A blockchain-based virtual power plant distributed energy collaborative scheduling method and system

Abstract

This application relates to the field of power system dispatching technology, and in particular to a blockchain-based method and system for collaborative dispatching of distributed energy resources in a virtual power plant. The method includes: acquiring node status files of each distributed energy resource node; determining the pre-dispatch output power of candidate nodes based on the target total power and the node status files; determining the superimposed expected peak current of the distribution feeder branch based on the pre-dispatch output power; determining the expected total load current of the distribution feeder branch based on the current load current and the superimposed expected peak current; and outputting a collaborative dispatching command for distributed energy resources based on the expected total load current of the distribution feeder branch. Through this technical solution, more rational collaborative dispatching of distributed energy resources managed by the virtual power plant can be achieved.

Description

Technical Field

[0001] This application relates to the field of power system dispatching technology, and in particular to a blockchain-based method and system for collaborative dispatching of distributed energy resources in a virtual power plant. Background Technology

[0002] A virtual power plant is a technology that uses information and communication technology and an energy management platform to aggregate and manage various distributed energy resources. The core components of a virtual power plant mainly include a distributed energy resource layer, a sensing and communication layer, a collaborative control layer, and a grid interaction layer. The distributed energy resource layer is the basic building block of a virtual power plant, which includes various distributed energy nodes connected to the distribution network, such as distributed photovoltaic, distributed wind power, energy storage devices, controllable loads, and electric vehicle charging piles.

[0003] Virtual power plants can achieve large-scale aggregation and refined management of distributed energy resources. They can integrate the output of scattered distributed energy sources to form a dispatchable and controllable equivalent power source, respond to the power regulation requirements issued by the upper-level power grid, participate in grid peak shaving, valley filling and backup ancillary services, smooth grid power fluctuations, and improve the stability and security of distribution network operation.

[0004] For example, Chinese invention patent application CN111563786A provides a blockchain-based virtual power plant control platform and operation method. It constructs a four-layer blockchain virtual power plant control platform that includes a physical layer, a data layer, a core layer, and an application layer. Through the registration of distributed energy units on the chain, the collection of operation information on the chain, the release of control requirements on the chain, market bidding and matching, and the issuance of control instructions and transaction settlement driven by smart contracts, it realizes decentralized, trusted, and collaborative scheduling of multiple participants in the virtual power plant.

[0005] The relevant technologies pose risks that the load current of the distribution feeder branches may exceed the safe current carrying capacity, causing line overload and distribution network operation safety accidents. Therefore, it is necessary to achieve more reasonable coordinated scheduling of distributed energy managed by virtual power plants. Summary of the Invention

[0006] To achieve more rational coordinated scheduling of distributed energy resources managed by virtual power plants, this application provides a blockchain-based method and system for coordinated scheduling of distributed energy resources in virtual power plants.

[0007] According to a first aspect of the embodiments of this application, a blockchain-based method for collaborative scheduling of distributed energy resources in a virtual power plant is provided, comprising: acquiring node status files of each distributed energy node; the node status files include feeder branch identifiers, distribution cable attribute parameters, and real-time operating parameters of each distributed energy node; acquiring the target total power issued by the upper-level power grid; determining the pre-scheduled output power of candidate nodes of the distributed energy nodes based on the target total power and the node status files; classifying candidate nodes located on the same distribution feeder branch according to the feeder branch identifier; and determining the superimposed expected current peak value of the distribution feeder branch in the pre-scheduled state based on the classification result and the pre-scheduled output power; determining the expected total load current of the distribution feeder branch based on the current load current of the distribution feeder branch and the superimposed expected current peak value; when the expected total load current is greater than or equal to a preset current carrying capacity, performing a consensus freezing operation on the candidate nodes on the distribution feeder branch to obtain a target frozen node; adjusting the action timing and output power allocation of the target frozen node based on historical discharge contribution data to obtain a scheduling scheme that makes the updated expected total load current less than the preset current carrying capacity; and outputting a distributed energy collaborative scheduling instruction according to the scheduling scheme.

[0008] This enables more rational coordinated scheduling of distributed energy resources managed by virtual power plants.

[0009] Optionally, the node status profile is constructed in the following way: the total path impedance of each distributed energy node to the busbar is determined according to the power distribution cable attribute parameters; a status mapping table between the blockchain hash address of each distributed energy node and the feeder branch identifier is constructed in the blockchain smart contract; the logical attribution characteristic information of the corresponding node is obtained by searching the status mapping table; and the total path impedance is associated with the real-time operating parameters and the logical attribution characteristic information to obtain the node status profile.

[0010] In this way, by forming logical attribution characteristic information, the communication interaction delay of cross-system topology verification is reduced, and the accuracy of virtual power plant constraint mapping to the underlying power distribution network is improved.

[0011] Optionally, the superimposed expected peak current is determined as follows: Based on the feeder branch identifiers contained in the node status file, candidate nodes are clustered into similar line groups to obtain a set of similar nodes with the same feeder branch identifier; based on the distribution network operating voltage standard and the power factor parameters corresponding to the distributed energy nodes, the pre-scheduled output power of each candidate node within the similar node set is calculated using AC / DC equivalent transformation to obtain the single-point expected injection current used to characterize the current fed into the grid by each candidate node; according to the distribution order of the total path impedance contained in the node status file, the single-point expected injection current on the same distribution feeder branch is sequentially accumulated in a directional manner, and the maximum value obtained from the accumulation is taken as the superimposed expected peak current of the distribution feeder branch.

[0012] In this way, the expected injection current at a single point is calculated and superimposed by directional accumulation processing according to the distribution order of the total path impedance. This avoids the local current overestimation error caused by simple algebraic addition and improves the reliability of the prediction results for distribution feeder branches.

