Source network load storage coordination control method, device, equipment, medium and program product

Abstract

This invention relates to the field of power system technology and discloses a source-grid-load-storage coordinated control method, device, equipment, medium, and program product. The source-grid-load-storage coordinated control method includes: predicting the power generation of power supply equipment to obtain a power generation prediction sequence for a prediction time period; predicting the load of load equipment to obtain a load prediction sequence for a prediction time period; constructing a first objective function and multiple constraints; and optimizing the control strategy of the source-grid-load-storage system based on the first objective function, multiple constraints, power generation prediction sequence, and load prediction sequence to obtain a first objective control strategy for a first preset time period, so as to perform source-grid-load-storage coordinated control according to the first objective control strategy. This invention obtains a more accurate source-grid-load-storage coordinated control strategy by predicting power generation and load, constructing an objective function and constraints, and optimizing the solution, thereby improving the accuracy of source-grid-load-storage coordinated control.

Description

Technical Field

[0001] This invention relates to the field of power system technology, specifically to methods, devices, equipment, media, and program products for coordinated control of power generation, grid, load, and storage. Background Technology

[0002] The source-grid-load-storage (PGS) model is a new energy system operation mode with the core objective of balancing energy supply and demand and achieving efficient utilization. It integrates four core elements—power source, grid, load, and energy storage—to achieve multi-stage synergistic interaction. Due to the intermittent and fluctuating nature of renewable energy sources, coordinated control of the source-grid-load-storage system is necessary during its operation.

[0003] In related technologies, the method for coordinated control of power generation, grid, load, and energy storage is hierarchical and zoned control. Shallow machine learning models are used for prediction on the power generation side, while time series forecasting methods are used on the load side, resulting in low prediction accuracy on both sides. At the coordinated control level, based on simple rule-based strategies or linear programming models, with grid security as the primary constraint, limited power balance scheduling is implemented. However, due to low prediction accuracy, simple rules and constraints, this approach does not conform to the complex operating conditions of power generation, grid, load, and energy storage, leading to low precision in coordinated control. Summary of the Invention

[0004] This invention provides a source-grid-load-storage coordinated control method, apparatus, equipment, medium, and program product to solve the problem of low coordinated control accuracy of source-grid-load-storage caused by related technologies.

[0005] In a first aspect, the present invention provides a source-grid-load-storage coordinated control method, comprising: predicting the power generation of the power equipment based on historical power data of the power equipment in the source-grid-load-storage system, weather forecast data of the area where the power equipment is located, and real-time meteorological data, to obtain a power generation prediction sequence for a prediction period; predicting the load of the load equipment based on historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type, to obtain a load prediction sequence for a prediction period; constructing a first objective function based on various cost information of the source-grid-load-storage system, and constructing multiple constraints based on the operating information of the source-grid-load-storage system; and optimizing the control strategy of the source-grid-load-storage system based on the first objective function, multiple constraints, power generation prediction sequence, and load prediction sequence to obtain a first objective control strategy for a first preset period, so as to perform source-grid-load-storage coordinated control according to the first objective control strategy.

[0006] The source-grid-load-storage coordinated control method of this invention predicts the power generation of power equipment based on historical power data, weather forecast data, and real-time meteorological data of the power equipment's location within the source-grid-load-storage system. This yields a power generation prediction sequence for the predicted time period. By combining weather forecast data and real-time meteorological data of the power equipment's location, this invention effectively improves the accuracy of power generation prediction, especially for new energy power equipment, as it considers the weather-dependent characteristics of power generation, making the prediction results more closely reflect actual power generation conditions. Furthermore, this invention predicts the load of load equipment based on historical load data, real-time load data, meteorological data of the load equipment's location, and date type, obtaining a load prediction sequence for the predicted time period. By integrating not only historical and real-time load data but also regional meteorological data and date type information, the load prediction sequence more closely matches actual electricity consumption patterns, effectively reducing load prediction errors. This invention constructs a first objective function based on various cost information of the power generation, grid, load, and storage system, and establishes multiple constraints based on the system's operational information. The first objective function uses cost as the core optimization objective, while the constraints cover technical and safety requirements such as grid security, equipment operating limits, and energy supply-demand balance. This avoids the risk of solely pursuing the lowest cost while neglecting system stability, providing a clear optimization direction and boundary for the scheduling of the power generation, grid, load, and storage system. Based on the first objective function, multiple constraints, power generation prediction sequences, and load prediction sequences, this invention optimizes the control strategy of the power generation, grid, load, and storage system to obtain a first objective control strategy for a first preset time period. Coordinated control of the power generation, grid, load, and storage system based on this first objective control strategy allows for a global consideration of the power supply capacity on the generation side, the electricity demand on the load side, economic cost objectives, and system safety constraints. The resulting first objective control strategy achieves optimal cost and efficient resource utilization for the entire system while ensuring safe and stable operation, thereby improving the accuracy of coordinated control of the power generation, grid, load, and storage system.

[0007] In one optional implementation, the source-grid-load-storage coordinated control method further includes: constructing a second objective function based on the power deviation information of the power supply equipment in the source-grid-load-storage system and the adjustment cost of adjustable resources; optimizing the control strategy of the source-grid-load-storage system based on the second objective function, the power generation prediction sequence, the load prediction sequence, the real-time operation information of the source-grid-load-storage system, the scheduling plan information, and the adjustable resource information to obtain a second objective control strategy for a second preset time period, and performing source-grid-load-storage coordinated control based on the second objective control strategy; the time interval of the second preset time period is shorter than the time interval of the first preset time period.

[0008] This invention constructs a second objective function based on the power deviation information of the power supply equipment in the power generation, grid, load, and storage system and the adjustment cost of adjustable resources. It incorporates the real-time deviation of system operation and economic adjustment costs into the optimization objective, focusing on the power imbalance problem in real-time operation. Based on the second objective function, the power generation prediction sequence, the load prediction sequence, the real-time operation information of the power generation, grid, load, and storage system, the scheduling plan information, and the adjustable resource information, the control strategy of the power generation, grid, load, and storage system is optimized and solved to obtain a second objective control strategy for a second preset time period. Through a short-cycle rolling optimization mechanism, the second objective control strategy can quickly respond to sudden fluctuations in power generation and load, and promptly call upon adjustable resources for dynamic balancing, thereby improving the accuracy and flexibility of the coordinated control of the power generation, grid, load, and storage system.

