Virtual power plant operation management and control method based on block chain

Abstract

The invention discloses a virtual power plant operation management and control method based on a block chain, original power data are compressed into Hash fingerprints to be chained through a collaborative architecture of an edge computing layer and an AI pre-decision module, the block chain network load is effectively reduced, and a dynamic elastic PBFT consensus mechanism adjusts the scale of verification nodes in real time based on transaction types, so that the efficiency is improved. And compared with a traditional fixed node consensus mechanism, the network communication overhead is reduced. And a hash value asynchronous pre-verification technology is adopted, so that the interaction rate of the multi-dimensional power data is improved, and the efficiency is improved compared with that of a conventional data uplink mode. Through the decoupling design of the intelligent contract execution flow and the consensus verification flow and in combination with the pre-decision data cache queue, the resource scheduling instruction issuing time delay is reduced. The block chain transaction throughput is improved and is effectively improved compared with a standard alliance chain architecture, and the technical problems that the consensus mechanism processing transaction speed of an existing block chain is limited and the high-frequency real-time data interaction requirement of a virtual power plant is difficult to meet are successfully solved.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plants, and in particular to a blockchain-based virtual power plant operation management and control method. Background Art

[0002] A virtual power plant (VPP) is a virtualized power management system that aggregates distributed energy resources through information communication and intelligent control technologies. It coordinates resources to participate in grid peak and frequency regulation, demand response, and market transactions, enhancing power supply security, promoting the integration of new energy, and improving grid flexibility. For example, by integrating into battery swap stations, energy storage can be used to regulate grid load; businesses can participate in flexible response to ensure production and alleviate power demand. With policy support for its development, the goal is to reach 20 million kilowatts of national regulation capacity by 2027, contributing to the "dual carbon" goals. Private enterprises are becoming a key player in this effort.

[0003] In the management and use of virtual power plants, blockchain, through its decentralized, tamper-proof nature and the automated execution of smart contracts, provides a transparent and secure distributed energy trading and coordinated scheduling mechanism, effectively addressing the trust issues and data credibility challenges inherent in multi-party collaboration. However, the blockchain consensus mechanism used in virtual power plants has limited transaction processing speeds, making it difficult to meet the high-frequency, real-time data exchange requirements of virtual power plants. Summary of the Invention

[0004] The purpose of the present invention is to provide a virtual power plant operation control method based on blockchain to solve the technical problem that the consensus mechanism of the existing blockchain has limited transaction processing speed and is difficult to meet the high-frequency real-time data interaction needs of the virtual power plant.

[0005] The technical solution of the present invention is achieved as follows:

[0006] A blockchain-based virtual power plant operation control method, comprising:

[0007] The distributed power data of the virtual power plant is collected and pre-processed in real time through edge devices, and transmitted to the AI ​​model for multi-dimensional prediction to generate pre-decision data;

[0008] Generate a unique hash value from the pre-decision data using a hash algorithm, and submit it to the blockchain transaction pool in the form of an encrypted transaction;

[0009] Verifying the unique hash value using a PBFT consensus mechanism, wherein the number of nodes of the PBFT consensus mechanism is dynamically adjusted according to the encrypted data transaction type;

[0010] Triggering a preset smart contract based on the verified unique hash value and automatically executing resource scheduling, transaction settlement or market response operations of the virtual power plant;

[0011] The execution results of the preset smart contract and the virtual power plant status are stored in the blockchain to generate an unalterable audit chain.

[0012] A further technical solution is to collect and pre-process the distributed power data of the virtual power plant in real time through edge devices, transmit it to the AI ​​model for multi-dimensional prediction, and generate pre-decision data, specifically including:

[0013] The photovoltaic panels, energy storage system, load equipment and meteorological sensors of the virtual power plant are all equipped with the edge computing device;

[0014] The edge computing device obtains first multi-source data on power generation, energy storage status, load power, and environmental parameters collected by the photovoltaic panels, energy storage system, load equipment, and meteorological sensors;

[0015] Performing localized data cleaning, standardization, and time series alignment on the first multi-source data to generate second multi-source data;

[0016] The AI ​​model is used to receive the second multi-source data, and combines the electricity price signal with the dispatch instruction to generate pre-decision data.

