Source network load storage coordinated carbon-energy coupling regulation and control system and method

Abstract

The invention discloses a source network load storage coordinated carbon-energy coupling regulation and control system and method, and relates to the technical field of intelligent energy regulation and control, and the method comprises the steps: collecting source network load storage multi-source data, carrying out the preprocessing, carrying out the feature extraction and time sequence prediction through a carbon-energy coupling prediction model, and generating a carbon-energy coupling prediction sequence; performing deviation comparison on the carbon-energy coupling prediction sequence and the preprocessed source network load storage multi-source data, monitoring the fluctuation of a carbon price signal, and when one of the deviation of the carbon-energy coupling prediction sequence and the fluctuation of the carbon price signal reaches a preset carbon-energy coupling deviation threshold, generating a recalculation request; on the basis of execution feedback, updating the carbon-energy coupling prediction model, and dynamically correcting the carbon-energy coupling control instruction set to obtain a carbon-energy optimization regulation and control scheme under source-network-load-storage cooperation; according to the method, distributed control is fused, and high-precision pre-judgment of a complex coupling relationship between source network load storage and carbon emission is realized through carbon-energy coupling prediction.

Description

Technical Field

[0001] This invention relates to distributed control, and more specifically to the field of intelligent energy regulation technology, and more specifically to a carbon-energy coupling regulation system and method for source-grid-load-storage coordination. Background Technology

[0002] With the advancement of global carbon neutrality goals, the green transformation of energy systems has become a research hotspot in academia and industry. In recent years, source-grid-load-storage coordinated regulation technology has gradually emerged, achieving efficient energy utilization and coordinated control of carbon emissions by coordinating the generation side, transmission and distribution network, load side, and energy storage system. Among these technologies, distributed control carbon-energy coupling prediction models combined with multi-source data feature extraction and time-series prediction techniques can accurately reflect the dynamic coupling relationship between power and carbon emissions in energy systems, providing a scientific basis for regulation strategies.

[0003] Existing technologies have certain limitations in terms of the accuracy of carbon emission prediction and the closed-loop mechanism for dynamic correction of control commands. These limitations are mainly reflected in insufficient real-time response capabilities, making it difficult to effectively adapt to rapidly changing carbon price fluctuations and load demands. Due to the multivariate time-varying characteristics and complex coupling relationships of carbon-energy coupled systems, existing prediction models cannot maintain high accuracy continuously, leading to the accumulation of prediction errors. This, in turn, affects the accuracy and timeliness of control commands. Furthermore, control commands lack dynamic correction capabilities and cannot be quickly adjusted based on execution feedback and real-time environmental changes, resulting in system lag and instability. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a source-grid-load-storage coordinated carbon-energy coupling regulation method to solve the problems of insufficient carbon emission prediction accuracy and lack of dynamic correction mechanism for regulation commands, which lead to system response lag and regulation instability.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a carbon-energy coupling regulation method for source-grid-load-storage coordination, comprising,

[0008] Multi-source data from sources, grid, load, and storage are collected, preprocessed, and then feature extraction and time-series prediction are performed using a carbon-energy coupling prediction model to generate a carbon-energy coupling prediction sequence.

[0009] The carbon-energy coupling prediction sequence is compared with the preprocessed multi-source data of source, grid, load and storage, and the fluctuation of carbon price signal is monitored. When either the carbon-energy coupling prediction sequence deviation or the carbon price signal fluctuation reaches the preset carbon-energy coupling deviation threshold, a recalculation request is generated.

[0010] Based on the recalculation request, the preprocessed multi-source data of source, grid, load and storage and the carbon-energy coupling prediction sequence are optimized and solved by the distributed scheduling method based on economic model prediction control to obtain the carbon-energy coupling control instruction set;

[0011] Through distributed communication channels and multi-path redundancy mechanisms, the carbon-energy coupling control instruction set is transmitted to the execution end, and execution feedback is collected in real time.

[0012] Based on execution feedback, the carbon-energy coupling prediction model is updated, and the carbon-energy coupling control instruction set is dynamically corrected to obtain an optimized carbon-energy regulation scheme under source-grid-load-storage coordination.

[0013] As a preferred embodiment of the carbon-energy coupling regulation method of source-grid-load-storage coordination described in this invention, the multi-source data of source-grid-load-storage includes source-side power generation, grid-side voltage and frequency, load-side electricity consumption behavior, and energy storage device status.

[0014] The preprocessing includes data cleaning, time synchronization, missing value imputation, and normalization.

[0015] As a preferred embodiment of the source-grid-load-storage coordinated carbon-energy coupling regulation method of the present invention, the specific steps for generating a carbon-energy coupling prediction sequence by performing feature extraction and time-series prediction through a carbon-energy coupling prediction model are as follows.

[0016] Using preprocessed historical source-grid-load-storage multi-source data, a long short-term memory network based on the fusion of multivariate time series modeling and attention mechanism is trained to obtain a carbon-energy coupling prediction model.

[0017] The preprocessed multi-source data from the source, grid, load, and storage are input into the carbon-energy coupling prediction model for feature extraction and multi-step time series prediction to obtain the carbon-energy coupling prediction sequence.

