Multi-energy comprehensive regulation optimization operation method and system for power system

Abstract

The invention provides a power system multi-energy comprehensive regulation optimization operation method and system, and relates to the technical field of power systems. The method comprises the following steps: firstly, determining multi-energy system data; secondly, mining a multi-energy system semantic vector corresponding to the multi-energy system data by using a first semantic mining network; then, performing semantic fusion on a multi-energy system semantic vector corresponding to the multi-energy system data and a pre-configured multi-energy scheme semantic vector corresponding to each reference multi-energy operation scheme by using a semantic fusion network, and outputting a multi-energy scheme semantic vector corresponding to the multi-energy system data; and finally, performing mapping output on the multi-energy scheme semantic vector corresponding to the multi-energy system data by using the first mapping output network to obtain a target multi-energy operation scheme. On the basis of the content, the problems that in the prior art, the reliability of multi-energy comprehensive adjustment optimization operation is low, and the labor cost is high can be solved.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular to a method and system for optimizing operation of multi-energy integrated regulation of power systems. Background Art

[0002] Traditional energy (such as coal, natural gas, nuclear energy, etc.) can provide stable power output, and renewable energy (wind energy, solar energy, hydropower, etc.) is a volatile and intermittent power source, which can effectively reduce greenhouse gas emissions, but its output is uncertain. Among them, the main goals of multi-energy system regulation and optimization include: minimizing the operating costs of the power system (such as fuel costs, operating costs, etc.); ensuring that the system can provide sufficient power supply to meet load requirements while ensuring the stability and safety of the system; reducing greenhouse gas emissions, promoting the use of green energy, and reducing dependence on traditional fossil energy. However, the inventors have found that in the prior art, the scheme for the comprehensive regulation and optimization of multi-energy in the power system is generally based on manual analysis by relevant personnel, so there are problems of low reliability and high labor costs. Summary of the invention

[0003] In view of this, the purpose of this application is to provide a method and system for multi-energy comprehensive regulation and optimization operation of an electric power system, so as to improve the problems of low reliability and high labor cost of multi-energy comprehensive regulation and optimization operation existing in the prior art.

[0004] To achieve the above objectives, this application adopts the following technical solutions:

[0005] A method for optimizing operation of multi-energy comprehensive regulation in a power system, comprising:

[0006] Determine multi-energy system data, wherein the multi-energy system data at least includes description data of multiple energy subsystems in the target power system and description data of power demand of the power consumption area corresponding to the target power system;

[0007] Using a first semantic mining network, mining a multi-energy system semantic vector corresponding to the multi-energy system data;

[0008] Using a semantic fusion network, semantically fuse the multi-energy system semantic vector corresponding to the multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in a pre-configured reference multi-energy operation solution set, and output the multi-energy solution semantic vector corresponding to the multi-energy system data, wherein the multi-energy solution semantic vector corresponding to the reference multi-energy operation solution is mined using a second semantic mining network;

[0009] The first mapping output network is used to map and output the multi-energy scheme semantic vector corresponding to the multi-energy system data to obtain a target multi-energy operation scheme corresponding to the multi-energy system data, wherein the target multi-energy operation scheme serves as a basis for the power generation operation of the various energy subsystems in the target power system.

[0010] In a preferred option of the present application, in the above-mentioned power system multi-energy comprehensive regulation optimization operation method, the semantic fusion network includes a first number of first connection units and a second number of first semantic fusion units, and the first connection units and the first semantic fusion units are alternately arranged in sequence, and the difference between the first number and the second number is equal to 1;

[0011] The step of using a semantic fusion network to semantically fuse the multi-energy system semantic vector corresponding to the multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in a pre-configured reference multi-energy operation solution set, and outputting the multi-energy solution semantic vector corresponding to the multi-energy system data includes:

[0012] Using the ath first connection unit, the multi-energy system input vector of the ath first connection unit is processed to form the multi-energy system output vector of the ath first connection unit, wherein the multi-energy system input vector of the first first connection unit includes the multi-energy solution semantic vector of the reference multi-energy operation solution, and the multi-energy system output vector of the last first connection unit includes the multi-energy solution semantic vector corresponding to the multi-energy system data;

[0013] Using the ath first semantic fusion unit, the multi-energy system input vector of the ath first semantic fusion unit is processed to form the multi-energy system output vector of the ath first semantic fusion unit, wherein the multi-energy system input vector of the ath first semantic fusion unit includes the multi-energy system output vector of the ath first connection unit and the multi-energy system semantic vector of the multi-energy system data, the multi-energy system input vector of the bth first connection unit includes the multi-energy system output vector of the ath first semantic fusion unit, and the bth first connection unit is the first connection unit after the ath first connection unit.

[0014] In a preferred option of the present application, in the above-mentioned method for optimizing operation of multi-energy integrated regulation in a power system, the first connection unit includes a high-dimensional mapping subunit and a first deep mining subunit;

[0015] The step of using the ath first connection unit to process the multi-energy system input vector of the ath first connection unit to form the multi-energy system output vector of the ath first connection unit includes:

[0016] Using the high-dimensional mapping subunit of the a-th first connection unit, the multi-energy system input vector of the a-th first connection unit is subjected to high-dimensional mapping, and a high-dimensional mapping vector is output;

[0017] The high-dimensional mapping vector is subjected to multi-stage deep mining by utilizing the first deep mining subunit to form a multi-energy system output vector of the a-th first connection unit.

[0018] In a preferred option of the present application, in the above-mentioned power system multi-energy comprehensive regulation and optimization operation method, the first connection unit also includes a second deep mining subunit and a third deep mining subunit;

[0019] The step of using the first deep mining subunit to perform multi-stage deep mining on the high-dimensional mapping vector to form the multi-energy system output vector of the a-th first connection unit includes:

[0020] Using the first deep mining subunit, the second deep mining subunit and the third deep mining subunit, the high-dimensional mapping vector is subjected to multi-stage deep mining to form a multi-energy system output vector of the a-th first connection unit;

[0021] Wherein, in each stage of deep mining, the first deep mining subunit is used to mine the multi-energy system input vector of the first deep mining subunit to form the multi-energy system output vector of the first deep mining subunit; the second deep mining subunit is used to mine the associated multi-energy solution semantic vector of the multi-energy system data to form the multi-energy system output vector of the second deep mining subunit; after the multi-energy system output vector of the first deep mining subunit and the multi-energy system output vector of the second deep mining subunit are aggregated, the third deep mining subunit is used to mine to form the multi-energy system output vector of the third deep mining subunit; and the multi-energy system input vector of the first deep mining subunit in the next stage of deep mining is determined based on the multi-energy system output vector of the third deep mining subunit and the multi-energy system input vector of the first deep mining subunit;

[0022] In the first stage of deep mining, the multi-energy system input vector of the first deep mining subunit is the high-dimensional mapping vector;

[0023] In the last stage of deep mining, the multi-energy system output vector of the third deep mining sub-unit is used as the multi-energy system output vector of the a-th first connection unit.

