A power coordination control method for a mountainous landscape and a same-site power collection circuit

Abstract

The application discloses a kind of mountain scenery same field power collection line power coordination control method, belong to new energy power generation control and power system operation technical field, specific steps include: relying on station sensing equipment and digital elevation model obtains multi-source fusion basic data set, wind and light basic predicted power is output through LSTM model and introduced mountain terrain compensation factor correction, obtain topographic adaptation's ultra-short-term predicted power;Dynamic power distribution optimization model considering topographic line loss is constructed to solve optimal output distribution value, preventive regulation instruction is generated after forward-looking verification and smooth correction by rolling time domain optimization mechanism, finally based on execution feedback data iterative optimization control model parameter.The application can significantly improve mountain scene power prediction accuracy and power collection line utilization efficiency, enhance system operation stability and station power generation benefit.

Description

Technical Field

[0001] This invention relates to the field of new energy power generation control and power system operation technology, and in particular to a method for coordinated power control of mountain wind and solar power collection lines. Background Technology

[0002] Existing wind and solar co-collection power control technology generally adopts a fixed weight ratio to allocate the output power of wind power and photovoltaic power. It relies on conventional power prediction models to provide output reference and uses passive reduction after the power exceeds the limit as the main adjustment means to realize the collection and transmission of wind and solar power and basic grid-connected operation. It can meet the basic operation requirements in the application scenarios of flat plains with gentle terrain and small output fluctuations.

[0003] However, when this type of power control technology is applied to complex mountainous terrain, it suffers from several adaptability defects and operational shortcomings. Fixed-ratio power allocation is ill-suited to the strong fluctuations and spatiotemporal differences in solar power output in mountainous areas, easily leading to overload tripping of collector lines or inefficient operation under light load, resulting in low overall line utilization efficiency. Conventional power prediction models are not specifically calibrated to take into account the terrain features such as altitude, slope, and shading, resulting in significant output prediction errors and failing to provide accurate and reliable data support for power regulation. Power regulation often adopts a reactive, post-event approach, making it difficult to cope with power ramp-up impacts caused by sudden changes in local weather conditions in mountainous areas, easily causing voltage fluctuations at the grid connection point to exceed limits, threatening system operational stability. Furthermore, it fails to consider the additional transmission losses in collector lines caused by the undulating mountainous terrain, resulting in significant deficiencies in overall transmission efficiency and operational economy. Summary of the Invention

[0004] The purpose of this invention is to provide a method for coordinated power control of mountainous landscape solar power collection lines, thereby solving the aforementioned technical problems.

[0005] To achieve the above objectives, the present invention provides a method for coordinated power control of mountainous wind and solar power collection lines, comprising the following steps: S1. Based on the sensing equipment and digital elevation model deployed at the site, the basic dataset of multi-source fusion is obtained by synchronously collecting wind and solar turbine operation data, on-site environmental data and extracting terrain feature parameters at each location. S2. Based on the basic dataset, the basic prediction power of wind and solar power is output through the LSTM basic prediction model, and the mountain terrain compensation factor is introduced to correct the basic prediction power to obtain the terrain-adapted ultra-short-term prediction power of wind and solar power. S3. Based on the ultra-short-term predicted power of wind and solar power, and according to the real-time operating parameters of the power collection line, the optimal wind and solar power output allocation value is obtained by constructing and solving a dynamic power allocation optimization model that takes into account terrain line loss. S4. Based on the optimal wind and solar power output allocation value, a rolling time domain optimization mechanism is used to perform forward verification and smoothing correction of power in future multi-period periods to obtain a preventive power regulation command that meets the ramp rate constraint. S5. Based on the actual execution feedback data of the preventive power regulation command, the control model parameters are obtained through adaptive iterative optimization by incrementally training the prediction model and dynamically correcting the compensation factor coefficients.