[0018] As a preferred embodiment of the source-grid-load-storage coordinated carbon-energy coupling regulation method of the present invention, the following steps are taken: The carbon-energy coupling prediction sequence is compared with the preprocessed source-grid-load-storage multi-source data to identify deviations, while simultaneously monitoring the fluctuations of the carbon price signal. When either the carbon-energy coupling prediction sequence deviation or the carbon price signal fluctuation reaches a preset carbon-energy coupling deviation threshold, a recalculation request is generated. The specific steps are as follows:

[0019] Real-time carbon price signals are acquired, and fluctuation amplitude is analyzed through a sliding time window to obtain carbon price signal fluctuations;

[0020] The carbon-energy coupling prediction sequence is matched with the preprocessed source-grid-load-storage multi-source data, and pairs of carbon-energy coupling prediction sequences and preprocessed source-grid-load-storage multi-source data at the same time point are extracted. Error analysis is performed to obtain the carbon-energy coupling prediction sequence deviation.

[0021] The carbon-energy coupling prediction sequence deviation and carbon price signal fluctuation are compared with a preset carbon-energy coupling deviation threshold. When one of them reaches the preset carbon-energy coupling deviation threshold, a recalculation request is generated.

[0022] As a preferred embodiment of the source-grid-load-storage coordinated carbon-energy coupling regulation method of the present invention, the step of optimizing and solving the preprocessed multi-source data of source-grid-load-storage and the carbon-energy coupling prediction sequence based on recalculation requests using a distributed scheduling method based on economic model predictive control to obtain a carbon-energy coupling control instruction set is as follows.

[0023] Using the recalculation request as a trigger signal, the distributed scheduling method based on economic model prediction control is driven to initialize the scheduling solution and obtain the initial state of the distributed scheduling problem solution.

[0024] The preprocessed multi-source data of source, grid, load and storage and the carbon-energy coupled prediction sequence are input into the distributed scheduling problem to solve the initial state, and a joint regulation and optimization problem of source, grid, load and storage is constructed.

[0025] By utilizing the joint regulation and optimization problem of source-grid-load-storage, a multi-objective optimization solution is performed on the preprocessed multi-source data of source-grid-load-storage and carbon-energy coupled prediction sequence to obtain the carbon-energy coupled control command set.

[0026] As a preferred embodiment of the source-grid-load-storage coordinated carbon-energy coupling control method of the present invention, the carbon-energy coupling control instruction set is transmitted to the execution end through a distributed communication channel and a multi-path redundancy mechanism, and execution feedback is collected in real time. The specific steps are as follows.

[0027] The carbon-energy coupling control instruction set is encrypted using a distributed communication channel encryption transmission mechanism to generate an encrypted control instruction set.

[0028] The encrypted control command set is input into a multi-path redundancy mechanism to build multiple independent transmission paths and transmit them to the execution end for execution, while the execution feedback generated by the execution operation is collected in real time.

[0029] As a preferred embodiment of the source-grid-load-storage coordinated carbon-energy coupling regulation method of the present invention, the steps of updating the carbon-energy coupling prediction model based on execution feedback and dynamically correcting the carbon-energy coupling control instruction set to obtain the optimized carbon-energy regulation scheme under source-grid-load-storage coordination are as follows.

[0030] Based on execution feedback, the parameters of the carbon-energy coupling prediction model are adjusted and optimized, and the carbon-energy coupling prediction sequence is regenerated.

[0031] Based on the regenerated carbon-energy coupling prediction sequence and execution feedback, the carbon-energy coupling control instruction set is dynamically corrected to obtain a carbon-energy optimization control scheme under source-grid-load-storage coordination.

[0032] Secondly, this invention provides a source-grid-load-storage coordinated carbon-energy coupling regulation system, comprising,

[0033] The prediction module is used to collect multi-source data from sources, grids, loads, and storage, perform preprocessing, and extract features and make time-series predictions through a carbon-energy coupling prediction model to generate a carbon-energy coupling prediction sequence.

[0034] The request module is used to compare the deviation of the carbon-energy coupling prediction sequence with the preprocessed multi-source data of source, grid, load and storage, and monitor the fluctuation of carbon price signal. When either the deviation of the carbon-energy coupling prediction sequence or the fluctuation of carbon price signal reaches the preset carbon-energy coupling deviation threshold, a recalculation request is generated.

[0035] The instruction module is used to optimize and solve the preprocessed multi-source data of source-grid-load-storage and carbon-energy coupling prediction sequence based on recalculation requests and through a distributed scheduling method based on economic model predictive control, so as to obtain a carbon-energy coupling control instruction set.

[0036] The feedback module is used to transmit the carbon-energy coupling control instruction set to the execution end through a distributed communication channel and a multi-path redundancy mechanism, and to collect execution feedback in real time.

[0037] The optimization module is used to update the carbon-energy coupling prediction model based on execution feedback and dynamically correct the carbon-energy coupling control instruction set to obtain a carbon-energy optimization control scheme under the coordination of source, grid, load and storage.

