Power grid control method, control device and electronic equipment

Abstract

The invention provides a power grid control method, a control device and electronic equipment. The method comprises the following steps: acquiring first data of a new energy power station in a first historical time period; processing the first data by using a prediction model to obtain a power generation prediction result of a first predetermined day after the first historical time period; and constructing a scheduling model of the power grid, at least inputting the power generation prediction result into the scheduling model to obtain a scheduling plan, and executing the scheduling plan to control the power grid, and the method solves the problem of low precision of power grid control in the prior art.

Description

Technical Field

[0001] The present application relates to the field of power grid dispatching, and in particular, to a power grid control method, a control device, a computer-readable storage medium, and an electronic device. Background Art

[0002] With the depletion of fossil energy and the deterioration of the ecological environment worldwide, the double pressure has promoted the rapid development of renewable energy. Establishing a high-quality, efficient and low-carbon energy interconnection system in the new era has become an inevitable choice for countries around the world to achieve sustainable energy development. Under the background of global energy transformation and the strong promotion of various countries, renewable energy power generation technology has been deeply developed. Renewable energy generation is mainly connected to the grid in a centralized or distributed access mode. It not only has flexibility in layout in terms of geographical location, but also can better disperse the power transmission demand in different regions according to actual conditions. At the same time, as a pollution-free clean energy, it can reduce the environmental pollution caused by traditional power generation methods and save power generation costs. Many advantages have made the proportion of distributed renewable energy generation connected to the distribution network show an increasing trend year by year. However, as an important supplement to centralized power generation, the increasing number of distributed renewable energy generation access will bring energy production and trading closer to the demand side. The original traditional radial network has been transformed into a complex network with multiple energy sources and multiple users interconnected. The previous source-grid-load energy flow no longer exists in a unidirectional form, but has been transformed into a bidirectional flow, which brings new challenges and tests to the power quality and voltage optimization control of the power grid.

[0003] There are technical solutions in the prior art for realizing grid control optimization through model prediction. For example, the prior art discloses a method for coordinated optimization control of active and reactive power of a distributed photovoltaic distribution network, the method comprising: based on the model prediction control method, dividing the distribution network system control process into long-time scale optimization control and short-time scale optimization control according to control objectives and control variables at different time scales, and establishing a long-time scale optimization control model and a short-time scale optimization control model; converting the problem of solving the long-time scale optimization control and the short-time scale optimization control into a second-order cone programming problem; the long-time scale optimization control model adopts multi-step rolling optimization to solve the active and reactive output of each controllable device based on photovoltaic output and load demand forecast information, and the short-time scale optimization control model uses the solution result of the long-time scale optimization control model as a reference value to roll the active and reactive output increments of each controllable device; however, when making predictions, the above scheme does not make a prediction of the power generation for the new energy type before performing coordinated optimization control, resulting in inaccurate optimization.

[0004] Therefore, a high-precision power grid control method is urgently needed. Summary of the invention

[0005] The main purpose of the present application is to provide a control method, a control device, a computer-readable storage medium and an electronic device for a power grid, so as to at least solve the problem of low accuracy of power grid control in the prior art.

[0006] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a method for controlling a power grid is provided, including obtaining first data of a new energy power station within a first historical time period, the new energy power station including a photovoltaic power station and a wind power station, the first data including at least a first power generation of the photovoltaic power station and a second power generation of the wind power station; using a prediction model to process the first data to obtain a power generation prediction result for a first scheduled day after the first historical time period, the prediction model being trained using multiple groups of historical data, each group of the historical data including second data within a second historical time period and third data on a second scheduled day, the second data including at least a third power generation of the photovoltaic power station and a fourth power generation of the wind power station on each day within the second historical time period, the third data including at least a fifth power generation of the photovoltaic power station and a sixth power generation of the wind power station on the second scheduled day, the power generation prediction result Including the seventh power generation of the photovoltaic power station and the eighth power generation of the wind power station, the first historical time period and the second historical time period are different historical time periods with the same duration, the time interval between the last day of the first historical time period and the first scheduled day is a first interval, the time interval between the last day of the second historical time period and the second scheduled day is a second interval, and the first interval is the same as the second interval; constructing a scheduling model for a power grid, the scheduling model includes an optimization target and a constraint condition, the optimization target is set at least according to the operation cost of the power grid, the constraint condition is set at least according to the power balance constraint, and the power grid is a device for connecting the power generated by the new energy power station to the grid; at least the power generation prediction result is input into the scheduling model to obtain a scheduling plan, and the scheduling plan is executed to control the power grid, and the scheduling plan is the amount of power that the seventh power generation and the eighth power generation need to be connected to the grid respectively.

[0007] Optionally, before using the prediction model to process the first data to obtain the power generation prediction result of the scheduled day after the historical time period, the method also includes an acquisition step of acquiring multiple groups of historical data; a preprocessing step of preprocessing the second data in the multiple groups of historical data to obtain multiple groups of fourth data; an input step of inputting the multiple groups of fourth data into a convolutional neural network model; and a training step of training the convolutional neural network model to obtain the prediction model.

[0008] Optionally, the preprocessing step includes performing data cleaning on the second data to obtain the cleaned second data; supplementing the missing data in the cleaned second data to obtain the supplemented second data; and performing data normalization on the supplemented second data to obtain the fourth data.

[0009] Optionally, the training step also includes at least setting the number of iterations to train the convolutional neural network, and obtaining at least the ninth power generation of the photovoltaic power station and the tenth power generation of the wind power station on the second scheduled day, and the corresponding loss function curve when the training number reaches the number of iterations, wherein each group of the fourth data includes a ninth power generation and a tenth power generation; a comparison step, respectively comparing the ninth power generation and the tenth power generation obtained by training each group of the fourth data with the third data corresponding to each group of the fourth data to obtain multiple first differences, and performing average calculation on the multiple first differences to obtain a second difference; a judgment step, determining whether the trained convolutional neural network model is the prediction model according to whether the second difference is within a preset range and whether the loss function curve is in a convergence state, and the preset range is set according to actual conditions.

[0010] Optionally, the judging step includes determining that the trained convolutional neural network model is the prediction model when the second difference is within the preset range and the loss function curve is in a convergence state; and adjusting the parameters of the convolutional neural network model when the second difference is not within the preset range or the loss function curve is not in a convergence state, and repeating the input step, the training step and the comparison step at least once in sequence until the second difference is within the preset range, determining that the trained convolutional neural network model is the prediction model, and the parameters include at least the number of iterations.

[0011] Optionally, constructing the optimization target of the dispatching model includes constructing the operation cost of the power grid according to the loss cost of the power grid and the power purchase cost and power sales income of the power grid, wherein the operation cost C of the power grid is w The calculation formula is C W =C K +C T , where C K is the loss cost of the power grid, C T is the electricity purchase cost and electricity sales income of the power grid; at least according to the power grid's ability to absorb the power generated by the photovoltaic power station and the power grid's ability to absorb the power generated by the wind power station, the new energy absorption cost of the power grid is constructed, wherein the new energy absorption cost C′Dis The calculation formula is Where P WT,dis (t) is the power grid's ability to absorb the power generated by the photovoltaic power station, P PV,dis (t) is the power grid's ability to absorb the power generated by the wind power station, and C(t) is the electricity price at time t; the optimization objective of the scheduling model is constructed based on the minimum value of the sum of the operating cost of the power grid and the new energy absorption cost.

