A day-ahead low-carbon dispatching method for distribution networks based on prediction optimization fusion learning

Abstract

A method for day-ahead low-carbon dispatch of distribution networks based on predictive optimization fusion learning includes: collecting and cleaning multi-source data through data acquisition and preprocessing steps to form a training dataset; predicting node carbon potential using a sequential neural network model that employs a Seq2Seq structure and GRU units to process time series data; constructing a low-carbon dispatch model for adjustable loads to minimize carbon emissions and meet grid operation constraints; integrating predictive optimization fusion learning with model training to achieve coordinated optimization of forecasting and dispatch decisions by constructing a hybrid decision loss function and optimizing it using a backpropagation algorithm; generating an optimized dispatch strategy for adjustable loads, and ensuring that it meets actual needs through parameter tuning, thereby achieving low-carbon and efficient dispatch of the distribution network. This method, based on predictive optimization fusion learning, achieves low-carbon emissions and efficient operation of the power system.

Description

Technical Field

[0001] The present invention relates to distribution network dispatching technology, and in particular to a day-ahead low-carbon dispatching method for distribution networks based on prediction optimization fusion learning. Background Art

[0002] Traditional low-carbon dispatch methods primarily adopt a "prediction first, optimization later" framework. This involves first predicting load, power generation, and carbon emissions based on historical data and statistical models, then developing dispatch plans using mathematical optimization techniques to minimize carbon emissions while ensuring power supply security and economic efficiency. The significance of this approach lies in achieving a balance between system operating indicators and carbon reduction targets by establishing predictive models and optimization algorithms. However, this phased approach fails to fully consider the impact of prediction errors on the final dispatch decision, which can easily lead to decision bias. Existing technologies are prone to prediction errors and inaccurate dispatch due to model limitations when dealing with the uncertainties of renewable energy and load fluctuations. Traditional dispatch methods typically rely on open-loop prediction and optimization strategies, making it difficult to achieve optimal carbon emission control in complex distribution network environments. With the integration of renewable energy and the increasing complexity of the system, how to comprehensively consider prediction errors and decision optimization in the dispatch process has become a major technical challenge that needs to be addressed in low-carbon dispatch.

[0003] In traditional low-carbon scheduling methods, the forecasting and scheduling optimization phases often operate independently. This phased approach fails to achieve information sharing and collaborative feedback. Forecasting models typically rely solely on historical data and statistical models when forecasting load, power generation, and carbon emissions, failing to fully incorporate lower-level decision-making requirements and actual operating conditions. This results in significant deviations at critical local moments, even when the overall forecast error is low. This separation not only makes it difficult to directly reflect forecast results in optimization decisions, but also magnifies small forecast errors during subsequent optimization, leading to significant deviations between the final scheduling plan and actual demand, thus impacting the economic and reliability of system operation. Furthermore, with the large-scale integration of renewable energy and the increasing complexity of load demand, traditional background technologies are struggling to manage uncertainty. Most existing forecasting methods rely on historical data statistics and fixed model parameters. They often struggle to capture the dynamic characteristics of renewable energy generation (such as wind and solar power), which is influenced by multiple factors such as climate conditions and equipment performance, as well as the random variations in user load. When faced with extreme climate or emergencies, the reliability and accuracy of the prediction results of this method are greatly reduced, resulting in the inability of scheduling decisions to adapt to on-site operating conditions in a timely manner, affecting the overall effect of the system's low-carbon scheduling.

[0004] It should be noted that the information disclosed in the above background technology section is only used to understand the background of this application, and therefore may include information that does not constitute prior art known to ordinary technicians in this field. Summary of the Invention

[0005] The main purpose of the present invention is to overcome the defects existing in the above-mentioned background technology and provide a day-ahead low-carbon scheduling method for distribution network based on prediction optimization fusion learning.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] A method for day-ahead low-carbon dispatching of distribution networks based on predictive optimization fusion learning includes the following steps:

[0008] S1. Data Collection and Preprocessing: Collect multi-source power system data, including historical load data, renewable energy output data, meteorological data, and carbon emissions data, and preprocess this data to form a high-quality training dataset;

[0009] S2. Sequential Neural Network Node Carbon Potential Prediction: This method utilizes a sequence-to-sequence (Seq2Seq) model to accurately predict the future carbon emission intensity of nodes. This method uses an encoder-decoder structure to process multi-source data, manage complex temporal dependencies, and employs gated recurrent units (GRUs) as computational units.

