Data processing method and device of energy internet marketing service system and storage medium

Abstract

The invention relates to an energy internet marketing service system data processing method and device and a storage medium, the system comprises a data medium station, a plurality of energy sales terminals and a plurality of distributed databases, and the plurality of distributed databases are electrically connected with the data medium station through corresponding database nodes. A plurality of energy sales terminals are electrically connected with a data center through corresponding scheduling nodes. The method comprises the following steps: acquiring data acquisition information on each database node; and importing a multi-mode fusion evaluation model, and evaluating the operation data information based on the multi-mode fusion evaluation model to obtain corresponding node evaluation information. Through comprehensive analysis of the characteristics, the most suitable database node for data operation can be more accurately selected through a multi-mode fusion evaluation model. And performing data adjustment on the data object based on the evaluation result so as to balance the operation condition of the database node.

Description

Technical Field

[0001] The present invention relates to the field of Internet technologies, and particularly to a data processing method, apparatus, and storage medium for an energy Internet marketing service system. Background Art

[0002] The energy Internet marketing service system is an important part of the energy Internet, aiming to achieve efficient coordination and market-oriented services in the entire energy production, transmission, and consumption chain through digital, intelligent, and networked technical means. It takes user needs as the core, integrates energy data, trading platforms, and service resources, promotes the precise matching of energy supply and demand, improves energy utilization efficiency, and contributes to the realization of the "dual carbon" goal.

[0003] With the rapid development of the energy Internet, the energy marketing service system faces the need for processing and updating massive amounts of data. Traditional data processing methods have problems such as uneven data storage, difficulty in ensuring data consistency, and heavy burdens on scheduling nodes, seriously affecting the performance and stability of the system. Therefore, a new data processing method is needed to solve these problems.

[0004] In the data center of the energy Internet marketing service system, traditional data update methods have problems such as uneven data storage, difficulty in ensuring data consistency, and heavy burdens on scheduling nodes. Although the patent No. CN119377240A proposes an update method for the data center of the energy Internet marketing service system, there are still many deficiencies in actual applications, such as the limitations of the graph neural network leading to the non-optimal selection of database nodes, potential risks in ensuring data consistency, and the burden on scheduling nodes not being completely eliminated. Therefore, a more optimized data processing method is needed to improve the performance and stability of the system. Summary of the Invention

[0005] Based on this, it is necessary to provide a data processing method, apparatus, and storage medium for an energy Internet marketing service system to address the problem of unstable system performance caused by uneven burdens on database nodes as described above.

[0006] A data processing method for an energy Internet marketing service system, the energy Internet marketing service system including a data center, a plurality of energy sales terminals, and a plurality of distributed databases, the plurality of distributed databases being electrically connected to the data center through corresponding database nodes respectively, and the plurality of energy sales terminals being electrically connected to the data center through corresponding scheduling nodes respectively, the method including:

[0007] Obtaining data collection information on each of the database nodes;

[0008] Import the data acquisition information into a multi-modal fusion evaluation model, and evaluate the operation data information based on the multi-modal fusion evaluation model to obtain corresponding node evaluation information;

[0009] Based on the node evaluation information, perform data adjustment on the data objects corresponding to the database nodes to balance the operation of the database nodes.

[0010] In one preferred embodiment, the importing the data acquisition information into a multi-modal fusion evaluation model and evaluating the operation data information based on the multi-modal fusion evaluation model to obtain corresponding node evaluation information includes:

[0011] Extract data feature information from the data acquisition information to obtain data feature information;

[0012] Import the data feature information into the multi-modal fusion evaluation model, where the multi-modal fusion evaluation model includes a preset graph neural network unit, a load balancing unit, and a reinforcement learning unit, to obtain the graph neural network information y GNN , load balancing information y RLB and reinforcement learning information y RL ;

[0013] Output the node evaluation information based on a node selection mechanism:

[0014] y = w1y GNN + w2y RLB + w3y RL

[0015] where w1, w1, w1 are weight coefficients, and y is the node evaluation information.

[0016] In one preferred embodiment, the obtaining the data acquisition information on each of the database nodes includes:

[0017] Collect data from database nodes by means of a distributed transaction mechanism.

