Photovoltaic-wind power-energy storage integrated power generation system based on intelligent optimization

Abstract

The invention discloses a photovoltaic-wind power-energy storage integrated power generation system based on intelligent optimization in the technical field of intelligent scheduling of energy storage systems. The photovoltaic-wind power-energy storage integrated power generation system comprises a power generation module; the hybrid energy prediction module is used for predicting the load of the power generation module by using a hybrid prediction model according to the meteorological data and the equipment operation state data acquired from the communication module; the optimal scheduling module is used for performing real-time power generation power compensation, short-term charging and discharging scheduling and long-term energy storage capacity proportion planning through multi-time scale cooperation; the energy storage collaboration module is used for constructing a full-life-cycle virtual mirror image of the wind and light storage system by using a digital twinning technology and establishing a time-sharing hierarchical energy storage architecture; a communication module and a feedback module. According to the invention, through power generation prediction, energy supply scheduling and energy storage ratio planning on multiple time scales, the overall operation efficiency and stability of the system are improved, the service life of equipment is optimized, the equipment, manual and leveling cost of kilowatt-hour electricity is reduced, and the shutdown loss of abandoning light and electricity is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent scheduling of energy storage systems, and specifically relates to an integrated photovoltaic-wind power-storage power generation system based on intelligent optimization. Background Art

[0002] With the intensification of the global energy crisis and growing awareness of environmental protection, the development and utilization of renewable energy has become a crucial component of national energy strategies. Photovoltaic and wind power, two major forms of renewable energy generation, are clean and renewable, playing a significant role in reducing fossil energy consumption and lowering carbon emissions. However, due to the intermittent and unstable nature of solar and wind energy, individual photovoltaic or wind farms alone cannot meet the grid's demand for a stable power supply. Energy storage technology can address the intermittent and unstable nature of renewable energy generation. By storing excess energy during periods of abundant sunlight or strong winds and releasing it during periods of low sunlight or weak winds, it smooths power output and improves energy efficiency. The introduction of energy storage systems enables photovoltaic and wind power generation systems to better adapt to grid needs and increase the proportion of renewable energy integrated into the grid.

[0003] With the rapid development of information technology, intelligent optimization technology has been widely applied in the energy sector. In integrated photovoltaic, wind power, and energy storage power generation systems, intelligent optimization technology can enable real-time monitoring, dynamic adjustment, and optimized control of the power generation system. For example, patent publication number CN110165690A demonstrates that intelligent algorithms, based on the power of renewable energy generation, control reverse power complementation and coordinated output. This improves the overall operational efficiency and stability of the system and addresses the technical shortcomings of renewable energy microgrids in terms of power supply fluctuations and instability.

[0004] However, existing technologies still find it difficult to solve the problem of unstable power generation caused by photovoltaic and wind power generation depending on weather conditions. Energy storage systems need to be frequently charged and discharged to balance supply and demand. Extreme weather (such as continuous cloudy days or windless periods) may lead to insufficient energy storage capacity, affecting power supply reliability. The high cost of power generation and energy storage equipment creates development obstacles for regions with limited funds.

[0005] To this end, it is necessary to propose a comprehensive system that can plan energy storage capacity ratios, dispatch energy storage equipment output, and conduct weather forecasts on multiple time scales. It can also charge energy storage equipment during price troughs to meet midday scheduling needs, thereby reducing operating costs, optimizing equipment life, and improving wind and solar power generation and grid output efficiency. This is an integrated photovoltaic-wind power-storage power generation system based on intelligent optimization. Summary of the Invention

[0006] In order to solve the above problems, the purpose of the present invention is to provide a photovoltaic-wind power-storage integrated power generation system based on intelligent optimization. Through power generation prediction, energy supply scheduling and energy storage ratio planning on multiple time scales, the efficiency and stability of the overall operation of the system can be improved, the life of the equipment can be optimized, the equipment, labor and levelized electricity costs can be reduced, and the downtime losses caused by "abandoned light and power" can be reduced.

[0007] To achieve the above objectives, the technical solution of the present invention is as follows: a photovoltaic-wind power-storage integrated power generation system based on intelligent optimization, comprising a power generation module, a hybrid energy prediction module, an optimization scheduling module, an energy storage coordination module, a communication module and a feedback module,

[0008] The power generation module is used to generate electricity in different ways and transmit it to the optimization scheduling module and energy storage coordination module;

[0009] The hybrid energy prediction module is used to predict the load of the power generation module and adjust the parameters of the power generation module using a hybrid prediction model based on the meteorological data and equipment operation status data obtained from the communication module;

[0010] An optimization scheduling module for real-time generation power compensation, short-term charge and discharge scheduling, and long-term energy storage capacity ratio planning through multi-time scale coordination;

[0011] The energy storage collaboration module uses digital twin technology to build a virtual mirror of the wind, solar, and energy storage system throughout its life cycle, and establishes a time-sharing and hierarchical energy storage architecture.

[0012] Communication module, used to obtain meteorological data from external databases, obtain equipment operating status data from equipment sensors, and transmit signals between the above modules;

[0013] The feedback module is used to perform rolling time domain optimization, taking the long-term planning results as the boundary conditions for short-term scheduling, and feeding back the decision results of the optimization scheduling module to resolve coupling conflicts on the time scale.

[0014] Furthermore, the power generation module includes a solar wind power generation unit and a fuel cell unit.

[0015] A solar-wind power generation unit is used to convert solar energy and wind kinetic energy into electrical energy using semiconductors and a transmission structure. The solar-wind power generation unit includes a support column, a sleeve is rotatably sleeved on the support column, and a plurality of rotating rods are vertically sleeved on the outer periphery of the sleeve. The ends of the rotating rods away from the sleeves are fixedly connected to inner photovoltaic panels, and one side of the inner photovoltaic panels is fixedly connected to outer photovoltaic panels. The inner photovoltaic panels and the outer photovoltaic panels are perpendicular to each other. A transmission rod is coaxially rotatably sleeved in the support column. The rotating rods all pass through the sleeves and are rotatably connected to the transmission rods. The outer periphery of the rotating rods near one end of the transmission rods is provided with rotating shaft teeth. The top of the transmission rod is fixedly connected to a rotating motor. The output shaft of the rotating motor is vertically upward and coaxially fixedly connected to a rotating gear disk. The rotating gear disk is meshed with the rotating shaft teeth.

[0016] Fuel cell unit that converts the chemical energy of hydrogen and oxygen directly into electrical energy.

