Methods and apparatus for optimizing day-ahead dispatching plans of integrated energy bases

Abstract

This application provides a method and apparatus for optimizing day-ahead dispatch plans of an integrated energy base. The method includes: acquiring day-ahead dispatch plan data and corresponding day-ahead power forecast data of the integrated energy base; inputting the day-ahead dispatch plan data and day-ahead power forecast data into a day-ahead dispatch plan optimization model, so that the day-ahead dispatch plan optimization model outputs day-ahead dispatch plan optimization data corresponding to the day-ahead dispatch plan data; the day-ahead dispatch plan optimization model is pre-trained based on historical day-ahead power forecast data of the integrated energy base, historical day-ahead dispatch plan data corresponding to the historical day-ahead power forecast data, and historical actual power output data, respectively. This application can effectively and reliably optimize day-ahead dispatch plans and improve the rationality and effectiveness of energy optimization allocation for the integrated energy base based on the optimization results, thereby enhancing the stability and safety of the integrated energy base operation.

Description

Technical Field

[0001] This application relates to the field of energy dispatching technology, and in particular to a method and apparatus for optimizing day-ahead dispatching plans for integrated energy bases. Background Technology

[0002] With the large-scale grid connection of renewable energy sources, some regions are constructing large-scale integrated energy bases to further enhance renewable energy transmission capacity and improve renewable energy consumption levels. These bases utilize ultra-high-voltage direct current (UHVDC) projects spanning multiple regions and provinces, bundling the power output of wind, solar, and thermal power before transmitting it via DC, thereby achieving multi-energy complementarity and cross-regional renewable energy consumption. Under the constraint of DC transmission capacity limitations, it is necessary to rationally allocate the output ratio of wind, solar, and thermal power to achieve a balance between operational economy and system security.

[0003] Currently, traditional integrated energy base dispatching methods only consider information such as day-ahead power forecast data and equipment status data. Based on the power forecast results, the output range of wind farms, photovoltaic power plants, and thermal power plants is determined as a constraint. A wind-solar-thermal power output allocation model is established, and the wind-solar-thermal ratio is optimized by setting an optimization objective function (such as economic indicators) to formulate an allocation plan. However, in reality, due to the uncertainty of wind and solar power forecasts, the final output ratio often deviates from the day-ahead plan. The information contained in this deviation is not effectively utilized in traditional methods, which affects the effectiveness and reliability of day-ahead dispatching plans and thus fails to improve the safe and stable operation capability of integrated energy bases. Summary of the Invention

[0004] In view of this, embodiments of this application provide a method and apparatus for optimizing the day-ahead scheduling plan of an integrated energy base, so as to eliminate or improve one or more defects existing in the prior art.

[0005] The first aspect of this application provides a method for optimizing day-ahead dispatching plans for an integrated energy base, comprising:

[0006] Obtain day-ahead dispatch plan data and corresponding day-ahead power forecast data for the integrated energy base;

[0007] The daytime scheduling plan data and the daytime power prediction data are input into the daytime scheduling plan optimization model so that the daytime scheduling plan optimization model outputs daytime scheduling plan optimization data corresponding to the daytime scheduling plan data.

[0008] The day-ahead scheduling optimization model is pre-trained based on the historical day-ahead power prediction data of the integrated energy base, the historical day-ahead scheduling plan data corresponding to the historical day-ahead power prediction data, and the historical actual power output data.

[0009] In some embodiments of this application, before inputting the day-ahead scheduling data and the day-ahead power prediction data into the day-ahead scheduling optimization model, the method further includes:

[0010] A dataset is generated based on the historical power forecast data of the integrated energy base, the historical daytime power output plan data and the historical actual power output data corresponding to the historical power forecast data, respectively.

[0011] A pre-defined deep neural network is trained using the dataset to obtain a day-ahead scheduling optimization model that outputs day-ahead scheduling optimization data based on the day-ahead scheduling plan data and corresponding power prediction data of the integrated energy base.

[0012] In some embodiments of this application, generating a dataset based on historical power forecast data of the integrated energy base, historical day-ahead power output plan data and historical actual power output data corresponding to the historical power forecast data respectively includes:

[0013] Acquire daily historical power prediction data samples corresponding to multiple historical sampling time points of the integrated energy base. Each daily historical power prediction data sample includes: daily historical photovoltaic power prediction data and daily historical wind power prediction data.