[0024] In a preferred choice of the present application, in the above-mentioned power system multi-energy comprehensive regulation optimization operation method, the step of using the a-th first semantic fusion unit to process the multi-energy system input vector of the a-th first semantic fusion unit to form the multi-energy system output vector of the a-th first semantic fusion unit includes:

[0025] Utilizing the a-th first semantic fusion unit, analyzing a vector similarity parameter between each local semantic vector in the multi-energy system output vector of the a-th first connection unit and the multi-energy system semantic vector of the multi-energy system data;

[0026] Based on the vector similarity parameters between the multi-energy system semantic vector of the multi-energy system data and the multi-energy scheme semantic vectors of each reference multi-energy operation scheme, the local semantic vector in the multi-energy system output vector of the a-th first connection unit that does not match the pre-configured first vector similarity parameter condition is updated to form the multi-energy system output vector of the a-th first semantic fusion unit.

[0027] In a preferred choice of the present application, in the above-mentioned power system multi-energy comprehensive regulation optimization operation method, the vector similarity parameter between the multi-energy system semantic vector based on the multi-energy system data and the multi-energy scheme semantic vector of each of the reference multi-energy operation schemes is used to update the local semantic vector in the multi-energy system output vector of the a-th first connection unit that does not match the pre-configured first vector similarity parameter condition to form the multi-energy system output vector of the a-th first semantic fusion unit, including:

[0028] Based on the vector similarity parameter between the multi-energy system semantic vector of the multi-energy system data and the multi-energy scheme semantic vector of each of the reference multi-energy operation schemes, the local semantic vector in the multi-energy system output vector of the a-th first connection unit that does not match the pre-configured first vector similarity parameter condition is updated to form an updated multi-energy system output vector;

[0029] Based on the updated multi-energy system output vector and the associated multi-energy solution semantic vector of the multi-energy system data, the multi-energy system output vector of the a-th first semantic fusion unit is formed.

[0030] In the preferred selection of the present application, in the above-mentioned power system multi-energy comprehensive regulation optimization operation method, the power system multi-energy comprehensive regulation optimization operation method also includes a training step of an operation scheme determination network, the operation scheme determination network includes a first semantic mining network, a second semantic mining network, a first mapping output network and a semantic fusion network, and the training step includes:

[0031] Extracting at least one first training data combination, wherein each first training data combination includes a corresponding training multi-energy system data and a multi-energy operation scheme label;

[0032] Using the first semantic mining network, mining the multi-energy system semantic vector corresponding to the training multi-energy system data;

[0033] Using the semantic fusion network, semantically fuse the multi-energy system semantic vector corresponding to the training multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in the reference multi-energy operation solution set, and output the multi-energy solution semantic vector corresponding to the training multi-energy system data;

[0034] Using the first mapping output network, mapping and outputting the multi-energy scheme semantic vector corresponding to the training multi-energy system data, to obtain the training multi-energy operation scheme corresponding to the training multi-energy system data;

[0035] Obtaining a corresponding first training error based on a difference between a multi-energy operation scheme label corresponding to the training multi-energy system data and a training multi-energy operation scheme corresponding to the training multi-energy system data;

[0036] Based on the first training error, the network parameters of the operation scheme determination network are updated to form an updated operation scheme determination network.

[0037] In a preferred option of the present application, in the above-mentioned power system multi-energy comprehensive regulation optimization operation method, the semantic fusion network includes a first number of first connection units and a second number of first semantic fusion units, the first connection units and the first semantic fusion units are alternately arranged in sequence, and the difference between the first number and the second number is equal to 1;

[0038] The step of using the semantic fusion network to semantically fuse the multi-energy system semantic vector corresponding to the training multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in the reference multi-energy operation solution set, and outputting the multi-energy solution semantic vector corresponding to the training multi-energy system data includes:

[0039] Using the ath first connection unit, the multi-energy system input vector of the ath first connection unit is processed to form the multi-energy system output vector of the ath first connection unit, wherein the multi-energy system input vector of the first first connection unit includes the multi-energy solution semantic vector of the reference multi-energy operation solution, and the multi-energy system output vector of the last first connection unit includes the multi-energy solution semantic vector corresponding to the training multi-energy system data;

[0040] Using the ath first semantic fusion unit, the multi-energy system input vector of the ath first semantic fusion unit is processed to form the multi-energy system output vector of the ath first semantic fusion unit, wherein the multi-energy system input vector of the ath first semantic fusion unit includes the multi-energy system output vector of the ath first connection unit and the multi-energy system semantic vector of the training multi-energy system data, and the multi-energy system input vector of the bth first connection unit includes the multi-energy system output vector of the ath first semantic fusion unit.

[0041] In a preferred embodiment of the present application, in the above-mentioned power system multi-energy comprehensive regulation optimization operation method, the step of updating the network parameters of the operation scheme determination network based on the first training error to form an updated operation scheme determination network includes:

[0042] Extracting at least one second training data combination, wherein each second training data combination includes a corresponding actual multi-energy operation scheme and a multi-energy system data label;

[0043] Using the second semantic mining network, mining the multi-energy solution semantic vector corresponding to the actual multi-energy operation solution;

[0044] Using the semantic fusion network, semantic fusion is performed on the multi-energy scheme semantic vector corresponding to the actual multi-energy operation scheme and the multi-energy system semantic vector corresponding to each reference multi-energy system data contained in the reference multi-energy system data set, and the multi-energy system semantic vector corresponding to the actual multi-energy operation scheme is output, wherein the multi-energy system semantic vector of the reference multi-energy system data is formed by mining using the first semantic mining network;

[0045] Utilizing the second mapping output network included in the operation scheme determination network, mapping and outputting the multi-energy system semantic vector corresponding to the actual multi-energy operation scheme, and obtaining the multi-energy system generation data corresponding to the actual multi-energy operation scheme;

[0046] Obtaining a corresponding second training error based on a difference between the multi-energy system data label corresponding to the actual multi-energy operation scheme and the multi-energy system generated data corresponding to the actual multi-energy operation scheme;

[0047] Based on the first training error and the second training error, network parameters of the operation scheme determination network are updated to form an updated operation scheme determination network.