[0018] In one preferred embodiment, the collecting data from database nodes by means of a distributed transaction mechanism includes:

[0019] Obtain data request information;

[0020] Write the operation record corresponding to the data request information into a log file;

[0021] Initiate a pre-commit request to the corresponding database node and receive a response;

[0022] When the request responses of all database nodes are received, initiate a formal request message to the corresponding database node.

[0023] In one preferred embodiment, the data adjustment of the data object corresponding to the database node based on the node evaluation information to balance the operation of the database node includes:

[0024] Establish a scheduling cluster, where the scheduling cluster includes multiple database nodes;

[0025] Based on the node evaluation information, perform data adjustment on the corresponding data objects of multiple database nodes in the scheduling cluster.

[0026] In one preferred embodiment, the method further includes:

[0027] Update the version of the adjusted data object.

[0028] In one preferred embodiment, the updating of the version of the adjusted data object includes:

[0029] Obtain the version information of the adjusted data object and update the version information;

[0030] Incorporate the current timestamp into the current data object.

[0031] The method disclosed in the above embodiments of the present invention evaluates the data operation situation on the database node, and through comprehensive analysis of these features, the multi-modal fusion evaluation model can more accurately select the most suitable database node for data operation. And based on the evaluation result, data adjustment is performed on the data object to balance the operation of the database node.

[0032] A data processing device for an energy Internet marketing service system, the energy Internet marketing service system includes a data middle platform, multiple energy sales terminals and multiple distributed databases, the multiple distributed databases are electrically connected to the data middle platform through corresponding database nodes respectively, and the multiple energy sales terminals are electrically connected to the data middle platform through corresponding scheduling nodes respectively. The device includes:

[0033] A data acquisition module for acquiring data acquisition information on each of the database nodes;

[0034] A node evaluation module for importing the data acquisition information into a multi-modal fusion evaluation model and evaluating the operation data information based on the multi-modal fusion evaluation model to obtain corresponding node evaluation information;

[0035] A data adjustment module for performing data adjustment on the data object corresponding to the database node based on the node evaluation information to balance the operation of the database node.

[0036] In one of them, the node evaluation module includes:

[0037] A feature extraction unit for extracting data features from the data acquisition information to obtain data feature information;

[0038] A feature evaluation unit for importing the data feature information into the multi-modal fusion evaluation model, which includes a preset graph neural network unit, a load balancing unit, and a reinforcement learning unit, to obtain the graph neural network information y GNN , load balancing information y RLB and reinforcement learning information y RL ;

[0039] Output the node evaluation information based on the node selection mechanism:

[0040] y = w1y GNN + w2y RLB + w3y RL

[0041] where w1, w1, w1 are weight coefficients, and y is the node evaluation information.

[0042] The device disclosed in the above embodiment of the present invention evaluates the data operation status on the database node as described above. By comprehensively analyzing these features, the multi-modal fusion evaluation model can more accurately select the most suitable database node for data operation. And based on the evaluation result, data adjustment is performed on the data object to balance the operation status of the database node.

[0043] A computer storage medium, on which a computer program is stored. When the computer program on the computer storage medium is executed by a processor, the method described in any one of the above is implemented.

[0044] The above computer storage medium of the present invention evaluates the data operation status on the database node as described above using the above method. By comprehensively analyzing these features, the multi-modal fusion evaluation model can more accurately select the most suitable database node for data operation. And based on the evaluation result, data adjustment is performed on the data object to balance the operation status of the database node. Description of the Drawings

[0045] Figure 1 It is a flowchart of a data processing method for an energy Internet marketing service system according to the first preferred embodiment of the present invention;

[0046] Figure 2 It is a detailed flowchart of step S10 of a data processing method for an energy Internet marketing service system according to the first preferred embodiment of the present invention;

[0047] Figure 3 It is a flowchart of the detailed steps of step S20 in a method for processing data of an energy Internet marketing service system according to the first preferred embodiment of the present invention;

[0048] Figure 4 It is a flowchart of the detailed steps of step S30 in a method for processing data of an energy Internet marketing service system according to the first preferred embodiment of the present invention;

[0049] Figure 5 It is a flowchart of a method in another embodiment of a method for processing data of an energy Internet marketing service system according to the first preferred embodiment of the present invention;

[0050] Figure 6 It is a flowchart of the detailed steps of step S40 in another embodiment of a method for processing data of an energy Internet marketing service system according to the first preferred embodiment of the present invention;

[0051] Figure 7 It is a schematic diagram of modules of a device for processing data of an energy Internet marketing service system according to the second preferred embodiment of the present invention. Specific Embodiments

[0052] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0053] It should be noted that when an element is referred to as being "disposed on" another element, it can be directly on the other element or there may also be an intermediate element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be an intermediate element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are only for the purpose of illustration and do not represent the only implementation.

