Privacy computing system for secure sharing of power market data

Abstract

The invention relates to the technical field of electric power data processing, and discloses an electric power market data security sharing privacy computing system, which performs secret sharing on declaration prices through a hybrid encryption module to obtain price fragments, performs homomorphic encryption on declaration electric quantity, and distributes processed data to a first service party and a second service party. Then, the security sorting module cooperatively executes distributed security sorting based on the price fragments, and a quotation index sequence is obtained in a privacy protection mode; and the privacy positioning module and the ciphertext aggregation module utilize the index sequence to determine a marginal unit in a ciphertext domain and complete aggregation calculation of clearing price and electric quantity. And finally, the decryption evidence storage module performs collaborative decryption and block chain evidence storage on the encryption result, and outputs a public and credible clearing result. Therefore, secret sharing and homomorphic encryption are deeply fused, accurate and traceable safe clearing is realized on the premise of not leaking original quotation, and data privacy and result credibility are both considered.

Description

Technical Field

[0001] This application relates to the field of power data processing technology, and more specifically, to a privacy computing system for secure sharing of power market data. Background Technology

[0002] Electricity trading data has become a core strategic resource, possessing immense value in scenarios such as unified clearing to optimize resource allocation. However, a profound conflict exists between data utilization and privacy protection: the declared prices and electricity volumes of power generators are their core trade secrets, and direct sharing would undermine market fairness. Therefore, how to achieve accurate and efficient market clearing calculations while protecting the sensitive data of all parties is a key technological challenge currently facing the electricity market. To address this, distributed privacy-preserving technologies, represented by federated learning, have been explored in the power sector. Their characteristic of data usability without visibility allows parties to collaboratively train machine learning models without leaving their local systems, initially solving the data silo problem in scenarios such as load forecasting.

[0003] However, existing technologies have fundamental limitations when applied to the precise computational scenario of market clearing. First, federated learning is a statistical paradigm used to solve machine learning problems, while clearing is a deterministic operations research computation involving sorting, comparison, and aggregation; their technical goals are mismatched. Second, while homomorphic encryption can aggregate electricity quantities, it cannot efficiently handle the core sorting logic. Therefore, the industry urgently needs a new privacy-preserving computation scheme that transcends the limitations of traditional machine learning frameworks, efficiently completing secure clearing calculations in the electricity market while absolutely guaranteeing data security and result accuracy. Its performance is far from meeting the minute-level real-time requirements of the electricity spot market. Similarly, differential privacy protects data by adding noise, but this sacrifices the accuracy of the settlement results, which is unacceptable for market clearing with strong financial attributes. Summary of the Invention

[0004] In view of the incompatibility of existing privacy protection technologies when applied to electricity market clearing, this application proposes a privacy computing system for secure sharing of electricity market data.

[0005] According to one aspect of this application, a privacy computing system for secure sharing of electricity market data is provided, comprising: a data acquisition module for acquiring the original bid prices and bid volumes of various power generators; a hybrid encryption module for hybrid encryption and distribution of the original bid prices and bid volumes of various power generators to obtain a price shard set held by a first service provider and a price shard set held by a second service provider; a secure sorting module for performing a distributed secure sorting based on secret sharing on the price shard sets held by the first and second service providers to obtain a sorted bid index sequence; a privacy positioning module for performing privacy positioning of the clearing critical point on the sorted bid index sequence based on total market demand to obtain an encrypted index of marginal units; a ciphertext aggregation module for performing ciphertext aggregation of clearing prices and volumes on the encrypted index of marginal units based on an original homomorphic ciphertext library to obtain an encrypted clearing result; and a decryption and evidence storage module for decrypting the encrypted clearing result and storing it on the blockchain to obtain a public clearing price, a private winning bid notification, and an on-chain evidence storage record.

[0006] In one possible implementation, the hybrid encryption module includes: a price sharding unit, used to perform random sharding and complementary sharding calculations on the original declared price to obtain a first price shard and a second price shard; an electricity encryption unit, used to perform homomorphic encryption on the declared electricity to obtain encrypted electricity; and an encapsulation and distribution unit, used to encapsulate and securely distribute the first price shard, the second price shard, and the encrypted electricity into a hybrid data packet to obtain a first encapsulated data packet and a second encapsulated data packet, wherein the first encapsulated data packet is stored on a first service provider, and the second encapsulated data packet is stored on a second service provider.

[0007] In one possible implementation, the encapsulation and distribution unit is further configured to: encapsulate the first price fragment and the encrypted power consumption to obtain a first encapsulated data packet; and encapsulate the second price fragment and the encrypted power consumption to obtain a second encapsulated data packet.

[0008] In one possible implementation, the secure sorting module includes: a multi-party comparison unit, used to perform a distributed secure multi-party comparison on the price shard set held by the first service provider and the price shard set held by the second service provider to obtain a secret sharing set of the comparison results; and a privacy sorting unit, used to perform privacy sorting on the secret sharing set of the comparison results based on unintentional exchange to obtain a secret sharing of the sorted index sequence as the sorted price index sequence.

[0009] In one possible implementation, the privacy positioning module includes: a ciphertext rearrangement unit, used to rearrange the secret sharing of the sorted index sequence and the original ciphertext list of electricity based on an unintentionally selected ciphertext sequence to obtain a rearranged ciphertext list of electricity; an accumulation comparison unit, used to perform iterative ciphertext accumulation and secure comparison on the rearranged ciphertext list of electricity based on the total market demand to obtain a comparison flag bit secret sharing list; and an index extraction unit, used to securely extract and encrypt the marginal unit index based on the comparison flag bit secret sharing list and the secret sharing of the rearranged index to obtain an encrypted index of the marginal unit.

[0010] In one possible implementation, the index extraction unit is further configured to: perform main edge point location and anchor price extraction on the comparison flag bit secret sharing list and the secret sharing of the rearranged index to obtain the anchor price sharing and the main edge point flag sharing vector; based on the anchor price sharing, perform price resonance detection and edge block identification on the secret sharing of the rearranged index to obtain the edge block flag sharing vector; and based on the edge block flag sharing vector and the comparison flag bit secret sharing list, determine the encrypted index of the edge unit.

[0011] In one possible implementation, the ciphertext aggregation module includes: a price determination unit, used to determine the homomorphic ciphertext of the marginal clearing price based on the encrypted index of the marginal unit, the secret sharing of the rearranged index, and the secret sharing list of the original price; an electricity quantity determination unit, used to determine the homomorphic ciphertext list of the winning electricity quantities of each original unit based on the encrypted index of the marginal unit and the secret sharing of the rearranged index; and a ciphertext settlement unit, used to perform homomorphic encryption-based settlement amount calculation on the homomorphic ciphertext list of the winning electricity quantities of each original unit and the homomorphic ciphertext of the marginal clearing price to obtain the encrypted clearing result.

