Self-adaptive regulation and control method for optical storage direct-flexible system of low-carbon transformer substation

Abstract

The invention relates to the technical field of self-adaptive regulation and control, in particular to a self-adaptive regulation and control method for an optical storage direct-flexible system of a low-carbon substation. The method comprises the following steps: S1, collecting real-time operation data from a photovoltaic panel, an energy storage device, a load end and a power grid interface; s2, establishing a photovoltaic power generation model based on the real-time operation data, and establishing a mathematical model of an energy storage system and a direct-current flexible power grid; and S3, based on the photovoltaic power generation model and the mathematical model, through a model prediction control algorithm, optimizing cooperative regulation and control of photovoltaic power generation, an energy storage system and a power grid. According to the design of the invention, by accurately predicting the photovoltaic power generation capacity and combining the mathematical models of the energy storage system and the direct current flexible power grid, the adaptive regulation and control method can optimize the cooperative work among photovoltaic power generation, energy storage charging and discharging and power grid dispatching. The utilization rate of renewable energy sources (such as solar energy) is increased, dependence on fossil fuel is reduced, and low carbon and environmental protection are achieved.

Description

Technical Field

[0001] The present invention relates to the field of adaptive control technology, and in particular to an adaptive control method for a photovoltaic storage direct-flexible system in a low-carbon substation. Background Art

[0002] As the global demand for clean energy increases, renewable energy such as solar photovoltaic (PV) has become an important part of electricity supply. However, these energy sources are intermittent and volatile. For example, PV power generation is affected by weather conditions, seasonal changes and time, which poses a great challenge to the stable operation of the power grid. Specifically, the traditional AC power grid design does not fully consider the need for large-scale access to unstable power sources, resulting in voltage instability and increased transmission loss when facing rapidly changing power generation output or load demand. In addition, when a large number of distributed energy resources (DERs) such as rooftop photovoltaic systems are connected to the grid, if there is no effective coordination mechanism, reverse flow may occur, affecting the power quality and the safety of the grid. Therefore, an adaptive control method for a low-carbon substation photovoltaic storage direct-flexible system is provided. Summary of the invention

[0003] The purpose of the present invention is to provide an adaptive control method for a low-carbon substation photovoltaic storage direct-flexible system to solve the problem that the traditional AC power grid design proposed in the above background technology does not fully consider the needs of large-scale access to unstable power sources and has no effective coordination mechanism.

[0004] To achieve the above object, the present invention aims to provide an adaptive control method for a photovoltaic storage direct-flexible system of a low-carbon substation, comprising the following steps: S1, collect real-time operation data from photovoltaic panels, energy storage devices, load terminals and grid interfaces; S2. Establish a photovoltaic power generation model based on real-time operation data, and establish a mathematical model of the energy storage system and DC flexible power grid; Among them, establishing a photovoltaic power generation model based on real-time operation data includes the following steps: S2.1. Clean and preprocess the collected real-time operation data; S2.2, perform regression analysis on the relationship between photovoltaic power generation and irradiance; S2.3. Use the real-time running data as a data set and divide the data set into a training set and a validation set. Use the training set to fit the model and use the validation set to evaluate the accuracy of the model. S2.4. Based on the fitting results, a photovoltaic power generation model is established, and the photovoltaic power generation model is optimized according to the angle of the photovoltaic panel; Furthermore, the photovoltaic power generation model is: ; in, Indicates the actual photovoltaic power generation; Represents the photovoltaic power generation under the reference irradiance; Indicates the actual irradiance; represents the reference irradiance; Indicates the current temperature of the photovoltaic panel; represents the temperature coefficient; Indicates the reference temperature; Optimize the photovoltaic power generation model according to the angle of the photovoltaic panel: ; ; ; in, It indicates the actual photovoltaic power generation after the angle factor of the photovoltaic panel is introduced; It represents the effective irradiance after taking angle correction into account; represents direct irradiance; represents the diffuse irradiance; represents the angle of incidence; Indicates the tilt angle of the photovoltaic panel; represents the solar altitude angle; Indicates the solar azimuth; Indicates the azimuth angle of the photovoltaic panel; S2.5. Input the new real-time data into the established photovoltaic power generation model to perform real-time power generation prediction; S3. Based on the photovoltaic power generation model and mathematical model, the model predictive control algorithm is used to optimize the coordinated regulation of photovoltaic power generation, energy storage system and power grid.