[0017] A further technical solution involves a long short-term memory network, wherein the AI ​​model is used to receive the second multi-source data and generate pre-decision data by combining the electricity price signal and the dispatch instruction, specifically including:

[0018] receiving the second multi-source data based on the long short-term memory network;

[0019] extracting long-term dependency features and short-term mutation features of the second multi-source data, the electricity price signal, and the scheduling instruction through a dynamic gating mechanism;

[0020] The long-term dependency characteristics and short-term mutation characteristics are integrated with the grid topology constraints to output the collaborative prediction results of power generation, load and electricity price in the future period.

[0021] A further technical solution is to further include:

[0022] Performing time series modeling on the historical data of the second multi-source data, the electricity price signal, and the scheduling instruction data based on the long short-term memory network to predict the prediction results of the second multi-source data, the electricity price signal, and the scheduling instruction data in a future time period;

[0023] Inputting the prediction result and the second multi-source data, the electricity price signal, and the real-time data of the scheduling instruction data into a deep learning model, and outputting an objective function of the optimal strategy through the real-time deep learning model;

[0024] Introducing the real-time objective function into the constraints of the edge computing device and executing the objective function of the real-time optimal strategy within the real-time constraints;

[0025] The objective function of the optimal strategy is sent to the power system, and real-time feedback data from the power system is received to continuously iterate the deep learning model.

[0026] A further technical solution is to generate a unique hash value from the pre-decision data through a hash algorithm and submit it to the blockchain transaction pool in the form of an encrypted transaction, specifically including:

[0027] Converting the pre-decision data into a unique and irreversible fixed-length hash value;

[0028] The hash value is signed based on the user's private key and is put into the blockchain transaction pool;

[0029] After verifying the legitimacy of the signature and confirming that there is no tampering, the transaction is packaged into the candidate block;

[0030] The transaction in the candidate is permanently written into the blockchain to generate an encrypted data record.

[0031] A further technical solution is that the unique hash value is verified using a PBFT consensus mechanism, wherein the number of nodes of the PBFT consensus mechanism is dynamically adjusted according to the type of encrypted data transaction, specifically including:

[0032] Using the PBFT consensus mechanism to complete consensus verification on the unique hash value of the pre-decision data in the blockchain transaction pool;

[0033] dividing the transaction type into a first transaction type and a second transaction type;

[0034] The number of nodes of the PBFT consensus mechanism is matched according to the first transaction type and the second transaction type.

[0035] A further technical solution is that the number of nodes of the PBFT consensus mechanism is matched according to the first transaction type and the second transaction type, specifically including:

[0036] Using a clustering algorithm, the number of nodes is divided into a first layer and a second layer according to an emergency state, and the nodes correspond to the first transaction type and the second transaction type;

[0037] The first layer is responsible for the first transaction type, and the second layer is responsible for the second transaction type.

[0038] A further technical solution is that the preset smart contract is triggered based on the verified unique hash value, and the resource scheduling, transaction settlement or market response operation of the virtual power plant is automatically executed, specifically including:

[0039] Verifying the unique hash value through the PBFT consensus mechanism;

[0040] Matching the verified unique hash value according to the preset conditions of the smart contract and triggering execution by obtaining external real-time data through an oracle;

[0041] Combining IoT devices with the unique hash value record for verification can automatically perform resource scheduling, transaction settlement, or market response.

[0042] A further technical solution is that the combination of the IoT device and the verification of the unique hash value record automatically performs resource scheduling, transaction settlement or market response, specifically including:

[0043] Uploading the original data and hash value to the blockchain network through the IoT device;

[0044] The smart contract calls the hash value stored on the blockchain and compares it with the hash value calculated in real time by the IoT device to generate a verification result of the unique hash value;

[0045] The resource scheduling, transaction settlement, and market response operations are automatically executed according to the verification results.

[0046] A further technical solution is to store the results of executing the preset smart contract and the virtual power plant status in the blockchain to generate an unalterable audit chain, specifically including:

[0047] The preset smart contract will be executed and combined with the real-time status data of the virtual power plant to form a proof unit;

[0048] Multi-signature verification of the evidence storage unit is performed by the evidence storage institution, the virtual power plant operator and the third-party arbitration node;

[0049] The evidence storage unit is written into the blockchain in timestamp order through the PBFT consensus mechanism, forming an audit chain including hash value, signature, contract address and execution time.