[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used in the description of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0055] Such as Figure 1As shown, the first preferred embodiment of the present invention discloses a data processing method for an energy Internet marketing service system. The energy Internet marketing service system includes a data center, multiple energy sales terminals, and multiple distributed databases. The multiple distributed databases are electrically connected to the data center through corresponding database nodes, and the multiple energy sales terminals are electrically connected to the data center through corresponding scheduling nodes. The method includes:

[0056] S10: Obtain the data collection information on each of the database nodes;

[0057] Specifically, in this step, the database nodes are data-collected by means of a distributed transaction mechanism. In a distributed database environment, a distributed data collection tool can be used to collect and integrate data from multiple nodes.

[0058] Distributed data collection is a technology that distributes data collection tasks to multiple nodes or devices to improve the efficiency and reliability of data collection. In the database field, distributed data collection can be used to collect data from multiple database nodes or servers for analysis, monitoring, and management.

[0059] Distributed data collection system architecture: Data collection nodes are responsible for collecting data from database nodes. These nodes can be servers, sensors, or other devices connected to the above databases. Data transmission: In this embodiment, the collected data can be transmitted through the network to a central server or data storage.

[0060] In the data update process of the above step, multiple database nodes involved are incorporated into a distributed transaction, and the two-phase commit (2PC) or three-phase commit (3PC) protocol is used to ensure data consistency. For example, when performing a data addition or modification operation, the operation record is first written to a log file, and then a pre-commit request is sent to all relevant database nodes.

[0061] More specifically, as shown in combination with Figure 1 and Figure 2 the above step S10 includes the following sub-steps:

[0062] S11: Obtain the data request information;

[0063] S12: Write the operation record corresponding to the data request information to a log file;

[0064] S13: Send a pre-commit request to the corresponding node of the database and receive a response;

[0065] S14: When the request responses of all database nodes are received, send a formal request message to the corresponding node of the database.

[0066] Assume that transaction T performs operations on node A and writes the operation records to the log file. Transaction T sends a pre-commit request to node B. Node B performs the operations and writes the operation records to the log file. Node B sends a ready response to transaction T. After transaction T receives the ready responses from all nodes, it sends a formal commit request. Node A and node B complete local commits and synchronize the results to other relevant nodes.

[0067] S20: Import the data acquisition information into the multi-modal fusion evaluation model, and evaluate the operation data information based on the multi-modal fusion evaluation model to obtain the corresponding node evaluation information;

[0068] In this embodiment, the above steps combine multiple models and algorithms such as graph neural networks, rule-based load balancing algorithms, and reinforcement learning algorithms to form a database node selection mechanism for multi-model fusion. For example, when the system load is low, the graph neural network is preferentially used for node selection; when the system load is high, it switches to a rule-based load balancing algorithm, such as the round-robin method or the least-connection method, to quickly disperse data update requests and avoid overloading a single database node.

[0069] Specifically, the above steps import the data acquisition information into the multi-modal fusion evaluation model, and use the load conditions (such as CPU usage rate, memory occupancy rate), storage capacity usage rate, network latency, data update rate, etc. of the corresponding data nodes as the input features of the graph neural network. By comprehensively analyzing these features, the graph neural network can more accurately select the most suitable database node for data operations.

[0070] Assume that the input feature vector of the graph neural network is X, then:

[0071] X = [x1, x2, x3, x4, x5]

[0072] Among them, x1 represents the connection relationship between the channel node and the database node, x2 represents the load condition of the database node, x3 represents the storage capacity usage rate, x4 represents the network latency, and x5 represents the data update frequency.

[0073] In detail, in combination with Figure 1 and Figure 3 as shown, this step includes the following sub-steps:

[0074] S21: Extract data features from the data acquisition information to obtain data feature information;

[0075] In this sub-step, the above data acquisition information is preprocessed, and data features are extracted from the preprocessed data acquisition information based on the feature extraction network to obtain the above data feature information.

