used to encrypt and transform the power plant operating topology feature set, ensuring that the encrypted power plant topology feature set maintains the algebraic structure of the operating features and supports computational operations in the encrypted state. This innovation resolves the contradiction between data privacy protection and collaborative data utilization, achieving effective collaborative optimization of cross-domain data while ensuring data security, and providing a secure and reliable technical solution for data sharing in the power industry. Regularly updating the homomorphic encryption key pair ensures the forward and backward security of the encryption system, further enhancing the privacy protection capabilities of data during long-term storage and transmission.

[0030] Topological features are extracted from generator set operation data, power grid dispatch data, and equipment status monitoring data to construct a complex structure of operation data. Through multi-scale coherence analysis, a coherence group sequence of operation data is generated. The birth time and death time of topological features are extracted to generate persistent features of operation status. This can capture the dynamic changes of power plant operation status more comprehensively and accurately, and provide rich feature information for subsequent prediction and anomaly detection. By associating and integrating the continuous characteristics of the operating status with the operating parameters of the equipment, a set of power plant operating topology features is generated, making the feature extraction closer to the actual operating conditions and improving the practicality and effectiveness of the features. In an encrypted state, the encrypted power plant topology feature set is fused to generate a global encrypted power plant topology feature map. By aligning the encrypted topology features from different subsystems in feature space, weighting and fusing them, and constructing the map structure, an encrypted representation of cross-system topology relationships is achieved, providing an effective tool for a comprehensive understanding of the power plant system's operating status. The initial global encrypted topology map is subjected to topology consistency verification and redundant feature removal to generate the final global power plant encrypted topology feature map, improving the quality and accuracy of the map. Simultaneously, operations such as analyzing cross-system topology association rules based on the map and generating a dynamic topology feature weight allocation scheme further optimize the fusion effect, enabling the map to better reflect the actual state of the system topology. Based on information geometry theory, a power plant operation prediction function is constructed under encrypted conditions. By establishing a parameterized probability distribution model of encrypted topological features, defining a metric tensor, and establishing a shortest path calculation method, state deduction and operation trend prediction are realized in the encrypted feature space. This avoids the privacy leakage risk caused by data decryption, while ensuring the accuracy and reliability of the prediction. By performing multi-scale expansion of the encrypted statistical manifold structure, dynamically adjusting the parameters of the encrypted Riemannian metric field, and optimizing the encrypted geodesic calculation function, the prediction function can better adapt to the spatiotemporal characteristics and real-time requirements of power plant operation data, thereby improving the accuracy and efficiency of prediction. Throughout the technical process, multiple dynamic optimization and adaptive adjustment mechanisms were established. These include analyzing cross-system topology association rules based on the final global encrypted topology feature map of the power plant and generating a dynamic topology feature weight allocation scheme; continuously monitoring topology structure changes in the initial global encrypted topology map and generating a topology evolution trend analysis report; dynamically adjusting topology consistency verification standards; and dynamically adjusting encrypted Riemannian metric field parameters and optimizing the computational efficiency of encrypted geodesic calculation functions based on the characteristics of power plant operating data. These mechanisms enable the technology to automatically adjust and optimize according to actual operating conditions, improving the system's adaptability and robustness. An anomaly detection model is constructed based on the deviation analysis between predicted and actual operating parameters, generating anomaly detection reports. This provides an effective means for equipment operating status assessment and fault early warning. Simultaneously, the topology feature extraction strategy is dynamically adjusted based on equipment status change information in the anomaly detection reports, forming a closed-loop system from data acquisition, feature extraction, predictive analysis to anomaly detection and strategy adjustment. This continuously improves the system's accuracy and reliability, enabling timely detection and handling of potential operational problems.

[0031] Figure 2 This is a schematic diagram of a privacy-preserving computation-based DCS cross-domain data collaborative optimization device 200 provided in an embodiment of the present invention. The privacy-preserving computation-based DCS cross-domain data collaborative optimization device 200 can vary significantly due to differences in configuration or performance. The device 200 includes a transmitter 201, a receiver 202, and a processor 203. The processor 203 can also be a controller. Figure 2 The device is referred to as "controller / processor 203". Optionally, the device 200 may also include a modem processor 205, wherein the modem processor 205 may include an encoder 206, a modulator 207, a decoder 208, and a demodulator 209.