[0012] Optionally, constructing the constraint condition of the scheduling model includes constructing a first constraint condition according to the maximum power generation of the photovoltaic power station and the seventh power generation, wherein the first constraint condition is P PV (t)≤P PV,pre (t), where P PV (t) is the maximum power generation of the photovoltaic power station, P PV,pre (t) is the seventh power generation; according to the maximum power generation of the wind power station and the eighth power generation, a second constraint condition is constructed, wherein the second constraint condition is P WT (t)≤P WT,pre (t), where P WT (t) is the maximum power generation of the wind power station, P WT,pre (t) is the eighth power generation; at least according to the purchased power of the power grid and the energy storage power on the user side, the power balance constraint is constructed, the first constraint condition, the second constraint condition and the power balance constraint constitute the constraint condition of the scheduling model, wherein the power balance constraint is P WT (t)+P PV (t)+P g (t) = P ESS (t)+P EV (t)+P load (t)+P loss (t), where P g (t) is the power purchased by the power grid, P ESS (t) is the energy storage power at the user side, P load (t) is the power of the fixed load on the user side, P loss (t) is the power lost on the user side, and the power lost on the user side is the power lost in the process of transmitting electricity to the user side.

[0013] According to another aspect of the present application, a control device for a power grid is provided, comprising:

[0014] An acquisition unit, acquiring first data of a new energy power station in a first historical time period, wherein the new energy power station includes a photovoltaic power station and a wind power station, and the first data includes at least a first power generation of the photovoltaic power station and a second power generation of the wind power station;

[0015] a processing unit, processing the first data using a prediction model to obtain a power generation prediction result for a first scheduled day after the first historical time period, the prediction model being obtained by training using multiple groups of historical data, each group of the historical data comprising second data in a second historical time period and third data on a second scheduled day, the second data comprising at least a third power generation of the photovoltaic power station and a fourth power generation of the wind power station on each day in the second historical time period, the third data comprising at least a fifth power generation of the photovoltaic power station and a sixth power generation of the wind power station on the second scheduled day, the power generation prediction result comprising a seventh power generation of the photovoltaic power station and an eighth power generation of the wind power station, the first historical time period and the second historical time period being different historical time periods and having the same duration, the time interval between the last day of the first historical time period and the first scheduled day being a first interval, the time interval between the last day of the second historical time period and the second scheduled day being a second interval, and the first interval being the same as the second interval;

[0016] A construction unit is configured to construct a dispatching model of a power grid, wherein the dispatching model includes an optimization target and a constraint condition, wherein the optimization target is set at least according to an operation cost of the power grid, and the constraint condition is set at least according to a power balance constraint, and the power grid is a device for connecting the power generated by the new energy power station to the grid;

[0017] An execution unit at least inputs the power generation prediction result into the scheduling model to obtain a scheduling plan, and executes the scheduling plan to control the power grid, wherein the scheduling plan is the amount of electricity that the seventh power generation and the eighth power generation need to be connected to the grid respectively.

[0018] According to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned power grid control methods.

[0019] According to another aspect of the present application, an electronic device is provided, comprising one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include a method for executing any one of the above-described power grid control methods.

[0020] Applying the technical solution of the present application, first, first data of a new energy power station within a first historical time period is obtained, the new energy power station including a photovoltaic power station and a wind power station, the first data including at least a first power generation of the photovoltaic power station and a second power generation of the wind power station; secondly, the first data is processed using a prediction model to obtain a power generation prediction result for a first scheduled day after the first historical time period, the prediction model is obtained by training using multiple groups of historical data, each group of the historical data including second data within a second historical time period and third data on a second scheduled day, the second data including at least a third power generation of the photovoltaic power station and a fourth power generation of the wind power station for each day within the second historical time period, the third data including at least a fifth power generation of the photovoltaic power station and a sixth power generation of the wind power station on the second scheduled day. power generation, the power generation prediction result includes the seventh power generation of the photovoltaic power station and the eighth power generation of the wind power station, the first historical time period and the second historical time period are different historical time periods with the same duration, the time interval between the last day of the first historical time period and the first scheduled day is the first interval, the time interval between the last day of the second historical time period and the second scheduled day is the second interval, and the first interval is the same as the second interval; constructing a scheduling model for a power grid, the scheduling model includes optimization objectives and constraints, and the power grid is a device for connecting the power generated by the new energy power station to the grid; at least the power generation prediction result is input into the scheduling model to obtain a scheduling plan, and the scheduling plan is executed to control the power grid, and the scheduling plan is the amount of power that the seventh power generation and the eighth power generation need to be connected to the grid respectively. Compared with the prior art, the solution of the present application utilizes the historical power generation of photovoltaic power stations and the historical power generation of wind power stations to train a convolutional neural network model to obtain a prediction model, and uses the prediction model to predict the power generation of photovoltaic power stations and wind power stations on the scheduled day respectively, to obtain prediction results, and uses the prediction results to make a scheduling plan for the power grid before the scheduled day, thereby achieving more accurate control of the power grid, thereby further solving the problem of low accuracy of power grid control in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0022] Figure 1 A hardware structure block diagram of a mobile terminal for executing a method for controlling a power grid provided in an embodiment of the present application is shown;

[0023] Figure 2 A schematic diagram of a flow chart of a method for controlling a power grid provided according to an embodiment of the present application is shown;

[0024] Figure 3 A structural block diagram of a control device for a power grid provided according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0025] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0026] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] As introduced in the background technology, the prior art has the problem of low accuracy in power grid control. To solve the above problem, the embodiments of the present application provide a power grid control method, a control device, a computer-readable storage medium and an electronic device.

[0029] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0030] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a mobile terminal of a power grid control method according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0031] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the display method of device information in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The transmission device 106 is used to receive or send data via a network. The above-mentioned specific examples of the network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0032] In this embodiment, a method for controlling a power grid running on a mobile terminal, a computer terminal or a similar computing device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0033] Figure 2 is a flow chart of a method for controlling a power grid according to an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0034] Step S201, obtaining first data of a new energy power station within a first historical time period, the new energy power station including a photovoltaic power station and a wind power station, and the first data at least includes a first power generation of the photovoltaic power station and a second power generation of the wind power station.

[0035] Specifically, the above-mentioned data for predicting new energy power stations include: historical power generation data of photovoltaic power generation in the power grid, historical power generation data of wind power generation, influencing factors of wind power generation, and influencing factors data of photovoltaic power generation, among which the influencing factors data of photovoltaic power generation include temperature data, humidity data, and radiation data; the influencing factors of wind turbine power generation include wind speed data and temperature data.

[0036] Step S202, using a prediction model to process the first data to obtain a power generation prediction result for the first scheduled day after the first historical time period, the prediction model is obtained by training using multiple groups of historical data, each group of historical data includes second data in a second historical time period and third data on a second scheduled day, the second data at least includes a third power generation of the photovoltaic power station and a fourth power generation of the wind power station on each day in the second historical time period, the third data at least includes a fifth power generation of the photovoltaic power station and a sixth power generation of the wind power station on the second scheduled day, the power generation prediction result includes a seventh power generation of the photovoltaic power station and an eighth power generation of the wind power station, the first historical time period and the second historical time period are different historical time periods and have the same duration, the time interval between the last day of the first historical time period and the first scheduled day is a first interval, the time interval between the last day of the second historical time period and the second scheduled day is a second interval, and the first interval is the same as the second interval.

[0037] Specifically, the prediction model is a deep learning model (convolutional neural network model), which consists of a data input layer, a hidden layer of a nonlinear activation function, and an output layer. The prediction ability of the model is improved by learning and training the historical data set; the training process is to input the data set into the convolutional neural network model, and the model will output a prediction result (photovoltaic predicted power generation and wind predicted power generation), compare the prediction result with the training set, and then the convolutional neural network model automatically adjusts various parameters in the convolutional neural network model until the model converges, indicating that the model training is completed, and then input the data of six days before the prediction day, and the convolutional neural network model will predict the power generation data of the prediction day.

