Flywheel energy storage and storage battery mixed UPS (Uninterrupted Power Supply) control method

Abstract

The invention relates to the technical field of flywheel energy storage, and discloses a flywheel energy storage and storage battery hybrid UPS control method, which comprises the steps of planning a hybrid energy storage control process based on load characteristics and energy storage equipment state parameters, and determining a dynamic adjustment algorithm and a cooperative control strategy; determining a working mode according to the real-time load demand and the energy storage equipment state parameter; generating a hybrid energy storage control strategy set and determining an execution sequence; generating a parameter configuration list and a strategy scheme list of each mode; and performing correlation matching to generate an overall control scheme. The invention further relates to the contents of dynamic control boundary determination, specific strategies in different working modes, control parameter updating, switching rule presetting, hybrid energy storage control database establishment and the like. According to the method, efficient cooperation of the flywheel and the storage battery is achieved, the adaptability of the UPS system to a load scene and the power supply reliability are improved, the working state of energy storage equipment is optimized, the service life of the storage battery is prolonged, and long-term stable operation of the system is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of flywheel energy storage, and in particular to a UPS control method combining flywheel energy storage with a battery. Background Art

[0002] In modern society, with the rapid development of information technology and the increasing demand for power supply reliability in critical equipment, the performance and reliability of uninterruptible power supply (UPS) systems, as core equipment for ensuring the continuity and stability of power supply, have attracted considerable attention. Traditional UPS systems primarily rely on batteries for energy storage, but batteries have numerous limitations. For one thing, batteries have relatively low energy density, making it difficult to respond quickly and provide sufficient energy support when faced with high-power, short-duration load shocks. This can cause power outages or voltage fluctuations, impacting the normal operation of equipment. Furthermore, batteries have a limited cycle life, and frequent charge and discharge processes accelerate performance degradation, increasing maintenance costs and replacement frequency. Furthermore, battery performance is susceptible to factors such as ambient temperature and humidity, significantly reducing reliability under extreme environmental conditions.

[0003] With the continuous development of energy storage technology, flywheel energy storage technology has gradually come into people's attention. Flywheel energy storage has the advantages of high power density, fast response speed, and long cycle life. It can quickly respond to transient changes in load and performs well in high-power scenarios. However, flywheel energy storage also has its own shortcomings. For example, its relatively low energy density makes it difficult to meet the demand for long-term continuous power supply. Therefore, combining flywheel energy storage with batteries to build a hybrid energy storage UPS system has become an important approach to address the limitations of traditional UPS systems.

[0004] Currently, a mature and comprehensive control system for hybrid energy storage UPS systems has yet to be established. Existing control strategies present numerous challenges in managing energy distribution and mode switching between different operating modes, as well as coping with complex load characteristics and environmental changes. For example, when loads fluctuate significantly, efficient coordination between the flywheel and battery is impossible, resulting in low energy efficiency. During mode switching, switching delays or instability are prone to occur, impacting power supply continuity. Furthermore, insufficient consideration of environmental factors and equipment health status makes it difficult to achieve optimal control and protection of energy storage equipment.

[0005] Furthermore, with the development of smart grids and distributed energy, UPS systems are required to collaborate with a wider range of energy devices and systems, placing higher demands on the control methods of hybrid energy storage UPS systems. Dynamically adjusting control strategies based on real-time load demand, energy storage device status, and environmental conditions to achieve optimal synergy between flywheels and batteries, thereby improving the overall performance and reliability of UPS systems, remains a pressing technical challenge. Therefore, developing an efficient and reliable UPS control method that combines flywheel energy storage with batteries has significant theoretical and practical application value. Summary of the Invention