[0017] The basic solution works as follows: the system achieves closed-loop control from data perception and predictive analysis to decision optimization through the organic collaboration of six core modules (power generation, forecasting, scheduling, energy storage, communication, and feedback). The communication module integrates meteorological data and equipment status data to provide a foundation for forecasting and scheduling. The hybrid energy forecasting module builds a dynamic forecasting model based on multi-source data to adjust power generation parameters in advance. The optimized scheduling module balances power fluctuations and energy storage requirements through multi-timescale collaboration (real-time, short-term, and long-term). The energy storage collaboration module uses digital twin technology to simulate the entire equipment lifecycle and optimize charging and discharging strategies to extend lifespan. The feedback module decouples long-term planning from short-term scheduling through rolling time domain optimization to resolve timescale conflicts.

[0018] The beneficial effects of the basic solution are: 1. Multi-time scale coordinated scheduling reduces power fluctuations, improves grid-connected power quality, and the energy storage hierarchical architecture realizes efficient energy storage and release, reducing the rate of curtailment of solar and wind power.

[0019] 2. The energy storage collaborative module optimizes the charging and discharging strategies, reduces the damage to the battery caused by deep charging and discharging, and dynamically adjusts parameters to reduce the risk of equipment overload and extend the service life of key components.

[0020] 3. Reducing energy loss directly reduces the levelized cost of electricity, extending equipment life reduces replacement frequency, reduces capital expenditures, and automated scheduling reduces manual intervention, thereby reducing operation and maintenance costs.

[0021] 4. The hybrid prediction model integrates meteorological and equipment data to improve prediction accuracy under complex working conditions. The rolling time domain optimization mechanism dynamically adjusts the decision boundary to enhance robustness to uncertainty. The angle between the photovoltaic panel and sunlight is adjusted through the light-adaptive posture of the solar-wind power generation unit to improve the absorption efficiency of solar energy and reduce the accumulation of dust on the photovoltaic panel, thus avoiding the decline in power generation efficiency of the photovoltaic panel with long-term use.

[0022] 5. Through energy storage coordination and multi-scale scheduling, the utilization rate of wind and solar resources can be maximized, the phenomenon of "abandoned solar and wind power" can be reduced, and the penetration rate of clean energy can be improved.

[0023] Furthermore, the hybrid energy prediction module includes a preprocessing unit, a multi-source data fusion unit and a prediction unit.

[0024] A preprocessing unit for processing outliers and missing values in meteorological data using dynamic thresholding and co-interpolation methods;

[0025] Multi-source data fusion unit, used to fuse meteorological data using the spatiotemporal alignment method to establish a spatiotemporal feature matrix including irradiance intensity, cloud thickness and wind direction turbulence coefficient;

[0026] The prediction unit is used to use the trained LSTM-Attention-WOA hybrid prediction model to predict short-term and long-term meteorological changes, including changes in irradiance intensity, cloud thickness, and wind turbulence coefficient.

[0027] The basic scheme has the following beneficial effects: 1. Through the dynamic threshold method and collaborative interpolation method, the preprocessing unit can effectively identify and correct outliers in meteorological data, fill in missing data, improve data quality, and provide reliable input for subsequent forecasts. The multi-source data fusion unit uses a spatiotemporal alignment method to integrate meteorological data from different sources and different temporal resolutions (such as irradiance intensity, cloud thickness, and wind turbulence coefficient) into a unified spatiotemporal feature matrix, capturing the spatial correlation and temporal dynamics between meteorological elements and enhancing the richness of input information for the forecast model.

[0028] 2. The prediction unit uses a hybrid LSTM-Attention-WOA model, combining the time series modeling capabilities of the long short-term memory (LSTM) network, the key feature extraction capabilities of the attention mechanism (Attention), and the global optimization capabilities of the Whale Optimization Algorithm (WOA) to achieve accurate forecasts of short-term (e.g., hourly) and long-term (e.g., daily / weekly) weather changes. The model output includes key parameters such as irradiance intensity, cloud thickness, and wind turbulence coefficient, which are directly related to photovoltaic power generation, wind power output, and energy storage requirements, providing high-value decision-making basis for optimized scheduling.

[0029] 3. Through high-precision weather forecasts, the hybrid energy forecasting module can more accurately predict load changes in the power generation module, reducing the impact of forecast errors on scheduling strategies and mitigating the risk of curtailed solar and wind power. Multi-scale weather forecasts provide the scheduling optimization module with richer information, enabling it to proactively adjust power generation parameters, optimize energy storage charging and discharging strategies, and improve the system's adaptability to weather fluctuations. Accurate forecasting and scheduling reduce equipment operating time under extreme conditions, minimizing fatigue damage. Combined with the digital twin technology of the energy storage collaboration module, this further extends equipment life.

[0030] 4. Dynamic thresholding and collaborative interpolation methods can adapt to the characteristics of meteorological data in different regions and seasons, improving the system's generalization capabilities. The layered structure of the hybrid energy forecasting module (preprocessing, fusion, and forecasting) facilitates the subsequent introduction of new algorithms or data sources, supporting system expansion and upgrades.

[0031] Furthermore, the LSTM-Attention-WOA hybrid prediction model includes a three-layer architecture. The first layer is a bidirectional long short-term memory network - LSTM, which is used to extract forward and / or backward time series feature dependencies;

[0032] The second layer is the attention mechanism used to dynamically assign meteorological factor weights;

[0033] The third layer is the Whale Optimization Algorithm (WOA), which is used to minimize the validation set.

[0034] The basic solution has the following benefits: 1. The bidirectional LSTM captures the bidirectional temporal dependencies of meteorological data, reducing the risk of unidirectional models missing unexpected meteorological changes (such as a sudden drop in irradiance caused by rapid cloud movement). The attention mechanism automatically focuses on key meteorological factors. For example, in photovoltaic power generation forecasts, cloud thickness and irradiance are prioritized; in wind power forecasts, the wind direction turbulence coefficient is weighted more strongly. WOA uses validation set error feedback to adjust model parameters, avoiding local optimality traps and improving long-term forecast stability.

[0035] 2. High-precision forecasts enable the optimized scheduling module to adjust power generation in real time, reducing frequent charging and discharging of energy storage due to forecast deviations and extending battery life. Long-term weather forecasts (such as seasonal irradiation variations) guide energy storage capacity planning, avoiding overinvestment or insufficient capacity. By anticipating sudden weather changes (such as a sudden drop in wind speed), the system can proactively activate backup power or adjust energy storage strategies, reducing losses from curtailed solar and wind power.