[0038] Specifically, the first historical time period and the second historical time period can be set to six days, specifically Monday to Saturday, then the first scheduled day and the second scheduled day can be set to Sunday, wherein the first interval and the second interval are Saturday to Sunday, and the values ​​of the first interval and the second interval remain the same; at the same time, the first historical time period and the second historical time period can also be set to thirty days, specifically the first to the thirtieth of each month, then the first scheduled day and the second scheduled day can be set to the thirty-first, wherein the first interval and the second interval are the thirtieth to the thirty-first, and the values ​​of the first interval and the second interval remain the same.

[0039] Step S203, constructing a dispatching model for the power grid, the dispatching model includes optimization objectives and constraints, the optimization objectives are set at least based on the operating cost of the power grid, the constraints are set at least based on the power balance constraint, and the power grid is a device for connecting the electricity generated by the new energy power station to the grid.

[0040] Specifically, the construction process of the power grid dispatching model needs to consider the following key steps: First, the optimization target needs to be determined. Generally, the optimization target of power grid dispatching is to minimize the operating cost of the power grid, including fuel cost, transmission loss, equipment operating cost, etc. According to the actual situation, other optimization targets can also be considered, such as maximizing the use of renewable energy power generation, reducing carbon emissions, etc.; secondly, after determining the optimization target, constraints need to be considered. Power balance constraint is the most basic constraint, which ensures the balance of supply and demand of the power grid and avoids overload or insufficient power supply; finally, a mathematical model is established: according to the determined optimization target and constraints, a mathematical model of power grid dispatching is established. Mathematical tools such as linear programming, integer programming, and mixed integer programming are usually used to describe the operating status and optimization problems of the power grid to achieve the optimal dispatching solution.

[0041] Step S204, at least input the above-mentioned power generation prediction result into the above-mentioned scheduling model to obtain a scheduling plan, and execute the above-mentioned scheduling plan to control the above-mentioned power grid. The above-mentioned scheduling plan is the amount of electricity that the above-mentioned seventh power generation and the above-mentioned eighth power generation need to be connected to the grid respectively.

[0042] Specifically, first, the power generation forecast results are input into the dispatch model, which generates a dispatch plan based on the forecast data and the grid demand. This dispatch plan includes the amount of electricity that needs to be connected to the grid for the seventh and eighth generation. Then, according to this dispatch plan, corresponding operations are performed, such as adjusting the output power of the generator and coordinating the operation between different power generation sources to ensure that the grid operates normally and meets demand. Through this process, the power generation forecast results can be effectively used to optimize the operation of the grid and improve the reliability and efficiency of the grid.

[0043] In the above-mentioned embodiment, firstly, first data of a new energy power station within a first historical time period is obtained, the above-mentioned new energy power station includes a photovoltaic power station and a wind power station, and the above-mentioned first data includes at least a first power generation of the above-mentioned photovoltaic power station and a second power generation of the above-mentioned wind power station; secondly, the above-mentioned first data is processed using a prediction model to obtain a power generation prediction result for the first scheduled day after the above-mentioned first historical time period, the above-mentioned prediction model is obtained by training using multiple groups of historical data, each group of the above-mentioned historical data includes second data within a second historical time period and third data on a second scheduled day, the above-mentioned second data includes at least a third power generation of the above-mentioned photovoltaic power station and a fourth power generation of the above-mentioned wind power station on each day within the above-mentioned second historical time period, and the above-mentioned third data includes at least a fifth power generation of the above-mentioned photovoltaic power station and a sixth power generation of the above-mentioned wind power station on the above-mentioned second scheduled day. The power generation prediction result includes the seventh power generation of the photovoltaic power station and the eighth power generation of the wind power station. The first historical time period and the second historical time period are different historical time periods with the same duration. The time interval between the last day of the first historical time period and the first scheduled day is the first interval. The time interval between the last day of the second historical time period and the second scheduled day is the second interval. The first interval is the same as the second interval. A scheduling model for a power grid is constructed. The scheduling model includes optimization objectives and constraints. The power grid is a device for connecting the power generated by the new energy power station to the grid. At least the power generation prediction result is input into the scheduling model to obtain a scheduling plan. The scheduling plan is executed to control the power grid. The scheduling plan is the amount of power that the seventh power generation and the eighth power generation need to be connected to the grid respectively. Compared with the prior art, the solution of the present application utilizes the historical power generation of photovoltaic power stations and the historical power generation of wind power stations to train a convolutional neural network model to obtain a prediction model, and uses the prediction model to predict the power generation of photovoltaic power stations and wind power stations on the scheduled day respectively, to obtain prediction results, and uses the prediction results to make a scheduling plan for the power grid before the scheduled day, thereby achieving more accurate control of the power grid, thereby further solving the problem of low accuracy of power grid control in the prior art.

[0044] In one embodiment of the present application, before the prediction model is used to process the first data to obtain the power generation prediction result of the scheduled day after the historical time period, the method further includes: an acquisition step to obtain multiple sets of the historical data; a preprocessing step to preprocess the second data in the multiple sets of historical data to obtain multiple sets of fourth data; an input step to input the multiple sets of fourth data into the convolutional neural network model; a training step to train the convolutional neural network model to obtain the prediction model. In this method, a prediction model is obtained by training with multiple sets of historical data, which can more accurately predict the power generation situation in a specific time period in the future.

[0045] In order to further improve the quality and availability of the second data, in one embodiment of the present application, the above-mentioned preprocessing step includes: performing data cleaning on the above-mentioned second data to obtain the cleaned above-mentioned second data; supplementing the missing data in the above-mentioned second data after cleaning to obtain the supplemented above-mentioned second data; performing data normalization on the supplemented above-mentioned second data to obtain the above-mentioned fourth data.

[0046] Specifically, data cleaning involves filtering out abnormal points in the data, deleting the data of the abnormal points, and then supplementing the data at that moment with the average value of the abnormal point at the previous moment and the value at the next moment. Data cleaning also includes processing duplicate data by deletion.

[0047] Specifically, data completion is as follows: for missing data, the average value of the value at the previous moment and the value at the next moment is used to complete the data at that moment.

[0048] Specifically, the formula for data normalization is:

[0049]

[0050] Among them, x in is the normalized data, x is the original data, and x min is the minimum value in the data, x max is the maximum value in the data.

[0051] In another embodiment, the training step further includes: at least setting the number of iterations, training the convolutional neural network, obtaining at least the ninth power generation of the photovoltaic power station and the tenth power generation of the wind power station on the second scheduled day, and the corresponding loss function curve when the training number reaches the number of iterations, wherein each group of the fourth data includes a ninth power generation and a tenth power generation; a comparison step, respectively comparing the ninth power generation and the tenth power generation obtained by training each group of the fourth data with the third data corresponding to each group of the fourth data, obtaining multiple first differences, performing average calculation on the multiple first differences, and obtaining a second difference; a judgment step, determining whether the trained convolutional neural network model is the prediction model according to whether the second difference is within a preset range and whether the loss function curve is in a convergence state, and the preset range is set according to the actual situation. In this method, through the steps of training and evaluating the model, it is further ensured that the model has higher prediction accuracy and stability.