[0038] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the source-grid-load-storage coordinated carbon-energy coupling regulation method as described in the first aspect of the present invention.

[0039] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the source-grid-load-storage coordinated carbon-energy coupling regulation method as described in the first aspect of the present invention.

[0040] The beneficial effects of this invention are as follows: By using a carbon-energy coupled prediction model for feature extraction and time-series prediction, it achieves high-precision prediction of the complex coupling relationship between power generation output, grid-side status, load demand, energy storage behavior, and carbon emissions, providing a reliable reference for regulation and control decisions; by executing feedback-driven adaptive updates of the carbon-energy coupled prediction model and dynamic correction of control commands, a real-time closed-loop adaptive mechanism is established to reduce prediction errors and execution deviations and continuously optimize scheduling schemes; ultimately, it improves the ability to respond quickly to sudden load fluctuations and carbon price fluctuations, enhances grid operation stability, reduces energy dispatch and carbon trading costs, and achieves synergistic optimization of economic benefits and carbon emission reduction. Attached Figure Description

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 The flowchart shows a carbon-energy coupling regulation method for source-grid-load-storage synergy.

[0043] Figure 2 This is a flowchart of a carbon-energy coupling regulation system that integrates source, grid, load, and storage.

[0044] Figure 3 A flowchart for generating a recalculation request.

[0045] Figure 4 A flowchart is provided to obtain a carbon-energy optimization control scheme under the coordinated operation of source, grid, load and storage. Detailed Implementation

[0046] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0047] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0048] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0049] Reference Figures 1-4 To provide a distributed control method for the present invention, a carbon-energy coupling regulation method with source-grid-load-storage coordination is provided, comprising the following steps:

[0050] S1. Collect and preprocess multi-source data from sources, grids, loads, and storage.

[0051] S1.1, The multi-source data of source, grid, load and storage includes source-side power generation, grid-side voltage and frequency, load-side electricity consumption behavior and energy storage device status.

[0052] Specifically, smart meters and power sensors are deployed on the source side to collect real-time power generation from photovoltaic, wind power, and gas-fired combined heat and power equipment to obtain source-side power generation; voltage and current monitoring devices are installed on the grid side to obtain voltage and frequency data from each transmission node to obtain grid-side voltage and frequency; on the load side, user-side smart electricity terminals and load identification modules are used to collect electricity consumption behavior characteristics to obtain load-side electricity consumption behavior; and battery management units (BMS) are set up at energy storage devices to collect charging and discharging status, power levels, and operating efficiency, achieving comprehensive and synchronous collection of multi-source data from the source, grid, load, and storage sides to obtain the status of energy storage devices.

[0053] S1.2 Preprocessing includes data cleaning, time synchronization, missing value imputation, and normalization.

[0054] Specifically, data cleaning includes: deleting records of abnormal zero values, negative values, and power exceeding the rated power in the source-side power generation; removing sampling points with continuous flat values ​​or sudden jumps in the grid-side voltage and frequency; removing electricity consumption curves with continuous missing values ​​or many periodic breaks in the load-side electricity consumption behavior; and clearing records of energy storage device states that have frequent state switching and a duration of less than 1 minute (as shown in the example).

[0055] Time synchronization: The power generation on the source side, the voltage and frequency on the grid side, the electricity consumption behavior on the load side, and the status of energy storage devices are synchronized according to the timestamp. The reference time axis is used to resample at equal intervals with a step size of 10 seconds, and the non-aligned time points are supplemented by linear interpolation.

[0056] Missing values ​​are imputed in the multi-source data of source, grid, load and storage using a linear interpolation method based on the mean of previous and subsequent times. When the missing values ​​are continuously missing for more than three sample periods, the mean of the same time point in the same typical daily sample is used for replacement.

[0057] Normalization processing: The source-side power generation, grid-side voltage frequency, load-side electricity consumption behavior and energy storage device status after time synchronization and missing value filling are normalized respectively. The maximum and minimum value normalization method is adopted. For example, the maximum and minimum values ​​of the monitoring indicators in the past 30 days are selected as the upper and lower limits of normalization. After normalization processing, all indicator values ​​are limited to the interval [0,1].

[0058] S2. Feature extraction and time-series prediction are performed using a carbon-energy coupling prediction model to generate a carbon-energy coupling prediction sequence.

[0059] S2.1. Using preprocessed historical source-grid-load-storage multi-source data, a long short-term memory network based on the fusion of multivariate time series modeling and attention mechanism is trained to obtain a carbon-energy coupling prediction model.

[0060] Specifically, by utilizing four features from the preprocessed historical multi-source data of source-grid-load-storage, namely, power generation on the source side, voltage and frequency on the grid side, electricity consumption behavior on the load side, and status of energy storage equipment, an exemplary 24-step sliding window method is used to extract the input sequence and the output label of the next time step.