[0009] In one optional implementation, the power generation of the power equipment in the source-grid-load-storage system is predicted based on historical power data, weather forecast data of the area where the power equipment is located, and real-time meteorological data, to obtain a power generation prediction sequence for the prediction period. This includes: inputting historical power data, weather forecast data of the area where the power equipment is located, and real-time meteorological data into a trained power prediction model to obtain a power generation prediction sequence for the prediction period; the power prediction model is a neural network model for power generation prediction built based on generative adversarial networks and attention mechanisms.

[0010] This invention uses a power prediction model based on generative adversarial networks and attention mechanisms to predict power generation. Generative adversarial networks can simulate complex power fluctuation patterns through adversarial training, while attention mechanisms can automatically focus on key meteorological factors and historical periods that have the greatest impact on power generation, effectively improving the prediction accuracy for the strong volatility and intermittency of new energy power generation.

[0011] In one optional implementation, the load of the load equipment in the source-grid-load-storage system is predicted based on historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type, to obtain a load prediction sequence for the prediction period. This includes: inputting historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type into a trained load prediction model to obtain a load prediction sequence for the prediction period; the load prediction model is a neural network model for load prediction built based on a variable attention mechanism.

[0012] This invention uses a load forecasting model based on a variable attention mechanism to forecast load. The variable attention mechanism can dynamically identify the different impacts of meteorological data and date type on load at different times, so that the load forecasting model can more accurately match actual electricity consumption behavior and reduce load forecasting errors.

[0013] In one optional implementation, multiple cost information includes variables such as electricity purchase cost, generator fuel cost, energy storage depreciation cost, wind and solar curtailment penalty cost, and green energy environmental benefit; multiple constraints include power balance constraints, output constraints, ramp rate constraints, and exchange power constraints; a first objective function is constructed based on the multiple cost information of the power generation, grid, load, and storage system; multiple constraints are constructed based on the operating information of the power generation, grid, load, and storage system, including: obtaining a summation result based on the sum of the electricity purchase cost, generator fuel cost, energy storage depreciation cost, and wind and solar curtailment penalty cost; and then, based on the summation result and... The difference between the environmental benefit variables of green energy is calculated to obtain the difference result; the difference result is minimized to obtain the first objective function; power balance constraints are constructed based on the sum of the conventional unit output of the power generation, grid-load-storage system, the predicted output of new energy, the curtailed power of new energy, the charging and discharging power of energy storage, and the exchange power equal to the predicted load value; output constraints are constructed based on the conventional unit output of the power generation, grid-load-storage system being within the equipment technical limit range; ramp rate constraints are constructed based on the ramp rate of the power generation, grid-load-storage system being within the preset ramp rate range; and exchange power constraints are constructed based on the exchange power of the power generation, grid-load-storage system being within the preset exchange power range.

[0014] The first objective function of this invention incorporates electricity purchase cost, generator fuel cost, energy storage depreciation cost, and wind / solar curtailment penalty cost into the total cost, while introducing green energy environmental benefits as a positive incentive, thus achieving bidirectional optimization of cost and environmental benefits. Multi-dimensional constraints cover the full-scenario safety requirements of the power generation, grid, load, and storage system, ensuring that the control strategy, while optimizing economic objectives, always meets the boundaries of stable grid operation.

[0015] In one optional implementation, the control strategy of the power generation, grid, load and storage system is optimized and solved based on the first objective function, multiple constraints, power generation prediction sequence and load prediction sequence to obtain the first objective control strategy for the first preset time period. This includes: taking the first objective function as the optimization objective, multiple constraints as constraints, and conventional unit output, energy storage charging and discharging power and power exchanged with the main grid as decision variables, inputting the power generation prediction sequence and load prediction sequence into a preset optimization solver for optimization and solving to obtain the first objective control strategy for the first preset time period.

[0016] This invention uses a preset optimization solver to efficiently solve the objective function and constraints, and can output the globally optimal control strategy in a short time. The automated optimization method greatly improves the accuracy and timeliness of scheduling decisions.

[0017] Secondly, the present invention provides a source-grid-load-storage coordinated control device, comprising: a power prediction module, used to predict the power generation of the power equipment based on historical power data of the power equipment in the source-grid-load-storage system, weather forecast data of the area where the power equipment is located, and real-time meteorological data, to obtain a power generation prediction sequence for a prediction period; a load prediction module, used to predict the load of the load equipment based on historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type, to obtain a load prediction sequence for a prediction period; a condition construction module, used to construct a first objective function based on various cost information of the source-grid-load-storage system, and construct multiple constraints based on the operating information of the source-grid-load-storage system; and a day-ahead strategy generation module, used to optimize and solve the control strategy of the source-grid-load-storage system based on the first objective function, multiple constraints, power generation prediction sequence, and load prediction sequence, to obtain a first objective control strategy for a first preset time period, so as to perform source-grid-load-storage coordinated control according to the first objective control strategy.

[0018] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the source-grid-load-storage coordinated control method of the first aspect or any corresponding embodiment described above.

[0019] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the source-grid-load-storage coordinated control method of the first aspect or any corresponding embodiment thereof.

[0020] Fifthly, the present invention provides a computer program product, including computer instructions, which are used to cause a computer to execute the source-grid-load-storage coordinated control method of the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0021] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of an application scenario according to an embodiment of the present invention;

[0023] Figure 2 This is a schematic diagram of the first type of source-grid-load-storage coordinated control method according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the second process of the source-grid-load-storage coordinated control method according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the third process of the source-grid-load-storage coordinated control method according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the fourth process of the source-grid-load-storage coordinated control method according to an embodiment of the present invention; Figure 6 This is a structural block diagram of a source-grid-load-storage coordinated control device according to an embodiment of the present invention; Figure 7 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0026] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0027] As an optional application scenario of this invention, such as Figure 1 As shown, the source-grid-load-storage coordinated control system may include at least one terminal device and at least one server. Figure 1 The system is illustrated in the example, which includes a computer 101, a mobile terminal 102, and a server 103, and the terminal devices such as the computer 101 and the mobile terminal 102 are connected to the server 103 through a network 110.