[0006] The object of the present invention is to provide a UPS control method that combines flywheel energy storage and batteries to solve the problems raised in the above background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a UPS control method for a hybrid flywheel energy storage and battery, the method comprising: Based on the load characteristics of the UPS system, the state parameters of the energy storage equipment, and the energy storage performance constraints, the hybrid energy storage control process is planned to determine the required dynamic adjustment algorithm and coordinated control strategy. The dynamic adjustment algorithm includes the charge and discharge state switching logic, and the coordinated control strategy includes the power distribution rules between the flywheel and the battery. Determine the operating mode of the hybrid energy storage system according to the real-time load demand of the UPS system and the state parameters of the energy storage device; Based on the working mode, dynamic adjustment algorithm and coordinated control strategy, a hybrid energy storage control strategy set is generated, and the execution order of the control strategy set is determined according to a priority sorting principle; Generate a parameter configuration list for each mode in each mode according to a preset control cycle, the working mode and the energy storage performance constraint condition, and generate a strategy solution list for each mode according to a preset switching rule, the working mode and the energy storage performance constraint condition; Based on the execution order of the control strategy set, the parameter configuration list, the strategy scheme list and the hybrid energy storage control content are associated and matched to generate an overall control scheme for the UPS system, and real-time control data and working mode information are synchronously recorded.

[0008] Preferably, the hybrid energy storage control process is planned based on the load characteristics of the UPS system, the state parameters of the energy storage device, and the energy storage performance constraints, including: For the current operating mode, determine the dynamic control boundary and preset response time window of the mode based on the load fluctuation history data of the mode, the energy storage response data of the adjacent modes, and the preset mode division threshold; extracting a subset of real-time control data of the mode according to the dynamic control boundary and the preset response time window; The mapping relationship between the dynamic adjustment algorithm and the coordination strategy related parameters in the real-time control data subset is determined as the core content of the control logic of the mode.

[0009] Preferably, determining the operating mode of the hybrid energy storage system according to the real-time load demand of the UPS system and the state parameters of the energy storage device includes: When it is detected that the load demand power exceeds the preset power threshold, it is determined to be in high-load main mode and the flywheel energy storage priority response mechanism is activated; When it is detected that the remaining capacity of the battery is lower than a preset capacity threshold, the flywheel speed margin and the power compensation capability are classified based on a first dynamic threshold algorithm, data of each compensation level is obtained and its response characteristic vector is calculated; Determine the power supply reliability level of the energy storage system based on the real-time grid status; When the matching degree between the response characteristic vector and the preset compensation level exceeds a set threshold, it is determined to be a mixed compensation mode; Otherwise, it is determined to be in normal power supply mode.

[0010] Preferably, the hybrid energy storage control content also includes environmental monitoring data and equipment health data, the working mode includes a main power supply mode, a backup power supply mode and a transition switching mode, and generating a hybrid energy storage control strategy set includes: The first classification algorithm is used to extract the working condition characteristics of the current environmental monitoring data to obtain the data of each environmental working condition category; For the current environmental condition category data, feature screening is performed based on the influence weight and change rate of each environmental parameter on energy storage efficiency to generate a set of key influencing factors; The environmental operating condition label is determined according to the correlation between the distribution characteristics of the key influencing factor set and the energy storage response index.

[0011] Preferably, after determining the working mode, the method further includes: When in the main power supply mode, the environmental condition label is matched with the flywheel dynamic response curve to generate a flywheel energy storage optimization strategy; When in backup power supply mode, a battery protection strategy is generated based on the battery charge and discharge characteristic curve and life attenuation model; When in transition switching mode, a seamless switching strategy is generated based on historical switching delay data and system stability requirements.

[0012] Preferably, after generating the hybrid energy storage control strategy set, the method further includes: Update the control parameters of each working mode based on the real-time collected load fluctuation data; Recalculate the adaptive working mode of the energy storage device according to the updated control parameters; The execution priority of the control strategy set is dynamically optimized based on the recalculated operating mode.

[0013] Preferably, the preset switching rules include: Dynamically adjust the sampling frequency of the control strategy based on the confidence level of the energy storage device's historical operating data; When the flywheel speed deviation or battery voltage abnormality exceeding the set tolerance is detected, the backup control module is automatically triggered.