[0048] The present application also provides a power system multi-energy comprehensive regulation and optimization operation system for implementing the above-mentioned method, including:

[0049] A description data determination module, used to determine multi-energy system data, wherein the multi-energy system data includes description data of multiple energy subsystems in the target power system and description data of power demand of the power consumption area corresponding to the target power system;

[0050] A semantic mining module, configured to mine a multi-energy system semantic vector corresponding to the multi-energy system data using a first semantic mining network;

[0051] A semantic fusion module, used for performing semantic fusion on the multi-energy system semantic vector corresponding to the multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in the pre-configured reference multi-energy operation solution set by using a semantic fusion network, and outputting the multi-energy solution semantic vector corresponding to the multi-energy system data, wherein the multi-energy solution semantic vector corresponding to the reference multi-energy operation solution is mined by using a second semantic mining network;

[0052] An operation plan output module is used to use a first mapping output network to map and output the multi-energy plan semantic vector corresponding to the multi-energy system data, so as to obtain a target multi-energy operation plan corresponding to the multi-energy system data, wherein the target multi-energy operation plan serves as a basis for the power generation operation of the various energy subsystems in the target power system.

[0053] The method and system for comprehensive regulation and optimization operation of multi-energy in power systems provided in the present application first determine the multi-energy system data; secondly, utilize a first semantic mining network to mine the multi-energy system semantic vector corresponding to the multi-energy system data; then, utilize a semantic fusion network to semantically fuse the multi-energy system semantic vector corresponding to the multi-energy system data and the multi-energy solution semantic vector corresponding to each pre-configured reference multi-energy operation solution, and output the multi-energy solution semantic vector corresponding to the multi-energy system data; finally, utilize a first mapping output network to map and output the multi-energy solution semantic vector corresponding to the multi-energy system data, and obtain a target multi-energy operation solution. Based on the above content, after the multi-energy system semantic vector corresponding to the multi-energy system data is mined, the multi-energy solution semantic vector corresponding to the reference multi-energy operation plan will be semantically fused, so that the multi-energy solution semantic vector corresponding to the formed multi-energy system data can not only characterize the semantic information of the multi-energy system data, but also carry the semantic information of the semantic space where the operation plan is located. In this way, when the target multi-energy operation plan is generated based on the multi-energy solution semantic vector corresponding to the multi-energy system data, the reliability of the generated target multi-energy operation plan can be higher because it carries the semantic information of the semantic space where the operation plan is located, thereby improving the problems of low reliability and high labor cost of multi-energy comprehensive regulation and optimization operation in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings.

[0055] Figure 1 A structural block diagram of an electronic device provided in an embodiment of the present application.

[0056] Figure 2 A flow chart of a method for comprehensive regulation and optimization of multi-energy sources in an electric power system provided in an embodiment of the present application.

[0057] Figure 3 A schematic diagram of a semantic fusion network provided in an embodiment of the present application.

[0058] Figure 4 A block diagram of a multi-energy integrated regulation and optimization operation system for a power system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0060] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0061] Embodiment 1

[0062] like Figure 1 As shown, an embodiment of the present application provides an electronic device, wherein the electronic device may include a memory, a processor, and a multi-energy comprehensive regulation and optimization operation system for a power system.

[0063] In detail, the memory and the processor are electrically connected directly or indirectly to realize data transmission or interaction. For example, the memory and the processor can be electrically connected through one or more communication buses or signal lines. The power system multi-energy comprehensive regulation and optimization operation system includes at least one software function module stored in the memory in the form of software or firmware. The processor is used to execute an executable computer program stored in the memory, for example, the software function modules and computer programs included in the power system multi-energy comprehensive regulation and optimization operation system, so as to realize the power system multi-energy comprehensive regulation and optimization operation method provided in the embodiment of the present application.

[0064] Optionally, the memory can be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), etc.

[0065] Furthermore, the processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a system on chip (SoC), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0066] Understandably, Figure 1 The structure shown is for illustration only, and the electronic device may also include Figure 1 More or fewer components as shown, or with Figure 1 The different configurations shown, for example, may also include a communication unit for exchanging information with other devices.

[0067] Combination Figure 2 The embodiment of the present application also provides a method for optimizing the operation of a power system with multiple energy sources integrated regulation, which can be applied to the above-mentioned electronic device. The method steps defined in the process related to the method for optimizing the operation of a power system with multiple energy sources integrated regulation can be implemented by the electronic device. Figure 2 The specific process shown is explained in detail.

[0068] Step S110, determining multi-energy system data.

[0069] In an embodiment of the present application, the electronic device can determine multi-energy system data. Among them, the multi-energy system data at least includes description data (such as the output power range of the unit, etc.) of multiple energy subsystems in the target power system (such as hydroelectric subsystems, thermal power subsystems, etc. with stable output, and wind power subsystems, solar power subsystems, etc. with unstable output) and description data of the power demand of the power consumption area corresponding to the target power system (such as power consumption, power stability demand, etc.). For example, part of the data in the multi-energy system data may be "the power output range of the hydroelectric subsystem is 50MW to 200MW, the power output range of the thermal power subsystem is 100MW to 500MW, the power output range of the wind power subsystem is 0MW to 150MW, the power output range of the solar power subsystem is 0 to 50MW, the power demand may reach 500MW, and a stable power supply is required". In addition, it may also include information such as the power generation cost of each energy subsystem. In a more detailed example, it also includes the water level of the reservoir, the water flow rate (m 3 / s) and water flow rate (m / s), fossil fuel reserves, estimated wind speed, estimated solar radiation intensity and duration, etc.

[0070] Step S120: using a first semantic mining network to mine a multi-energy system semantic vector corresponding to the multi-energy system data.

[0071] In an embodiment of the present application, after determining the multi-energy system data, the electronic device can use the first semantic mining network to mine the multi-energy system semantic vector corresponding to the multi-energy system data, that is, to mine the semantic features in the multi-energy system data and express them as vectors. Specifically, the first semantic mining network may include a word embedding network, which is used to perform word embedding processing on the multi-energy system data to form a corresponding multi-energy system semantic vector. For example, for "the power output range of the hydroelectric power generation electronic system is 50MW to 200MW", the formed vector is:

[0072] "hydraulicity": [0.32, -0.12, 0.45, 0.27, -0.14, ...];

[0073] "Generation": [0.29, 0.15, -0.34, 0.48, -0.12, ...];

[0074] "subsystem": [0.12, -0.26, 0.54, -0.17, 0.13, ...];

[0075] "power": [0.40, 0.07, 0.22, 0.13, -0.20, ...];

[0076] "output": [0.33, -0.12, -0.46, 0.08, 0.24, ...];

[0077] "interval": [0.25, 0.09, 0.17, -0.35, 0.17, ...];

[0078] "50MW": [0.15, 0.22, -0.13, 0.31, 0.54,...];

[0079] "to": [0.07, -0.02, 0.01, 0.23, -0.25, ...];

[0080] "200MW": [0.30, 0.17, 0.08, 0.40, 0.46, ...].

[0081] In addition, after obtaining the word embedding vector, multiple word embedding vectors can be spliced ​​together, and then the obtained spliced ​​vector is processed through the self-attention network and the feedforward neural network to obtain the final multi-energy system semantic vector.