[0076] S22: Import the data feature information into the multi-modal fusion evaluation model, which includes a pre-set graph neural network unit, a load balancing unit, and a reinforcement learning unit, to obtain the graph neural network information y GNN , the load balancing information y RLB , and the reinforcement learning information y RL ;

[0077] In the above sub-steps, the multi-modal fusion evaluation model includes a pre-set graph neural network unit, a load balancing unit, and a reinforcement learning unit. The graph neural network unit uses a graph neural network to analyze the data feature information to obtain the graph neural network information y GNN , the load balancing unit analyzes the data feature information based on a rule-based load balancing algorithm to obtain the load balancing information y RLB , and the reinforcement learning unit analyzes the data feature information based on the reinforcement learning algorithm to obtain the reinforcement learning information y RL .

[0078] S23: Output the node evaluation information based on the node selection mechanism:

[0079] y = w1y GNN + w2y RLB + w3y RL

[0080] where w1, w1, w1 are weight coefficients, and y is the node evaluation information.

[0081] In this sub-step, the weight coefficients w1, w1, w1 can be set by the operator according to the usage scenario.

[0082] S30: Based on the node evaluation information, perform data adjustment on the data objects corresponding to the database nodes to balance the operating conditions of the database nodes.

[0083] Combined with Figure 1 and Figure 4 shown, the above step S30 includes the following sub-steps:

[0084] S31: Establish a scheduling cluster, which includes multiple database nodes;

[0085] S32: Based on the node evaluation information, perform data adjustment on the corresponding data objects of multiple database nodes in the scheduling cluster.

[0086] In the above step S30, a distributed scheduling architecture is adopted, and the functions of the scheduling nodes are dispersed to multiple nodes to form a scheduling cluster. Each scheduling node is responsible for processing the operation requests of the energy sales terminals of a part of the channels and regularly communicates with other scheduling nodes in the cluster to synchronize the system status and load information.

[0087] Suppose the set of scheduling nodes is:

[0088] S = {S1, S2, S3,..., S n}

[0089] The selection of the above scheduling nodes can be implemented through the consistent hashing algorithm, and the specific formula is as follows:

[0090] f: T × N → N

[0091] Where T represents the set of tasks, N represents the set of nodes, and f represents the distributed scheduling algorithm.

[0092] Combined with Figure 1 and Figure 5 shown, in another embodiment of the above method, the above further includes:

[0093] S40: Perform version update on the adjusted data object.

[0094] In this step, a version number is imported for each data object. When the above data modification operation is performed, this step automatically records the timestamp and version number of the operation. When the user modifies the same data object multiple times, this step automatically assigns a unique version number to each version and stores different versions of the data object in the database.

[0095] More specifically, combined with Figure 6 shown, the above step S40 includes the following sub-steps:

[0096] S41: Obtain the version information of the adjusted data object and update the version information;

[0097] S42: Incorporate the current timestamp into the current data object.

[0098] More specifically, if the version number of the above data object is V and the timestamp is T, then when the data is modified, the update rules for the version number and timestamp are as follows:

[0099] V new = V old + 1

[0100] T new = current time

[0101] In this step, during the process of adding timestamps to each data object as described above, when the user performs an addition or modification operation, the system automatically records the timestamp of the operation. When operating on the same data object, in this step, the timestamps of each operation are compared, and the operations are executed in chronological order.

[0102] In this method, a caching mechanism is introduced at the nodes of the database to cache frequently accessed data and operation requests. For example, the judgment results of common operation types and operation data identifiers are cached, and when the same request is encountered again, the result can be directly obtained from the cache.

[0103] The method disclosed in the above embodiment of the present invention evaluates the data operation situation through the data on the database nodes. By comprehensively analyzing these features, the multi-modal fusion evaluation model can more accurately select the most suitable database node for data operation. And based on the evaluation result, data adjustment is performed on the data object to balance the operation situation of the database nodes.

[0104] As Figure 7 shown, the second preferred embodiment of the present invention discloses a data processing device 100 for an energy Internet marketing service system. The energy Internet marketing service system includes a data middle platform, multiple energy sales terminals, and multiple distributed databases. The multiple distributed databases are respectively electrically connected to the data middle platform through corresponding database nodes, and the multiple energy sales terminals are respectively electrically connected to the data middle platform through corresponding scheduling nodes. The device 100 includes a data acquisition module 110, a node evaluation module 120, and a data adjustment module 130.