[0012] Compared to existing technologies, the privacy-preserving computation system for secure sharing of electricity market data provided in this application firstly addresses the fundamental limitation of semi-homomorphic encryption in its inability to perform comparative operations by introducing a secure sorting module and a privacy positioning module built with a secret sharing mechanism. This makes it possible to complete the entire market clearing process under privacy protection. Secondly, compared to the exponential computational latency of fully homomorphic encryption schemes, this system entrusts the computational bottleneck (sorting) to a secret sharing protocol with significantly lower communication overhead and computational complexity. This allows the overall clearing time to be controlled within minutes or even seconds, perfectly adapting to the high-frequency trading needs of the electricity spot market. Furthermore, while ensuring absolute accuracy of the calculation results, this system provides a high level of security and trustworthiness. The dual-server architecture ensures that, unless both parties collude, neither can reconstruct the true price data. The decryption and evidence storage module utilizes the immutability and decentralization of blockchain to provide an undeniable audit record for each clearing calculation that can be jointly verified by all participants, effectively avoiding excessive reliance on a single centralized institution. Attached Figure Description

[0013] The above and other objects, features, and advantages of this application will become more apparent from the more detailed description of the embodiments of this application in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain this application and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0014] Figure 1 The illustration shows a schematic block diagram of a privacy computing system for secure sharing of electricity market data according to an embodiment of this application.

[0015] Figure 2 The illustration shows a schematic data flow diagram of a privacy computing system for secure sharing of electricity market data according to an embodiment of this application.

[0016] Figure 3 The illustration shows a schematic block diagram of a hybrid encryption module in a privacy computing system for secure sharing of electricity market data according to an embodiment of this application.

[0017] Figure 4 The illustration shows a schematic block diagram of a secure sorting module in a privacy computing system for secure sharing of electricity market data according to an embodiment of this application.

[0018] Figure 5 The illustration shows a schematic block diagram of a privacy positioning module in a privacy computing system for secure sharing of electricity market data according to an embodiment of this application.

[0019] Figure 6The illustration shows a schematic block diagram of a ciphertext aggregation module in a privacy computing system for secure sharing of electricity market data according to an embodiment of this application. Detailed Implementation

[0020] Hereinafter, exemplary embodiments according to this application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein.

[0021] like Figure 1 and Figure 2 As shown, this application provides a privacy computing system 100 for secure sharing of electricity market data, including: a data acquisition module 110 for acquiring the original bid prices and bid volumes of each power generator; a hybrid encryption module 120 for hybrid encryption and distribution of the original bid prices and bid volumes of each power generator to obtain a price shard set held by a first service provider and a price shard set held by a second service provider; a secure sorting module 130 for performing a distributed secure sorting based on secret sharing on the price shard sets held by the first and second service providers to obtain a sorted bid index sequence; a privacy positioning module 140 for performing privacy positioning of the clearing threshold on the sorted bid index sequence based on total market demand to obtain an encrypted index of marginal units; a ciphertext aggregation module 150 for performing ciphertext aggregation of clearing prices and volumes on the encrypted index of marginal units based on the original homomorphic ciphertext library to obtain an encrypted clearing result; and a decryption and evidence storage module 160 for decrypting the encrypted clearing result and storing it on the blockchain to obtain a public clearing price, a private winning bid notification, and an on-chain evidence storage record.

[0022] For example, the data acquisition module 110 acquires the original declared prices and declared electricity volumes of each power generator. It should be understood that this raw data is the fundamental basis for constructing the market supply curve, determining the market equilibrium point, and calculating the final revenue of each participant; its completeness and authenticity directly determine the validity of the entire market clearing result. More importantly, this raw data, especially the declared prices, directly reflects the power generator's production costs, market strategies, and profit expectations, and constitutes its highest level of trade secrets. Therefore, the data acquisition module does not merely collect data, but securely introduces this highly sensitive business data from various dispersed and distrustful market participants into a unified and trusted computing environment within a complete and rigorous privacy protection framework.

[0023] Specifically, the data acquisition module, acting as a secure data aggregation interface, is responsible for receiving data from various power generator clients. The system connects to the local power trading and pricing systems deployed by each power generator through a dedicated data interface based on standard secure transmission protocols such as TLS / HTTPS. At the start of a clearing cycle, each power generator, as a data producer, submits its declared price and declared electricity data pairs to this system via its client software or API calls. Upon receiving this data, the cloud-based or data center data acquisition module first performs preliminary data format and range checks to ensure the basic validity and integrity of the data; for example, confirming that the price and electricity are valid numerical types and within reasonable business scope. Once successful reception is confirmed, the data acquisition module immediately transmits the acquired original declared price and declared electricity as an indivisible data unit directly in memory to the subsequent hybrid encryption module, without any form of local plaintext storage or logging. This seamless data processing paradigm ensures that highly sensitive original bid plaintext data does not accumulate or remain on any storage node in the system, thus eliminating the risk of data leakage due to server attacks or malicious access to internal data at the source.

[0024] In one specific embodiment, a power generator (e.g., labeled generator i) submits its bid data to the privacy computing system of this invention through its internal power trading bidding system in order to participate in the next round of electricity spot market clearing. This data specifically includes the original bid price P. i (For example, 0.5 yuan / kWh) and declared electricity volume Q i (For example, 100 megawatt-hours). The data acquisition module of this system receives this set of data from P through a preset secure data interface. i and Q i The plaintext data is constructed. After confirming that the data format is correct, the module immediately sends this pair of P... i and Q i The value is passed directly as input to the hybrid encryption module as a whole.

[0025] For example, in the hybrid encryption module 120, the original declared prices and declared electricity quantities of each power generator are hybrid encrypted and distributed to obtain a price shard set held by the first service provider and a price shard set held by the second service provider. It should be understood that the core calculation logic of electricity market clearing involves two completely different operations: comparing and sorting declared prices, and accumulating and aggregating declared electricity quantities. These two operations place drastically different demands on cryptographic tools. On the one hand, additive homomorphic encryption naturally supports addition operations within the ciphertext field, making it an efficient tool for secure electricity aggregation; however, it cannot or is extremely difficult to efficiently support comparisons of ciphertext sizes, thus it is not suitable for price sorting. On the other hand, secure multi-party computation protocols based on secret sharing are very adept at performing secure comparison, exchange, and other logical operations, making them an ideal choice for achieving price privacy sorting; however, if used for large-scale numerical accumulation, their communication overhead may be higher than homomorphic encryption. Therefore, this application abandons the limitations of a single cryptographic tool and instead adopts a hybrid strategy, applying the most suitable cryptographic transformation to different types of data (prices and electricity quantities) before the calculation begins, based on the specific subsequent calculation requirements.