[0005] As a further improvement of the technical solution, in S1, the real-time operating data includes: photovoltaic power generation power of photovoltaic panels, light irradiance, charging and discharging power of energy storage devices, voltage, load and power of the power grid.

[0006] As a further improvement of the technical solution, in S2, establishing a mathematical model of the energy storage system includes the following steps: S2.6. Define the basic variables and parameters of the energy storage system; S2.7. Establish an energy dynamic model of the energy storage system; S2.8. Set constraints for energy storage charging and discharging.

[0007] As a further improvement of the technical solution, in S2.7, the energy dynamic model of the energy storage system is: ; in, Indicates time Energy storage system energy; Indicates time; Indicates time Energy storage system energy; Represents the charging efficiency coefficient of the energy storage system; Indicates time Charging power of the energy storage system; Indicates time Discharge power of the energy storage system; Represents the discharge efficiency coefficient of the energy storage system.

[0008] As a further improvement of the technical solution, in S2, establishing a mathematical model of the DC flexible power grid includes the following steps: S2.9. Define the basic variables of the DC grid; S2.10. According to Kirchhoff's law, establish the current balance equation for each node of the power grid: ; in, Indicates the inflow node The current, represents the elements of the connectivity matrix, Indicates passing through the branch The current, Represents the node index; S2.11. Describe the power flow model based on the voltage difference and the resistance between nodes, and establish the power balance model of the power grid through the relationship between the total power supply and demand; S2.12. Set the constraints of the power grid.

[0009] As a further improvement of the technical solution, in S2.11, the power flow model and the power balance model are: The power flow model is: ; The power balance model is: ; in, Representation Node and nodes The resistance of the line between them; Represents a slave node and nodes The power flow; Representation Node Voltage; Representation Node Voltage; Representation Node The net power injection of Representation Node The power generation at Representation Node The load consumption at the .

[0010] As a further improvement of the technical solution, in S3, the coordinated regulation of photovoltaic power generation, energy storage system and power grid is optimized based on the photovoltaic power generation model and the mathematical model through the model predictive control algorithm, including the following steps: S3.1. Define the optimization objective function and constraints; S3.2, predicting the system status within a future period of time t through the mathematical model of photovoltaic power generation model, energy storage system and DC flexible power grid; S3.3, use real-time data as input for prediction; S3.4, set the prediction time window g and calculate the power flow and energy storage state at each moment in the future; S3.5. At each control time t1, solve the optimization problem based on the current system state and the prediction results; S3.6, use optimization algorithms to solve optimization problems; S3.7, calculate the error between the actual operating value and the predicted value, and update the model parameters; S3.8. Re-optimize the control strategy based on the new forecast and adjust the energy storage charging and discharging power and grid dispatching.

[0011] As a further improvement of the technical solution, in S3.1, the optimization objective function is: ; in, Represents the sum of the optimization objectives during the forecast period; Indicates the length of the forecast period; The weight coefficient representing the difference between grid power, PV generation and energy storage discharge power; Represents the weight coefficient of energy storage charging power; Represents the weight coefficient of energy storage discharge power; Indicates time The load requires power.

[0012] Compared with the prior art, the present invention has the following beneficial effects: 1. In the adaptive control method of the photovoltaic storage direct-current flexible system of the low-carbon substation, by accurately predicting the photovoltaic power generation and combining the mathematical model of the energy storage system and the DC flexible power grid, the adaptive control method can optimize the coordinated work between photovoltaic power generation, energy storage charging and discharging, and power grid dispatching. This not only improves the utilization rate of renewable energy (such as solar energy), reduces dependence on fossil fuels, and achieves low-carbon and environmental protection, but also enhances the flexibility and response speed of the power system, helps to maintain the stable operation of the power grid, and ensures the reliability of power supply even when the light conditions change or the load fluctuates.