[0082] Step S130, using a semantic fusion network, semantically fuses the multi-energy system semantic vector corresponding to the multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in a pre-configured reference multi-energy operation solution set, and outputs the multi-energy solution semantic vector corresponding to the multi-energy system data.

[0083] In an embodiment of the present application, after mining the semantic vector of the multi-energy system, the electronic device can use a semantic fusion network to semantically fuse the multi-energy system semantic vector corresponding to the multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in the pre-configured reference multi-energy operation solution set, and output the multi-energy solution semantic vector corresponding to the multi-energy system data, that is, the multi-energy solution semantic vector corresponding to the multi-energy system data carries the semantic information represented by the multi-energy system semantic vector corresponding to the multi-energy system data, and carries the semantic information represented by the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution. Among them, the multi-energy solution semantic vector corresponding to the reference multi-energy operation solution is mined using the second semantic mining network. In addition, the reference multi-energy operation solution can be the historical operation solution of the target power system or other power systems, which can be a whole or a part, such as the operation solution of a certain energy subsystem, as long as the semantic information of the operation solution semantic space can be integrated, so as to facilitate the reliability of the subsequent mapping output.

[0084] Step S140: Mapping and outputting the multi-energy scheme semantic vector corresponding to the multi-energy system data using the first mapping output network to obtain a target multi-energy operation scheme corresponding to the multi-energy system data.

[0085] In an embodiment of the present application, after forming the multi-energy scheme semantic vector corresponding to the multi-energy system data, the electronic device can use the first mapping output network to map and output the multi-energy scheme semantic vector corresponding to the multi-energy system data, and obtain the target multi-energy operation scheme corresponding to the multi-energy system data. Among them, the target multi-energy operation scheme serves as the basis for the power generation operation of the various energy subsystems in the target power system, such as the excitation current of the generator in the hydroelectric power generation subsystem is 400A and the excitation voltage is 200V, and the excitation current of the generator in the thermal power generation subsystem is 500A and the excitation voltage is 250V. Exemplarily, the first mapping output network can be a decoder neural network, so that each word in the target multi-energy operation scheme corresponding to the multi-energy system data can be generated in sequence.

[0086] Based on the above content, after the multi-energy system semantic vector corresponding to the multi-energy system data is mined, the multi-energy solution semantic vector corresponding to the reference multi-energy operation plan will be semantically fused, so that the multi-energy solution semantic vector corresponding to the formed multi-energy system data can not only characterize the semantic information of the multi-energy system data, but also carry the semantic information of the semantic space where the operation plan is located. In this way, when the target multi-energy operation plan is generated based on the multi-energy solution semantic vector corresponding to the multi-energy system data, the reliability of the generated target multi-energy operation plan can be higher because it carries the semantic information of the semantic space where the operation plan is located, thereby improving the problems of low reliability and high labor cost of multi-energy comprehensive regulation and optimization operation in the prior art.

[0087] It should be noted that for step S120 and step S140, the semantic mining and mapping output can refer to the relevant existing technologies, which will not be described one by one here. In the embodiments of the present application (as described later), the semantic fusion in step S130 will be explained in detail.

[0088] For example, in an alternative implementation, in order to fully integrate the multi-energy system semantic vector corresponding to the multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in the pre-configured reference multi-energy operation solution set, and to ensure that the obtained multi-energy solution semantic vector corresponding to the multi-energy system data has better semantic representation ability, combined with Figure 3 , the semantic fusion network includes a first number of first connection units and a second number of first semantic fusion units, and the first connection units and the first semantic fusion units are arranged alternately in sequence, and the difference between the first number and the second number is equal to 1. The above-mentioned step S130 can further include step S131 and step S132, and the specific contents are described as follows.

[0089] Step S131: using the ath first connection unit, processing the multi-energy system input vector of the ath first connection unit to form the multi-energy system output vector of the ath first connection unit.

[0090] In an embodiment of the present application, the ath first connection unit can be used to process the multi-energy system input vector of the ath first connection unit to form the multi-energy system output vector of the ath first connection unit. The multi-energy system input vector of the first first connection unit includes the multi-energy solution semantic vector of the reference multi-energy operation solution, and the multi-energy system output vector of the last first connection unit includes the multi-energy solution semantic vector corresponding to the multi-energy system data. In other words, the multi-energy solution semantic vector of the reference multi-energy operation solution serves as the input of the first first connection unit, and the multi-energy solution semantic vector corresponding to the multi-energy system data serves as the output of the last first connection unit.

[0091] Step S132: using the a-th first semantic fusion unit, processing the multi-energy system input vector of the a-th first semantic fusion unit to form the multi-energy system output vector of the a-th first semantic fusion unit.

[0092] In an embodiment of the present application, after forming the multi-energy system output vector of the a-th first connection unit, the a-th first semantic fusion unit can be used to process the multi-energy system input vector of the a-th first semantic fusion unit to form the multi-energy system output vector of the a-th first semantic fusion unit. Wherein, the multi-energy system input vector of the a-th first semantic fusion unit includes the multi-energy system output vector of the a-th first connection unit and the multi-energy system semantic vector of the multi-energy system data, the multi-energy system input vector of the b-th first connection unit includes the multi-energy system output vector of the a-th first semantic fusion unit, and the b-th first connection unit is the first connection unit after the a-th first connection unit. In this way, the arrangement relationship of the first first connection unit, the first first semantic fusion unit, the second first connection unit, the second first semantic fusion unit, ..., the second to last first connection unit, the last first semantic fusion unit, and the last first connection unit can be formed.

[0093] It can be understood that in the above-mentioned step S131, the specific method of processing performed by the first connection unit is not limited and can be selected according to actual needs. For example, in an alternative embodiment, in order to make the semantic representation ability of the multi-energy system output vector formed by the first connection unit better, the first connection unit may include a high-dimensional mapping subunit and a first deep mining subunit. Based on this, the above-mentioned step S131 may further include step S131a and step S131b. The specific content of each step is described as follows.

[0094] Step S131a, using the high-dimensional mapping subunit of the a-th first connection unit, performs high-dimensional mapping on the multi-energy system input vector of the a-th first connection unit, and outputs a high-dimensional mapping vector.

[0095] In an embodiment of the present application, the high-dimensional mapping subunit of the a-th first connection unit can be used to perform high-dimensional mapping on the multi-energy system input vector of the a-th first connection unit, and output a high-dimensional mapping vector. For example, the data of m (equal to the number of reference multi-energy operation schemes) × 2560 can be adjusted to m × 160 × 16, and the specific dimensions can be configured according to actual needs, such as making the multi-energy scheme semantic vector of the reference multi-energy operation scheme match the multi-energy scheme semantic vector corresponding to the multi-energy system data.

[0096] Step S131b: Utilize the first deep mining subunit to perform multi-stage deep mining on the high-dimensional mapping vector to form a multi-energy system output vector of the a-th first connection unit.