[0061] Based on four features, two layers of long short-term memory network are stacked. The first layer of long short-term memory network has 64 memory units, and the second layer of long short-term memory network has 32 memory units. An attention mechanism is applied to the output of the second layer of long short-term memory network at all time steps to calculate a weighted average, generate a context vector, and concatenate the context vector with the output of the second layer of long short-term memory network at the last time step. The result is then input into a fully connected layer to output the prediction result.

[0062] The mean squared error is used as the loss function. The network is trained with the Adam optimization algorithm with an exemplary learning rate of 0.001. The learning is done in batches of 32 with an exemplary batch size. The validation loss is monitored using an exemplary 20% validation set. An exemplary early stopping strategy with no improvement for 5 consecutive rounds is triggered. Finally, the long short-term memory network weights with the lowest validation loss are saved to obtain the carbon-energy coupling prediction model.

[0063] S2.2 Input the preprocessed multi-source data of source, grid, load and storage into the carbon-energy coupling prediction model for feature extraction and multi-step time series prediction to obtain the carbon-energy coupling prediction sequence.

[0064] Specifically, the preprocessed multi-source data from the source, grid, load, and storage are converted into input matrices in batches according to an exemplary 24-step sliding window and input into the carbon-energy coupling prediction model; in the first layer of the long short-term memory network, the hidden state sequence of each batch of input matrices is calculated step by step and output.

[0065] The hidden state sequence of the first long short-term memory (LSM) network layer is input into the second LSM network layer to iteratively update the hidden state. An attention mechanism is applied to the hidden state sequence of the second LSM network layer at all time steps to calculate a weighted average and generate a context vector. The context vector is concatenated with the hidden state of the second LSM network layer at the last time step and then input into a fully connected layer for forward computation to output the predicted power and carbon emissions for the next time step. The above operation is repeated until the prediction for multiple future time steps is completed, and all predicted values ​​are summarized to form a carbon-energy coupled prediction sequence.

[0066] It should be noted that the expression for progressively calculating and outputting the hidden state sequence at each time step for each batch of input matrices in the first Long Short-Term Memory (LSTM) network layer is as follows:

[0067]

[0068] Among them, [h1,h2,…,h T [] is a hidden state sequence of length T, h T h1 is the hidden state vector at the T-th time step, h2 is the hidden state vector at the first time step, and h3 is the hidden state vector at the second time step. t It is the output gate vector at time step t, ⊙ is the element-wise multiplication operator, f t It is the forgetting gate vector at time step t, c t-1 It is the cell state vector at time step t-1, c t It is the cell state vector at time step t, i t It is the input gate vector at time step t. Let be the candidate cell state vector at time step t, where T is the total number of time steps in the hidden state sequence, and t is the index of the current time step.

[0069] S3. Compare the carbon-energy coupling prediction sequence with the preprocessed source-grid-load-storage multi-source data, and monitor the fluctuation of carbon price signal. When either the carbon-energy coupling prediction sequence deviation or the carbon price signal fluctuation reaches the preset carbon-energy coupling deviation threshold, generate a recalculation request.

[0070] S3.1. Acquire carbon price signals in real time and analyze the fluctuation amplitude through a sliding time window to obtain carbon price signal fluctuations.

[0071] Specifically, the system calls the carbon trading market data interface to obtain the latest carbon price signals, for example, parsing the carbon price signals returned by the interface every 60 seconds; it configures the sliding time window length to an example of 5 minutes, reads the carbon price signals arranged in ascending order of time within the window in sequence, and calculates the difference between the maximum and minimum values ​​as the carbon price signal fluctuation.

[0072] S3.2 Match the carbon-energy coupling prediction sequence with the preprocessed source-grid-load-storage multi-source data, extract the carbon-energy coupling prediction sequence and the preprocessed source-grid-load-storage multi-source data pairs at the same time point, perform error analysis, and obtain the carbon-energy coupling prediction sequence deviation.

[0073] Specifically, the carbon-energy coupling prediction sequence and the preprocessed source-grid-load-storage multi-source data are read and matched according to the timestamp field. Each carbon-energy coupling prediction value and the corresponding original value of the preprocessed source-grid-load-storage multi-source data are extracted at the same time point. The absolute difference is calculated for each pair of carbon-energy coupling prediction values ​​and the original values ​​of the preprocessed source-grid-load-storage multi-source data. All calculation results are statistically summarized, and the carbon-energy coupling prediction sequence deviation is output.

[0074] S3.4. Compare the carbon-energy coupling prediction sequence deviation and carbon price signal fluctuation with the preset carbon-energy coupling deviation threshold respectively. When one of them reaches the preset carbon-energy coupling deviation threshold, generate a recalculation request.

[0075] Specifically, the process of setting the carbon-energy coupling prediction sequence deviation and carbon price signal fluctuation is as follows: Select any 30 consecutive time periods for the carbon-energy coupling prediction sequence deviation and carbon price signal fluctuation, calculate the 95th percentile values ​​of the carbon-energy coupling prediction sequence deviation and carbon price signal fluctuation, and use the smaller 95th percentile value as an example carbon-energy coupling deviation threshold, with an example value of 0.05 yuan / kWh; after the end of each sliding time window period, read the carbon-energy coupling prediction sequence deviation and carbon price signal fluctuation in the current period, and compare whether either of the above carbon-energy coupling prediction sequence deviation and carbon price signal fluctuation values ​​exceeds the set carbon-energy coupling deviation threshold; if any value exceeds the example carbon-energy coupling deviation threshold, immediately output a recalculation request at the current moment.

