Power system grid optimization method and apparatus

Abstract

The application relates to the technical field of low-carbon power systems, in particular to a power grid optimization method and device for a power system, wherein the method comprises the following steps: generating a data set of system dynamic emission factors through power system operation simulation and carbon emission flow calculation; training a feedforward neural network model by using the data set, obtaining a trained neural network, converting the trained neural network into a mixed integer programming constraint, and constructing a double-layer optimization model considering user low-carbon demand response behavior to perform carbon emission reduction on the demand side of power consumption. Therefore, the problem that, in the related art, the dynamic carbon emission factor lacks load interaction capability in the demand response of the power system, thus causing a large error of demand guidance and affecting the carbon reduction effect of the power system is solved.

Description

Technical Field

[0001] This application relates to the field of low-carbon power system technology, and in particular to a power grid optimization method and apparatus for a power system. Background Technology

[0002] In related technologies, in order to achieve carbon emission reduction, low-carbon demand response methods provide guidance signals on the source side, and the load side flexibly adjusts the power consumption according to the guidance signals. The dynamic emission factor calculated based on carbon emission flow improves the temporal and spatial granularity, and guides demand response more effectively.

[0003] However, in related technologies, as the load changes, the generator output will also be adjusted, and the carbon emission factor will also change accordingly. Therefore, as the adjustment space on the load side increases in the future, the carbon emission factor obtained by operation simulation based on the original load forecast may produce large errors or even generate incorrect signals after the load changes, which urgently needs to be improved. Summary of the Invention

[0004] This application provides a power grid optimization method and apparatus for a power system to address the problem in related technologies where the dynamic carbon emission factor lacks the ability to interact with the load in the power system demand response, resulting in large errors in demand guidance and affecting the carbon reduction effect of the power system.

[0005] The first aspect of this application provides a power grid optimization method for a power system, comprising the following steps: generating a dataset of dynamic emission factors of the system through power system operation simulation and carbon emission flow calculation; training a feedforward neural network model using the dataset to obtain a trained neural network; converting the trained neural network into mixed integer programming constraints to construct a two-layer optimization model that considers the low-carbon demand response behavior of users, so as to carry out carbon emission reduction on the demand side of electricity consumption.

[0006] Through the above technical solutions, the embodiments of this application can generate a dynamic emission factor dataset of the system through power system operation simulation and carbon emission flow calculation, providing a data foundation for subsequent operations. Then, this dataset is used to train a feedforward neural network model. After obtaining the trained neural network, it is transformed into mixed integer programming constraints, and then a two-layer optimization model considering the user's low-carbon demand response behavior is constructed. This achieves carbon emission reduction on the electricity demand side. The whole process, from data generation and model training to building an optimization model, forms a complete logic, which helps to effectively promote the power system to achieve carbon emission reduction targets on the electricity demand side, in line with the needs of low-carbon development.

[0007] Optionally, in one embodiment of this application, the step of generating a dataset of dynamic emission factors of the system through power system operation simulation and carbon emission flow calculation includes: generating the system load curve of the power system to obtain the result of economic dispatch through power system operation simulation; calculating the dynamic carbon emission factor of each node under the corresponding scenario using carbon emission flow theory; and obtaining the dynamic carbon emission factor dataset including massive power system operation scenarios based on the result of economic dispatch and the dynamic carbon emission factor of each node.

[0008] Through the above technical solutions, the embodiments of this application can generate the system load curve of the power system and obtain the economic dispatch results by means of power system operation simulation, laying the foundation for subsequent operations. Then, the dynamic carbon emission factor of each node under the corresponding scenario is calculated by using carbon emission flow theory. Finally, based on the economic dispatch results and the dynamic carbon emission factors of each node, a dynamic carbon emission factor dataset containing a large number of power system operation scenarios is integrated. This dataset can comprehensively and meticulously consider a variety of operation scenarios, making the generated dataset more representative and practical. This provides high-quality and realistic data support for subsequent operations such as training models and building optimization models, ensuring that the power grid optimization work of the entire power system is carried out more scientifically and accurately.