[0028] Specifically, the terminal device can be a smartphone, tablet, laptop, PDA, desktop computer, game console, smart TV, smart wearable device, in-vehicle terminal, VR (Virtual Reality) device, AR (Augmented Reality) device, etc. Server 103 can be a standalone physical server, a server cluster, a distributed system, or a cloud server providing cloud services. Network 110 can be a wired or wireless network, examples of which include, but are not limited to, the Internet, corporate intranet, local area network, wide area network, mobile communication network, and combinations thereof.

[0029] With the increasingly widespread application of new energy sources such as wind and solar power in the power system, industrial parks, as important units of energy consumption, are gradually transforming from single load centers into integrated energy entities that combine power generation, power consumption, energy storage, and regulation. The integrated operation model of power generation, grid, load, and storage has emerged as a result.

[0030] In related technologies, the coordinated control of power generation, grid, load, and storage at the park level typically employs a hierarchical and zoned control architecture. On the power supply side, statistical methods based on numerical weather prediction and historical power data, or shallow machine learning models, are commonly used for power forecasting. On the load side, classical time-series-based forecasting methods (such as trend extrapolation and seasonal index methods) or combinations with expert systems are frequently used for load forecasting. At the coordinated control level, most related systems are based on simple rule-based strategies (such as the "wind and solar curtailment" principle) or linear programming models, with grid security as the primary constraint, to perform limited power balance scheduling.

[0031] However, in the coordinated control of power generation, grid, load, and storage at the park level in related technologies, traditional renewable energy power prediction methods rely on accurate geographic and meteorological models, which are complex and difficult to adapt to complex terrain and micro-meteorological changes. Conventional statistical methods (such as time series and shallow neural networks) lack the ability to learn deep spatiotemporal correlation features in historical data, resulting in large prediction errors under sudden weather changes or extreme conditions. The refresh frequency and accuracy of ultra-short-term (0-4 hours) predictions in related technologies are insufficient to meet the needs of real-time control. Load prediction models in related technologies often ignore the diversity and randomness of user behavior within the park, as well as the nonlinear coupling relationship with factors such as meteorology and holidays. Simple time series models cannot effectively handle multi-source heterogeneous data, resulting in significant prediction deviations under special scenarios such as holidays and extreme weather. The control strategies of related technologies are mostly based on grid security as a rigid constraint, and their ability to aggregate and schedule massive distributed, small-capacity adjustable resources (such as distributed photovoltaics, user-side energy storage, electric vehicles, and interruptible loads) is insufficient. Control models are mostly centralized and deterministic optimizations, which are difficult to adapt to the randomness and volatility of both the source and load sides, and lack flexible interaction mechanisms that consider the interests of multiple stakeholders. In related technologies, load and energy storage are often treated as passive objects of regulation, failing to be treated as equal market players and failing to stimulate their initiative to participate in system regulation through price signals or incentive mechanisms. As a result, the flexibility and economy of system operation need to be improved.

[0032] The subsystems of source, network, load, and storage are often built independently, resulting in severe data silos and information islands. The lack of a unified data platform to integrate, govern, and mine massive amounts of data across multiple temporal and spatial scales leads to insufficient decision-making basis. The overall system remains at a rudimentary stage of data display and manual decision-making, or simple rule control, failing to achieve a fully intelligent closed loop from perception and prediction to decision-making and execution.

[0033] This invention provides a source-grid-load-storage coordinated control method. By predicting power generation and load, constructing objective functions and constraints, and optimizing the solution, a more accurate source-grid-load-storage coordinated control strategy is obtained, thereby improving the accuracy of source-grid-load-storage coordinated control.

[0034] According to an embodiment of the present invention, a source-grid-load-storage coordinated control method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0035] This embodiment provides a source-grid-load-storage coordinated control method, which can be used in computer equipment. Figure 2 This is a first flowchart of the source-grid-load-storage coordinated control method according to an embodiment of the present invention, as follows: Figure 2 As shown, the process includes the following steps: Step S201: Based on the historical power data of the power supply equipment in the power source-grid-load-storage system, the weather forecast data of the area where the power supply equipment is located, and the real-time meteorological data, the power generation power of the power supply equipment is predicted to obtain the power generation prediction sequence for the prediction period.

[0036] Among them, the source-grid-load-storage system is an integrated energy system that includes power supply, grid, load and energy storage. For example, power supply refers to power supply equipment, including photovoltaic units, wind turbines and conventional units, load refers to load equipment, including user electrical equipment, and energy storage refers to energy storage systems, including lithium batteries, pumped storage equipment, etc.

[0037] In some optional implementations, the historical power data of the power supply equipment is the actual power generation data of the power supply equipment in the past period; the weather forecast data (Numerical Weather Prediction, NWP) of the area where the power supply equipment is located is the weather forecast data of the area where the power supply equipment is located in the future period; and the real-time weather data is the weather data at the current moment, including measured weather information such as sunshine, wind speed, and temperature.

[0038] In some alternative implementations, the prediction time period can be set according to the actual situation. For example, the prediction time period can be a 48-hour period of the next day.

[0039] In some alternative implementations, the power generation prediction sequence is a set of power generation prediction values ​​arranged in chronological order.

[0040] In some optional implementations, a preset power neural network model is used to predict the power generation of the power supply equipment to obtain a power generation prediction sequence for the prediction period.

[0041] Step S202: Based on the historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type of the load equipment in the source-grid-load-storage system, the load of the load equipment is predicted to obtain the load prediction sequence for the prediction period.

[0042] Among them, the historical load data of the load equipment is the actual power consumption data of the load equipment in the past time period, and the real-time load data is the actual power consumption of the load equipment at the current moment; the meteorological data of the area where the load equipment is located is the weather forecast data of the area where the load equipment is located; the date type is an identifier used to distinguish between weekdays, weekends and holidays; and the load forecast sequence is a set of hourly power load forecast values ​​for the forecast period.

[0043] In some optional implementations, a preset load neural network model is used to predict the load of the load equipment to obtain a load prediction sequence for the prediction period.

[0044] Step S203: Construct a first objective function based on various cost information of the source-grid-load-storage system, and construct multiple constraints based on the operation information of the source-grid-load-storage system.

