Wind-solar-energy-storage cooperative power system suitable for Sagomean area with deficient wind resources

Abstract

The invention discloses a wind and light storage cooperative power system suitable for a Sagomean area with deficient wind resources. The wind and light storage cooperative power system comprises a photovoltaic power generation unit, a floating type high-altitude wind power generation unit, a large-scale electrochemical energy storage unit and a central control module. The photovoltaic power generation unit serves as a basic power source, the wind power generation unit captures high-altitude stable wind energy through the floating technology to achieve complementary power supply, and the energy storage unit solves the problems of wind and light output fluctuation and load demand imbalance. And the central control module realizes efficient utilization of energy and stable operation of the system through a wind power prediction module, a multi-target collaborative scheduling module and a system operation optimization module in combination with an equipment health state evaluation and preventive maintenance strategy. The system breaks through the limitation of ground wind resources, improves the absorption capability of renewable energy sources, adapts to extreme environments, and guarantees the reliable supply of power in the Saggob area.

Description

Technical Field

[0001] This application belongs to the field of power systems, and specifically relates to a wind-solar-storage co-generation power system suitable for desert areas with scarce wind resources. Background Technology

[0002] my country's vast desert and Gobi regions possess abundant solar energy resources, making them ideal locations for large-scale photovoltaic (PV) bases. However, these areas typically lack ground-based wind power, and traditional wind power technologies are economically unviable. Simultaneously, the power grid infrastructure is weak, and load levels are low. The inherent intermittency and volatility of PV power generation result in a very poor match with local loads, leading to large-scale curtailment of solar power and posing a severe challenge to the stable operation of the power system and reliable power supply.

[0003] Currently, the absorption of new energy in desert and Gobi regions mainly relies on large-scale energy storage or the construction of long-distance transmission lines, which are costly. Limited by resource conditions, traditional ground-mounted wind power is difficult to deploy effectively in such areas to achieve complementarity. Therefore, there is an urgent need for a new power generation method that can overcome the limitations of ground-mounted wind resources and form efficient synergy with photovoltaics, fundamentally optimizing the energy supply structure of desert and Gobi regions.

[0004] Aerial wind power generation technology utilizes tethered aerostat platforms to reach altitudes of hundreds to thousands of meters above the ground, directly harnessing the stable and robust wind energy in that airspace to generate electricity. This technology is not constrained by the scarcity of ground-based wind resources, and its power output characteristics (especially at night and in winter) naturally complement photovoltaics. Combining aerial wind power generation technology with photovoltaics and energy storage to construct a "wind-solar-storage" synergistic system is an innovative path to solving the energy development dilemma in desert and Gobi regions. The key to achieving efficient system synergy lies in accurately predicting high-altitude wind power output, which is a core prerequisite for ensuring the reliability of grid dispatch and the economic viability of the system. Summary of the Invention

[0005] This application provides a wind-solar-storage integrated power system suitable for desert areas with scarce wind energy resources. By introducing floating high-altitude wind power generation technology, it breaks through the limitations of surface wind energy and forms a strong complement to photovoltaic power generation. With the help of accurate high-altitude wind power prediction and multi-objective coordinated scheduling, it realizes the optimized allocation and efficient utilization of multiple energy sources within the system, and ultimately significantly improves the stability, reliability and power supply guarantee capability of the power system in desert areas.

[0006] To achieve the above objectives, this application provides a wind-solar-storage integrated power system suitable for desert areas with scarce wind resources, including a photovoltaic power generation unit, a wind power generation unit, an energy storage unit, and a central control module;

[0007] The photovoltaic power generation unit consists of a large-scale photovoltaic array and supporting inverter equipment, which serves as the basic power source and is connected to the energy storage unit. The wind power generation unit adopts a floating high-altitude wind turbine, which serves as a complementary power source and is connected to the energy storage unit.

[0008] The central control module is used to collect the output data of the photovoltaic power generation unit and the wind power generation unit, as well as the state of charge of the energy storage unit. By using real-time ambient wind speed and light intensity, it dynamically adjusts the working status of the photovoltaic power generation unit and the wind power generation unit, and predicts and optimizes the charging and discharging strategy of the energy storage unit.

[0009] In one embodiment, the output power model of the photovoltaic power generation unit is as follows:

[0010] ; ①

[0011] In formula ①, Let t represent the output power of the photovoltaic array. Indicates the conversion efficiency of photovoltaic modules; Indicates the total area of ​​the photovoltaic array; Represents the solar irradiance at time t; Indicates the power temperature coefficient; This represents the temperature of the photovoltaic panel at time t; Indicates the reference temperature.

[0012] In one embodiment, the output power model of the wind power generation unit is as follows:

[0013] ; ②

[0014] In formula ②, This represents the power generation at time t at height h; This represents the air density at height h; Indicates the wind energy utilization coefficient; Indicates the swept area of ​​the wind turbine; This represents the wind speed at time t at height h; Indicates the cut-in, rated, and cut-out wind speeds; Indicates the rated power.