[0014] Obtain the historical day-ahead power output plan data sample and historical actual power output data sample corresponding to each of the aforementioned daily historical power forecast data samples for the next day. Each of the historical day-ahead power output plan data samples includes: historical photovoltaic planned power output ratio, historical wind power planned power output ratio, and historical thermal power planned power output ratio; each of the historical actual power output data samples includes: historical photovoltaic actual power output ratio, historical wind power actual power output ratio, and historical thermal power actual power output ratio.

[0015] Each sample pair is generated based on the correspondence between each of the daily historical power prediction data samples and the historical daily power output plan data samples and historical actual power output data samples, so as to obtain a dataset containing multiple sample pairs.

[0016] In some embodiments of this application, before training a preset deep neural network using the dataset, the following steps are further included:

[0017] Construct an objective function to represent the difference between the day-ahead scheduling optimization data output by the deep neural network and the historical actual output data;

[0018] Correspondingly, training a preset deep neural network using the dataset includes:

[0019] A preset deep neural network is trained based on the dataset with the goal of minimizing the objective function.

[0020] In some embodiments of this application, the historical power prediction data, historical day-ahead power output plan data, and historical actual power output data are all time series data;

[0021] Correspondingly, the deep neural network includes: LSTM model.

[0022] In some embodiments of this application, the LSTM model includes: an optimizer consisting of multiple LSTM layers connected sequentially;

[0023] The optimizer is used to optimize historical day-ahead power output plan data based on the input historical power prediction data and historical actual power output data.

[0024] The second aspect of this application provides a day-ahead dispatching plan optimization device for an integrated energy base, comprising:

[0025] The data acquisition module is used to acquire the day-ahead scheduling plan data and the corresponding day-ahead power forecast data of the integrated energy base;

[0026] The model prediction module is used to input the day-ahead scheduling plan data and the day-ahead power prediction data into the day-ahead scheduling plan optimization model, so that the day-ahead scheduling plan optimization model outputs day-ahead scheduling plan optimization data corresponding to the day-ahead scheduling plan data;

[0027] The day-ahead scheduling optimization model is pre-trained based on the historical day-ahead power prediction data of the integrated energy base, the historical day-ahead scheduling plan data corresponding to the historical day-ahead power prediction data, and the historical actual power output data.

[0028] In some embodiments of this application, it also includes:

[0029] The dataset generation module is used to generate a dataset based on the historical power forecast data of the integrated energy base, the historical day-ahead power output plan data and the historical actual power output data corresponding to the historical power forecast data, respectively.

[0030] The model training module is used to train a preset deep neural network using the dataset to obtain a day-ahead scheduling optimization model that outputs day-ahead scheduling optimization data based on the day-ahead scheduling plan data and the corresponding power prediction data of the integrated energy base.

[0031] A third aspect of this application provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the day-ahead scheduling optimization method for the integrated energy base.

[0032] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the day-ahead scheduling optimization method for the integrated energy base.

[0033] The day-ahead scheduling plan optimization method for integrated energy bases provided in this application acquires day-ahead scheduling plan data and corresponding day-ahead power forecast data for the integrated energy base; inputs the day-ahead scheduling plan data and the day-ahead power forecast data into a day-ahead scheduling plan optimization model, so that the day-ahead scheduling plan optimization model outputs day-ahead scheduling plan optimization data corresponding to the day-ahead scheduling plan data; wherein, the day-ahead scheduling plan optimization model is pre-trained based on the historical day-ahead power forecast data of the integrated energy base, the historical day-ahead scheduling plan data corresponding to the historical day-ahead power forecast data, and the historical actual power output data, which can effectively reduce the deviation between planned power output allocation and actual power output allocation, can reasonably utilize the deviation between the historical actual power output ratio and the day-ahead planned allocation, and can effectively and reliably optimize the day-ahead scheduling plan, and can improve the rationality and effectiveness of energy optimization allocation for the integrated energy base based on the optimization results, thereby improving the stability and safety of the integrated energy base operation.

[0034] Additional advantages, objectives, and features of this application will be set forth in part in the description which follows, and will in part become apparent to those skilled in the art upon review of the following description, or may be learned by practice of the application. The objectives and other advantages of this application can be realized and obtained by means of the structures specifically pointed out in the specification and drawings.

[0035] Those skilled in the art will understand that the purposes and advantages that can be achieved with this application are not limited to those specifically described above, and that the above and other purposes that this application can achieve will be more clearly understood from the following detailed description. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. 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.