[0006] Preferably, step S1 specifically includes: By deploying sensing terminals at wind turbines, photovoltaic arrays, and power collection lines, and using a 5m resolution digital elevation model of the site, operational data including real-time wind turbine output, real-time photovoltaic output, and power collection line current, voltage, and power are collected, along with environmental data including wind speed, wind direction, ambient temperature, solar irradiance, and component temperature. Simultaneously, terrain parameters such as altitude, slope angle, aspect angle, terrain curvature, and terrain shading height at each location are extracted to obtain a multi-source basic dataset containing operational data, environmental data, and terrain parameters.

[0007] Preferably, the mountain terrain compensation factor in step S2 includes the wind power terrain compensation factor, and the calculation formula for the wind power terrain compensation factor is as follows: ; In the formula, For wind power terrain compensation factors; This is the altitude correction factor, reflecting the change in air density with altitude; This is the slope correction factor, which reflects the impact of slope airflow tilt on wind energy capture. This is the terrain acceleration coefficient, reflecting the wind speed gain due to the valley funnel effect; The corrected wind power forecast satisfies: ; in, This is the corrected wind power forecast. This represents the predicted wind power base output from the LSTM model.

[0008] Preferably, the mountain terrain compensation factor in step S2 includes the photovoltaic terrain compensation factor, and the calculation formula for the photovoltaic terrain compensation factor is as follows: ; In the formula, This is a photovoltaic terrain compensation factor; This is the tilt angle correction factor, which reflects the impact of the deviation between the slope tilt angle and the optimal tilt angle on irradiation reception. This is the aspect correction factor, reflecting the irradiance loss when the slope aspect deviates from due south. This is the terrain shading correction factor, reflecting the shading effect of surrounding mountains on the photovoltaic array; The corrected photovoltaic power forecast satisfies: ; in, The corrected photovoltaic power forecast. This represents the base-predicted photovoltaic power output by the LSTM model.

[0009] Preferably, the dynamic power allocation optimization model considering terrain line loss in step S3 has the following objective function expression: ; In the formula, The target value for comprehensive transmission benefits; To improve the transmission efficiency of the power collection line, additional line losses caused by mountainous terrain should be considered. This refers to the allocated output power of the fan; The allocated photovoltaic output power; This is the penalty coefficient for hill-climbing fluctuations; This represents the total output change between adjacent control cycles; The constraints of the optimization model include line capacity constraints, upper and lower limits of output constraints, and voltage deviation constraints at the grid connection point.

[0010] Preferably, the transmission efficiency of the collector line is... The calculation formula is: ; in, The resistance per unit length of the conductor; This is the total length of the collector line; The terrain loss coefficient is calculated from the elevation difference and terrain undulation of the line. This is the line's rated voltage.

[0011] Preferably, in step S3, the dynamic power allocation optimization model is solved by an adaptive genetic algorithm. Specifically, based on the fitness value of individuals in the population, the crossover probability and mutation probability are dynamically adjusted. Individuals with fitness lower than the population average are given a higher crossover and mutation probability, while individuals with fitness higher than the population average are given a lower crossover and mutation probability, thereby obtaining the globally optimal wind and solar power output allocation solution.

[0012] Preferably, step S4 specifically includes: Based on a rolling time-domain optimization framework with a 30-minute prediction window and a 5-minute control step, the power ramp rate constraint is embedded into the allocation solution process by successively verifying the power change rate of the next 6 control steps. The power sequence that exceeds the limit is smoothed and corrected to obtain a preventive power regulation command that meets the ramp rate constraint. The formula for the slope rate constraint is: ; In the formula, For the first Total output per step length; To control the step size; This represents the maximum permissible gradeability.

[0013] Preferably, step S5 specifically includes: Based on real-time collected data on actual wind and solar power output and line operation, the prediction error and allocation deviation are calculated. The LSTM prediction model is incrementally trained every 24 hours using the latest operating data. At the same time, the coefficient parameters of the mountain terrain compensation factor are dynamically calibrated to obtain the iteratively optimized prediction model and compensation factor parameters.