[0010] S3. Construction of an Adjustable Load Low-Carbon Dispatch Model: Build an adjustable load low-carbon dispatch model to minimize system operation carbon emissions while meeting constraints such as power balance, equipment operation limits, and congestion.

[0011] S4. Prediction Optimization Fusion Learning and Model Training: This approach combines prediction error and decision error to construct a hybrid decision loss function. This is then used through neural network training to optimize the overall low-carbon scheduling strategy. Backpropagation is used to perform gradient descent optimization on the hybrid decision loss function to reduce both prediction and decision errors. The weight coefficients are dynamically adjusted based on actual system requirements to improve prediction accuracy and decision effectiveness.

[0012] S5. Generate an optimized scheduling strategy for adjustable loads: Generate an optimized scheduling strategy for adjustable loads based on the predicted node carbon potential and the adjustable load low-carbon scheduling model, check whether the scheduling decision meets the requirements, and perform parameter tuning until the requirements are met.

[0013] S6. Result acquisition: Obtain the optimal scheduling strategy for adjustable loads and the predicted node carbon potential to achieve low-carbon scheduling of the distribution network on the day before.

[0014] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the day-ahead low-carbon dispatching method for a distribution network based on predictive optimization fusion learning.

[0015] A computer program product includes a computer program, which, when executed by a processor, implements the method for day-ahead low-carbon dispatching of a distribution network based on predictive optimization fusion learning.

[0016] The present invention has the following beneficial effects:

[0017] The present invention proposes a day-ahead low-carbon dispatching method for distribution networks based on predictive optimization fusion learning. The predictive optimization fusion learning method effectively solves the dispatching decision deviation problem caused by prediction error transmission in the traditional "predict first, optimize later" framework. The prediction error and decision error are collaboratively optimized by constructing a hybrid decision loss function. The sequence-to-sequence model (Seq2Seq) and the gated recurrent unit (GRU) are combined to model multi-source time series data, accurately capture the complex dynamic characteristics of node carbon potential, and significantly improve the accuracy of carbon emission prediction. At the same time, the dynamic weight coefficient is used to balance the prediction and decision goals, and the adjustable load low-carbon dispatching model is combined to perform multi-dimensional collaborative optimization of power balance, flow constraints and operating boundaries, so as to minimize carbon emissions while ensuring the safe and stable operation of the power grid. The real-time feedback mechanism dynamically adapts to the fluctuations of new energy and load uncertainties, enhances the robustness and responsiveness of the system in a complex distribution network environment, and provides an innovative solution for building a low-carbon smart grid with high-precision prediction, adaptive optimization and multi-constraint collaboration.

[0018] The prediction optimization fusion learning method in the present invention effectively solves the defect that the prediction error has a great influence on the decision results in the traditional prediction-first-then-optimization framework, and provides a dynamic adaptive, accurate and efficient decision optimization solution for the low-carbon dispatch of power systems, which has significant technological advancement and broad application prospects.

[0019] Other beneficial effects of the embodiments of the present invention will be further described below. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is an overall flow chart of a method for low-carbon day-ahead dispatching of distribution networks based on prediction optimization fusion learning according to an embodiment of the present invention.

[0021] Figure 2 This is a flowchart of a day-ahead low-carbon scheduling framework based on prediction optimization fusion learning according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The following is a detailed description of the embodiments of the present invention. It should be emphasized that the following description is only exemplary and is not intended to limit the scope of the present invention and its application.

