Low-carbon power system optimization method, device, computer equipment and storage medium

Abstract

The present application relates to a low-carbon power system optimization method, device, computer equipment and storage medium, and relates to the field of power system optimization technology. The method includes: determining the historical data of carbon emission factors in a target area within a preset time; inputting the historical data of carbon emission factors into a preconfigured carbon emission factor generation model to obtain a carbon emission factor baseline for the target area within the preset time; the carbon emission factor generation model is used to map the historical data of carbon emission factors to a low-dimensional space, and perform adversarial learning in the low-dimensional space to obtain an output result after fitting the time series characteristics corresponding to the historical data of carbon emission factors; simulating the target area based on the carbon emission factor baseline and a preconfigured power system time series simulation model to obtain a simulation result that meets multi-objective optimization conditions; the multi-objective optimization conditions at least include optimal network loss, optimal unit economic cost and optimal carbon emissions.

Description

Technical Field

[0001] The present application relates to the technical field of power system optimization, and in particular to a low-carbon power system optimization method, apparatus, computer equipment, storage medium, and computer program product. Background Art

[0002] Currently, regional power grids can plan the output of each unit through low-carbon power systems, taking into account factors such as power generation costs, power generation, network losses, and carbon emissions to achieve optimal results, thereby improving the economic efficiency and stability of the power grid. Related technologies can be optimized through various pre-configured optimization models for low-carbon power systems.

[0003] However, carbon emissions are primarily measured through carbon metering, which assesses the carbon emissions of regional power grids by measuring energy consumption and the corresponding carbon emission factor. Due to the intermittent and volatile nature of renewable energy, when the output of new energy units is insufficient to meet current electricity demand, energy storage systems are used to provide power support, causing fluctuations in the unit's carbon emission factor. However, the optimization models for low-carbon power systems in related technologies do not account for fluctuations in carbon emission factors, resulting in inaccurate optimization results when low-carbon power systems are optimized using these optimization models. Summary of the Invention

[0004] Based on this, it is necessary to provide a low-carbon power system optimization method, device, computer equipment, computer-readable storage medium and computer program product that can improve optimization accuracy in response to the above technical problems.

[0005] In a first aspect, the present application provides a low-carbon power system method. The method comprises:

[0006] Determine historical carbon emission factor data for the target area within a preset time period; the historical carbon emission factor data is determined based on the access power and static carbon emission factor of the new energy units, and the access power and static carbon emission factor of the energy storage system;

[0007] Inputting the historical carbon emission factor data into a preconfigured carbon emission factor generation model to obtain a carbon emission factor baseline for the target area within a preset time period; the carbon emission factor generation model is used to map the historical carbon emission factor data into a low-dimensional space, and perform adversarial learning in the low-dimensional space to obtain an output result after fitting the time series features corresponding to the historical carbon emission factor data;

[0008] Based on the carbon emission factor baseline and the preconfigured power system timing simulation model, simulation is performed on the target area to obtain simulation results that meet multi-objective optimization conditions; the multi-objective optimization conditions include at least optimal network loss, optimal unit economic cost and optimal carbon emissions; the simulation results include the access power of each new energy unit in the target area within a preset time.

[0009] In one embodiment, the carbon emission factor generation model is a time series generative adversarial network; the time series generative adversarial network includes a hidden Markov module, an embedding recovery module, a discrimination module and a generation module;

[0010] Inputting the carbon emission factor historical data into a pre-configured carbon emission factor generation model to obtain a carbon emission factor baseline for the target area within a preset time period includes:

[0011] Mapping the carbon emission factor historical data based on the embedding recovery module of the temporal generative adversarial network to obtain a low-dimensional output mapped to a low-dimensional space;

[0012] Determine a generation layer output corresponding to a hidden space based on a generation module of the temporal generative adversarial network; the generation layer output is obtained by Gaussian sampling of the hidden space;

[0013] Based on the hidden Markov module of the temporal generative adversarial network, the low-dimensional output and the generating layer output are processed to obtain a first hidden Markov state parameter corresponding to the low-dimensional output and a second hidden Markov state parameter corresponding to the generating layer output;

[0014] Training the hidden Markov module, the embedding recovery module, the discrimination module, and the generation module based on the first hidden Markov state parameter, the second hidden Markov state parameter, and the low-dimensional output to obtain a trained temporal generative adversarial network;

[0015] Based on the generation module corresponding to the trained time series generative adversarial network, the carbon emission factor baseline corresponding to the carbon emission factor historical data is determined, and is determined as the carbon emission factor baseline of the target area within a preset time.

[0016] In one embodiment, the training of the hidden Markov module, the embedding recovery module, the discrimination module, and the generation module based on the first hidden Markov state parameter, the second hidden Markov state parameter, and the low-dimensional output to obtain a trained temporal generative adversarial network includes:

[0017] Determining a hidden Markov loss corresponding to the hidden Markov module based on the first hidden Markov state parameter and the second hidden Markov state parameter;

[0018] Determining an embedding recovery loss corresponding to the embedding recovery module based on a predicted carbon emission factor baseline generated by the embedding recovery module and the carbon emission factor historical data; the predicted carbon emission factor baseline is obtained by processing the low-dimensional output and the second hidden Markov state parameter by the embedding recovery module;

[0019] Processing the low-dimensional output and the second hidden Markov state parameter by the discrimination module to obtain a discrimination result, and determining a discrimination loss corresponding to the discrimination module based on the discrimination result;

[0020] Determining a generation loss corresponding to the generation module based on the discrimination result and the predicted carbon emission factor baseline;

[0021] Based on the hidden Markov loss, the embedding recovery loss, the discrimination loss and the generation loss, the corresponding hidden Markov module, the embedding recovery module, the discrimination module, the generation module are trained respectively to obtain a trained temporal generative adversarial network.

[0022] In one embodiment, the power system timing simulation model includes a power simulation constraint sub-model and a multi-objective optimization sub-model;

[0023] The target area is simulated based on the carbon emission factor baseline and the preconfigured power system time series simulation model to obtain simulation results that meet multi-objective optimization conditions, including:

[0024] Initializing a power simulation constraint sub-model corresponding to the target area; wherein the power simulation constraint sub-model includes safety constraints, electrical nodes, and model variables; the electrical nodes include at least new energy generators; and the model variables include access power, voltage, and phase angle of each electrical node in the power system;

[0025] The power simulation constraint sub-model is imported into the multi-objective optimization model, and based on the network loss optimal objective function, the unit economic cost optimal objective function and the carbon emission optimal objective function in the multi-objective optimization sub-model, the power simulation constraint sub-model is solved to obtain a simulation result that meets the multi-objective optimization conditions; the carbon emission optimal objective function is determined based on the carbon emission factor baseline.

[0026] In one embodiment, the safety constraints include power safety sufficiency constraints, generator output constraints, voltage amplitude constraints, power angle constraints, and line safety constraints;

[0027] Initializing the power simulation constraint sub-model corresponding to the target area includes:

[0028] Determining the new energy generating units contained in the target area as electrical nodes, and determining the access power, voltage, and phase angle corresponding to each of the electrical nodes as model variables; the access power includes active power and reactive power;

[0029] Configuring the power security adequacy constraint and generator output constraint corresponding to the access power; the power security adequacy constraint is used to limit the value range of the remaining access power of the electrical node after the load is mounted;

[0030] Configuring a voltage amplitude constraint corresponding to the voltage; the voltage amplitude constraint is used to limit the value range of the voltage;

[0031] Configuring a power angle constraint corresponding to the phase angle; the power angle constraint is used to limit the value range of the phase angle;

[0032] A line safety constraint corresponding to the active power of the branch of the electrical node is configured; the line safety constraint is used to limit the value range of the remaining active power of the branch after the load is mounted.

[0033] In one embodiment, the power simulation constraint sub-model is imported into the multi-objective optimization model, and based on the network loss optimal objective function, the unit economic cost optimal objective function, and the carbon emission optimal objective function in the multi-objective optimization sub-model, the power simulation constraint sub-model is solved to obtain a simulation result that meets the multi-objective optimization conditions, including:

[0034] Determining a network loss optimal objective function based on the access power of the electrical node and the mounted load of the electrical node;

[0035] Determining an optimal objective function for the economic cost of the unit based on the access power of the electrical node;

[0036] Determining an optimal carbon emission objective function based on the carbon emission factor baseline of the electrical node and the access power;

[0037] Determine the differences between the network loss optimal objective function, the unit economic cost optimal objective function, and the carbon emission optimal objective function and their corresponding single-objective optimal solutions, and perform weighted summation of the differences to obtain a multi-objective optimization function;

[0038] The power simulation constraint sub-model is solved based on the multi-objective optimization function to obtain a simulation result that meets the multi-objective optimization conditions.

[0039] In one embodiment, determining the historical data of carbon emission factors of the target area within a preset time period includes:

[0040] Obtaining a first access power and a first static carbon emission factor of the new energy generating unit, and a second access power and a second static carbon emission factor of the energy storage system;

[0041] Determining, for each time point within a preset time, based on the first access power and the second access power, a first power proportion corresponding to the first access power and a second power proportion corresponding to the second access power;

[0042] The first power proportion, the second power proportion, and the corresponding first static carbon emission factor, the second static carbon emission factor are weighted and summed respectively to obtain the carbon emission factor historical data corresponding to the time point.