[0013] Optionally, a consensus freezing operation is performed on candidate nodes on the distribution feeder branch to obtain the target frozen node, including: generating a conflict blocking signal for the distribution feeder branch whose expected total load current is greater than or equal to the preset current carrying capacity; issuing a consensus freezing instruction to the blockchain hash address corresponding to all candidate nodes mounted on the distribution feeder branch according to the conflict blocking signal; blocking the pre-scheduled output power corresponding to the candidate node from entering the block packaging verification sequence according to the consensus freezing instruction, and determining the corresponding candidate node as the target frozen node.

[0014] In this way, by introducing a blocking control mechanism based on consensus-freezing instructions to block the pre-scheduled output power from entering the block packaging verification sequence, dangerous power injection behaviors that may cause grid overload can be stopped at the smart contract level, thus ensuring the safety of the power system.

[0015] Optionally, the timing of actions and the allocation of output power for target frozen nodes are adjusted based on historical discharge contribution data, including: using the cumulative actual discharge of the target frozen nodes within a preset time period as historical discharge contribution data; performing a descending sorting operation on the target frozen nodes on the same distribution feeder branch based on the historical discharge contribution data; maintaining the pre-scheduled output power and configuring a first action timestamp for target frozen nodes located in the first preset interval; and delaying the response action start time by a preset buffer time to obtain a second action timestamp for target frozen nodes located in the second preset interval; the ranking in the first preset interval is higher than the ranking in the second preset interval, and the first and second preset intervals do not overlap.

[0016] In this way, by configuring different action timestamps in different preset intervals based on historical discharge contribution data, the time dimension peak shaving mechanism enables multiple nodes on the same distribution feeder branch to stagger power output peaks on the time axis, which not only eliminates the risk of superimposed load current exceeding limits, but also ensures the priority response of high contribution nodes.

[0017] Optionally, the method further includes: when it is determined that the second action timestamp exceeds the time window corresponding to the target total power, for the target frozen node located in the second preset interval, performing a reduction operation on the pre-scheduled output power of the target frozen node according to a preset ratio parameter to obtain the corrected output power.

[0018] Optionally, the scheduling scheme is obtained by: merging the state data corresponding to the candidate nodes that have not performed the consensus freeze operation with the adjusted data corresponding to the target frozen node obtained after adjusting the timing of the execution action and the output power allocation to obtain a global node set; and integrating and packaging the global node set together with the corresponding timestamp and the adjusted output power to obtain the scheduling scheme.

[0019] Optionally, the scheduling scheme includes action timestamps and power limit information for each distributed energy node; outputting distributed energy collaborative scheduling instructions based on the scheduling scheme, including: monitoring the connectivity status information of the encrypted communication tunnel between the control center and each distributed energy node; if the connectivity status information is determined to be normal, encapsulating and packaging the scheduling scheme to obtain a scheduling proposal block; after the blockchain network verifies the scheduling proposal block, extracting action timestamps and power limit information for each distributed energy node from the scheduling scheme; encapsulating and generating control messages based on the action timestamps and power limit information, and issuing the control messages as distributed energy collaborative scheduling instructions to the controllers of each distributed energy node.

[0020] Optionally, the real-time operating parameters of the distributed energy nodes are obtained in the following ways: when a target node in the distributed energy nodes is detected to be in a state of communication loss, a set of neighboring nodes with the same meteorological associated area label as the target node and in a normal communication state are retrieved from the blockchain network; the external environment measurement data and the real-time change slope of the output power of the neighboring node set are obtained; based on the external environment measurement data and the real-time change slope of the output power, the basic capacity of the target node before the communication loss is calculated to obtain the estimated capacity data, and the estimated capacity data is used as the real-time operating parameters of the target node.

[0021] In this way, when communication is lost, the estimated capacity data of the target node can be inferred by using a set of adjacent nodes with the same meteorological associated area labels. This can maintain the continuity of coordinated scheduling in a weak communication network environment and improve the robustness of the virtual power plant perception and control system.

[0022] According to a second aspect of the embodiments of this application, a blockchain-based virtual power plant distributed energy collaborative scheduling system is provided, comprising: a processor and a memory, wherein the memory stores computer program instructions, and the computer program instructions, when executed by the processor, implement the steps of the blockchain-based virtual power plant distributed energy collaborative scheduling method provided in the first aspect of this application.

[0023] The technical solutions provided by the embodiments of this application may include the following beneficial effects: Based on the node status files of different distributed energy sources in the virtual power plant and the topological association with the distribution network, the expected peak current generated by power injection in each distribution feeder branch is calculated in advance during the pre-scheduling stage. This can identify potential local line over-limit risks in advance. By introducing consensus freezing and allocation adjustment operations in a blockchain environment, serious overload damage to the underlying distribution cables is avoided while ensuring that the overall target total power response of the upper-level power grid meets the standard. This improves the security of the distributed energy cluster collaborative scheduling and the operational stability of the distribution network, and achieves more reasonable collaborative scheduling of the distributed energy sources managed by the virtual power plant.

[0024] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating a blockchain-based distributed energy collaborative scheduling method for virtual power plants, according to an exemplary embodiment. Figure 2 This is a schematic diagram comparing the peak current at the junction points of the coordinated scheduling strategy; Figure 3 This is a schematic diagram illustrating the structure of a blockchain-based distributed energy collaborative dispatch system for virtual power plants, according to an exemplary embodiment. Detailed Implementation

[0026] To achieve more reasonable coordinated scheduling of distributed energy resources managed by virtual power plants, this application provides a blockchain-based method and system for coordinated scheduling of distributed energy resources in virtual power plants. The blockchain-based method for coordinated scheduling of distributed energy resources in virtual power plants provided in this application can be applied to the cloud scheduling master station equipment of virtual power plants. Specifically, it can be operated through a cloud-edge-device collaborative computing architecture composed of the cloud scheduling master station equipment, edge computing gateway, and underlying inverter.

[0027] The cloud-based scheduling master station equipment is equipped with a blockchain master node cluster in the form of a consortium blockchain. The blockchain master node cluster runs a distributed shared ledger system and smart contract execution environment based on Ethereum or Hyperledger architecture. The underlying equipment such as photovoltaic power generation units and energy storage battery packs deployed on the user side are connected to the communication network through edge computing gateways. The edge computing gateways participate in decentralized data interaction as light nodes of the blockchain.