[0023] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0024] See Figure 1 and Figure 2 The embodiment of the present invention provides a method for day-ahead low-carbon dispatching of a distribution network based on prediction optimization fusion learning, comprising the following steps:

[0025] Step S1. Data collection and preprocessing: Collect multi-source data from the power system, including historical load data, renewable energy output data, meteorological data, and carbon emission data, and perform preprocessing such as cleaning, normalization, and feature extraction on the data to form a high-quality training dataset;

[0026] Step S2. Sequential neural network node carbon potential prediction: Utilize the node carbon potential prediction method based on the sequence-to-sequence (Seq2Seq) model, process multi-source data through the encoder-decoder structure, manage complex time dependencies, and use the gated recurrent unit (GRU) as the computing unit to achieve accurate prediction of the future carbon emission intensity of the node.

[0027] In a preferred embodiment, step S2 employs a prediction method based on a sequence-to-sequence (Seq2Seq) model. The model comprises an encoder and decoder structure. The encoder extracts features from the input multi-source data and generates a fixed-length context vector. The decoder generates a node carbon potential prediction sequence for the future time period based on the context vector. The model, using a gated recurrent unit (GRU) as a computational unit, effectively captures long- and short-term dependencies in time series data and simulates the nonlinear relationships between different predictive factors. Specifically, the encoder processes the input sequence by updating and resetting gate vectors to generate candidate hidden states, and then updates the current hidden state based on the updated gate vector and the candidate hidden states, ultimately generating a context vector. The decoder initializes the hidden state using the context vector and gradually generates an output sequence. The output sequence is then subjected to a probability distribution calculation using a softmax function to ultimately generate a node carbon potential prediction result.

[0028] More specifically, the encoder's processing can include the following steps: dynamically controlling the retention and forgetting of historical information in the input sequence through update gates and reset gates to generate update gate vectors and reset gate vectors; calculating candidate hidden states based on the current input, reset gate vectors, and the previous hidden state; fusing the candidate hidden states with the previous hidden state using the update gate vectors, and updating the current hidden state time-step by time-step; using the hidden state at the final time-step as a context vector to convey the global features of the input sequence to the decoder. The decoder's processing can include the following steps: initializing the decoder's hidden state based on the context vector generated by the encoder; updating the current hidden state time-step by combining the previous hidden state, the previous prediction output, and the context vector; mapping the current hidden state to an output probability distribution and generating a prediction result for the current time-step by maximizing the probability; iteratively generating a complete output sequence to achieve time-series prediction of node carbon potential.

[0029] Step S3. Construction of an adjustable load low-carbon dispatch model: Construct an adjustable load low-carbon dispatch model with the goal of minimizing the system operation carbon emissions while meeting the constraints of power balance, equipment operation restrictions, and congestion of the power grid.

[0030] In some embodiments, in step S3, the objective function is to minimize the carbon emissions of the distribution network, and dynamic optimization scheduling is achieved through the following constraints: a) adjustable load constraints, ensuring that the adjusted load power meets power invariance, adjustment unidirectionality, adjustment upper and lower bounds, and time series continuity; b) flow constraints, using a linearized model to characterize the active and reactive power injection relationship of the grid bus; c) power balance constraints, ensuring the supply and demand balance of power generation, power purchase, new energy grid connection and load power in the distribution network; d) operation boundary constraints, limiting branch flow, voltage amplitude, phase angle and generator output range to meet the requirements of safe and stable operation of the power grid. In a further embodiment, the adjustable load constraint optimizes load distribution by adjusting the load power translation, and limits the cumulative deviation of power adjustment, the mutual exclusivity of adjustment directions, and the power change rate in adjacent time periods; the flow constraint constructs a linearized active and reactive power injection relationship based on line impedance parameters and bus voltage amplitude and phase angle; the operating boundary constraint includes upper and lower limit constraints on branch flow, voltage amplitude, phase angle, thermal power unit output, and new energy grid-connected power to ensure that the system operates within a safe threshold.