[0014] Preferably, the preset switching rules further include: A control signal calibration mechanism is established to perform error compensation calibration on the control loop based on the test values of the reference voltage source and standard load before each mode switching.

[0015] Preferably, after generating the overall control scheme, the process further includes: Establishing a hybrid energy storage control database to store the parameter configuration list, strategy solution list, real-time control data and working mode records; Regularly verify the validity of historical data in the database and remove redundant data; Based on the time series analysis algorithm, the control effects of the energy storage system are compared longitudinally to evaluate the long-term stability of the control strategy.

[0016] Preferably, the first classification algorithm is a K-means clustering algorithm, and the specific steps of classifying the environmental monitoring data include: Calculate the Euclidean distance between environmental data samples and initialize the cluster center; Divide the samples into the nearest neighbor cluster centers through iterative updates; The final environmental condition classification result is determined based on the preset cluster number or silhouette coefficient threshold.

[0017] Compared with the prior art, the present invention has the following beneficial effects: By planning the hybrid energy storage control process based on the UPS system's load characteristics, energy storage device status parameters, and energy storage performance constraints, and determining a dynamic adjustment algorithm and collaborative control strategy, the system achieves efficient coordinated operation of the flywheel and battery. The system flexibly adjusts the operating mode based on real-time load demand and energy storage device status, improving the hybrid energy storage system's adaptability to different load scenarios. When the load demand power exceeds the preset power threshold, the system is identified as high-load primary mode and the flywheel energy storage priority response mechanism is activated. This fully leverages the flywheel energy storage's high power density and fast response speed, enabling rapid response to high-power load shocks and ensuring the stability and continuity of power supply.

[0018] The operating mode determination process takes into account multiple factors, including the remaining battery capacity, flywheel speed margin, power compensation capability, and grid status. Scientific judgment rules and algorithms ensure the accuracy and rationality of the operating mode. For example, when the remaining battery capacity falls below a preset capacity threshold, flywheel-related parameters are classified and calculated based on a first dynamic threshold algorithm. This allows the flywheel's energy storage capacity to be rationally utilized when battery performance degrades, compensating for battery deficiencies and improving system power supply reliability.

[0019] By classifying and filtering environmental monitoring data, we generate a collection of key influencing factors and environmental condition labels. We then implement corresponding optimization strategies based on different operating modes, such as flywheel energy storage optimization in primary power mode, battery protection in backup power mode, and seamless switching in transition mode. This fully considers the impact of environmental factors and device characteristics on system performance. This not only optimizes the operating state of energy storage equipment and improves energy storage efficiency, but also effectively extends battery life and reduces maintenance costs.

[0020] After generating the hybrid energy storage control strategy set, by updating the control parameters based on real-time collected load fluctuation data, recalculating the adaptive working mode of the energy storage equipment, and dynamically optimizing the execution priority of the control strategy set, the system can adapt to load changes in real time and maintain the optimal working state, further improving the flexibility and reliability of the system.

[0021] The preset switching rules include dynamically adjusting the sampling frequency of the control strategy based on the confidence level of the energy storage device's historical operating data, automatically triggering the backup control module, and the control signal calibration mechanism. This ensures the accuracy and stability of the control strategy, reduces delays and errors during mode switching, and improves the safety and reliability of the system.

[0022] Establishing a hybrid energy storage control database to store, verify and analyze relevant data can provide data support for system optimization and improvement. Through the analysis and evaluation of historical data, problems in the control strategy can be discovered in a timely manner, and the control strategy can be continuously optimized to improve the long-term stability of the energy storage system control effect and ensure the continuous and reliable operation of the UPS system. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a working principle diagram of the UPS control method for hybrid flywheel energy storage and battery according to the present invention; Figure 2 A flow chart for planning the hybrid energy storage control process based on load characteristics and energy storage device state parameters; Figure 3 Flowcharts generated for energy storage control strategies in different operating modes; Figure 4 Flowchart of the control signal calibration mechanism in the preset switching rule. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0025] See also Figure 1-Figure 4 The present invention relates to a UPS control method that combines flywheel energy storage and batteries. The specific implementation steps are as follows: Based on the UPS system's load characteristics, the energy storage device's state parameters, and the energy storage performance constraints, the hybrid energy storage control process is planned, determining the required dynamic adjustment algorithm and coordinated control strategy. The dynamic adjustment algorithm includes the charge-discharge state switching logic, and the coordinated control strategy includes the power distribution rules between the flywheel and the battery. During the planning process, the current UPS system load, such as load size and fluctuation frequency, as well as the energy storage device's state parameters, such as flywheel speed, remaining battery capacity, and voltage, must be comprehensively considered to develop a control algorithm and strategy suitable for the current system operation.