[0037] Figure 1 This is a flowchart illustrating a method for optimizing the day-ahead scheduling plan of an integrated energy base according to an embodiment of this application.

[0038] Figure 2 This is another flowchart illustrating the day-ahead scheduling optimization method for an integrated energy base according to one embodiment of this application.

[0039] Figure 3 This is a schematic diagram illustrating the specific execution flow of steps 010 to 020 in the day-ahead scheduling plan optimization method for an integrated energy base according to an embodiment of this application.

[0040] Figure 4 This is a schematic diagram of a day-ahead scheduling optimization device for an integrated energy base, as described in another embodiment of this application.

[0041] Figure 5 This is another schematic diagram of the day-ahead scheduling optimization device for an integrated energy base in another embodiment of this application.

[0042] Figure 6 This is a flowchart illustrating the day-ahead scheduling optimization method for an integrated energy base provided in the application example of this application.

[0043] Figure 7 This is a schematic diagram illustrating an example of a photovoltaic power prediction curve provided in an application example of this application.

[0044] Figure 8 This is a schematic diagram illustrating an example of a wind power prediction curve provided in an application example of this application.

[0045] Figure 9 This is an example diagram illustrating the comparison between the day-ahead scheduling plan and the actual output provided in the application example of this application.

[0046] Figure 10 This is a schematic diagram illustrating the structure of the LSTM model in an application example of this application. Detailed Implementation

[0047] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0048] It should also be noted that, in order to avoid obscuring this application with unnecessary details, only the structures and / or processing steps closely related to the scheme according to this application are shown in the accompanying drawings, while other details that are not closely related to this application are omitted.

[0049] It should be emphasized that the term "including / comprises" as used herein refers to the presence of a feature, element, step, or component, but does not exclude the presence or addition of one or more other features, elements, steps, or components.

[0050] It should also be noted that, unless otherwise specified, the term "connection" in this article can refer not only to a direct connection, but also to an indirect connection involving an intermediary.

[0051] In the following description, embodiments of the present application will be illustrated with reference to the accompanying drawings. In the drawings, the same reference numerals represent the same or similar parts, or the same or similar steps.

[0052] To address the shortcomings of existing day-ahead scheduling generation methods, such as the inability to guarantee the effectiveness and reliability of day-ahead scheduling plans, the inability to effectively optimize day-ahead scheduling plans, and the inability to improve the safe and stable operation of integrated energy bases, this application first considers that integrated energy bases utilize a large amount of historical data containing information, and there is still room for optimization. If the deviation between the final output ratio and the day-ahead plan can be effectively utilized, the day-ahead scheduling plan can be optimized. However, directly using historical information requires a significant amount of manpower and time, as a considerable amount of time is needed to compile historical data each time a day-ahead scheduling plan needs to be optimized.

[0053] Based on this, in order to achieve effective optimization of the day-ahead scheduling plan without affecting its optimization and execution efficiency, this application provides a method for optimizing the day-ahead scheduling plan of an integrated energy base. This method employs a deep neural network and trains it using historical day-ahead power prediction data, historical day-ahead scheduling plan data corresponding to the historical day-ahead power prediction data, and historical actual power output data. The deep neural network is then trained to output a day-ahead scheduling plan generation model that outputs optimized day-ahead scheduling plan data based on the day-ahead scheduling plan data and the day-ahead power prediction data. Subsequently, this day-ahead scheduling plan generation model can be directly used to quickly and effectively optimize the day-ahead scheduling plan data. This not only ensures the optimization efficiency of the day-ahead scheduling plan but also effectively improves the training effectiveness and reliability of the day-ahead scheduling plan optimization model. It enables effective and reliable optimization of the day-ahead scheduling plan and enhances the rationality and effectiveness of energy optimization allocation for the integrated energy base based on the optimization results, thereby improving the stability and safety of the integrated energy base's operation.

[0054] The following examples will provide a detailed description.

[0055] This application provides a method for optimizing the day-ahead scheduling plan of an integrated energy base, which can be implemented by a day-ahead scheduling plan optimization device for an integrated energy base. See [link to relevant documentation]. Figure 1 The day-ahead scheduling optimization method for the integrated energy base specifically includes the following:

[0056] Step 100: Obtain the day-ahead dispatch plan data and the corresponding day-ahead power forecast data of the integrated energy base.