[0045] The various cost information includes variables such as electricity purchase cost, generator fuel cost, energy storage depreciation cost, wind and solar curtailment penalty cost, and green energy environmental benefits; and multiple constraints include power balance constraints, output constraints, ramp rate constraints, and exchange power constraints.

[0046] Specifically, the electricity purchase cost variable is the cost of purchasing electricity from the main grid, the unit fuel cost variable is the cost of coal or gas for thermal power units, the energy storage depreciation cost variable is the depreciation and amortization of energy storage equipment, the wind and solar curtailment penalty cost variable is the cost variable of not consuming clean energy, and the green energy environmental benefit variable is the environmental benefit variable of consuming clean energy.

[0047] In some alternative implementations, power balance constraints are used to control the supply and demand balance between power generation and load, output constraints are used to limit the maximum or minimum output of equipment, ramp rate constraints are used to limit the rate of change of unit output, and exchange power constraints are used to limit the range of power exchange with the main grid.

[0048] Step S204: Based on the first objective function, multiple constraints, power generation prediction sequence, and load prediction sequence, optimize the control strategy of the source-grid-load-storage system to obtain the first objective control strategy for the first preset time period, and perform source-grid-load-storage coordinated control according to the first objective control strategy.

[0049] The first target control strategy is a scheduling scheme for the output of conventional units, the charging and discharging power of energy storage, and the power exchanged with the main grid. The first preset time period can be set according to the actual situation. For example, the first preset time period can be the time period corresponding to 96 time points of the next day, with each time point being 15 minutes.

[0050] In some optional implementations, the particle swarm optimization algorithm is used to optimize the control strategy of the source-grid-load-storage system to obtain the first target control strategy for the first preset time period.

[0051] The source-grid-load-storage coordinated control method provided in this embodiment predicts the power generation of the power equipment based on historical power data, weather forecast data, and real-time meteorological data of the power equipment's location within the source-grid-load-storage system. This yields a power generation prediction sequence for the predicted time period. By combining weather forecast data and real-time meteorological data of the power equipment's location, this embodiment effectively improves the accuracy of power generation prediction, especially for new energy power equipment, as it considers the weather-dependent characteristics of power generation, making the prediction results more closely reflect actual power generation conditions. Furthermore, this embodiment predicts the load of the load equipment based on historical load data, real-time load data, meteorological data of the load equipment's location, and date type within the source-grid-load-storage system. This not only utilizes historical and real-time load data but also incorporates regional meteorological data and date type information. By integrating these factors, the load prediction sequence more closely matches actual electricity consumption patterns, effectively reducing load prediction errors. This invention constructs a first objective function based on various cost information of the power generation, grid, load, and storage system, and establishes multiple constraints based on the system's operational information. The first objective function uses cost as the core optimization objective, while the constraints cover technical and safety requirements such as grid security, equipment operating limits, and energy supply-demand balance. This avoids the risk of solely pursuing the lowest cost while neglecting system stability, providing a clear optimization direction and boundary for the scheduling of the power generation, grid, load, and storage system. Based on the first objective function, multiple constraints, power generation prediction sequences, and load prediction sequences, this invention optimizes the control strategy of the power generation, grid, load, and storage system to obtain a first objective control strategy for a first preset time period. Coordinated control of the power generation, grid, load, and storage system based on this first objective control strategy allows for a global consideration of the power generation capacity, load demand, economic cost objectives, and system safety constraints. The resulting first objective control strategy achieves optimal cost and efficient resource utilization for the entire system while ensuring safe and stable operation, thereby improving the accuracy of coordinated control of the power generation, grid, load, and storage system.

[0052] This embodiment provides a source-grid-load-storage coordinated control method, which can be used in computer equipment. Figure 3 This is a second flowchart of the source-grid-load-storage coordinated control method according to an embodiment of the present invention, as follows: Figure 3 As shown, the process includes the following steps: Step S301: Based on the historical power data of the power supply equipment in the power source-grid-load-storage system, the weather forecast data of the area where the power supply equipment is located, and the real-time meteorological data, the power generation power of the power supply equipment is predicted to obtain the power generation prediction sequence for the prediction period.

[0053] Specifically, step S301 includes: Step S3011: Input the historical power data of the power supply equipment of the power source-grid-load-storage system, the weather forecast data of the area where the power supply equipment is located, and the real-time meteorological data into the trained power prediction model to obtain the power generation prediction sequence for the prediction period.

[0054] In some optional implementations, the source-grid-load-storage coordinated control method of this invention is applicable to park-level source-grid-load-storage systems. Sensing devices and data acquisition devices are deployed at each node of the park-level source-grid-load-storage system. The system periodically (e.g., every 5 minutes) acquires different data. For example, at the power supply equipment, wind turbine / photovoltaic inverter data acquisition units and weather stations (irradiance meters, anemometers, temperature and humidity sensors) are deployed; on the grid side, smart meters, PMUs (Phasor Measurement Units), and SCADA (Supervisory Control and Data Acquisition) systems are deployed; at the load equipment, smart meters and non-intrusive load monitoring units are deployed; and at the energy storage system, a battery management system is deployed to monitor SOC (State of Charge), charging and discharging power, etc.

[0055] In some optional implementations, the acquired data is preprocessed, specifically by data cleaning (removing outliers and filling in missing values), normalization, and alignment by time scale to form a standard dataset that can be used as input to the model. The data used for power prediction and load prediction are the processed data.

[0056] In some alternative implementations, the power prediction model is a neural network model for predicting power generation based on generative adversarial networks and attention mechanisms.

[0057] The generative adversarial network (GAN) consists of a generator and a discriminator. The generator generates a power generation prediction sequence based on the input data, while the discriminator judges the similarity between the generated power generation prediction sequence and the real historical sequence. Through repeated adversarial training, the generator's prediction output continuously approaches the real value, thereby obtaining the power generation prediction sequence and confidence interval for the prediction period. The attention mechanism is used to capture key features and improve the accuracy of the power generation prediction sequence.

[0058] Step S302: Based on the historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type of the load equipment in the source-grid-load-storage system, the load of the load equipment is predicted to obtain the load prediction sequence for the prediction period.