[0050] The beneficial effects of the present invention are:

[0051] The present invention compresses the original power data into a hash fingerprint and uploads it to the chain through the collaborative architecture of the edge computing layer and the AI ​​pre-decision module, effectively reducing the load of the blockchain network. The dynamic elastic PBFT consensus mechanism adjusts the scale of verification nodes in real time based on the transaction type, reducing network communication overhead compared to the traditional fixed node consensus mechanism. The use of hash value asynchronous pre-verification technology increases the rate of multi-dimensional power data interaction, which is more efficient than conventional data on-chain methods. Through the decoupling design of the smart contract execution flow and the consensus verification flow, combined with the pre-decision data cache queue, the delay in issuing resource scheduling instructions is reduced. The blockchain transaction throughput is improved, which is an effective improvement over the standard alliance chain architecture, successfully solving the technical problem that the consensus mechanism of the existing blockchain has limited transaction processing speed and is difficult to meet the high-frequency real-time data interaction needs of virtual power plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a flowchart of a blockchain-based virtual power plant operation control method of the present invention;

[0053] Figure 2 This is a flow chart of step S1 of a blockchain-based virtual power plant operation control method of the present invention;

[0054] Figure 3 This is a flow chart of step S2 of a blockchain-based virtual power plant operation control method of the present invention;

[0055] Figure 4 This is a flow chart of step S3 of a blockchain-based virtual power plant operation control method of the present invention;

[0056] Figure 5 This is a flow chart of step S4 of a blockchain-based virtual power plant operation control method of the present invention;

[0057] Figure 6 This is a flow chart of step S5 of a blockchain-based virtual power plant operation control method of the present invention. DETAILED DESCRIPTION

[0058] In order to better understand the technical content of the present invention, specific embodiments are provided below, and the present invention is further described in conjunction with the accompanying drawings.

[0059] See also Figure 1 The present invention provides a blockchain-based virtual power plant operation management and control method, comprising:

[0060] Step S1: Use edge devices to collect and pre-process distributed power data of the virtual power plant in real time, transmit it to the AI ​​model for multi-dimensional prediction, and generate pre-decision data;

[0061] Step S2: Generate a unique hash value from the pre-decision data using a hash algorithm and submit it to the blockchain transaction pool in the form of an encrypted transaction;

[0062] Step S3: Use the PBFT consensus mechanism to verify the unique hash value, where the number of nodes in the PBFT consensus mechanism is dynamically adjusted according to the encrypted data transaction type;

[0063] Step S4: triggering a preset smart contract based on the verified unique hash value, and automatically executing resource scheduling, transaction settlement, or market response operations of the virtual power plant;

[0064] Step S5: The execution of the preset smart contract and the virtual power plant status results are stored in the blockchain to generate an unalterable audit chain.

[0065] In an embodiment of the present invention, an edge device can be a physical device deployed close to a data source or end user, and its core function is to process, transmit, and store data near the data source or user end to reduce latency, save bandwidth, and improve real-time performance; distributed power data can be operating data of photovoltaics, energy storage, and charging piles; pre-decision data can be power scheduling, resource aggregation, or fault warning data.

[0066] Distributed power data is collected in real time at the local end close to the data source through edge devices, and the data dimension is reduced through preprocessing. The distributed power data is then transmitted to the AI ​​model for load forecasting, new energy output forecasting, and demand response optimization analysis, ultimately generating pre-decision data.

[0067] The SHA-256 hash algorithm is used to generate a unique, irreversible, fixed-length hash of the pre-decision data, ensuring that any data modification will result in a change in the hash value. The hash value is then packaged with a timestamp and the identities of the participating parties into an encrypted transaction (usually signed using asymmetric cryptography to ensure the authenticity of the source) and submitted to the blockchain network's transaction pool.

[0068] The PBFT consensus mechanism is used to verify hash values. Its core is to reach consensus through multi-node voting, ensuring data integrity and the PBFT consensus mechanism's attack resistance. The number of nodes is dynamically adjusted based on transaction type requirements. For example, nodes are divided into different levels or groups through a layered architecture or clustering algorithm, and the number of nodes participating in the consensus is dynamically selected based on the transaction type.