[0009] Optionally, in one embodiment of this application, the economic dispatch model of the power system, with the objective of minimizing the total system cost, is defined as follows:

[0010] min C = c T P G Δt,

[0011] Where c is the unit output cost of the generator set, and P G Δt represents the generator output vector, and Δt is the unit time interval.

[0012] Through the above technical solutions, the embodiments of this application can quantify and control the key element of total system cost, clearly and intuitively pointing out the optimization direction for the economic dispatch of the power system. This helps to accurately weigh various related factors during the dispatch process in order to reduce costs as much as possible. As a result, the power system can achieve more efficient resource allocation and improve economic benefits during operation, laying a solid foundation for the stable and economical operation of the entire power system and ensuring that subsequent related operations can be carried out in an orderly manner under the premise of better cost.

[0013] Optionally, in one embodiment of this application, the formula for calculating the dynamic carbon emission factor is:

[0014]

[0015] Among them, E N Let P be the dynamic carbon emission factor vector for each node.N Let P be the active flux matrix. B Inject a matrix into the node, P G E is the generator output vector. G This represents the unit carbon emission vector for the generator set.

[0016] Through the above technical solutions, the embodiments of this application can accurately measure the specific value of dynamic carbon emission factors, making the analysis of carbon emissions in various aspects of the power system more scientific and accurate. This provides reliable data for subsequent operations such as generating datasets, training models based on carbon emission-related information, and building optimization models. It helps to control carbon emissions in the power system more meticulously and reasonably, thereby better serving the low-carbon optimization of the entire power system.

[0017] Optionally, in one embodiment of this application, wherein,

[0018] The objective function of the upper-level optimization in the two-level optimization model is:

[0019]

[0020] Among them, c g This represents the unit output cost vector of the generator set. It is the power output vector of the unit at time t, where Δt is the unit time interval;

[0021] The objective function of the lower-level optimization in the two-level optimization model is:

[0022]

[0023] Among them, c L,t It is the vector of unit electricity cost of load l at time t. It is the power consumption vector of the load at time t, where Δt is the unit time interval.

[0024] Through the above technical solutions, the embodiments of this application can start from the overall power system, and by minimizing the total cost including unit operation and carbon emission costs, guide the power system to comprehensively consider economic and environmental factors during unit operation. This enables units to allocate resources in a more optimized manner under the conditions of meeting upper and lower output limits, power flow safety constraints, and carbon emission flow calculation formulas, thereby improving system operating efficiency, reducing total system costs, and achieving a balance between economic and environmental benefits. The lower-level optimization objective function starts from the perspective of individual users, guiding users to optimize their own demand response behavior based on dynamic carbon emission factors. Under the condition of meeting linear constraints for load types, it minimizes the total electricity cost. This helps users adjust their electricity consumption strategies according to carbon emission costs, increases users' enthusiasm for participating in demand response, and thus achieves carbon emission reduction on the demand side of electricity consumption. At the same time, it also ensures the rationality of users' economic costs, promotes source-load interaction, and improves the low-carbon operation effect of the entire power system.

[0025] A second aspect of this application provides a power grid optimization device for a power system, comprising: a calculation module for generating a dataset of dynamic emission factors of the system through power system operation simulation and carbon emission flow calculation; a training module for training a feedforward neural network model using the dataset to obtain a trained neural network; and an optimization module for converting the trained neural network into mixed integer programming constraints to construct a two-layer optimization model that considers the low-carbon demand response behavior of users, so as to carry out carbon emission reduction on the demand side of electricity consumption.

[0026] Through the above technical solutions, the embodiments of this application can generate a dynamic emission factor dataset of the system through power system operation simulation and carbon emission flow calculation, providing a data foundation for subsequent operations. Then, this dataset is used to train a feedforward neural network model. After obtaining the trained neural network, it is transformed into mixed integer programming constraints, and then a two-layer optimization model considering the user's low-carbon demand response behavior is constructed. This achieves carbon emission reduction on the electricity demand side. The whole process, from data generation and model training to building an optimization model, forms a complete logic, which helps to effectively promote the power system to achieve carbon emission reduction targets on the electricity demand side, in line with the needs of low-carbon development.