[0026] In one embodiment, such as Figure 3 As shown, the hybrid encryption module 120 includes: a price sharding unit 121, used to perform random sharding and complementary sharding calculations on the original declared price to obtain a first price shard and a second price shard; an electricity encryption unit 122, used to perform homomorphic encryption on the declared electricity to obtain encrypted electricity; and an encapsulation and distribution unit 123, used to encapsulate and securely distribute hybrid data packets of the first price shard, the second price shard, and the encrypted electricity to obtain a first encapsulated data packet and a second encapsulated data packet, wherein the first encapsulated data packet is stored on a first service provider, and the second encapsulated data packet is stored on a second service provider.

[0027] Specifically, firstly, the price sharding unit performs random sharding and complementary sharding calculations on the original declared prices. For each generator's original declared price P obtained from the data acquisition module... i This unit utilizes an additive secret-sharing technique to split a finite field defined by a pre-defined, sufficiently large prime number M into two random partitions. Specifically, the unit first... M Randomly select a value as the first price slice. Then, based on this random partition, a second price partition paired with it is calculated. This calculation strictly follows the reconstruction rules of additive secret sharing, ensuring that the sum of the two pieces equals the original price modulo M. The calculation formula is as follows: in, Refers to the original declared price value submitted by the distributor i; Refers to the original price A random shard generated after being secretly shared does not reveal anything about the shard itself. Information; Reference and The corresponding other random partition, both satisfy the following conditions. ; This is a preset, sufficiently large prime number, used as the modulus for calculation. Through this step, each original price is split into two independent random numbers that do not contain any original price information, preparing for subsequent distributed secure sorting.

[0028] Secondly, the electricity encryption unit performs homomorphic encryption on the declared electricity amount. This is related to the declared price P. i Paired reported power Q i The unit uses an encryption algorithm that supports additive homomorphism (such as the Paillier encryption algorithm) to encrypt the plaintext. The unit obtains a public key pk from the system, which contains parameters (n, g). During encryption, the unit chooses a random integer r coprime to n to ensure the probabilistic nature of the encryption, meaning that the same plaintext will produce different ciphertexts each time it is encrypted. Subsequently, the Paillier encryption algorithm is applied to encrypt the plaintext quantity Q. i Convert to ciphertext Enc(Q) i The calculation formula is as follows: in, Refers to the declared electricity volume Ciphertext generated after encryption using the Paillier scheme; It is part of the Paillier public key and is a specific generator; This is the original declared electricity consumption figure; It is a randomly selected integer used to implement probabilistic encryption; This is the key parameter of the Paillier public key, which consists of the product of two large prime numbers; This indicates a modulo operation. The output of this step is an encrypted quantity that supports ciphertext addition, providing a foundation for subsequent ciphertext aggregation calculations.

[0029] Finally, the encapsulation and secure distribution of the hybrid data packets are performed by the encapsulation and distribution unit. This unit assembles and dispatches the outputs of the first two sub-steps. Specifically, it processes the first price fragment... and encrypted power Encapsulation is performed to obtain the first encapsulated data packet; simultaneously, the second price fragment is processed. and encrypted power Encapsulation is performed to obtain a second encapsulated data packet. After encapsulation, the unit sends the first encapsulated data packet to the first service provider for storage and the second encapsulated data packet to the second service provider for storage via a secure, end-to-end encrypted communication channel (e.g., TLS / SSL). Thus, both service providers hold the complete homomorphically encrypted electricity data, but for the price, each service provider only holds an incomplete, randomized fragment, and neither party can independently reconstruct the original declared price.

[0030] In one specific embodiment, the original declared price submitted by generator i is P. i =0.5, the declared electricity consumption is Q i =100. The hybrid encryption module first checks the price P. i It processes the data and randomly generates a price shard. And calculate complementary fragments. This ensures that the sum of the two equals 0.5 modulo M. Simultaneously, this module uses the Paillier public key to control the energy Q. i Encryption is performed to obtain the encrypted electricity amount Enc(100). Subsequently, the module segments the price. The encrypted electricity value Enc(100) is encapsulated into a first data packet and sent to the first service provider. Simultaneously, the price is fragmented. The encrypted power Enc(100) is encapsulated into a second data packet and sent to the second service provider.

[0031] For example, in the secure sorting module 130, a distributed secure sorting based on secret sharing is performed on the price shard set held by the first service provider and the price shard set held by the second service provider to obtain a sorted bid index sequence. It should be understood that the unified clearing mechanism of the electricity market strictly follows the price priority principle, that is, selecting generating units in ascending order of bid prices until the total market demand is met. This requires the system to be able to accurately sort the bids of all generators. However, in the hybrid encryption and distribution steps, to protect trade secrets, the original bid prices have been split into random shards with no discernible pattern and stored separately by two untrusted service providers. Any single service provider, based solely on its shards, cannot know the true value or relative size of any original price, and therefore cannot independently complete the sorting task. In the secure sorting module, a distributed collaborative computing protocol is designed so that the two service providers can securely and correctly determine the relative size relationships of all bids without reconstructing or disclosing any original price plaintext, and thereby generate a sorted index sequence pointing to the original bids.

[0032] In one embodiment, such as Figure 4As shown, the secure sorting module 130 includes: a multi-party comparison unit 131, used to perform a distributed secure multi-party comparison on the price shard set held by the first service provider and the price shard set held by the second service provider to obtain a secret sharing set of the comparison results; and a privacy sorting unit 132, used to perform privacy sorting on the secret sharing set of the comparison results based on unintentional exchange to obtain a secret sharing of the sorted index sequence as the sorted price index sequence.

[0033] Specifically, firstly, a distributed secure multi-party comparison is performed by a multi-party comparison unit. The goal of this unit is to perform a comparison between any two offers P. i and P j The secret fragments securely compute their comparison results. (For example, =1 indicates that P i <P j , = indicates P i ≥P j Furthermore, the comparison results are also shared in a secret manner. Specifically, regarding the price shards held by the first service provider... , Price shards held by the second service provider , The two service providers first calculate the difference between the shards they hold locally to obtain the difference. = Secret Sharing Subsequently, the two service providers interact to execute a secure symbol determination protocol, which can be performed without disclosing [the symbol's identity]. In the case of a specific value, securely determine its sign and output the secret share of the comparison result. By repeating this process for all the quote pairs that need to be compared, a secret set of comparison results is eventually obtained.