[0014] Preferably, a power coordination control system for mountainous landscape power collection lines includes: a data acquisition layer, a data analysis layer, a control decision layer, and an execution feedback layer that are sequentially connected in communication. The data acquisition layer is distributed across key nodes of wind turbine locations, photovoltaic block arrays, and power collection lines. It includes wind speed sensors, solar irradiance meters, electrical power monitoring devices, and terrain data input modules. These are used to collect operational data, environmental data, and terrain parameters, and then upload them to the data analysis layer via fiber optic communication. The data analysis layer is deployed on the central control server of the site and includes a data fusion preprocessing module, an LSTM basic prediction module, and a mountain terrain compensation calculation module. It is used to complete data cleaning, basic power prediction and terrain correction calculation, and output the terrain-adapted wind and solar power prediction to the control decision layer. The control decision layer is deployed on the central control server of the station and includes a dynamic power allocation optimization module, a rolling time domain adjustment module, and an adaptive algorithm solution module, which are used to generate the optimal power allocation scheme and preventive adjustment instructions, and send them to the execution feedback layer. The execution feedback layer includes a wind turbine pitch control system and a photovoltaic inverter control system, which are deployed in the wind turbine nacelle and the photovoltaic array inverter room, respectively. They are used to receive control commands to adjust the wind and solar power output and to feed back the actual operating status to the data analysis layer and the control decision layer in real time, forming a closed-loop control link.

[0015] Therefore, the present invention employs the above-mentioned method for coordinated power control of mountainous landscape and solar power collection lines, which has the following beneficial effects: 1. By constructing a dual-dimensional terrain compensation factor for wind power and photovoltaic power, the wind power prediction results are corrected from the perspectives of altitude, slope, and terrain acceleration effect, while the photovoltaic prediction results are corrected from the perspectives of tilt angle, slope aspect, and terrain shading. The terrain adaptability calibration of the basic predicted power can offset the prediction bias caused by local topography and meteorological characteristics in mountainous areas, improve the accuracy of wind and solar power prediction in mountainous scenarios, and provide a reliable data foundation for subsequent power allocation and regulation.

[0016] 2. With the goal of maximizing the comprehensive transmission benefits of the power collection line, the additional losses due to mountainous terrain are incorporated into the line loss calculation model. At the same time, multiple constraints such as line capacity, output range, and grid connection point voltage are taken into account. An adaptive genetic algorithm is used to solve the optimal wind and solar power output allocation scheme, giving full play to the complementary characteristics of wind and solar power output. Under the premise of ensuring the safe operation of the line, the effective transmission power is maximized, the utilization efficiency of the power collection line is improved, the transmission loss is reduced, and the economic efficiency of the station operation is improved.

[0017] 3. By adopting a rolling time-domain optimization mechanism, setting multi-step prediction windows and control cycles, the power ramp-up rate constraint is embedded into a multi-time-period look-ahead verification process. This allows for the early identification of power surge risks and the dispersion of impacts through multi-cycle smoothing corrections. The power fluctuation rate is limited to a safe threshold, effectively avoiding the power ramp-up impact caused by sudden changes in mountain wind speed and irradiance, reducing line overload risks, smoothing voltage fluctuations at grid connection points, and significantly enhancing the stability and anti-disturbance capability of the system operation.

[0018] 4. By establishing a full-cycle closed-loop feedback mechanism, real-time data collection of actual wind and solar power output and line operation is used to calculate prediction errors and distribution deviations. The prediction model is incrementally trained regularly, and the coefficient parameters of the terrain compensation factor are dynamically calibrated. This enables the control model to continuously adapt to changes in weather, terrain, and equipment status at the station, ensuring the long-term stability of control accuracy, reducing the workload of manual parameter calibration, and improving the environmental adaptability and long-term operational reliability of the control system.

[0019] 5. By constructing a four-layer system architecture with sequential communication connections, the deployment location, connection relationship and functional positioning of each layer are clarified. Real-time data interaction across the entire link is achieved through the communication network, forming a complete closed-loop control link of perception-analysis-decision-execution. It is suitable for the characteristics of the distributed layout of mountain stations and has the advantages of flexible deployment, rapid response and reliable operation. It provides a stable and implementable system carrier for the whole set of power collaborative control methods, ensuring the efficient execution and feedback iteration of control strategies.