[0028] Figure 1 This is a flowchart illustrating a blockchain-based distributed energy collaborative scheduling method for virtual power plants, according to an exemplary embodiment. Figure 1 As shown, the method includes the following steps.

[0029] In step S101, the node status files of each distributed energy node are obtained.

[0030] The node status file includes the feeder branch identifier, power distribution cable attribute parameters, and real-time operating parameters of each distributed energy node; the control program drives the edge computing gateway to send status query messages to the underlying inverter through a serial communication interface or an industrial Ethernet interface, and the underlying inverter reads the real-time monitoring values ​​of its internal sensors after receiving the status query message.

[0031] The edge computing gateway polls and collects real-time monitoring values ​​according to a preset data collection frequency. For example, the data collection frequency is set to perform a complete data sampling and reporting operation every 5 seconds. The edge computing gateway packages the collected data into data frames and sends them to the cloud scheduling master station device through a wireless network.

[0032] After receiving multiple data frames, the cloud-based dispatching master station can strip the protocol headers of the data frames to extract the data payload content that carries the actual valid information. The cloud-based dispatching master station can also access the power geographic information database deployed on the local server to retrieve the topology connection table corresponding to the distributed energy nodes.

[0033] Based on the topology connection table, the cloud-based dispatch master station equipment extracts the feeder branch identifier, which represents the connection hierarchy. The feeder branch identifier is used to uniquely indicate the specific branch line number of the distributed energy node connected to the distribution network.

[0034] Power distribution cable attribute parameters can be extracted from the power geographic information database. These parameters include the cross-sectional area, material resistivity, and line length of the power distribution cable. These parameters reflect the material limits and inherent electrical characteristics of the power transmission channel. Based on these power distribution cable attribute parameters, the cloud-based dispatch master station equipment can assess the theoretical upper limit of the current load carrying capacity of different line sections.

[0035] It can integrate real-time operating parameters obtained from the edge computing gateway, including the current available remaining capacity and battery state of charge of the distributed energy node, as well as the real-time output voltage of the photovoltaic panel.

[0036] The feeder branch identifier, power distribution cable attribute parameters, and real-time operating parameters are combined and encapsulated according to a standardized data dictionary format to generate a node status file that corresponds one-to-one with each distributed energy node.

[0037] In one embodiment, the node status profile is constructed as follows: the total path impedance of each distributed energy node to the busbar is determined based on the power distribution cable attribute parameters; a state mapping table between the blockchain hash address of each distributed energy node and the feeder branch identifier is constructed in the blockchain smart contract; the logical attribution characteristic information of the corresponding node is obtained by searching the state mapping table; and the total path impedance is associated with the real-time operating parameters and the logical attribution characteristic information to obtain the node status profile.

[0038] The cloud-based dispatching master station extracts the material resistivity and line laying length parameters from the power distribution cable attribute parameters, multiplies the material resistivity and line laying length parameters, and then divides them by the cross-sectional area parameters of the power distribution cable to obtain the ohmic resistance value of a single cable segment.

[0039] Based on the power grid topology, the ohmic resistance values ​​of all single cable segments passing through the distributed energy node to the substation busbar are calculated by accumulating them step by step to obtain the total path impedance that characterizes the electrical distance between the nodes; the total path impedance can characterize the degree of voltage loss of electrical energy during transmission.

[0040] The cloud-based scheduling master station device calls the secure hash algorithm module to perform encrypted hashing operations on the identity registration information of distributed energy nodes, such as generating a 256-bit blockchain hash address. A mapping table between the blockchain hash address of the distributed energy node and the feeder branch identifier can be built in the blockchain smart contract. The logical affiliation characteristic information of the corresponding node can be obtained by retrieving the mapping table. By embedding the feeder branch identifier into the blockchain hash address, the topological location of the corresponding distributed energy node can be obtained simultaneously while verifying the source of the data block.

[0041] A relational data table can be constructed, with the completed logical attribution feature information used as the primary key index field of the data table; the total path impedance and the collected real-time operating parameters are stored in the data table as attribute fields within the data rows, so that the logical attribution feature information, the total path impedance, and the real-time operating parameters form a bound association mapping relationship at the database level, and the data table records containing the complete mapping relationship are exported as node status archives, which improves the efficiency of multi-dimensional feature data joint retrieval in the subsequent scheduling calculation stage.

[0042] In step S102, the superimposed expected peak current of the distribution feeder branch is determined.

[0043] Obtain the target total power issued by the upper-level power grid, determine the pre-scheduled output power of candidate nodes of distributed energy nodes based on the target total power and node status files, classify candidate nodes located in the same distribution feeder branch according to the feeder branch identifier, and determine the superimposed expected peak current of the distribution feeder branch in the pre-scheduled state based on the classification results and the pre-scheduled output power.

[0044] It can continuously monitor dispatch messages from the provincial power grid dispatch center through a data transmission protocol that conforms to power system standards. After capturing the dispatch message, it performs identity authentication and decryption operations, extracts the active power peak shaving or valley filling value that the virtual power plant needs to respond to within the current dispatch cycle specified in the instruction message, and uses the extracted active power value as the target total power.

[0045] It can traverse all node status files stored locally, extract nodes that are in normal online status and have remaining capacity greater than the set minimum response threshold to form a candidate node set for distributed energy nodes; read the real-time operating parameters corresponding to each node in the candidate node set, and use a capacity ratio allocation algorithm to divide and calculate the target total power according to the proportion of available capacity of each candidate node by mathematical division and multiplication, and use the power quota obtained by the division calculation as the pre-scheduled output power of each candidate node.