[0105] The above data acquisition module 110 acquires data acquisition information on each of the database nodes;

[0106] Specifically, in this step, the distributed transaction mechanism is adopted to collect data from the database nodes. In a distributed database environment, distributed data collection tools can be used to collect and integrate data from multiple nodes.

[0107] Distributed data collection is a technology that distributes data collection tasks to multiple nodes or devices to improve the efficiency and reliability of data collection. In the database field, distributed data collection can be used to collect data from multiple database nodes or servers for analysis, monitoring, and management.

[0108] Distributed data collection system architecture: Data collection nodes are responsible for collecting data from database nodes. These nodes can be servers, sensors, or other devices connected to the above databases. Data transmission: In this embodiment, the collected data can be transmitted to a central server or data storage through a network.

[0109] In the above-mentioned step during data update, multiple database nodes involved are incorporated into a distributed transaction, and the two-phase commit (2PC) or three-phase commit (3PC) protocol is used to ensure data consistency. For example, when performing data addition or modification operations, the operation records are first written into a log file, and then a pre-commit request is sent to all relevant database nodes.

[0110] More specifically, the above-mentioned data acquisition module 110 specifically includes: obtaining data request information; writing the operation records corresponding to the data request information into a log file; sending a pre-commit request to the corresponding database node and receiving a response; when receiving the request responses from all database nodes, sending a formal request message to the corresponding database node.

[0111] Suppose transaction T performs an operation on node A and writes the operation record into a log file. Transaction T sends a pre-commit request to node B. Node B performs the operation and writes the operation record into a log file. Node B sends a ready response to transaction T. After transaction T receives the ready responses from all nodes, it sends a formal commit request. Node A and node B complete local commits and synchronize the results to other relevant nodes.

[0112] The above-mentioned node evaluation module 120 imports the data acquisition information into a multi-modal fusion evaluation model, and evaluates the operation data information based on the multi-modal fusion evaluation model to obtain corresponding node evaluation information;

[0113] In this embodiment, the above steps combine multiple models and algorithms such as graph neural networks, rule-based load balancing algorithms, and reinforcement learning algorithms to form a database node selection mechanism for multi-model fusion. For example, when the system load is low, the graph neural network is preferentially used for node selection; when the system load is high, it switches to a rule-based load balancing algorithm, such as the round-robin method or the least connection method, to quickly disperse data update requests and avoid overloading a single database node.

[0114] Specifically, the above steps import the data acquisition information into a multi-modal fusion evaluation model, and use the load conditions (such as CPU usage rate, memory occupancy rate), storage capacity utilization rate, network latency, data update rate, etc. of the corresponding data nodes as the input features of the graph neural network. By comprehensively analyzing these features, the graph neural network can more accurately select the most suitable database node for data operations.

[0115] Suppose the input feature vector of the graph neural network is X, then:

[0116] X = [x1, x2, x3, x4, x5]

[0117] Among them, x1 represents the connection relationship between the channel node and the database node, x2 represents the load condition of the database node, x3 represents the storage capacity utilization rate, x4 represents the network latency, and x5 represents the data update frequency.

[0118] Specifically, the above node evaluation module 120 includes a feature extraction unit 121 and a feature evaluation unit 122.

[0119] The feature extraction unit 121 extracts data features from the data acquisition information to obtain data feature information; preprocesses the above data acquisition information, and based on the feature extraction network, extracts features from the preprocessed data acquisition information, and obtains the above data feature information. The feature evaluation unit 122 imports the data feature information into the multi-modal fusion evaluation model, and the multi-modal fusion evaluation model includes a preset graph neural network unit, a load balancing unit, and a reinforcement learning unit, and obtains the graph neural network information y GNN , load balancing information y RLB and reinforcement learning information y RL ; the above multi-modal fusion evaluation model includes a preset graph neural network unit, a load balancing unit, and a reinforcement learning unit. The above graph neural network unit uses a graph neural network to analyze the above data feature information to obtain graph neural network information y GNN , the above load balancing unit analyzes the above data feature information based on a rule-based load balancing algorithm to obtain load balancing information y RLB , the above reinforcement learning unit analyzes the above data feature information based on the above reinforcement learning algorithm to obtain reinforcement learning information y Rl . Based on the node selection mechanism, the node evaluation information is output:

[0120] y = w1y gnn + w2y RlB + w3y RL

[0121] Among them, w1, w1, w1 are weight coefficients, and y is the node evaluation information. The above weight coefficients w1, w1, w1 can be set by the operator according to the usage scenario.