[0013] 2. In the adaptive control method of the photovoltaic storage direct-flexible system of the low-carbon substation, the model predictive control algorithm based on real-time data allows the system to dynamically adjust the energy storage strategy to avoid unnecessary battery charge and discharge cycles, thereby reducing the aging rate of the energy storage equipment and extending its service life. At the same time, by optimizing the matching degree between power generation and power demand, unnecessary energy loss is reduced and the cost expenditure in the power transmission and distribution process is minimized. This method effectively realizes the effective management of energy and the optimal allocation of resources, providing a guarantee for the long-term economic operation of the substation. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The figure is a flow chart of the overall method of the present invention. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] Example: See Figure 1 As shown, this embodiment provides an adaptive control method for a low-carbon substation photovoltaic storage direct-flexible system, including the following steps: S1. Collect real-time operating data from photovoltaic panels, energy storage devices, load terminals, and grid interfaces. Photovoltaic panels can convert solar energy into electrical energy and are an important part of renewable energy. Energy storage devices can store excess electrical energy and release it when needed, which helps balance supply and demand, improve grid stability, and increase the utilization rate of renewable energy. In this embodiment, the real-time operation data includes: photovoltaic power generation of photovoltaic panels, light irradiance, charging and discharging power of energy storage devices, voltage, load and power of the power grid.