[0059] Specifically, step S302 includes: Step S3021: Input the historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type of the load equipment into the trained load prediction model to obtain the load prediction sequence for the prediction period.

[0060] The load forecasting model is a neural network model for load forecasting built based on a variable attention mechanism.

[0061] In some alternative implementations, the load forecasting model automatically calculates the attention weights of each input variable. For example, in the summer afternoon, the weights of temperature and solar intensity increase significantly. Based on the weighted features, regression prediction is performed through a deep neural network to output a load forecast sequence for the forecast period.

[0062] Step S303: Construct a first objective function based on various cost information of the source-grid-load-storage system, and construct multiple constraints based on the operation information of the source-grid-load-storage system.

[0063] Specifically, step S303 includes: Step S3031: Based on the sum of the electricity purchase cost variable, the generator fuel cost variable, the energy storage depreciation cost variable, and the wind and solar curtailment penalty cost variable, obtain the summation result; based on the difference between the summation result and the green energy environmental benefit variable, obtain the difference result; minimize the difference result to obtain the first objective function.

[0064] For example, the expression for the first objective function is:

[0065] in, Let the first objective function be... for The variable of electricity purchase cost at a given point in time. for Point-in-time unit fuel cost variables for Point-in-time energy storage depreciation cost variable for The timing of wind and solar power curtailment penalties is a variable. for Green energy environmental benefits variables at a given time point The goal is to find the variable value that minimizes the value of the first objective function.

[0066] Step S3032: Based on the fact that the sum of the conventional unit output of the power source, grid, load and storage system, the predicted output of new energy sources, the abandoned power of new energy sources, the charging and discharging power of energy storage and the exchange power equals the predicted load value, a power balance constraint is constructed.

[0067] For example, the power balance constraint can be expressed as:

[0068] in, To provide power to conventional units, To contribute to the forecasting of new energy sources, For the abandoned power of new energy sources, For the charging and discharging power of energy storage, To exchange power, This represents the load forecast.

[0069] Step S3033: Based on the fact that the output of conventional units in the power generation, grid, load and storage system is within the technical limit range of the equipment, output constraints are constructed.

[0070] For example, the output constraint can be expressed as:

[0071] in, To provide power to conventional units, This represents the lower limit of the equipment's technical limit range. This represents the upper limit of the equipment's technical limits.

[0072] Step S3034: Based on the fact that the ramp rate of the source-grid-load-storage system is within the preset ramp rate range, a ramp rate constraint is constructed.

[0073] For example, the gradeability constraint can be expressed as:

[0074] in, for The output of the conventional units at a given time point, for The output of the conventional units at a given time point, This is the lower limit of the preset gradient rate range. This is the upper limit of the preset gradient rate range.

[0075] Step S3035: Based on the fact that the switching power of the source-grid-load-storage system is within a preset switching power range, a switching power constraint is constructed.

[0076] For example, the switching power constraint can be expressed as:

[0077] in, To exchange power, This is the lower limit of the preset switching power range. This is the upper limit of the preset switching power range.

[0078] Step S304: Based on the first objective function, multiple constraints, power generation prediction sequence, and load prediction sequence, optimize the control strategy of the source-grid-load-storage system to obtain the first objective control strategy for the first preset time period, and perform source-grid-load-storage coordinated control according to the first objective control strategy.

[0079] Specifically, step S304 includes: Step S3041: Using the first objective function as the optimization objective, multiple constraints as constraints, and conventional unit output, energy storage charging and discharging power, and power exchanged with the main grid as decision variables, the power generation prediction sequence and load prediction sequence are input into a preset optimization solver for optimization and solution to obtain the first objective control strategy for the first preset time period.

[0080] The preset optimized solver can be Gurobi, CPLEX, FICO Xpress, etc.

[0081] In some alternative implementations, a preset optimization solver is invoked to solve the mixed-integer linear / quadratic programming problem, resulting in a sequence of decision variables that satisfies all constraints and minimizes the first objective function.

[0082] Specifically, the pre-defined optimization solver first performs preprocessing to simplify the problem structure, eliminate redundant constraints, and narrow the range of variable values. Second, it decomposes the original problem into multiple subproblems and gradually approximates the optimal integer solution by solving the relaxed solutions of the subproblems. Third, it adds linear constraints to tighten the boundaries of the relaxed solutions and reduce the search space. Finally, it quickly finds feasible solutions during the solution process, accelerating the convergence of the optimal solution.

[0083] In some optional implementations, the first target control strategy is the output value of conventional units, the charging and discharging power value of energy storage, and the power exchange value with the main grid. The output value of conventional units represents the power generation arrangement of controllable power sources (such as gas turbines). The charging and discharging power value of energy storage is positive for discharging and negative for charging, which is used to smooth out peaks and valleys and smooth fluctuations. The power exchange value with the main grid is positive for purchasing electricity and negative for selling electricity, which is used to balance power deficits or surpluses and achieve economic interaction.

[0084] In this embodiment of the invention, the first target control strategy is converted into control commands and sent to the corresponding devices for source-grid-load-storage coordinated control.

[0085] The source-grid-load-storage coordinated control method provided in this embodiment predicts power generation using a power prediction model built on generative adversarial networks (GANs) and attention mechanisms. The GANs simulate complex power fluctuation patterns through adversarial training, while the attention mechanism automatically focuses on key meteorological factors and historical periods that have the greatest impact on power generation, effectively improving the prediction accuracy for the strong volatility and intermittency of renewable energy power generation. Load prediction is performed using a load prediction model built on a variable attention mechanism. This mechanism dynamically identifies the differentiated impacts of meteorological data and date types on load at different times, enabling the load prediction model to more accurately reflect actual electricity consumption behavior and reduce load prediction errors. The first objective function incorporates electricity purchase costs, generator fuel costs, energy storage depreciation costs, and wind and solar curtailment penalties into the total cost, while introducing green energy environmental benefits as a positive incentive, achieving bidirectional optimization of cost and environmental benefits. Multi-dimensional constraints cover the full-scenario safety requirements of the source-grid-load-storage system, ensuring that the control strategy, while optimizing economic objectives, always meets the boundaries of stable grid operation. By using a pre-set optimization solver to efficiently solve the objective function and constraints, the system can output the globally optimal control strategy in a short time. This automated optimization method significantly improves the accuracy and timeliness of scheduling decisions.