[0034] Secondly, a privacy-based sorting unit sorts the original indexes of the quotes based on the comparison results. This unit typically operates based on a predefined sorting network topology (e.g., a bitonic sorting network or an odd-even sorting network) consisting of a series of fixed comparison-swapping units. For each comparison-swapping unit in the network, the service providers secretly share the two input indexes (e.g., […]). ]and[ Perform an oblivious swap operation. The core of this operation is to use the secret sharing of the comparison result... As a control switch, a switching control term is safely calculated. , and then update the secret sharing of the index through local addition and subtraction operations to achieve conditional swapping. The core calculation formula can be expressed as: In the above formula, refers to the secret sharing of the difference between two indexes; refers to the secret sharing of the swap control item, and its calculation needs to be completed through a secure multi-party multiplication protocol; and are the updated index secret sharings after an oblivious swap operation. The essence of this process is that only when the reconstructed value of the comparison result indicates that a swap is needed, the reconstructed value of is non-zero, thus effectively realizing the swapping of index positions. By iteratively performing this oblivious swap operation in the entire sorting network, the original index sequence is finally rearranged into a sorted index sequence that is consistent with the order of prices from low to high, and this final sequence itself also exists in the form of secret sharing.

[0035] In a specific embodiment, there are the quotations of three power generators, and their original indexes are {1, 2, 3}. The first service provider and the second service provider respectively hold the secret shards of the prices of these three quotations. In the secure sorting module, the service providers first determine that P3 < P1 < P2 through a secure comparison protocol. Subsequently, the privacy sorting unit performs an oblivious swap. For example, when comparing P1 and P2, since P1 < P2, the swap control item is 0, and the positions of indexes {1, 2} remain unchanged. When comparing P2 and P3, since P3 < P2, the swap control item is non-zero, triggering an oblivious swap, which makes the positions of indexes 2 and 3 swapped, and the sequence becomes {1, 3, 2}. After subsequent comparisons and swaps, finally, the two service providers jointly obtain a sequence in the form of secret sharing. The plaintext after decrypting this sequence is {3, 1, 2}, which is the sorted quotation index sequence, and it correctly reflects the sorting result that P3 is the lowest, P1 is the second, and P2 is the highest. Throughout the process, neither of the two service providers knows the true price of any quotation.

[0036] For example, in the privacy positioning module 140, based on the total market demand, the sorted bid index sequence is privacy-based to determine the clearing threshold, thus obtaining the encrypted index of marginal units. It should be understood that although the relative price order of all bids has been determined in a privacy-preserving manner through a distributed secure sorting based on secret sharing, simply knowing the order is insufficient for clearing. The essence of clearing is to accumulate the declared electricity volume one by one along this price-to-high order until the accumulated supply first meets or exceeds the total market demand. The generating unit corresponding to this threshold point that precisely meets demand is the marginal unit, and its bid will determine the unified clearing price for all winning units in this round. Since the sorted index sequence and the original electricity volume value corresponding to each index are in an encrypted or secret-shared state, a cryptographic protocol for finding the boundary is designed in the privacy positioning module, which operates entirely in a secret state. The protocol must be able to accurately locate the clearing threshold without revealing the specific sequence of the sorted index, the declared electricity of any unit, or any intermediate accumulated electricity value, and output the index of the marginal unit in encrypted form.

[0037] In one embodiment, such as Figure 5 As shown, the privacy positioning module 140 includes: a ciphertext rearrangement unit 141, used to rearrange the secret sharing of the sorted index sequence and the original ciphertext list of electricity based on an unintentionally selected ciphertext sequence to obtain a rearranged ciphertext list of electricity; an accumulation comparison unit 142, used to perform iterative ciphertext accumulation and secure comparison on the rearranged ciphertext list of electricity based on the total market demand to obtain a comparison flag bit secret sharing list; and an index extraction unit 143, used to securely extract and encrypt the marginal unit index based on the comparison flag bit secret sharing list and the secret sharing of the rearranged index to obtain an encrypted index of the marginal unit.

[0038] Specifically, firstly, the ciphertext rearrangement unit securely rearranges the original ciphertext list of electrical quantities. The input to this unit is a sorted index sequence that exists in the form of a secret sharing mechanism. and the list of ciphertext values ​​of the electricity generated during the hybrid encryption phase. This unit executes a cryptographic protocol based on unintentional selection to permutate the original ciphertext list of electricity charges according to the instructions of the sorted index sequence, thereby obtaining a rearranged ciphertext list of electricity charges whose order is completely consistent with the price from low to high. Throughout the entire reordering process, the service provider has no way of knowing the plaintext of the sorted index sequence, nor can it know which original ciphertext of the battery level was moved to which new position.

[0039] Secondly, the accumulation and comparison unit performs iterative ciphertext accumulation and secure comparison. This unit uses the rearranged list of ciphertext electrical quantities. and publicly available total market demand For input. It starts from an encrypted accumulator with an initial value of zero. It begins by iterating along the rearranged list of ciphertext for battery power. In the k-th iteration, it utilizes the addition property of homomorphic encryption to add the ciphertext for the current battery power. This is added to the accumulated ciphertext from the previous step; this is a local calculation, and the formula is: in, The homomorphic ciphertext refers to the total amount of electricity accumulated up to the k-th sorting position; It is the encrypted data of the accumulated battery power from the previous step; This represents the addition operation in homomorphic encryption. After accumulation, the service provider interactively executes a secure comparison protocol to compare the current encrypted accumulation. Compared with the total publicly disclosed demand The comparison is performed to determine whether the cumulative supply has met demand. The output of this comparison protocol is a comparison flag in the form of a secret-shared identifier. For example, when Its reconstruction value is 1 if k is true, otherwise it is 0. By iterating over all k, a secret sharing list with comparison flags is eventually generated. .