[0015] In one embodiment, the core of the energy storage unit adopts a large-scale electrochemical energy storage system to achieve spatiotemporal energy transfer and solve the instantaneous imbalance between wind and solar power output fluctuations and load demand. Its dynamic model is as follows:

[0016] ; ③

[0017] In formula ③, Indicates the state of charge of the stored energy at time t; Indicates charging efficiency and discharging efficiency; Indicates charging power and discharging power; Δt represents the rated capacity of the energy storage; Δt represents the time step.

[0018] In one embodiment, the central control module incorporates a wind power prediction module, a multi-objective collaborative scheduling module, and a system operation optimization module, forming an intelligent control hub that integrates perception, decision-making, optimization, and execution.

[0019] In one embodiment, the core fusion prediction formula of the wind power prediction module is as follows:

[0020] ; ④

[0021] In formula ④, This represents the fusion prediction result at time t+Δt; Represents the time-varying weights of the i-th model; This represents the i-th prediction model; This represents the input feature vector; Indicates model parameters;

[0022] The weights in the fusion prediction formula adopt an adaptive adjustment mechanism based on exponential weighting, as shown below:

[0023] ⑤

[0024] In formula ⑤, It is represented as the mean squared error of the i-th model in the most recent time window; It is represented as an adjustment coefficient.

[0025] In one embodiment, the multi-objective collaborative scheduling module is used to address the dynamic imbalance between the output fluctuations and load demands of the wind power generation unit and the photovoltaic power generation unit, and to construct a multi-scale collaborative optimization scheduling strategy of day-ahead scheduling layer - intraday rolling optimization layer - real-time.

[0026] The current scheduling layer uses a 24-hour optimization period, and its objective function is as follows:

[0027] ⑥

[0028] In formula ⑥, This represents the unit price of electricity purchased from the main grid at time t. This represents the power purchased from the main grid at time t; Indicates the penalty coefficient for abandoning wind and solar power; This represents the amount of photovoltaic power that is discarded at time t; This represents the amount of high-altitude wind power that is being wasted at time t; This represents the cost of energy storage equipment losses due to charging and discharging at time t.

[0029] Formula ⑥ achieves a comprehensive optimization of electricity purchase cost, wind and solar curtailment penalties, and energy storage losses by satisfying power balance constraints, energy storage operation constraints, and grid interaction constraints; the power balance constraints are as follows:

[0030] ; ⑦

[0031] In formula ⑦, This represents the actual power output of the photovoltaic system at time t; This represents the actual power output of the high-altitude wind power at time t; This represents the output power (kW) of the stored energy at time t; when Time indicates discharge. This indicates that the device is charging. This represents the total load power at time t;

[0032] The energy storage constraints are as follows:

[0033] ; ⑧

[0034] In formula ⑧, This represents the state of charge of the energy storage unit at time t, and represents the ratio of the remaining energy to the rated capacity. This indicates the minimum state of charge that the energy storage unit is allowed to operate, and is usually set to avoid over-discharge. This indicates the highest state of charge that the energy storage unit is allowed to operate, and is usually set to avoid overcharging.

[0035] The power grid interaction constraints are as follows:

[0036] ; ⑨

[0037] In formula ⑨, This represents the minimum power that can be purchased from the grid at time t; This represents the maximum power that can be purchased from the grid at time t.

[0038] The intraday rolling optimization layer employs a model predictive control method, with a 4-hour optimization window, and the target position is:

[0039] ; ⑩

[0040] In formula 10, k represents the time index within the rolling optimization window, starting from the current time t; N represents the prediction time domain; These represent the weighting coefficients for balancing the curtailment penalty, the energy storage state deviation penalty, and the grid power fluctuation penalty, respectively. This represents the total power curtailment at the predicted time k; The reference state of charge for energy storage is usually set to an ideal value (such as 50%) to allow sufficient charge and discharge space for the energy storage. This represents the change in grid interaction power at the predicted time k.

[0041] In one embodiment, the system operation optimization module dynamically scores operational performance based on a comprehensive status evaluation system with multi-dimensional indicators, and triggers adaptive switching of operating modes accordingly; the operational status indicators of the comprehensive status evaluation system are as follows:

[0042] ; ⑪

[0043] In formula ⑪, Indicates the overall status index; δ represents the weighting coefficient; This represents the predicted value of short-term high-altitude wind power output at time t. This indicates the rated installed capacity of the high-altitude wind power unit; This represents the predicted short-term photovoltaic output at time t. This indicates the rated installed capacity of the photovoltaic unit; This indicates a load fluctuation index.

[0044] In one embodiment, based on comprehensive state index Establish a three-level operation switching mechanism:

[0045] when At that time, we will enter an efficiency-first mode, focusing on improving the absorption capacity of renewable energy;

[0046] when When this happens, the system enters a safe and stable mode, prioritizing the reliability and stability of system operation.

[0047] when When this happens, the system enters grid support mode, prioritizing the fulfillment of grid ancillary service needs.