[0036] 3. The bidirectional LSTM-Attention-WOA model is more robust to sudden meteorological changes (such as extreme weather), ensuring stable system operation under adverse conditions. The three-tier architecture supports the introduction of new features (such as the air quality index) or the replacement of optimization algorithms (such as particle swarm optimization) to adapt to future needs. The model can be ported to different geographic regions or energy types (such as hydropower and tidal power), reducing development costs.

[0037] Furthermore, the optimization scheduling module includes a real-time compensation module, a short-term scheduling module and a long-term planning module.

[0038] A real-time compensation unit for controlling the power generation voltage and frequency using virtual synchronous machine technology to suppress instantaneous fluctuations in power generation parameters;

[0039] The short-term dispatch unit uses deep reinforcement learning technology to schedule the charging and discharging of the energy storage collaborative module during the valley and peak periods of the power grid to meet daily load demand.

[0040] The long-term planning unit is used to build a virtual model of the energy storage equipment of the energy storage synergy module using digital twin simulation technology, and adjust the wind-solar-storage capacity ratio based on the long-term meteorological changes predicted by the prediction unit to adapt to seasonal changes in sunlight and wind speed.

[0041] Furthermore, the model construction objects of the digital twin simulation technology include solar and wind power generation units, fuel cell units and energy storage equipment of the energy storage collaborative module.

[0042] The basic solution offers the following benefits: 1. The real-time compensation unit simulates the inertia of synchronous generators through virtual synchronous generator technology (VSG), rapidly responding to transient fluctuations in voltage and frequency. This reduces power oscillations caused by sudden changes in wind and solar resources (such as cloud cover and sudden changes in wind speed), improving grid-connected power quality. VSG technology enables renewable energy generation systems to have frequency and voltage regulation capabilities similar to those of traditional power sources, reducing impact on the grid and improving stability when a high proportion of renewable energy is connected to the grid.

[0043] 2. The short-term dispatch unit utilizes deep reinforcement learning (DRL) technology to dynamically adjust energy storage charging and discharging strategies based on peak and valley periods of grid power. Charging is prioritized during low-load periods (such as nighttime) to reduce electricity purchase costs; energy storage is released during peak load periods (such as daytime) to smooth the load curve and reduce grid pressure. Through DRL's real-time decision-making capabilities, it accurately matches intraday load fluctuations, reduces idle energy storage or excessive charging and discharging, and improves energy storage utilization.

[0044] 3. The long-term planning unit uses digital twin simulation technology to construct virtual models of photovoltaic, wind power, fuel cells, and energy storage equipment. This model simulates the impact of seasonal changes in sunlight and wind speed on power generation, dynamically adjusting the ratio of wind, solar, and energy storage capacity: in seasons with abundant sunlight, photovoltaic capacity is increased to reduce reliance on wind power; in seasons with stable wind speeds, the synergy between wind power and energy storage is optimized to reduce wind curtailment. Through simulations, equipment aging and performance degradation are predicted, allowing for pre-planned equipment replacement or expansion, avoiding power generation losses or overinvestment due to insufficient capacity.

[0045] 4. Digital twin technology covers solar and wind power generation units, fuel cell units, and energy storage devices, enabling dynamic simulation of the entire system and capturing the coupling effects between different devices (such as wind-solar complementarity and synergy between energy storage and fuel cells). Through virtual models, equipment failure scenarios (such as aging photovoltaic panels and wear of wind turbine blades) can be simulated in advance to develop preventative maintenance strategies and reduce downtime. This allows for rapid assessment of the impact of new technologies (such as new energy storage materials) or new strategies (such as demand response) on the system, accelerating technological iteration.

[0046] Furthermore, the energy storage coordination module includes a fast energy storage unit, a short-term energy storage unit and a long-term energy storage unit.

[0047] The fast energy storage unit includes a supercapacitor energy storage device for rapid response to real-time compensation of power generation parameter fluctuations;

[0048] The short-term energy storage unit includes lithium iron phosphate battery energy storage equipment, which is used for short-term coordinated scheduling of charging and discharging of energy storage equipment;

[0049] The long-term energy storage unit includes hydrogen energy storage equipment, which is used for long-term energy storage on a year-round time scale, and is used to coordinate seasonal adjustments to the ratio of wind and solar energy storage capacity.

[0050] Furthermore, the hydrogen energy storage device includes an electrolytic hydrogen generator, a hydrogen compressor and a hydrogen storage tank, and the hydrogen storage tank is connected to the fuel cell unit through a control valve.

[0051] The basic solution offers the following benefits: 1. The millisecond-level response capability of the fast energy storage unit (supercapacitor) mitigates instantaneous fluctuations in wind and solar power generation (such as cloud cover and sudden changes in wind speed) in real time, ensuring grid frequency stability. The high-frequency charge and discharge capabilities of the short-term energy storage unit (lithium iron phosphate battery) work in conjunction with the short-term scheduling module to smooth the daily load curve, reducing electricity purchase costs and improving equipment utilization. The cross-seasonal energy storage capacity of the long-term energy storage unit (hydrogen storage) addresses the seasonal mismatch between wind and solar resources. Through hydrogen production by electrolysis, hydrogen storage tanks, and fuel cells, long-term energy storage and conversion are achieved.

[0052] 2. Matching different timescale requirements based on the characteristics of energy storage technology (response speed, cycle life, and cost) avoids the limitations of a single technology and reduces initial investment and operation and maintenance costs. Through digital twin simulation and seasonal weather forecasting, the ratio of wind, solar, and storage capacity can be optimized to reduce the risk of overinvestment or insufficient capacity.

[0053] 3. Wind, solar, and hydrogen storage synergize to form a multi-level energy buffer system encompassing "on-demand, short-term storage, and long-term reserve," enhancing the system's adaptability to extreme weather and equipment failures. Rapid energy storage and short-term energy storage units can participate in ancillary services such as grid peak regulation and frequency regulation, increasing revenue streams and improving grid stability.

[0054] 4. Utilizing excess wind and solar power to produce hydrogen enables secondary storage of clean energy and zero-carbon power generation, reducing wasted energy and contributing to decarbonization in the transportation and industrial sectors. System-wide optimization improves wind and solar resource utilization, reduces reliance on fossil fuels, and significantly reduces carbon emissions per unit of electricity.