[0057] In step 100, the day-ahead scheduling plan optimization device of the integrated energy base can receive the latest day-ahead scheduling plan data from the integrated energy base and extract the day-ahead power prediction data corresponding to the day-ahead scheduling plan data.

[0058] It is understood that the day-ahead power forecast data and the corresponding day-ahead scheduling plan data specifically refer to the day-ahead scheduling plan data and the day-ahead power forecast data of the day preceding the day-ahead scheduling plan data.

[0059] Step 200: Input the day-ahead scheduling plan data and the day-ahead power prediction data into the day-ahead scheduling plan optimization model so that the day-ahead scheduling plan optimization model outputs day-ahead scheduling plan optimization data corresponding to the day-ahead scheduling plan data; wherein, the day-ahead scheduling plan optimization model is pre-trained based on the historical day-ahead power prediction data of the integrated energy base, the historical day-ahead scheduling plan data corresponding to the historical day-ahead power prediction data, and the historical actual power output data respectively.

[0060] In one or more embodiments of this application, the day-ahead scheduling plan optimization model refers to a machine learning model used to optimize the day-ahead scheduling plan. The architecture of this model is the same as its deep neural network architecture. That is, the day-ahead scheduling plan optimization model refers to the trained deep neural network. Subsequently, according to actual application needs, the latest updated historical day-ahead power prediction data, historical day-ahead scheduling plan data and historical actual output data can be collected periodically from the database of the integrated energy base. Then, the day-ahead scheduling plan optimization model can be optimized and iterated using these updated historical day-ahead power prediction data, historical day-ahead scheduling plan data and historical actual output data to obtain an updated day-ahead scheduling plan optimization model, so as to further improve the reliability and effectiveness of the optimization results output by the model, and to be more suitable for the state changes of the integrated energy base.

[0061] As described above, the day-ahead scheduling plan optimization method for integrated energy bases provided in this application embodiment obtains day-ahead scheduling plan data and corresponding day-ahead power prediction data for the integrated energy base; inputs the day-ahead scheduling plan data and the day-ahead power prediction data into a day-ahead scheduling plan optimization model, so that the day-ahead scheduling plan optimization model outputs day-ahead scheduling plan optimization data corresponding to the day-ahead scheduling plan data; wherein, the day-ahead scheduling plan optimization model is pre-trained based on the historical day-ahead power prediction data of the integrated energy base, the historical day-ahead scheduling plan data corresponding to the historical day-ahead power prediction data, and the historical actual power output data, which can effectively reduce the deviation between planned power output allocation and actual power output allocation, can reasonably utilize the deviation between the historical actual power output ratio and the day-ahead planned allocation, effectively and reliably optimize the day-ahead scheduling plan, and can improve the rationality and effectiveness of energy optimization allocation for the integrated energy base based on the optimization results, thereby improving the stability and safety of the integrated energy base operation.

[0062] To further improve the effectiveness of reducing the deviation between planned and actual power output allocation, a day-ahead scheduling optimization method for an integrated energy base is provided in this application embodiment, see [link to relevant documentation]. Figure 2 The method for optimizing the day-ahead scheduling plan of the integrated energy base includes the following steps prior to step 100:

[0063] Step 010: Generate a dataset based on the historical power forecast data of the integrated energy base, the historical day-ahead power output plan data and the historical actual power output data corresponding to the historical power forecast data.

[0064] It is understood that the dataset stores the correspondence between historical power forecast data, historical day-ahead power output plan data, and historical actual power output data. This correspondence means that the dataset not only stores the content of historical power forecast data, historical day-ahead power output plan data, and historical actual power output data, but also stores the one-to-one correspondence between historical power forecast data, historical day-ahead power output plan data, and historical actual power output data.

[0065] Step 020: Train a preset deep neural network using the dataset to obtain a day-ahead scheduling plan optimization model that outputs day-ahead scheduling plan optimization data for day-ahead power output plan data based on the day-ahead scheduling plan data and corresponding power prediction data of the integrated energy base.

[0066] In step 020, the dataset is used to train a preset deep neural network. Specifically, the dataset can be used as a training set to train the deep neural network; the dataset can also be divided into a training set, a validation set, and a test set, etc., so that after training the deep neural network with the training set, the deep neural network can be further optimized with the validation set and the test set, etc., to improve the reliability and effectiveness of the model training results. The specific settings can be set according to the actual application situation.