[0048] In one embodiment, the method further includes establishing a device health status assessment model by constructing a device failure rate model, as shown below:

[0049] ; ⑫

[0050] In formula ⑫, This indicates the health status of device i at time t; Indicates the initial health status; A function representing the failure rate related to operating conditions;

[0051] Based on the declining trend of equipment health status, a preventative maintenance strategy based on remaining useful life prediction is adopted, as shown below:

[0052] ; ⑬

[0053] In formula ⑬, τ represents the remaining lifetime of device i at time t; τ represents the future time increment starting from the current time t. This represents the predicted health status value of device i at a future time t+τ; This indicates the threshold for health status failure.

[0054] Compared with the prior art, the beneficial effects of this application are:

[0055] 1. By efficiently utilizing the stable high-altitude wind energy in the desert region through floating power generation technology, the problem of scarce ground wind resources in the region has been completely solved. Furthermore, it forms a natural strong complementarity with photovoltaic power generation in terms of timing, which greatly optimizes the energy supply structure.

[0056] 2. Relying on accurate high-altitude wind power prediction technology and combined with multi-timescale collaborative scheduling strategies, the renewable energy absorption capacity has been significantly improved, while reducing system operating costs and dependence on traditional power grid support.

[0057] 3. With the help of real-time intelligent mode switching and multi-objective optimization mechanism, the system can adapt to the extreme environment and complex operating conditions in the desert region, ensuring the reliability of power supply and system resilience.

[0058] 4. The system's embedded data-driven analysis and strategy adaptive architecture endow it with continuous self-learning and self-optimization capabilities, effectively coping with equipment aging and changes in the external environment, and maximizing its value throughout the entire lifecycle. Attached Figure Description

[0059] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1 This application provides a schematic diagram of a wind-solar-storage co-generation power system applicable to desert areas with scarce wind resources;

[0061] Figure 2 A schematic diagram of the workflow of a high-altitude wind power prediction module for a wind-solar-storage co-generation power system applicable to wind-scarce desert areas where wind resources are scarce, provided for this application;

[0062] Figure 3 A schematic diagram of a multi-timescale collaborative optimization scheduling framework for a wind-solar-storage integrated power system applicable to desert and Gobi areas with scarce wind resources, provided in this application;

[0063] Figure 4 A schematic diagram of the system operation status assessment and mode switching logic flow of a wind-solar-storage co-generation power system applicable to desert areas with scarce wind resources, provided for this application;

[0064] Figure 5 This application provides a schematic diagram of the implementation process for a full life-cycle health management strategy for wind-solar-storage integrated power systems applicable to desert and barren areas with scarce wind resources. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions in the embodiments of this application are described clearly and completely below. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are also within the scope of protection of this application.

[0066] See Figures 1 to 5 As shown, this application provides a wind-solar-storage integrated power system suitable for desert areas with scarce wind resources, including a photovoltaic power generation unit, a wind power generation unit, an energy storage unit, and a central control module;

[0067] The photovoltaic power generation unit consists of a large-scale photovoltaic array and supporting inverter equipment, which serves as the basic power source and is connected to the energy storage unit. The wind power generation unit adopts a floating high-altitude wind turbine, which serves as a complementary power source and is connected to the energy storage unit.

[0068] The central control module is used to collect the output data of the photovoltaic power generation unit and the wind power generation unit, as well as the state of charge of the energy storage unit. By using real-time ambient wind speed and light intensity, it dynamically adjusts the working status of the photovoltaic power generation unit and the wind power generation unit, and predicts and optimizes the charging and discharging strategy of the energy storage unit.

[0069] Optionally, the photovoltaic power generation unit consists of a large-scale photovoltaic array, which is the main power output unit of the system under sufficient sunlight conditions during the day. Its power generation capacity is affected by solar irradiance, ambient temperature, and component degradation characteristics. Utilizing the abundant solar resources of desert and Gobi regions, the photovoltaic array can achieve high-efficiency power generation, and its output power model is shown below:

[0070] ; ①

[0071] In formula ①, Let t represent the output power of the photovoltaic array. Indicates the conversion efficiency of photovoltaic modules; Indicates the total area of ​​the photovoltaic array; Represents the solar irradiance at time t; Indicates the power temperature coefficient; This represents the temperature of the photovoltaic panel at time t; Indicates the reference temperature.

[0072] Optionally, the wind power generation unit employs a floating high-altitude wind turbine as a complementary power source for the system. It utilizes a helium-filled aerostat to carry the wind power generation device, suspending it at an altitude of 300 to 3000 meters to capture stable wind energy resources. Compared to ground-based wind power generation, high-altitude wind speeds are higher and fluctuate less, significantly improving power generation efficiency and output stability. Its output power model is shown below:

[0073] ; ②

[0074] In formula ②, This represents the power generation at time t at height h; This represents the air density at height h; Indicates the wind energy utilization coefficient; Indicates the swept area of ​​the wind turbine; This represents the wind speed at time t at height h; Indicates the cut-in, rated, and cut-out wind speeds; Indicates the rated power.