[0055] 5. Digital twin technology builds a virtual model of the energy storage equipment, providing early warning of failure risks and reducing downtime. A control valve connects the hydrogen storage tank and fuel cell unit, automating the management of hydrogen delivery and power generation. The multi-layer energy storage architecture provides safety redundancy to ensure continuous system operation.

[0056] Furthermore, the communication module includes a data acquisition unit and a signal transmission unit,

[0057] A data acquisition unit is used to obtain meteorological data from an external server and collect equipment operating status data of the above modules. The meteorological data includes radiation intensity, cloud thickness, wind direction and wind force parameters, temperature and humidity. The equipment operating status data includes the DC voltage / current, AC power, IGBT temperature and power of the solar wind power generation unit, the pitch angle, gearbox oil temperature, vibration spectrum and generator speed of the wind power unit, and the SOC / SOH, single cell voltage, temperature difference and number of charge and discharge cycles of the energy storage device in the energy storage coordination module;

[0058] The signal transmission unit is used to transmit signals between the above modules.

[0059] The basic solution offers the following benefits: 1. A data acquisition unit acquires meteorological parameters such as irradiance intensity, cloud thickness, wind direction and force, temperature and humidity from an external server, providing high-precision input for the prediction model and improving the accuracy of wind and solar power generation forecasts. Real-time collection of key parameters such as solar and wind power units (DC voltage / current, IGBT temperature), wind turbine units (pitch angle, gearbox oil temperature), and energy storage devices (SOC / SOH, cell voltage-temperature difference) enables comprehensive awareness of device health, providing a data foundation for fault warning and lifespan prediction.

[0060] 2. The signal transmission unit ensures real-time synchronization of meteorological data and equipment status data to forecasting, scheduling, and energy storage modules, supporting dynamic system responses (such as rapidly adjusting energy storage charging and discharging strategies in the event of sudden changes in cloud cover). Based on real-time data, the optimized scheduling module dynamically matches wind and solar power output with load demand, while the energy storage module precisely controls charging and discharging power, improving overall system efficiency and stability.

[0061] 3. By monitoring parameters such as IGBT temperature, gearbox oil temperature, and vibration spectrum, potential equipment failures (such as photovoltaic inverter overheating and wind turbine gearbox wear) can be detected in advance, reducing unplanned downtime. Based on equipment status data (such as the number of charge and discharge cycles and the voltage and temperature difference between cells), differentiated maintenance plans can be developed to reduce operation and maintenance costs and equipment replacement frequency.

[0062] 4. Real-time data collection (such as energy storage device temperature difference and photovoltaic DC voltage) can be set with multi-level safety thresholds. When an anomaly is triggered, faulty equipment is automatically isolated to prevent the escalation of the incident. The communication module supports cross-validation of multi-source data (such as meteorological data and power generation), improving data credibility and avoiding misjudgments caused by errors in a single data source.

[0063] 5. The data acquisition unit supports multi-protocol compatibility (such as Modbus and IEC 61850), enabling flexible integration of new devices (such as new energy storage technologies and smart sensors), reducing system upgrade costs. Standardized signal transmission protocols facilitate integration with external platforms (such as power grid dispatch systems and meteorological service providers), expanding data application scenarios (such as participating in power market transactions and sharing meteorological services).

[0064] Furthermore, the feedback module includes a rolling optimization unit and a correction unit,

[0065] A rolling optimization unit, used to continuously optimize energy storage scheduling and planning using online rolling optimization;

[0066] The correction unit is used to compensate for prediction errors using closed-loop correction to adapt to the uncertainties of wind and solar fluctuations, load changes and equipment aging.

[0067] The basic solution offers the following benefits: 1. Through online rolling optimization, the system continuously adjusts energy storage charging and discharging strategies and capacity planning based on the latest meteorological data, equipment status, and load demand, ensuring that the dispatch plan remains adaptable to dynamic changes. Rolling optimization considers multi-timescale uncertainties (such as sudden changes in wind and solar power and load forecast deviations), generating a more robust dispatch plan and reducing the risk of system instability caused by sudden disturbances. Dynamic optimization of energy storage charging and discharging power reduces peak-valley price arbitrage errors, thereby improving energy storage utilization and profitability.

[0068] 2. Based on the deviation between actual power generation data and predicted values, the correction unit adjusts prediction model parameters (such as LSTM-Attention-WOA model weights) in real time to reduce prediction errors and improve wind and solar power forecast accuracy. To address high-frequency disturbances such as rapid cloud movement and sudden changes in wind speed, the correction unit uses a feedback mechanism to quickly correct scheduling instructions to avoid overcharging or over-discharging energy storage. By combining actual load changes with equipment performance degradation data (such as energy storage SOC deviation and photovoltaic panel efficiency decline), long-term planning parameters are dynamically adjusted to extend equipment life and reduce the risk of capacity shortages.

[0069] 3. Rolling optimization and correction units work together to form a closed loop of "forecasting-dispatch-feedback-re-optimization," effectively addressing multiple uncertainties such as intermittent wind and solar resources, load fluctuations, and equipment aging. By correcting the dispatch strategy in real time, the system can maintain critical functions (such as prioritizing power supply to critical loads) in the event of sudden equipment failures or communication interruptions, reducing the scope of the incident.

[0070] 4. The calibration unit supports real-time updates of forecast model parameters to adapt to new meteorological patterns (such as frequent extreme weather events) or changes in equipment characteristics (such as the integration of new energy storage technologies). The rolling optimization unit can quickly respond to policy adjustments (such as changes in peak and valley electricity prices) or market demand (such as the participation of virtual power plants in the electricity market), improving system adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 Schematic diagram of a photovoltaic-wind power-storage integrated power generation system based on intelligent optimization in an embodiment of the present invention.

[0072] Figure 2 Schematic diagram of a power generation module in an embodiment of the present invention.

[0073] Figure 3 Schematic diagram of a hybrid energy prediction module in an embodiment of the present invention.

[0074] Figure 4 Schematic diagram of an optimization scheduling module in an embodiment of the present invention.

[0075] Figure 5 Schematic diagram of the energy storage coordination module in an embodiment of the present invention.

[0076] Figure 6 Schematic diagram of a communication module in an embodiment of the present invention.

[0077] Figure 7 Schematic diagram of a feedback module in an embodiment of the present invention.

[0078] Figure 8 It is an axonometric cross-sectional view of a solar-wind power generation unit in an embodiment of the present invention.