[0017] S2. Establish a photovoltaic power generation model (for predicting power generation) based on real-time operation data, and establish mathematical models of energy storage systems and DC flexible power grids to simulate the charging and discharging efficiency of the energy storage system and the power flow in the power grid, respectively. The mathematical model supports DC power transmission and flexible power grid operation. The DC flexible power grid allows more flexible control of power flow, optimizes energy distribution, and is more suitable for integration with renewable energy and energy storage systems, thereby improving the flexibility and efficiency of the entire power system. In this embodiment, a photovoltaic power generation model is established based on real-time operation data (the main purpose of establishing a photovoltaic power generation model based on real-time operation data is to accurately predict and optimize the output performance of the photovoltaic power generation system), including the following steps: S2.1. Clean and preprocess the collected real-time operation data. The preprocessing includes removing outliers, filling missing values, and standardizing or normalizing the data; S2.2. Perform regression analysis on the relationship between photovoltaic power generation and irradiance. The principle of regression analysis on the relationship between photovoltaic power generation and irradiance is based on statistical methods. By establishing a mathematical model to describe the dependency between the two, it is used to understand the relationship between variables and provide a basis for the subsequent selection of appropriate modeling methods. S2.3. Use the real-time running data as a data set and divide the data set into a training set and a validation set. Use the training set to fit the model and use the validation set to evaluate the accuracy of the model to ensure the generalization ability of the model. S2.4. Based on the fitting results, a photovoltaic power generation model is established, and the photovoltaic power generation model is optimized according to the angle of the photovoltaic panel; Furthermore, the photovoltaic power generation model is: ; in, Indicates the actual photovoltaic power generation; Indicates the photovoltaic power generation under reference irradiance (measured under standard test conditions); Indicates the actual irradiance; represents the reference irradiance; Indicates the current temperature of the photovoltaic panel; It represents the temperature coefficient, which indicates the effect of temperature on the output power of the photovoltaic panel; Indicates the reference temperature, usually 25°C; The angle of the photovoltaic panel (including the tilt angle and azimuth) directly affects the amount of solar radiation received. By optimizing the angle, it is possible to ensure that the photovoltaic panel can receive as much solar radiation as possible at different times of the day and different seasons of the year, thereby maximizing the energy collection efficiency. For example, in high-latitude areas, when the sun is low in winter, appropriately increasing the tilt angle can help the photovoltaic panel get more light; accurate prediction of photovoltaic power generation is crucial to optimizing the entire photovoltaic storage direct and flexible system. The angle change of the photovoltaic panel will cause the irradiance it receives to be significantly different, which in turn affects the power generation. If these factors are not considered, the prediction model may have large errors. Therefore, incorporating the angle factor into the model can improve the prediction accuracy, making the energy storage system, load scheduling and grid interaction more intelligent and efficient; low-carbon substations are usually equipped with energy storage devices to balance the supply and demand of electricity. By more accurately predicting photovoltaic power generation, energy storage charging and discharging strategies can be better planned, reducing unnecessary battery cycles, extending battery life, and reducing operating costs. In addition, this also helps to more accurately match power generation and power demand, and improve the overall energy efficiency of the system; Optimize the photovoltaic power generation model according to the angle of the photovoltaic panel: ; ; ; in, It indicates the actual photovoltaic power generation after the angle factor of the photovoltaic panel is introduced; It represents the effective irradiance after taking angle correction into account; represents direct irradiance; represents the diffuse irradiance; represents the angle of incidence, i.e. the angle between the sun's rays and the normal to the photovoltaic panel surface (degrees); Indicates the tilt angle of the photovoltaic panel, relative to the horizontal plane (degrees); represents the solar altitude angle (degrees), that is, the angle of the center of the sun above the horizon; Indicates the solar azimuth (degrees), which refers to the angle of the sun relative to due south, with positive values ​​toward the east and negative values ​​toward the west; Indicates the azimuth angle of the photovoltaic panel (degrees), which refers to the angle of the photovoltaic panel relative to the south direction, with positive values ​​to the east and negative values ​​to the west; S2.5. Input new real-time data (such as real-time irradiance and temperature) into the established photovoltaic power generation model to perform real-time power generation prediction; Among them, establishing a mathematical model of the energy storage system includes the following steps: S2.6. Define the basic variables and parameters of the energy storage system, which include energy state variables, charge and discharge power, charge and discharge efficiency, energy storage device capacity and maximum charge and discharge power; S2.7. Establish an energy dynamic model of the energy storage system. The main purpose of establishing an energy dynamic model of the energy storage system is to accurately describe and predict the behavior of the energy storage system under different operating conditions, so as to optimize its charging and discharging strategy and ensure the stability and efficient operation of the power system; Furthermore, the energy dynamic model of the energy storage system is: ; in, Indicates time Energy storage system energy; Indicates time; Indicates time Energy storage system energy; It represents the charging efficiency coefficient of the energy storage system, which indicates the energy loss during the charging process of the energy storage system; Indicates time Charging power of the energy storage system; Indicates time Discharge power of the energy storage system; Represents the discharge efficiency coefficient of the energy storage system; S2.8. Set constraints for energy storage charging and discharging (the energy state of the energy storage battery must be maintained between the maximum and minimum capacity; the charging and discharging power of the energy storage are both limited by the maximum power; at any time, the energy storage device can only choose to charge or discharge, not both at the same time); Furthermore, the main purpose of establishing a mathematical model of a DC flexible grid is to accurately describe and optimize the power flow, voltage level, and power balance within the DC grid, thereby ensuring the safe, stable, and efficient operation of the grid; Establishing a mathematical model of a DC flexible power grid includes the following steps: S2.9. Define the basic variables of the DC grid, which include node voltage, power flow, node current, grid connection matrix and grid impedance; S2.10. According to Kirchhoff's law, a current balance equation is established for each node of the power grid to ensure that the total current flowing into a node is equal to the total current flowing out of the node: ; in, Indicates the inflow node The current, represents the elements of the connectivity matrix, Indicates passing through the branch The current, Represents the node index; S2.11. Describe the power flow model based on the voltage difference and the resistance (impedance) between nodes (in a DC grid, power flow is mainly driven by the voltage difference between nodes, while the resistance of the line (for a DC system, impedance is almost equivalent to resistance) determines the degree of power flow. By modeling based on voltage difference and resistance, the amount of power flow in each branch can be accurately calculated, which is crucial for understanding the energy distribution within the grid). Establish a power balance model for the grid through the relationship between total power supply and demand (to ensure energy conservation of the entire system, it is necessary to establish a power balance equation for each node to ensure that the sum of all power flowing into a node is equal to the sum of power flowing out of the node plus the net consumption of the node); Furthermore, the power flow model and the power balance model are: The power flow model is: ; The power balance model is: ; in, Representation Node and nodes The resistance of the line between them; Represents a slave node and nodes The power flow; Representation Node Voltage; Representation Node Voltage; Representation Node of net power injection (generation minus load); Representation Node The power generated at the location (e.g. contribution from photovoltaic panels or energy storage systems); Representation Node Load consumption at S2.12. Set the constraints of the power grid (the voltage of each node has certain upper and lower limits, usually to ensure the safe operation of the power grid and avoid damage to equipment due to excessively high or low voltage; the power flow must meet the capacity limitations of the power grid; the flow of power in the power grid is usually unidirectional, that is, in some cases, power can only flow from the power generation node to the load node or energy storage node, so it is necessary to ensure that the direction of power flow is correct).