[0020] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0021] Figure 1 A flowchart of a method for coordinated power control of mountainous landscape co-current power collection lines provided by the present invention; Figure 2 The present invention provides a structural block diagram of a power coordinated control system for mountainous landscape co-current power collection lines. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages disclosed in the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the embodiments of the present invention and are not intended to limit the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of this application. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout.

[0023] It should be noted that the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, such as a process, method, system, product, or server that includes a series of steps or units, not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or device.

[0024] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0025] Based on the above analysis, this invention is designed. (See appendix.) Figures 1-2 ,include: Example 1: This embodiment is applied to a 35kV wind and solar power collection line unit in a loess gully mountainous area. The station adopts a four-layer collaborative control system architecture, namely the data acquisition layer, data analysis layer, control decision layer, and execution feedback layer. Each layer realizes bidirectional data interaction through an optical fiber communication network, forming a complete closed-loop control link of perception-analysis-decision-execution.

[0026] The data acquisition layer is distributed across key nodes of wind turbine locations, photovoltaic array blocks, and power collection lines. It includes wind speed sensors, wind direction sensors, temperature sensors, solar irradiance meters, electrical power monitoring devices, and terrain data input modules. It is used to collect wind and solar turbine operating data, on-site environmental data, and terrain parameters. The collected raw data is uploaded to the data analysis layer in real time via fiber optic communication.

[0027] The data analysis layer is deployed on the central control server of the site and includes a data fusion preprocessing module, an LSTM basic prediction module, and a mountain terrain compensation calculation module. The data fusion preprocessing module is responsible for cleaning, aligning and extracting features from the raw data. The LSTM basic prediction module is responsible for outputting the basic wind and solar power prediction. The mountain terrain compensation calculation module is responsible for calculating the terrain compensation factor and correcting the prediction power. Finally, the terrain-adapted ultra-short-term wind and solar power prediction is output to the control decision layer.

[0028] The control decision layer is deployed on the central control server of the site and includes a dynamic power allocation optimization module, a rolling time-domain adjustment module, and an adaptive algorithm solution module. The dynamic power allocation optimization module is used to build an allocation optimization model that takes into account terrain and line loss. The adaptive algorithm solution module is used to solve the optimal allocation solution. The rolling time-domain adjustment module is used to complete the look-ahead verification and smoothing correction of power in multiple time periods. Finally, it generates a preventive power adjustment command and sends it to the execution feedback layer.

[0029] The execution feedback layer includes the wind turbine pitch control system and the photovoltaic inverter control system, which are deployed in the wind turbine nacelle and the photovoltaic array inverter room, respectively. They are used to receive control commands and adjust the output of the wind and solar units, while feeding back the actual operating status to the data analysis layer and the control decision layer in real time to achieve closed-loop iterative optimization.

[0030] The specific control implementation steps in this embodiment are as follows: S1. By deploying sensing terminals at wind turbines, photovoltaic arrays, and power collection lines, and using a 5m resolution digital elevation model of the site, operational data including real-time wind turbine output, real-time photovoltaic output, and power collection line current, voltage, and power are collected, as well as environmental data including wind speed, wind direction, ambient temperature, solar irradiance, and component temperature. At the same time, terrain parameters such as altitude, slope angle, aspect angle, terrain curvature, and terrain shading height at each location are extracted to obtain a multi-source fusion basic dataset containing operational data, environmental data, and terrain parameters.

[0031] S2. Based on the basic dataset, the basic prediction power of wind and solar power is output through the LSTM basic prediction model, and the mountain terrain compensation factor is introduced to correct the basic prediction power to obtain the terrain-adapted ultra-short-term prediction power of wind and solar power.

[0032] The mountain terrain compensation factor includes the wind power terrain compensation factor and the photovoltaic terrain compensation factor: The formula for calculating the wind power terrain compensation factor is: ; In the formula, For wind power terrain compensation factors; This is the altitude correction factor, reflecting the change in air density with altitude; This is the slope correction factor, which reflects the impact of slope airflow tilt on wind energy capture. This is the terrain acceleration coefficient, reflecting the wind speed gain due to the valley funnel effect.