[0075] Optionally, the core of the energy storage unit adopts a large-scale electrochemical energy storage system to achieve spatiotemporal energy transfer and solve the instantaneous imbalance between wind and solar power output fluctuations and load demand. Its dynamic model is as follows:

[0076] ; ③

[0077] In formula ③, Indicates the state of charge of the stored energy at time t; Indicates charging efficiency and discharging efficiency; Indicates charging power and discharging power; Δt represents the rated capacity of the energy storage; Δt represents the time step.

[0078] Optionally, the central control module incorporates a wind power prediction module, a multi-objective collaborative scheduling module, and a system operation optimization module, forming an intelligent control hub integrating perception, decision-making, optimization, and execution. Addressing the uncertainty and volatility of upper-level wind energy in the desert region, the system employs an intelligent prediction method based on multi-model fusion, combining historical data with real-time meteorological information to improve the accuracy of short-term and ultra-short-term power predictions.

[0079] Optionally, the wind power prediction module integrates a multi-model fusion architecture based on Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and Temporal Convolutional Network (TCN) to construct a prediction system with strong generalization capabilities, effectively addressing the nonlinearity and random fluctuations of high-altitude wind speeds. Its core fusion prediction formula is shown below:

[0080] ; ④

[0081] In formula ④, This represents the fusion prediction result at time t+Δt; Represents the time-varying weights of the i-th model; This represents the i-th prediction model; This represents the input feature vector; Indicates model parameters.

[0082] Simultaneously, by collecting real-time multi-dimensional feature data such as upper-level wind speed, wind direction, and temperature, and combining this with historical operational data, a dynamically updated input feature vector is established. The time-varying weights of each prediction model... An adaptive adjustment mechanism based on exponential weighting is adopted, and its formula is shown below:

[0083] ⑤

[0084] In formula ⑤, It is represented as the mean squared error of the i-th model in the most recent time window; This is represented as the adjustment coefficient. This mechanism dynamically allocates weights based on the model's prediction accuracy within the sliding time window, ensuring the prediction system can adapt to prediction needs under different meteorological conditions, significantly improving prediction accuracy and stability under complex meteorological conditions.

[0085] Optionally, the multi-objective collaborative scheduling module is used to address the dynamic imbalance between the output fluctuations and load demands of the wind power generation unit and the photovoltaic power generation unit, and to construct a multi-scale collaborative optimization scheduling strategy of day-ahead scheduling layer - intraday rolling optimization layer - real-time.

[0086] The current scheduling layer uses a 24-hour optimization period, and its objective function is as follows:

[0087] ⑥

[0088] In formula ⑥, This represents the unit price of electricity purchased from the main grid at time t. This represents the power purchased from the main grid at time t; Indicates the penalty coefficient for abandoning wind and solar power; This represents the amount of photovoltaic power that is discarded at time t; This represents the amount of high-altitude wind power that is being wasted at time t; This represents the cost of energy storage equipment losses due to charging and discharging at time t.

[0089] Formula ⑥ achieves a comprehensive optimization of electricity purchase cost, wind and solar curtailment penalties, and energy storage losses by satisfying power balance constraints, energy storage operation constraints, and grid interaction constraints; the power balance constraints are as follows:

[0090] ; ⑦

[0091] In formula ⑦, This represents the actual power output of the photovoltaic system at time t; This represents the actual power output of the high-altitude wind power at time t; This represents the output power (kW) of the stored energy at time t; when Time indicates discharge. This indicates that the device is charging. This represents the total load power at time t;

[0092] The energy storage constraints are as follows:

[0093] ; ⑧

[0094] In formula ⑧, This represents the state of charge of the energy storage unit at time t, and represents the ratio of the remaining energy to the rated capacity. This indicates the minimum state of charge that the energy storage unit is allowed to operate, and is usually set to avoid over-discharge. This indicates the highest state of charge that the energy storage unit is allowed to operate at, and is typically set to avoid overcharging.

[0095] The power grid interaction constraints are as follows:

[0096] ; ⑨

[0097] In formula ⑨, This represents the minimum power that can be purchased from the grid at time t; This represents the maximum power that can be purchased from the grid at time t.

[0098] Under the constraints of formulas ⑦, ⑧, and ⑨, the day-ahead scheduling scheme is optimized to coordinate the matching relationship between wind and solar power output forecasts and load demand, improve system economy and operational reliability, and formulate the economically optimal scheduling plan. The intraday rolling optimization layer adopts a model predictive control method with a 4-hour optimization window, and the optimization target is:

[0099] ; ⑩

[0100] In formula 10, k represents the time index within the rolling optimization window, starting from the current time t; N represents the prediction time domain; These represent the weighting coefficients for balancing the curtailment penalty, the energy storage state deviation penalty, and the grid power fluctuation penalty, respectively. This represents the total power curtailment at the predicted time k; The reference state of charge for energy storage is usually set to an ideal value (such as 50%) to allow sufficient charge and discharge space for the energy storage. This represents the change in grid interaction power at the predicted time k. The optimization objective focuses on reducing wind and solar curtailment rates, energy storage state of charge deviation, and grid interaction power fluctuations, achieving dynamic correction of day-ahead plans. Furthermore, it aims to achieve precise power control with second-level response, ensuring the system maintains real-time power balance under varying operating conditions.