[0018] S3. Based on the photovoltaic power generation model and mathematical model, the model predictive control (MPC) algorithm is used to optimize the coordinated regulation of photovoltaic power generation, energy storage system and power grid, which directly promotes the effective use of resources and reduces unnecessary energy loss and dependence on fossil fuels; In this embodiment, the prediction model is used to estimate the future state, and the control strategy is adjusted in real time through the optimization algorithm to achieve the best energy distribution and the optimization of system performance; by combining real-time operation data and historical trends, MPC can more accurately predict future power generation and load demand, so as to better plan power dispatch and resource allocation; MPC can quickly respond to changing environmental conditions and load demand, dynamically adjust the charging and discharging strategy of the energy storage system and the operation mode of the power grid, and ensure that the system is always in the optimal state; accurate prediction and intelligent regulation make more full use of renewable energy such as photovoltaics, reduce dependence on traditional fossil energy, and promote green transformation; Based on the photovoltaic power generation model and mathematical model, the model predictive control algorithm is used to optimize the coordinated regulation of photovoltaic power generation, energy storage system and power grid, including the following steps: S3.1. Define the optimization objective function and constraints. The constraints include: voltage constraint (ensuring that the system voltage remains within the specified range), energy storage capacity constraint (charging and discharging of the energy storage system must be within its capacity range), power flow constraint (grid power flow needs to meet physical constraints, such as the maximum transmission capacity of the line), and load balance constraint (the power provided by the grid must be equal to the load demand minus photovoltaic power generation and energy storage discharge); Furthermore, the optimization objective function is: ; in, represents the sum of the optimization objectives (minimized error) during the forecast period; Indicates the length of the forecast period; The weight coefficient representing the difference between grid power, PV generation and energy storage discharge power; Represents the weight coefficient of energy storage charging power; Represents the weight coefficient of energy storage discharge power; Indicates time The load demand power; S3.2. Predict the system status (such as photovoltaic power generation, energy storage status, grid power flow, etc.) within a future period of time t through the mathematical model of photovoltaic power generation model, energy storage system and DC flexible power grid; S3.3. Use real-time data (such as light intensity, temperature, etc.) as input for prediction; S3.4, set the prediction time window g and calculate the power flow and energy storage state at each moment in the future; S3.5. At each control time t1, according to the current system state and the prediction results, solve the optimization problem. The optimization goal is to minimize the total cost (such as grid power, energy storage loss, etc.) and satisfy the physical constraints of the system. and When it increases, the charging and discharging power amplitude decreases (experimental data: From 0.1 to 0.5, the number of daily charge and discharge times increases from 2.1 to 1.2 times). Dominant PV absorption rate (when α changes from 1 to 5, the absorption rate changes from 85% to 94%); S3.6. Use an optimization algorithm (gradient descent method) to solve the optimization problem. The gradient descent method iteratively adjusts parameters to minimize the objective function and gradually approaches the optimal solution using the negative gradient direction. The main purpose of using the gradient descent method to solve the optimization problem is to find the minimum (or maximum, depending on the setting of the problem) value of the objective function, that is, to find a set of parameters that minimizes the error or loss of the model on a given data set; S3.7, calculate the error between the actual operating value and the predicted value, update the model parameters, and ensure that the system can adapt to changes in the environment; S3.8. Re-optimize the control strategy based on the new prediction, adjust the energy storage charging and discharging power and grid dispatch, and adjust the energy storage charging and discharging strategy based on the optimization result of the objective function: if the newly predicted photovoltaic power generation is higher than the load demand, or the grid voltage is too high, the energy storage system absorbs excess electricity by charging to avoid grid overload. Conversely, if the grid load is large, the energy storage needs to support the grid demand by discharging. When the newly predicted photovoltaic power generation is insufficient or the grid load is large, the energy storage system will discharge to make up for the shortage of photovoltaic power generation or meet the additional power demand of the grid. This process needs to be carried out within the battery charging capacity range through the energy dynamic model of the energy storage system. The grid power dispatch is to optimize the grid operation by adjusting the power flow and energy storage status in the grid, maintain the balance between the grid power supply and load demand during the optimization process, and ensure that the grid voltage fluctuates within a reasonable range to prevent excessive or low voltage.