[0046] The feeder branch identifier recorded in the status file of each node in the candidate node set can be read sequentially. A string matching algorithm is used to compare the read feeder branch identifier with the name of the initialized data set. The data pointer of the candidate node that matches successfully is written into the corresponding name data set. After the comparison operation of all candidate nodes is completed, the candidate nodes located in the same distribution feeder branch are centrally classified and stored.

[0047] In one embodiment, the superimposed expected peak current is determined as follows: Candidate nodes are clustered into homogeneous lines based on the feeder branch identifiers contained in the node status file to obtain a set of homogeneous nodes with the same feeder branch identifier; based on the distribution network operating voltage standard and the power factor parameters corresponding to the distributed energy nodes, the pre-scheduled output power of each candidate node within the homogeneous node set is calculated using AC / DC equivalent transformation to obtain the single-point expected injection current characterizing the current fed into the grid by each candidate node; according to the distribution order of the total path impedance contained in the node status file, the single-point expected injection current on the same distribution feeder branch is sequentially accumulated in a directional manner, and the maximum value obtained from the accumulation is taken as the superimposed expected peak current of the distribution feeder branch.

[0048] The cloud-based scheduling master station device starts the unsupervised clustering algorithm program in the data mining module. It uses the feeder branch identifier contained in the node status file as the input dimension of the clustering algorithm for calculation. It divides the clusters by determining whether the feeder branch identifier strings are completely consistent. Candidate nodes belonging to the same cluster are combined into a set of homogeneous nodes with the same feeder branch identifier. The underlying devices with the same electrical topology root node are treated as an independent set to provide a boundary for subsequent local line over-limit early warning calculation.

[0049] The cloud-based dispatching master station equipment can retrieve the distribution network operating voltage standards stored in the system configuration library. For example, it can read that the standard line voltage parameter of the current distribution line is rated at 10 kV. Based on the distribution network operating voltage standards and the power factor parameters corresponding to the distributed energy nodes, it performs AC / DC equivalent transformation calculations on the pre-scheduled output power of each candidate node in the same source node set to obtain the single-point expected injection current used to characterize the current fed into the grid by each candidate node. The power factor is equal to the ratio of active power to apparent power in the AC grid, and is used to measure the degree of effective utilization of electricity by power generation equipment or power consumption equipment.

[0050] The cloud-based scheduling master station extracts the total path impedance value corresponding to each candidate node in the set of same-source nodes, and sorts the expected injection current of a single point in descending order from the largest to the smallest in the memory array according to the magnitude of the total path impedance value.

[0051] The process of current flowing towards the busbar can be simulated sequentially along the direction of decreasing impedance. Vector addition is performed on the expected injection current at each intersection node along the way to achieve directional accumulation. The peak current data that appears during the directional accumulation process is recorded and used as the superimposed expected peak current of the distribution feeder branch, ensuring that the prediction results conform to the current accumulation law of the distribution network trunk line.

[0052] In step S103, the target frozen node is determined from the distribution feeder branch.

[0053] Based on the current load current of the distribution feeder branch and the superimposed expected peak current, the expected total load current of the distribution feeder branch is determined. When the expected total load current is greater than or equal to the preset current carrying capacity, a consensus freezing operation is performed on the candidate nodes on the distribution feeder branch to obtain the target frozen node.

[0054] The current transformer measurement feedback data installed on the intelligent circuit breaker at the head end of the distribution feeder branch is retrieved through the distribution automation system interface. After filtering, the current transformer measurement feedback data is used as the current load current of the distribution feeder branch. The absolute value or vector superposition result of the difference between the superimposed expected current peak value and the current load current is calculated, and the calculated net power flow current peak value is used as the expected total load current of the distribution feeder branch.

[0055] Read the safe heating limit current parameter set in the nameplate ledger database of power distribution cable equipment, and multiply the safe heating limit current parameter by a safety margin factor of, for example, 0.9 to obtain the preset current carrying capacity; The calculated expected total load current is compared with the preset current carrying capacity. When the expected total load current is determined to be greater than or equal to the preset current carrying capacity, the cloud scheduling master station equipment triggers a consensus freeze operation command to execute power response suspension for candidate nodes in abnormal branches to obtain the target frozen node, thereby avoiding actual power grid line burnout faults.

[0056] In one embodiment, performing a consensus freeze operation on candidate nodes on a distribution feeder branch to obtain a target frozen node includes: generating a conflict blocking signal for the distribution feeder branch whose expected total load current is greater than or equal to a preset current carrying capacity; issuing a consensus freeze command to the blockchain hash address corresponding to all candidate nodes mounted on the distribution feeder branch according to the conflict blocking signal; blocking the pre-scheduled output power corresponding to the candidate node from entering the block packaging verification sequence according to the consensus freeze command; and determining the corresponding candidate node as the target frozen node.

[0057] The cloud-based dispatch master station equipment can generate a special interrupt code instruction with the highest execution priority within the risk assessment module. This special interrupt code instruction is then encoded and combined with the corresponding feeder branch number to generate a conflict blocking signal for distribution feeder branches whose expected total load current is greater than or equal to the preset current carrying capacity.

[0058] It can activate the circuit breaker logic pre-set in the smart contract, and accurately locate the blocking control target to the blockchain hash address corresponding to all candidate nodes mounted on the power distribution feeder branch according to the conflict blocking signal. The cloud scheduling master station equipment sends a consensus freeze command to the target blockchain hash address through the blockchain peer-to-peer network broadcast channel.

[0059] After receiving the consensus freeze instruction, the blockchain master node searches the memory pool for power scheduling proposal transaction data packets that contain the target blockchain hash address. The blockchain master node attaches a denial-of-service tag to the successfully matched power scheduling proposal transaction data packets and refuses to process such transaction data according to the consensus freeze instruction, thereby blocking the pre-scheduled output power of the candidate node from entering the block packaging and verification sequence.

[0060] The cloud-based scheduling master station will mark all candidate nodes that have been blocked from participating in the current round of consensus scheduling process, and determine the marked candidate nodes as target frozen nodes, cutting off the authorized path of over-limit power output at the decentralized ledger level.