[0097] In an embodiment of the present application, after obtaining a high-dimensional mapping vector, the first deep mining sub-unit can be used to perform multi-stage deep mining on the high-dimensional mapping vector to form a multi-energy system output vector of the a-th first connection unit. In this way, the semantic information of the multi-energy system output vector can be enriched, thereby improving the reliability of subsequent analysis.

[0098] It can be understood that, in the above step S131b, the specific method of performing multi-stage deep mining is not limited and can be selected according to actual needs. For example, in an alternative implementation, in order to improve the reliability of deep mining, the first connection unit also includes a second deep mining subunit and a third deep mining subunit. Based on this, the above step S131b can further include the following specific implementation contents:

[0099] The first deep mining subunit, the second deep mining subunit and the third deep mining subunit may be used to perform multi-stage deep mining on the high-dimensional mapping vector to form a multi-energy system output vector of the a-th first connection unit;

[0100] Among them, in each stage of deep mining, the first deep mining subunit is used to mine the multi-energy system input vector of the first deep mining subunit (including self-attention processing, nonlinear activation processing and full connection processing) to form the multi-energy system output vector of the first deep mining subunit; the second deep mining subunit is used to mine the associated multi-energy scheme semantic vector of the multi-energy system data (for example, obtained by semantic mining the operation scheme of the similar power system of the target power system through the second semantic mining network) (including self-attention processing, nonlinear activation processing and full connection processing) to form the multi-energy system output vector of the second deep mining subunit; after the multi-energy system output vector of the first deep mining subunit and the multi-energy system output vector of the second deep mining subunit are aggregated (for example, the multi-energy system output vector of the second deep mining subunit can be aggregated with the multi-energy system output vector of the first deep mining subunit The multi-energy system output vector of the multi-energy system of the mining subunit is multiplied by the transposed vector to obtain the corresponding similarity parameter, and the similarity parameter is multiplied by the multi-energy system output vector of the first deep mining subunit to obtain the result of vector aggregation), and the third deep mining subunit is used to perform mining (including self-attention processing, nonlinear activation processing and full connection processing) to form the multi-energy system output vector of the third deep mining subunit; and the multi-energy system input vector of the first deep mining subunit in the subsequent stage of deep mining is determined based on the multi-energy system output vector of the third deep mining subunit and the multi-energy system input vector of the first deep mining subunit (for example, the multi-energy system output vector of the third deep mining subunit and the multi-energy system input vector of the first deep mining subunit can be added as the multi-energy system input vector of the first deep mining subunit in the subsequent stage of deep mining);

[0101] In addition, in the first stage of deep mining, the multi-energy system input vector of the first deep mining sub-unit is the high-dimensional mapping vector; and, in the last stage of deep mining, the multi-energy system output vector of the third deep mining sub-unit is used as the multi-energy system output vector of the ath first connection unit; the specific number of multiple stages is also not limited and can be selected according to actual needs, such as 2, 3, 4, 5, 6, etc.

[0102] It can be understood that, in the above-mentioned step S132, the specific method of processing performed by the first semantic fusion unit is not limited and can be selected according to actual needs. For example, in an alternative embodiment, in order to ensure the reliability of the output vector of the formed multi-energy system, the above-mentioned step S132 can further include step S132a and step S132b, the specific contents of which are as follows.

[0103] Step S132a, using the a-th first semantic fusion unit, analyze the vector similarity parameters between each local semantic vector in the multi-energy system output vector of the a-th first connection unit and the multi-energy system semantic vector of the multi-energy system data.

[0104] In an embodiment of the present application, the ath first semantic fusion unit can be used to analyze the vector similarity parameters between each local semantic vector in the multi-energy system output vector of the ath first connection unit and the multi-energy system semantic vector of the multi-energy system data. Exemplarily, each local semantic vector in the multi-energy system output vector of the first connection unit corresponds one-to-one to each reference multi-energy operation scheme, so that the cosine similarity between each local semantic vector and the multi-energy system semantic vector of the multi-energy system data can be calculated respectively.

[0105] Step S132b, based on the vector similarity parameters between the multi-energy system semantic vector of the multi-energy system data and the multi-energy scheme semantic vectors of each reference multi-energy operation scheme, the local semantic vector in the multi-energy system output vector of the a-th first connection unit that does not match the pre-configured first vector similarity parameter condition is updated to form the multi-energy system output vector of the a-th first semantic fusion unit.

[0106] In an embodiment of the present application, after obtaining the vector similarity parameters, the local semantic vector in the multi-energy system output vector of the a-th first connection unit that does not match the pre-configured first vector similarity parameter condition can be updated based on the vector similarity parameters between the multi-energy system semantic vector of the multi-energy system data and the multi-energy scheme semantic vector of each reference multi-energy operation scheme (such as the cosine similarity between vectors) to form the multi-energy system output vector of the a-th first semantic fusion unit, so as to improve the reliability of the multi-energy system output vector.

[0107] It can be understood that, in the above step S132b, the specific method of updating the local semantic vector is not limited and can be selected according to actual needs. For example, in an alternative implementation, in order to further improve the reliability of the output vector of the multi-energy system, the above step S132b can further include the following specific implementation contents:

[0108] First, based on the vector similarity parameters between the multi-energy system semantic vector of the multi-energy system data and the multi-energy scheme semantic vectors of each of the reference multi-energy operation schemes, the local semantic vectors in the multi-energy system output vector of the a-th first connection unit that do not match the pre-configured first vector similarity parameter condition are updated to form an updated multi-energy system output vector; exemplarily, the vector similarity parameters between each local semantic vector in the multi-energy system output vector of the a-th first connection unit and the multi-energy system semantic vector of the multi-energy system data are analyzed, such as obtaining the vector similarity parameters between the local semantic vector 1 corresponding to the reference multi-energy operation scheme 1 and the multi-energy system of the multi-energy system data. The vector similarity parameter 1 between the local semantic vector 2 corresponding to the multi-energy operation plan 2 and the multi-energy system semantic vector of the multi-energy system data is used as the reference; if the vector similarity parameter 1 is greater than the threshold, it is considered to match the pre-configured first vector similarity parameter condition; if the vector similarity parameter 2 is not greater than the threshold, it is considered to not match the pre-configured first vector similarity parameter condition. In this way, it can be considered that the reliability of the local semantic vector 2 is not high. Therefore, it can be directly replaced with a multi-energy plan semantic vector that refers to the multi-energy operation plan and has the largest vector similarity parameter with the multi-energy system semantic vector of the multi-energy system data, thereby forming an updated multi-energy system output vector.

[0109] Secondly, based on the updated multi-energy system output vector and the associated multi-energy scheme semantic vector of the multi-energy system data (such as camera operation or weighted mean calculation, etc.), the multi-energy system output vector of the ath first semantic fusion unit is formed.