[0019] Furthermore, through simulation comparison experiments and key indicators, we can quantitatively explain how to avoid unnecessary battery charge and discharge cycles, reduce the aging rate of energy storage equipment, and achieve effective energy management and optimal resource allocation: Compare the implementation designs: Experimental scenario: Location: a low-carbon substation (installed capacity: 1MW photovoltaic, 500kWh / 1MWh energy storage, 800kW DC load peak); Time range: 7 consecutive days of operation data (including sunny, cloudy, rainy and other weather conditions); Comparison group: control based on fixed SOC threshold (in this embodiment: stop charging when SOC>80%, stop discharging when SOC<20%), dynamic optimization based on model prediction (MPC control) (prediction window K=4 hours, rolling optimization cycle 15 minutes); Experimental indicators: Energy storage life related: Daily charge and discharge cycles: the number of complete charge and discharge cycles (0%→100%→0%) in a single day; Average cycle depth (DoD): the proportion of single discharge capacity to total capacity (shallow charge and discharge can extend life); Equivalent full cycle number (EFC): the equivalent number of cycles converted to 100% according to DoD (the lower the EFC, the longer the life); Energy management efficiency: PV consumption rate: the proportion of PV power generation that is locally consumed (load + energy storage); Energy storage life extension effect: The number of daily charge and discharge cycles under traditional rule control is 3.2 times / day, and under MPC control is 1.5 times / day, which is a 53% reduction. The verification steps are: Optimize the objective function weight in step S3.5 and Suppress frequent charging and discharging; The average cycle depth under traditional rule control is: 65%, and under MPC control is: 32%, a decrease of 51%. The verification steps are: the penalty term in the objective function in step S3.5 and Limit high-power charging and discharging; The equivalent full cycle number under traditional rule control is: 2.08 times / day, and under MPC control is: 0.48 times / day, which is a reduction of 77%. The verification steps are: the error feedback in step S3.8 adjusts the model parameters to optimize the charge and discharge smoothness; Through the optimization objective function design of step S3.5 (penalizing high-power charging and discharging) and the dynamic parameter update of step S3.8, the energy storage action frequency and cycle depth are significantly reduced: if the battery life is inversely proportional to EFC (the capacity of a lithium battery decays to 80% when EFC=4000 times), then the traditional control: life = 4000 / 2.08≈1923 days (5.3 years), MPC control: life = 4000 / 0.48≈8333 days (22.8 years), the life is extended by 4.3 times compared with the traditional control; In terms of improving energy management efficiency: Step S3.5 predicts photovoltaic output (step S2.5) and load demand, and schedules energy storage charging and discharging in advance. The traditional rule control efficiency is 78%, and the MPC control efficiency of this solution is 93%, and the photovoltaic absorption rate is increased by 15%. Based on the above data, this solution avoids unnecessary battery charging and discharging cycles, reduces the aging rate of energy storage equipment, and realizes effective energy management and optimal resource allocation.

[0020] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only preferred examples of the present invention, and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. An adaptive control method for a low-carbon substation photovoltaic storage direct-flexible system, characterized in that: The following steps are involved: S1, collect real-time operation data from photovoltaic panels, energy storage devices, load terminals and grid interfaces; S2. Establish a photovoltaic power generation model based on real-time operation data, and establish a mathematical model of the energy storage system and DC flexible power grid; Among them, establishing a photovoltaic power generation model based on real-time operation data includes the following steps: S2.

1. Clean and preprocess the collected real-time operation data; S2.2, perform regression analysis on the relationship between photovoltaic power generation and irradiance; S2.

3. Use the real-time running data as a data set and divide the data set into a training set and a validation set. Use the training set to fit the model and use the validation set to evaluate the accuracy of the model. S2.

4. Based on the fitting results, a photovoltaic power generation model is established, and the photovoltaic power generation model is optimized according to the angle of the photovoltaic panel; Among them, the photovoltaic power generation model is: ； in, Indicates the actual photovoltaic power generation; Represents the photovoltaic power generation under the reference irradiance; Indicates the actual irradiance; represents the reference irradiance; Indicates the current temperature of the photovoltaic panel; represents the temperature coefficient; Indicates the reference temperature; Optimize the photovoltaic power generation model according to the angle of the photovoltaic panel: ； ； ； in, It indicates the actual photovoltaic power generation after the angle factor of the photovoltaic panel is introduced; It represents the effective irradiance after taking angle correction into account; represents direct irradiance; represents the diffuse irradiance; represents the angle of incidence; Indicates the tilt angle of the photovoltaic panel; represents the solar altitude angle; Indicates the solar azimuth; Indicates the azimuth angle of the photovoltaic panel; S2.