[0033] The corrected wind power forecast satisfies: ; in, This is the corrected wind power forecast. This represents the predicted wind power base output from the LSTM model. The formula for calculating the photovoltaic terrain compensation factor is: ; In the formula, This is a photovoltaic terrain compensation factor; This is the tilt angle correction factor, which reflects the impact of the deviation between the slope tilt angle and the optimal tilt angle on irradiation reception. This is the aspect correction factor, reflecting the irradiance loss when the slope aspect deviates from due south. This is the terrain shading correction factor, reflecting the shading effect of surrounding mountains on the photovoltaic array.

[0034] The corrected photovoltaic power forecast satisfies: ; in, The corrected photovoltaic power forecast. This represents the base-predicted photovoltaic power output by the LSTM model.

[0035] S3. Based on the ultra-short-term predicted power of wind and solar power, and according to the real-time operating parameters of the power collection line, the optimal wind and solar power output allocation value is obtained by constructing and solving a dynamic power allocation optimization model that takes into account terrain line loss.

[0036] The objective function expression of the dynamic power allocation optimization model is: ; In the formula, The target value for comprehensive transmission benefits; To improve the transmission efficiency of the power collection line, additional line losses caused by mountainous terrain should be considered. This refers to the allocated output power of the fan; The allocated photovoltaic output power; This is the penalty coefficient for hill-climbing fluctuations; This represents the total output change in adjacent control cycles.

[0037] Among them, the transmission efficiency of the collector line The calculation formula is: ; In the formula, The resistance per unit length of the conductor; This is the total length of the collector line; The terrain loss coefficient is calculated from the elevation difference and terrain undulation of the line. This is the line's rated voltage.

[0038] The constraints of the optimization model include line capacity constraints, upper and lower limits of output constraints, and voltage deviation constraints at the grid connection point.

[0039] This step uses an adaptive genetic algorithm to solve the dynamic power allocation optimization model: based on the fitness value of individuals in the population, the crossover and mutation probabilities are dynamically adjusted. Individuals with fitness lower than the population average are given a larger crossover and mutation probability, while individuals with fitness higher than the population average are given a smaller crossover and mutation probability, thus obtaining the globally optimal wind and solar power output allocation solution.

[0040] S4. Based on the optimal wind and solar power output allocation value, a rolling time-domain optimization mechanism is used to perform forward verification and smoothing correction of power in future multi-period periods to obtain a preventive power regulation command that meets the ramp rate constraint.

[0041] Specifically, based on a rolling time-domain optimization framework with a 30-minute prediction window and a 5-minute control step, the power ramp rate constraint is embedded into the allocation solution process by successively verifying the power change rate of the next 6 control steps. The power sequence that exceeds the limit is smoothed and corrected to obtain a preventive power regulation command that meets the ramp rate constraint.

[0042] The formula for the slope rate constraint is: ; In the formula, For the first Total output per step length; To control the step size; This represents the maximum permissible gradeability.

[0043] S5. Based on the actual execution feedback data of the preventive power regulation command, the control model parameters are obtained through adaptive iterative optimization by incrementally training the prediction model and dynamically correcting the compensation factor coefficients.

[0044] Specifically, based on real-time collected actual power output of wind and solar power and line operation data, the prediction error and allocation deviation are calculated, and the LSTM prediction model is incrementally trained every 24 hours using the latest operation data. At the same time, the coefficient parameters of the mountain terrain compensation factor are dynamically calibrated to obtain the iteratively optimized prediction model and compensation factor parameters.

[0045] This embodiment was run continuously for 3 months, and the running data was compared with the traditional fixed ratio method and the ordinary dynamic allocation method. The core performance indicators are shown in Table 1 below: Table 1: Performance Indicator Parameter Table