[0069] Using a unique, verified hash value on the blockchain as a trigger, a pre-set smart contract (such as a resource scheduling contract, transaction settlement contract, or market response contract) is activated. Once the hash value verifies data integrity through the PBFT consensus mechanism, the smart contract automatically executes the pre-set logic. For example, it can dynamically adjust the scheduling strategy for energy storage, charging stations, and industrial loads within the virtual power plant based on real-time electricity prices or load demand; automatically distribute excess load reduction revenue to user accounts in a tiered manner according to market trading rules; or respond to emergency grid dispatch orders to rapidly aggregate distributed resources to participate in the ancillary services market. The entire process, based on the decentralized execution and conditional automation characteristics of blockchain, requires no human intervention.

[0070] By hashing the smart contract's resource scheduling instructions, transaction settlement results, and the virtual power plant's equipment operating parameters and real-time energy output data, a unique data fingerprint is generated. This hash value, along with a timestamp and operator identity information, is recorded on the blockchain. The blockchain's distributed ledger ensures that all nodes store this data synchronously, forming a chronological audit chain. Each status update is automatically triggered by the smart contract, and the data authenticity is verified using the PBFT consensus mechanism, ensuring that the record is tamper-proof and traceable.

[0071] The present invention compresses the original power data into a hash fingerprint and uploads it to the chain through the collaborative architecture of the edge computing layer and the AI ​​pre-decision module, effectively reducing the load of the blockchain network. The dynamic elastic PBFT consensus mechanism adjusts the scale of verification nodes in real time based on the transaction type, reducing network communication overhead compared to the traditional fixed node consensus mechanism. The use of hash value asynchronous pre-verification technology increases the rate of multi-dimensional power data interaction, which is more efficient than conventional data on-chain methods. Through the decoupling design of the smart contract execution flow and the consensus verification flow, combined with the pre-decision data cache queue, the delay in issuing resource scheduling instructions is reduced. The blockchain transaction throughput is improved, which is an effective improvement over the standard alliance chain architecture, successfully solving the technical problem that the consensus mechanism of the existing blockchain has limited transaction processing speed and is difficult to meet the high-frequency real-time data interaction needs of virtual power plants.

[0072] Preferably, see Figure 2 , the specific steps of step S1 are:

[0073] Step S11: The photovoltaic panels, energy storage system, load equipment and meteorological sensors of the virtual power plant are all equipped with edge computing equipment;

[0074] Step S12: The edge computing device obtains first multi-source data on power generation, energy storage status, load power, and environmental parameters collected by photovoltaic panels, energy storage systems, load equipment, and meteorological sensors;

[0075] Step S13: performing localized data cleaning, standardization processing, and time series alignment on the first multi-source data to generate second multi-source data;

[0076] Step S14: Utilize the AI ​​model to receive the second multi-source data, and combine the electricity price signal with the dispatch instruction to generate pre-decision data.

[0077] In an embodiment of the present invention, real-time data processing and intelligent decision-making of a virtual power plant are achieved through edge computing devices. Edge computing devices (such as sensors, controllers) are directly deployed next to photovoltaic panels, energy storage systems, load equipment and meteorological sensors to collect multi-source data such as power generation, energy storage status, load power, environmental parameters, etc. in real time, and form first multi-source data. Subsequently, the edge device performs local processing on the data: second multi-source data is generated by denoising the data, filling missing values, unifying the format and dimension, and aligning the time series. Finally, the AI ​​model at the edge combines real-time electricity price signals with grid dispatch instructions to analyze the second multi-source data, predict power generation and load trends, and generate optimized energy storage charging and discharging strategies and load dispatch plan pre-decision data. The entire process relies on the low-latency processing of edge computing and the intelligent modeling of AI to achieve a closed loop of data collection-processing-decision-making.

[0078] Preferably, step S14 involves a long short-term memory network, and the specific steps include:

[0079] Step S141: receiving second multi-source data based on the long short-term memory network;

[0080] Step S142: extracting long-term dependency features and short-term mutation features of the second multi-source data, electricity price signals, and dispatch instructions through a dynamic gating mechanism;

[0081] Step S143: Fuse the long-term dependency characteristics and the short-term mutation characteristics with the grid topology constraints to output the collaborative prediction results of power generation, load, and electricity price in the future period.