[0079] The figure marks in the drawings of the specification include: 1. support column; 2. transmission rod; 3. sleeve; 4. rotating motor; 5. rotating gear plate; 6. rotating rod; 7. rotating shaft gear; 8. inner photovoltaic panel; 9. outer photovoltaic panel. DETAILED DESCRIPTION

[0080] The following is further described in detail through specific implementation methods:

[0081] Example 1

[0082] Basically as attached Figure 1 、 Figure 2 、 Figure 3 、 Figure 4 、 Figure 5 、 Figure 6 、 Figure 7 and Figure 8 Shown: A photovoltaic-wind power-energy storage integrated power generation system based on intelligent optimization, including a power generation module, a hybrid energy prediction module, an optimization scheduling module, an energy storage coordination module, a communication module and a feedback module.

[0083] The power generation module is used to generate electricity in different ways and transmit it to the optimization scheduling module and the energy storage coordination module. The power generation module includes a wind power generation unit and a fuel cell unit. The solar wind power generation unit is used to convert solar energy and wind kinetic energy into electrical energy using semiconductors and transmission structures. The solar wind power generation unit includes a support column 1. A sleeve 3 is rotatably sleeved on the support column 1. Several rotating rods 6 are vertically sleeved on the outer periphery of the sleeve 3. The ends of the rotating rods 6 away from the sleeve 3 are fixedly connected to the inner photovoltaic panel 8. One side of the inner photovoltaic panel 8 is fixedly connected to the outer photovoltaic panel 9. The inner photovoltaic panel 8 and the outer photovoltaic panel 9 are perpendicular to each other. A transmission rod 2 is coaxially rotatably sleeved in the support column 1. The rotating rods 6 all pass through the sleeve 3 and are rotatably connected to the transmission rod 2. The outer periphery of the rotating rod 6 near one end of the transmission rod 2 is provided with rotating shaft teeth 7. The top of the transmission rod 2 is fixedly connected to a rotating motor 4. The output shaft of the rotating motor 4 is vertically upward and coaxially fixedly connected to a rotating gear disc 5. The rotating gear disc 5 is engaged with the rotating shaft teeth 7; the fuel cell unit is used to directly convert the chemical energy of hydrogen and oxygen into electrical energy.

[0084] The hybrid energy prediction module is used to predict the load of the power generation module and adjust the parameters of the power generation module using a hybrid prediction model based on the meteorological data and equipment operation status data obtained from the communication module.

[0085] The optimization scheduling module is used to coordinate real-time power generation compensation, short-term charge and discharge scheduling, and long-term energy storage capacity ratio planning through multiple time scales.

[0086] The energy storage coordination module is used to use digital twin technology to build a virtual mirror of the entire life cycle of the wind, solar and storage system, and establish a time-sharing and hierarchical energy storage architecture. The energy storage coordination module includes fast energy storage units, short-term energy storage units and long-term energy storage units. The fast energy storage unit includes supercapacitor energy storage equipment, which is used for real-time compensation of rapid response to fluctuations in power generation parameters; the short-term energy storage unit includes lithium iron phosphate battery energy storage equipment, which is used for short-term coordinated scheduling of charging and discharging of energy storage equipment; the long-term energy storage unit includes hydrogen energy storage equipment, which is used for long-term energy storage on a multi-year time scale, and to cooperate with seasonal wind and solar energy storage capacity ratio adjustments. The hydrogen energy storage equipment includes an electrolytic hydrogen generator, a hydrogen compressor and a hydrogen storage tank. The hydrogen storage tank is connected to the fuel cell unit through a control valve.

[0087] The communication module is used to obtain meteorological data from an external database, obtain equipment operating status data from equipment sensors, and transmit signals between the above modules. The communication module includes a data acquisition unit and a signal transmission unit. The data acquisition unit is used to obtain meteorological data from an external server and collect the equipment operating status data of the above modules. The meteorological data includes irradiation intensity, cloud thickness, wind direction and wind force parameters, temperature and humidity. The equipment operating status data includes the DC voltage / current, AC power, IGBT temperature and power of the solar wind power generation unit, the pitch angle, gearbox oil temperature, vibration spectrum and generator speed of the wind power unit, the SOC / SOH, single cell voltage, temperature difference and number of charge and discharge cycles of the energy storage device in the energy storage collaborative module. The signal transmission unit is used to transmit signals between the above modules.

[0088] The feedback module is used to perform rolling time domain optimization, taking the long-term planning results as the boundary conditions for short-term scheduling, and feeding back the decision results of the optimization scheduling module to resolve coupling conflicts on the time scale.

[0089] The specific implementation process is as follows: This system is equipped with a solar wind power generation unit, and is also equipped with a fuel cell for using the stored hydrogen to generate electricity and store energy to compensate for the lack of power. While generating electricity, the communication module first obtains meteorological data and transmits it to the hybrid energy prediction module. The communication module establishes a connection with the meteorological satellite and the ground meteorological station through the 5G private network, and simultaneously obtains detailed meteorological data for the next 72 hours, including three-dimensional wind speed fields updated every 10 minutes, cloud optical thickness distribution maps, and solar radiation spectrum data. At the same time, the smart sensors deployed in the photovoltaic array collect the backplane temperature of the monocrystalline silicon module (typical value 45±5℃), string current fluctuations (±2% tolerance range), and the junction temperature of the inverter IGBT module (monitored by thermistors, the threshold is set to 120℃) at a frequency of 10 times per second. The vibration sensor in the wind turbine tower continuously monitors the vibration spectrum in the 0.1-100Hz frequency band. When the amplitude at 2.5Hz is detected to exceed 8mm / s 2 When the gearbox fault warning is triggered.

[0090] The hybrid energy prediction module predicts long-term and short-term weather changes, thereby adjusting the opening of photovoltaic and wind power generation devices in the solar and wind power generation units, such as Figure 8As shown, while rotating to generate wind power, the rotating motor 4 drives the rotation of the rotating gear disc 5 and then causes the rotating rod 6 to rotate axially, adjusting the angle of the inner photovoltaic panel 8 to cater to direct sunlight, and reducing the accumulation of dust on the photovoltaic panel through continuous angle adjustment and rotation, which can avoid the idle consumption of the power generation device when the relevant weather is bad, and reduce the phenomenon of abandoned light and wind when the power generation device cannot be turned on in time when the sunshine or wind is sufficient, thereby dynamically improving the power generation efficiency, and the inner photovoltaic panel 8 and the outer photovoltaic panel 9 that are perpendicular to each other prevent the photovoltaic panel from affecting the wind power generation efficiency during the angle adjustment process, so that the photovoltaic panels always have a suitable angle for wind power generation.