[0031] Step S4. Prediction optimization fusion learning and model training: Combine the prediction error and decision error to construct a hybrid decision loss function, and achieve overall optimization of the low-carbon scheduling strategy through neural network training. Use the backpropagation algorithm to perform gradient descent optimization on the hybrid decision loss function to reduce the errors in both prediction and decision making. Dynamically adjust the weight coefficient according to actual system requirements to improve prediction accuracy and decision-making effect;

[0032] In a preferred embodiment, in step S4, the prediction error and the decision error are weightedly fused by constructing a hybrid decision loss function, the hybrid decision loss function includes a prediction error term and a decision error term, and the loss function is gradient-descent optimized by a back-propagation algorithm to reduce the errors in both prediction and decision-making. The weight coefficient is dynamically adjusted according to the actual system requirements to improve the prediction accuracy and decision-making effect. Specifically, the collaborative optimization of prediction and scheduling decisions is achieved through the following mechanisms: a) constructing a hybrid decision loss function, weighting and fusion the prediction error term and the decision error term according to a dynamic weight coefficient, the prediction error term represents the deviation between the predicted value and the actual value of the node carbon potential, and the decision error term represents the deviation between the scheduling decision based on the predicted value and the optimal decision based on the actual value; b) using the back-propagation algorithm to perform gradient-descent optimization on the hybrid decision loss function, synchronously updating the parameters of the prediction model and the scheduling decision model, and gradually reducing the prediction error and decision error; c) dynamically adjusting the weight coefficient according to system requirements to balance the optimization goals of prediction accuracy and decision effect. More preferably, the decision error term is gradient solvable by constructing a substitute function, where the substitute function is constructed based on the difference between the optimal decision and the actual decision, and the nonlinear decision error is converted into a differentiable form through a maximization operation to support backpropagation training of the neural network.

[0033] Step S5. Generate an optimized scheduling strategy for adjustable loads: Generate an optimized scheduling strategy for adjustable loads based on the predicted node carbon potential and the adjustable load low-carbon scheduling model, check whether the scheduling decision meets the requirements, and perform parameter tuning until it meets the requirements.

[0034] Step S6. Result acquisition: Obtain the adjustable load optimization scheduling strategy and predicted node carbon potential to achieve day-ahead low-carbon scheduling of the distribution network.

[0035] The following further describes an algorithm example and experimental verification of a specific embodiment of the present invention.

[0036] The present invention proposes a day-ahead low-carbon dispatching method for distribution networks based on predictive optimization fusion learning. Due to the dual uncertainty of source and load of new energy access on the power generation side and flexible resources on the load side, the trend of node carbon potential has complex characteristics. Considering the lower-level adjustable load dispatching model, a reasonable arrangement and optimization scheduling strategy is made for the factors affecting the carbon emission trajectory, and a carbon potential prediction model for the lower-level low-carbon dispatching nodes is constructed. Combined with a sequential neural network to capture the carbon emission trend, low-carbon dispatching based on predictive optimization fusion learning is realized on the day-ahead basis.

[0037] The method specifically comprises the following steps:

[0038] 1. Establish a node carbon potential prediction model based on sequential neural network to achieve day-ahead node carbon potential output.

[0039] Nodal carbon potential refers to the carbon emission intensity of each node in the power system, reflecting the carbon emissions associated with the transmission and use of electricity at a specific node. Accurately predicting nodal carbon potential is crucial for developing low-carbon dispatch strategies, helping to optimize power resource allocation and reduce overall carbon emissions.

[0040] To this end, a node carbon potential prediction method based on the sequence-to-sequence (Seq2Seq) model was established. By collecting multi-source data in the power system and forming input features, it can accurately predict the future carbon emission intensity of each node, thereby providing a reliable data basis for low-carbon scheduling decisions.

[0041] According to one embodiment of the present invention, a Seq2Seq model with an encoder-decoder structure is used to process multi-source data such as historical load data, renewable energy output data, meteorological data, and carbon emission data. It is capable of managing complex time dependencies, effectively integrating heterogeneous input variables, and simulating nonlinear relationships between different predictive factors. To demonstrate the relationship between variables, the general form of the prediction model is:

[0042]

[0043] Among them, the future node carbon potential is determined by the prediction model generate; is the selected input feature sequence, with a length of ; is the generated output sequence, with length .