[0043] In a second aspect, the present application also provides a low-carbon power system optimization device. The device comprises:

[0044] a data determination module for determining historical carbon emission factor data for a target area within a preset time period; the historical carbon emission factor data is determined based on the access power and static carbon emission factor of the new energy generating unit, and the access power and static carbon emission factor of the energy storage system;

[0045] A baseline determination module is configured to input the carbon emission factor historical data into a preconfigured carbon emission factor generation model to obtain a carbon emission factor baseline for the target area within a preset time period; the carbon emission factor generation model is configured to map the carbon emission factor historical data into a low-dimensional space and perform adversarial learning in the low-dimensional space to obtain an output result after fitting the time series features corresponding to the carbon emission factor historical data;

[0046] A simulation module is used to simulate the target area based on the carbon emission factor baseline and a preconfigured power system timing simulation model to obtain simulation results that meet multi-objective optimization conditions; the multi-objective optimization conditions at least include optimal network loss, optimal unit economic cost and optimal carbon emissions; the simulation results include the access power of each new energy unit in the target area within a preset time.

[0047] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described in the first aspect when executing the computer program.

[0048] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0049] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0050] The above-mentioned low-carbon power system optimization method, device, computer equipment, storage medium and computer program product can determine the historical data of carbon emission factors of the target area within a preset time, and the historical data of carbon emission factors are determined based on the access power and static carbon emission factors corresponding to the new energy units and energy storage systems, and input the historical data of carbon emission factors into a pre-configured carbon emission factor generation model to obtain the carbon emission factor baseline of the target area within the preset time. The carbon emission factor generation model can map the historical data of carbon emission factors to a low-dimensional space and perform adversarial learning in the low-dimensional space. After the learning is completed, the output result of fitting the time series characteristics corresponding to the historical data of carbon emission factors is obtained. Finally, based on the carbon emission factor baseline and the preset The configured power system timing simulation model is simulated for the target area to obtain simulation results that at least meet the multi-objective optimization conditions of optimal network loss, optimal unit economic cost and optimal carbon emissions, thereby obtaining the access power of each new energy unit in the target area within the preset time. It can combine the access power and static carbon emission factor of the new energy units and energy storage systems within the preset time to reduce the fluctuation of the carbon emission factor when the output of the new energy units is insufficient, and improve the accuracy of determining the carbon emission factor baseline. Based on the carbon emission factor baseline and multi-objective optimization conditions, the power system timing simulation model is simulated, which can take into account the impact of the carbon emission factor baseline on the power system timing simulation model, thereby improving the accuracy of the simulation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0052] Figure 1 This is a diagram of an application environment of a low-carbon power system optimization method in one embodiment;

[0053] Figure 2 A schematic flow chart of a low-carbon power system optimization method according to an embodiment;

[0054] Figure 3 A schematic flow chart of the steps for obtaining a carbon emission factor baseline in one embodiment;

[0055] Figure 4A schematic flow chart of a low-carbon power system optimization method according to another embodiment;

[0056] Figure 5 A schematic diagram of the network structure of a temporal generative adversarial network in one embodiment;

[0057] Figure 6 Schematic diagram of an optimization framework of a power system timing simulation model in one embodiment;

[0058] Figure 7 A schematic diagram of a carbon emission factor baseline in one embodiment;

[0059] Figure 8 is a schematic diagram of power flow voltage and carbon emission factor in one embodiment;

[0060] Figure 9 A structural block diagram of a low-carbon power system optimization device in one embodiment;

[0061] Figure 10 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0063] The low-carbon power system optimization method provided in the embodiment of the present application can be applied to Figure 1In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. The terminal 102 can obtain the access power and static carbon emission factor of each new energy unit in the target area within a preset time, as well as the access power and static carbon emission factor of the energy storage system, and send the above-mentioned access power and static carbon emission factor to the server 104. The server can determine the historical carbon emission factor data of the target area within the preset time based on the access power and static carbon emission factor of each new energy unit and energy storage system. The server 104 can input the historical carbon emission factor data into the preconfigured carbon emission factor generation model to generate a carbon emission factor baseline for the target area within the preset time. The carbon emission factor generation model can perform adversarial learning based on the historical carbon emission factor data, so that the carbon emission factor generation model can generate a carbon emission factor baseline for the historical carbon emission factor data. Server 104 can simulate the power system in the target area based on the carbon emission factor baseline and the preconfigured power system timing simulation model, so as to obtain simulation results that meet the optimal network loss, optimal unit economic cost and optimal carbon emissions, and obtain the access power of each new energy unit in the target area within the preset time.

[0064] The terminal 102 may be, but is not limited to, various personal computers, laptop computers, smart phones, and tablet computers. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers.

[0065] In an exemplary embodiment, Figure 2 As shown, a carbon power system optimization method is provided, which is applied to Figure 1 The server 104 in the example is used as an example to illustrate the process, including the following steps S202 to S206.

[0066] Step S202: determining the historical data of the carbon emission factor of the target area within a preset time. The historical data of the carbon emission factor is determined based on the access power and static carbon emission factor of the new energy unit, and the access power and static carbon emission factor of the energy storage system.

[0067] Among them, the target area is a pre-configured geographical area, which can include multiple new energy units and energy storage systems. The target area can be, for example, a city, an area, a building, etc., which is not specifically limited in the embodiment of this application. The preset time can be a time interval formed by two time points, for example, a time interval corresponding to a month, a time interval corresponding to a day, and a time interval corresponding to a year, etc., which is not specifically limited in the embodiment of this application. In an exemplary embodiment, the server can determine the historical data of the carbon emission factor of a city in each year.

[0068] Historical carbon emission factor data includes the actual carbon emission factors recorded for the target region at each point in time during a preset period. Because new energy generators in the power system can experience unstable output, the energy storage system will output when these generators are unstable. The static carbon emission factor of the energy storage system differs from that of the new energy generators, resulting in different recorded carbon emission factors at different points in time. Based on this, the server can dynamically determine historical carbon emission factor data at different times based on the connected power and static carbon emission factors of the new energy generators, as well as the connected power and static carbon emission factors of the energy storage system, within the preset period.

[0069] It should be understood that a new energy generator set is a group of equipment that generates electricity through renewable energy, such as wind power and hydropower generation. An energy storage system can store electricity generated by various power generation methods. It can also generate electricity in real time through non-renewable energy methods and provide this stored or generated energy to a target area. For example, an energy storage system can provide electricity to the power system in a target area through thermal power generation or energy storage equipment. Access power refers to the power provided to the power system in real time by the new energy generator set and the energy storage system.

[0070] In one example, a server can obtain the new energy generators and energy storage systems that provide electricity to a target region, as well as the connected power of the new energy generators within a preset time period and the corresponding static carbon emission factors of the new energy generators. Furthermore, the connected power of the energy storage system within a preset time period and the corresponding static carbon emission factors of the energy storage system can be obtained. The static carbon emission factors corresponding to the new energy generators are typically greater than the static carbon emission factors. The static carbon emission factors can be fixed values ​​that do not change over time. For example, the static carbon emission factors of each type of generator in the new energy generators are first fixed values, while the static carbon emission factors of each type of generator or energy storage device in the energy storage system are second fixed values. When the new energy generators and energy storage system are simultaneously generating power, the first and second fixed values ​​coexist. Based on the respective output ratios (the ratios of connected power), the first and second fixed values, the overall historical carbon emission factor data for the target region within the preset time period is determined.

[0071] In step S204, the carbon emission factor historical data is input into a preconfigured carbon emission factor generation model to obtain a carbon emission factor baseline for the target area within a preset time period; the carbon emission factor generation model is used to map the carbon emission factor historical data to a low-dimensional space, and perform adversarial learning in the low-dimensional space to obtain an output result after fitting the time series features corresponding to the carbon emission factor historical data.

[0072] Among them, the carbon emission factor generation model can be a preconfigured fitting model or a prediction model. The carbon emission factor generation model can be fitted based on the historical data of the carbon emission factor to obtain a carbon emission factor baseline that conforms to the time series characteristics of the historical data of the carbon emission factor.

[0073] In one example, the server can input the carbon emission factor historical data into the carbon emission factor generation model. The carbon emission factor generation model can map the carbon emission factor historical data to a low-dimensional space and perform adversarial learning in the low-dimensional space. The server can continuously perform adversarial learning on the time series features in the low-dimensional space through the generator and the discriminator. The generator can generate a carbon emission factor, and the discriminator can discriminate the generated carbon emission factor based on the carbon emission factor historical data, thereby continuously optimizing the generator and the discriminator to obtain a trained carbon emission factor generation model, and through the trained generator, generate a carbon emission factor baseline for the target area within a preset time.

[0074] Step S206: Simulate the target region based on the carbon emission factor baseline and a preconfigured power system time-series simulation model to obtain simulation results that meet multi-objective optimization criteria. These criteria include at least optimal network loss, optimal unit economic cost, and optimal carbon emissions. The simulation results include the access power of each renewable energy unit in the target region within a preset timeframe.