[0061] In step S104, a distributed energy collaborative scheduling instruction is output.

[0062] Based on historical discharge contribution data, the timing of actions and the distribution of output power of the target frozen node are adjusted to obtain a scheduling scheme that makes the updated expected total load current less than the preset current carrying capacity. Distributed energy collaborative scheduling instructions are output according to the scheduling scheme.

[0063] Because the original scheduling plan had the risk of overload in the distribution network, some nodes were frozen and blocked. The cloud scheduling master station equipment started the recalculation engine to retrieve the past performance information of each node recorded in the cloud ledger data warehouse as historical discharge contribution data.

[0064] Based on historical discharge contribution data, the reputation priority of the target frozen node can be assessed, and the output rights of nodes with good historical performance can be guaranteed first within the safety boundary. The cloud scheduling master station equipment can adjust the timing of actions and the distribution of output power of the target frozen node under the action of delay trigger parameters and attenuation output coefficient.

[0065] After the cloud-based dispatch master station equipment performs the application allocation and adjustment operation, it recalculates the current state parameters of the line using the same pre-calculation logic to ensure that the updated expected total load current after parameter adjustment can smoothly fall back and be less than the preset current carrying capacity.

[0066] The cloud-based dispatching master station can merge the adjustment results that meet the security constraint verification with the regular dispatching instructions of the unfrozen nodes and save them as a dispatching scheme. The cloud-based dispatching master station then digitally signs and encrypts the dispatching scheme to generate distributed energy collaborative dispatching instructions, which are then sent to the underlying control terminal to execute specific switching actions.

[0067] In one embodiment, adjusting the timing of actions and the allocation of output power for target frozen nodes based on historical discharge contribution data includes: using the cumulative actual discharge amount of the target frozen nodes completing scheduling tasks within a preset time period as historical discharge contribution data; performing a descending sorting operation on target frozen nodes on the same distribution feeder branch based on the historical discharge contribution data; maintaining the pre-scheduled output power and configuring a first action timestamp for target frozen nodes ranked in the first preset interval; and delaying the response action start time by a preset buffer time to obtain a second action timestamp for target frozen nodes ranked in the second preset interval; the ranking in the first preset interval is higher than the ranking in the second preset interval, and the first preset interval and the second preset interval do not overlap.

[0068] The cloud-based scheduling master station device can, for example, set the past 30 days of calendar time span as a preset time period parameter, and call the blockchain explorer interface to query the historical transaction ledger records of each target frozen node that have been confirmed by the smart contract within the preset time period.

[0069] The cloud-based scheduling master station adds up the cleared electricity values ​​recorded in the historical transaction ledger, and uses the cumulative actual discharge electricity of the target frozen node that completes the scheduling task within the preset time period as historical discharge contribution data to represent its active participation in peak shaving interaction.

[0070] The cloud-based dispatching master station equipment can perform a descending sorting operation on all target frozen nodes on the same distribution feeder branch based on historical discharge contribution data. For example, the top 30% of the descending sorting results can be used as the first preset interval, and the target frozen nodes in the first preset interval can maintain their original calculated pre-scheduled output power.

[0071] The cloud-based scheduling master station can read the current standard time value provided by the local network time protocol server and configure the current standard time value as the first action timestamp for immediately executing scheduling instructions to the high-ranking nodes.

[0072] The ranking from 30% to the last position in the descending sort results is taken as the second preset interval. Target nodes in the second preset interval are frozen, and the priority of response scheduling is reduced. The start time of the response action is delayed on the time axis by adding a preset buffer time value, such as adding 1 minute, to obtain the second action timestamp that postpones the triggering of the output task.

[0073] The index number of the first preset interval in the sorting algorithm array is lower than the index number of the second preset interval, and the boundary node values ​​of the first preset interval and the second preset interval do not overlap. By controlling the peak time, the simultaneous large-scale discharge to the distribution network is effectively avoided, thus preventing the potential problem of a surge in local feeder current.

[0074] In one embodiment, if it is determined that the second action timestamp exceeds the time window corresponding to the target total power, the pre-scheduled output power of the target frozen node located in the second preset interval can be reduced according to a preset ratio parameter to obtain the corrected output power.

[0075] The cloud-based dispatch master station equipment can parse the specified task end time parameters from the dispatch messages issued by the provincial power grid as the time window cutoff boundary corresponding to the target total power.

[0076] The cloud-based scheduling master station will calculate the difference between the calculated second action timestamp and the task end time parameter. If the value of the second action timestamp is greater than the task end time parameter, it indicates that the time window has been exceeded. This means that adopting the time extension strategy will cause the node to miss the current scheduling cycle. For target frozen nodes that are in extreme situations and are ranked in the second preset interval, the cloud-based scheduling master station can call a preset ratio parameter with a value of 0.5 from the system operation parameter library.

[0077] Compression processing is achieved by multiplying the preset ratio parameter with the original pre-scheduled output power of the target frozen node, and a reduction operation is performed on the pre-scheduled output power to reduce the peak output intensity, ultimately obtaining the corrected output power after weakening.

[0078] In one embodiment, the scheduling scheme is obtained by merging the state data corresponding to the candidate nodes that have not performed consensus freezing operations with the adjusted data corresponding to the target frozen nodes obtained after adjusting the timing of the execution actions and the output power allocation to obtain a global node set; and integrating and packaging the global node set together with the corresponding timestamps and the adjusted output power allocation to obtain the scheduling scheme.

[0079] A connection query statement can be established in the relational database to extract the state data of candidate nodes that have passed the traffic security check and have not performed the consensus freeze operation into the memory buffer area.

[0080] The cloud-based scheduling master station synchronously extracts the adjusted data corresponding to the target frozen node after the allocation and adjustment operations such as action timing delay and output power reduction. The cloud-based scheduling master station uses a data splicing function to merge the above two batches of data fragments from different sources at the data structure level to obtain a global node set covering all equipment objects participating in this coordinated response.