[0101] Optionally, the system operation optimization module is based on a comprehensive state evaluation system with multi-dimensional indicators, and key parameters such as the real-time monitoring of the energy storage system's state of charge, wind and solar power generation forecasts, and load fluctuation characteristics.

[0102] The system dynamically scores operational performance and triggers adaptive switching of operating modes accordingly. The operational status indicators of the comprehensive status assessment system are shown below:

[0103] ; ⑪

[0104] In formula ⑪, Indicates the overall status index; δ represents the weighting coefficient; This represents the predicted value of short-term high-altitude wind power output at time t. This indicates the rated installed capacity of the high-altitude wind power unit; This represents the predicted short-term photovoltaic output at time t. This indicates the rated installed capacity of the photovoltaic unit; This indicates a load fluctuation index. The indicators comprehensively consider multiple factors such as the adequacy of system energy reserves, the sufficiency of power generation capacity, and the stability of load demand. By weighted summation, the overall operating status of the system is quantified, enabling an accurate depiction of the operating status under complex conditions.

[0105] This indicator dynamically reflects the degree of supply-demand matching and the reserve of adjustment resources in the system, providing a decision-making basis for multi-mode adaptive switching. Furthermore, based on this comprehensive status indicator... Establish a three-level operation switching mechanism:

[0106] when At that time, we will enter an efficiency-first mode, focusing on improving the absorption capacity of renewable energy;

[0107] when When this happens, the system enters a safe and stable mode, prioritizing the reliability and stability of system operation.

[0108] when When this happens, the system enters grid support mode, prioritizing the fulfillment of grid ancillary service needs.

[0109] Through the aforementioned three-level operation switching mechanism, the system can dynamically adjust its operating strategy based on real-time comprehensive status indicators, achieving seamless mode switching under different operating conditions. When the comprehensive status indicators are below the threshold, it automatically enters the efficiency-first mode to maximize the utilization of wind and solar energy resources and reduce power curtailment; when there are moderate fluctuations, it activates the safety and stability mode to strengthen energy storage scheduling and power balance control and prevent operational risks; when there are disturbances in the power grid or changes in dispatch instructions, it quickly switches to the grid support mode to provide auxiliary services such as frequency regulation and voltage regulation, enhancing the system's support capability for the power grid.

[0110] Optionally, it also includes establishing a device health status assessment model by constructing a device failure rate model, as shown below:

[0111] ; ⑫

[0112] In formula ⑫, This indicates the health status of device i at time t; Indicates the initial health status; This represents the failure rate function related to operating conditions.

[0113] This model comprehensively considers factors such as operating environment, workload, and historical maintenance records to assess the health status of critical equipment in real time.

[0114] Based on the declining trend of equipment health status, a preventative maintenance strategy based on remaining useful life prediction is adopted, as shown below:

[0115] ; ⑬

[0116] In formula ⑬, τ represents the remaining lifetime of device i at time t; τ represents the future time increment starting from the current time t. This represents the predicted health status value of device i at a future time t+τ; This indicates the threshold for health status failure.

[0117] This strategy improves the accuracy of maintenance decisions by dynamically revising remaining service life predictions based on real-time tracking of equipment health degradation and combining historical operating data with environmental stress factors. Simultaneously, it formulates optimal maintenance plans before significant equipment performance degradation. Through an established equipment health status threshold early warning mechanism, combined with remaining service life predictions, it optimizes maintenance resource allocation, reduces the risk of sudden failures, and extends equipment lifespan, thereby improving the system's overall lifecycle reliability and economic efficiency. This complete intelligent control strategy system, through the coordinated operation of its modules, ensures the safe and stable operation of the system in the complex environment of the desert region, significantly improving power supply reliability and overall lifecycle efficiency.

[0118] It should be noted that in practical applications, this equipment also possesses online self-learning capabilities. Based on continuously accumulated operational data, it can automatically optimize the health status assessment model and remaining service life prediction algorithm, automatically optimize control parameters and maintenance strategy thresholds, and adapt to equipment aging trends and environmental changes. This closed-loop optimization mechanism ensures that the system maintains optimal operating conditions throughout its entire lifecycle, while dynamically adjusting the operating mode in conjunction with real-time weather forecasts and power grid dispatch instructions.

[0119] To adapt to the extreme climatic conditions of the "Shagolan" region and ensure the long-term reliable operation of equipment under harsh environments such as high temperature, high humidity, and wind and sand erosion, the following key measures are implemented:

[0120] 1. **High-Strength Sealing and Positive Pressure Protection System:** The energy storage equipment enclosure employs multi-layer high-efficiency sealing technology, with special sealing treatment applied to key connection points. During system operation, the air supply system intelligently maintains a slightly positive pressure environment inside the enclosure, effectively preventing the disorderly intrusion of external sand and dust. Both the inlet and outlet are equipped with high-efficiency filters (such as HEPA grade) with self-cleaning functions. Based on real-time monitoring of the pressure difference data before and after the filters, the system accurately determines the blockage status, automatically prompts for replacement cycles, or triggers the reverse pulse jet cleaning function to ensure continuous and stable ventilation efficiency.