[0075] Among them, the power system timing simulation model can model the power system in the target area and simulate the power system according to multi-objective optimization conditions. By simulating the access power of each new energy unit in the target area within the preset time, the network loss, unit economic cost and carbon emissions in the power system can be determined. The access power of each new energy unit is continuously optimized in three directions: optimal network loss, optimal unit economic cost and optimal carbon emissions, so as to obtain simulation results that meet the multi-objective optimization conditions.

[0076] Specifically, for a target region and within a preset timeframe, the server can input the corresponding carbon emission factor baseline into a preconfigured power system time-series simulation model. The power system time-series simulation model can then determine the access power of each renewable energy unit within the preset timeframe through simulation, and can also determine the power system's carbon emissions based on the carbon emission factor baseline. Through continuous simulation, the server can obtain simulation results that meet multi-objective optimization criteria, thereby determining the access power of each renewable energy unit in the target region within the preset timeframe.

[0077] In the above-mentioned low-carbon power system optimization method, the historical data of the carbon emission factor of the target area within the preset time can be determined, and the historical data of the carbon emission factor is determined based on the access power and static carbon emission factor corresponding to the new energy unit and the energy storage system, and the historical data of the carbon emission factor is input into the pre-configured carbon emission factor generation model to obtain the carbon emission factor baseline of the target area within the preset time. The carbon emission factor generation model can map the historical data of the carbon emission factor to a low-dimensional space and perform adversarial learning in the low-dimensional space. After the learning is completed, the output result of fitting the time series characteristics corresponding to the historical data of the carbon emission factor is obtained. Finally, based on the carbon emission factor baseline and the pre-configured power system time series simulation The model simulates the target area and obtains simulation results that at least meet the multi-objective optimization conditions of optimal network loss, optimal unit economic cost and optimal carbon emissions, thereby obtaining the access power of each new energy unit in the target area within the preset time. It can combine the access power and static carbon emission factor of the new energy units and energy storage systems within the preset time to reduce the fluctuation of the carbon emission factor when the output of the new energy units is insufficient, improve the accuracy of determining the carbon emission factor baseline, and simulate the power system timing simulation model based on the carbon emission factor baseline and multi-objective optimization conditions. It can take into account the impact of the carbon emission factor baseline on the power system timing simulation model, thereby improving the accuracy of the simulation results.

[0078] In an exemplary embodiment, Figure 3 As shown, the carbon emission factor generation model is a time series generative adversarial network; the time series generative adversarial network includes a hidden Markov module, an embedding recovery module, a discrimination module, and a generation module; the specific implementation process of the step "inputting historical carbon emission factor data into the pre-configured carbon emission factor generation model to obtain the carbon emission factor baseline of the target area within a preset time" includes steps S302 to S310. Among them:

[0079] Step S302 : Mapping the historical data of carbon emission factors based on the embedding recovery module of the time series generative adversarial network to obtain a low-dimensional output mapped to a low-dimensional space.

[0080] The carbon emission factor generation model can be a time series generative adversarial network, which includes a hidden Markov module, an embedding recovery module, a discriminant module, and a generation module. The hidden Markov module can process observations using a statistical model of a Markov process with hidden unknown parameters to obtain corresponding hidden state parameters. The observations can be historical carbon emission factor data mapped to a low-dimensional space or generated data generated by the generator in the generation module. The hidden state parameters can be the hidden Markov state parameters corresponding to the historical carbon emission factor data and the hidden Markov state parameters corresponding to the generated data.

[0081] The embedding recovery module can include an embedding layer and a recovery layer respectively. The server can map the historical data of the carbon emission factor to a low-dimensional space through the embedding layer, and the server can map the time series features of the low-dimensional space back to the real space through the recovery layer, thereby obtaining the fitted carbon emission factor baseline.

[0082] The discriminator module may include a discriminator, which compares the historical carbon emission factor data in the low-dimensional space with the generated data generated by the generator to determine the discrimination result of the generated data. The generation module may include a generator, which can sample random parameters in the latent space to obtain the generated data.

[0083] Specifically, the server can input the carbon emission factor historical data into the input layer of the embedding recovery module, send the data of the input layer to the embedding layer, and map the carbon emission factor historical data to a low-dimensional space through the embedding layer to obtain a low-dimensional output corresponding to the emission factor historical data.

[0084] Step S304: Determine the generation layer output corresponding to the latent space based on the generation module of the temporal generation adversarial network; the generation layer output is obtained by Gaussian sampling of the latent space.

[0085] The latent space generator converts the input noise into an intermediate representation space of observable data, and the latent space can contain noise. The generative layer can Gaussian sample the noise in the latent space to obtain noise data that conforms to the Gaussian distribution and serves as the output of the generative layer.

[0086] Step S306: Based on the hidden Markov module of the temporal generative adversarial network, the low-dimensional output and the generation layer output are processed to obtain the first hidden Markov state parameter corresponding to the low-dimensional output and the second hidden Markov state parameter corresponding to the generation layer output.

[0087] In one example, the server processes the low-dimensional output and the generating layer output through the hidden Markov layer in the hidden Markov module to determine the first hidden Markov state parameter and the second hidden Markov state parameter corresponding to the low-dimensional output and the generating layer output, respectively.

[0088] Step S308: Based on the first hidden Markov state parameter, the second hidden Markov state parameter and the low-dimensional output, the hidden Markov module, the embedding recovery module, the discrimination module and the generation module are trained to obtain a trained temporal generative adversarial network.

[0089] Specifically, the first hidden Markov state parameter, the second hidden Markov state parameter, and the low-dimensional output can be used as parameters of the loss function corresponding to the hidden Markov module, the embedding recovery module, the discriminant module, and the generator module, thereby determining the loss function corresponding to each module. Based on the first hidden Markov state parameter, the second hidden Markov state parameter, and the low-dimensional output in each round of training, the loss values ​​corresponding to the hidden Markov module, the embedding recovery module, the discriminant module, and the generator module are continuously updated, thereby adjusting the internal parameters of each module, thereby completing the training of the hidden Markov module, the embedding recovery module, the discriminant module, and the generator module, and then training the temporal generative adversarial network. When the trained temporal generative adversarial network meets the training end condition, a trained temporal generative adversarial network is obtained.

[0090] Step S310: Based on the generation module corresponding to the trained time series generative adversarial network, a carbon emission factor baseline corresponding to the carbon emission factor historical data is determined, and the carbon emission factor baseline of the target area within a preset time is determined.

[0091] Specifically, the server can call upon the generation module corresponding to the trained time-series generative adversarial network to generate a carbon emission factor baseline for the target region within a preset timeframe. Because the generator in the generation module is continuously trained based on historical carbon emission factor data, the generated carbon emission factor baseline conforms to the temporal characteristics of this historical data.

[0092] In this embodiment, the time series generative adversarial network is trained using historical data of carbon emission factors, and based on the historical data of carbon emission factors, the hidden Markov module, embedding recovery module, discrimination module and generation module in the time series generative adversarial network are trained separately through adversarial training to obtain a trained time series generative adversarial network, and then the corresponding carbon emission factor baseline is generated through the trained generation module, which can improve the accuracy of the carbon emission factor baseline.

[0093] In an exemplary embodiment, the specific implementation process of the step of "training the hidden Markov module, the embedding recovery module, the discrimination module, and the generation module based on the first hidden Markov state parameter, the second hidden Markov state parameter, and the low-dimensional output to obtain a trained temporal generative adversarial network" includes:

[0094] Step S402: Determine the hidden Markov loss corresponding to the hidden Markov module based on the first hidden Markov state parameter and the second hidden Markov state parameter.

[0095] Specifically, the server may determine the hidden Markov loss based on the difference between the first hidden Markov state parameter and the second hidden Markov state parameter, and the difference between the first hidden Markov state parameter and the second hidden Markov state parameter at all time points before the current time point.

[0096] In one example, the server can generate a hidden Markov loss function based on the difference between the first hidden Markov state parameter and the second hidden Markov state parameter, as well as the difference between the first hidden Markov state parameter and the second hidden Markov state parameter at all time points before the current time point, and after obtaining the first hidden Markov state parameter and the second hidden Markov state parameter corresponding to each time point, determine the hidden Markov loss based on the hidden Markov loss function.

[0097] In an example, the hidden Markov layer loss function can be expressed as:

[0098] ℒ s = 1 t ∑ [ ( h * hmm − g * hmm ) 2 + ( h * − g * ) 2 ]

[0099] in, is a constant, is the first hidden Markov state parameter, is the second hidden Markov state parameter, is the first hidden Markov state parameter at all moments before t-1, is the second hidden Markov state parameter of all moments before t-1.

[0100] Step S404: determining the embedded recovery loss corresponding to the embedded recovery module based on the predicted carbon emission factor baseline generated by the embedded recovery module and the carbon emission factor historical data.

[0101] Among them, the predicted carbon emission factor baseline is obtained by embedding the recovery module to process the low-dimensional output and the second hidden Markov state parameters.

[0102] Specifically, the server processes the low-dimensional output and the second hidden Markov state parameters through the recovery layer of the embedding recovery module to obtain a predicted carbon emission factor baseline. The predicted carbon emission factor baseline is obtained by mapping the processed low-dimensional output and the second hidden Markov state parameters back to the original space. Based on this, the server constructs an embedding recovery loss function based on the predicted carbon emission factor baseline and historical carbon emission factor data and calculates the loss value corresponding to the embedding recovery loss function.