[0067] To further improve the effectiveness and reliability of the data foundation used for training the model, in the day-ahead scheduling plan optimization method for an integrated energy base provided in this application embodiment, see [link to relevant documentation]. Figure 3 Step 010 of the day-ahead scheduling plan optimization method for the integrated energy base specifically includes the following:

[0068] Step 011: Obtain daily historical power prediction data samples corresponding to multiple historical sampling time points of the integrated energy base. Each daily historical power prediction data sample includes daily historical photovoltaic power prediction data and daily historical wind power prediction data.

[0069] Step 012: Obtain the historical day-ahead power output plan data sample and historical actual power output data sample corresponding to each of the daily historical power prediction data samples for the next day. Each historical day-ahead power output plan data sample includes: historical photovoltaic planned power output ratio, historical wind power planned power output ratio, and historical thermal power planned power output ratio; each historical actual power output data sample includes: historical photovoltaic actual power output ratio, historical wind power actual power output ratio, and historical thermal power actual power output ratio.

[0070] Step 013: Generate each sample pair according to the correspondence between each of the daily historical power prediction data samples and the historical daily power output plan data samples and historical actual power output data samples, so as to obtain a dataset containing multiple sample pairs.

[0071] Specifically, a dataset can be created by sampling historical daytime power forecast data and the ratio of planned to actual daytime power output in segments of ΔT. An example of ΔT segmented sampling is sampling 24-hour forecast data every hour.

[0072] It is understandable that the planned output of photovoltaic, wind power, and thermal power is derived from a comprehensive analysis of DC transmission capacity, upper and lower limits of thermal power output, and wind and solar power prediction results. The aforementioned conventional scheduling method can be used here, or other scheduling methods can be used, but it is necessary to ensure that all samples use the same scheduling method.

[0073] To further improve the application effectiveness of the trained model, in the day-ahead scheduling plan optimization method for an integrated energy base provided in this application embodiment, see [link to relevant documentation]. Figure 3 The method for optimizing the day-ahead scheduling plan of the integrated energy base, prior to step 020, specifically includes the following:

[0074] Step 030: Construct an objective function to represent the difference between the day-ahead scheduling plan optimization data output by the deep neural network and the historical actual output data.

[0075] In step 030, the difference between the day-ahead scheduling plan optimization data output by the deep neural network and the historical actual output data can be determined based on the average of the squared differences between the day-ahead scheduling plan optimization data and the historical actual output data.

[0076] Correspondingly, see Figure 3 Step 020 of the day-ahead scheduling plan optimization method for the integrated energy base specifically includes the following:

[0077] Step 021: With the objective function as the target, train a preset deep neural network based on the dataset to obtain a day-ahead scheduling plan optimization model for outputting day-ahead scheduling plan optimization data for day-ahead power output plan data based on the day-ahead scheduling plan data and corresponding power prediction data of the integrated energy base.

[0078] To further improve the reliability of model mapping learning, in the day-ahead scheduling plan optimization method for an integrated energy base provided in this application embodiment, the historical power prediction data, historical day-ahead output plan data, and historical actual output data are all time series data; correspondingly, the deep neural network includes an LSTM model.

[0079] It is understandable that the LSTM model refers to the Long Short-Term Memory (LSTM) model.

[0080] To further improve the reliability of model training and the effectiveness of application, in the day-ahead scheduling plan optimization method for an integrated energy base provided in this application embodiment, the LSTM model may specifically include: an optimizer composed of multiple LSTM layers connected in sequence;

[0081] The optimizer is used to optimize historical day-ahead power output plan data based on the input historical power prediction data and historical actual power output data.

[0082] From a software perspective, this application also provides an apparatus for optimizing the day-ahead scheduling plan of an integrated energy base, comprising all or part of the day-ahead scheduling plan optimization method for the integrated energy base. See [link to relevant documentation]. Figure 4 The day-ahead scheduling optimization device of the integrated energy base is connected to the database of the integrated energy base and the user's client device to obtain data from the database and send the optimization results to the client device for the user to view. The day-ahead scheduling optimization device of the integrated energy base specifically includes the following:

[0083] Data acquisition module 10 is used to acquire day-ahead scheduling plan data and corresponding day-ahead power forecast data of the integrated energy base;

[0084] The model prediction module 20 is used to input the day-ahead scheduling plan data and the day-ahead power prediction data into the day-ahead scheduling plan optimization model, so that the day-ahead scheduling plan optimization model outputs day-ahead scheduling plan optimization data corresponding to the day-ahead scheduling plan data;

[0085] The day-ahead scheduling optimization model is pre-trained based on the historical day-ahead power prediction data of the integrated energy base, the historical day-ahead scheduling plan data corresponding to the historical day-ahead power prediction data, and the historical actual power output data.