[0040] Finally, the index extraction unit securely extracts the index of the marginal group based on the comparison flag bit secret sharing list. This flag bit list is characterized by being a sequence starting with several 1s followed by several 0s, with the position of the first 0 indicating the clearing threshold. By performing differential calculations on adjacent flag bit secret shares, a marginal flag bit secret sharing vector can be obtained. The calculation formula is as follows: in, The secret is shared and defined as 1. This vector Only in the position corresponding to the marginal unit At the top, its reconstructed value is 1, while all other positions are 0. Subsequently, through a secure multi-dot product operation, the flag bit vector is... Shared with the index representing the sorted position By multiplying and summing the shares of the public index {1,..,N}, the secret share of the marginal unit's position index in the sorted list can be precisely extracted. The calculation formula is as follows: in, This indicates a secure multiplication protocol requiring server interaction. Ultimately, through a secret-to-homomorphic encryption conversion protocol, the secret-shared form of the index is transformed into a homomorphic encrypted form, yielding the encrypted index of the edge group. .

[0041] However, the calculation model for locating marginal units in the above embodiments has a key technical reason that leads to its incomplete functionality in certain scenarios: its processing logic is overly simplified and fails to effectively consider and handle the special supply and demand relationship of price matching that is common in the electricity spot market. The core issue is that this mechanism assumes differential calculation... This will inevitably produce a one-hot secret sharing vector, meaning that only one element is reconstructed to a value of 1.

[0042] However, when multiple power generators submit identical bids, if these units at the same price happen to cover the total market demand, the cumulative supply may remain constant within a price bracket, which could lead to a change in the comparison threshold. The sequence reaches a plateau, which then disrupts... The one-hot characteristic of this model presents a challenge. In a real-world market scenario, the original mechanism might arbitrarily identify only one of many units with the same price as the sole marginal unit. However, according to market rules, all units that collectively constitute that marginal price should be treated equally. Therefore, this atomized marginal point positioning model, due to its insufficient information granularity, cannot identify a complete marginal price block, but can only locate a vague marginal point. This could lead to logical ambiguities and unfair allocation in subsequent complex settlement or power curtailment allocation processes.

[0043] To address the aforementioned technical challenges, a security identification mechanism for marginal blocks based on price resonance detection is proposed. This mechanism no longer seeks a single atomized marginal point, but instead accurately identifies and defines a complete marginal block with the same price through a series of secure computational steps.

[0044] Specifically, in a preferred embodiment, the index extraction unit is further configured to: perform main edge point location and anchor price extraction on the comparison flag bit secret sharing list and the secret sharing of the rearranged index to obtain the anchor price sharing and the main edge point flag sharing vector; based on the anchor price sharing, perform price resonance detection and edge block identification on the secret sharing of the rearranged index to obtain the edge block flag sharing vector; and based on the edge block flag sharing vector and the comparison flag bit secret sharing list, determine the encrypted index of the edge unit.

[0045] Specifically, the index extraction unit first performs main edge point location and anchor price extraction. It should be understood that before block probing, a price benchmark on the verge of market clearing must first be established.

[0046] Specifically, firstly, using the differential calculation of the original mechanism, a secret sharing of a principal boundary point marker vector is securely generated. This is used to locate the first unit that allows supply to meet or exceed market demand, and is represented as: in, The secret sharing of the k-th element of the sharing vector is indicated by the main boundary point. This is a secret sharing of a comparison flag at point k indicating whether the cumulative supply is less than the total demand; 1 represents yes, and 0 represents no.

[0047] Subsequently, using this flag vector as a safety selector, a single safe multi-dot product operation is performed to select from the sorted price sharing list. Extracting the price quote for the primary marginal unit and sharing the secrets to anchoring the price. , is represented as: in, The secret sharing refers to the extracted principal marginal price; The secret sharing refers to the k-th price in the sorted price list; This indicates a secure multiplication protocol that requires server interaction between two secret shared values; the ∑ symbol represents the summation of values ​​within the secret shared domain, which is an addition operation for local server shards.

[0048] In this way, the subsequent detection process is anchored to a known price level that is precisely at the supply-demand critical point. The purpose and effect of this is to provide a precise and privacy-protected price reference standard for subsequent resonance detection.

[0049] Next, the index extraction unit performs price resonance detection and marginal block identification. After obtaining the anchor price, all other offers equal to that price must be identified in order to fully construct the marginal block.

[0050] Specifically, each server first performs local calculations and shares each price in the sorted price list. Share with the anchor price obtained in the previous step The secret to safely obtaining a price difference vector by performing differencing. , is represented as: in, The secret share refers to the difference between the k-th sorted price and the anchor price; the - symbol indicates subtraction under the secret share domain, which is a local operation on the server.

[0051] Subsequently, for each element in the difference vector, the servers interactively execute a secure zero-value test protocol, the output of which is a new marginal block flag share. If and only if its input When the reconstructed value is zero, its reconstructed value is one, which is represented as: in, The secret sharing of the k-th element of the marginal block token sharing vector; Refers to a cryptographic sub-protocol that takes a secret shared value as input and securely outputs a flag shared to indicate whether the input value is zero.

[0052] Price resonance detection and marginal block identification achieve price resonance detection in a privacy-preserving environment by securely calculating the difference between each bid and the anchor price and performing a zero-value test. In this way, all bid groups that resonate with the anchor price (i.e., are equal to it) are accurately marked, thus expanding an isolated point into a complete block, perfectly solving the challenges posed by the special market relationship of identical price bids.

[0053] Next, the index extraction unit precisely defines the final marginal set. It should be understood that simply identifying groups with the same price is insufficient; groups that, although having the same price, are ranked higher in the block and have already been fully cleared must also be removed from this set.

[0054] The execution process involves the server sharing the boundary block flag generated in the previous step through a single safe multiplication operation. Share with the inverse flag of the supply and demand comparison flag By combining these elements, we obtain the final marginal unit identifier sharing vector. , is represented as: in, This refers to the secret sharing of the k-th element of the final marginal unit's shared vector.

[0055] Finally, to determine the encrypted index of the marginal unit, the index extraction unit further performs the following steps: First, the unit performs a secure multi-party dot product operation to interactively calculate the final marginal unit identifier sharing vector with a secret sharing vector representing the index of each unit (e.g., {1, 2, 3, ..., N}), thereby securely extracting a new secret sharing vector. This new vector retains the true index value only at the position corresponding to the marginal unit, with the remaining positions being zero. Next, each non-zero index sharing in this new vector is passed to the subsequent ciphertext aggregation module for processing. The service providers collaboratively execute a secret sharing to homomorphic encryption conversion protocol, transforming the secret sharing form into its corresponding homomorphic encrypted form. Finally, all successfully converted encrypted indexes are collected, forming a set of encrypted indexes for a marginal unit, and serve as the final output of the index extraction unit.