[0103] In an example, the embedding recovery loss function can be expressed as:

[0104]

[0105] in, is the historical data of carbon emission factors, It is the baseline for predicting carbon emission factors.

[0106] Step S406: Process the low-dimensional output and the second hidden Markov state parameter through the discrimination module to obtain a discrimination result, and determine the discrimination loss corresponding to the discrimination module based on the discrimination result.

[0107] Specifically, the server can use the discriminant module to discriminate the low-dimensional output and the second hidden Markov state parameter to obtain a discrimination result, which can be true or false. If the discrimination result is true, a first discrimination loss function can be determined, and if the discrimination result is false, a second discrimination loss function can be determined. The server can obtain the first discrimination loss corresponding to the discriminant module through the first discrimination loss function, and can obtain the second discrimination loss corresponding to the discriminant module by providing the second discrimination loss function.

[0108] In one example, the first discriminant loss function and the second discriminant loss function can be expressed as follows:

[0109] ℒ d − r e a l = − 1 t ∑ [ ones ( y real )log( y real ) + [ 1 − ones ( y real )]log( 1 − y real )] ℒ d − f a k e = − 1 t ∑ [ zeros ( y fake )log( y fake ) + [ 1 − zeros ( y fake )]log( 1 − y fake )]

[0110] Among them, ones() represents a matrix with a value of 1, zeros() represents a matrix with a value of 0, and y real Indicates that the judgment result is true, y fake Indicates that the judgment result is false.

[0111] Step S408: Determine the generation loss corresponding to the generation module based on the discrimination result and the predicted carbon emission factor baseline.

[0112] Specifically, the server can determine a first generation loss function and a second generation loss function based on the discrimination results of the discrimination module and the predicted carbon emission factor baseline output by the recovery layer. The first generation loss function can determine the loss value of the generation layer, and the second generation loss function can determine the loss value between the generated data (predicted carbon emission factor baseline) and the initial data (historical carbon emission factor data).

[0113] In one example, the first generation loss function and the second generation loss function can be expressed as follows:

[0114] ℒ g 1 = − 1 t ∑ [ ones ( y fake )log( y fake ) + [ 1 − ones ( y fake )]log( 1 − y fake )] ℒ g 2 = 1 t ∑ [ m ( x , x * ) + s 2 ( x , x * ) ]

[0115] Among them, ones() represents a matrix with a value of 1, and zeros() represents a matrix with a value of 0. is the historical data of carbon emission factors, It is the baseline for predicting carbon emission factors. and denote the mean and variance operators respectively.

[0116] In step S410, based on the hidden Markov loss, embedding recovery loss, discrimination loss and generation loss, the corresponding hidden Markov module, embedding recovery module, discrimination module, generation module are trained respectively to obtain a trained temporal generative adversarial network.

[0117] In one example, for a round of training, the server can perform Gaussian sampling on the hidden space through the generation layer of the generation module to obtain the generation layer output g, and the server can determine the low-dimensional output h corresponding to the historical data of the carbon emission factor through the embedding layer in the embedding recovery module. Based on this, the server can input the generation layer output g and the low-dimensional output h into the hidden Markov module to obtain the hidden Markov state parameters corresponding to the generation layer output g and the low-dimensional output h, which can include the first hidden Markov state parameter , the second hidden Markov state parameter , the first hidden Markov state parameter of all moments before t-1 , the second hidden Markov state parameter of all moments before t-1 The server can embed the recovery layer of the recovery module to recover the low-dimensional output h and the second hidden Markov state parameter Map to the original space to obtain the predicted carbon emission factor baseline .

[0118] Based on this, for the hidden Markov module, the server can determine the hidden Markov loss based on the hidden Markov layer loss function. For the embedding recovery module, the server can determine the embedding recovery loss based on the embedding recovery loss function. For the discriminant module, the server can determine the first discriminant loss and the second discriminant loss based on the first discriminant loss function and the second discriminant loss function. For the generation module, the server can determine the first generation loss and the second generation loss based on the first generation loss function and the second generation loss function.

[0119] For the above-mentioned hidden Markov loss, embedding recovery loss, discriminant loss, and generation loss, the stochastic gradient descent method can be used to adjust the module parameters of the hidden Markov module, the module parameters of the embedding recovery module, the module parameters of the discriminant module, and the module parameters of the generation module respectively. After the adjustment, the updated low-dimensional output is obtained by re-passing the generation layer of the generation module. After that, each round of training process described above is repeated, and the module parameters of each module are continuously updated through the hidden Markov loss, embedding recovery loss, discriminant loss, and generation loss, so that the hidden Markov loss, embedding recovery loss, discriminant loss, and generation loss are reduced to the loss value range configured by the training end condition. After multiple rounds of training, and when the preset training end condition is met, a trained temporal generative adversarial network is obtained.

[0120] In this embodiment, the module parameters of each module are updated by determining the hidden Markov loss, embedding recovery loss, discrimination loss and generation loss, and the hidden Markov module, embedding recovery module, discrimination module and generation module are continuously updated through iteration, so as to train the time series generative adversarial network, thereby improving the training efficiency and training convergence speed, and improving the fitting accuracy of the time series generative adversarial network.

[0121] In an exemplary embodiment, the power system timing simulation model includes a power simulation constraint sub-model and a multi-objective optimization sub-model; the specific implementation process of the step of "simulating a target region based on a carbon emission factor baseline and a preconfigured power system timing simulation model to obtain a simulation result that meets the multi-objective optimization conditions" includes:

[0122] Step S502 : Initialize the power simulation constraint sub-model corresponding to the target area.

[0123] Among them, the power system timing simulation model includes a power simulation constraint sub-model and a multi-objective optimization sub-model; the power simulation constraint sub-model includes safety constraints, electrical nodes and model variables; the electrical nodes include at least new energy units; the model variables include the access power, voltage and phase angle of each electrical node in the power system.

[0124] Specifically, the server can configure safety constraints, electrical nodes, and model variables corresponding to the target region. The server can identify nodes corresponding to the new energy generators and energy storage systems in the target region as electrical nodes. The server can configure the access power, voltage, and phase angle of each electrical node at each time point as model variables. The server can also configure safety constraints for each model variable, thereby limiting the value range of the model variable during simulation using the multi-objective optimization submodel.

[0125] In step S504, the power simulation constraint sub-model is imported into the multi-objective optimization model, and based on the network loss optimal objective function, the unit economic cost optimal objective function and the carbon emission optimal objective function in the multi-objective optimization sub-model, the power simulation constraint sub-model is solved to obtain a simulation result that meets the multi-objective optimization conditions; the carbon emission optimal objective function is determined based on the carbon emission factor baseline.

[0126] Specifically, the server can configure the network loss optimal objective function, the unit economic cost optimal objective function, and the carbon emission optimal objective function at each time point within a preset time. For example, the time point can be a moment. The server can simulate each electrical node in the power simulation constraint submodel and the model variables of the electrical node through the multi-objective optimization model, and make the model variables meet the safety constraints during the simulation, thereby obtaining multiple sets of simulation results. The server can determine the final target value corresponding to each set of simulation results through the network loss optimal objective function, the unit economic cost optimal objective function, and the carbon emission optimal objective function in the multi-objective optimization submodel, and determine whether the final target value meets the multi-objective optimization conditions. In one example, minimizing the final target value can generally be used as the multi-objective optimization condition, and it can be determined that when the final target value is less than the pre-configured minimum threshold, the obtained simulation result meets the multi-objective optimization condition, thereby obtaining a simulation result that meets the multi-objective optimization condition, that is, the access power of each new energy unit in the target area within the preset time.

[0127] In this embodiment, by initializing the power simulation constraint sub-model corresponding to the target area, and importing the power simulation constraint sub-model into the multi-objective optimization model, and solving the power simulation constraint sub-model based on the network loss optimal objective function, the unit economic cost optimal objective function and the carbon emission optimal objective function in the multi-objective optimization sub-model, a simulation result that meets the multi-objective optimization conditions is obtained, and multiple objective functions can be integrated so that the simulation results obtained by simulation meet the multi-objective optimization conditions to the greatest extent, thereby improving the accuracy and rationality of the simulation results.

[0128] In an exemplary embodiment, the safety constraints include power safety and sufficiency constraints, generator output constraints, voltage amplitude constraints, power angle constraints, and line safety constraints. The specific implementation process of the step "initializing the power simulation constraint sub-model corresponding to the target area" includes:

[0129] Step S602: determine the new energy generating units included in the target area as electrical nodes, and determine the access power, voltage and phase angle corresponding to each electrical node as model variables.

[0130] The input power includes active power and reactive power. Active power refers to the power actually consumed and converted into electrical energy, and is the portion of the power system that actually performs work. Reactive power can be the power used to create electric or magnetic fields. This power is exchanged back and forth between the power source and energy storage elements (such as inductors and capacitors) and does not directly perform work.

[0131] Specifically, the server can obtain the new energy units included in the target area and identify the new energy units as electrical nodes. The server can also identify the energy storage system corresponding to the target area as an electrical node. The server can configure the access power, voltage, and phase angle of the electrical node as model variables.