[0081] It can scan the information of each underlying node record in the global node set, extract the device communication addressing routing address embedded in the record, and convert the addressing address in the global node set, together with the timestamp calculated for each device, and the output power corrected due to the current limiting and derating calculation, into a common text data exchange format document.

[0082] The cloud-based scheduling master station device performs compression algorithms on text data exchange format documents to achieve integrated packaging and obtain a final scheduling scheme data packet with reduced size and rigorous structure.

[0083] In one embodiment, after obtaining the corrected output power, the scheduling scheme can also be obtained in the following ways: Based on the difference in output power before and after the reduction operation of the target frozen node located in the second preset interval, obtain the single-point power reduction amount; accumulate the single-point power reduction amounts to obtain the peak-shaving power gap; from the global node set, select distribution feeder branches whose difference between the preset current carrying capacity and the expected total load current is greater than the safety margin as healthy feeder branches; extract candidate nodes connected to the healthy feeder branches and whose remaining available capacity is greater than the pre-scheduled output power to form a compensation node set; according to the proportion of the remaining available capacity of each candidate node in the compensation node set, proportionally allocate the peak-shaving power gap to obtain the supplementary compensation power of the corresponding candidate nodes and superimpose it to update the pre-scheduled output power; perform the load current safety verification operation again based on the updated pre-scheduled output power; after the verification is passed, integrate and package the data of the compensation node set and the target frozen node to obtain the final scheduling scheme.

[0084] By using a horizontal transfer compensation mechanism to address the peak-shaving power gap, the power reduced by the frozen nodes is precisely transferred to the surplus nodes of healthy feeder branches, thus avoiding the risk of local line overload and filling the peak-shaving gap, achieving a closed-loop response to the target total power of the power grid.

[0085] In one embodiment, the scheduling scheme includes action timestamps and power limit information for each distributed energy node; outputting distributed energy collaborative scheduling instructions according to the scheduling scheme, including: monitoring the connectivity status information of the encrypted communication tunnel between the control center and each distributed energy node; if the connectivity status information is determined to be normal, encapsulating and packaging the scheduling scheme to obtain a scheduling proposal block; after the blockchain network verifies the scheduling proposal block, extracting action timestamps and power limit information for each distributed energy node from the scheduling scheme; encapsulating the action timestamps and power limit information to generate control messages, and issuing the control messages as distributed energy collaborative scheduling instructions to the controllers of each distributed energy node.

[0086] It can send probe and diagnostic data packets to the underlying virtual private network gateway to continuously monitor the connectivity status between the control center server interface and the edge computing gateways of each distributed energy node.

[0087] Upon receiving, for example, three consecutive response data packets, thus confirming that the connectivity status is normal with no packet loss and the latency within an acceptable range, the cloud scheduling master station injects the specific values ​​of the scheduling plan into the standard blockchain transaction structure template, adds the initiator's timestamp anti-counterfeiting signature, encapsulates and packages the data, and generates a scheduling proposal block to be confirmed. The scheduling proposal block is then broadcast and disseminated to the peer node network of the consortium blockchain, and each participating master node independently verifies and votes on the legality of the plan.

[0088] After the blockchain network verifies the scheduling proposal block and anchors it to the main ledger chain, the cloud scheduling master station equipment can read the blockchain ledger data after the rights have been confirmed. From the scheduling scheme that has been stored on the chain, it can extract action timestamp parameters such as the first action timestamp or the second action timestamp required for the execution of the underlying hardware of each distributed energy node, as well as the power limit information corresponding to the corrected output power.

[0089] The cloud-based scheduling master station can convert the extracted action timestamps and power limit information into register read / write instruction codes that can be directly recognized by the underlying inverter, and encapsulate them using binary encoding rules to generate standardized control messages.

[0090] The cloud-based dispatch master station equipment uses control messages with clear guiding significance as unalterable distributed energy collaborative dispatch instructions, and sends them through the communication link to the underlying microprocessor hardware controllers of each distributed energy node to drive the completion of closed-loop actions.

[0091] Figure 2 This is a schematic diagram comparing the peak current at the confluence points of the coordinated scheduling strategy, as shown below. Figure 2 As shown, under the existing strategy evaluation using the traditional single-unit independent control mode, the peak value of the equivalent grid-connected current at the junction point of distribution feeders 01 to 05 all exceeds the network security threshold, which can easily cause severe overload and heat damage to local distribution feeder branches.

[0092] like Figure 2 As shown, after adopting the blockchain-based collaborative scheduling strategy provided in this application embodiment, by accurately predicting the expected total load current in the pre-scheduling stage, and when it is greater than or equal to the preset current carrying capacity, a consensus freeze operation is triggered and the timing of actions and the distribution adjustment of output power based on historical discharge contribution data are executed, such as extending the action timestamp or performing a derating operation, so that the expected total load current of distribution feeder 01 to distribution feeder 05 is ultimately controlled below the network security threshold.

[0093] Meanwhile, the load current of distribution feeders 06 to 15 is maintained within a safe range. The solution proposed in this application effectively avoids the risk of distribution cables exceeding limits while ensuring the distributed energy cluster responds to the grid target power, thus improving the operational stability and security of the distribution network.

[0094] In one embodiment, the real-time operating parameters of the distributed energy node are obtained as follows: when a target node in the distributed energy node is detected to be in a communication loss state, a set of neighboring nodes with the same meteorological associated area label as the target node and in a normal communication state are retrieved from the blockchain network; the external environment measurement data and the real-time change slope of the output power of the neighboring node set are obtained; based on the external environment measurement data and the real-time change slope of the output power, a projection operation is performed on the basic capacity of the target node before the communication loss to obtain the estimated capacity data; and the estimated capacity data is used as the real-time operating parameters of the target node.

[0095] The cloud-based scheduling master station can periodically check the arrival time interval of heartbeat packets uploaded by edge nodes. When it detects that the heartbeat interval timeout parameter exceeds the preset tolerance threshold and detects that the target node in the distributed energy node has lost communication and is offline, the cloud-based scheduling master station can trigger a data completion mechanism.