[0027] Optionally, in one embodiment of this application, the calculation module includes: a simulation unit for generating the system load curve of the power system to obtain the result of economic dispatch through power system operation simulation; a calculation unit for calculating the dynamic carbon emission factor of each node under the corresponding scenario using carbon emission flow theory; and a generation unit for obtaining the dynamic carbon emission factor dataset including massive power system operation scenarios based on the result of economic dispatch and the dynamic carbon emission factor of each node.

[0028] Through the above technical solutions, the embodiments of this application can generate the system load curve of the power system and obtain the economic dispatch results by means of power system operation simulation, laying the foundation for subsequent operations. Then, the dynamic carbon emission factor of each node under the corresponding scenario is calculated by using carbon emission flow theory. Finally, based on the economic dispatch results and the dynamic carbon emission factors of each node, a dynamic carbon emission factor dataset containing a large number of power system operation scenarios is integrated. This dataset can comprehensively and meticulously consider a variety of operation scenarios, making the generated dataset more representative and practical. This provides high-quality and realistic data support for subsequent operations such as training models and building optimization models, ensuring that the power grid optimization work of the entire power system is carried out more scientifically and accurately.

[0029] Optionally, in one embodiment of this application, the economic dispatch model of the power system, with the objective of minimizing the total system cost, is defined as follows:

[0030] min C = c T P G Δt,

[0031] Where c is the unit output cost of the generator set, and P G Δt represents the generator output vector, and Δt is the unit time interval.

[0032] Through the above technical solutions, the embodiments of this application can quantify and control the key element of total system cost, clearly and intuitively pointing out the optimization direction for the economic dispatch of the power system. This helps to accurately weigh various related factors during the dispatch process in order to reduce costs as much as possible. As a result, the power system can achieve more efficient resource allocation and improve economic benefits during operation, laying a solid foundation for the stable and economical operation of the entire power system and ensuring that subsequent related operations can be carried out in an orderly manner under the premise of better cost.

[0033] Optionally, in one embodiment of this application, the formula for calculating the dynamic carbon emission factor is:

[0034]

[0035] Among them, E N Let P be the dynamic carbon emission factor vector for each node. N Let P be the active flux matrix. B Inject a matrix into the node, P G E is the generator output vector. G This represents the unit carbon emission vector for the generator set.

[0036] Through the above technical solutions, the embodiments of this application can accurately measure the specific value of dynamic carbon emission factors, making the analysis of carbon emissions in various aspects of the power system more scientific and accurate. This provides reliable data for subsequent operations such as generating datasets, training models based on carbon emission-related information, and building optimization models. It helps to control carbon emissions in the power system more meticulously and reasonably, thereby better serving the low-carbon optimization of the entire power system.

[0037] Optionally, in one embodiment of this application, wherein,

[0038] The objective function of the upper-level optimization in the two-level optimization model is:

[0039]

[0040] Among them, c g This represents the unit output cost vector of the generator set. It is the power output vector of the unit at time t, where Δt is the unit time interval;

[0041] The objective function of the lower-level optimization in the two-level optimization model is:

[0042]

[0043] Among them, c l,t It is the vector of unit electricity cost of load l at time t. It is the power consumption vector of the load at time t, where Δt is the unit time interval.

[0044] Through the above technical solutions, the embodiments of this application can start from the overall power system, and by minimizing the total cost including unit operation and carbon emission costs, guide the power system to comprehensively consider economic and environmental factors during unit operation. This enables units to allocate resources in a more optimized manner under the conditions of meeting upper and lower output limits, power flow safety constraints, and carbon emission flow calculation formulas, thereby improving system operating efficiency, reducing total system costs, and achieving a balance between economic and environmental benefits. The lower-level optimization objective function starts from the perspective of individual users, guiding users to optimize their own demand response behavior based on dynamic carbon emission factors. Under the condition of meeting linear constraints for load types, it minimizes the total electricity cost. This helps users adjust their electricity consumption strategies according to carbon emission costs, increases users' enthusiasm for participating in demand response, and thus achieves carbon emission reduction on the demand side of electricity consumption. At the same time, it also ensures the rationality of users' economic costs, promotes source-load interaction, and improves the low-carbon operation effect of the entire power system.