[0132] In one example, the model variables of the power simulation constraint sub-model can be expressed as:

[0133] { P i , t = P i , t G − P i , t D Q i , t = Q i , t G − Q i , t D P i , t = ∑ j V i , t V j , t [ G ij cos( d i , t − d j , t ) + B ij sin( d i , t − d j , t )] Q i , t = ∑ j V i , t V j , t [ G ij sin( d i , t − d j , t ) − B ij cos( d i , t − d j , t )]

[0134] The above formula is the model of node injection power balance constraint. 、 denote the injected active power and reactive power of the node respectively, 、 They represent the active power and reactive power of the units connected to the node, 、 They represent the active power and reactive power of the node's load respectively. 、 Represent the voltage and phase angle of the unit respectively, 、 They represent the conductance and susceptance of the branch ij of the unit respectively, N is the set of system nodes, and the nodes ,time .

[0135] Step S604: configuring power security and sufficiency constraints and generator output constraints corresponding to the access power.

[0136] Among them, the power security and sufficiency constraint is used to limit the range of the remaining access power of the electrical node after the load is mounted.

[0137] In one example, at time t, the power security adequacy constraint is as follows:

[0138]

[0139] in, is the grid loss (network loss) of the power system at time t. Indicates the active power of the unit connected to the node, Indicates the active power of the load mounted on the node.

[0140] Step S606: configuring a voltage amplitude constraint corresponding to the voltage.

[0141] The voltage amplitude constraint is used to limit the voltage value range.

[0142] Step S608: configure the power angle constraint corresponding to the phase angle.

[0143] Among them, the power angle constraint is used to limit the value range of the phase angle.

[0144] Step S610: configuring line safety constraints corresponding to the active power of the branches of the electrical node.

[0145] Among them, the line safety constraint is used to limit the value range of the remaining active power of the branch after the load is mounted.

[0146] In an example, the generator output constraints, voltage amplitude constraints, power angle constraints, and line safety constraints are as follows:

[0147]

[0148] Among them, the subscripts "min" and "max" represent the minimum and maximum values ​​of the corresponding electrical parameters respectively. is the active power limit of the branch. 、 denote the injected active power and reactive power of the node respectively, 、 They represent the active and reactive power of the units connected to the node, 、 They represent the active and reactive power of the node's load respectively. 、 Represent the voltage and phase angle of the unit respectively.

[0149] In this embodiment, by configuring the electrical nodes, model variables, and safety constraints in the power simulation constraint sub-model, the integrity and richness of the power simulation constraint sub-model can be improved.

[0150] In an exemplary embodiment, the specific implementation process of the step of "importing the power simulation constraint sub-model into the multi-objective optimization model, and solving the power simulation constraint sub-model based on the network loss optimal objective function, the unit economic cost optimal objective function, and the carbon emission optimal objective function in the multi-objective optimization sub-model to obtain a simulation result that meets the multi-objective optimization conditions" includes:

[0151] Based on the access power of the electrical node and the mounted load of the electrical node, the optimal objective function of network loss is determined; based on the access power of the electrical node, the optimal objective function of unit economic cost is determined; based on the carbon emission factor baseline and access power of the electrical node, the optimal objective function of carbon emission is determined; the difference between the optimal objective function of network loss, the optimal objective function of unit economic cost and the optimal objective function of carbon emission and their corresponding single-objective optimal solutions is determined, and the differences are weighted and summed to obtain a multi-objective optimization function; based on the multi-objective optimization function, the power simulation constraint sub-model is solved to obtain simulation results that meet the multi-objective optimization conditions.

[0152] In one example, the network loss optimal objective function is It can be expressed as:

[0153]

[0154] Optimal objective function of unit economic cost It can be expressed as:

[0155]

[0156] Among them, a i 、b i and r i They respectively represent the quadratic term, linear term and constant term of the polynomial modeling of the unit output economic cost.

[0157] Optimal objective function for carbon emissions It can be expressed as:

[0158]

[0159] In one example, the specific implementation process of the step of "determining the differences between the network loss optimal objective function, the unit economic cost optimal objective function, and the carbon emission optimal objective function and their corresponding single-objective optimal solutions, and weighted summing the differences to obtain the multi-objective optimization function" can be expressed as:

[0160] { minutes f t BCP = min[ l loss , l cost , l CO 2 ] [ f t loss − f t ,min loss f t cost − f t ,min cost f t cost − f t ,min CO 2 ] st l loss + l cost + l CO 2 = 1

[0161] Where, 、 、 The server combines the objective functions of network loss optimization, economic optimization and carbon emission optimization with their corresponding single-objective optimal solutions. 、 、 The differences are calculated in sequence, and then linearly summed, while the corresponding weights are configured to achieve overall optimization of the three sets of objective functions.

[0162] In this embodiment, by solving the power simulation constraint sub-model based on the multi-objective optimization function, a simulation result that meets the multi-objective optimization conditions is obtained, which can improve the efficiency of the optimization and the accuracy and rationality of the simulation results obtained by the optimization.

[0163] In an exemplary embodiment, the specific implementation process of the step of “determining historical data of carbon emission factors of the target area within a preset time” includes:

[0164] Obtain the first access power and the first static carbon emission factor of the new energy unit, as well as the second access power and the second static carbon emission factor of the energy storage system; for each time point within the preset time, determine the first power ratio corresponding to the first access power and the second power ratio corresponding to the second access power based on the first access power and the second access power; respectively perform weighted summation of the first power ratio and the second power ratio with the corresponding first static carbon emission factor and the second static carbon emission factor to obtain the historical data of the carbon emission factor corresponding to the time point.

[0165] In one example, the ideal carbon emission factor for an energy storage system is defined as follows:

[0166]

[0167] Among them, at time t, for new energy unit i, 、 Respectively represent the access power of the new energy unit and the energy storage system, namely the first access power and the second access power, 、 Represent the first static carbon emission factor and the second static carbon emission factor of the new energy unit and the energy storage system respectively. From this, we can get the historical data of the carbon emission factor of the new energy unit i at time t By calculating the carbon emission factor over the entire time series, we can obtain a historical data set, denoted as X, and its data feature space is denoted as F.

[0168] In this embodiment, by weighted summing the first power ratio, the second power ratio and the corresponding first static carbon emission factor, the second static carbon emission factor, the historical data of the carbon emission factor corresponding to the time point is obtained, which can improve the accuracy of the historical data of the carbon emission factor.

[0169] In an exemplary embodiment, the solution of the embodiment of the present application includes two contents: (1) a method for generating a time-series dynamic carbon emission factor based on a generative framework; (2) a time-series multi-objective optimization modeling and framework for a low-carbon power system based on carbon emission factors and compromise planning. The following are respectively introduced:

[0170] (1) New energy sources have intermittent and volatile characteristics, and their carbon emission factors also have temporal volatility. Based on a time-series generative deep learning framework, the dynamic characteristics of energy storage systems are integrated to capture the intermittent and volatile characteristics of historical carbon emission factor data and generate a carbon emission factor baseline.

[0171] (2) Based on the carbon emission factor baseline generated in (1), combined with actual grid parameters and provincial grid planning schemes, a multi-objective optimization framework for the power system based on the Basic Compromise Programming (BCP) theory was constructed. The open source simulation framework PYOMO and the Interior-Point Optimizer (IPOPT) were combined to obtain the timing simulation results of the low-carbon power system.

[0172] This embodiment, by integrating the power system carbon emission factor baseline of the energy storage dynamic system and the multi-objective optimization simulation framework, can be used to solve the multi-dimensional optimization problem integrating economy, safety, and low carbon in actual engineering projects, and promote the application of electricity-carbon coupling technology in energy companies.

[0173] like Figure 4 As shown, the specific execution process of the above-mentioned low-carbon power system optimization method is described in detail below with reference to a specific embodiment, including the following steps:

[0174] Step 1: Construct a historical dataset of carbon emission factors for power systems integrated with energy storage systems. Step 2: Build a carbon emission factor training model based on a generative framework. Step 3: Complete training and export the trained model and carbon emission factor baseline; for example, export the trained model and carbon emission factor baseline from a trained neural network model. Step 4: Construct a low-carbon power system time-series simulation calculation model and perform iterative solution. For example, based on the PYOMO open source architecture, build a basic compromise programming (BCP) power system multi-objective time-series simulation calculation model and perform iterative solution. Step 5: Output the optimal system output, economic cost, and carbon emissions results.

[0175] Specifically, step 1: Construct a historical dataset of carbon emission factors for the power system that integrates the energy storage system. The ideal carbon emission factor for the energy storage system is defined as follows:

[0176] (1)

[0177] Among them, at time t, for new energy unit i, 、 Respectively represent the access power of the new energy unit and the energy storage system, namely the first access power and the second access power, 、 Represent the first static carbon emission factor and the second static carbon emission factor of the new energy unit and the energy storage system respectively. From this, we can get the historical data of the carbon emission factor of the new energy unit i at time t By calculating the carbon emission factor over the entire time series, we can obtain a historical data set, denoted as X, and its data feature space is denoted as .

[0178] Step 2: Build a carbon emission factor training model based on a generative framework. Figure 5 As shown in the figure, a temporal generative adversarial network with hidden Markov model (HMM-TSGAN) is selected as the generative artificial intelligence framework, including: embedding recovery module, generation module, discrimination module and hidden Markov module.