[0076] S4. Based on the recalculation request, the preprocessed multi-source data of source, grid, load and storage and the carbon-energy coupling prediction sequence are optimized and solved by the distributed scheduling method based on economic model prediction control to obtain the carbon-energy coupling control instruction set.

[0077] S4.1 Using the recalculation request as a trigger signal, drive the distributed scheduling method based on economic model prediction control to perform scheduling solution initialization, and obtain the initial state of the distributed scheduling problem solution.

[0078] Specifically, it receives recalculation requests, reads the carbon-energy coupling prediction sequence of the previous period within the current time window period, and writes it into the input sequence of the distributed scheduling method based on economic model prediction control in chronological order.

[0079] The connection relationships, transmission power limits, and marginal transmission costs between various energy nodes (photovoltaic power generation, wind power generation, natural gas cogeneration, energy storage, electrical load, heat load, and cooling load) are read synchronously and used as constraint parameters of the energy node topology network structure. These parameters are then written into the distributed scheduling method based on economic model predictive control to construct the joint regulation and optimization problem for the current cycle. The optimization variables and constraint structure are initialized, and the scheduling solution process is initiated.

[0080] Read the scheduling decision results from the previous cycle, extract the termination values ​​of each scheduling variable (including the power generation of energy nodes, the charging and discharging power of energy storage, and the load adjustment) at the end of the previous cycle, use them as the initial scheduling variable values ​​for the current solution cycle, and write them to the corresponding scheduling variable positions; set the scheduling cycle length and scheduling time step, for example, a total cycle of 60 minutes and a single step time interval of 5 minutes, calculate that one cycle contains 12 time steps, read the current timestamp and mark it as the start time of the distributed scheduling solution, so as to obtain the initial state of the distributed scheduling problem solution.

[0081] S4.2. Input the preprocessed multi-source data of source, grid, load and storage and the carbon-energy coupled prediction sequence into the distributed scheduling problem to solve the initial state, and construct the joint regulation and optimization problem of source, grid, load and storage.

[0082] Specifically, the preprocessed multi-source data (source, grid, load, and storage) and carbon-energy coupling prediction sequences are read, aligned to the time dimension by the timestamp field, and sorted in chronological order for all time points. These are then sequentially written into the input data structure of the initial state for solving the distributed scheduling problem. Based on the energy nodes (photovoltaic power generation, wind power generation, natural gas cogeneration, energy storage, electrical load, thermal load, and cooling load) and operating parameters, the feature values ​​corresponding to the multi-source data (source-side power generation, grid-side voltage and frequency, load-side electricity consumption behavior, and energy storage device status) are filled into the data based on the economic model. In the constraints and objective function of the joint regulation optimization problem in the distributed scheduling method of predictive control (constraints include energy balance, operation restrictions, power transmission restrictions and carbon emission control, and objective function includes minimizing the overall operating cost and total carbon emissions to achieve the goal of low-carbon and high-efficiency operation), the predicted electricity load value and the corresponding carbon emission factor of the energy type at each time step are extracted from the carbon-energy coupled prediction sequence and filled into the load demand field and carbon emission parameter field of each time step in the joint regulation optimization problem in chronological order to ensure that the prediction parameters in the joint regulation optimization problem correspond to the time series input data.

[0083] Based on the exemplary scheduling cycle length and time step, the time intervals covered by the preprocessed multi-source data of source, grid, load and storage and the carbon-energy coupled prediction sequence are adjusted to match the scheduling solution time window, thus obtaining the joint regulation and optimization problem of source, grid, load and storage.

[0084] S4.3. Using the source-grid-load-storage joint regulation and optimization problem, multi-objective optimization is performed on the preprocessed source-grid-load-storage multi-source data and carbon-energy coupling prediction sequence to obtain the carbon-energy coupling control command set.

[0085] Specifically, the objective function and constraints in the joint regulation and optimization problem of source-grid-load-storage are read, and a distributed scheduling method based on economic model predictive control is called. The multi-source data of source-grid-load-storage at different time steps and the carbon-energy coupled prediction sequence are input into the scheduling solver through a time-series rolling method. In each scheduling cycle, based on constraints such as energy balance, operation restrictions, power transmission restrictions and carbon emission control, the upper and lower limits of power generation on the source side, the power flow transmission capacity on the grid side, the charging and discharging range of energy storage, the boundary of electrical load, the range of heat load demand and the boundary of cooling load demand are set.

[0086] The objective function is set as a weighted sum of total carbon emissions and operating costs, with the weighting factors set as follows: operating cost weight 0.7 and carbon emission factor weight 0.3. Using the rolling optimization mechanism in the distributed scheduling method based on economic model predictive control, the joint regulation and optimization problem of source-grid-load-storage in all scheduling cycles is solved sequentially. The scheduling variables corresponding to the optimal solution of the objective function in each scheduling cycle are used to extract power generation, energy storage charging and discharging plan, grid power flow, various load allocation schemes and carbon emission distribution sequences as a carbon-energy coupling control instruction set.