[0032] In one example, transmitter 201 modulates (e.g., analog-to-analog conversion, filtering, amplification, and up-conversion, etc.) the output sample and generates an uplink signal, which is transmitted via an antenna to an access network device. On the downlink, the antenna receives the downlink signal transmitted by the access network device. Receiver 202 modulates (e.g., filtering, amplification, down-conversion, and digitization, etc.) the signal received from the antenna and provides an input sample. In modem processor 205, encoder 206 receives traffic data and signaling messages to be transmitted on the uplink and processes (e.g., formatting, encoding, and interleaving) the traffic data and signaling messages. Modulator 207 further processes (e.g., symbol mapping and modulation) the encoded traffic data and signaling messages and provides an output sample. Demodulator 209 processes (e.g., demodulates) the input sample and provides a symbol estimate. Decoder 208 processes (e.g., deinterleaving and decoding) the symbol estimate and provides decoded data and signaling messages to device 200. Encoder 206, modulator 207, demodulator 209, and decoder 208 can be implemented by a combined modem processor 205. These units process data according to the radio access technology used by the radio access network (e.g., LTE and other evolved systems access technologies). It should be noted that when device 200 does not include modem processor 205, the aforementioned functions of modem processor 205 can also be performed by processor 203.

[0033] The processor 203 controls and manages the operation of the device 200, and is used to execute the processing procedures performed by the device 200 in the above embodiments of this disclosure. For example, the processor 203 is also used to execute various steps of the transmitting or receiving device in the above method embodiments, and / or other steps of the technical solutions described in the embodiments of this disclosure.

[0034] Furthermore, the device 200 may also include a memory 204 for storing program code and data for the device 200.

[0035] Understandable, Figure 2 Only a simplified design of device 200 is shown. In practical applications, device 200 can include any number of transmitters, receivers, processors, modem processors, memory, etc., and all devices that can implement the embodiments of this disclosure are within the protection scope of the embodiments of this disclosure.

[0036] The present invention also provides a privacy-based computing-based DCS cross-domain data collaborative optimization device, the privacy-based computing-based DCS cross-domain data collaborative optimization device including a memory and a processor, the memory storing computer-readable instructions, when the computer-readable instructions are executed by the processor, causing the processor to perform the steps of the privacy-based computing-based DCS cross-domain data collaborative optimization method in the above embodiments.

[0037] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when the instructions are executed on a computer, cause the computer to perform the steps of the privacy-based computing-based DCS cross-domain data collaborative optimization method.

[0038] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0039] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0040] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A DCS cross-domain data collaborative optimization method based on privacy computing, characterized in that, include: Each power plant subsystem extracts topological features from generator unit operation data, power grid dispatch data, and equipment status monitoring data to generate a set of power plant operation topological features. The power plant operation topology feature set is encrypted using a homomorphic encryption algorithm to generate an encrypted power plant topology feature set. The encrypted power plant topology feature set is fused under encrypted conditions to generate a global encrypted power plant topology feature map. Using the global encrypted topology feature map of power plants, a power plant operation prediction function under encrypted conditions is constructed; When a request for querying operating status is received, the encrypted topology features generated from the query operating parameters are input into the power plant operating prediction function. Under encrypted conditions, the function performs matching calculations with the global encrypted topology feature map of the power plant, outputs the encrypted prediction result, and authorizes decryption into the predicted operating status value.

2. The DCS cross-domain data collaborative optimization method based on privacy computing according to claim 1, characterized in that, include: Based on generator set vibration data, temperature gradient data, and pressure fluctuation data, a complex of operational data is constructed to generate the operational data complex structure. Multi-scale homology analysis is performed on the complex structure of the running data to generate a homology group sequence of the running data. The birth and death times of topological features are extracted from the homology group sequence of the running data to generate persistent features of the running state; By associating and integrating the continuous characteristics of the operating status with the operating parameters of the equipment, a set of power plant operating topology features is generated.

3. The DCS cross-domain data collaborative optimization method based on privacy computing according to claim 2, characterized in that, include: Based on polynomial ring theory, generate the public and private keys required for homomorphic encryption, and generate encryption key pairs; The topological invariants and persistent patterns in the power plant's operational topological feature set are algebraically encoded to generate algebraically encoded topological features. Using the public key in the encryption key pair, homomorphic encryption is performed on the algebraic encoded topological features to generate primary encrypted topological features; Algebraic operations are performed to verify the primary encrypted topology features, generating a set of encrypted power plant topology features.

4. The DCS cross-domain data collaborative optimization method based on privacy computing according to claim 3, characterized in that, Based on the power plant's operational safety level and time cycle factors, homomorphic encryption key pairs are updated periodically to generate key update rules; The encrypted power plant topology feature set is cross-validated using a secure multi-party computation protocol to generate a verified secure topology feature. Based on the data security level requirements of power plants, the verified security topology features are managed in a hierarchical storage manner to generate a layered encrypted storage architecture. Integrity verification information is added to the encrypted topology features in the hierarchical encrypted storage architecture to generate encrypted topology features with integrity protection; Secure transmission records are generated using encrypted topology features with integrity protection through a secure channel protocol.