[0091] While the solar and wind power generation units are generating electricity, the power and voltage of the power generation devices fluctuate due to changes in meteorological parameters. The optimization scheduling module compensates and controls the power generation voltage and frequency on three time scales: real-time, short-term, and long-term, suppresses instantaneous power generation parameter fluctuations, reduces transmission power fluctuations, improves power supply quality, maintains continuous output of the power generation network, avoids the reduction of power generation during peak power supply periods, and improves the economic benefits of the power generation system and the stability of power supply to the grid. The optimization scheduling module establishes a virtual model of the power generation device and energy storage equipment through digital twin simulation, calculates the changes in energy storage conditions and the loss rate of the equipment according to the algorithm, so that the system can actually grasp the status of each device and optimize energy storage scheduling or remind staff to repair equipment and replace spare equipment according to the loss rate, reduce system downtime due to wear and aging over a long period of time, and update hardware in time for continuous power generation, thereby improving the economic benefits of the power generation system.

[0092] The energy storage collaborative module runs synchronously with the physical system through the OPC UA protocol. The digital twin maps in real time the RC equivalent circuit parameters of the supercapacitor (with a self-updated resistance value accuracy of 0.1mΩ), the three-dimensional thermal field distribution of the lithium iron phosphate battery (calculated using the finite element method, with a grid size of ≤2mm), and the defect evolution process of the hydrogen storage tank (based on the phase field model to simulate hydrogen-induced cracking). In the time-sharing and graded control strategy, the supercapacitor is responsible for suppressing power fluctuations above 10Hz (using a bandpass filter to extract high-frequency components), the lithium iron phosphate battery smoothes the 15-minute energy imbalance, and the hydrogen energy storage system starts the 5MW electrolyzer (DC efficiency 82%) during the daily low load period, with a speed of 1200Nm 3 / h, hydrogen is compressed to 35MPa in four stages and then stored in composite tanks. When sustained high wind power generation causes the bus voltage to rise by 0.5%, the Energy Management System (EMS) automatically switches to hydrogen production mode, converting the excess electricity into chemical energy for storage. Over longer periods of time, such as seasonal changes, large amounts of electricity can be stored in the form of hydrogen to compensate for insufficient photovoltaic and wind power generation, thereby curbing the long-term power generation shortages caused by seasonal meteorological changes. In shorter periods of time, fast-responding energy storage devices can be used to promptly compensate for and suppress fluctuations in power generation voltage and frequency, improving power supply stability.

[0093] The feedback module provides rolling feedback on power supply parameters and actual meteorological parameter changes, optimizes the algorithm parameters of the hybrid energy prediction module and the optimization scheduling module, and adapts to the uncertainty of model prediction caused by meteorological parameter changes, load changes and equipment aging, improves the real-time performance of the overall system in prediction and optimization, always maintains the stability of power generation, energy storage and power supply functions, maintains the system's power generation efficiency, and improves economic benefits.

[0094] For example, before a thunderstorm is about to occur, the communication module receives a cumulonimbus cloud warning from a meteorological satellite through the 5G private network, and the hybrid energy prediction module predicts that the local solar radiation value will increase from 800W / m 2 Dropped to 50W / m 2 At this time, the sensors on the photovoltaic array detected that the temperature of the monocrystalline silicon module backplane dropped from 48°C to 32°C within 2 minutes, and the string current fluctuation exceeded the ±2% tolerance range (fluctuating sharply from 12.5A to 9.8A). The system immediately responded: the supercapacitor detected high-frequency power oscillations above 10Hz (with an amplitude of 15% of the rated power) and injected 180kW of buffer power through the bidirectional DC / DC converter within 20ms, controlling the bus voltage fluctuation within ±0.2%. The lithium iron phosphate battery pack took over the load within 5 seconds based on the SOC calculated by the digital twin (from 65% to 58%), releasing 2.3MWh of electricity to fill the photovoltaic gap. At the same time, the inverter IGBT junction temperature was controlled at 105°C (below the 120°C threshold) due to the load dump. The EMS monitored that wind power was over-generated due to the pre-storm (the power of a single wind turbine was over-allocated by 15%) and immediately started the hydrogen production mode, converting the excess 3.6MW of electricity into hydrogen (increasing production by 432Nm per hour). 3 ), and predict the crack growth rate of the hydrogen storage tank through the phase field model, and automatically reduce the inflation pressure to 30MPa to delay the hydrogen embrittlement effect.

[0095] Example 2

[0096] The difference from the above embodiment is that, as shown in the attached Figure 1 、 Figure 3 、 Figure 4 、 Figure 6 and Figure 7 As shown, the hybrid energy prediction module includes a preprocessing unit, a multi-source data fusion unit, and a prediction unit. The preprocessing unit uses dynamic thresholding and collaborative interpolation to handle outliers and missing values in meteorological data. The multi-source data fusion unit uses a spatiotemporal alignment method to fuse meteorological data and establish a spatiotemporal feature matrix containing irradiation intensity, cloud thickness, and wind turbulence coefficient. The prediction unit uses a trained LSTM-Attention-WOA hybrid prediction model to predict short-term and long-term meteorological changes, including changes in irradiation intensity, cloud thickness, and wind turbulence coefficient. The LSTM-Attention-WOA hybrid prediction model has a three-layer architecture. The first layer is a bidirectional long short-term memory network (LSTM) for extracting forward and / or backward time series feature dependencies; the second layer is an attention mechanism for dynamically assigning meteorological factor weights; and the third layer is a whale optimization algorithm (WOA) for minimizing the validation set.

[0097] The optimization scheduling module includes a real-time compensation module, a short-term scheduling module and a long-term planning module. The real-time compensation unit is used to use virtual synchronous machine technology to control the power generation voltage and frequency and suppress the instantaneous fluctuations of power generation parameters; the short-term scheduling unit is used to use deep reinforcement learning technology to utilize the valley and peak segments of the power grid to schedule the charging and discharging of the energy storage coordination module to meet the daily load demand; the long-term planning unit is used to use digital twin simulation technology to build a virtual model of the energy storage equipment of the energy storage coordination module, and adjust the wind-solar energy storage capacity ratio according to the long-term meteorological change results predicted by the prediction unit to adapt to the seasonal changes in light and wind speed. The model construction objects of the digital twin simulation technology include solar-wind power generation units, fuel cell units and energy storage equipment of the energy storage coordination module.