[0087] S5. Through distributed communication channels and multi-path redundancy mechanisms, the carbon-energy coupling control instruction set is transmitted to the execution end, and execution feedback is collected in real time.

[0088] S5.1. The carbon-energy coupling control instruction set is encrypted through the encrypted transmission mechanism of the distributed communication channel to generate an encrypted control instruction set.

[0089] Specifically, the carbon-energy coupling control instruction set corresponding to each time step within the scheduling cycle is formatted into a structured instruction data packet. This structured instruction data packet includes a scheduling timestamp, source-side power generation instructions, energy storage charging and discharging instructions, grid power flow instructions, load allocation instructions, heat load supply instructions, cooling load supply instructions, and carbon emission control instructions. To ensure data security and integrity during transmission, the structured instruction data packet is encrypted using an asymmetric encryption method, such as elliptic curve cryptography. Before encryption, a fixed-length digest is generated from the structured instruction data packet using a hash digest function, serving as an integrity check code; an example hash algorithm is SHA-256. After concatenating the digest with the structured instruction data packet, asymmetric encryption is performed using a public key pre-allocated at the receiving end to form the encrypted control instruction set.

[0090] S5.2 Input the encrypted control instruction set into the multi-path redundancy mechanism to build multiple independent transmission paths and transmit them to the execution end for execution, and collect the execution feedback generated by the execution operation in real time.

[0091] Specifically, the encrypted control command set is input into the multi-path redundancy mechanism, the physical address information and available bandwidth parameters of the communication nodes are extracted based on the distributed communication channel topology, and the average round-trip time and packet loss rate of the candidate paths are calculated according to the maximum transmission distance, minimum signal attenuation and current load level of the communication path. The communication path set is obtained by selecting paths that simultaneously meet the criteria of an average round-trip time of less than 10 milliseconds and a packet loss rate of less than 1%.

[0092] The redundancy level of each path in the communication path set is obtained. The redundancy level is comprehensively evaluated based on the number of backup nodes, the number of independent links, and fault tolerance. For example, a level 3 redundancy level requires at least 3 backup nodes or 3 independent links and can tolerate the failure of two links. The paths are sorted from high to low redundancy level. If the redundancy levels are the same, the average round-trip time is compared. If the time is similar, the packet loss rate is compared. The path with lower time and packet loss rate is selected first. Finally, the paths with a redundancy level of not less than level 3 are selected to form an independent transmission path set. Each independent transmission path is assigned a unique identifier.

[0093] The encrypted control command set is divided into multiple data fragments bit by bit. Each data fragment is attached with a path identifier and a fragment sequence number label, and each is sent to its corresponding independent transmission path for parallel transmission. At the receiving end, all received data fragments are reassembled according to their fragment sequence numbers, duplicate fragments are removed, and verification is performed to ensure that the reassembled encrypted control command set is consistent with the original encrypted control command set. The reassembled encrypted control command set is submitted to the execution end, which retrieves source-side power generation commands, energy storage charging and discharging commands, grid power flow commands, electrical load allocation commands, heat load supply commands, cooling load supply commands, and carbon emission control commands based on the scheduling timestamps in the structured command data packet, and calls the corresponding execution equipment to perform various operations. The execution feedback of the equipment is collected through monitoring terminals deployed on each execution equipment.

[0094] S6. Based on execution feedback, the carbon-energy coupling prediction model is updated, and the carbon-energy coupling control instruction set is dynamically corrected to obtain a carbon-energy optimization control scheme under source-grid-load-storage coordination.

[0095] S6.1 Based on execution feedback, the parameters of the carbon-energy coupling prediction model are adjusted and optimized, and the carbon-energy coupling prediction sequence is regenerated.

[0096] Specifically, the process involves reading the running status parameters and corresponding timestamps contained in the execution feedback, organizing the correspondence between the running status parameters and the input variables of the preset carbon-energy coupling prediction model according to the time series, comparing the deviation between the output of the carbon-energy coupling prediction model and the running status parameters, and gradually adjusting the key parameters in the carbon-energy coupling prediction model, such as the prediction weight coefficient and the time window length, using gradient descent or Bayesian optimization methods. For example, if the prediction error decreases by less than 0.01, it is determined whether the adjustment stopping condition is met.

[0097] The preprocessed real-time multi-source data of source, grid, load and storage is used as input to the updated carbon-energy coupling prediction model. The updated weight coefficients and time window length parameters are used to extract features and make time-series predictions on the historical input data, and the carbon-energy coupling prediction sequence is regenerated.

[0098] S6.1 Based on the regenerated carbon-energy coupling prediction sequence and execution feedback, the carbon-energy coupling control instruction set is dynamically corrected to obtain a carbon-energy optimization control scheme under source-grid-load-storage coordination.