5. The DCS cross-domain data collaborative optimization method based on privacy computing according to claim 1, characterized in that, include: The encrypted power plant topology feature set from the power generation system, transmission and distribution system, and power consumption system is aligned in feature space to generate aligned encrypted topology features; The aligned encrypted topological features are weighted and fused using homomorphic encryption addition to generate a preliminary fused encrypted topological feature. Based on the initially fused encrypted topological features, a graph structure containing node associations and feature similarities is constructed to generate a primary global encrypted topological graph; The initial global encrypted topology map is subjected to topology consistency verification and redundant feature removal to generate a global power plant encrypted topology feature map.

6. The DCS cross-domain data collaborative optimization method based on privacy computing according to claim 5, characterized in that, Based on the final global encrypted topology feature map of the power plant, the correlation between the topology features of the power generation system, transmission and distribution system and power consumption system is analyzed, and cross-system topology association rules are generated. Based on the cross-system topology association rules and the importance assessment of the operating status of each subsystem, a dynamic topology feature weight allocation scheme is generated. Based on the dynamic topology feature weight allocation scheme, the initially fused encrypted topology features are weighted and optimized in the encrypted state to generate optimized encrypted fused features. Continuously monitor the topological changes of the primary global encrypted topology map and generate a topology evolution trend analysis report; Based on the topology evolution trend analysis report, the topology consistency verification standard is dynamically adjusted to generate an adaptively optimized global power plant encrypted topology feature map.

7. The DCS cross-domain data collaborative optimization method based on privacy computing according to claim 1, characterized in that, include: Based on the global power plant encrypted topology feature map, a parameterized probability distribution model of encrypted topology features is established to generate encrypted statistical manifold structure; Define a metric tensor on a cryptographic statistical manifold structure to generate a cryptographic Riemannian metric field; Based on the encrypted Riemannian metric field, a method for calculating the shortest path in the encrypted feature space is established, and an encrypted geodesic calculation function is generated. By integrating the encrypted geodesic calculation function with the topology feature matching algorithm, a power plant operation prediction function under encrypted conditions is generated.

8. The DCS cross-domain data collaborative optimization method based on privacy computing according to claim 7, characterized in that, Based on the spatiotemporal characteristics of power plant operation data, the encrypted statistical manifold structure is extended to a multi-scale representation, generating a multi-scale encrypted statistical manifold. Based on the changing characteristics of the power plant's operating status, the parameter settings of the encrypted Riemann metric field are dynamically adjusted to generate an adaptive encrypted Riemann metric field. Based on the real-time requirements of power plant operation data, the computational efficiency of the densified geodesic calculation function is optimized to generate an efficient densified geodesic calculation function. Based on the power plant operation prediction function, an uncertainty quantification function for the prediction results is added to generate a prediction function with confidence assessment; Based on continuous comparative analysis of actual operating data and prediction results, the parameter configuration of the prediction function with confidence assessment is dynamically optimized to generate a self-learning prediction function.

9. The DCS cross-domain data collaborative optimization method based on privacy computing according to claim 1, characterized in that, include: Based on real-time collected query operation parameters, topological features are extracted using algebraic topology theory to generate a query topological feature set. The query topology feature set is encrypted using a homomorphic encryption algorithm to generate encrypted query topology features; The encrypted query topology features are input into the power plant operation prediction function, and similarity matching is performed with the global power plant encrypted topology feature map in the encrypted state to generate encrypted matching results. Using the encrypted matching results, state deduction is performed under the encrypted state through manifold interpolation and geodesic distance calculation methods to generate encrypted prediction results; The encrypted prediction results are authorized and decrypted to generate the final running state prediction value.

10. The DCS cross-domain data collaborative optimization method based on privacy computing according to claim 1, characterized in that, Also includes: Based on the deviation analysis between the predicted operating status and the actual operating parameters, an anomaly detection model is constructed to generate an anomaly detection report; Based on the equipment status change information in the anomaly detection report, the topology feature extraction strategy is dynamically adjusted to generate updated topology feature extraction rules. Based on the evaluation results of encryption processing efficiency and security, the parameter configuration of the homomorphic encryption algorithm is dynamically optimized to generate an optimized set of encryption parameters. By comparing and analyzing the decrypted prediction results with the actual operating data, the parameter settings of the power plant operation prediction function are adjusted to generate an optimized version of the prediction function. Record the entire process of encrypted computation operations from topological feature extraction to prediction result output, and generate a complete privacy protection audit log.

Images