[0026] The hybrid energy storage system's operating mode is determined based on the UPS system's real-time load requirements and the energy storage device's status parameters. The system monitors the UPS system's load power and the energy storage device's status in real time. For example, when the load power changes, the system's operating mode is determined to ensure the system can stably and efficiently power the load.

[0027] Based on the determined operating mode, dynamic adjustment algorithm, and coordinated control strategy, a hybrid energy storage control strategy set is generated, and the execution order of this control strategy set is determined according to the priority sorting principle. Different operating modes require different control strategies. It is necessary to integrate various control strategies into a strategy set and arrange them in a certain priority order so that the corresponding strategy can be quickly called according to the actual situation during system operation.

[0028] Based on the preset control cycle, operating mode, and energy storage performance constraints, a parameter configuration list for each mode is generated. Furthermore, based on the preset switching rules, operating mode, and energy storage performance constraints, a strategy list for each mode is generated. The preset control cycle can be a time interval pre-set based on the system's operational requirements. Within each cycle, the corresponding parameter configuration and strategy are determined based on the current operating mode and energy storage performance constraints.

[0029] Based on the execution order of the control strategy set, the parameter configuration list, strategy solution list, and hybrid energy storage control content are correlated and matched to generate the overall control solution for the UPS system. By organically combining the contents of each part, a complete control solution is formed that can guide the operation of the UPS system.

[0030] The present invention will be further described below in conjunction with Examples 1 to 5: Example 1: When planning the hybrid energy storage control process, it is necessary to determine the dynamic control boundaries, extract a subset of real-time control data, and clarify the core content of the control logic for the current operating mode. Specifically, it is necessary to first collect historical load fluctuation data for the current operating mode. This data covers information such as the magnitude, amplitude, and frequency of load power changes during different time periods under this mode. For example, in a specific operating mode, the load power may show periodic fluctuations throughout the day, such as high and frequent load power fluctuations during daytime hours and relatively stable load power at night. At the same time, energy storage performance constraints must be clearly defined, including the maximum safe flywheel speed (e.g., 20,000 rpm), the maximum battery charge / discharge depth (discharge depth ≤ 80%), and the lower limit of the system's total energy storage capacity (no less than 20% of the design capacity). These serve as fundamental constraint parameters for planning the hybrid energy storage control process. Energy storage response data for adjacent modes is also required. These might include high-load prime mode versus conventional power supply mode, or conventional power supply mode versus backup power supply mode. This data includes the flywheel's speed adjustment rate and the battery's charge / discharge current variations. For example, when switching from conventional power supply mode to high-load prime mode, the flywheel must increase its speed within a short period of time to provide additional power. Data such as the speed increase rate and the time it takes to reach a steady state constitute the energy storage response data for the adjacent mode. The preset mode division thresholds are pre-set based on the UPS system's design parameters and actual operating experience to define the range of different operating modes. For example, when the load power exceeds a certain threshold, it is classified as high-load prime mode, while when it is below the threshold, it is classified as conventional power supply mode.