[0044] Specifically, the prediction model consists of an encoder and a decoder. The encoder extracts features from the input sequence and converts it into a fixed-length context vector. The decoder, based on this context vector, generates a predicted output sequence describing the carbon potential of a node in the future. This model structure fully captures the long-term and short-term dependencies in time series data, overcoming the limitations of traditional methods when dealing with inconsistent input and output sequence lengths. This allows for accurate prediction of node carbon potential. The implementation uses a gated recurrent unit (GRU) as the computational unit.

[0045] In the encoder, the input sequence Converted to hidden state The update process of the calculation unit is as follows:

[0046]

[0047] in, is the update gate vector, is the reset gate vector; is a candidate hidden state. 、 、 is the input weight matrix; 、 、 is the weight matrix of the hidden state; 、 、 is the bias vector. represents Hadamard multiplication; Is a sigmoid activation function; is the hyperbolic tangent activation function. The final hidden state of the encoder is Often used as context vectors .

[0048] In the decoder, the decoder decodes the context vector into a length of The output sequence The decoder uses the context vector to initialize the hidden state and then gradually generates the output sequence. The calculation formula is as follows:

[0049]

[0050] Among them, the decoder GRU uses the previous hidden state , previous output and context vector At each time step Update its hidden state. is the weight matrix of the output layer; Is the bias vector of the output layer. Final node carbon potential prediction Depend on express.

[0051] In the present invention, data collection includes comprehensive collection of historical loads, renewable energy generation, meteorological conditions and carbon emission records in the power system, followed by cleaning, normalization and feature extraction of the data to construct a high-quality training dataset.

[0052] 2. Establish an adjustable load low-carbon scheduling model and optimize the load scheduling strategy based on predicted carbon potential.

[0053] A low-carbon dispatch model for adjustable loads dynamically optimizes the dispatch of adjustable load resources in the power system, minimizing carbon emissions while meeting the grid's requirements for safe, stable, and economical operation. This model fully leverages the control capabilities of adjustable loads, rationally allocating power demand and optimizing the overall load distribution of the system, thereby reducing fossil energy consumption and carbon emissions, with significant environmental and economic benefits.

[0054] In one embodiment of the present invention, adjustable loads are defined as power-consuming devices and energy storage devices that can actively adjust their power consumption within a certain range based on dispatch instructions. First, the basic parameters of all adjustable loads in the power system (including control range requirements) are comprehensively collected, and a low-carbon dispatch optimization model is constructed. Its objective function is to minimize system operating carbon emissions while satisfying constraints such as power balance, equipment operating limits, and congestion.

[0055] To achieve the above goals, the present invention constructs an adjustable load optimization scheduling model with minimizing carbon emissions as the objective function, and its formula is as follows:

[0056]

[0057] in, represents the carbon emissions of the distribution network, is the total number of distribution network nodes, To predict the nodal carbon potential, is the adjustable load power after adjustment.

[0058] The adjustable load optimization scheduling model is composed of the following constraints, including adjustable load constraints, power flow constraints, power balance constraints, and operation boundary constraints, as follows:

[0059] (a) Adjustable load constraints

[0060] The adjustable load represents the power relationship of the adjustable load before and after adjustment, ensuring the invariance of the adjustable load power, the unidirectionality of the adjustment, the upper and lower limits of the adjustment, and the continuity of the time sequence, as follows:

[0061]

[0062]

[0063] in, is the adjustable load power before adjustment, Allowable power deviation, Is the adjustment set Single load adjustment action in It is an adjustment parameter that constrains the upper and lower bounds of load adjustment. On the one hand, it limits the power deviation and alleviates the mismatch of user demand. On the other hand, it also limits the user's tolerance. It is the maximum power variation to avoid step or sudden changes.

[0064] (b) Power flow constraints

[0065] The power flow constraint is expressed using a linearized power flow model, as follows:

[0066]

[0067] in, , Busbar Linear expressions for active and reactive power injection. , is the voltage amplitude and voltage phase angle; is the line impedance coefficient.