[0045] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the power grid optimization method for a power system as described in the above embodiments.

[0046] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described power grid optimization method for a power system.

[0047] A fifth aspect of this application provides a computer program product, including a computer program that, when executed, implements the above-described power grid optimization method for a power system.

[0048] This application embodiment can generate system load curves and obtain economic dispatch results through power system operation simulation. It uses carbon emission flow theory to calculate dynamic carbon emission factors, generating a massive dataset of operational scenarios to provide high-quality data support for subsequent operations. This comprehensive approach considers various scenarios, making the data more practical. Next, it quantifies and controls the total system cost, providing direction for economic dispatch and aiding in resource optimization and improved economic efficiency. It can also accurately measure dynamic carbon emission factors, providing a reliable basis for low-carbon optimization. Finally, a two-layer optimization model is constructed. The upper layer minimizes the total cost from the perspective of the entire system, comprehensively considering multiple factors to optimize unit operation. The lower layer guides users to adjust their strategies based on carbon emission costs from the perspective of individual users, minimizing electricity costs under constraints, increasing user enthusiasm, and promoting source-load interaction. The entire system, from data to model, is comprehensively coordinated, effectively promoting carbon emission reduction on the demand side of the power system, achieving a balance between economic and environmental benefits, and ensuring stable, scientific, accurate, and low-carbon system operation.

[0049] Additional aspects and advantages of this application will be set forth in the description which follows, and in some respects will become apparent from the description or may be learned by practice of this application. Attached Figure Description

[0050] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0051] Figure 1 This is a flowchart of a power grid optimization method for a power system according to an embodiment of this application;

[0052] Figure 2 This is a schematic diagram of the structure of a power grid optimization device for a power system according to an embodiment of this application;

[0053] Figure 3 This is a structural example diagram of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0054] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0055] The following describes a power grid optimization method and apparatus for a power system according to embodiments of this application, with reference to the accompanying drawings. Addressing the problem mentioned in the background art where dynamic carbon emission factors lack interaction with load in power system demand response, leading to significant errors in demand guidance and affecting the carbon reduction effect of the power system, this application provides a power grid optimization method for a power system. In this method, a dynamic emission factor dataset is generated through power system operation simulation and carbon emission flow calculation, providing a data foundation for subsequent operations. This dataset is then used to train a feedforward neural network model. The trained neural network is then transformed into mixed-integer programming constraints, thereby constructing a two-layer optimization model that considers users' low-carbon demand response behavior. This achieves carbon emission reduction on the demand side of electricity consumption. The entire process, from data generation and model training to building the optimization model, forms a complete logical system, which helps to effectively promote the achievement of carbon emission reduction targets on the demand side of the power system, aligning with the needs of low-carbon development. Therefore, this solves the problem in related technologies where the lack of interaction between dynamic carbon emission factors and load in power system demand response leads to significant errors in demand guidance, affecting the carbon reduction effect of the power system.

[0056] Specifically, Figure 1 This is a flowchart illustrating a power grid optimization method for a power system provided in an embodiment of this application.

[0057] like Figure 1 As shown, the power grid optimization method for this power system includes the following steps:

[0058] In step S101, a dataset of dynamic emission factors of the system is generated through power system operation simulation and carbon emission flow calculation.

[0059] Understandably, a dataset refers to a collection of dynamic emission factor data of the system obtained through power system operation simulation and carbon emission flow calculation. This data includes dynamic emission factor information of each node under different operating scenarios (such as, but not limited to, different system load curves, different unit combinations, etc.).

[0060] Optionally, in one embodiment of this application, generating a dataset of dynamic emission factors of the system through power system operation simulation and carbon emission flow calculation includes: generating the system load curve of the power system to obtain the result of economic dispatch through power system operation simulation; calculating the dynamic carbon emission factor of each node under the corresponding scenario using carbon emission flow theory; and obtaining a dynamic carbon emission factor dataset including massive power system operation scenarios based on the result of economic dispatch and the dynamic carbon emission factor of each node.