[0096] The cloud-based dispatch master station sends structured query statements to the ledger storage nodes of the blockchain network through the application programming interface. From the node registration metadata stored in the blockchain network, it retrieves nodes in the same region that have the same meteorological associated area label as the target node that has failed and are currently maintaining a normal communication heartbeat, forming a set of adjacent nodes.

[0097] Using the same meteorological associated regional labels for retrieval ensures that the selected neighboring nodes have a very high spatial similarity to the target node in terms of meteorological boundary conditions such as solar radiation intensity and ambient temperature.

[0098] The cloud-based scheduling master station sends a data access request to the set of adjacent nodes to obtain external environmental measurement data collected and transmitted by the light intensity sensor and temperature sensor of the set of adjacent nodes. At the same time, it calculates the real-time slope of the output power of the set of adjacent nodes over the past 10 minutes relative to the output power over time.

[0099] The cloud-based scheduling master station uses the target node's last valid remaining power value stored in its local memory before the network outage and communication loss as the starting point for the extrapolation. The cloud-based scheduling master station substitutes external environmental measurement data into the photovoltaic power generation efficiency conversion formula model to calculate the theoretical power generation change. Combined with the load consumption trend represented by the real-time output power change slope, the system uses integral calculation to perform a forward extrapolation calculation on the target node's basic capacity before the communication loss, obtaining estimated capacity data including simulation calculation results. This estimated capacity data directly replaces the missing sensor readings as the real-time operating parameters of the target node.

[0100] In one embodiment, after determining that the communication link of the target node has returned to normal, the actual state capacity data recorded in the local device of the target node can be extracted, the actual state capacity data can be compared with the estimated capacity data on the chain to obtain the compensation feature coefficient, and the compensation feature coefficient can be added to the pre-scheduled output power allocation stage of the target node in the next scheduling cycle.

[0101] After receiving the registration handshake request data packet re-initiated by the target node, the cloud scheduling master station device releases the offline suspension state of the network protocol stack and opens the bidirectional data transmission permission after confirming that the communication link connection of the target node has been restored to normal; it issues a historical log reading instruction to the target node to extract the absolute actual state capacity data recorded by the built-in storage chip of the target node in the local device during the communication interruption.

[0102] The estimated capacity data generated by the algorithm in the cloud and the extracted actual capacity data are submitted to the blockchain smart contract execution and verification environment. Subtraction and percentage calculation are performed on the two to compare the differences on the chain. The on-chain difference comparison can reflect the error range of the inference algorithm and obtain the compensation feature coefficient used to characterize the degree of energy deviation.

[0103] When the system enters a new round of power allocation, the cloud-based scheduling master station equipment can use the compensation characteristic coefficient containing energy deviation information as a multiplication coefficient to add to the calculation formula of the target node in the pre-scheduled output power allocation stage of the next scheduling cycle. By weakening or strengthening the allocation weight of the target node, the calculation error can be compensated and repaired.

[0104] In one embodiment, before outputting distributed energy collaborative scheduling instructions according to the scheduling scheme, the regional aggregation terminal deployed on the transformer side of the distribution area can be controlled to receive measurement status data with digital signatures pushed unidirectionally by the underlying sensing nodes. After the digital signature of the underlying sensing nodes is verified, the measurement status data received inside the distribution area is compressed and recombined to obtain a recombined status data set. The regional aggregation terminal is then controlled to participate in the consensus operation of the blockchain network as a single independent node based on the recombined status data set.

[0105] The cloud-based dispatch master station sends topology configuration commands to the edge-side network management system to control the regional aggregation terminals deployed in the low-voltage side equipment room of the transformer in the distribution network area, and to enable Bluetooth Low Energy communication or ZigBee short-range wireless communication receiving ports.

[0106] The regional aggregation terminal utilizes the underlying sensing nodes distributed in the user's home meter box to unidirectionally push measurement status data with a digital signature that has non-repudiation characteristics after encryption; the regional aggregation terminal can call the public key files corresponding to each underlying sensing node in the local certificate store to perform decryption and verification operations.

[0107] After the digital signature verification of the consensus operation confirms that the data source is authentic and has not been maliciously tampered with during transmission, the regional aggregation terminal can start the data filtering program to remove duplicates and compress the scattered and redundant measurement status data received within the distribution area using Huffman coding. The compression and reassembly operation is then performed to fuse and encapsulate the scattered data frames into a data carrier to obtain a reassembled status data set.

[0108] The cloud-based scheduling master station can control the regional aggregation terminals, representing the entire transformer's micro-topology network. This enables the regional aggregation terminals to cooperate as independent nodes based on the reorganized state dataset, participating in the transaction broadcasting and consensus operation process of the cloud consortium blockchain network. This avoids network congestion failures that may be caused by the underlying devices to the cloud main chain.

[0109] Figure 3 This is a schematic diagram illustrating the structure of a blockchain-based virtual power plant distributed energy collaborative dispatch system 1000 according to an exemplary embodiment. (Refer to...) Figure 3 The blockchain-based virtual power plant distributed energy collaborative scheduling system 1000 includes a processor 1100 and a memory 1200. The memory 1200 stores computer program instructions, which, when executed by the processor 1100, implement all or part of the steps of the blockchain-based virtual power plant distributed energy collaborative scheduling method in this application.

[0110] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only.