[0052] In order to more effectively improve the performance of the model, in an embodiment of the present application, the above-mentioned judgment step includes: when the above-mentioned second difference is within the above-mentioned preset range and the above-mentioned loss function curve is in a convergence state, determining that the trained convolutional neural network model is the above-mentioned prediction model; when the above-mentioned second difference is not within the above-mentioned preset range or the above-mentioned loss function curve is not in a convergence state, adjusting the parameters of the above-mentioned convolutional neural network model, and repeating the above-mentioned input step, the above-mentioned training step and the above-mentioned comparison step at least once in sequence until the above-mentioned second difference is within the above-mentioned preset range, determining that the trained convolutional neural network model is the above-mentioned prediction model, and the above-mentioned parameters at least include the above-mentioned number of iterations.

[0053] Specifically, when the convolutional neural network model converges, at least three indicators, namely, mean square error, mean absolute percentage error, and prediction accuracy, meet the requirements. Among them, mean square error is a commonly used indicator for evaluating the difference between the model's predicted value and the actual value. The smaller the value, the smaller the difference between the model's predicted value and the actual value, and the higher the degree of model convergence; mean absolute percentage error is an indicator for evaluating the percentage difference between the model's predicted value and the actual value. The smaller the value, the smaller the percentage difference between the model's predicted value and the actual value, and the higher the degree of model convergence; prediction accuracy is a comprehensive indicator, usually including accuracy, recall, F1 value and other indicators. The higher the prediction accuracy, the higher the accuracy of the model in classification or regression tasks, and the better the degree of model convergence.

[0054] In another embodiment, constructing the optimization target of the dispatching model includes: constructing the operation cost of the power grid according to the loss cost of the power grid and the power purchase cost and power sales income of the power grid, wherein the operation cost C of the power grid is w The calculation formula is C W =C K +C T , where C K is the loss cost of the above power grid, C T is the electricity purchase cost and electricity sales income of the above-mentioned power grid; at least according to the above-mentioned power grid's consumption capacity for the power generation of the above-mentioned photovoltaic power station and the above-mentioned power grid's consumption capacity for the power generation of the above-mentioned wind power station, the new energy consumption cost of the above-mentioned power grid is constructed, wherein the above-mentioned new energy consumption cost C′ Dis The calculation formula is Where P WT,dis (t) is the power grid’s ability to absorb the power generated by the photovoltaic power station, P PV,dis (t) is the power grid's ability to absorb the power generated by the wind power station, and C(t) is the electricity price at time t; the optimization target of the dispatch model is constructed based on the minimum value of the sum of the operating cost of the power grid and the new energy absorption cost. In this method, the operating efficiency and reliability of the power grid are further improved by setting the optimization target.

[0055] Specifically, first of all, the optimization goal is to minimize the total grid operation cost and give priority to the consumption of new energy. The total grid operation cost is composed of grid operation cost, loss cost, power purchase cost and power sales income. The consumption cost of new energy is jointly determined by the photovoltaic power generation consumption capacity, wind power generation consumption capacity and electricity price.

[0056] In order to further improve the efficiency of the power grid system, in one embodiment of the present application, constructing the above-mentioned constraint conditions of the above-mentioned scheduling model includes: constructing a first constraint condition according to the maximum power generation of the above-mentioned photovoltaic power station and the above-mentioned seventh power generation, wherein the above-mentioned first constraint condition is P PV (t)≤P PV,pre (t), where P PV (t) is the maximum power generation of the above photovoltaic power station, P PV,pre (t) is the seventh power generation; according to the maximum power generation of the wind power station and the eighth power generation, a second constraint condition is constructed, wherein the second constraint condition is P WT (t)≤P WT,pre (t), where P WT (t) is the maximum power generation of the wind power station, P WT,pre(t) is the eighth power generation; at least according to the purchased power of the power grid and the energy storage power on the user side, the power balance constraint is constructed, and the first constraint condition, the second constraint condition and the power balance constraint constitute the constraint condition of the scheduling model, wherein the power balance constraint is P WT (t)+P PV (t)+P g (t) = P ESS (t)+P EV (t)+P load (t)+P loss (t), where P g (t) is the power purchased by the above power grid, P ESS (t) is the energy storage power on the user side, P load (t) is the power of the fixed load on the user side, P loss (t) is the power lost on the user side, which is the power lost during the transmission of electricity to the user side.

[0057] Specifically, the constraints include photovoltaic output constraints, wind output constraints and power balance constraints. Photovoltaic output constraints ensure that photovoltaic output does not exceed the forecast data, wind output constraints ensure that wind output does not exceed the forecast data, and power balance constraints ensure the balance between power purchased by the grid, user-side energy storage power, fixed load power and loss power.

[0058] In order to enable those skilled in the art to more clearly understand the technical solution of the present application, the implementation process of the power grid control method of the present application will be described in detail below in conjunction with specific embodiments.

[0059] This embodiment relates to a specific power grid control method, including the following steps:

[0060] Step 1: Obtain multiple sets of historical data of new energy power stations, where the new energy power stations include photovoltaic power stations and wind power stations;

[0061] Step 2: Preprocess multiple sets of historical data to obtain data sets for model training;

[0062] Step 3: Establish a convolutional neural network model to predict renewable energy power generation;

[0063] Step 4: Input the training set obtained in step 2 into the convolutional neural network model in step 3, and train the convolutional neural network model to obtain a prediction model;

[0064] Step 5: using the prediction model to predict the first data of the new energy power station in the first historical time period to obtain the power generation of the new energy power station on the first scheduled day;

[0065] Step 6: Construct the dispatch model of the power grid;

[0066] Step 7: Input the power generation of the new energy power station on the first scheduled day obtained in step 5 into the dispatching model of the power grid to obtain the dispatching plan of the power grid, thereby realizing the control of the power grid.

[0067] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0068] The present application also provides a control device for a power grid. It should be noted that the control device for the power grid in the embodiment of the present application can be used to execute the control method for the power grid provided in the embodiment of the present application. The device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions thereof will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and contemplated.

[0069] The following is an introduction to the control device of the power grid provided in the embodiment of the present application.

[0070] Figure 3 Schematic diagram of a control device for a power grid according to an embodiment of the present application. Figure 3 As shown, the device comprises:

[0071] The acquisition unit 10 acquires first data of a new energy power station within a first historical time period, wherein the new energy power station includes a photovoltaic power station and a wind power station, and the first data at least includes a first power generation of the photovoltaic power station and a second power generation of the wind power station.

[0072] Specifically, the above-mentioned data for predicting new energy power stations include: historical power generation data of photovoltaic power generation in the power grid, historical power generation data of wind power generation, influencing factors of wind power generation, and influencing factors data of photovoltaic power generation, among which the influencing factors data of photovoltaic power generation include temperature data, humidity data, and radiation data; the influencing factors of wind turbine power generation include wind speed data and temperature data.

[0073] The processing unit 20 processes the first data using a prediction model to obtain a power generation prediction result for a first scheduled day after the first historical time period, wherein the prediction model is obtained by training using multiple groups of historical data, each group of the historical data includes second data in a second historical time period and third data on a second scheduled day, the second data at least includes a third power generation of the photovoltaic power station and a fourth power generation of the wind power station on each day in the second historical time period, the third data at least includes a fifth power generation of the photovoltaic power station and a sixth power generation of the wind power station on the second scheduled day, the power generation prediction result includes a seventh power generation of the photovoltaic power station and an eighth power generation of the wind power station, the first historical time period and the second historical time period are different historical time periods and have the same duration, the time interval between the last day of the first historical time period and the first scheduled day is a first interval, the time interval between the last day of the second historical time period and the second scheduled day is a second interval, and the first interval is the same as the second interval.