5. Input the new real-time data into the established photovoltaic power generation model to perform real-time power generation prediction; S3. Based on the photovoltaic power generation model and mathematical model, the model predictive control algorithm is used to optimize the coordinated regulation of photovoltaic power generation, energy storage system and power grid.

2. The adaptive control method of the low-carbon substation photovoltaic storage direct-flexible system according to claim 1 is characterized by: In S1, the real-time operation data includes: photovoltaic power generation of photovoltaic panels, light irradiance, charging and discharging power of energy storage devices, voltage, load and power of the power grid.

3. The adaptive control method of the low-carbon substation photovoltaic storage direct-flexible system according to claim 2 is characterized by: In S2, a mathematical model of the energy storage system is established, including the following steps: S2.

6. Define the basic variables and parameters of the energy storage system; S2.

7. Establish an energy dynamic model of the energy storage system; S2.

8. Set constraints for energy storage charging and discharging.

4. The adaptive control method of the low-carbon substation photovoltaic storage direct-flexible system according to claim 3 is characterized by: In S2.7, the energy dynamic model of the energy storage system is: ； in, Indicates time Energy storage system energy; Indicates time; Indicates time Energy storage system energy; Represents the charging efficiency coefficient of the energy storage system; Indicates time Charging power of the energy storage system; Indicates time Discharge power of the energy storage system; Represents the discharge efficiency coefficient of the energy storage system.

5. The adaptive control method of the low-carbon substation photovoltaic storage direct-flexible system according to claim 4 is characterized by: In S2, a mathematical model of a DC flexible power grid is established, including the following steps: S2.

9. Define the basic variables of the DC grid; S2.

10. According to Kirchhoff's law, establish the current balance equation for each node of the power grid: ； in, Indicates the inflow node The current, represents the elements of the connectivity matrix, Indicates passing through the branch The current, Represents the node index; S2.

11. Describe the power flow model based on the voltage difference and the resistance between nodes, and establish the power balance model of the power grid through the relationship between the total power supply and demand; S2.

12. Set the constraints of the power grid.

6. The adaptive control method of the low-carbon substation photovoltaic storage direct-flexible system according to claim 5 is characterized by: In S2.11, the power flow model and the power balance model are: The power flow model is: ； The power balance model is: ； in, Representation Node and nodes The resistance of the line between them; Represents a slave node To Node The power flow; Representation Node Voltage; Representation Node Voltage; Representation Node The net power injection of Representation Node The power generation at Representation Node The load consumption at the .

7. The adaptive control method of the low-carbon substation photovoltaic storage direct-flexible system according to claim 6 is characterized by: In S3, the coordinated regulation of photovoltaic power generation, energy storage system and power grid is optimized based on the photovoltaic power generation model and the mathematical model through the model predictive control algorithm, including the following steps: S3.

1. Define the optimization objective function and constraints; S3.2, predicting the system status within a future period of time t through the mathematical models of photovoltaic power generation model, energy storage system and DC flexible power grid; S3.3, use real-time data as input for prediction; S3.4, set the prediction time window g and calculate the power flow and energy storage state at each moment in the future; S3.

5. At each control time t1, solve the optimization problem based on the current system state and the prediction results; S3.6, use optimization algorithms to solve optimization problems; S3.7, calculate the error between the actual operating value and the predicted value, and update the model parameters; S3.

8. Re-optimize the control strategy based on the new forecast and adjust the energy storage charging and discharging power and grid dispatching.

8. The adaptive control method of the low-carbon substation photovoltaic storage direct-flexible system according to claim 7 is characterized by: In S3.1, the optimization objective function is: ； in, Represents the sum of the optimization objectives during the forecast period; Indicates the length of the forecast period; The weight coefficient representing the difference between grid power, PV generation and energy storage discharge power; Represents the weight coefficient of energy storage charging power; Represents the weight coefficient of energy storage discharge power; Indicates time The load requires power.

Images