[0098] The feedback module includes a rolling optimization unit and a correction unit. The rolling optimization unit is used to continuously optimize energy storage scheduling and planning using online rolling optimization; the correction unit is used to compensate for prediction errors using closed-loop correction to adapt to the uncertainties of wind and solar fluctuations, load changes and equipment aging.

[0099] The specific implementation process is as follows: After acquiring meteorological data, the hybrid energy prediction module preprocesses and integrates multiple sources. Cloud reflectivity data from weather radar (updated every 5 minutes) and turbulence intensity parameters from ground-based observation stations (sampled every 1 minute) are spatially interpolated using kriging to generate a 100-meter gridded feature field. Combined with satellite-derived irradiance data (Himawari-8 imagery every 10 minutes), a spatiotemporally aligned fusion matrix is constructed. This matrix is then fed into a trained bidirectional LSTM-Attention-WOA hybrid prediction model for short-term 48-hour and long-term 3-day weather forecasts. For PV output forecasting, the model first calculates theoretical irradiance based on solar altitude, then corrects the actual irradiance using a cloud attenuation model (using a Monte Carlo ray tracing algorithm). Finally, the model, combined with the temperature attenuation coefficient of the PV modules (-0.35% / °C), outputs a 15-minute power forecast curve. Wind power forecasts utilize the WRF (Weather Research and Forecasting) model for downscaling, interpolating 2km gridded meteorological data to turbine hub height. This data is then combined with each turbine's power curve (cut-off wind speed of 3m / s, rated wind speed of 10m / s) to generate probabilistic output ranges. Forecasts are dynamically refined using a Kalman filter, keeping the mean absolute error (MAE) for four-hour forward forecasts below 5.8%.

[0100] For each time step t, the output of the forward LSTM is:

[0101]

[0102] Among them, i, f and o are the input gate, forget gate and output gate respectively, σ is the sigmoid function, and ⊙ is element-by-element multiplication.

[0103] The LSTM calculation of the back propagation is symmetric with the forward propagation, and we get And merged in both directions into:

[0104]

[0105] Where d is the number of hidden units in a single layer of LSTM.

[0106] The attention mechanism can weight parameters in two ways: time attention and feature attention. The time dimension weighting can be expressed as:

[0107]

[0108] in, is a learnable parameter, d a The attention dimension.

[0109] Multivariate feature weighting can be expressed as:

[0110] ui =q T σ(W f x i +U f h time +b f )

[0111]

[0112] in, is a learnable parameter and m is the input feature dimension.

[0113] The optimization scheduling module initiates three levels of collaborative optimization based on the prediction results. At the second-level real-time control layer, when a high-speed data acquisition unit (sampling rate 10kHz) deployed on the DC bus detects a 2MW drop in photovoltaic power, the supercapacitor energy storage system (composed of 120 3kV / 3000F modules) responds within 80ms, injecting 1.8MW of compensation power through a bidirectional DC / DC converter (peak efficiency 98.7%). It also adjusts the output current slope of the proton exchange membrane fuel cell (PEMFC) (limited to 5A / s) to avoid sudden changes in hydrogen pipeline pressure. At the 15-minute scheduling layer, the model comprehensively considers the lithium iron phosphate battery's state of charge (SOC) (maintaining the optimal range of 40-60%), cycle life loss (each 1% of capacity charge or discharge corresponds to 0.0003% capacity decay), and electricity price period to formulate an energy storage charging and discharging plan. It then uses a particle swarm optimization algorithm to solve the unit combination solution with the lowest operating cost. The annual planning layer, based on historical meteorological big data, uses Copula functions to analyze the complementary characteristics of wind and solar resources, dynamically adjusts the capacity ratio of the hydrogen energy storage system (configured wind power installed capacity × 1200 equivalent full-load hours in typical scenarios), and uses a digital twin model to simulate the fatigue crack growth rate of hydrogen storage tanks under a cyclic pressure of 35MPa (calculated based on the Paris formula) to formulate preventive maintenance strategies. This achieves the coordination of power generation, power supply, and energy storage at multiple time scales, and can quickly provide compensatory power in emergencies such as short-term photovoltaic power drops, maintaining system stability and improving power supply quality and reliability. It can also formulate energy storage charging and discharging plans and unit combination plans to achieve optimal resource allocation, help reduce operating costs, improve energy utilization efficiency, and enable the pre-establishment of preventive maintenance strategies to improve system safety and reliability.

[0114] The feedback module performs a rolling horizon optimization (Receding Horizon Optimization) every 15 minutes. The model predictive control (MPC) algorithm uses the maximum annual turnover of hydrogen energy storage determined in the annual plan (set to 45 times) as a constraint condition to dynamically adjust the short-term scheduling plan. When the actual wind and solar power output deviates from the predicted value by more than 10% for three consecutive periods, the adaptive correction mechanism is triggered: first, the local meteorological field model is updated by the Kriging interpolation method, and then the recursive least squares method with forgetting factor (FFRLS) is used to update the prediction model parameters online, and finally the corrected prediction sequence is fed back to the optimization scheduling module for re-solving. At the same time, the digital twin system continuously compares the virtual battery aging model with the actual SOH (health status) data. When the deviation between the two exceeds 2%, the Kalman filter calibration is automatically triggered to update the activation energy parameter in the Arrhenius aging equation, which helps to update the algorithm parameters and improve the prediction accuracy of the model.

[0115] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0116] The above is only an embodiment of the present invention. Common knowledge such as the known specific structures and characteristics in the scheme is not described in detail here. Ordinary technicians in the field are aware of all common technical knowledge in the technical field of the invention before the application date or priority date, can obtain all existing technologies in the field, and have the ability to apply conventional experimental means before that date. Ordinary technicians in the field can improve and implement this scheme in combination with their own abilities under the inspiration given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several variations and improvements can be made, which should also be regarded as the scope of protection of the present invention. These will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A photovoltaic-wind power-storage integrated power generation system based on intelligent optimization, characterized by: It includes power generation module, hybrid energy prediction module, optimization scheduling module, energy storage coordination module, communication module and feedback module. The power generation module is used to generate electricity in different ways and transmit it to the optimization scheduling module and energy storage coordination module; The hybrid energy prediction module is used to predict the load of the power generation module and adjust the parameters of the power generation module using a hybrid prediction model based on the meteorological data and equipment operation status data obtained from the communication module; An optimization scheduling module for real-time power generation compensation, short-term charge and discharge scheduling, and long-term energy storage capacity ratio planning through multi-time scale collaboration; The energy storage collaboration module uses digital twin technology to build a virtual mirror of the wind, solar, and energy storage system throughout its life cycle, and establishes a time-sharing and hierarchical energy storage architecture. Communication module, used to obtain meteorological data from external databases, obtain equipment operating status data from equipment sensors, and transmit signals between the above modules; The feedback module is used to perform rolling time domain optimization, taking the long-term planning results as the boundary conditions for short-term scheduling, and feeding back the decision results of the optimization scheduling module to resolve coupling conflicts on the time scale.