[0074] Specifically, the prediction model is a deep learning model (convolutional neural network model), which consists of a data input layer, a hidden layer of a nonlinear activation function, and an output layer. The prediction ability of the model is improved by learning and training the historical data set; the training process is to input the data set into the convolutional neural network model, and the model will output a prediction result (photovoltaic predicted power generation and wind predicted power generation), compare the prediction result with the training set, and then the convolutional neural network model automatically adjusts various parameters in the convolutional neural network model until the model converges, indicating that the model training is completed, and then input the data of six days before the prediction day, and the convolutional neural network model will predict the power generation data of the prediction day.

[0075] Specifically, the first historical time period and the second historical time period can be set to six days, specifically Monday to Saturday, then the first scheduled day and the second scheduled day can be set to Sunday, wherein the first interval and the second interval are Saturday to Sunday, and the values ​​of the first interval and the second interval remain the same; at the same time, the first historical time period and the second historical time period can also be set to thirty days, specifically the first to the thirtieth of each month, then the first scheduled day and the second scheduled day can be set to the thirty-first, wherein the first interval and the second interval are the thirtieth to the thirty-first, and the values ​​of the first interval and the second interval remain the same.

[0076] Construction unit 30 constructs a dispatching model for a power grid, the dispatching model includes an optimization target and constraints, the optimization target is set at least based on the operating cost of the power grid, the constraints are set at least based on the power balance constraint, and the power grid is a device for connecting the electricity generated by the new energy power station to the grid.

[0077] Specifically, the construction process of the power grid dispatching model needs to consider the following key steps: First, the optimization target needs to be determined. Generally, the optimization target of power grid dispatching is to minimize the operating cost of the power grid, including fuel cost, transmission loss, equipment operating cost, etc. According to the actual situation, other optimization targets can also be considered, such as maximizing the use of renewable energy power generation, reducing carbon emissions, etc.; secondly, after determining the optimization target, constraints need to be considered. Power balance constraint is the most basic constraint, which ensures the balance of supply and demand of the power grid and avoids overload or insufficient power supply; finally, a mathematical model is established: according to the determined optimization target and constraints, a mathematical model of power grid dispatching is established. Mathematical tools such as linear programming, integer programming, and mixed integer programming are usually used to describe the operating status and optimization problems of the power grid to achieve the optimal dispatching solution.

[0078] The execution unit 40 at least inputs the above power generation prediction result into the above scheduling model to obtain a scheduling plan, and executes the above scheduling plan to control the above power grid. The above scheduling plan is the amount of electricity that the above seventh power generation and the above eighth power generation need to be connected to the grid respectively.

[0079] Specifically, first, the power generation forecast results are input into the dispatch model, which generates a dispatch plan based on the forecast data and the grid demand. This dispatch plan includes the amount of electricity that needs to be connected to the grid for the seventh and eighth generation. Then, according to this dispatch plan, corresponding operations are performed, such as adjusting the output power of the generator and coordinating the operation between different power generation sources to ensure that the grid operates normally and meets demand. Through this process, the power generation forecast results can be effectively used to optimize the operation of the grid and improve the reliability and efficiency of the grid.

[0080] In the above-mentioned embodiment, firstly, first data of a new energy power station within a first historical time period is obtained, the above-mentioned new energy power station includes a photovoltaic power station and a wind power station, and the above-mentioned first data includes at least a first power generation of the above-mentioned photovoltaic power station and a second power generation of the above-mentioned wind power station; secondly, the above-mentioned first data is processed using a prediction model to obtain a power generation prediction result for the first scheduled day after the above-mentioned first historical time period, the above-mentioned prediction model is obtained by training using multiple groups of historical data, each group of the above-mentioned historical data includes second data within a second historical time period and third data on a second scheduled day, the above-mentioned second data includes at least a third power generation of the above-mentioned photovoltaic power station and a fourth power generation of the above-mentioned wind power station on each day within the above-mentioned second historical time period, and the above-mentioned third data includes at least a fifth power generation of the above-mentioned photovoltaic power station and a sixth power generation of the above-mentioned wind power station on the above-mentioned second scheduled day. The power generation prediction result includes the seventh power generation of the photovoltaic power station and the eighth power generation of the wind power station. The first historical time period and the second historical time period are different historical time periods with the same duration. The time interval between the last day of the first historical time period and the first scheduled day is the first interval. The time interval between the last day of the second historical time period and the second scheduled day is the second interval. The first interval is the same as the second interval. A scheduling model for a power grid is constructed. The scheduling model includes optimization objectives and constraints. The power grid is a device for connecting the power generated by the new energy power station to the grid. At least the power generation prediction result is input into the scheduling model to obtain a scheduling plan. The scheduling plan is executed to control the power grid. The scheduling plan is the amount of power that the seventh power generation and the eighth power generation need to be connected to the grid respectively. Compared with the prior art, the solution of the present application utilizes the historical power generation of photovoltaic power stations and the historical power generation of wind power stations to train a convolutional neural network model to obtain a prediction model, and uses the prediction model to predict the power generation of photovoltaic power stations and wind power stations on the scheduled day respectively, to obtain prediction results, and uses the prediction results to make a scheduling plan for the power grid before the scheduled day, thereby achieving more accurate control of the power grid, thereby further solving the problem of low accuracy of power grid control in the prior art.

[0081] In one embodiment of the present application, the acquisition unit further includes an acquisition module, a preprocessing module, an input module and a training module, wherein the acquisition module is used to execute the acquisition step to acquire multiple sets of the above historical data; the processing module is used to execute the preprocessing step to preprocess the above second data in the multiple sets of the above historical data to obtain multiple sets of fourth data; the input module is used to execute the input step to input the multiple sets of the above fourth data into the convolutional neural network model; the training module is used to execute the training step to train the above convolutional neural network model to obtain the above prediction model. In this solution, a prediction model is obtained by training with multiple sets of historical data, which can more accurately predict the power generation situation in a specific time period in the future.

[0082] In order to further improve the quality and availability of the second data, in one embodiment of the present application, the above-mentioned preprocessing module includes a data cleaning submodule, a supplementation submodule and a data normalization submodule, wherein the data cleaning submodule is used to perform data cleaning on the above-mentioned second data to obtain the cleaned second data; the supplementation submodule is used to supplement the missing data in the cleaned second data to obtain the supplemented second data; the data normalization submodule is used to perform data normalization on the supplemented second data to obtain the above-mentioned fourth data.

[0083] Specifically, data cleaning involves filtering out abnormal points in the data, deleting the data of the abnormal points, and then supplementing the data at that moment with the average value of the abnormal point at the previous moment and the value at the next moment. Data cleaning also includes processing duplicate data by deletion.

[0084] Specifically, data completion is as follows: for missing data, the average value of the value at the previous moment and the value at the next moment is used to complete the data at that moment.

[0085] Specifically, the formula for data normalization is:

[0086]

[0087] Among them, x in is the normalized data, x is the original data, and x min is the minimum value in the data, x max is the maximum value in the data.

[0088] In another embodiment, the training module further includes a training submodule, a comparison submodule and a judgment submodule, wherein the training submodule is used to set at least the number of iterations, train the convolutional neural network, and obtain at least the ninth power generation of the photovoltaic power station and the tenth power generation of the wind power station on the second scheduled day, and the corresponding loss function curve when the training number reaches the number of iterations, wherein each group of the fourth data includes one ninth power generation and one tenth power generation; the training submodule is used to perform a comparison step, respectively comparing the ninth power generation and the tenth power generation obtained by training each group of the fourth data with the third data corresponding to each group of the fourth data, to obtain multiple first differences, and performing an average operation on the multiple first differences to obtain a second difference; the judgment submodule is used to perform a judgment step, and determines whether the trained convolutional neural network model is the prediction model according to whether the second difference is within a preset range and whether the loss function curve is in a convergence state, and the preset range is set according to the actual situation. In this scheme, through the steps of training and evaluating the model, it is further ensured that the model has higher prediction accuracy and stability.