[0082] In an embodiment of the present invention, long-term dependency characteristics may be seasonal patterns of photovoltaic output and historical trends of load; short-term mutation characteristics may be the impact of weather changes on power generation and load mutations caused by electricity price fluctuations; and grid topology constraints may be line capacity, node voltage limits, and equipment physical constraints.

[0083] The LSTM network receives the second multi-source data after edge computing processing and automatically filters key time series features using its forget gate, input gate, and output gate. Subsequently, a dynamic gating mechanism further separates and enhances two types of features: long-term dependency and short-term mutation. Dynamic adjustment of gating weights captures temporal dependency and mutation signals in the data. The LSTM network combines these long-term dependency and short-term mutation features with grid topology constraints to generate collaborative forecasts for future time periods that meet the physical rules of the grid, including multi-objective forecasts of power generation, load demand, and electricity prices.

[0084] Preferably, step S14 further includes:

[0085] Step S1411: performing time series modeling on the historical data of the second multi-source data, the electricity price signal, and the dispatch instruction data based on the long short-term memory network to predict the prediction results of the second multi-source data, the electricity price signal, and the dispatch instruction data in a future time period;

[0086] Step S1412: Input the prediction results and the real-time data of the second multi-source data, the electricity price signal, and the dispatch instruction data into the deep learning model, and output the objective function of the optimal strategy through the real-time deep learning model;

[0087] Step S1413: Introduce the real-time objective function into the constraints of the edge computing device, and execute the objective function of the real-time optimal strategy within the real-time constraints;

[0088] Step S1414: Send the objective function of the optimal strategy to the power system, and receive real-time feedback data from the power system to continuously iterate the deep learning model.

[0089] In an embodiment of the present invention, a long short-term memory network is used to perform time series modeling on the historical data of the second multi-source data, electricity prices, and dispatch instructions to capture long-term dependencies and predict multi-target data of photovoltaic output and electricity price fluctuations in future time periods. The prediction results and the second multi-source data collected in real time are input into the deep Q network together, and the objective function of the optimal strategy is generated by combining the optimization goals of minimum cost and maximum benefit. The objective function is verified and executed through real-time constraints such as the energy storage SOC limit of the edge computing device and the grid topology constraint to ensure the feasibility of the strategy. The strategy is sent to the power system for execution, and the actual operation data of the device response and deviation value fed back by the system in real time are transmitted back to the deep Q network. The model parameters are continuously updated through online learning to form a closed-loop optimization of prediction-decision-execution-feedback.

[0090] Preferably, see Figure 3 , step S2 specifically includes the following steps:

[0091] Step S21: converting the pre-decision data into a unique and irreversible fixed-length hash value;

[0092] Step S22: The hash value is signed based on the user's private key and is put into the blockchain transaction pool;

[0093] Step S23: After verifying the validity of the signature and confirming that there is no tampering, the transaction is packaged into the candidate block;

[0094] Step S24: Permanently write the candidate transactions into the blockchain to generate encrypted data records.

[0095] In an embodiment of the present invention, the power generation scheduling plan and load adjustment instruction pre-decision data are generated using the SHA-256 hash algorithm to generate a unique and irreversible fixed-length hash value to ensure the uniqueness of the data fingerprint. The user uses the private key to digitally sign the hash value, and the signed transaction is broadcast to the blockchain transaction pool for verification. The nodes in the network verify the legitimacy of the signature through proof of stake or PBFT consensus mechanism. For example, the signature is decrypted using the user's public key and compared with the original hash value. If they are consistent, it is confirmed that the data has not been tampered with. The verified transactions are packaged into candidate blocks, and finally the candidate blocks are written to the blockchain through proof of work, forming a permanent record linked in chronological order. Its hash value is bound to the previous block to ensure the integrity and irreversibility of the data chain.

[0096] Preferably, see Figure 4 , step S3 specifically includes:

[0097] Step S31: Use the PBFT consensus mechanism to complete consensus verification on the unique hash value of the pre-decision data in the blockchain transaction pool;

[0098] Step S32: Classify the transaction type into a first transaction type and a second transaction type;

[0099] Step S33: Match the number of nodes of the PBFT consensus mechanism according to the first transaction type and the second transaction type.

[0100] In an embodiment of the present invention, the first transaction type may be a regular dispatch instruction or low-risk data, and the second transaction type may be an emergency power grid adjustment or a high-value transaction.