[0086] This embodiment provides a source-grid-load-storage coordinated control method, which can be used in computer equipment. Figure 4 This is a third flowchart of the source-grid-load-storage coordinated control method according to an embodiment of the present invention, such as... Figure 4 As shown, the process includes the following steps: Step S401: Based on historical power data of the power supply equipment in the power generation, grid, load, and storage system, weather forecast data for the area where the power supply equipment is located, and real-time meteorological data, the power generation capacity of the power supply equipment is predicted to obtain a power generation prediction sequence for the prediction period. For details, please refer to [link to details]. Figure 2 Step S201 of the illustrated embodiment will not be described again here.

[0087] Step S402: Based on the historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type of the load equipment in the source-grid-load-storage system, the load of the load equipment is predicted to obtain a load prediction sequence for the prediction period. For details, please refer to [link to details]. Figure 2 Step S202 of the illustrated embodiment will not be described again here.

[0088] Step S403: Construct a second objective function based on the power deviation information of the power supply equipment in the source-grid-load-storage system and the adjustment cost of adjustable resources.

[0089] For example, the expression for the second objective function can be:

[0090] in, The second objective function is... for The power deviation between actual power and planned power at a given time point. for The adjustment cost of time-adjustable resources As the first weighting coefficient, This is the second weighting coefficient. The goal is to find the variable value that minimizes the value of the first objective function.

[0091] Step S404: Based on the second objective function, the power generation prediction sequence, the load prediction sequence, the real-time operation information of the power generation, grid, load and storage system, the scheduling plan information and the adjustable resource information, the control strategy of the power generation, grid, load and storage system is optimized and solved to obtain the second objective control strategy for the second preset time period, so as to carry out coordinated control of power generation, grid, load and storage according to the second objective control strategy.

[0092] The second preset time period is shorter than the first preset time period. For example, the second preset time period can be the next 1-4 hours.

[0093] In some optional implementations, the second objective control strategy for the second preset time period is obtained by quickly solving the second objective function, the power generation prediction sequence, the load prediction sequence, the real-time operation information of the power generation, grid, load and storage system, the scheduling plan information, and the adjustable resource information using mixed integer linear programming or heuristic algorithms.

[0094] Specifically, the second objective function, the power generation prediction sequence, the load prediction sequence, the real-time operation information of the power generation, grid, load and storage system, the scheduling plan information, and the adjustable resource information are input into the mixed-integer linear programming process or the heuristic algorithm process to obtain the second objective control strategy. The mixed-integer linear programming solver quickly searches for the optimal solution in the constraint space through algorithms such as branch and bound and cutting plane. The heuristic algorithm process generates feasible solutions through iterative optimization, and approximates the optimal solution as closely as possible while meeting the real-time requirements.

[0095] In some optional implementations, the second target control strategy is to issue precise active power control commands for each time period within the next 1–4 hours, including the charging and discharging power of energy storage and the amount of interruptible loads to be cut off. These commands are sent to the corresponding devices for control via a hybrid network of power fiber optic private network and 5G wireless network. For example, on the power supply side: active power setpoints are issued to control the output of conventional units or the active power control of new energy power plants. On the energy storage side: charging / discharging power and timing commands are issued. On the load side: adjustment signals or price signals are sent to interruptible loads (such as air conditioners and charging piles) through the demand response platform.

[0096] In some optional implementations, the source-grid-load-storage coordinated control method also includes providing operation and maintenance personnel with functions such as panoramic system monitoring, visualization of prediction results, simulation of control strategies, and report generation.

[0097] The source-grid-load-storage coordinated control method provided in this embodiment constructs a second objective function based on the power deviation information of the power supply equipment in the source-grid-load-storage system and the adjustment cost of adjustable resources. It incorporates the real-time deviation of system operation and economic adjustment cost into the optimization objective, focusing on the power imbalance problem in real-time operation. Based on the second objective function, the power generation prediction sequence, the load prediction sequence, the real-time operation information of the source-grid-load-storage system, the scheduling plan information, and the adjustable resource information, the control strategy of the source-grid-load-storage system is optimized and solved to obtain the second objective control strategy for a second preset time period. Through a short-cycle rolling optimization mechanism, the second objective control strategy can quickly respond to sudden fluctuations in power generation and load, and promptly call upon adjustable resources for dynamic balancing, thereby improving the accuracy and flexibility of the source-grid-load-storage system coordinated control.

[0098] This embodiment provides a source-grid-load-storage coordinated control method, which can be used in computer equipment. Figure 5 This is a fourth flowchart of the source-grid-load-storage coordinated control method according to an embodiment of the present invention, as follows: Figure 5 As shown, the process includes the following steps: Real-time acquisition and preprocessing of multi-source data; execution of high-precision power and load forecasting; collaborative optimization decision-making based on forecast results, specifically including day-ahead optimization layer and real-time rolling optimization layer.

[0099] In some alternative implementations, edge computing nodes are deployed in the park's substations to handle real-time data preprocessing and the issuance of ultra-short-term control commands. A cloud data center is configured to store and manage massive amounts of historical data, as well as to train and compute short-term prediction models and optimization algorithms.

[0100] The source-grid-load-storage coordinated control method of this invention, employing advanced deep learning models such as GANs and attention mechanisms, can more accurately characterize the complex nonlinear laws of renewable energy output and load changes. Simulation verification shows that the average absolute percentage error of ultra-short-term wind and solar power prediction can be reduced to below 8%, and the load prediction error can be reduced to below 3%, providing a reliable data foundation for subsequent optimized control. Through a two-layer optimization model, cross-timescale optimized resource allocation is achieved. The day-ahead planning ensures global economic optimality, while real-time rolling optimization effectively smooths fluctuations and reduces the use of expensive adjustment resources. Simultaneously, demand response incentives stimulate the adjustment potential on the load side at a lower cost, replacing some expensive energy storage investments. Comprehensive calculations indicate that the overall energy utilization efficiency of the industrial park can be improved by approximately 15%, and the average electricity cost reduced by approximately 10%. This invention treats source-grid-load-storage as a whole for coordinated control, enabling rapid response to system power imbalances. When potential power deficits or voltage exceedance risks are predicted, the system can proactively activate energy storage discharge or reduce flexible loads, effectively preventing line overloads and voltage exceedances, reducing the risk of grid accidents. High-precision power forecasting allows the system to more aggressively schedule renewable energy generation, while the flexible interactive control strategy provides a wealth of adjustment methods for absorbing fluctuating renewable energy. This invention can effectively control the wind and solar curtailment rates in the industrial park from over 10% to below 3%, greatly promoting local absorption of renewable energy. Based on a cloud-edge collaborative architecture, the algorithm model can continuously learn and evolve in the cloud, adapting to changes in the park's energy structure.