[0061] Specifically, the economic dispatch model of a power system, with the objective of minimizing the total system cost, is defined as follows:

[0062] min C = c T P G Δt,

[0063] Where c is the unit output cost of the generator set, and P G Δt represents the generator output vector, and Δt is the unit time interval.

[0064] The model must satisfy the upper and lower limits of unit output and power flow safety constraints, as shown in the following formula:

[0065]

[0066]

[0067] in, For generator sets, For the collection of transmission lines, and These are the upper and lower limits of the output of the i-th unit, respectively. and These represent the upper and lower limits of power flow for the first line. Based on the economic dispatch model, the output of each unit and the power flow of the lines are obtained.

[0068] Furthermore, the dynamic carbon emissions at each node in this scenario are calculated using carbon emission flow theory. The calculation formula is as follows:

[0069]

[0070] Among them, E N Let P be the dynamic carbon emission factor vector for each node. N Let P be the active flux matrix. B Inject a matrix into the node, P G E is the generator output vector. G This represents the unit carbon emission vector for the generator set.

[0071] Repeat the above process to generate a dynamic carbon emission factor dataset that includes massive power system operation scenarios.

[0072] The embodiments of this application can generate system load curves for operation simulation, obtain economic dispatch results with the goal of minimizing total cost under various constraints, and provide basic data for subsequent operations; calculate dynamic carbon emission factors using carbon emission flow theory; and generate massive scenario datasets through repeated operations, providing rich and diverse data for neural network training, which can promote the optimization of the power system from the source and achieve low-carbon and efficient operation.

[0073] In step S102, a feedforward neural network model is trained using the dataset to obtain the trained neural network.

[0074] Understandably, a feedforward neural network is a simple yet effective artificial neural network structure, characterized by a unidirectional flow of information from the input layer to the output layer, with no feedback connections in between.

[0075] In actual implementation, the system unit output and carbon emission intensity are used as input data, while the dynamic carbon emission factors of each node in the system are used as the desired output. The network contains multiple hidden layers, each consisting of several neurons, which are connected by weights. and paranoia They are interconnected and perform information transmission and computation (where l is the layer number, i is the neuron number of the hidden layer, and j is the neuron number of the previous layer).

[0076] Further, the feedforward neural network is trained using the dataset from step S101. The goal of training is to adjust the weights and biases in the network so that when the system's unit output and carbon emission intensity are input, the network can output the corresponding dynamic carbon emission factors for each node of the system as accurately as possible. During training, the backpropagation algorithm can be used to calculate the gradient, and the weights and biases can be updated according to the gradient descent method. Specifically, for each sample in the dataset, the input data is fed into the network, and the predicted dynamic carbon emission factor is calculated through forward propagation. Then, the predicted value is compared with the actual dynamic carbon emission factor to calculate the error. Next, the error can be propagated back from the output layer to the input layer using the backpropagation algorithm. The weights and biases of the connections between each neuron are adjusted based on the error to reduce the prediction error.

[0077] In some implementations, to improve training effectiveness and accelerate convergence, the ReLU (Rectified Linear Unit) function can be used as the activation function in the neural network. It is computationally simple and efficient, and effectively alleviates the vanishing gradient problem, allowing gradients to be propagated more efficiently during backpropagation, thus helping the neural network converge faster. Simultaneously, the ReLU function's output exhibits some sparsity; the output is 0 when the input is less than 0, which helps reduce computational cost and allows the network to learn more representative features.

[0078] Specifically, after multiple iterations of training, the training process ends when the network's prediction error reaches a certain convergence criterion or the number of training iterations reaches a preset value. At this point, the trained neural network is obtained, which contains the optimized and adjusted parameters of each hidden layer. and The hidden layer parameters store information learned by the network during training about the relationship between system unit output, unit carbon emission intensity and dynamic carbon emission factors of each node in the system. This allows the model to be used in subsequent power system demand response models to accurately predict the system dynamic carbon emission factors under different unit operating conditions, thus providing important support for achieving low-carbon optimized operation of the power system.