[0110] It should be noted that in order to ensure that the above-mentioned networks can have reliable processing capabilities, the power system multi-energy comprehensive regulation and optimization operation method also includes a training step of an operation plan determination network, and the operation plan determination network includes the first semantic mining network, the second semantic mining network, the first mapping output network and the semantic fusion network. In detail, the training step of the operation plan determination network may include step S151, step S152, step S153, step S154, step S155 and step S156. The specific content of each step is described as follows.

[0111] Step S151: extract at least one first training data combination.

[0112] In the embodiment of the present application, at least one first training data combination can be extracted, wherein each first training data combination includes a corresponding training multi-energy system data (such as multi-energy system data) and a multi-energy operation scheme label (ie, the actual operation scheme).

[0113] Step S152: using the first semantic mining network, mining the multi-energy system semantic vector corresponding to the training multi-energy system data, as in step S120.

[0114] Step S153, using the semantic fusion network, semantically fuse the multi-energy system semantic vector corresponding to the training multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in the reference multi-energy operation solution set, and output the multi-energy solution semantic vector corresponding to the training multi-energy system data, as in step S130.

[0115] Step S154, using the first mapping output network, maps and outputs the multi-energy solution semantic vector corresponding to the training multi-energy system data to obtain the training multi-energy operation solution (i.e., the generated operation solution) corresponding to the training multi-energy system data, as in step S140.

[0116] Step S155, based on the difference between the multi-energy operation scheme label corresponding to the training multi-energy system data and the training multi-energy operation scheme corresponding to the training multi-energy system data, obtain the corresponding first training error (the specific error calculation method can be selected according to needs).

[0117] Step S156: Based on the first training error, the network parameters of the operation scheme determination network are updated to form an updated operation scheme determination network. For example, the network parameters of the operation scheme determination network can be updated in a direction of reducing the first training error until the error converges or the number of updates reaches a threshold, thereby obtaining an updated operation scheme determination network.

[0118] It can be understood that, in an alternative implementation, the semantic fusion network includes a first number of first connection units and a second number of first semantic fusion units, the first connection units and the first semantic fusion units are alternately arranged in sequence, and the difference between the first number and the second number is equal to 1, as described above, based on this, the above step S153 can further include the following implementation content:

[0119] First, the ath first connection unit can be used to process the multi-energy system input vector of the ath first connection unit to form the multi-energy system output vector of the ath first connection unit, wherein the multi-energy system input vector of the first first connection unit includes the multi-energy solution semantic vector of the reference multi-energy operation solution, and the multi-energy system output vector of the last first connection unit includes the multi-energy solution semantic vector corresponding to the training multi-energy system data;

[0120] Secondly, the multi-energy system input vector of the a-th first semantic fusion unit is processed using the a-th first semantic fusion unit to form a multi-energy system output vector of the a-th first semantic fusion unit, wherein the multi-energy system input vector of the a-th first semantic fusion unit includes the multi-energy system output vector of the a-th first connection unit and the multi-energy system semantic vector of the training multi-energy system data, and the multi-energy system input vector of the b-th first connection unit includes the multi-energy system output vector of the a-th first semantic fusion unit.

[0121] It is understandable that, in the above step S156, the specific method of updating based on the first training error is not limited and can be selected according to actual needs. For example, in an alternative implementation, in order to improve the reliability of training, the above step S156 can further include the following implementation content:

[0122] First, at least one second training data combination can be extracted, wherein each second training data combination includes a corresponding actual multi-energy operation scheme (i.e., an actual operation scheme) and a multi-energy system data label;

[0123] Secondly, the second semantic mining network can be used to mine the multi-energy solution semantic vector corresponding to the actual multi-energy operation solution;

[0124] Then, the semantic fusion network can be used to perform semantic fusion on the multi-energy scheme semantic vector corresponding to the actual multi-energy operation scheme and the multi-energy system semantic vector corresponding to each reference multi-energy system data contained in the reference multi-energy system data set, and output the multi-energy system semantic vector corresponding to the actual multi-energy operation scheme, wherein the multi-energy system semantic vector of the reference multi-energy system data (which may be local data of a certain energy subsystem or description data of electricity demand, etc.) is formed by mining using the first semantic mining network;

[0125] Afterwards, the second mapping output network included in the operation scheme determination network can be used to map and output the multi-energy system semantic vector corresponding to the actual multi-energy operation scheme to obtain the multi-energy system generation data corresponding to the actual multi-energy operation scheme;

[0126] Further, based on the difference between the multi-energy system data label corresponding to the actual multi-energy operation scheme and the multi-energy system generated data corresponding to the actual multi-energy operation scheme, a corresponding second training error is obtained (the specific error calculation method can be selected according to actual needs);

[0127] Finally, based on the first training error and the second training error, the network parameters of the operation scheme determination network can be updated to form an updated operation scheme determination network. For example, the first training error and the second training error can be summed to obtain a total error, and then the network parameters can be updated in a direction to reduce the total error. In addition, the network parameters can also be updated based on the first training error and the second training error, respectively.

[0128] Combination Figure 4 The embodiment of the present application also provides a power system multi-energy comprehensive regulation and optimization operation system that can be applied to the above electronic device. The power system multi-energy comprehensive regulation and optimization operation system can include a description data determination module, a semantic mining module, a semantic fusion module and an operation plan output module.

[0129] The description data determination module is used to determine the multi-energy system data, wherein the multi-energy system data includes description data of multiple energy subsystems in the target power system and description data of the power demand of the power consumption area corresponding to the target power system. Figure 2 As shown in step S110, the relevant contents of the description data determination module can refer to the above description of step S110.

[0130] The semantic mining module is used to mine the multi-energy system semantic vector corresponding to the multi-energy system data using the first semantic mining network. In the embodiment of the present application, the semantic mining module can be used to perform Figure 2 As shown in step S120, the relevant contents of the semantic mining module can refer to the description of step S120 in the previous text.

[0131] The semantic fusion module is used to use the semantic fusion network to perform semantic fusion on the multi-energy system semantic vector corresponding to the multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in the pre-configured reference multi-energy operation solution set, and output the multi-energy solution semantic vector corresponding to the multi-energy system data, wherein the multi-energy solution semantic vector corresponding to the reference multi-energy operation solution is mined using the second semantic mining network. In the embodiment of the present application, the semantic fusion module can be used to perform Figure 2 As shown in step S130, the relevant contents of the semantic fusion module can refer to the above description of step S130.

[0132] The operation plan output module is used to use the first mapping output network to map and output the multi-energy plan semantic vector corresponding to the multi-energy system data to obtain the target multi-energy operation plan corresponding to the multi-energy system data, wherein the target multi-energy operation plan is used as the basis for the power generation operation of the various energy subsystems in the target power system. In the embodiment of the present application, the operation plan output module can be used to execute Figure 2 As shown in step S140, the relevant contents of the operation plan output module can refer to the above description of step S140.