[0099] Specifically, based on the regenerated carbon-energy coupling prediction sequence and execution feedback, the predicted load and carbon emission parameters corresponding to each time step, as well as the operating status parameters fed back during the actual execution process, are extracted. The difference between the predicted values ​​and the actual feedback values ​​is compared to obtain the error index. According to the error index, the power generation command, energy storage charging and discharging command, grid power flow command, electricity load allocation command, heat load supply command, cooling load supply command, and carbon emission control command in the original carbon-energy coupling control command set are adjusted using a weighted adjustment strategy. The exemplary adjustment range does not exceed 10%. It is then determined whether the adjustment meets the stability condition. If not, iterative correction continues until the adjustment range meets the requirement that the exemplary adjustment range does not exceed 10%. The adjusted carbon-energy coupling control command set is then updated to form a carbon-energy optimized regulation scheme under the coordination of source, grid, load, and storage.

[0100] This embodiment also provides a source-grid-load-storage coordinated carbon-energy coupling regulation system, including:

[0101] The prediction module is used to collect multi-source data from sources, grids, loads, and storage, perform preprocessing, and extract features and make time-series predictions through a carbon-energy coupling prediction model to generate a carbon-energy coupling prediction sequence.

[0102] The request module is used to compare the deviation of the carbon-energy coupling prediction sequence with the preprocessed multi-source data of source, grid, load and storage, and monitor the fluctuation of carbon price signal. When either the deviation of the carbon-energy coupling prediction sequence or the fluctuation of carbon price signal reaches the preset carbon-energy coupling deviation threshold, a recalculation request is generated.

[0103] The instruction module is used to optimize and solve the preprocessed multi-source data of source-grid-load-storage and carbon-energy coupling prediction sequence based on recalculation requests and through a distributed scheduling method based on economic model predictive control, so as to obtain a carbon-energy coupling control instruction set.

[0104] The feedback module is used to transmit the carbon-energy coupling control instruction set to the execution end through a distributed communication channel and a multi-path redundancy mechanism, and to collect execution feedback in real time.

[0105] The optimization module is used to update the carbon-energy coupling prediction model based on execution feedback and dynamically correct the carbon-energy coupling control instruction set to obtain a carbon-energy optimization control scheme under the coordination of source, grid, load and storage.

[0106] This embodiment also provides a computer device applicable to the source-grid-load-storage coordinated carbon-energy coupling regulation method, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the source-grid-load-storage coordinated carbon-energy coupling regulation method proposed in the above embodiment.

[0107] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0108] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the carbon-energy coupling control method for source-grid-load-storage coordination as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0109] In summary, this invention achieves high-precision prediction of the complex coupling relationship between power generation output, grid-side status, load demand, energy storage behavior, and carbon emissions by using a carbon-energy coupled prediction model for feature extraction and time-series prediction, providing a reliable reference for regulation and decision-making. Furthermore, by implementing feedback-driven adaptive updates of the carbon-energy coupled prediction model and dynamic correction of control commands, a real-time closed-loop adaptive mechanism is established to reduce prediction errors and execution deviations and continuously optimize scheduling schemes. Ultimately, this invention enhances the ability to respond quickly to sudden load fluctuations and carbon price fluctuations, strengthens grid operation stability, reduces energy dispatch and carbon trading costs, and achieves synergistic optimization of economic benefits and carbon emission reduction.

[0110] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A carbon-energy coupling regulation method for source-grid-load-storage synergy, characterized in that: The application relates to a carbon-energy coupling optimization method and system. The carbon-energy coupling prediction sequence is compared with the pretreated source-grid-load-storage multi-source data, and the fluctuation of the carbon price signal is monitored, and when one of the carbon-energy coupling prediction sequence deviation and the carbon price signal fluctuation reaches a preset carbon-energy coupling deviation threshold, a recalculation request is generated. Based on the recalculation request, the pretreated source-grid-load-storage multi-source data and the carbon-energy coupling prediction sequence are optimized and solved by using a distributed scheduling method based on economic model predictive control to obtain a carbon-energy coupling control instruction set. The carbon-energy coupling control instruction set is transmitted to an execution end through a distributed communication channel and a multi-path redundancy mechanism, and execution feedback is collected in real time. Based on the execution feedback, the carbon-energy coupling prediction model is updated, and the carbon-energy coupling control instruction set is dynamically corrected to obtain a carbon-energy optimization regulation scheme under the source-grid-load-storage cooperation. The source-grid-load-storage multi-source data includes source-side power generation, grid-side voltage frequency, load-side power consumption behavior and storage equipment state.

2. The method of claim 1, wherein the source-sink-load storage coordination carbon-energy coupling regulation method is characterized by: The pretreatment includes data cleaning, time synchronization, missing value filling and normalization processing. The carbon-energy coupling prediction sequence is generated through the carbon-energy coupling prediction model, and the specific steps are as follows, 3. The method of claim 2, wherein the source-sink-load storage coordination carbon-energy coupling regulation method is characterized by: The pretreated historical source-grid-load-storage multi-source data is used to train a long short-term memory network based on multivariate time series modeling and attention mechanism fusion to obtain a carbon-energy coupling prediction model. The pretreated source-grid-load-storage multi-source data is input into the carbon-energy coupling prediction model for feature extraction and multi-step time series prediction to obtain a carbon-energy coupling prediction sequence. The carbon-energy coupling prediction sequence is compared with the pretreated source-grid-load-storage multi-source data, and the fluctuation of the carbon price signal is monitored, and when one of the carbon-energy coupling prediction sequence deviation and the carbon price signal fluctuation reaches a preset carbon-energy coupling deviation threshold, a recalculation request is generated.