[0121] 2. **Intelligent Environmental Sensing and Operation Mode Switching:** Dust concentration and humidity sensors are deployed at key locations within the power plant and inside the enclosure. When a dust storm or excessively high ambient humidity (posing a risk of condensation) is detected, the control system automatically executes preset safety strategies: reducing or completely shutting off the fresh air volume, switching to internal circulation mode, and prioritizing the cleanliness and insulation safety of the internal environment of core equipment such as battery clusters and PCS. In this mode, the pursuit of maximum power output performance can be temporarily postponed.

[0122] 3. **Multi-parameter Coupled Thermal Management Model:** A thermodynamic model is constructed based on the coupling of multiple parameters, including real-time ambient temperature, solar radiation intensity, battery internal core temperature, and current charge / discharge rate. This model can dynamically predict the short-term heat load change trend of the battery cluster, enabling forward-looking and precise control of the cooling / heating system and effectively avoiding drastic temperature fluctuations.

[0123] 4. **Graded Temperature Control and System Energy Consumption Optimization:** A gradient temperature control strategy based on environmental conditions is implemented. In mild spring and autumn weather or at night, the system prioritizes low-energy-consumption natural cooling with fresh air. During hot summer months or when high-power charging and discharging cause a surge in heat generation, the system automatically activates forced cooling. Based on thermal model predictions, the system dynamically optimizes the cooling temperature setpoint, minimizing the thermal management system's own energy consumption and improving overall system energy efficiency while ensuring the battery operates within its optimal temperature range (e.g., 15℃-35℃).

[0124] 5. **Low-Temperature Self-Heating and Insulation Management:** For extremely cold winter environments, the battery pack integrates a distributed heating film or liquid thermal circulation system. Before a charging command is issued, a preheating program is intelligently activated based on the ambient temperature, rapidly raising the cell temperature above the high-efficiency charging threshold. Simultaneously, the energy storage container is wrapped with high-performance insulation materials, significantly reducing heat loss within the compartment and lowering heating energy consumption.

[0125] **Lifetime Awareness Control: Adaptive Charge / Discharge Strategy**

[0126] To optimize the balance between short-term performance and long-term lifespan of energy storage systems, an adaptive charging and discharging strategy based on health status awareness is implemented:

[0127] 1. **Online SOH Estimation and Remaining Life Prediction:**

[0128] Based on massive amounts of data from real-time monitoring of battery operation, including internal resistance changes, capacity decay curves, cumulative cycle counts, and historical operating temperature profiles, an online high-precision estimation of the battery pack's state of health (SOH) is achieved by integrating electrochemical models and data-driven algorithms. Furthermore, machine learning models are used to predict the remaining useful life (RUL) of the energy storage system under different future operating conditions (such as average depth of discharge and charge / discharge rate), providing forward-looking data support for optimized scheduling.

[0129] 2. **Dynamically Adjustable SOC Operating Window and Intelligent Power Allocation**:

[0130] **Dynamic SOC Operation Window Management:** This eliminates fixed upper and lower limits for SOC (State of Charge) operation (e.g., 20%-90%). The control system dynamically adjusts the optimized SOC operating range based on the current aggregated SOH level of the energy storage system, future wind and solar power forecast curves, and grid dispatch requirements. For example, during periods of extremely abundant wind and solar resources requiring full absorption, the upper limit of SOC can be temporarily relaxed to 98% under certain conditions; during periods of stable system performance and low load demand, the operating window is tightened to 40%-85%, reducing stress on the battery at high / low SOC levels and effectively delaying lifespan degradation.

[0131] **Health-Oriented Cluster Power Allocation:** This approach revolutionizes traditional average power allocation strategies in energy storage power stations composed of multiple parallel battery clusters. The central controller implements unbalanced power scheduling based on real-time assessments of health indicators such as state of harmonics (SOH) and internal resistance differences among the battery clusters. Battery clusters with better health are assigned more frequent fluctuation adjustments and high-power throughput tasks; those with poor health are given a "protective" usage strategy, allocating more stable and gentle power tasks, thereby extending the overall lifespan and maximizing the value of the power station cluster.

[0132] **Real-time optimized control in conjunction with wind and solar forecasting and dispatching commands**

[0133] Ensure that energy storage systems act as grid "stabilizers" and "regulators," achieving seamless integration with front-end forecasting and dispatching strategies:

[0134] 1. **Smoothing and Tracking Control Based on Ultra-Short-Term Forecasting**: Real-time reception of minute- to hour-level high-precision forecast data from "Accurate High-Altitude Wind Power Forecasting" and photovoltaic ultra-short-term forecasting systems. Calculation of combined wind and solar power output fluctuation components, generating corresponding energy storage system charging and discharging power commands, smoothing the total power output fluctuations of new energy power plants, and meeting grid connection technical requirements (such as volatility and ramp rate limitations). Rapid and accurate tracking of Automatic Generation Control (AGC) or Automatic Voltage Control (AVC) commands issued by the grid dispatch center, utilizing the millisecond-level response characteristics of the energy storage system to effectively compensate for the inertia of wind and solar power generation, enhancing grid frequency and voltage stability.