[0079] This application embodiment constructs a feedforward neural network by using the system unit output and carbon emission intensity as inputs and the dynamic carbon emission factors of each node as outputs. The weights and biases are adjusted using a dataset for training, and backpropagation and gradient descent are used for optimization. The ReLU activation function is combined to improve the training effect, and finally a trained neural network containing effective hidden layer parameters is obtained. This neural network can accurately capture the complex relationship between input and output, effectively predict the dynamic carbon emission factors under different unit states, provide key support for the low-carbon optimization operation of the power system, improve system operating efficiency and environmental protection, and is computationally efficient and converges quickly, which can reduce the amount of computation and enhance the model's feature learning ability.

[0080] In step S103, the trained neural network is transformed into a mixed integer programming constraint, and a two-layer optimization model that considers the user's low-carbon demand response behavior is constructed to carry out carbon emission reduction on the electricity demand side.

[0081] It is understood that the power system demand response model embedded in the carbon emission flow calculation in the embodiments of this application is applied to this model after the trained neural network is obtained.

[0082] In actual implementation, the trained neural network is transformed into a mixed integer programming constraint containing 0 and 1 variables, as follows:

[0083]

[0084]

[0085]

[0086]

[0087]

[0088] Where l is the layer number and i is the neuron number of that hidden layer. By setting the input of this constraint as the unit output and the output as the carbon emission factor, the piecewise linearized formula for calculating carbon emission flow can be obtained.

[0089] Furthermore, a two-layer demand response optimization model is established. The upper layer of the model optimizes unit operation from the perspective of the power system, while the lower layer optimizes the user's own demand response behavior from the user's perspective. The objective function of the upper layer optimization in the two-layer model is:

[0090]

[0091] Among them, c g This represents the unit output cost vector of the generator set. It is the unit's output vector at time t, Δt is the unit time interval, and c in the cost is... g This includes unit operating costs and carbon emission costs. Constraints are satisfied:

[0092]

[0093] in, This represents the carbon emission flow calculation formula extracted from the neural network.

[0094] The objective function of the lower-level optimization in the two-level optimization model is:

[0095]

[0096] Among them, c L,t P is the vector of unit electricity cost of load l at time t. Lt It is the power consumption vector of the load at time t, where Δt is the unit time interval. L,t The electricity cost for load L at time t mainly includes the basic electricity purchase cost and the carbon emission cost determined based on the dynamic carbon emission factor. The constraints that need to be satisfied are:

[0097]

[0098] in, This is a linear constraint condition for load type j. For example, if a data center batch processing load requires the computation task to be completed within a specified time (such as 12 hours), then the constraint can be expressed as the batch processing load received at time t=1 needs to be completed within the time period from t=1 to t=12, but the power consumption in each time period can be freely adjusted.

[0099] In some embodiments, since the lower-level optimization problem is a linear optimization, the KKT conditions can be used to transform the bi-level optimization problem into a single-level optimization problem for solution.

[0100] This application's embodiments can integrate neural networks into the power system demand response model through transformation, resulting in an accurate carbon emission flow calculation formula. The two-layer optimization model optimizes unit operation at the upper layer, considering operating and carbon emission costs, and combining the carbon emission flow formula with various constraints to achieve system-level optimization; the lower layer optimizes demand response from the user's perspective based on dynamic carbon emission factors, considering electricity costs and load type constraints to enhance user participation. Utilizing KKT conditional transformation for solution promotes source-load interaction, effectively achieving carbon emission reduction on the demand side and improving the low-carbon economic operation level of the power system.

[0101] According to the power grid optimization method for power systems proposed in this application, a dynamic emission factor dataset of the system can be generated through power system operation simulation and carbon emission flow calculation, providing a data foundation for subsequent operations. Then, a feedforward neural network model is trained using this dataset. After obtaining the trained neural network, it is transformed into mixed integer programming constraints, and then a two-layer optimization model considering the low-carbon demand response behavior of users is constructed. This achieves carbon emission reduction on the demand side of electricity consumption. The whole process, from data generation and model training to the construction of the optimization model, forms a complete logic, which helps to effectively promote the power system to achieve carbon emission reduction targets on the demand side of electricity consumption, in line with the needs of low-carbon development.