[0046] The comparison results show that the terrain compensation prediction method adopted in this embodiment reduces the power prediction error in mountainous scenarios, providing a precise data foundation for subsequent power allocation. The dynamic collaborative allocation strategy fully adapts to the dynamic changes in wind and solar power output, while incorporating terrain-related loss optimization, significantly improving the utilization efficiency of the power collection lines and reducing transmission losses. The rolling time-domain preventive adjustment disperses power fluctuations into multi-cycle smoothing, greatly reducing the number of line overloads and smoothing voltage fluctuations at the grid connection point. The closed-loop iteration mechanism ensures the long-term stability of control accuracy. The overall solution effectively improves the power generation efficiency and operational stability of the power station. In summary, this embodiment, through a four-layer closed-loop collaborative control system architecture, combined with a five-step control process of terrain correction prediction, terrain line loss optimization allocation, rolling time-domain look-ahead adjustment, and closed-loop adaptive iteration, effectively solves the problems of low prediction accuracy, poor allocation adaptability, lag response, and high transmission losses in existing power control methods in mountainous scenarios. Under the premise of ensuring the safe and stable operation of the system, it significantly improves the operating efficiency of the power collection lines and the power generation efficiency of the power station, demonstrating good practicality and promotion value.

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for coordinated power control of mountainous landscape power collection lines, characterized in that, Includes the following steps: S1. Based on the sensing equipment and digital elevation model deployed at the site, the basic dataset of multi-source fusion is obtained by synchronously collecting wind and solar turbine operation data, on-site environmental data and extracting terrain feature parameters at each location. S2. Based on the basic dataset, the basic prediction power of wind and solar power is output through the LSTM basic prediction model, and the mountain terrain compensation factor is introduced to correct the basic prediction power to obtain the terrain-adapted ultra-short-term prediction power of wind and solar power. S3. Based on the ultra-short-term predicted power of wind and solar power, and according to the real-time operating parameters of the power collection line, the optimal wind and solar power output allocation value is obtained by constructing and solving a dynamic power allocation optimization model that takes into account terrain line loss. S4. Based on the optimal wind and solar power output allocation value, a rolling time domain optimization mechanism is used to perform forward verification and smoothing correction of power in future multi-period periods to obtain a preventive power regulation command that meets the ramp rate constraint. S5. Based on the actual execution feedback data of the preventive power regulation command, the control model parameters are obtained through adaptive iterative optimization by incrementally training the prediction model and dynamically correcting the compensation factor coefficients.

2. The method for coordinated power control of mountainous landscape and solar power collection lines according to claim 1, characterized in that: Step S1 is as follows: By deploying sensing terminals at wind turbines, photovoltaic arrays, and power collection lines, and using a 5m resolution digital elevation model of the site, operational data including real-time wind turbine output, real-time photovoltaic output, and power collection line current, voltage, and power are collected, along with environmental data including wind speed, wind direction, ambient temperature, solar irradiance, and component temperature. Simultaneously, terrain parameters such as altitude, slope angle, aspect angle, terrain curvature, and terrain shading height at each location are extracted to obtain a multi-source basic dataset containing operational data, environmental data, and terrain parameters.

3. The method for coordinated power control of mountainous landscape and solar power collection lines according to claim 2, characterized in that: In step S2, the mountain terrain compensation factor includes the wind power terrain compensation factor, and the calculation formula for the wind power terrain compensation factor is as follows: ； In the formula, For wind power terrain compensation factors; This is the altitude correction factor, reflecting the change in air density with altitude; This is the slope correction factor, which reflects the impact of slope airflow tilt on wind energy capture. This is the terrain acceleration coefficient, reflecting the wind speed gain due to the valley funnel effect; The corrected wind power forecast satisfies: ; in, This is the corrected wind power forecast. This represents the predicted wind power base output from the LSTM model.

4. The method for coordinated power control of mountainous landscape and solar power collection lines according to claim 3, characterized in that: The mountain terrain compensation factor in step S2 includes the photovoltaic terrain compensation factor, and the calculation formula for the photovoltaic terrain compensation factor is as follows: ； In the formula, This is a photovoltaic terrain compensation factor; This is the tilt angle correction factor, which reflects the impact of the deviation between the slope tilt angle and the optimal tilt angle on irradiation reception. This is the aspect correction factor, reflecting the irradiance loss when the slope aspect deviates from due south. This is the terrain shading correction factor, reflecting the shading effect of surrounding mountains on the photovoltaic array; The corrected photovoltaic power forecast satisfies: ; in, The corrected photovoltaic power forecast. This represents the base-predicted photovoltaic power output by the LSTM model.