[0089] In order to more effectively improve the performance of the model, in one embodiment of the present application, the above-mentioned judgment submodule includes a determination submodule and a repeated execution submodule, wherein the judgment submodule is used to determine that the trained convolutional neural network model is the above-mentioned prediction model when the above-mentioned second difference is within the above-mentioned preset range and the above-mentioned loss function curve is in a convergence state; the repeated execution submodule is used to adjust the parameters of the above-mentioned convolutional neural network model when the above-mentioned second difference is not within the above-mentioned preset range or the above-mentioned loss function curve is not in a convergence state, and repeatedly execute the above-mentioned input step, the above-mentioned training step and the above-mentioned comparison step at least once in sequence until the above-mentioned second difference is within the above-mentioned preset range, and determine that the trained convolutional neural network model is the above-mentioned prediction model, and the above-mentioned parameters include at least the above-mentioned number of iterations.

[0090] Specifically, when the convolutional neural network model converges, at least three indicators, namely, mean square error, mean absolute percentage error, and prediction accuracy, meet the requirements. Among them, mean square error is a commonly used indicator for evaluating the difference between the model's predicted value and the actual value. The smaller the value, the smaller the difference between the model's predicted value and the actual value, and the higher the degree of model convergence; mean absolute percentage error is an indicator for evaluating the percentage difference between the model's predicted value and the actual value. The smaller the value, the smaller the percentage difference between the model's predicted value and the actual value, and the higher the degree of model convergence; prediction accuracy is a comprehensive indicator, usually including accuracy, recall, F1 value and other indicators. The higher the prediction accuracy, the higher the accuracy of the model in classification or regression tasks, and the better the degree of model convergence.

[0091] In another embodiment, the construction unit includes a first construction module, a second construction module and a third construction module, wherein the first construction module is used to construct the operating cost of the above-mentioned power grid according to the loss cost of the above-mentioned power grid and the power purchase cost and power sales income of the above-mentioned power grid, wherein the operating cost C of the above-mentioned power grid w The calculation formula is C W =C K +C T , where C K is the loss cost of the above power grid, C T is the electricity purchase cost and electricity sales income of the above-mentioned power grid; the second construction module is used to construct the new energy consumption cost of the above-mentioned power grid according to at least the consumption capacity of the above-mentioned power grid for the power generation of the above-mentioned photovoltaic power station and the consumption capacity of the above-mentioned power grid for the power generation of the above-mentioned wind power station, wherein the new energy consumption cost C′ Dis The calculation formula is Where P WT,dis (t) is the power grid’s ability to absorb the power generated by the photovoltaic power station, P PV,dis (t) is the power grid's ability to absorb the power generated by the wind power station, and C(t) is the electricity price at time t; the third construction module is used to construct the optimization target of the scheduling model according to the minimum value of the sum of the operation cost of the power grid and the new energy absorption cost. In this solution, the operation efficiency and reliability of the power grid are further improved by setting the optimization target.

[0092] Specifically, first of all, the optimization goal is to minimize the total grid operation cost and give priority to the consumption of new energy. The total grid operation cost is composed of grid operation cost, loss cost, power purchase cost and power sales income. The consumption cost of new energy is jointly determined by the photovoltaic power generation consumption capacity, wind power generation consumption capacity and electricity price.

[0093] In order to further improve the efficiency of the power grid system, in one embodiment of the present application, the construction unit includes a fourth construction module, a fifth construction module and a sixth construction module, wherein the fourth construction module is used to construct a first constraint condition according to the maximum power generation of the photovoltaic power station and the seventh power generation, wherein the first constraint condition is P PV (t)≤P PV,pre (t), where P PV (t) is the maximum power generation of the above photovoltaic power station, P PV,pre (t) is the seventh power generation; the fifth building module is used to build a second constraint condition according to the maximum power generation of the wind power station and the eighth power generation, wherein the second constraint condition is P WT (t)≤P WT,pre (t), where PWT (t) is the maximum power generation of the wind power station, P WT,pre (t) is the eighth power generation; the sixth construction module is used to construct the power balance constraint according to at least the power purchase power of the power grid and the energy storage power on the user side, the first constraint condition, the second constraint condition and the power balance constraint constitute the constraint condition of the scheduling model, wherein the power balance constraint is P WT (t)+P PV (t)+P g (t) = P ESS (t)+P EV (t)+P load (t)+P loss (t), where P g (t) is the power purchased by the above power grid, P ESS (t) is the energy storage power on the user side, P load (t) is the power of the fixed load on the user side, P loss (t) is the power lost on the user side, which is the power lost during the transmission of electricity to the user side.

[0094] Specifically, the constraints include photovoltaic output constraints, wind output constraints and power balance constraints. Photovoltaic output constraints ensure that photovoltaic output does not exceed the forecast data, wind output constraints ensure that wind output does not exceed the forecast data, and power balance constraints ensure the balance between power purchased by the grid, user-side energy storage power, fixed load power and loss power.

[0095] The control device of the power grid includes a processor and a memory. The acquisition unit, processing unit, construction unit and execution unit are all stored in the memory as program units, and the processor executes the program units stored in the memory to implement corresponding functions. The modules are all located in the same processor; or, the modules are located in different processors in any combination.

[0096] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be provided, and the problem of low precision of power grid control in the prior art is solved by adjusting kernel parameters.

[0097] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0098] An embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the power grid control method, the method comprising:

[0099] Step S201, obtaining first data of a new energy power station within a first historical time period, the new energy power station including a photovoltaic power station and a wind power station, and the first data at least includes a first power generation of the photovoltaic power station and a second power generation of the wind power station.

[0100] Step S202, using a prediction model to process the first data to obtain a power generation prediction result for the first scheduled day after the first historical time period, the prediction model is obtained by training using multiple groups of historical data, each group of historical data includes second data in a second historical time period and third data on a second scheduled day, the second data at least includes a third power generation of the photovoltaic power station and a fourth power generation of the wind power station on each day in the second historical time period, the third data at least includes a fifth power generation of the photovoltaic power station and a sixth power generation of the wind power station on the second scheduled day, the power generation prediction result includes a seventh power generation of the photovoltaic power station and an eighth power generation of the wind power station, the first historical time period and the second historical time period are different historical time periods and have the same duration, the time interval between the last day of the first historical time period and the first scheduled day is a first interval, the time interval between the last day of the second historical time period and the second scheduled day is a second interval, and the first interval is the same as the second interval.

[0101] Step S203, constructing a dispatching model for the power grid, the dispatching model includes optimization objectives and constraints, the optimization objectives are set at least based on the operating cost of the power grid, the constraints are set at least based on the power balance constraint, and the power grid is a device for connecting the electricity generated by the new energy power station to the grid.

[0102] Step S204, at least input the above-mentioned power generation prediction result into the above-mentioned scheduling model to obtain a scheduling plan, and execute the above-mentioned scheduling plan to control the above-mentioned power grid. The above-mentioned scheduling plan is the amount of electricity that the above-mentioned seventh power generation and the above-mentioned eighth power generation need to be connected to the grid respectively.

[0103] An embodiment of the present invention provides a processor, the processor is used to run a program, wherein when the program is run, the control method of the power grid is executed, the method comprising:

[0104] Step S201, obtaining first data of a new energy power station within a first historical time period, the new energy power station including a photovoltaic power station and a wind power station, and the first data at least includes a first power generation of the photovoltaic power station and a second power generation of the wind power station.