[0102] Next, the power grid optimization device for a power system proposed according to an embodiment of this application is described with reference to the accompanying drawings.

[0103] Figure 2 This is a block diagram of a power grid optimization device for a power system according to an embodiment of this application.

[0104] like Figure 2 As shown, the power grid optimization device 10 of the power system includes: a calculation module 100, a training module 200 and an optimization module 300.

[0105] Specifically, the calculation module 100 is used to generate a dataset of dynamic emission factors of the system through power system operation simulation and carbon emission flow calculation.

[0106] Training module 200 is used to train a feedforward neural network model using a dataset to obtain the trained neural network.

[0107] The optimization module 300 is used to transform the trained neural network into mixed integer programming constraints and construct a two-layer optimization model that considers users' low-carbon demand response behavior in order to carry out carbon emission reduction on the electricity demand side.

[0108] Optionally, in one embodiment of this application, the calculation module 100 includes: a simulation unit, a calculation unit, and a generation unit.

[0109] The simulation unit is used to generate the system load curve of the power system, so as to obtain the results of economic dispatch through power system operation simulation.

[0110] The computing unit is used to calculate the dynamic carbon emission factor of each node in the corresponding scenario using carbon emission flow theory.

[0111] The generation unit is used to obtain a dynamic carbon emission factor dataset that includes massive power system operation scenarios based on the results of economic scheduling and the dynamic carbon emission factors of each node.

[0112] Optionally, in one embodiment of this application, the economic dispatch model of the power system, with the objective of minimizing the total system cost, is defined as follows:

[0113] min C = c T P G Δt,

[0114] Where c is the unit output cost of the generator set, and P G Δt represents the generator output vector, and Δt is the unit time interval.

[0115] Optionally, in one embodiment of this application, the formula for calculating the dynamic carbon emission factor is:

[0116]

[0117] Among them, E N Let P be the dynamic carbon emission factor vector for each node. N Let P be the active flux matrix. B Inject a matrix into the node, P G E is the generator output vector. G This represents the unit carbon emission vector of the generator set.

[0118] Optionally, in one embodiment of this application, wherein,

[0119] The objective function of the upper-level optimization in the two-level optimization model is:

[0120]

[0121] Among them, c g This represents the unit output cost vector of the generator set. It is the power output vector of the unit at time t, where Δt is the unit time interval;

[0122] The objective function of the lower-level optimization in the two-level optimization model is:

[0123]

[0124] Among them, c L,t It is the vector of unit electricity cost of load l at time t. It is the power consumption vector of the load at time t, where Δt is the unit time interval.

[0125] It should be noted that the foregoing explanation of the power grid optimization method embodiment also applies to the power grid optimization device of the power system in this embodiment, and will not be repeated here.

[0126] According to the power grid optimization device for the power system proposed in this application, a dynamic emission factor dataset of the system can be generated through power system operation simulation and carbon emission flow calculation, providing a data foundation for subsequent operations. Then, a feedforward neural network model is trained using this dataset. After obtaining the trained neural network, it is transformed into mixed integer programming constraints, and then a two-layer optimization model considering the user's low-carbon demand response behavior is constructed. This achieves carbon emission reduction on the demand side of electricity consumption. The whole process, from data generation and model training to the construction of the optimization model, forms a complete logic, which helps to effectively promote the power system to achieve carbon emission reduction targets on the demand side of electricity consumption, in line with the needs of low-carbon development.

[0127] Figure 3 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include:

[0128] The memory 301, the processor 302, and the computer program stored on the memory 301 and capable of running on the processor 302.

[0129] When processor 302 executes the program, it implements the power grid optimization method for the power system provided in the above embodiments.

[0130] Furthermore, electronic devices also include:

[0131] Communication interface 303 is used for communication between memory 301 and processor 302.

[0132] The memory 301 is used to store computer programs that can run on the processor 302.