[0086] To further improve the effectiveness of reducing the deviation between planned and actual power output allocation, a day-ahead scheduling optimization device for an integrated energy base, as provided in this application embodiment, is described below. Figure 5 The day-ahead scheduling optimization device for the integrated energy base also specifically includes the following components:

[0087] The dataset generation module 01 is used to generate a dataset based on the historical power forecast data of the integrated energy base, the historical day-ahead power output plan data and the historical actual power output data corresponding to the historical power forecast data respectively;

[0088] Model training module 02 is used to train a preset deep neural network using the dataset to obtain a day-ahead scheduling plan optimization model that outputs day-ahead scheduling plan optimization data for day-ahead power output plan data based on the day-ahead scheduling plan data and corresponding power prediction data of the integrated energy base.

[0089] The embodiments of the day-ahead scheduling plan optimization device for integrated energy bases provided in this application can be used to execute the processing flow of the embodiments of the day-ahead scheduling plan optimization method for integrated energy bases in the above embodiments. Its functions will not be repeated here, but can be referred to the detailed description of the embodiments of the day-ahead scheduling plan optimization method for integrated energy bases.

[0090] As can be seen from the above description, the day-ahead scheduling plan optimization device for integrated energy bases provided in this application embodiment can effectively reduce the deviation between planned output allocation and actual output allocation, can reasonably utilize the deviation between historical actual output ratio and day-ahead plan, can effectively and reliably optimize the day-ahead scheduling plan, and can improve the rationality and effectiveness of energy optimization allocation for integrated energy bases based on the optimization results, thereby improving the stability and safety of integrated energy base operation.

[0091] It is understood that the optimization of the day-ahead scheduling plan of the integrated energy base by the day-ahead scheduling plan optimization device can be completed in the client device. Specifically, the selection can be made based on the processing capacity of the client device and the limitations of the user's usage scenario. This application does not impose any limitations in this regard. If all operations are completed in the client device, the client device may further include a processor for the specific processing of the day-ahead scheduling plan optimization of the integrated energy base.

[0092] The aforementioned client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission. The server may include a server on the task scheduling center side; in other implementation scenarios, it may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, a server cluster consisting of multiple servers, or a distributed server structure.

[0093] The server and the client device can communicate using any suitable network protocol, including those not yet developed as of the date of this application. Such network protocols may include, for example, TCP / IP, UDP / IP, HTTP, HTTPS, etc. Furthermore, such network protocols may also include RPC (Remote Procedure Call Protocol) and REST (Representational State Transfer Protocol) protocols used on top of the aforementioned protocols.

[0094] To further illustrate this solution, this application also provides a specific application example of a day-ahead scheduling plan optimization method for integrated energy bases. This method addresses the problems existing in day-ahead scheduling plan generation methods, such as the discrepancy between planned and actual power output allocation caused by various uncertainties in wind and solar power prediction and operation control. The information contained in this discrepancy is not effectively utilized in traditional methods. Specifically, a deep learning network is built to learn from historical data, thereby optimizing the established scheduling plan and reducing the discrepancy between planned and actual power output allocation.

[0095] See Figure 6 The day-ahead scheduling optimization method for integrated energy bases provided in this application example specifically includes the following:

[0096] S1: Sample historical data and create a dataset.

[0097] 1) Photovoltaic power forecast data n equals the number of samples. The photovoltaic power forecast curve is shown below, representing the data volume for one day. Figure 7 As shown.

[0098]

[0099] In the formula, m is the number of sampling points in a day. If the sampling interval is 1 hour, then m is 24. If the sampling interval is 15 minutes, then m is 96.

[0100] 2) Similar to photovoltaic power forecast data, wind power forecast data, n equals the number of samples. The wind power forecast curve is shown below, representing the data volume for one day. Figure 8 As shown.

[0101]

[0102] In the formula, m is the number of sampling points in a day. If the sampling interval is 1 hour, then m is 24. If the sampling interval is 15 minutes, then m is 96.

[0103] 3) Current planned data on the ratio of wind, solar and thermal power output.