2. The photovoltaic-wind power-storage integrated power generation system based on intelligent optimization according to claim 1, characterized in that: The power generation module includes a solar wind power generation unit and a fuel cell unit. A solar-wind power generation unit is used to convert solar energy and wind kinetic energy into electrical energy using semiconductors and a transmission structure. The solar-wind power generation unit comprises a support column (1), a sleeve (3) is rotatably sleeved on the support column (1), a plurality of rotating rods (6) are vertically sleeved on the outer periphery of the sleeve (3), the ends of the rotating rods (6) away from the sleeve (3) are fixedly connected to an inner photovoltaic panel (8), one side of the inner photovoltaic panel (8) is fixedly connected to an outer photovoltaic panel (9), and the inner photovoltaic panel (8) and the outer photovoltaic panel (9) are perpendicular to each other, a transmission rod (2) is coaxially rotatably sleeved in the support column (1), the rotating rods (6) all pass through the sleeve (3) and are rotatably connected to the transmission rod (2), the outer periphery of the rotating rod (6) near one end of the transmission rod (2) is provided with rotating shaft teeth (7), the top end of the transmission rod (2) is fixedly connected to a rotating motor (4), the output shaft of the rotating motor (4) is vertically upward and coaxially fixedly connected to a rotating gear disc (5), and the rotating gear disc (5) and the rotating shaft teeth (7) are all meshed; Fuel cell unit that converts the chemical energy of hydrogen and oxygen directly into electrical energy.

3. The photovoltaic-wind power-storage integrated power generation system based on intelligent optimization according to claim 2, characterized in that: The hybrid energy prediction module includes a preprocessing unit, a multi-source data fusion unit and a prediction unit. A preprocessing unit for processing outliers and missing values in meteorological data using dynamic thresholding and co-interpolation methods; Multi-source data fusion unit, used to fuse meteorological data using the spatiotemporal alignment method to establish a spatiotemporal feature matrix including irradiance intensity, cloud thickness and wind direction turbulence coefficient; The prediction unit is used to use the trained LSTM-Attention-WOA hybrid prediction model to predict short-term and long-term meteorological changes, including changes in irradiance intensity, cloud thickness, and wind turbulence coefficient.

4. The photovoltaic-wind power-storage integrated power generation system based on intelligent optimization according to claim 3 is characterized in that: The LSTM-Attention-WOA hybrid prediction model consists of a three-layer architecture. The first layer is a bidirectional long short-term memory network (LSTM) used to extract forward and / or backward time series feature dependencies. The second layer is the attention mechanism used to dynamically assign meteorological factor weights; The third layer is the Whale Optimization Algorithm (WOA), which is used to minimize the validation set.

5. The photovoltaic-wind power-storage integrated power generation system based on intelligent optimization according to claim 4 is characterized in that: The optimization scheduling module includes real-time compensation module, short-term scheduling module and long-term planning module. A real-time compensation unit for controlling the power generation voltage and frequency using virtual synchronous machine technology to suppress instantaneous fluctuations in power generation parameters; The short-term dispatch unit uses deep reinforcement learning technology to schedule the charging and discharging of the energy storage collaborative module during the valley and peak periods of the power grid to meet daily load demand. The long-term planning unit is used to build a virtual model of the energy storage equipment of the energy storage synergy module using digital twin simulation technology, and adjust the wind-solar-storage capacity ratio based on the long-term meteorological changes predicted by the prediction unit to adapt to seasonal changes in sunlight and wind speed.

6. The photovoltaic-wind power-storage integrated power generation system based on intelligent optimization according to claim 5, characterized in that: The model construction objects of digital twin simulation technology include solar and wind power generation units, fuel cell units and energy storage equipment of energy storage collaborative modules.

7. The photovoltaic-wind power-storage integrated power generation system based on intelligent optimization according to claim 6, characterized in that: The energy storage collaborative module includes fast energy storage units, short-term energy storage units and long-term energy storage units. The fast energy storage unit includes a supercapacitor energy storage device for rapid response to real-time compensation of power generation parameter fluctuations; The short-term energy storage unit includes lithium iron phosphate battery energy storage equipment, which is used for short-term coordinated scheduling of charging and discharging of energy storage equipment; The long-term energy storage unit includes hydrogen energy storage equipment, which is used for long-term energy storage on a year-round time scale, and is used to coordinate seasonal adjustments to the ratio of wind and solar energy storage capacity.

8. The photovoltaic-wind power-storage integrated power generation system based on intelligent optimization according to claim 7, characterized in that: The hydrogen energy storage device includes an electrolytic hydrogen generator, a hydrogen compressor and a hydrogen storage tank. The hydrogen storage tank is connected to the fuel cell unit through a control valve.

9. The photovoltaic-wind power-storage integrated power generation system based on intelligent optimization according to claim 8, characterized in that: The communication module includes a data acquisition unit and a signal transmission unit. A data acquisition unit is used to obtain meteorological data from an external server and collect equipment operating status data of the above modules. The meteorological data includes radiation intensity, cloud thickness, wind direction and wind force parameters, temperature and humidity. The equipment operating status data includes the DC voltage / current, AC power, IGBT temperature and power of the solar wind power generation unit, the pitch angle, gearbox oil temperature, vibration spectrum and generator speed of the wind power unit, and the SOC / SOH, single cell voltage, temperature difference and number of charge and discharge cycles of the energy storage device in the energy storage coordination module; The signal transmission unit is used to transmit signals between the above modules.

10. The photovoltaic-wind power-storage integrated power generation system based on intelligent optimization according to claim 9, characterized in that: The feedback module includes a rolling optimization unit and a correction unit. A rolling optimization unit, used to continuously optimize energy storage scheduling and planning using online rolling optimization; The correction unit is used to compensate for prediction errors using closed-loop correction to adapt to the uncertainties of wind and solar fluctuations, load changes and equipment aging.

Images