[0101] The source-grid-load-storage coordinated control method of this invention applies generative adversarial networks (GANs) to power forecasting. The generator in the GAN simulates the complex probability distribution of new energy output sequences, while the discriminator continuously forces the generator to output more realistic predictions by comparing them with real historical data. This method is suitable for processing highly random wind and solar power data. A spatiotemporal attention mechanism is introduced, enabling the power forecasting model to automatically learn and focus on key time points and spatial regions (such as the correlation between adjacent wind farm clusters) that have the greatest impact on future forecast times from historical data, thereby significantly improving forecast accuracy, especially during critical periods such as the passage of weather systems. The load forecasting model based on the variable attention mechanism can dynamically evaluate and quantify the weights of different input variables (such as temperature, humidity, wind speed, solar radiation intensity, date type, and holiday markers) on the load. It features a dual-layer optimization control architecture with a day-ahead optimization layer and a real-time rolling optimization layer. The day-ahead optimization layer is based on high-precision short-term forecasts and aims to minimize the daily operating cost and the curtailment rate of wind and solar power to formulate unit combination and energy storage charging and discharging plans. The real-time rolling optimization layer is based on ultra-short-term forecasts and aims to minimize the power balance deviation and the adjustment cost to perform minute-level fine-tuning of controllable resources.

[0102] This embodiment also provides a source-grid-load-storage coordinated control device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0103] This embodiment provides a source-grid-load-storage coordinated control device, such as... Figure 6 As shown, it includes: The power prediction module 601 is used to predict the power generation of the power equipment based on the historical power data of the power equipment in the power source-grid-load-storage system, the weather forecast data of the area where the power equipment is located, and the real-time meteorological data, so as to obtain the power generation prediction sequence for the prediction period.

[0104] The load forecasting module 602 is used to forecast the load of the load equipment based on the historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type of the load equipment in the source-grid-load-storage system, and to obtain the load forecast sequence for the forecast period.

[0105] The condition construction module 603 is used to construct a first objective function based on various cost information of the source-grid-load-storage system, and to construct multiple constraints based on the operation information of the source-grid-load-storage system.

[0106] The strategy generation module 604 is used to optimize the control strategy of the source-grid-load-storage system based on the first objective function, multiple constraints, power generation prediction sequence and load prediction sequence, to obtain the first objective control strategy for the first preset time period, so as to carry out source-grid-load-storage coordinated control according to the first objective control strategy.

[0107] In some alternative implementations, the power prediction module 601 includes: The power prediction unit is used to input historical power data of the power equipment in the power generation, grid, load and storage system, weather forecast data of the area where the power equipment is located and real-time meteorological data into the trained power prediction model to obtain the power generation prediction sequence for the prediction period. The power prediction model is a neural network model for power generation prediction based on generative adversarial network and attention mechanism.

[0108] In some alternative implementations, the load forecasting module 602 includes: The load forecasting unit is used to input historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type into the trained load forecasting model to obtain the load forecasting sequence for the forecast period. The load forecasting model is a neural network model for load forecasting built based on the variable attention mechanism.

[0109] In some alternative implementations, the condition construction module 603 includes: The objective function construction unit is used to obtain a summation result based on the sum of the electricity purchase cost variable, the generator fuel cost variable, the energy storage depreciation cost variable, and the wind and solar curtailment penalty cost variable; to obtain a difference result based on the difference between the summation result and the green energy environmental benefit variable; and to minimize the difference result to obtain the first objective function.

[0110] The power balance constraint construction unit is used to construct power balance constraints based on the sum of the conventional unit output of the power source-grid-load-storage system, the predicted output of new energy sources, the curtailed power of new energy sources, the charging and discharging power of energy storage, and the exchange power, which equals the predicted load value.

[0111] The output constraint construction unit is used to construct output constraints based on the fact that the output of conventional units in the power source-grid-load-storage system is within the technical limit range of the equipment.

[0112] The ramp rate constraint construction unit is used to construct ramp rate constraints based on the ramp rate of the source-grid-load-storage system being within a preset ramp rate range.

[0113] The switching power construction unit is used to construct switching power constraints based on the fact that the switching power of the source-grid-load-storage system is within a preset switching power range.

[0114] In some optional implementations, the source-grid-load-storage coordinated control device further includes: The objective function construction module is used to construct a second objective function based on the power deviation information of the power supply equipment in the source-grid-load-storage system and the adjustment cost of adjustable resources.

[0115] The real-time strategy generation module is used to optimize the control strategy of the source-grid-load-storage system based on the second objective function, the power generation prediction sequence, the load prediction sequence, the real-time operation information of the source-grid-load-storage system, the scheduling plan information, and the adjustable resource information, so as to obtain the second objective control strategy for the second preset time period, and to carry out source-grid-load-storage coordinated control according to the second objective control strategy; the time interval of the second preset time period is shorter than the time interval of the first preset time period.

[0116] The source-grid-load-storage coordinated control device provided in this embodiment of the invention can execute the source-grid-load-storage coordinated control method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the various modules and units described above are the same as in the corresponding embodiments described above, and will not be repeated here.

[0117] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0118] The following is a detailed reference. Figure 7 This diagram illustrates a suitable structural schematic for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 701, which can perform various appropriate actions and processes based on a program stored in read-only memory (ROM) 702 or a program loaded from memory 708 into random access memory (RAM) 703. The RAM 703 also stores various programs and data required for the operation of the electronic device. The processor 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0119] Typically, the following devices can be connected to I / O interface 705: input devices 706 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 707 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 708 including, for example, magnetic tapes, hard disks, etc.; and communication devices 709. Communication device 709 allows electronic devices to exchange data via wireless or wired communication with other devices. Although Figure 7 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0120] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 709, or installed from a memory 708, or installed from a ROM 702. When the computer program is executed by the processor 701, it performs the functions defined in the source-grid-load-storage coordinated control method of the embodiments of the present invention.