[0056] The precise definition of the final marginal set integrates price and supply-demand balance information. From a broad price range, it accurately identifies those generating units that are truly cleared out due to market demand—that is, all units with the same price located within the intersection of the supply and demand curves. Its purpose and effect is to produce a complete and unambiguous description of the multi-hot marginal set, providing a solid computational foundation for subsequent complex settlement rules such as proportional allocation of awarded electricity.

[0057] In this preferred embodiment, the limitations of the original scheme in handling the scenario of identical price declarations in the electricity market are overcome. It can safely and completely identify the set of all generating entities that together constitute the marginal price of the market, rather than a vague single marginal point. By introducing the steps of price resonance detection and final set definition, the output of this mechanism is a multi-hot flag vector with finer information granularity that can accurately reflect the true market clearing state. This output provides a complete and unambiguous data foundation for subsequent settlement, constraint processing, and possible imbalanced electricity allocation in downstream processes, thereby greatly improving the logical rigor, fairness, and robustness of the entire safe market clearing process.

[0058] In one specific embodiment, the total market demand is 250 MWh. The sorted price list includes four generating units G1, G2, G3, and G4, with electricity volumes of 100, 120, 30, and 50 MWh respectively, and prices of 0.4, 0.45, 0.5, and 0.5 respectively. After the accumulation comparison unit executes, the plaintext of the flag list is {1, 1, 0, 0} (because 100 < 250, 100 + 120 = 220 < 250, 220 + 30 = 250 ≥ 250, 250 + 50 > 250). The index extraction unit first... The calculation locates the first position that sets the flag to 0, namely the 3rd bit G3, thus obtaining the primary edge point. Then, the price of G3, 0.5, is safely extracted as the anchor price. Next, the price resonance detection step marks all units with a price of 0.5 (i.e., G3 and G4), resulting in the plaintext of the edge block flag vector as {0, 0, 1, 1}. Finally, through... After precise definition, the final marginal unit flag vector in plaintext is {0, 0, 1, 1}, accurately identifying G3 and G4 as the marginal unit set, and finally outputting its encrypted index set. To further illustrate the advantages of this preferred embodiment in the same-price bidding scenario, consider another case: if the total market demand is 260 MWh. In this case, the flag list (plaintext) obtained after cumulative comparison will become {1, 1, 1, 0}. According to the precise definition step of the final marginal set, the comparison flag list is {0, 0, 0, 1}. This is then securely multiplied with the marginal block flag vector {0, 0, 1, 1} obtained from price resonance detection, resulting in the final marginal unit flag vector in plaintext as {0, 0, 0, 1}. This result accurately identifies G4 as the unique marginal unit. This shows that the mechanism of the preferred embodiment can clearly distinguish which units among the same-price units truly meet the marginal market demand, effectively avoiding uncertainty in the clearing process and ensuring the fairness and accuracy of the final settlement.

[0059] For example, in the ciphertext aggregation module 150, based on the original homomorphic ciphertext library, the encrypted index of the marginal units is aggregated with ciphertext for clearing price and electricity to obtain an encrypted clearing result. It should be understood that although the encrypted index of the marginal units determining the market price has been successfully located in a privacy-preserving manner, the clearing price has not actually been calculated, nor has the specific winning electricity volume and final settlement amount for each winning unit been determined. In the ciphertext aggregation module, using existing encrypted data and the newly obtained encrypted index, the final two core steps of market clearing are securely executed without decrypting any intermediate values: First, based on the encrypted index of the marginal unit, the unit's bid is securely extracted as the unified market clearing price; second, based on this clearing price, the winning electricity volume of each winning unit is securely determined, and its final settlement amount is calculated. Only after completing this series of ciphertext calculations can a complete final encrypted result package be obtained.

[0060] In one embodiment, such as Figure 6As shown, the ciphertext aggregation module 150 includes: a price determination unit 151, used to determine the homomorphic ciphertext of the marginal clearing price based on the encrypted index of the marginal unit, the secret sharing of the rearranged index, and the secret sharing list of the original price; an electricity quantity determination unit 152, used to determine the homomorphic ciphertext list of the winning electricity quantity of each original unit based on the encrypted index of the marginal unit and the secret sharing of the rearranged index; and a ciphertext settlement unit 153, used to perform homomorphic encryption-based settlement amount calculation on the homomorphic ciphertext list of the winning electricity quantity of each original unit and the homomorphic ciphertext of the marginal clearing price to obtain the encrypted clearing result.

[0061] Specifically, firstly, the price determination unit securely determines the homomorphic ciphertext of the marginal clearing price from the secret-shared list of the original prices, based on the encrypted index of the marginal units. According to the unified clearing pricing principle, the settlement price for all winning units is equal to the bid price of the marginal units. This unit first needs to use the encrypted index of the marginal unit (pointing to its position in the sorted list) to find its true index in the original bid list through a secure lookup protocol. Then, a secret-shared selector vector of the same length as the original bid list is constructed. The value is reconstructed to 1 only at the position corresponding to the actual index. This selector vector is then combined with the original price secret sharing list through a secure multi-dot product operation. The secret to accurately extracting marginal prices through multiplication. The calculation formula is as follows: Finally, through a secure conversion protocol from secret sharing to homomorphic encryption, Convert to its homomorphic encryption form This is for use in subsequent settlements.

[0062] Secondly, the power volume determination unit securely determines the homomorphic ciphertext of the winning bid volume for each original unit based on the encrypted index of the marginal units. This unit needs to calculate the winning bid volume for each unit according to its relative position in the price ranking (ranked before the marginal units, exactly a marginal unit, or ranked after the marginal units). For all units ranked before the marginal units, they win the bid in full, and the winning bid volume is their original declared volume. For all units ranked after the marginal units, they do not win the bid, and their winning bid volume is zero. For the marginal units themselves, their winning bid volume equals the total market demand. Subtract the sum of the electricity generated by all units ranked before it. All these judgments and calculations are performed within a encrypted or secret shared domain. For example, calculating a portion of the winning bid electricity for marginal units. The formula is: in This involves the secret sharing of the total energy accumulated up to the position preceding the edge unit. Ultimately, this unit will output a homomorphic ciphertext containing the target energy for each original unit j in the system. .