[0133] To summarize, the method and system for comprehensive regulation and optimization operation of multi-energy in power systems provided in the present application first determine the multi-energy system data; secondly, use the first semantic mining network to mine the multi-energy system semantic vector corresponding to the multi-energy system data; then, use the semantic fusion network to semantically fuse the multi-energy system semantic vector corresponding to the multi-energy system data and the multi-energy scheme semantic vector corresponding to each pre-configured reference multi-energy operation scheme, and output the multi-energy scheme semantic vector corresponding to the multi-energy system data; finally, use the first mapping output network to map and output the multi-energy scheme semantic vector corresponding to the multi-energy system data to obtain the target multi-energy operation scheme. Based on the above content, after the multi-energy system semantic vector corresponding to the multi-energy system data is mined, the multi-energy solution semantic vector corresponding to the reference multi-energy operation plan will be semantically fused, so that the multi-energy solution semantic vector corresponding to the formed multi-energy system data can not only characterize the semantic information of the multi-energy system data, but also carry the semantic information of the semantic space where the operation plan is located. In this way, when the target multi-energy operation plan is generated based on the multi-energy solution semantic vector corresponding to the multi-energy system data, the reliability of the generated target multi-energy operation plan can be higher because it carries the semantic information of the semantic space where the operation plan is located, thereby improving the problems of low reliability and high labor cost of multi-energy comprehensive regulation and optimization operation in the prior art.

[0134] In several embodiments provided in the embodiments of the present application, it should be understood that the disclosed device and method can also be implemented in other ways. The device and method embodiments described above are merely schematic, for example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the device, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0135] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0136] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, an electronic device, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk. It should be noted that, in this article, the term "include", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such a process, method, article or device. Without more constraints, an element defined by the phrase "comprising a..." does not exclude the existence of other identical elements in the process, method, article or apparatus comprising the element.

[0137] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for optimizing the operation of multi-energy comprehensive regulation in a power system, characterized in that: include: Determine multi-energy system data, wherein the multi-energy system data at least includes description data of multiple energy subsystems in the target power system and description data of power demand of the power consumption area corresponding to the target power system; Using a first semantic mining network, mining a multi-energy system semantic vector corresponding to the multi-energy system data; Using a semantic fusion network, semantically fuse the multi-energy system semantic vector corresponding to the multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in a pre-configured reference multi-energy operation solution set, and output the multi-energy solution semantic vector corresponding to the multi-energy system data, wherein the multi-energy solution semantic vector corresponding to the reference multi-energy operation solution is mined using a second semantic mining network; The first mapping output network is used to map and output the multi-energy scheme semantic vector corresponding to the multi-energy system data to obtain a target multi-energy operation scheme corresponding to the multi-energy system data, wherein the target multi-energy operation scheme serves as a basis for the power generation operation of the various energy subsystems in the target power system.

2. The method for optimizing operation of multi-energy comprehensive regulation in a power system according to claim 1 is characterized in that: The semantic fusion network includes a first number of first connection units and a second number of first semantic fusion units, and the first connection units and the first semantic fusion units are alternately arranged in sequence, and the difference between the first number and the second number is equal to 1; The step of using a semantic fusion network to semantically fuse the multi-energy system semantic vector corresponding to the multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in a pre-configured reference multi-energy operation solution set, and outputting the multi-energy solution semantic vector corresponding to the multi-energy system data includes: Using the ath first connection unit, the multi-energy system input vector of the ath first connection unit is processed to form the multi-energy system output vector of the ath first connection unit, wherein the multi-energy system input vector of the first first connection unit includes the multi-energy solution semantic vector of the reference multi-energy operation solution, and the multi-energy system output vector of the last first connection unit includes the multi-energy solution semantic vector corresponding to the multi-energy system data; Using the ath first semantic fusion unit, the multi-energy system input vector of the ath first semantic fusion unit is processed to form the multi-energy system output vector of the ath first semantic fusion unit, wherein the multi-energy system input vector of the ath first semantic fusion unit includes the multi-energy system output vector of the ath first connection unit and the multi-energy system semantic vector of the multi-energy system data, the multi-energy system input vector of the bth first connection unit includes the multi-energy system output vector of the ath first semantic fusion unit, and the bth first connection unit is the first connection unit after the ath first connection unit.

3. The method for optimizing operation of multi-energy comprehensive regulation in power system according to claim 2 is characterized in that: The first connection unit includes a high-dimensional mapping subunit and a first deep mining subunit; The step of using the ath first connection unit to process the multi-energy system input vector of the ath first connection unit to form the multi-energy system output vector of the ath first connection unit includes: Using the high-dimensional mapping subunit of the a-th first connection unit, the multi-energy system input vector of the a-th first connection unit is subjected to high-dimensional mapping, and a high-dimensional mapping vector is output; The high-dimensional mapping vector is subjected to multi-stage deep mining by utilizing the first deep mining subunit to form a multi-energy system output vector of the a-th first connection unit.

4. The method for optimizing operation of multi-energy comprehensive regulation in power system according to claim 3 is characterized in that: The first connection unit further includes a second depth digging subunit and a third depth digging subunit; The step of using the first deep mining subunit to perform multi-stage deep mining on the high-dimensional mapping vector to form the multi-energy system output vector of the a-th first connection unit includes: Using the first deep mining subunit, the second deep mining subunit and the third deep mining subunit, the high-dimensional mapping vector is subjected to multi-stage deep mining to form a multi-energy system output vector of the a-th first connection unit; Wherein, in each stage of deep mining, the first deep mining subunit is used to mine the multi-energy system input vector of the first deep mining subunit to form the multi-energy system output vector of the first deep mining subunit; the second deep mining subunit is used to mine the associated multi-energy solution semantic vector of the multi-energy system data to form the multi-energy system output vector of the second deep mining subunit; after the multi-energy system output vector of the first deep mining subunit and the multi-energy system output vector of the second deep mining subunit are aggregated, the third deep mining subunit is used to mine to form the multi-energy system output vector of the third deep mining subunit; and the multi-energy system input vector of the first deep mining subunit in the next stage of deep mining is determined based on the multi-energy system output vector of the third deep mining subunit and the multi-energy system input vector of the first deep mining subunit; In the first stage of deep mining, the multi-energy system input vector of the first deep mining subunit is the high-dimensional mapping vector; In the last stage of deep mining, the multi-energy system output vector of the third deep mining sub-unit is used as the multi-energy system output vector of the a-th first connection unit.