[0121] Figure 7The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0122] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the source-grid-load-storage coordinated control method shown in the above embodiments is implemented.

[0123] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0124] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A source-grid-load-storage coordinated control method, characterized in that, The method includes: Based on the historical power data of the power equipment in the power source-grid-load-storage system, the weather forecast data of the area where the power equipment is located, and the real-time meteorological data, the power generation of the power equipment is predicted to obtain a power generation prediction sequence for the prediction period. Based on the historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type of the load equipment in the source-grid-load-storage system, the load of the load equipment is predicted to obtain the load prediction sequence for the prediction period. A first objective function is constructed based on various cost information of the source-grid-load-storage system, and multiple constraints are constructed based on the operation information of the source-grid-load-storage system. Based on the first objective function, multiple constraints, the power generation prediction sequence, and the load prediction sequence, the control strategy of the source-grid-load-storage system is optimized and solved to obtain a first target control strategy for a first preset time period, so as to carry out source-grid-load-storage coordinated control according to the first target control strategy.

2. The method according to claim 1, characterized in that, The method further includes: A second objective function is constructed based on the power deviation information of the power supply equipment in the source-grid-load-storage system and the adjustment cost of adjustable resources; Based on the second objective function, the power generation prediction sequence, the load prediction sequence, the real-time operation information of the power generation, grid, load and storage system, the scheduling plan information, and the adjustable resource information, the control strategy of the power generation, grid, load and storage system is optimized and solved to obtain a second objective control strategy for a second preset time period, so as to carry out coordinated control of power generation, grid, load and storage according to the second objective control strategy; the time interval of the second preset time period is shorter than the time interval of the first preset time period.

3. The method according to claim 1 or 2, characterized in that, The process involves predicting the power generation capacity of the power equipment based on historical power data of the power generation system, weather forecast data for the area where the power equipment is located, and real-time meteorological data, to obtain a power generation prediction sequence for the prediction period, including: The historical power data of the power supply equipment in the source-grid-load-storage system, the weather forecast data of the area where the power supply equipment is located, and the real-time meteorological data are input into the trained power prediction model to obtain the power generation prediction sequence for the prediction period; the power prediction model is a neural network model for power generation prediction built based on generative adversarial networks and attention mechanisms.

4. The method according to claim 1 or 2, characterized in that, The method of predicting the load of the load equipment based on historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type, to obtain a load prediction sequence for the prediction period, includes: The historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type of the load equipment in the source-grid-load-storage system are input into the trained load prediction model to obtain the load prediction sequence for the prediction period; the load prediction model is a neural network model for load prediction built based on the variable attention mechanism.

5. The method according to claim 1 or 2, characterized in that, The various cost information includes variables such as electricity purchase cost, generator fuel cost, energy storage depreciation cost, wind and solar curtailment penalty cost, and green energy environmental benefit variable; the various constraints include power balance constraints, output constraints, ramp rate constraints, and exchange power constraints; the first objective function is constructed based on the various cost information of the source-grid-load-storage system, and the various constraints are constructed based on the operating information of the source-grid-load-storage system, including: The summation result is obtained by summing the electricity purchase cost variable, the generator fuel cost variable, the energy storage depreciation cost variable, and the wind and solar curtailment penalty cost variable; the difference result is obtained by the difference between the summation result and the green energy environmental benefit variable; the difference result is minimized to obtain the first objective function. The power balance constraint is constructed based on the sum of the conventional unit output, the predicted output of new energy sources, the abandoned power of new energy sources, the charging and discharging power of energy storage, and the exchange power of the source-grid-load-storage system, which equals the predicted load value. Based on the fact that the conventional unit output of the source-grid-load-storage system is within the equipment's technical limit range, the output constraint is constructed. Based on the fact that the ramp rate of the source-grid-load-storage system is within a preset ramp rate range, the ramp rate constraint is constructed. The switching power constraint is constructed based on the fact that the switching power of the source-grid-load-storage system is within a preset switching power range.

6. The method according to claim 1 or 2, characterized in that, The step of optimizing the control strategy of the power generation, grid, load, and storage system based on the first objective function, multiple constraints, the power generation prediction sequence, and the load prediction sequence to obtain a first objective control strategy for a first preset time period includes: Using the first objective function as the optimization objective, multiple constraints as constraints, and conventional unit output, energy storage charging and discharging power, and power exchanged with the main grid as decision variables, the power generation prediction sequence and the load prediction sequence are input into a preset optimization solver for optimization and solution to obtain the first objective control strategy for the first preset time period.

7. A source-grid-load-storage coordinated control device, characterized in that, The device includes: The power prediction module is used to predict the power generation of the power equipment based on the historical power data of the power equipment in the source-grid-load-storage system, the weather forecast data of the area where the power equipment is located, and the real-time meteorological data, so as to obtain the power generation prediction sequence for the prediction period. The load forecasting module is used to forecast the load of the load equipment based on the historical load data, real-time load data, meteorological data of the area where the load equipment is located, and date type of the load equipment in the source-grid-load-storage system, so as to obtain the load forecasting sequence for the forecasting period. The condition construction module is used to construct a first objective function based on various cost information of the source-grid-load-storage system, and to construct multiple constraint conditions based on the operation information of the source-grid-load-storage system. The strategy generation module is used to optimize the control strategy of the source-grid-load-storage system based on the first objective function, multiple constraints, the power generation prediction sequence, and the load prediction sequence, to obtain a first target control strategy for a first preset time period, so as to perform source-grid-load-storage coordinated control according to the first target control strategy.

8. An electronic device, characterized in that, include: A memory and a processor are interconnected, the memory stores computer instructions, and the processor executes the source-grid-load-storage coordinated control method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the source-grid-load-storage coordinated control method according to any one of claims 1 to 6.

10. A computer program product, characterized in that, It includes computer instructions for causing a computer to execute the source-grid-load-storage coordinated control method according to any one of claims 1 to 6.

Images