[0111] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A blockchain-based distributed energy collaborative scheduling method for virtual power plants, applied to virtual power plants, characterized in that: The method includes: Obtain the node status profile of each distributed energy node; the node status profile includes the feeder branch identifier, power distribution cable attribute parameters and real-time operating parameters of each distributed energy node; Obtain the target total power issued by the upper-level power grid, determine the pre-scheduled output power of candidate nodes of distributed energy nodes based on the target total power and node status files, classify candidate nodes located in the same distribution feeder branch according to the feeder branch identifier, and determine the superimposed expected peak current of the distribution feeder branch in the pre-scheduled state based on the classification results and the pre-scheduled output power. Based on the current load current of the distribution feeder branch and the superimposed expected peak current, the expected total load current of the distribution feeder branch is determined. When the expected total load current is greater than or equal to the preset current carrying capacity, a consensus freezing operation is performed on the candidate nodes on the distribution feeder branch to obtain the target frozen node. Based on historical discharge contribution data, the timing of actions and the distribution of output power of the target frozen node are adjusted to obtain a scheduling scheme that makes the updated expected total load current less than the preset current carrying capacity. Distributed energy collaborative scheduling instructions are output according to the scheduling scheme.

2. The blockchain-based distributed energy collaborative scheduling method for virtual power plants according to claim 1, characterized in that, The node state profile is constructed in the following way: The total path impedance from each distributed energy node to the busbar is determined based on the attribute parameters of the power distribution cable. A state mapping table between the blockchain hash address of each distributed energy node and the feeder branch identifier is constructed in the blockchain smart contract. By searching the state mapping table, the logical attribution characteristic information of the corresponding node is obtained. The total path impedance is associated with the real-time operating parameters and the logical attribution characteristic information to obtain the node state profile.

3. The blockchain-based distributed energy collaborative scheduling method for virtual power plants according to claim 1, characterized in that, The expected peak current is determined in the following way: Based on the feeder branch identifiers contained in the node status files, the candidate nodes are clustered into nodes with the same feeder branch identifiers to obtain a set of nodes with the same feeder branch identifiers. Based on the distribution network operating voltage standard and the power factor parameters corresponding to the distributed energy nodes, the pre-scheduled output power of each candidate node in the same source node set is calculated by AC / DC equivalent transformation to obtain the single-point expected injection current used to characterize the current fed into the grid by each candidate node. According to the distribution order of the total path impedance contained in the node status file, the single-point expected injection current on the same distribution feeder branch is sequentially accumulated in a directional manner, and the maximum value obtained by accumulation is taken as the superimposed expected current peak value of the distribution feeder branch.

4. The blockchain-based distributed energy collaborative scheduling method for virtual power plants according to claim 1, characterized in that, A consensus freeze operation is performed on candidate nodes on the distribution feeder branch to obtain the target frozen node, including: Generate a conflict blocking signal for the distribution feeder branch whose expected total load current is greater than or equal to the preset current carrying capacity, and issue a consensus freeze command to the blockchain hash address corresponding to all candidate nodes mounted on the distribution feeder branch based on the conflict blocking signal. According to the consensus freeze command, the pre-scheduled output power of the candidate node is blocked from entering the block packaging verification sequence, and the corresponding candidate node is determined as the target freeze node.

5. The blockchain-based distributed energy collaborative scheduling method for virtual power plants according to claim 1, characterized in that, Based on historical discharge contribution data, the timing of actions and the allocation of output power for the target frozen node are adjusted, including: The cumulative actual discharge volume of the target frozen node that completes the scheduling task within the preset time period is used as historical discharge contribution data. Based on the historical discharge contribution data, the target frozen nodes on the same distribution feeder branch are sorted in descending order. For target frozen nodes located in the first preset interval, the pre-scheduled output power is kept unchanged and a first action timestamp is configured; for target frozen nodes located in the second preset interval, the response action start time is delayed by a preset buffer time to obtain a second action timestamp; the ranking in the first preset interval is higher than the ranking in the second preset interval, and the first preset interval and the second preset interval do not overlap.

6. The blockchain-based distributed energy collaborative scheduling method for virtual power plants according to claim 5, characterized in that, The method further includes: If the second action timestamp exceeds the time window corresponding to the target total power, for the target frozen node located in the second preset interval, the pre-scheduled output power of the target frozen node is reduced according to the preset ratio parameter to obtain the corrected output power.

7. The blockchain-based distributed energy collaborative scheduling method for virtual power plants according to claim 1, characterized in that, The scheduling scheme is obtained through the following methods: The state data corresponding to the candidate nodes that have not performed the consensus freeze operation is merged with the adjusted data corresponding to the target frozen node obtained after adjusting the timing of the execution action and the distribution of output power to obtain the global node set. The global node set, along with its corresponding timestamp and the adjusted output power, are integrated and packaged to obtain the scheduling scheme.

8. The blockchain-based distributed energy collaborative scheduling method for virtual power plants according to claim 1, characterized in that, The scheduling scheme includes action timestamps and power limit information for each distributed energy node; Output distributed energy collaborative dispatch instructions according to the dispatch scheme, including: The monitoring and control center monitors the connectivity status of the encrypted communication tunnel between the control center and each distributed energy node. If the connectivity status is determined to be normal, the scheduling scheme is encapsulated and packaged to obtain the scheduling proposal block. After the blockchain network verifies the scheduling proposal block, the action timestamps and power limit information for each distributed energy node are extracted from the scheduling scheme. The control message is generated by encapsulating the action timestamp and power limit information, and then sent to the controller of each distributed energy node as a distributed energy collaborative scheduling instruction.

9. The blockchain-based distributed energy collaborative scheduling method for virtual power plants according to claim 1, characterized in that, The real-time operating parameters of distributed energy nodes are obtained through the following methods: When a target node in a distributed energy node is detected to be out of communication, the set of neighboring nodes with the same meteorological associated area label as the target node and in normal communication status is retrieved from the blockchain network. The system acquires external environmental measurement data and real-time output power change slope of the adjacent node set. Based on the external environmental measurement data and real-time output power change slope, it performs a simulation operation on the basic capacity of the target node before the communication loss to obtain the estimated capacity data. The estimated capacity data is then used as the real-time operating parameter of the target node.

10. A blockchain-based distributed energy collaborative dispatch system for virtual power plants, characterized in that, include: A processor and a memory, the memory storing computer program instructions, which, when executed by the processor, implement the blockchain-based distributed energy collaborative scheduling method for virtual power plants according to any one of claims 1-9.

Images