[0068] (c) Power balance constraints

[0069]

[0070] in, Purchase electricity from the main grid for distribution network operators, The amount of electricity connected to the grid by new energy sources, The power generation of thermal power units. is the collection of inflow and outflow nodes of the busbar.

[0071] (d) Operational boundary constraints

[0072]

[0073] Among them, upper and lower limit constraints on branch currents, voltage phase angles, generator sets and new energy units are stipulated.

[0074] 3. Through the prediction optimization fusion learning method, the carbon potential prediction model for the lower-level low-carbon scheduling nodes is trained to realize day-ahead low-carbon scheduling based on prediction optimization fusion learning.

[0075] This method, based on a fusion learning approach based on forecast optimization, combines forecast error and decision error to construct a hybrid decision loss function, achieving overall optimization of low-carbon dispatch strategies through neural network training. This method addresses the problem of forecast error propagation in the traditional "forecast first, optimize later" process. By introducing decision error feedback during the model training phase, it achieves coordinated optimization of forecasting and dispatch decisions, thereby improving the accuracy and adaptability of power grid dispatch solutions.

[0076] In one embodiment of the present invention, the historical data, real-time data and external environmental variables in the power system are first collected, cleaned and normalized to form a high-quality training data set. Then, a node carbon potential prediction method based on a sequence-to-sequence (Seq2Seq) model is used to make a preliminary prediction of the node carbon potential, and the prediction error is calculated. Subsequently, the prediction result is input into an adjustable load optimization scheduling model constructed with minimizing carbon emissions as the objective function. The preliminary scheduling plan generated based on the optimization algorithm is compared with the preset target scheduling result to calculate the scheduling decision error. Based on the above two errors, the present invention constructs a hybrid decision loss function, which is composed of a weighted sum of a prediction error term and a decision error term according to a certain weight, wherein the weight coefficient can be dynamically adjusted according to the actual system requirements, so as to take into account both prediction accuracy and decision effect during the training process.

[0077] First, define the decision error as described above. The decision error is obtained by the difference between the optimal decision based on the observed value and the actual decision based on the predicted value, as shown below:

[0078]

[0079] During neural network training, a backpropagation algorithm is used to perform gradient descent optimization on the hybrid decision-making loss function, continuously updating model parameters and gradually reducing both prediction and decision-making errors. After training, the neural network not only outputs highly accurate predictions but also provides real-time feedback and correction of prediction errors during scheduling decisions, ensuring the robustness and cost-effectiveness of the low-carbon scheduling strategy in actual operation.

[0080] The decision error is unsolvable in gradient descent and is replaced by a class decision error. The calculation method is as follows:

[0081]

[0082] Secondly, for the traditional mean square error loss function, its expression is as follows:

[0083]

[0084] The hybrid decision loss function is a weighted sum of the above-mentioned class decision error loss function and the mean square error loss function, and its expression is as follows:

[0085]

[0086] in, is the decision mixing coefficient.

[0087] In summary, the prediction optimization fusion learning method of the present invention effectively solves the defect that the prediction error has a great impact on the decision results in the traditional prediction-first-then-optimization framework, and provides a dynamic adaptive, accurate and efficient decision optimization solution for the low-carbon scheduling of power systems, which has significant technological advancement and broad application prospects.

[0088] Based on the above process, a comprehensive low-carbon scheduling framework based on forecast optimization and fusion learning is proposed, such as Figure 2 shown.

[0089] This invention proposes a method for day-ahead low-carbon dispatch of distribution networks based on predictive optimization and fusion learning. By employing this method, the invention achieves high-precision prediction of key parameters in the power system and coordinated optimization of dispatch decisions. Specifically, the invention constructs a hybrid decision loss function that combines prediction error and decision error according to certain weights and utilizes neural network training to effectively reduce dispatch errors caused by prediction bias in traditional "predict first, optimize later" methods. Compared with existing technologies, the node carbon potential prediction method based on the sequence-to-sequence model of the present invention demonstrates significant advantages in capturing complex time series dynamics, providing more accurate and timely data support for low-carbon dispatch decisions, and improving the accuracy and robustness of low-carbon dispatch in the power system. Through a real-time feedback mechanism, the model can dynamically adapt to changes in system operating conditions and constraints, ensuring high response speed and robustness even in complex operating environments with large-scale new energy integration and large load fluctuations. This provides reliable data support and technical guarantees for the safe and economical operation of the power grid. Furthermore, due to its wide applicability, the invention can be promoted and applied in various power systems and dispatch scenarios, thus providing an innovative and efficient solution for building a green, low-carbon smart grid.