[0105] Step S202, using a prediction model to process the first data to obtain a power generation prediction result for the first scheduled day after the first historical time period, the prediction model is obtained by training using multiple groups of historical data, each group of historical data includes second data in a second historical time period and third data on a second scheduled day, the second data at least includes a third power generation of the photovoltaic power station and a fourth power generation of the wind power station on each day in the second historical time period, the third data at least includes a fifth power generation of the photovoltaic power station and a sixth power generation of the wind power station on the second scheduled day, the power generation prediction result includes a seventh power generation of the photovoltaic power station and an eighth power generation of the wind power station, the first historical time period and the second historical time period are different historical time periods and have the same duration, the time interval between the last day of the first historical time period and the first scheduled day is a first interval, the time interval between the last day of the second historical time period and the second scheduled day is a second interval, and the first interval is the same as the second interval.

[0106] Step S203, constructing a dispatching model for the power grid, the dispatching model includes optimization objectives and constraints, the optimization objectives are set at least based on the operating cost of the power grid, the constraints are set at least based on the power balance constraint, and the power grid is a device for connecting the electricity generated by the new energy power station to the grid.

[0107] Step S204, at least input the above-mentioned power generation prediction result into the above-mentioned scheduling model to obtain a scheduling plan, and execute the above-mentioned scheduling plan to control the above-mentioned power grid. The above-mentioned scheduling plan is the amount of electricity that the above-mentioned seventh power generation and the above-mentioned eighth power generation need to be connected to the grid respectively.

[0108] An embodiment of the present invention provides a device, the device comprising a processor, a memory, and a program stored in the memory and executable on the processor, and when the processor executes the program, at least the following steps are implemented:

[0109] Step S201, obtaining first data of a new energy power station within a first historical time period, the new energy power station including a photovoltaic power station and a wind power station, and the first data at least includes a first power generation of the photovoltaic power station and a second power generation of the wind power station.

[0110] Step S202, using a prediction model to process the first data to obtain a power generation prediction result for the first scheduled day after the first historical time period, the prediction model is obtained by training using multiple groups of historical data, each group of historical data includes second data in a second historical time period and third data on a second scheduled day, the second data at least includes a third power generation of the photovoltaic power station and a fourth power generation of the wind power station on each day in the second historical time period, the third data at least includes a fifth power generation of the photovoltaic power station and a sixth power generation of the wind power station on the second scheduled day, the power generation prediction result includes a seventh power generation of the photovoltaic power station and an eighth power generation of the wind power station, the first historical time period and the second historical time period are different historical time periods and have the same duration, the time interval between the last day of the first historical time period and the first scheduled day is a first interval, the time interval between the last day of the second historical time period and the second scheduled day is a second interval, and the first interval is the same as the second interval.

[0111] Step S203, constructing a dispatching model for the power grid, the dispatching model includes optimization objectives and constraints, the optimization objectives are set at least based on the operating cost of the power grid, the constraints are set at least based on the power balance constraint, and the power grid is a device for connecting the electricity generated by the new energy power station to the grid.

[0112] Step S204, at least input the above-mentioned power generation prediction result into the above-mentioned scheduling model to obtain a scheduling plan, and execute the above-mentioned scheduling plan to control the above-mentioned power grid. The above-mentioned scheduling plan is the amount of electricity that the above-mentioned seventh power generation and the above-mentioned eighth power generation need to be connected to the grid respectively.

[0113] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0114] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program for initializing at least the following method steps:

[0115] Step S201, obtaining first data of a new energy power station within a first historical time period, the new energy power station including a photovoltaic power station and a wind power station, and the first data at least includes a first power generation of the photovoltaic power station and a second power generation of the wind power station.

[0116] Step S202, using a prediction model to process the first data to obtain a power generation prediction result for the first scheduled day after the first historical time period, the prediction model is obtained by training using multiple groups of historical data, each group of historical data includes second data in a second historical time period and third data on a second scheduled day, the second data at least includes a third power generation of the photovoltaic power station and a fourth power generation of the wind power station on each day in the second historical time period, the third data at least includes a fifth power generation of the photovoltaic power station and a sixth power generation of the wind power station on the second scheduled day, the power generation prediction result includes a seventh power generation of the photovoltaic power station and an eighth power generation of the wind power station, the first historical time period and the second historical time period are different historical time periods and have the same duration, the time interval between the last day of the first historical time period and the first scheduled day is a first interval, the time interval between the last day of the second historical time period and the second scheduled day is a second interval, and the first interval is the same as the second interval.

[0117] Step S203, constructing a dispatching model for the power grid, the dispatching model includes optimization objectives and constraints, the optimization objectives are set at least based on the operating cost of the power grid, the constraints are set at least based on the power balance constraint, and the power grid is a device for connecting the electricity generated by the new energy power station to the grid.

[0118] Step S204, at least input the above-mentioned power generation prediction result into the above-mentioned scheduling model to obtain a scheduling plan, and execute the above-mentioned scheduling plan to control the above-mentioned power grid. The above-mentioned scheduling plan is the amount of electricity that the above-mentioned seventh power generation and the above-mentioned eighth power generation need to be connected to the grid respectively.

[0119] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0120] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0121] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0122] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0123] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0124] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0125] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0126] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0127] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0128] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0129] In the above-mentioned embodiment, firstly, first data of a new energy power station within a first historical time period is obtained, the above-mentioned new energy power station includes a photovoltaic power station and a wind power station, and the above-mentioned first data includes at least a first power generation of the above-mentioned photovoltaic power station and a second power generation of the above-mentioned wind power station; secondly, the above-mentioned first data is processed using a prediction model to obtain a power generation prediction result for the first scheduled day after the above-mentioned first historical time period, the above-mentioned prediction model is obtained by training using multiple groups of historical data, each group of the above-mentioned historical data includes second data within a second historical time period and third data on a second scheduled day, the above-mentioned second data includes at least a third power generation of the above-mentioned photovoltaic power station and a fourth power generation of the above-mentioned wind power station on each day within the above-mentioned second historical time period, and the above-mentioned third data includes at least a fifth power generation of the above-mentioned photovoltaic power station and a sixth power generation of the above-mentioned wind power station on the above-mentioned second scheduled day. The power generation prediction result includes the seventh power generation of the photovoltaic power station and the eighth power generation of the wind power station. The first historical time period and the second historical time period are different historical time periods with the same duration. The time interval between the last day of the first historical time period and the first scheduled day is the first interval. The time interval between the last day of the second historical time period and the second scheduled day is the second interval. The first interval is the same as the second interval. A scheduling model for a power grid is constructed. The scheduling model includes optimization objectives and constraints. The power grid is a device for connecting the power generated by the new energy power station to the grid. At least the power generation prediction result is input into the scheduling model to obtain a scheduling plan. The scheduling plan is executed to control the power grid. The scheduling plan is the amount of power that the seventh power generation and the eighth power generation need to be connected to the grid respectively. Compared with the prior art, the solution of the present application utilizes the historical power generation of photovoltaic power stations and the historical power generation of wind power stations to train a convolutional neural network model to obtain a prediction model, and uses the prediction model to predict the power generation of photovoltaic power stations and wind power stations on the scheduled day respectively, to obtain prediction results, and uses the prediction results to make a scheduling plan for the power grid before the scheduled day, thereby achieving more accurate control of the power grid, thereby further solving the problem of low accuracy of power grid control in the prior art.