[0101] After the hash value of the pre-decision data is submitted to the transaction pool, it is verified through the three-stage consensus process of pre-preparation, preparation, and submission of the PBFT consensus mechanism: the master node broadcasts the hash value to other replica nodes, and the nodes verify the signature 2f+1 times to ensure that the data has not been tampered with and reach a consensus. Among them, transactions are divided into two categories according to business characteristics:

[0102] The first transaction type is sensitive to latency but has low security requirements. It uses a small-scale node group (such as N1=3f1+1, where f1 is small) to quickly reach consensus.

[0103] The second transaction type requires higher security and uses a large-scale node group (such as N2=3f2+1, f2 is larger) to resist more malicious nodes.

[0104] Finally, the number of nodes corresponding to the transaction type is dynamically matched: different node groups are allocated through load balancing strategies to ensure low latency for the first type of transactions and strong resistance to PBFT consensus mechanism attacks for the second type of transactions.

[0105] Preferably, step S33 specifically includes:

[0106] Step S331: Use a clustering algorithm to divide the number of nodes into a first layer and a second layer according to the emergency state, and correspond to the first transaction type and the second transaction type;

[0107] Step S332: The first layer is responsible for the first transaction type, and the second layer is responsible for the second transaction type.

[0108] In an embodiment of the present invention, the first layer may be emergency / high priority and the second layer may be regular / high security.

[0109] Based on the key characteristics of the node, such as processing speed, load, geographical location, and fault tolerance, a feature vector is constructed, and the nodes are divided into two layers using a clustering algorithm:

[0110] The first layer: A clustering algorithm is used to select a group of nodes with fast response speed and low latency, which is responsible for the first transaction type of emergency dispatch instructions.

[0111] Second layer: The remaining nodes form another cluster with high redundancy and anti-attack capabilities, responsible for the second transaction type of ordinary transactions or non-urgent instructions.

[0112] It is then automatically routed to the corresponding layer based on the transaction type: the first-tier nodes prioritize urgent transactions by simplifying the verification steps, while the second-tier nodes use multi-node redundancy to ensure the security of non-urgent transactions.

[0113] Preferably, see Figure 5 , step S4 specifically includes the following steps:

[0114] Step S41: Verify the unique hash value through the PBFT consensus mechanism;

[0115] Step S42: Match the verified unique hash value according to the preset conditions of the smart contract, and obtain external real-time data through the oracle to trigger execution;

[0116] Step S43: Combine the IoT device with the unique hash value record to automatically perform resource scheduling, transaction settlement, or market response.

[0117] In the embodiment of the present invention, the preset conditions may be a price threshold and a resource requirement.

[0118] The unique hash value in the transaction pool is verified using the PBFT consensus mechanism: the master node broadcasts the hash value to replica nodes, which then complete signature verification through a three-phase consensus process, ensuring that the virtual power plant's energy dispatch instructions have not been tampered with and are consistent. The smart contract matches the verified hash value against pre-set conditions and obtains real-time electricity prices and device status from an oracle. For example, if the oracle receives real-time data from IoT sensors indicating excessive grid load, the smart contract will trigger automatic energy allocation adjustments. IoT devices interact with the blockchain through IoT cards. On the one hand, the real-time electricity consumption and device status uploaded by the device are verified by hashing on the blockchain and used as the basis for execution. On the other hand, the results of the smart contract's resource dispatch instructions are automatically distributed to the device via the IoT gateway, completing transaction settlement or market response (such as automatic settlement of distributed energy transactions).

[0119] Preferably, step S43 specifically includes:

[0120] Step S431: Upload the original data and hash value to the blockchain network through the IoT device;

[0121] Step S432: The hash value stored on the blockchain is compared with the hash value calculated in real time by the IoT device through the smart contract to generate a unique hash value verification result;

[0122] Step S433: Automatically execute resource scheduling, transaction settlement, and market response operations based on the verification results.

[0123] In the embodiment of the present invention, the IoT device may be a sensor or a smart meter; the raw data may be temperature or electricity consumption.