[0135] 2. **Multi-Timescale Power Allocation and Operating Mode Switching:** Receives planning instructions (such as day-ahead energy plans and intraday frequency regulation capacity) issued by the "multi-timescale collaborative optimization scheduling" strategy. Within the energy storage system, macro-level planning instructions are decomposed into control objectives at different time scales: second-level / minute-level (highest priority) response to primary frequency regulation and instantaneous fluctuation smoothing; hourly-level objectives to achieve energy time shifting and peak-valley electricity price arbitrage. Based on the comprehensive instructions issued by the "system operation status assessment and mode switching strategy," the energy storage system can seamlessly switch operating modes: pursuing economic efficiency in "normal mode"; prioritizing the reliability of power supply to critical loads in "extreme weather supply guarantee mode"; and performing supportive operations stipulated by grid mandatory standards in "fault ride-through mode."

[0136] **Clustered Intelligent Management and Fault Early Warning**

[0137] Improve the availability, reliability, and maintainability of energy storage power stations at the system level:

[0138] 1. **Early Fault Warning Based on Big Data Analytics:** Continuously collect operational data such as voltage, temperature, and insulation resistance of each battery module to build a battery health big data platform. Utilize machine learning algorithms (such as isolated forest and cluster analysis) to analyze the data stream in real time, identifying early abnormal characteristics (such as sudden divergence in voltage consistency, abnormal expansion of temperature differences between modules, and gradual changes in internal resistance), enabling early warning of serious safety hazards such as thermal runaway and internal short circuits, providing a critical time window for predictive maintenance.

[0139] 2. **Modular Redundancy and Plug-and-Play Design:** The system employs a standardized, modular PCS (converter) and battery cluster design. When a single module fails, the control system can immediately isolate the fault online without affecting the normal operation of other modules. The system supports plug-and-play replacement of battery modules and PCS modules, greatly simplifying the operation and maintenance process in remote, barren areas, improving efficiency, reducing total lifecycle maintenance costs, and forming an effective closed loop with the "full lifecycle performance management" strategy.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A wind-solar-storage co-generation power system suitable for desert and Gobi areas with scarce wind resources, characterized in that: It includes photovoltaic power generation units, wind power generation units, energy storage units, and a central control module; The photovoltaic power generation unit consists of a large-scale photovoltaic array and supporting inverter equipment, which serves as the basic power source and is connected to the energy storage unit. The wind power generation unit adopts a floating high-altitude wind turbine, which serves as a complementary power source and is connected to the energy storage unit. The central control module is used to collect the output data of the photovoltaic power generation unit and the wind power generation unit, as well as the state of charge of the energy storage unit. By using real-time ambient wind speed and light intensity, it dynamically adjusts the working status of the photovoltaic power generation unit and the wind power generation unit, and predicts and optimizes the charging and discharging strategy of the energy storage unit.

2. The wind-solar-storage co-generation power system suitable for wind-scarce desert areas as described in claim 1, characterized in that: The output power model of the photovoltaic power generation unit is shown below: ； ① In formula ①, Let t represent the output power of the photovoltaic array. Indicates the conversion efficiency of photovoltaic modules; Indicates the total area of ​​the photovoltaic array; Represents the solar irradiance at time t; Indicates the power temperature coefficient; This represents the temperature of the photovoltaic panel at time t; Indicates the reference temperature.

3. A wind-solar-storage co-generation power system suitable for desert and Gobi areas with scarce wind resources, as described in claim 1, is characterized in that: The output power model of the wind power generation unit is shown below: ； ② In formula ②, This represents the power generation at time t at height h; This represents the air density at height h; Indicates the wind energy utilization coefficient; Indicates the swept area of ​​the wind turbine; This represents the wind speed at time t at height h; Indicates the cut-in, rated, and cut-out wind speeds; Indicates the rated power.

4. A wind-solar-storage co-generation power system suitable for desert and Gobi areas with scarce wind resources, as described in claim 1, is characterized in that: The core of the energy storage unit adopts a large-scale electrochemical energy storage system to achieve spatiotemporal energy transfer and solve the instantaneous imbalance between the fluctuation of wind and solar power output and load demand. Its dynamic model is as follows: ； ③ In formula ③, Indicates the state of charge of the stored energy at time t; Indicates charging efficiency and discharging efficiency; Indicates charging power and discharging power; Δt represents the rated capacity of the energy storage; Δt represents the time step.

5. A wind-solar-storage co-generation power system suitable for desert and Gobi areas with scarce wind resources, as described in claim 1, is characterized in that: The central control module incorporates a wind power prediction module, a multi-objective collaborative scheduling module, and a system operation optimization module, forming an intelligent control hub that integrates perception, decision-making, optimization, and execution.