[0104]

[0105] in The planned contribution percentage is calculated using the following formula:

[0106]

[0107]

[0108]

[0109] P′ Pij 、P′ Wij 、P′ Tij The planned output for photovoltaic, wind power, and thermal power is determined by a comprehensive analysis of DC transmission capacity, upper and lower limits of thermal power output, and wind and solar power prediction results. The aforementioned conventional scheduling method can be used here, or other scheduling methods can be used, but it is necessary to ensure that all samples use the same scheduling method.

[0110] 4) Actual output ratio of photovoltaic, wind power, and thermal power

[0111]

[0112] R Pij R Wij R Tij The actual output ratio of photovoltaic, wind power, and thermal power, and the comparison between planned output and actual output. Figure 9 As shown.

[0113] S2: Model Building

[0114] Establish input as The target output is R i To achieve this goal, the model can be built using different neural networks. For example, LSTM can be used through... Figure 10 The LSTM model structure shown implements this model, where, Figure 10 The dashed coil portion forms an optimizer, which performs optimization on... Optimization, making Get as close to R as possible i .

[0115] S3: Construct the objective function to guide model training

[0116]

[0117] In the formula The output power percentage of the model is represented by n, where n is the total number of samples.

[0118] S4: Model Training

[0119] The model was trained using historical datasets to establish... The mapping.

[0120] S5: Output results using the model

[0121] To determine the day-ahead wind and solar power forecasts and the day-ahead power output to be optimized, this data is used as input to the model trained in section 3.3 to obtain the final output. That is The optimized scheduling plan.

[0122] In summary, the day-ahead scheduling optimization method for integrated energy bases provided in this application example can generate a dataset that matches historical power forecast data, historical day-ahead scheduling plan data, and historical actual scheduling data. It can realize a deep network structure that takes time-series data of power forecast and day-ahead scheduling plan as input and outputs time-series data of optimized day-ahead scheduling plan. By training the deep learning network using the above dataset and training the deep network using historical dataset, the function of day-ahead scheduling plan optimization can be realized. That is, the function of taking power forecast data and the generated day-ahead scheduling plan as input and outputting the optimized day-ahead scheduling plan can be realized.

[0123] This application also provides an electronic device (i.e., an electronic device), which may include a processor, a memory, a receiver, and a transmitter. The processor is used to execute the day-ahead scheduling plan optimization method for integrated energy bases mentioned in the above embodiments. The processor and memory can be connected via a bus or other means, taking a bus connection as an example. The receiver can be connected to the processor and memory via wired or wireless means.

[0124] The processor can be a central processing unit (CPU). The processor can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips.

[0125] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the day-ahead scheduling plan optimization method for integrated energy bases in the embodiments of this application. The processor executes various functional applications and data processing by running the non-transitory software programs, instructions, and modules stored in the memory, thereby implementing the day-ahead scheduling plan optimization method for integrated energy bases in the above method embodiments.

[0126] The memory may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the processor, etc. Furthermore, the memory may include high-speed random access memory and non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include memory remotely located relative to the processor, which can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0127] The one or more modules are stored in the memory, and when executed by the processor, the day-ahead scheduling plan optimization method of the integrated energy base in the embodiment is executed.

[0128] In some embodiments of this application, the user equipment may include a processor, a memory, and a transceiver unit. The transceiver unit may include a receiver and a transmitter. The processor, memory, receiver, and transmitter may be connected via a bus system. The memory is used to store computer instructions, and the processor is used to execute the computer instructions stored in the memory to control the transceiver unit to send and receive signals.

[0129] As one implementation method, the functions of the receiver and transmitter in this application can be implemented by transceiver circuits or dedicated transceiver chips, and the processor can be implemented by dedicated processing chips, processing circuits or general-purpose chips.

[0130] As another implementation approach, the server provided in this application embodiment can be implemented using a general-purpose computer. That is, the program code implementing the processor, receiver, and transmitter functions is stored in memory, and the general-purpose processor implements the processor, receiver, and transmitter functions by executing the code in memory.

[0131] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned integrated energy base day-ahead scheduling plan optimization method. The computer-readable storage medium can be a tangible storage medium, such as random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, floppy disks, hard disks, removable storage disks, CD-ROMs, or any other form of storage medium known in the art.

[0132] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave.

[0133] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0134] In this application, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.