4. The source-network-load-storage coordinated carbon-energy coupling regulation method of claim 3, wherein: The carbon price signal is obtained in real time, and the fluctuation amplitude is analyzed through a sliding time window to obtain the carbon price signal fluctuation. The carbon-energy coupling prediction sequence is compared with the pretreated source-grid-load-storage multi-source data, and the fluctuation of the carbon price signal is monitored, and when one of the carbon-energy coupling prediction sequence deviation and the carbon price signal fluctuation reaches a preset carbon-energy coupling deviation threshold, a recalculation request is generated. The carbon-energy coupling prediction sequence deviation and the carbon price signal fluctuation are compared with the preset carbon-energy coupling deviation threshold, and when one of them reaches the preset carbon-energy coupling deviation threshold, a recalculation request is generated. The recalculation request is used as a trigger signal to drive the distributed scheduling method based on the economic model predictive control to initialize scheduling and solving to obtain an initial state of distributed scheduling problem solving.

5. The method of claim 4, wherein the source-sink-load storage coordination carbon-energy coupling regulation method is characterized by: The recalculation request is used as a trigger signal to drive the distributed scheduling method based on the economic model predictive control to initialize scheduling and solving to obtain an initial state of distributed scheduling problem solving. ​ The pre-processed source grid storage multi-source data and the carbon-energy coupling prediction sequence are input into an initial state of a distributed scheduling problem to construct a source grid storage joint regulation optimization problem; The pre-processed source grid storage multi-source data and the carbon-energy coupling prediction sequence are subjected to multi-objective optimization solving by using the source grid storage joint regulation optimization problem to obtain a carbon-energy coupling control instruction set.

6. The source-network-load-storage coordinated carbon-energy coupling regulation method of claim 5, wherein: The carbon-energy coupling control instruction set is transmitted to an execution end through a distributed communication channel and a multi-path redundancy mechanism, and execution feedback is collected in real time, and the specific steps are as follows, The carbon-energy coupling control instruction set is subjected to encryption processing through an encryption transmission mechanism of the distributed communication channel to generate an encrypted control instruction set; The encrypted control instruction set is input into the multi-path redundancy mechanism to construct multiple independent transmission paths and is transmitted to the execution end for execution, and execution feedback generated by the execution operation is collected in real time.

7. The source-network-load-storage coordinated carbon-energy coupling regulation method of claim 6, wherein: Based on the execution feedback, the carbon-energy coupling prediction model is updated, and the carbon-energy coupling control instruction set is dynamically corrected to obtain a carbon-energy optimization regulation scheme under the cooperation of the source grid storage, and the specific steps are as follows, Based on the execution feedback, the carbon-energy coupling prediction model is subjected to parameter adjustment and optimization, and a carbon-energy coupling prediction sequence is regenerated; Based on the regenerated carbon-energy coupling prediction sequence and the execution feedback, the carbon-energy coupling control instruction set is dynamically corrected to obtain a carbon-energy optimization regulation scheme under the cooperation of the source grid storage.

8. A source-grid-load-storage collaborative carbon-energy coupling regulation system based on the source-grid-load-storage collaborative carbon-energy coupling regulation method of any one of claims 1-7. It comprises, A prediction module is configured to collect source grid storage multi-source data for preprocessing, and perform feature extraction and time series prediction through a carbon-energy coupling prediction model to generate a carbon-energy coupling prediction sequence; A request module is configured to compare the carbon-energy coupling prediction sequence with the pre-processed source grid storage multi-source data for deviation, and monitor fluctuations in carbon price signals, and generate a recalculation request when one of the carbon-energy coupling prediction sequence deviation and the carbon price signal fluctuation reaches a preset carbon-energy coupling deviation threshold; An instruction module is configured to perform optimization solving on the pre-processed source grid storage multi-source data and the carbon-energy coupling prediction sequence based on the recalculation request through a distributed scheduling method based on an economic model prediction control to obtain a carbon-energy coupling control instruction set; A feedback module is configured to transmit the carbon-energy coupling control instruction set to an execution end through a distributed communication channel and a multi-path redundancy mechanism, and collect execution feedback in real time; An optimization module is configured to update the carbon-energy coupling prediction model based on the execution feedback, and dynamically correct the carbon-energy coupling control instruction set to obtain a carbon-energy optimization regulation scheme under the cooperation of the source grid storage. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to realize the steps of the source grid storage collaborative carbon-energy coupling regulation method of any one of claims 1-7.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the source grid storage collaborative carbon-energy coupling regulation method of any one of claims 1-7.

Images