[0122] The above data adjustment module 130 adjusts the data objects corresponding to the database nodes based on the node evaluation information to balance the operation conditions of the database nodes.

[0123] Specifically, the above-mentioned data adjustment module 130 establishes a scheduling cluster, and the scheduling cluster includes multiple database nodes; based on the node evaluation information, data adjustment is performed on the corresponding data objects of the multiple database nodes in the scheduling cluster. The above-mentioned distributed scheduling architecture is adopted, and the functions of the scheduling nodes are dispersed to multiple nodes to form a scheduling cluster. Each scheduling node is responsible for processing the operation requests of the energy sales terminals of a part of the channels, and regularly communicates with other scheduling nodes in the cluster to synchronize the system status and load information.

[0124] Assume that the set of scheduling nodes is:

[0125] S = {S1, S2, S3,..., S n}

[0126] The selection of the above-mentioned scheduling nodes can be implemented through the consistent hashing algorithm, and the specific formula is as follows:

[0127] f: T×N→N

[0128] Among them, T represents the set of tasks, N represents the set of nodes, and f represents the distributed scheduling algorithm.

[0129] The above-mentioned system 100 includes a version update module 140, and the version update module 140 performs version update on the adjusted data objects.

[0130] Import a version number for each data object. When the above-mentioned data modification operation is performed, this step automatically records the operation timestamp and version number. When the user modifies the same data object multiple times, this step automatically assigns a unique version number to each version and stores different versions of the data object in the database.

[0131] More specifically, the version update module 140 obtains the version information of the adjusted data object and updates the version information; incorporates the current timestamp into the current data object.

[0132] More specifically, if the version number of the above-mentioned data object is V and the timestamp is T, then when data modification is performed, the update rules for the version number and timestamp are as follows:

[0133] V new = V old + 1

[0134] T new = current time

[0135] During the process of adding a timestamp to each data object above, when the user performs an add or modify operation, the system automatically records the operation timestamp. When operating on the same data object, this step compares the timestamps of each operation and executes the operations in chronological order.

[0136] Introduce a caching mechanism on the nodes of the database to cache frequently accessed data and operation requests. For example, cache the judgment results of common operation types and operation data identifiers. When the same request is encountered again, the result can be directly obtained from the cache.

[0137] The device disclosed in the above embodiment of the present invention evaluates the data operation conditions on the database nodes. By comprehensively analyzing these features, the multi-modal fusion evaluation model can more accurately select the most suitable database node for data operation. And based on the evaluation result, data adjustment is performed on the data object to balance the operation conditions of the database nodes.

[0138] A computer storage medium, on which a computer program is stored. When the computer program on the computer storage medium is executed by a processor, the method described in any one of the above is implemented.

[0139] The above computer storage medium of the present invention evaluates the data operation conditions on the database nodes by using the above method. By comprehensively analyzing these features, the multi-modal fusion evaluation model can more accurately select the most suitable database node for data operation. And based on the evaluation result, data adjustment is performed on the data object to balance the operation conditions of the database nodes.

[0140] It should be noted that the computer storage medium described above in the present disclosure can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium can also be any computer storage medium other than a computer-readable storage medium, and this computer-readable signal medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer storage medium can be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0141] In some embodiments, the client and the server can communicate using any currently known or future-developed network protocol such as HTTP (Hyper Text Transfer Protocol), and can be interconnected with digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include local area networks ("LANs"), wide area networks ("WANs"), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0142] The above computer storage medium can be included in an electronic device; it can also exist separately without being assembled into the electronic device.

[0143] The above computer storage medium carries one or more programs, and when the one or more programs are executed by the electronic device, it causes the electronic device to...

[0144] The technical features of the above-mentioned embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0145] The above-mentioned embodiments only express several implementation manners of the present invention. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the invention patent. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention patent shall be subject to the appended claims.