[0063] Finally, the ciphertext settlement unit performs homomorphic encryption-based settlement amount calculation on the homomorphic ciphertext list of the winning bid volume for each original generating unit and the homomorphic ciphertext of the marginal clearing price. The core task of this unit is to perform multiplication operations within the ciphertext field, i.e., calculating the winning bid volume × clearing price. Since standard additive homomorphic encryption schemes (such as Paillier) do not support direct multiplication between two ciphertexts, this step needs to be completed through a secure multi-party multiplication protocol that requires interaction between service providers. For each generating unit j, this unit calculates the homomorphic ciphertext of its winning bid volume... Homomorphic ciphertext with a unified marginal clearing price As input, the homomorphic ciphertext of the final settlement revenue is calculated through this protocol. The illustrative calculation formula is as follows: in, This represents a complex, interactive, homomorphic ciphertext secure multiplication protocol. By repeating this process for all units, the unit ultimately outputs a list containing the cryptographic gains of all units (i.e., the cryptographic clearing results), providing the final computational output for an end-to-end privacy-preserving closed loop.

[0064] In one specific embodiment, the total market demand is 250 MWh, with four generating units G1, G2, G3, and G4, whose electricity volumes are 100, 120, 30, and 50 MWh respectively, and whose prices are 0.4, 0.45, 0.5, and 0.5 respectively. G3 and G4 have been identified as common marginal units. The ciphertext aggregation module starts working after receiving the encrypted index set of G3 and G4. The price determination unit first securely extracts the marginal unit's bid of 0.5 as the uniform clearing price and outputs its homomorphic ciphertext Enc(0.5). Next, the electricity volume determination unit securely determines that G1 and G2 are ranked before the marginal units, therefore they win the full bid, with winning volumes of 100 and 120 MWh respectively; for G3 and G4, the two units in the marginal block, the remaining market demand needs to be allocated according to the proportion of their bid electricity volume. The remaining demand is 250 - (100 + 120) = 30 MWh. The total declared electricity volume for G3 and G4 is 30 + 50 = 80 MWh. Therefore, the winning bid volume for G3 is 30 * (30 / 80) = 11.25 MWh, and the winning bid volume for G4 is 30 * (50 / 80) = 18.75 MWh. This unit finally outputs a homomorphic ciphertext list of the winning bid volumes for each unit. Finally, the ciphertext settlement unit settles the bids for each unit: for G1, the secure multiplication protocol is executed to calculate its encrypted gain Enc(50); similarly, Enc(60) is calculated for G2; Enc(5.625) (i.e., 11.25 * 0.5) is calculated for G3; and Enc(9.375) (i.e., 18.75 * 0.5) is calculated for G4. Finally, the module encapsulates these encrypted gains into encrypted clearing results {Enc(50), Enc(60), Enc(5.625), Enc(9.375)} and outputs them.

[0065] For example, in the decryption and evidence storage module 160, the encrypted clearing result is decrypted and stored on the blockchain to obtain a publicly disclosed clearing price, a private winning bid notification, and an on-chain evidence record. It should be understood that all preceding steps operate in an opaque ciphertext world, and the resulting encrypted results are not directly interpretable or usable by market participants. Therefore, a secure and controllable mechanism is necessary to reveal these results. However, the server performing the calculations is not entirely trustworthy, and there is a risk that it may tamper with the results during the decryption stage or subsequently deny the calculation results. The decryption and evidence storage module addresses this trust issue by introducing a decentralized third-party arbitration and recording mechanism. By implementing a "commit first, reveal later, verify later" process, it can be ensured that the finally published plaintext result is strictly consistent with the result obtained through ciphertext calculation, forming a legally binding final product: a publicly disclosed clearing price, a private winning bid notification, and an on-chain evidence record with the highest credibility. The results obtained are strictly consistent. Meanwhile, the decryption and evidence storage module also needs to address the need for differentiated information disclosure: the unified market clearing price needs to be made public to ensure market transparency, while the specific winning electricity volume and settlement amount for each power generator are their trade secrets and need to be communicated privately. Only through the decryption and evidence storage module can the confidential calculation results be securely and differentiatedly transformed into public consensus and private credentials.

[0066] First, before decrypting any data, an on-chain hash commitment is performed on the encrypted result. This step aims to create an immutable fingerprint of the computation result. Specifically, both the first and second service providers locally create a list containing the encrypted settlement amounts for all crews. Perform deterministic serialization, and then apply a standard cryptographic hash function (such as SHA-256) to compute a unique hash digest. The calculation formula is as follows: in, It is the calculated commitment hash value; It is a cryptographic hash function; It is a deterministic serialization function that converts input structured data into a unique byte stream; It is a cryptographic clearing result containing multiple homomorphic ciphertexts. After calculation, a service provider (or through a multi-party signature mechanism) calls a smart contract pre-set on the blockchain to process the hash digest. The hash is submitted on the blockchain as transaction data. After receiving the hash, the smart contract associates it with the current clearing round ID and stores it, thus forming a commitment that no one can tamper with.

[0067] Secondly, distributed threshold decryption is performed. It should be understood that no single service provider can independently decrypt the clearing result; collaborative operation is necessary to recover the plaintext. For encrypted clearing results... For each encrypted value C (e.g., the encrypted revenue of a certain unit), the first and second service providers respectively fragment it using their private keys. and Partial decryption is performed. For example, in a Paillier-based threshold decryption scheme, each server calculates a partially decrypted value, illustrated by the following formula: in, and Each of the two servers calculates a portion of the decrypted value; This is the ciphertext to be decrypted; and Fragment the private keys held by the two servers respectively; This is the public key parameter. After computation, both parties securely exchange the two decrypted values ​​and merge them locally, ultimately reconstructing the original plaintext result using a specific formula. This is achieved through... This process is repeated for all encrypted items, and both service providers eventually obtain the complete plaintext clearing result. .

[0068] Finally, the execution result is verified, settled, and the event is distributed on-chain. The service provider will then distribute the plaintext clearing result obtained in the previous step. Together with the original encrypted result used for verification This information is submitted to the smart contract on the blockchain. During execution, the smart contract first uses the exact same hash algorithm as the commitment phase to hash the received information. Perform a hash calculation to obtain a verification hash. Then, the contract will... Compared with the previously stored commitment hash A comparison is performed. The contract will only continue to execute if the two are completely identical; otherwise, the transaction will fail and roll back, effectively preventing the submission of forged plaintext results. After successful verification, the smart contract automatically executes according to preset logic: updating the clearing price, which can be viewed by all market participants, to a public state variable of the contract; for each winning generator, triggering an on-chain event containing its private settlement information (such as the winning electricity volume and settlement amount), which only the generator itself can easily listen to and obtain; and simultaneously, the entire call process is permanently recorded on the blockchain as a transaction.