[0090] An embodiment of the present invention further provides a storage medium for storing a computer program, which at least performs the above method when executed.

[0091] An embodiment of the present invention further provides a control device, comprising a processor and a storage medium for storing a computer program; wherein the processor is configured to execute at least the method described above when executing the computer program.

[0092] An embodiment of the present invention further provides a processor, which executes a computer program and at least performs the method described above.

[0093] The storage medium can be implemented by any type of non-volatile storage device, or a combination thereof. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); the magnetic surface memory can be a magnetic disk memory or a magnetic tape memory. The storage medium described in the embodiments of the present invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0094] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0095] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0096] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0097] Those skilled in the art will appreciate that all or part of the steps of the above-mentioned method embodiments may be implemented by hardware associated with program instructions, and the aforementioned program may be stored in a computer-readable storage medium. When the program is executed, the program executes the steps of the above-mentioned method embodiments. The aforementioned storage medium includes various media that can store program codes, such as mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0098] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0099] The methods disclosed in the several method embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new method embodiments.

[0100] The features disclosed in several product embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new product embodiments.

[0101] The features disclosed in several method or device embodiments provided by the present invention can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0102] The above is a further detailed description of the present invention in conjunction with specific preferred embodiments, and the specific implementation of the present invention should not be considered to be limited to these descriptions. Those skilled in the art will recognize that, without departing from the scope of the present invention, several equivalent substitutions or obvious variations can be made, and the performance or use of the same should be considered to fall within the scope of protection of the present invention.

Claims

1. A method for low-carbon dispatching of distribution network based on prediction optimization fusion learning, characterized by: The following steps are involved: S1. Data Acquisition and Preprocessing: Collect multi-source power system data, including historical load data, renewable energy output data, meteorological data, and carbon emissions data, and preprocess this data to form a high-quality training dataset; S2. Sequential Neural Network Node Carbon Potential Prediction: This approach utilizes a Sequence-to-Sequence (Seq2Seq) model to accurately predict the future carbon emission intensity of nodes. This approach uses an encoder-decoder architecture to process multi-source data, manage complex temporal dependencies, and employs a gated recurrent unit (GRU) as the computational unit. S3. Construction of an Adjustable Load Low-Carbon Dispatch Model: Develop an adjustable load low-carbon dispatch model to minimize system operating carbon emissions while satisfying grid power balance, equipment operating restrictions, and congestion constraints. S4. Prediction Optimization Fusion Learning and Model Training: This approach combines prediction error and decision error to construct a hybrid decision loss function. This is then used through neural network training to optimize the overall low-carbon scheduling strategy. Backpropagation is used to perform gradient descent optimization on the hybrid decision loss function to reduce both prediction and decision errors. The weight coefficients are dynamically adjusted based on actual system requirements to improve prediction accuracy and decision effectiveness. S5. Generate an optimized scheduling strategy for adjustable loads: Based on the predicted node carbon potential and the adjustable load low-carbon scheduling model, generate an optimized scheduling strategy for adjustable loads. Check whether the scheduling decision meets the requirements and perform parameter tuning until the requirements are met. S6. Results Acquisition: Obtain an optimal scheduling strategy for adjustable loads and predict node carbon potential to achieve day-ahead low-carbon scheduling of the distribution network. In step S4, the prediction error and the decision error are weightedly fused by constructing a hybrid decision loss function, wherein the hybrid decision loss function includes a prediction error term and a decision error term, and the loss function is gradient-descent optimized by a back-propagation algorithm to reduce the errors in both prediction and decision. The weight coefficient is dynamically adjusted according to the actual system requirements to improve the prediction accuracy and decision effect; In step S4, the collaborative optimization of prediction and scheduling decisions is achieved through the following mechanisms: a) Constructing a hybrid decision loss function, weightedly fusing the prediction error term and the decision error term according to a dynamic weight coefficient. The prediction error term represents the deviation between the predicted value and the actual value of the node carbon potential, and the decision error term represents the deviation between the scheduling decision based on the predicted value and the optimal decision based on the actual value. b) performing gradient descent optimization on the hybrid decision loss function using a backpropagation algorithm, synchronously updating the parameters of the prediction model and the scheduling decision model, and gradually reducing the prediction error and the decision error; c) dynamically adjusting the weight coefficients according to system requirements to balance the optimization goals of prediction accuracy and decision-making effect; The decision error term is gradient-solvable by constructing a surrogate function based on the difference between the optimal decision and the actual decision. The surrogate function is then converted into a differentiable form, i.e., a quasi-decision error loss function, through a maximization operation to support backpropagation training of the neural network. The expression of the hybrid decision loss function is as follows: ; in, is the class decision error loss function, is the mean square error loss function, that is, the prediction error term, is the decision mixing coefficient.