5. The method for coordinated power control of mountainous landscape and solar power collection lines according to claim 4, characterized in that: The dynamic power allocation optimization model considering terrain line loss in step S3 has the following objective function expression: ； In the formula, The target value for comprehensive transmission benefits; To improve the transmission efficiency of the power collection line, additional line losses caused by mountainous terrain should be considered. This refers to the allocated output power of the fan; The allocated photovoltaic output power; This is the penalty coefficient for hill-climbing fluctuations; This represents the total output change between adjacent control cycles; The constraints of the optimization model include line capacity constraints, upper and lower limits of output constraints, and voltage deviation constraints at the grid connection point.

6. The method for coordinated power control of mountainous landscape power collection lines according to claim 5, characterized in that: Transmission efficiency of collector lines The calculation formula is: ； in, The resistance per unit length of the conductor; This is the total length of the collector line; The terrain loss coefficient is calculated from the elevation difference and terrain undulation of the line. This is the line's rated voltage.

7. The method for coordinated power control of mountainous landscape and solar power collection lines according to claim 5, characterized in that: In step S3, the dynamic power allocation optimization model is solved by an adaptive genetic algorithm. Specifically, based on the fitness value of individuals in the population, the crossover probability and mutation probability are dynamically adjusted. Individuals with fitness lower than the population average are given a higher crossover and mutation probability, while individuals with fitness higher than the population average are given a lower crossover and mutation probability, thus obtaining the globally optimal wind and solar power output allocation solution.

8. The method for coordinated power control of mountainous landscape power collection lines according to claim 7, characterized in that: Step S4 is as follows: Based on a rolling time-domain optimization framework with a 30-minute prediction window and a 5-minute control step, the power ramp rate constraint is embedded into the allocation solution process by successively verifying the power change rate of the next 6 control steps. The power sequence that exceeds the limit is smoothed and corrected to obtain a preventive power regulation command that meets the ramp rate constraint. The formula for the slope rate constraint is: ； In the formula, For the first Total output per step length; To control the step size; This represents the maximum permissible gradeability.

9. A method for coordinated power control of mountainous landscape and solar power collection lines according to claim 8, characterized in that: Step S5 is as follows: Based on real-time collected data on actual wind and solar power output and line operation, the prediction error and allocation deviation are calculated. The LSTM prediction model is incrementally trained every 24 hours using the latest operating data. At the same time, the coefficient parameters of the mountain terrain compensation factor are dynamically calibrated to obtain the iteratively optimized prediction model and compensation factor parameters.

10. A method for coordinated power control of mountainous landscape and solar power collection lines according to any one of claims 1-9, characterized in that: A power coordination control system for mountainous and solar power collection lines used to implement the above method includes: a data acquisition layer, a data analysis layer, a control decision layer, and an execution feedback layer that are connected in sequence via communication. The data acquisition layer is distributed across key nodes of wind turbine locations, photovoltaic block arrays, and power collection lines. It includes wind speed sensors, solar irradiance meters, electrical power monitoring devices, and terrain data input modules. These are used to collect operational data, environmental data, and terrain parameters, and then upload them to the data analysis layer via fiber optic communication. The data analysis layer is deployed on the central control server of the site and includes a data fusion preprocessing module, an LSTM basic prediction module, and a mountain terrain compensation calculation module. It is used to complete data cleaning, basic power prediction and terrain correction calculation, and output the terrain-adapted wind and solar power prediction to the control decision layer. The control decision layer is deployed on the central control server of the station and includes a dynamic power allocation optimization module, a rolling time domain adjustment module, and an adaptive algorithm solution module, which are used to generate the optimal power allocation scheme and preventive adjustment instructions, and send them to the execution feedback layer. The execution feedback layer includes a wind turbine pitch control system and a photovoltaic inverter control system, which are deployed in the wind turbine nacelle and the photovoltaic array inverter room, respectively. They are used to receive control commands to adjust the wind and solar power output and to feed back the actual operating status to the data analysis layer and the control decision layer in real time, forming a closed-loop control link.

Images