6. A wind-solar-storage co-generation power system suitable for wind-scarce desert areas as described in claim 5, characterized in that: The core fusion prediction formula of the wind power prediction module is shown below: ; ④ In formula ④, This represents the fusion prediction result at time t+Δt; Represents the time-varying weights of the i-th model; This represents the i-th prediction model; This represents the input feature vector; Indicates model parameters; The weights in the fusion prediction formula adopt an adaptive adjustment mechanism based on exponential weighting, as shown below: ； ⑤ In formula ⑤, It is represented as the mean squared error of the i-th model in the most recent time window; It is represented as an adjustment coefficient.

7. A wind-solar-storage co-generation power system suitable for desert and Gobi areas with scarce wind resources, as described in claim 5, is characterized in that: The multi-objective collaborative scheduling module is used to address the dynamic imbalance between the output fluctuations and load demands of the wind power generation unit and the photovoltaic power generation unit, and to construct a multi-scale collaborative optimization scheduling strategy of day-ahead scheduling layer - intraday rolling optimization layer - real-time. The current scheduling layer uses a 24-hour optimization period, and its objective function is as follows: ；⑥ In formula ⑥, This represents the unit price of electricity purchased from the main grid at time t. This represents the power purchased from the main grid at time t; Indicates the penalty coefficient for abandoning wind and solar power; This represents the amount of photovoltaic power that is discarded at time t; This represents the amount of high-altitude wind power that is being wasted at time t; This represents the energy storage device loss cost caused by charging and discharging at time t. Formula ⑥ achieves a comprehensive optimization of electricity purchase cost, wind and solar curtailment penalties, and energy storage losses by satisfying power balance constraints, energy storage operation constraints, and grid interaction constraints; the power balance constraints are as follows: ； ⑦ In formula ⑦, This represents the actual power output of the photovoltaic system at time t; This represents the actual power output of the high-altitude wind power at time t; This represents the output power (kW) of the stored energy at time t; when Time indicates discharge. This indicates that the device is charging. This represents the total load power at time t; The energy storage constraints are as follows: ； ⑧ In formula ⑧, This represents the state of charge of the energy storage unit at time t, and represents the ratio of the remaining energy to the rated capacity. This indicates the minimum state of charge that the energy storage unit is allowed to operate, and is usually set to avoid over-discharge. This indicates the highest state of charge that the energy storage unit is allowed to operate, and is usually set to avoid overcharging. The power grid interaction constraints are as follows: ； ⑨ In formula ⑨, This represents the minimum power that can be purchased from the grid at time t; This represents the maximum power that can be purchased from the grid at time t; The intraday rolling optimization layer employs a model predictive control method, with a 4-hour optimization window, and the target position is: ； ⑩ In formula 10, k represents the time index within the rolling optimization window, starting from the current time t; N represents the prediction time domain; These represent the weighting coefficients for balancing the curtailment penalty, the energy storage state deviation penalty, and the grid power fluctuation penalty, respectively. This represents the total power curtailment at the predicted time k; The reference state of charge for energy storage is usually set to an ideal value to allow sufficient charge and discharge space for the energy storage. This represents the change in grid interaction power at the predicted time k.

8. A wind-solar-storage co-generation power system suitable for wind-scarce desert areas as described in claim 5, characterized in that: The system operation optimization module, based on a comprehensive state assessment system with multi-dimensional indicators, monitors key parameters such as the energy storage system's state of charge, predicted wind and solar power generation, and load fluctuation characteristics in real time. It dynamically scores operational efficiency and triggers adaptive switching of operating modes accordingly. The operational state indicators of the comprehensive state assessment system are shown below: ； ⑪ In formula ⑪, Indicates the overall status index; δ represents the weighting coefficient; This represents the predicted value of short-term high-altitude wind power output at time t. This indicates the rated installed capacity of the high-altitude wind power unit; This represents the predicted short-term photovoltaic output at time t. This indicates the rated installed capacity of the photovoltaic unit; This indicates a load fluctuation index.

9. A wind-solar-storage co-generation power system suitable for wind-scarce desert areas as described in claim 8, characterized in that: Based on comprehensive status indicators Establish a three-level operation switching mechanism: when At that time, we will enter an efficiency-first mode, focusing on improving the absorption capacity of renewable energy; when When this happens, the system enters a safe and stable mode, prioritizing the reliability and stability of system operation. when When this happens, the system enters grid support mode, prioritizing the fulfillment of grid ancillary service needs.

10. A wind-solar-storage co-generation power system suitable for desert and Gobi areas with scarce wind resources, as described in any one of claims 1-9, characterized in that: This also includes establishing a device health status assessment model by constructing a device failure rate model, as shown below: ； ⑫ In formula ⑫, This indicates the health status of device i at time t; Indicates the initial health status; A function representing the failure rate related to operating conditions; Based on the declining trend of equipment health status, a preventative maintenance strategy based on remaining useful life prediction is adopted, as shown below: ； ⑬ In formula ⑬, τ represents the remaining lifetime of device i at time t; τ represents the future time increment starting from the current time t. This represents the predicted health status value of device i at a future time t+τ; This indicates the threshold for health status failure.

Images