2. The method for low-carbon dispatching of distribution network based on prediction optimization fusion learning according to claim 1 is characterized in that: In step S2, a prediction method based on a sequence-to-sequence (Seq2Seq) model is adopted. The model includes an encoder and a decoder structure. The encoder is used to extract features from the input multi-source data and generate a context vector of fixed length. The decoder generates a node carbon potential prediction sequence in the future time period based on the context vector. The model uses a gated recurrent unit (GRU) as a computing unit, which can effectively capture the long-term and short-term dependencies in time series data and simulate the nonlinear relationship between different prediction factors.

3. The method for low-carbon dispatching of distribution network based on prediction optimization fusion learning according to claim 2 is characterized in that: The encoder processes the input sequence by updating the gate vector and resetting the gate vector to generate candidate hidden states, and updates the current hidden state based on the update gate vector and the candidate hidden state, and finally generates a context vector; the decoder initializes the hidden state through the context vector and gradually generates an output sequence, and the output sequence is calculated by the softmax function for probability distribution, and finally generates the node carbon potential prediction result.

4. The method for low-carbon dispatching of distribution network based on prediction optimization fusion learning according to claim 1 is characterized in that: In step S3, the objective function is to minimize the carbon emissions of the distribution network, and dynamic optimization scheduling is achieved through the following constraints: a) Adjustable load constraints ensure that the adjusted load power meets power invariance, adjustment unidirectionality, adjustment upper and lower bounds, and time sequence continuity; b) Power flow constraints, using a linearized model to characterize the relationship between active and reactive power injection into the grid bus; c) Power balance constraints to ensure the balance of power supply and demand within the distribution network, including power generation, power purchase, new energy grid connection, and load power; d) Operational boundary constraints, which limit branch currents, voltage amplitudes, phase angles, and generator output ranges to meet the requirements for safe and stable operation of the power grid.

5. The method for day-ahead low-carbon dispatching of distribution networks based on prediction optimization fusion learning according to claim 4 is characterized in that: The adjustable load constraint optimizes load distribution by adjusting the load power translation and limits the cumulative deviation of power adjustment, the mutual exclusivity of adjustment directions, and the power change rate in adjacent time periods; The power flow constraint is based on the line impedance parameters and the bus voltage amplitude and phase angle to construct a linear active and reactive power injection relationship; The operating boundary constraints include upper and lower limit constraints on branch currents, voltage amplitudes, phase angles, thermal power unit outputs, and renewable energy grid-connected power to ensure that the system operates within a safe threshold.

6. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for day-ahead low-carbon dispatching of a distribution network based on predictive optimization fusion learning as described in any one of claims 1 to 5 is implemented.

7. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for day-ahead low-carbon dispatching of a distribution network based on predictive optimization fusion learning as described in any one of claims 1 to 5 is implemented.

Images