[0124] The IoT device uploads the collected raw data and the calculated hash value to the blockchain network via an encrypted communication protocol. The hash value serves as the data's unique fingerprint and is stored on the chain as an immutable and trusted record. Next, the smart contract proactively calls the historical hash value stored on the blockchain and compares it with the hash value verified by the SHA-256 algorithm calculated in real time by the IoT device. If the hash values ​​match, the data has not been tampered with and a verification result is generated. If not, a notification to the operations and maintenance system or a freezing transaction abnormality alarm mechanism is triggered. Finally, the smart contract automatically executes the preset actions based on the verification results:

[0125] Successful verification triggers adjustments to grid load, automated supplier payments, or adjustments to energy procurement strategies based on real-time electricity prices. For example, in a supply chain, if temperature sensor data passes verification, the smart contract automatically releases payment. Failed verification halts the operation and logs the anomaly, such as suspending device permissions or initiating an audit.

[0126] Preferably, see Figure 6 , step S5 specifically includes the following steps:

[0127] Step S51: Execute the preset smart contract and combine it with the real-time status data of the virtual power plant to form a proof unit;

[0128] Step S52: Multi-signature verification is performed on the evidence storage unit by the evidence storage institution, the virtual power plant operator, and the third-party arbitration node;

[0129] Step S53: Write the evidence unit into the blockchain in timestamp order through the PBFT consensus mechanism, forming an audit chain including hash value, signature, contract address and execution time.

[0130] In an embodiment of the present invention, the preset smart contract may be an electricity trading contract; the evidence depository may be a notary office; the audit chain includes data fingerprints, trusted credentials, execution basis, and operation sequence, supporting full life cycle traceability.

[0131] After the pre-set smart contract is automatically triggered, the system integrates the transaction amount, resource allocation contract execution results, and the virtual power plant's power generation, load demand, and equipment status data to generate a proof unit containing a hash value, execution timestamp, and contract address. After the data is hashed using the SHA-256 algorithm, it is stored separately from the original data to ensure data integrity. Multi-party signature verification is initiated: the proof unit is jointly verified by the proof organization, the virtual power plant operator, and a third-party arbitration node. For example, in the event of a power transaction dispute, the arbitration node must verify the hash value with the original data, the operator must confirm the compliance of the transaction, and the proof organization must verify the legitimacy of the timestamp. After all three parties sign, the proof unit enters the consensus process. The proof unit is written to the blockchain in timestamp order using the PBFT consensus mechanism. Nodes in the PBFT consensus mechanism reach consensus through three phases: pre-prepare, prepare, and commit. The master node broadcasts the proof unit, replica nodes verify the signature and data legitimacy, and after collecting sufficient signatures, it is submitted to the chain, ultimately forming an immutable audit chain.

[0132] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A blockchain-based virtual power plant operation management and control method, characterized in that: include: The distributed power data of the virtual power plant is collected and pre-processed in real time through edge devices, and transmitted to the AI ​​model for multi-dimensional prediction to generate pre-decision data; Generate a unique hash value from the pre-decision data using a hash algorithm, and submit it to the blockchain transaction pool in the form of an encrypted transaction; Verifying the unique hash value using a PBFT consensus mechanism, wherein the number of nodes of the PBFT consensus mechanism is dynamically adjusted according to the encrypted data transaction type; Triggering a preset smart contract based on the verified unique hash value and automatically executing resource scheduling, transaction settlement or market response operations of the virtual power plant; The execution results of the preset smart contract and the virtual power plant status are stored in the blockchain to generate an unalterable audit chain.

2. A blockchain-based virtual power plant operation management and control method according to claim 1, characterized in that: The distributed power data of the virtual power plant is collected and pre-processed in real time by edge devices, transmitted to the AI ​​model for multi-dimensional prediction, and generated pre-decision data, specifically including: The photovoltaic panels, energy storage system, load equipment and meteorological sensors of the virtual power plant are all equipped with the edge computing device; The edge computing device obtains first multi-source data on power generation, energy storage status, load power, and environmental parameters collected by the photovoltaic panels, energy storage system, load equipment, and meteorological sensors; Performing localized data cleaning, standardization, and time series alignment on the first multi-source data to generate second multi-source data; The AI ​​model is used to receive the second multi-source data, and combines the electricity price signal with the dispatch instruction to generate pre-decision data.