[0179] The embedding layer maps the original high-dimensional information to a low-dimensional space, reducing the dimensionality while improving the ability to learn core features. The recovery layer can restore the true sequence in the initial state space. The generative layer and the discriminative layer use a game to perform adversarial learning on Gaussian sampling in the latent space, allowing the generative layer model to simulate the distribution of the generated real space. The supervision layer sets up a supervised learning model with the true distribution as the label, fitting the time series features of the generative layer distribution and the true distribution in the low-dimensional space. For carbon emission factor data, the embedding recovery module model is:

[0180] (2)

[0181] Among them, h and r are the embedding layer and recovery layer models respectively. 、 For the corresponding parameters. is the low-dimensional output of the embedding layer mapping, is the output of the recovery layer restored to the real space. Assume that the low-dimensional space of the embedding layer is recorded as , the mathematical form of model optimization is:

[0182] { ℒ h =  x : ℱ , h : ℋ | | x − x ^ | | 2 m i n ℒ h ⇐ ∇ [ ∂ ℒ h ∂ ( i h + i r ) ] (3)

[0183] in, represents the loss function of the embedding-restoration pooling module, which is regularized by the L2 norm. Represents the stochastic gradient descent method for model parameters and Update. Generate layer instance as follows:

[0184] (4)

[0185] Among them, s, g, d and 、 、 They represent the instantiated hidden Markov layer, generation layer, and discrimination layer as well as the corresponding model parameters respectively. represents the Gaussian sampling z in the latent space Z. g, 、 denote the corresponding outputs of the generating layer, hidden Markov layer and recovery layer respectively, represents the output of the discriminative layer, represents the hidden Markov state parameter of the generating layer. The model is optimized as follows:

[0186] { ℒ g 1 =  z :  l o g ( 1 − y fake ) ℒ g 2 =  x , x * : ℱ [ m ( x , x * ) + s 2 ( x , x * ) ] (5)

[0187] m i n ( ℒ g 2 + ℒ g 1 ) ⇐ ∇ [ ∂ ( k 1 ℒ g 1 + k 2 ℒ g 2 ) ∂ ( i g + i r ) ] (6)

[0188] The above formula, represents the loss of the generated layer, represents the joint loss of generated data and initial data, where and denote the mean and variance operators respectively. Equation (6) gives the optimization solution process of the joint loss, where k1 and k2 are the hyperparameters of the joint loss.

[0189] The discriminant layer instance and optimization process are as follows:

[0190] { y real = d ( h ; i d ) ℒ d − r e a l = −  h : ℋ l o g ( y real ) ℒ d − f a k e =  z :  log( 1 − y fake ) min( ℒ d − r e a l + ℒ d − f a k e ) ⇐∇ [ ∂ ( ℒ d − r e a l + ℒ d − f a k e ) ∂ i d ] (7)

[0191] Where, represents the output of the discriminative layer, and Respectively represent the classification learning loss of the discriminator for different output labels. For a data with time series distribution characteristics, the state at this moment should be determined by the hidden states of multiple previous moments, such as:

[0192] (8)

[0193] Among them, p(s t ) indicates that time t is at s t The probability of the state. Then p(s t-1 ,s t-2 ,...) represents the probability of the hidden state at all previous moments. For parameters Supervised learning model , assuming that at time t, all the previous t-1 moments are Markov hidden state parameters denoted as , then the instantiation of the hidden Markov-supervised model is as follows:

[0194] (9)

[0195] The above formula Represents the output of the hidden Markov layer. Through supervised learning in the above formula, the generation layer can use the low-dimensional mapping of the initial data as the learning label, thereby improving the learning efficiency of the generation model for the true distribution.

[0196] Step 3: Complete training and export the trained model and carbon emission factors .

[0197] Step 4-1: Construct a low-carbon power system time series simulation model and perform iterative solution. Let N be the set of system nodes, and node ,time . Power flow constraints:

[0198] { P i , t = P i , t G − P i , t D Q i , t = Q i , t G − Q i , t D P i , t = ∑ j V i , t V j , t [ G ij cos( d i , t − d j , t ) + B ij sin( d i , t − d j , t )] Q i , t = ∑ j V i , t V j , t [ G ij sin( d i , t − d j , t ) − B ij cos( d i , t − d j , t )] (10)

[0199] The above formula is the model of node injection power balance constraint. 、 denote the injected active power and reactive power of the node respectively, 、 They represent the active power and reactive power of the units connected to the node, 、 They represent the active power and reactive power of the node's load respectively. 、 Represent the voltage and phase angle of the unit respectively, 、 They represent the conductance and susceptance of the branch ij of the unit respectively.

[0200] At time t, the power security and sufficiency constraints are as follows:

[0201] (11)

[0202] in, is the grid loss (network loss) of the power system at time t. Indicates the active power of the unit connected to the node, Indicates the active power of the load mounted on the node.

[0203] Generator output constraints, voltage amplitude constraints, power angle constraints, and line safety constraints are as follows:

[0204] (12)

[0205] Among them, the subscripts "min" and "max" represent the minimum and maximum values ​​of the corresponding electrical parameters respectively. is the active power limit of the branch. 、 denote the injected active power and reactive power of the node respectively, 、 They represent the active and reactive power of the units connected to the node, 、 They represent the active and reactive power of the node's load respectively. 、 Represent the voltage and phase angle of the unit respectively.

[0206] Optimal objective function of network loss at time t It can be expressed as:

[0207] (13)

[0208] The optimal objective function of the unit economic cost at time t It can be expressed as:

[0209] (14)

[0210] Among them, a i 、b i and r i They respectively represent the quadratic term, linear term and constant term of the polynomial modeling of the unit output economic cost.

[0211] Optimal objective function of carbon emissions at time t It can be expressed as:

[0212] (15)

[0213] From step (1), we can know that the carbon emission factor of the unit is It will fluctuate over time. The carbon emission factor here is taken from the learning results of step 2 and step 3. Note that the combined formula (10)-(12) will establish a non-convex nonlinear power simulation constraint model, and combined with the single objective function (13)-(15) to build a multi-objective optimization model. Figure 6 As shown in the figure, the multi-objective optimization model of power sequence based on BCP theory is as follows:

[0214] { minutes f t BCP = min[ l loss , l cost , l CO 2 ] [ f t loss − f t ,min loss f t cost − f t ,min cost f t cost − f t ,min CO 2 ] st l loss + l cost + l CO 2 = 1 (16)

[0215] Where, 、 、 The server combines the objective functions of network loss optimization, economic optimization and carbon emission optimization with their corresponding single-objective optimal solutions. 、 、 The differences are calculated in sequence, and then linearly summed, while the corresponding weights are configured to achieve overall optimization of the three sets of objective functions.

[0216] Step 4-2: Multi-objective optimization framework of low-carbon power system based on PYOMO-IPOPT, such as Figure 6 shown.

[0217] Step 5: Output the optimal system output, economic cost and carbon emissions results.

[0218] In an exemplary embodiment, the grid parameter information and carbon emission factor modeling settings are as follows:

[0219] Table 1: Static carbon emission factors for various types of units

[0220]

[0221] The simulation example uses the basic grid structure of a provincial power grid operating in extreme summer conditions in a particular year. The test system consists of 466 nodes, including 42 generator nodes and 79 load nodes, with 406 branches and 298 transformer branches.

[0222] like Figure 7 As shown in the figure, the horizontal axis represents the time period from 00:00 to 23:00. The gray dashed line represents the actual variation curve of the wind power and photovoltaic power carbon emission factors throughout 2022. The blue solid line represents the daily average variation curve of the wind power and photovoltaic power carbon emission factors. The green thick solid line represents the time series characteristic distribution curve generated by the HMM-TSGAN model. To quantitatively compare the two sets of curves, the standard deviation of the carbon emission factor within a 24-hour day is used as the volatility indicator. The daily average baseline volatility is 1.7 kg∙CO2 / MW∙h, the daily average volatility of the historical data is 5.6 kg∙CO2 / MW∙h, and the volatility of the generated characteristic curve is 3.2 kg∙CO2 / MW∙h. The generated characteristic curve simultaneously captures the stationarity and volatility of the historical data, achieving a balanced result that takes into account historical extreme values ​​and other data.

[0223] To better reflect the actual operational needs of the power grid, the 2025 planning scheme for Province A was selected as the basic business scenario for this simulation. According to the planning scheme, the peak load in 2025 is 10,500 MW, with a total installed capacity of 10,400 MW for conventional coal and gas turbines, 2,600 MW for nuclear power, 930 MW for hydropower, 600 MW for pumped storage, 4,000 MW for wind power, 6,500 MW for photovoltaic systems, and 1,200 MW for energy storage. Furthermore, coal-fired power, gas-fired power, and nuclear power plants are each allocated 15%-20% reserve capacity, while hydropower and pumped storage are allocated 30%-40% reserve capacity. Based on local hydrological characteristics, the dry season is defined from January to May, and the rest of the year is defined as the wet season. The available capacity of hydropower and pumped storage is adjusted to 85%-95% during the wet season and 15%-40% during the dry season. The grid loss boundary of the system is set to 200MW. The quadratic and linear terms of the economic cost model of traditional units are taken as [0.05, 0.15] (yuan / MW2) and [20, 60] (yuan / MW), respectively. For new energy sources such as wind power and photovoltaic power, the one-time installation cost and 20-year service life are estimated, and the constant term range is [57000, 62000] (yuan).