[0069] In one specific embodiment, the decryption and evidence storage module first calculates the hash value of the result. The system then sends this information to a blockchain smart contract for commitment. Subsequently, the first and second service providers collaboratively decrypt the plaintext result by exchanging partial decryption information; for example, the clearing price is 0.5 yuan / kWh, and power generator A wins the bid for 100 megawatt-hours. The system then combines this plaintext result with the original... The transaction is submitted to the smart contract. The contract first verifies if the received result hash matches the previously promised hash. If the verification passes, the contract updates its current clearing price state variable to 0.5 and triggers an event that only generator A can resolve, containing the content "Winning bid electricity 100 MWh, settlement amount xxx". The entire process is packaged into a single transaction and recorded on the blockchain. In summary, the privacy-preserving computation system for secure sharing of electricity market data provided in this application has been clarified. It abandons the approach of applying machine learning paradigms such as federated learning to deterministic operations research, instead constructing a hybrid computation framework that deeply integrates multiple cryptographic primitives. Specifically, this system recognizes that the core computation of market clearing can be decomposed into two parts with different characteristics: ranking comparison and numerical aggregation. For the former, namely the ranking of the declared prices of each power generator—a step highly unfriendly to semi-homomorphic encryption—this system innovatively adopts a secure multi-party computation technique based on secret sharing. This technique splits the price data into random fragments containing no original information, which are then handed over to two non-colluding service parties for efficient distributed secure comparison and ranking. For the latter, namely the accumulation and aggregation of declared electricity quantities, additive homomorphic encryption technology is used to fully utilize its inherent advantage of performing addition operations directly in the ciphertext domain. This "hybrid encryption and computation" model breaks down the complex clearing process into the most suitable cryptographic protocols for execution, and uses the secret sharing of the sorting results as an index for subsequent homomorphic ciphertext aggregation operations, thereby seamlessly linking the two technologies. Finally, blockchain technology is used to store and settle the trusted decrypted results, forming a logically closed-loop privacy computing system.

[0070] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0071] The block diagrams of devices, apparatuses, devices, and systems involved in this application are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0072] It should also be noted that in the apparatus, equipment, and methods of this application, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of this application.

[0073] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of this application. Therefore, this application is not intended to be limited to the aspects shown herein, but rather to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0074] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this application to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A privacy computing system for secure sharing of power market data, characterized in that, The method comprises the following steps: A data collection module is used to collect the original bidding price and bidding capacity of each power supplier; A hybrid encryption module is used to encrypt and distribute the original bidding price and bidding capacity of each power supplier to obtain a first service provider-held price shard set and a second service provider-held price shard set; A secure sorting module is used to perform secret sharing-based distributed secure sorting on the first service provider-held price shard set and the second service provider-held price shard set to obtain a sorted bidding index sequence; A privacy positioning module is used to perform a clearing critical point privacy positioning on the sorted bidding index sequence based on the total market demand to obtain an encrypted index of marginal units; A ciphertext aggregation module is used to perform a clearing price and capacity ciphertext aggregation on the encrypted index of marginal units based on an original homomorphic ciphertext library to obtain an encrypted clearing result; A decryption and storage module is used to decrypt and store the encrypted clearing result on a blockchain to obtain a public clearing price, a private bid notice and a chain storage record.

2. The privacy computing system for electricity market data security sharing according to claim 1, wherein, The hybrid encryption module comprises: A price shard unit is used to perform random sharding and complementary sharding calculation on the original bidding price to obtain a first price shard and a second price shard; An encrypted capacity unit is used to perform homomorphic encryption on the bidding capacity to obtain encrypted capacity; An encapsulation and distribution unit is used to encapsulate and securely distribute the first price shard and the second price shard and the encrypted capacity into hybrid data packets to obtain a first encapsulated data packet and a second encapsulated data packet, wherein the first encapsulated data packet is stored in the first service provider and the second encapsulated data packet is stored in the second service provider. 3.The privacy computing system for secure sharing of power market data according to claim 2, wherein, The encapsulation and distribution unit is further used to: Encapsulate the first price shard and the encrypted capacity to obtain the first encapsulated data packet; Encapsulate the second price shard and the encrypted capacity to obtain the second encapsulated data packet. 4.The privacy computing system for secure sharing of power market data according to claim 1, wherein, The secure sorting module comprises: A multi-party comparison unit is used to perform distributed secure multi-party comparison on the first service provider-held price shard set and the second service provider-held price shard set to obtain a secret sharing set of comparison results; A privacy sorting unit is used to perform oblivious exchange-based privacy sorting on the secret sharing set of comparison results to obtain a secret sharing of the sorted index sequence as the sorted bidding index sequence.

5. The privacy computing system for electricity market data security sharing according to claim 1, wherein, The privacy positioning module comprises: A ciphertext rearrangement unit is used to perform ciphertext sequence rearrangement based on oblivious selection on the secret sharing of the sorted index sequence and the original capacity ciphertext list to obtain a rearranged capacity ciphertext list; An accumulation comparison unit is used to perform iterative ciphertext accumulation and secure comparison on the rearranged capacity ciphertext list based on the total market demand to obtain a comparison flag bit secret sharing list; An index extraction unit is used to perform secure extraction and encryption of the marginal unit index based on the comparison flag bit secret sharing list and the secret sharing of the rearranged index to obtain the encrypted index of the marginal unit.

6. The privacy computing system for electricity market data security sharing according to claim 5, wherein, The index extraction unit is further used to: Perform main marginal point positioning and anchor price extraction on the comparison flag bit secret sharing list and the secret sharing of the rearranged index to obtain an anchor price share and a main marginal point flag share vector; Based on the anchor price sharing, the secret sharing of the rearranged index is subjected to price resonance detection and marginal block identification to obtain a marginal block flag sharing vector; Based on the marginal block flag sharing vector and a comparison flag bit secret sharing list, an encrypted index of a marginal unit is determined.

7. The privacy computing system for electricity market data security sharing according to claim 1, wherein, The ciphertext aggregation module comprises: A price determination unit configured to determine a homomorphic ciphertext of a marginal clearing price based on the encrypted index of the marginal unit, the secret sharing of the rearranged index and a secret sharing list of an original price; An electric quantity determination unit configured to determine a homomorphic ciphertext list of a winning electric quantity in each original unit based on the encrypted index of the marginal unit and the secret sharing of the rearranged index; A ciphertext settlement unit configured to perform homomorphic encryption-based settlement amount calculation on the homomorphic ciphertext list of the winning electric quantity in each original unit and the homomorphic ciphertext of the marginal clearing price to obtain the encrypted clearing result.

Images