Claims

1. A data processing method for an energy Internet marketing service system, characterized in that, The energy Internet marketing service system includes a data center, multiple energy sales terminals, and multiple distributed databases. The multiple distributed databases are electrically connected to the data center through corresponding database nodes, and the multiple energy sales terminals are electrically connected to the data center through corresponding scheduling nodes. The method includes: Obtain the data collection information on each of the database nodes; Import the data collection information into a multi-modal fusion evaluation model, and evaluate the operation data information based on the multi-modal fusion evaluation model to obtain corresponding node evaluation information; Based on the node evaluation information, perform data adjustment on the data objects corresponding to the database nodes to balance the operation of the database nodes.

2. The data processing method of the energy Internet marketing service system according to claim 1, wherein The step of importing the data collection information into a multi-modal fusion evaluation model and evaluating the operation data information based on the multi-modal fusion evaluation model to obtain corresponding node evaluation information includes: Extract data feature information from the data collection information; Import the data feature information into the multi-modal fusion evaluation model, which includes a preset graph neural network unit, a load balancing unit, and a reinforcement learning unit, to obtain the graph neural network information y GNN , load balancing information y RLB , and reinforcement learning information y RL ; Output the node evaluation information based on a node selection mechanism: y = w1y GNN + w2y RLB + w3y RL where w1, w1, w1 are weight coefficients, and y is the node evaluation information.

3. The data processing method of the energy Internet marketing service system according to claim 1, wherein The step of obtaining the data collection information on each of the database nodes includes: Collect data from the database nodes by using a distributed transaction mechanism.

4. The data processing method of the energy Internet marketing service system according to claim 3, characterized in that The step of collecting data from the database nodes by using a distributed transaction mechanism includes: Obtain data request information; Write the operation record corresponding to the data request information into a log file; Initiate a pre-commit request to the corresponding database node and receive a response; When the request responses of all database nodes are received, initiate a formal request message to the corresponding database node.

5. The data processing method of the energy Internet marketing service system according to claim 1, characterized in that, The step of performing data adjustment on the data objects corresponding to the database nodes based on the node evaluation information to balance the operation of the database nodes includes: Establish a scheduling cluster, where the scheduling cluster includes multiple database nodes; Based on the node evaluation information, perform data adjustment on the corresponding data objects of the multiple database nodes in the scheduling cluster.

6. The data processing method of the energy Internet marketing service system according to claim 1, characterized in that The method further includes: Update the version of the adjusted data object.

7. The data processing method of the energy Internet marketing service system according to claim 6, characterized in that The step of updating the version of the adjusted data object includes: Obtain the version information of the adjusted data object and update the version information; Incorporate the current timestamp into the current data object.

8. A data processing device for an energy Internet marketing service system, characterized in that, The energy Internet marketing service system includes a data center, multiple energy sales terminals, and multiple distributed databases. The multiple distributed databases are electrically connected to the data center through corresponding database nodes, and the multiple energy sales terminals are electrically connected to the data center through corresponding scheduling nodes. The device includes: A data collection module for obtaining the data collection information on each of the database nodes; A node evaluation module for importing the data collection information into a multi-modal fusion evaluation model and evaluating the operation data information based on the multi-modal fusion evaluation model to obtain corresponding node evaluation information; A data adjustment module, configured to perform data adjustment on the data object corresponding to the database node based on the node evaluation information, so as to balance the operation conditions of the database node.

9. The data processing method of the energy Internet marketing service system according to claim 8, characterized in that, The node evaluation module includes: A feature extraction unit, configured to perform data feature extraction on the data collection information to obtain data feature information; A feature evaluation unit for importing the data feature information into the multi-modal fusion evaluation model, which includes a preset graph neural network unit, a load balancing unit, and a reinforcement learning unit, to obtain the graph neural network information y GNN , load balancing information y RLB , and reinforcement learning information y RL ; Output the node evaluation information based on the node selection mechanism: y = w1y GNN + w2y RLB + w3y RL Wherein, w1, w1, w1 are weight coefficients, and y is the node evaluation information.

10. A computer storage medium, characterized in that, A computer program is stored on the computer storage medium, and when the computer program on the computer storage medium is executed by a processor, the method according to any one of claims 1-7 is implemented.

Images