[0133] The memory 301 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0134] If the memory 301, processor 302, and communication interface 303 are implemented independently, then the communication interface 303, memory 301, and processor 302 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0135] Optionally, in a specific implementation, if the memory 301, processor 302, and communication interface 303 are integrated on a single chip, then the memory 301, processor 302, and communication interface 303 can communicate with each other through an internal interface.

[0136] Processor 302 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0137] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described power grid optimization method for a power system.

[0138] This application also provides a computer program product, including a computer program, which, when executed, is used to implement the above-described power grid optimization method for the power system.

[0139] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0140] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0141] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0142] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0143] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or more of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0144] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments.

[0145] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0146] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A power grid optimization method for a power system, characterized in that, Includes the following steps: A dataset of dynamic emission factors for the system is generated through power system operation simulation and carbon emission flow calculation; The feedforward neural network model is trained using the dataset to obtain the trained neural network; The trained neural network is transformed into a mixed integer programming constraint, and a two-layer optimization model that considers users' low-carbon demand response behavior is constructed to carry out carbon emission reduction on the electricity demand side. The dataset for generating system dynamic emission factors through power system operation simulation and carbon emission flow calculation includes: Generate the system load curve of the power system to obtain the results of economic dispatch through power system operation simulation; The dynamic carbon emission factor of each node under the corresponding scenario is calculated using carbon emission flow theory; Based on the results of the economic dispatch and the dynamic carbon emission factors of each node, a dynamic carbon emission factor dataset including power system operation scenarios is obtained. The mixed-integer programming constraints include: in, For the number of floors, The sequence number of the neurons in the hidden layer; The objective function of the upper-level optimization in the two-level optimization model is: ， in, This represents the unit output cost vector of the generator set. It is the power output vector of the unit at time t. The unit time interval; The objective function of the lower-level optimization in the two-level optimization model is: ， in, It is the vector of unit electricity cost of load l at time t. It is the power consumption vector of the load at time t. The unit time interval.

2. The method according to claim 1, characterized in that, The economic dispatch model of the power system, with the objective of minimizing the total system cost, is defined as follows: ， in, Cost per unit output of generator set This is the generator output vector. The unit time interval.

3. The method according to claim 1, characterized in that, The formula for calculating the dynamic carbon emission factor is as follows: ， in, This represents the dynamic carbon emission factor vector for each node. The active flux matrix, Inject a matrix into the node. This is the generator output vector. This represents the unit carbon emission vector of the generator set.

4. A power grid optimization device for a power system, characterized in that, include: The calculation module is used to generate a dataset of dynamic emission factors for the system through power system operation simulation and carbon emission flow calculation; The training module is used to train a feedforward neural network model using the dataset to obtain the trained neural network. The optimization module is used to transform the trained neural network into mixed integer programming constraints and construct a two-layer optimization model that considers users' low-carbon demand response behavior in order to carry out carbon emission reduction on the electricity demand side. The computing module includes: The simulation unit is used to generate the system load curve of the power system so as to obtain the results of economic dispatch through power system operation simulation; The computing unit is used to calculate the dynamic carbon emission factor of each node in the corresponding scenario using carbon emission flow theory; The generation unit is used to obtain the dynamic carbon emission factor dataset, which includes the power system operation scenario, based on the results of the economic dispatch and the dynamic carbon emission factors of each node. The mixed-integer programming constraints include: in, For the number of floors, The sequence number of the neurons in the hidden layer; The objective function of the upper-level optimization in the two-level optimization model is: ， in, This represents the unit output cost vector of the generator set. It is the power output vector of the unit at time t. The unit time interval; The objective function of the lower-level optimization in the two-level optimization model is: ， in, It is the vector of unit electricity cost of load l at time t. It is the power consumption vector of the load at time t. The unit time interval.

5. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the power grid optimization method for a power system as described in any one of claims 1-3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the power grid optimization method for the power system as described in any one of claims 1-3.

7. A computer program product, comprising a computer program, characterized in that, The computer program is executed to implement the power grid optimization method for the power system as described in any one of claims 1-3.

Images