[0224] like Figure 8 As shown in Figure 2, the power flow, voltage distribution, and carbon potential distribution of each voltage level of the power grid in Province A are given at a certain time in the winter of 2025, when the weather is low wind and no sunlight. The carbon potential calculation is derived from the literature [1] and will not be elaborated here. At this time, the system load is 6444MW, the system unit output is 6586MW, and the system network loss is 135.06MW. Figure 8 Part (a) details the optimal branch power flow and voltage results for the 500 kV grid. For readability, the power flow calculation results for voltage levels 200 kV and below are not presented in detail. Corresponding to (a), part (b) provides a detailed distribution of the 500 kV node carbon potential and carbon potential histograms for grids at other voltage levels. At the current moment, the total node carbon potential for the test system is 76,724.66 kg∙CO₂ / MW∙h, of which the total node carbon potential for the 500 kV voltage level is 1,049.92 kg∙CO₂ / MW∙h, the total node carbon potential for the 220 kV voltage level is 29,331.38 kg∙CO₂ / MW∙h, and the total carbon potential for voltage levels 110 kV and below is 46,298.36 kg∙CO₂ / MW∙h.

[0225] Table 2: Seasonal comparison of multi-objective optimization results

[0226]

[0227] Table 2 compares the results of the BCP method with those of an unoptimized example. Based on the BCP model, the optimal annual CO2 equivalent emission result is 18.5746 million tons, the optimal annual economic cost of power generation is 6.417 billion yuan, and the minimum grid loss is 109.59 MW. Carbon emissions are reduced by an average of 23.96% over the entire period, and economic efficiency is improved by 14.25%.

[0228] This embodiment can generate a time-series carbon emission factor baseline for the power system, which can be used to optimize the quality of basic carbon emission data for new power systems. This embodiment proposes a time-series low-carbon power system simulation method based on BCP and an open-source simulation framework, and combines it with actual planning schemes to accurately assess the transformation path of the future power system and quantify economic, safety, and carbon emission indicators, making the project highly feasible.

[0229] It should be understood that, although the steps in the flowcharts of the above embodiments are shown in sequence as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be performed in other orders. Moreover, at least a portion of the steps in the flowcharts of the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily to be performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0230] Based on the same inventive concept, the present application also provides a low-carbon power system optimization device for implementing the low-carbon power system optimization method mentioned above. The implementation solution provided by this device is similar to the implementation solution described in the above method. Therefore, the specific limitations in one or more embodiments of the low-carbon power system optimization device provided below can be found in the above-mentioned limitations on the low-carbon power system optimization method and will not be repeated here.

[0231] In an exemplary embodiment, Figure 9 As shown, a low-carbon power system optimization device 900 is provided, comprising: a data determination module 901, a baseline determination module 902 and a simulation module 903, wherein:

[0232] Data determination module 901 is used to determine the historical carbon emission factor data of the target area within a preset time period; the historical carbon emission factor data is determined based on the access power and static carbon emission factor of the new energy unit, and the access power and static carbon emission factor of the energy storage system;

[0233] Baseline determination module 902 is used to input the historical carbon emission factor data into a pre-configured carbon emission factor generation model to obtain the carbon emission factor baseline for the target area within a preset time period; the carbon emission factor generation model is used to map the historical carbon emission factor data into a low-dimensional space and perform adversarial learning in the low-dimensional space to obtain an output result after fitting the time series features corresponding to the historical carbon emission factor data;

[0234] Simulation module 903 is used to simulate the target area based on the carbon emission factor baseline and the preconfigured power system timing simulation model to obtain simulation results that meet multi-objective optimization conditions; the multi-objective optimization conditions include at least optimal network loss, optimal unit economic cost and optimal carbon emissions; the simulation results include the access power of each new energy unit in the target area within a preset time.

[0235] Furthermore, the carbon emission factor generation model is a time-series generative adversarial network; the time-series generative adversarial network includes a hidden Markov module, an embedding recovery module, a discrimination module and a generation module; the baseline determination module 902 is specifically used to: map the historical data of the carbon emission factor based on the embedding recovery module of the time-series generative adversarial network to obtain a low-dimensional output mapped to the low-dimensional space; determine the generation layer output corresponding to the hidden space based on the generation module of the time-series generative adversarial network; the generation layer output is obtained by Gaussian sampling of the hidden space; based on the hidden Markov module of the time-series generative adversarial network, process the low-dimensional output and the generation layer output to obtain the first hidden Markov state parameter corresponding to the low-dimensional output and the second hidden Markov state parameter corresponding to the generation layer output; based on the first hidden Markov state parameter, the second hidden Markov state parameter and the low-dimensional output, train the hidden Markov module, the embedding recovery module, the discrimination module and the generation module to obtain a trained time-series generative adversarial network; determine the carbon emission factor baseline corresponding to the historical data of the carbon emission factor based on the generation module corresponding to the trained time-series generative adversarial network, and determine it as the carbon emission factor baseline of the target area within the preset time.

[0236] Furthermore, the baseline determination module 902 is specifically used to: determine the hidden Markov loss corresponding to the hidden Markov module based on the first hidden Markov state parameter and the second hidden Markov state parameter; determine the embedding recovery loss corresponding to the embedding recovery module based on the predicted carbon emission factor baseline generated by the embedding recovery module and the historical data of the carbon emission factor; the predicted carbon emission factor baseline is obtained by processing the low-dimensional output and the second hidden Markov state parameter by the embedding recovery module; the low-dimensional output and the second hidden Markov state parameter are processed by the discriminant module to obtain a discrimination result, and the discrimination loss corresponding to the discriminant module is determined based on the discrimination result; based on the discrimination result and the predicted carbon emission factor baseline, the generation loss corresponding to the generation module is determined; based on the hidden Markov loss, embedding recovery loss, discrimination loss and generation loss, the corresponding hidden Markov module, embedding recovery module, discrimination module, generation module are trained respectively to obtain a trained temporal generative adversarial network.

[0237] Furthermore, the power system timing simulation model includes a power simulation constraint sub-model and a multi-objective optimization sub-model, and the simulation module 903 is specifically used to: initialize the power simulation constraint sub-model corresponding to the target area; wherein, the power simulation constraint sub-model includes safety constraints, electrical nodes and model variables; the electrical nodes include at least new energy units; the model variables include the access power, voltage and phase angle of each electrical node in the power system; import the power simulation constraint sub-model into the multi-objective optimization model, and based on the network loss optimal objective function, the unit economic cost optimal objective function and the carbon emission optimal objective function in the multi-objective optimization sub-model, solve the power simulation constraint sub-model to obtain a simulation result that meets the multi-objective optimization conditions; the carbon emission optimal objective function is determined based on the carbon emission factor baseline.

[0238] Furthermore, the safety constraints include power safety sufficiency constraints, generator output constraints, voltage amplitude constraints, power angle constraints and line safety constraints. The simulation module 903 is specifically used to: determine the new energy units contained in the target area as electrical nodes, and determine the access power, voltage and phase angle corresponding to each electrical node as model variables; access power includes active power and reactive power; configure power safety sufficiency constraints and generator output constraints corresponding to access power; power safety sufficiency constraints are used to limit the value range of the remaining access power of the electrical node after the load is mounted; configure voltage amplitude constraints corresponding to voltage; voltage amplitude constraints are used to limit the value range of voltage; configure power angle constraints corresponding to phase angle; power angle constraints are used to limit the value range of phase angle; configure line safety constraints corresponding to the active power of the branch of the electrical node; line safety constraints are used to limit the value range of the remaining active power of the branch after the load is mounted.

[0239] Furthermore, the simulation module 903 is specifically used to: determine the optimal objective function of network loss based on the access power of the electrical node and the mounted load of the electrical node; determine the optimal objective function of unit economic cost based on the access power of the electrical node; determine the optimal objective function of carbon emission based on the carbon emission factor baseline and access power of the electrical node; determine the difference between the optimal objective function of network loss, the optimal objective function of unit economic cost and the optimal objective function of carbon emission and their corresponding single-objective optimal solutions, and weightedly sum the differences to obtain a multi-objective optimization function; solve the power simulation constraint sub-model based on the multi-objective optimization function to obtain a simulation result that meets the multi-objective optimization conditions.

[0240] Furthermore, the data determination module 901 is specifically used to: obtain the first access power and the first static carbon emission factor of the new energy unit, and the second access power and the second static carbon emission factor of the energy storage system; for each time point within the preset time, based on the first access power and the second access power, determine the first power ratio corresponding to the first access power and the second power ratio corresponding to the second access power; and weightedly sum the first power ratio and the second power ratio with the corresponding first static carbon emission factor and the second static carbon emission factor to obtain the historical data of the carbon emission factor corresponding to the time point.

[0241] Each module in the aforementioned low-carbon power system optimization device may be implemented in whole or in part through software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0242] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 10 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O) and a communication interface. The processor, memory and input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store historical data of carbon emission factors. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a low-carbon power system optimization method is implemented.

[0243] Those skilled in the art will understand that Figure 10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0244] In an exemplary embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0245] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0246] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0247] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), data processing logic devices based on quantum computing, and the like.