[0130] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for controlling a power grid, characterized in that: include: Acquire first data of a new energy power station in a first historical time period, the new energy power station including a photovoltaic power station and a wind power station, the first data including at least a first power generation of the photovoltaic power station and a second power generation of the wind power station; Processing the first data using a prediction model to obtain a power generation prediction result for a first scheduled day after the first historical time period, the prediction model being obtained by training using multiple groups of historical data, each group of the historical data comprising second data in a second historical time period and third data on a second scheduled day, the second data comprising at least a third power generation of the photovoltaic power station and a fourth power generation of the wind power station on each day in the second historical time period, the third data comprising at least a fifth power generation of the photovoltaic power station and a sixth power generation of the wind power station on the second scheduled day, the power generation prediction result comprising a seventh power generation of the photovoltaic power station and an eighth power generation of the wind power station, the first historical time period and the second historical time period being different historical time periods and having the same duration, the time interval between the last day of the first historical time period and the first scheduled day being a first interval, the time interval between the last day of the second historical time period and the second scheduled day being a second interval, and the first interval being the same as the second interval; Constructing a dispatch model of a power grid, the dispatch model comprising an optimization target and a constraint condition, the optimization target being set at least according to the operation cost of the power grid, the constraint condition being set at least according to the power balance constraint, and the power grid being a device for connecting the power generated by the new energy power station to the grid; At least the power generation prediction result is input into the scheduling model to obtain a scheduling plan, and the scheduling plan is executed to control the power grid, wherein the scheduling plan is the amount of electricity that the seventh power generation and the eighth power generation need to be connected to the grid respectively.

2. The control method of the power grid according to claim 1, characterized in that: Before processing the first data using the prediction model to obtain a power generation prediction result for a predetermined day after the historical time period, the method further includes: An acquisition step of acquiring multiple sets of the historical data; a preprocessing step of performing data preprocessing on the second data in the plurality of groups of historical data to obtain a plurality of groups of fourth data; An input step of inputting a plurality of sets of the fourth data into a convolutional neural network model; The training step is to train the convolutional neural network model to obtain the prediction model.

3. The control method of the power grid according to claim 2, characterized in that: The pre-processing step comprises: performing data cleaning on the second data to obtain cleaned second data; Complementing missing data in the cleaned second data to obtain the supplemented second data; The supplemented second data is normalized to obtain the fourth data.

4. The control method of the power grid according to claim 2, characterized in that: The training step also includes: At least setting the number of iterations, training the convolutional neural network, obtaining at least a plurality of ninth power generation of the photovoltaic power station and a tenth power generation of the wind power station on the second predetermined day, and a corresponding loss function curve when the number of training times reaches the number of iterations, wherein each set of the fourth data includes a ninth power generation and a tenth power generation; a comparison step of comparing the ninth power generation and the tenth power generation obtained by training each group of the fourth data with the third data corresponding to each group of the fourth data to obtain a plurality of first differences, and performing an average operation on the plurality of first differences to obtain a second difference; A judgment step is performed to determine whether the trained convolutional neural network model is the prediction model according to whether the second difference is within a preset range and whether the loss function curve is in a convergence state, and the preset range is set according to actual conditions.

5. The control method of the power grid according to claim 4, characterized in that: The judging step comprises: When the second difference is within the preset range and the loss function curve is in a convergence state, determining that the trained convolutional neural network model is the prediction model; When the second difference is not within the preset range or the loss function curve is not in a convergence state, adjust the parameters of the convolutional neural network model, and repeat the input step, the training step and the comparison step at least once in sequence until the second difference is within the preset range, determine that the trained convolutional neural network model is the prediction model, and the parameters include at least the number of iterations.

6. The control method of the power grid according to claim 1, characterized in that: Constructing the optimization goal of the scheduling model includes: The operation cost of the power grid is constructed according to the loss cost of the power grid and the power purchase cost and power sales income of the power grid, wherein the operation cost C of the power grid is w The calculation formula is C W =C K +C T , where C K is the loss cost of the power grid, C T The electricity purchase cost and electricity sales income of the power grid; At least according to the power grid's ability to absorb the power generated by the photovoltaic power station and the power grid's ability to absorb the power generated by the wind power station, the power grid's new energy absorption cost is constructed, wherein the calculation formula of the new energy absorption cost is C′ Dis for Where P WT,dis (t) is the power grid's ability to absorb the power generated by the photovoltaic power station, P PV,dis (t) is the power grid’s ability to absorb the power generated by the wind power station, and C(t) is the electricity price at time t; The optimization target of the scheduling model is constructed according to the lowest value of the sum of the operating cost of the power grid and the new energy consumption cost.

7. The control method of the power grid according to claim 1, characterized in that: The constraint conditions for constructing the scheduling model include: According to the maximum power generation of the photovoltaic power station and the seventh power generation, a first constraint condition is constructed, wherein the first constraint condition is P PV (t)≤P PV,pre (t), where P PV (t) is the maximum power generation of the photovoltaic power station, P PV,pre (t) is the seventh power generation; According to the maximum power generation of the wind power station and the eighth power generation, a second constraint condition is constructed, wherein the second constraint condition is P WT (t)≤P WT,pre (t), where P WT (t) is the maximum power generation of the wind power station, P WT,pre (t) is the eighth power generation; At least according to the purchased power of the power grid and the energy storage power on the user side, the power balance constraint is constructed, and the first constraint condition, the second constraint condition and the power balance constraint constitute the constraint condition of the scheduling model, wherein the power balance constraint is P WT (t)+P PV (t)+P g (t) = P ESS (t)+P EV (t)+P load (t)+P loss (t), where P g (t) is the power purchased by the power grid, P ESS (t) is the energy storage power at the user side, P load (t) is the power of the fixed load on the user side, P loss (t) is the power lost on the user side, and the power lost on the user side is the power lost in the process of transmitting electricity to the user side.

8. A control device for a power grid, characterized in that: include: An acquisition unit, which acquires first data of a new energy power station in a first historical time period, wherein the new energy power station includes a photovoltaic power station and a wind power station, and the first data includes at least a first power generation of the photovoltaic power station and a second power generation of the wind power station; a processing unit, processing the first data using a prediction model to obtain a power generation prediction result for a first scheduled day after the first historical time period, the prediction model being obtained by training using multiple groups of historical data, each group of the historical data comprising second data in a second historical time period and third data on a second scheduled day, the second data comprising at least a third power generation of the photovoltaic power station and a fourth power generation of the wind power station on each day in the second historical time period, the third data comprising at least a fifth power generation of the photovoltaic power station and a sixth power generation of the wind power station on the second scheduled day, the power generation prediction result comprising a seventh power generation of the photovoltaic power station and an eighth power generation of the wind power station, the first historical time period and the second historical time period being different historical time periods and having the same duration, the time interval between the last day of the first historical time period and the first scheduled day being a first interval, the time interval between the last day of the second historical time period and the second scheduled day being a second interval, and the first interval being the same as the second interval; A construction unit is configured to construct a dispatching model of a power grid, wherein the dispatching model includes an optimization target and a constraint condition, wherein the optimization target is set at least according to an operation cost of the power grid, and the constraint condition is set at least according to a power balance constraint, and the power grid is a device for connecting the power generated by the new energy power station to the grid; An execution unit at least inputs the power generation prediction result into the scheduling model to obtain a scheduling plan, and executes the scheduling plan to control the power grid, wherein the scheduling plan is the amount of electricity that the seventh power generation and the eighth power generation need to be connected to the grid respectively.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the power grid control method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing the control method of the power grid described in any one of claims 1 to 7.

Images