[0135] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to the embodiments of this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for day-ahead scheduling optimization of an integrated energy base, characterized in that, include: Obtain day-ahead dispatch plan data and corresponding day-ahead power forecast data for the integrated energy base; The daytime scheduling plan data and the daytime power prediction data are input into the daytime scheduling plan optimization model so that the daytime scheduling plan optimization model outputs daytime scheduling plan optimization data corresponding to the daytime scheduling plan data. The day-ahead scheduling plan optimization model is pre-trained based on the historical day-ahead power prediction data of the integrated energy base, the historical day-ahead scheduling plan data corresponding to the historical day-ahead power prediction data, and the historical actual power output data. Before inputting the day-ahead scheduling plan data and the day-ahead power prediction data into the day-ahead scheduling plan optimization model, the method further includes: A dataset is generated based on the historical power forecast data of the integrated energy base, the historical daytime power output plan data and the historical actual power output data corresponding to the historical power forecast data, respectively. A pre-set deep neural network is trained using the dataset to obtain a day-ahead scheduling optimization model that outputs day-ahead scheduling optimization data based on the day-ahead scheduling plan data and the corresponding power prediction data of the integrated energy base. The dataset is generated based on historical power forecast data of the integrated energy base, historical day-ahead power output plan data and historical actual power output data corresponding to the historical power forecast data, and includes: Acquire daily historical power prediction data samples corresponding to multiple historical sampling time points of the integrated energy base. Each daily historical power prediction data sample includes: daily historical photovoltaic power prediction data and daily historical wind power prediction data. Obtain the historical day-ahead power output plan data sample and historical actual power output data sample corresponding to each of the aforementioned daily historical power forecast data samples for the next day. Each of the historical day-ahead power output plan data samples includes: historical photovoltaic planned power output ratio, historical wind power planned power output ratio, and historical thermal power planned power output ratio; each of the historical actual power output data samples includes: historical photovoltaic actual power output ratio, historical wind power actual power output ratio, and historical thermal power actual power output ratio. Each sample pair is generated based on the correspondence between each of the daily historical power prediction data samples and the historical daily power output plan data samples and historical actual power output data samples, so as to obtain a dataset containing multiple sample pairs.

2. The day-ahead scheduling optimization method of an integrated energy community according to claim 1, characterized in that, Before training the preset deep neural network using the dataset, the method further includes: Construct an objective function to represent the difference between the day-ahead scheduling optimization data output by the deep neural network and the historical actual output data; Correspondingly, training a preset deep neural network using the dataset includes: A preset deep neural network is trained based on the dataset with the goal of minimizing the objective function.

3. The day-ahead scheduling optimization method of an integrated energy community of claim 1, wherein, The historical power forecast data, historical day-ahead power output plan data, and historical actual power output data are all time series data. Correspondingly, the deep neural network includes: LSTM model.

4. The method for optimizing the day-ahead scheduling plan of an integrated energy base according to claim 3, characterized in that, The LSTM model includes an optimizer consisting of multiple LSTM layers connected sequentially. The optimizer is used to optimize historical day-ahead power output plan data based on the input historical power prediction data and historical actual power output data.

5. A day-ahead scheduling plan optimization device for an integrated energy base, implementing the day-ahead scheduling plan optimization method for an integrated energy base as described in any one of claims 1 to 4, characterized in that, include: The data acquisition module is used to acquire the day-ahead scheduling plan data and the corresponding day-ahead power forecast data of the integrated energy base; The model prediction module is used to input the day-ahead scheduling plan data and the day-ahead power prediction data into the day-ahead scheduling plan optimization model, so that the day-ahead scheduling plan optimization model outputs day-ahead scheduling plan optimization data corresponding to the day-ahead scheduling plan data; The day-ahead scheduling optimization model is pre-trained based on the historical day-ahead power prediction data of the integrated energy base, the historical day-ahead scheduling plan data corresponding to the historical day-ahead power prediction data, and the historical actual power output data.

6. The day-ahead scheduling optimization device for an integrated energy base according to claim 5, characterized in that, Also includes: The dataset generation module is used to generate a dataset based on the historical power forecast data of the integrated energy base, the historical day-ahead power output plan data and the historical actual power output data corresponding to the historical power forecast data, respectively. The model training module is used to train a preset deep neural network using the dataset to obtain a day-ahead scheduling optimization model that outputs day-ahead scheduling optimization data based on the day-ahead scheduling plan data and the corresponding power prediction data of the integrated energy base.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the day-ahead scheduling plan optimization method for an integrated energy base as described in any one of claims 1 to 4.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the day-ahead scheduling optimization method for an integrated energy base as described in any one of claims 1 to 4.

Images