[0248] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0249] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A low-carbon power system optimization method, characterized in that: The method comprises: Determining historical data of carbon emission factors for a target area within a preset time period includes: obtaining a first access power and a first static carbon emission factor of a new energy generating unit, and a second access power and a second static carbon emission factor of an energy storage system; determining, for each time point within the preset time period, a first power ratio corresponding to the first access power and a second power ratio corresponding to the second access power based on the first access power and the second access power; and weightedly summing the first power ratio and the second power ratio with the corresponding first static carbon emission factor and second static carbon emission factor, respectively, to obtain historical data of carbon emission factors corresponding to the time point; Inputting the historical carbon emission factor data into a preconfigured carbon emission factor generation model to obtain a carbon emission factor baseline for the target area within a preset time period; the carbon emission factor generation model is used to map the historical carbon emission factor data into a low-dimensional space, and perform adversarial learning in the low-dimensional space to obtain an output result after fitting the time series features corresponding to the historical carbon emission factor data; Based on the carbon emission factor baseline and the preconfigured power system timing simulation model, simulation is performed on the target area to obtain simulation results that meet multi-objective optimization conditions; the multi-objective optimization conditions include at least optimal network loss, optimal unit economic cost and optimal carbon emissions; the simulation results include the access power of each new energy unit in the target area within a preset time.

2. The method according to claim 1, characterized in that The carbon emission factor generation model is a time series generative adversarial network; the time series generative adversarial network includes a hidden Markov module, an embedding recovery module, a discrimination module and a generation module; Inputting the carbon emission factor historical data into a pre-configured carbon emission factor generation model to obtain a carbon emission factor baseline for the target area within a preset time period includes: Mapping the carbon emission factor historical data based on the embedding recovery module of the temporal generative adversarial network to obtain a low-dimensional output mapped to a low-dimensional space; Determine a generation layer output corresponding to a hidden space based on a generation module of the temporal generative adversarial network; the generation layer output is obtained by Gaussian sampling of the hidden space; Based on the hidden Markov module of the temporal generative adversarial network, the low-dimensional output and the generating layer output are processed to obtain a first hidden Markov state parameter corresponding to the low-dimensional output and a second hidden Markov state parameter corresponding to the generating layer output; Training the hidden Markov module, the embedding recovery module, the discrimination module, and the generation module based on the first hidden Markov state parameter, the second hidden Markov state parameter, and the low-dimensional output to obtain a trained temporal generative adversarial network; Based on the generation module corresponding to the trained time series generative adversarial network, the carbon emission factor baseline corresponding to the carbon emission factor historical data is determined, and is determined as the carbon emission factor baseline of the target area within a preset time.

3. The method according to claim 2, characterized in that The method of training the hidden Markov module, the embedding recovery module, the discrimination module, and the generation module based on the first hidden Markov state parameter, the second hidden Markov state parameter, and the low-dimensional output to obtain a trained temporal generative adversarial network includes: Determining a hidden Markov loss corresponding to the hidden Markov module based on the first hidden Markov state parameter and the second hidden Markov state parameter; Determining an embedding recovery loss corresponding to the embedding recovery module based on a predicted carbon emission factor baseline generated by the embedding recovery module and the carbon emission factor historical data; the predicted carbon emission factor baseline is obtained by processing the low-dimensional output and the second hidden Markov state parameter by the embedding recovery module; Processing the low-dimensional output and the second hidden Markov state parameter by the discrimination module to obtain a discrimination result, and determining a discrimination loss corresponding to the discrimination module based on the discrimination result; Determining a generation loss corresponding to the generation module based on the discrimination result and the predicted carbon emission factor baseline; Based on the hidden Markov loss, the embedding recovery loss, the discrimination loss and the generation loss, the corresponding hidden Markov module, the embedding recovery module, the discrimination module, the generation module are trained respectively to obtain a trained temporal generative adversarial network.

4. The method according to claim 1, wherein The power system timing simulation model includes a power simulation constraint sub-model and a multi-objective optimization sub-model; The target area is simulated based on the carbon emission factor baseline and the preconfigured power system time series simulation model to obtain simulation results that meet multi-objective optimization conditions, including: Initializing a power simulation constraint sub-model corresponding to the target area; wherein the power simulation constraint sub-model includes safety constraints, electrical nodes, and model variables; the electrical nodes include at least new energy generators; and the model variables include access power, voltage, and phase angle of each electrical node in the power system; The power simulation constraint sub-model is imported into the multi-objective optimization sub-model, and based on the network loss optimal objective function, the unit economic cost optimal objective function and the carbon emission optimal objective function in the multi-objective optimization sub-model, the power simulation constraint sub-model is solved to obtain a simulation result that meets the multi-objective optimization conditions; the carbon emission optimal objective function is determined based on the carbon emission factor baseline.

5. The method according to claim 4, characterized in that The safety constraints include power safety and sufficiency constraints, generator output constraints, voltage amplitude constraints, power angle constraints and line safety constraints; Initializing the power simulation constraint sub-model corresponding to the target area includes: Determining the new energy generating units contained in the target area as electrical nodes, and determining the access power, voltage, and phase angle corresponding to each of the electrical nodes as model variables; the access power includes active power and reactive power; Configuring the power security adequacy constraint and generator output constraint corresponding to the access power; the power security adequacy constraint is used to limit the value range of the remaining access power of the electrical node after the load is mounted; Configuring a voltage amplitude constraint corresponding to the voltage; the voltage amplitude constraint is used to limit the value range of the voltage; Configuring a power angle constraint corresponding to the phase angle; the power angle constraint is used to limit the value range of the phase angle; A line safety constraint corresponding to the active power of the branch of the electrical node is configured; the line safety constraint is used to limit the value range of the remaining active power of the branch after the load is mounted.

6. The method according to claim 4, characterized in that The power simulation constraint sub-model is imported into the multi-objective optimization sub-model, and based on the network loss optimal objective function, the unit economic cost optimal objective function, and the carbon emission optimal objective function in the multi-objective optimization sub-model, the power simulation constraint sub-model is solved to obtain a simulation result that meets the multi-objective optimization conditions, including: Determining a network loss optimal objective function based on the access power of the electrical node and the mounted load of the electrical node; Determining an optimal objective function for the economic cost of the unit based on the access power of the electrical node; Determining an optimal carbon emission objective function based on the carbon emission factor baseline of the electrical node and the access power; Determine the differences between the network loss optimal objective function, the unit economic cost optimal objective function, and the carbon emission optimal objective function and their corresponding single-objective optimal solutions, and perform weighted summation of the differences to obtain a multi-objective optimization function; The power simulation constraint sub-model is solved based on the multi-objective optimization function to obtain a simulation result that meets the multi-objective optimization conditions.

7. A low-carbon power system optimization device, characterized in that: The device comprises: a data determination module for determining historical data of carbon emission factors in a target area within a preset time, including: obtaining a first access power and a first static carbon emission factor of the new energy generating unit, and a second access power and a second static carbon emission factor of the energy storage system; determining, for each time point within the preset time point, a first power ratio corresponding to the first access power and a second power ratio corresponding to the second access power based on the first access power and the second access power; and performing a weighted summation of the first power ratio and the second power ratio with the corresponding first static carbon emission factor and second static carbon emission factor, respectively, to obtain historical data of carbon emission factors corresponding to the time point; A baseline determination module is configured to input the carbon emission factor historical data into a preconfigured carbon emission factor generation model to obtain a carbon emission factor baseline for the target area within a preset time period; the carbon emission factor generation model is configured to map the carbon emission factor historical data into a low-dimensional space and perform adversarial learning in the low-dimensional space to obtain an output result after fitting the time series features corresponding to the carbon emission factor historical data; A simulation module is used to simulate the target area based on the carbon emission factor baseline and a preconfigured power system timing simulation model to obtain simulation results that meet multi-objective optimization conditions; the multi-objective optimization conditions at least include optimal network loss, optimal unit economic cost and optimal carbon emissions; the simulation results include the access power of each new energy unit in the target area within a preset time.

8. The device according to claim 7, characterized in that The carbon emission factor generation model is a time series generative adversarial network; the time series generative adversarial network includes a hidden Markov module, an embedding recovery module, a discrimination module and a generation module; The baseline determination module is also used to Mapping the carbon emission factor historical data based on the embedding recovery module of the temporal generative adversarial network to obtain a low-dimensional output mapped to a low-dimensional space; Determine a generation layer output corresponding to a hidden space based on a generation module of the temporal generative adversarial network; the generation layer output is obtained by Gaussian sampling of the hidden space; Based on the hidden Markov module of the temporal generative adversarial network, the low-dimensional output and the generating layer output are processed to obtain a first hidden Markov state parameter corresponding to the low-dimensional output and a second hidden Markov state parameter corresponding to the generating layer output; Training the hidden Markov module, the embedding recovery module, the discrimination module, and the generation module based on the first hidden Markov state parameter, the second hidden Markov state parameter, and the low-dimensional output to obtain a trained temporal generative adversarial network; Based on the generation module corresponding to the trained time series generative adversarial network, the carbon emission factor baseline corresponding to the carbon emission factor historical data is determined, and is determined as the carbon emission factor baseline of the target area within a preset time.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images