3. A blockchain-based virtual power plant operation management and control method according to claim 2, characterized in that: Involving a long short-term memory network, the AI ​​model is used to receive the second multi-source data and generate pre-decision data by combining the electricity price signal and the dispatch instruction, specifically including: receiving the second multi-source data based on the long short-term memory network; extracting long-term dependency features and short-term mutation features of the second multi-source data, the electricity price signal, and the scheduling instruction through a dynamic gating mechanism; The long-term dependency characteristics and short-term mutation characteristics are integrated with the grid topology constraints to output the collaborative prediction results of power generation, load and electricity price in the future period.

4. A blockchain-based virtual power plant operation control method according to claim 3, characterized in that: Also includes: Performing time series modeling on the historical data of the second multi-source data, the electricity price signal, and the scheduling instruction data based on the long short-term memory network to predict the prediction results of the second multi-source data, the electricity price signal, and the scheduling instruction data in a future time period; Inputting the prediction result and the second multi-source data, the electricity price signal, and the real-time data of the scheduling instruction data into a deep learning model, and outputting an objective function of the optimal strategy through the real-time deep learning model; Introducing the real-time objective function into the constraints of the edge computing device and executing the objective function of the real-time optimal strategy within the real-time constraints; The objective function of the optimal strategy is sent to the power system, and real-time feedback data from the power system is received to continuously iterate the deep learning model.

5. The method for controlling and managing virtual power plant operations based on blockchain according to claim 1, characterized in that: The method of generating a unique hash value from the pre-decision data through a hash algorithm and submitting the pre-decision data to the blockchain transaction pool in the form of an encrypted transaction specifically includes: Converting the pre-decision data into a unique and irreversible fixed-length hash value; The hash value is signed based on the user's private key and is put into the blockchain transaction pool; After verifying the legitimacy of the signature and confirming that there is no tampering, the transaction is packaged into the candidate block; The transaction in the candidate is permanently written into the blockchain to generate an encrypted data record.

6. The method for controlling the operation of a virtual power plant based on blockchain according to claim 1, characterized in that: The unique hash value is verified using a PBFT consensus mechanism, wherein the number of nodes of the PBFT consensus mechanism is dynamically adjusted according to the encrypted data transaction type, specifically including: Using the PBFT consensus mechanism to complete consensus verification on the unique hash value of the pre-decision data in the blockchain transaction pool; dividing the transaction type into a first transaction type and a second transaction type; The number of nodes of the PBFT consensus mechanism is matched according to the first transaction type and the second transaction type.

7. A blockchain-based virtual power plant operation management and control method according to claim 6, characterized in that: The number of nodes matching the PBFT consensus mechanism according to the first transaction type and the second transaction type specifically includes: Using a clustering algorithm, the number of nodes is divided into a first layer and a second layer according to an emergency state, and the nodes correspond to the first transaction type and the second transaction type; The first layer is responsible for the first transaction type, and the second layer is responsible for the second transaction type.

8. The method for controlling the operation of a virtual power plant based on blockchain according to claim 1, characterized in that: The triggering of a preset smart contract based on the verified unique hash value and the automatic execution of resource scheduling, transaction settlement or market response operations of the virtual power plant specifically include: Verifying the unique hash value through the PBFT consensus mechanism; Matching the verified unique hash value according to the preset conditions of the smart contract and triggering execution by obtaining external real-time data through an oracle; Combining IoT devices with the unique hash value record for verification can automatically perform resource scheduling, transaction settlement, or market response.

9. A blockchain-based virtual power plant operation management and control method according to claim 8, characterized in that: The combination of the IoT device and the verification of the unique hash value record to automatically perform resource scheduling, transaction settlement or market response specifically includes: Uploading the original data and hash value to the blockchain network through the IoT device; The smart contract calls the hash value stored on the blockchain and compares it with the hash value calculated in real time by the IoT device to generate a verification result of the unique hash value; The resource scheduling, transaction settlement, and market response operations are automatically executed according to the verification results.

10. The method for controlling and managing virtual power plant operations based on blockchain according to claim 1, characterized in that: The execution of the preset smart contract and the virtual power plant status result are stored in the blockchain to generate an unalterable audit chain, specifically including: The preset smart contract will be executed and combined with the real-time status data of the virtual power plant to form a proof unit; Multi-signature verification of the evidence storage unit is performed by the evidence storage institution, the virtual power plant operator and the third-party arbitration node; The evidence storage unit is written into the blockchain in timestamp order through the PBFT consensus mechanism, forming an audit chain including hash value, signature, contract address and execution time.

Images