5. The method for optimizing operation of multi-energy comprehensive regulation in power system according to claim 2, characterized in that: The step of using the ath first semantic fusion unit to process the multi-energy system input vector of the ath first semantic fusion unit to form the multi-energy system output vector of the ath first semantic fusion unit includes: Utilizing the a-th first semantic fusion unit, analyzing a vector similarity parameter between each local semantic vector in the multi-energy system output vector of the a-th first connection unit and the multi-energy system semantic vector of the multi-energy system data; Based on the vector similarity parameters between the multi-energy system semantic vector of the multi-energy system data and the multi-energy scheme semantic vectors of each reference multi-energy operation scheme, the local semantic vector in the multi-energy system output vector of the a-th first connection unit that does not match the pre-configured first vector similarity parameter condition is updated to form the multi-energy system output vector of the a-th first semantic fusion unit.

6. The method for optimizing operation of multi-energy comprehensive regulation in power system according to claim 5 is characterized in that: The step of updating the local semantic vector in the multi-energy system output vector of the a-th first connection unit that does not match the pre-configured first vector similarity parameter condition based on the vector similarity parameter between the multi-energy system semantic vector of the multi-energy system data and the multi-energy scheme semantic vector of each of the reference multi-energy operation schemes to form the multi-energy system output vector of the a-th first semantic fusion unit comprises: Based on the vector similarity parameter between the multi-energy system semantic vector of the multi-energy system data and the multi-energy scheme semantic vector of each of the reference multi-energy operation schemes, the local semantic vector in the multi-energy system output vector of the a-th first connection unit that does not match the pre-configured first vector similarity parameter condition is updated to form an updated multi-energy system output vector; Based on the updated multi-energy system output vector and the associated multi-energy solution semantic vector of the multi-energy system data, the multi-energy system output vector of the a-th first semantic fusion unit is formed.

7. The method for optimizing operation of multi-energy comprehensive regulation in a power system according to any one of claims 1 to 6, characterized in that: The method for optimizing operation of multi-energy comprehensive regulation of a power system further comprises a training step of an operation scheme determination network, wherein the operation scheme determination network comprises a first semantic mining network, a second semantic mining network, a first mapping output network and a semantic fusion network, and the training step comprises: Extracting at least one first training data combination, wherein each first training data combination includes a corresponding training multi-energy system data and a multi-energy operation scheme label; Using the first semantic mining network, mining the multi-energy system semantic vector corresponding to the training multi-energy system data; Using the semantic fusion network, semantically fuse the multi-energy system semantic vector corresponding to the training multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in the reference multi-energy operation solution set, and output the multi-energy solution semantic vector corresponding to the training multi-energy system data; Using the first mapping output network, mapping and outputting the multi-energy scheme semantic vector corresponding to the training multi-energy system data, to obtain the training multi-energy operation scheme corresponding to the training multi-energy system data; Obtaining a corresponding first training error based on a difference between a multi-energy operation scheme label corresponding to the training multi-energy system data and a training multi-energy operation scheme corresponding to the training multi-energy system data; Based on the first training error, the network parameters of the operation scheme determination network are updated to form an updated operation scheme determination network.

8. The method for optimizing operation of multi-energy comprehensive regulation in power system according to claim 7 is characterized in that: The semantic fusion network includes a first number of first connection units and a second number of first semantic fusion units, the first connection units and the first semantic fusion units are alternately arranged in sequence, and the difference between the first number and the second number is equal to 1; The step of using the semantic fusion network to semantically fuse the multi-energy system semantic vector corresponding to the training multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in the reference multi-energy operation solution set, and outputting the multi-energy solution semantic vector corresponding to the training multi-energy system data includes: Using the ath first connection unit, the multi-energy system input vector of the ath first connection unit is processed to form the multi-energy system output vector of the ath first connection unit, wherein the multi-energy system input vector of the first first connection unit includes the multi-energy solution semantic vector of the reference multi-energy operation solution, and the multi-energy system output vector of the last first connection unit includes the multi-energy solution semantic vector corresponding to the training multi-energy system data; Using the ath first semantic fusion unit, the multi-energy system input vector of the ath first semantic fusion unit is processed to form the multi-energy system output vector of the ath first semantic fusion unit, wherein the multi-energy system input vector of the ath first semantic fusion unit includes the multi-energy system output vector of the ath first connection unit and the multi-energy system semantic vector of the training multi-energy system data, and the multi-energy system input vector of the bth first connection unit includes the multi-energy system output vector of the ath first semantic fusion unit.

9. The method for optimizing operation of multi-energy comprehensive regulation in power system according to claim 7, characterized in that: The step of updating the network parameters of the operation scheme determination network based on the first training error to form an updated operation scheme determination network includes: Extracting at least one second training data combination, wherein each second training data combination includes a corresponding actual multi-energy operation scheme and a multi-energy system data label; Using the second semantic mining network, mining the multi-energy solution semantic vector corresponding to the actual multi-energy operation solution; Using the semantic fusion network, semantic fusion is performed on the multi-energy scheme semantic vector corresponding to the actual multi-energy operation scheme and the multi-energy system semantic vector corresponding to each reference multi-energy system data contained in the reference multi-energy system data set, and the multi-energy system semantic vector corresponding to the actual multi-energy operation scheme is output, wherein the multi-energy system semantic vector of the reference multi-energy system data is formed by mining using the first semantic mining network; Utilizing the second mapping output network included in the operation scheme determination network, mapping and outputting the multi-energy system semantic vector corresponding to the actual multi-energy operation scheme, and obtaining the multi-energy system generation data corresponding to the actual multi-energy operation scheme; Obtaining a corresponding second training error based on a difference between the multi-energy system data label corresponding to the actual multi-energy operation scheme and the multi-energy system generated data corresponding to the actual multi-energy operation scheme; Based on the first training error and the second training error, network parameters of the operation scheme determination network are updated to form an updated operation scheme determination network.

10. A multi-energy comprehensive regulation and optimization operation system for a power system, characterized in that: The method for implementing any one of claims 1 to 9 comprises: A description data determination module, used to determine multi-energy system data, wherein the multi-energy system data includes description data of multiple energy subsystems in the target power system and description data of power demand of the power consumption area corresponding to the target power system; A semantic mining module, configured to mine a multi-energy system semantic vector corresponding to the multi-energy system data using a first semantic mining network; A semantic fusion module, used for performing semantic fusion on the multi-energy system semantic vector corresponding to the multi-energy system data and the multi-energy solution semantic vector corresponding to each reference multi-energy operation solution included in the pre-configured reference multi-energy operation solution set by using a semantic fusion network, and outputting the multi-energy solution semantic vector corresponding to the multi-energy system data, wherein the multi-energy solution semantic vector corresponding to the reference multi-energy operation solution is mined by using a second semantic mining network; An operation plan output module is used to use a first mapping output network to map and output the multi-energy plan semantic vector corresponding to the multi-energy system data, so as to obtain a target multi-energy operation plan corresponding to the multi-energy system data, wherein the target multi-energy operation plan serves as a basis for the power generation operation of the various energy subsystems in the target power system.

Images