[0031] Based on the above-mentioned load fluctuation historical data, the energy storage response data of adjacent modes and the preset mode division threshold, the dynamic control boundary of the mode is determined through data analysis and processing methods. The dynamic control boundary is not fixed, but is dynamically adjusted as the system operating status and data change. For example, when analyzing the load fluctuation historical data, statistical methods can be used to calculate the mean, variance and other statistical quantities of the load power to determine the normal fluctuation range of the load power. Combined with the power variation range that the flywheel and battery can withstand in the energy storage response data of the adjacent mode, the dynamic control boundary of the current mode is finally determined. This boundary clarifies the reasonable variation range of the load power and energy storage device status parameters in the current mode. When this range is exceeded, the system may need to switch to the adjacent mode or take other control measures.

[0032] The preset response time window is set based on the load type—for example, 50ms for medical equipment loads and 100ms for industrial motor loads. This ensures that the real-time control data subset covers the critical response period to load changes. After determining the dynamic control boundary, the real-time control data subset for that mode is extracted based on the preset response time window. The preset response time window is a time range set to ensure that the system can respond promptly to load changes and energy storage device status changes. For example, a 100-ms window means that the system must complete the corresponding control decisions and adjustments within 100 milliseconds of detecting a load or energy storage device status change. The real-time control data subset is extracted as follows: UPS system data such as load power, flywheel speed, remaining battery capacity, voltage, and current are collected in real time. The dynamic control boundary is then used to filter data within this boundary. Combined with the preset response time window, only data within 100 milliseconds prior to the current time point is retained to form the real-time control data subset. This approach avoids processing excessive amounts of irrelevant or outdated data, improving data processing efficiency and control accuracy.

[0033] The mapping relationship between the dynamic adjustment algorithm and the coordinated strategy parameters within the real-time control data subset is the core of this mode's control logic. Dynamic adjustment algorithm parameters include the trigger conditions and switching time for charge / discharge state switching, while coordinated strategy parameters include the power allocation ratio and priority between the flywheel and battery. Within the real-time control data subset, the relationship between changes in load power, flywheel speed adjustments, and battery charge / discharge currents is analyzed. For example, when the load power suddenly increases, the flywheel speed may decrease, and the battery may need to increase discharge current to compensate for the power shortfall. By establishing a mapping relationship between these parameters, it is possible to clearly define how the dynamic adjustment algorithm and coordinated strategy should be executed under different real-time control data conditions. For example, if the real-time control data subset shows a 10% increase in load power and a 5% decrease in flywheel speed within a short period of time, the dynamic adjustment algorithm may trigger accelerated charging of the flywheel based on this mapping relationship. Simultaneously, the coordinated strategy will adjust the power allocation ratio between the flywheel and battery, increasing the flywheel's output power share to quickly respond to changes in load demand.

[0034] In practical applications, the above process needs to be handled separately for different operating modes. For example, in high-load primary mode, load fluctuations are typically more severe, so the determination of the dynamic control boundary must prioritize the protection of the flywheel and battery to prevent them from operating at extreme limits for extended periods. In conventional power supply mode, the load is relatively stable, and the dynamic control boundary can be set wider to allow for a certain degree of load fluctuation without frequent switching of operating modes. The frequency of extracting real-time control data subsets can also be adjusted according to the requirements of different operating modes. In high-load primary mode, the frequency of data collection and subset extraction can be increased to achieve faster response, such as shortening the preset response time window to 50 milliseconds. In conventional power supply mode, the frequency can be appropriately reduced to reduce the system's computational burden.

[0035] Furthermore, the impact of the energy storage device's state parameters on the core control logic must be considered. For example, when the battery's remaining capacity is low, even in conventional power supply mode, the coordinated strategy must adjust power distribution to reduce battery discharge and prioritize flywheel energy storage to prevent overdischarge, which can impact battery life and performance. In this case, the battery's remaining capacity parameter becomes a significant factor influencing the mapping within the real-time control data subset, and the dynamic adjustment algorithm may trigger premature charging of the battery to maintain its capacity within a reasonable range.

[0036] After determining the core content of the control logic, it needs to be applied to the actual control of the hybrid energy storage system. By monitoring data changes within the real-time control data subset in real time, the corresponding dynamic adjustment algorithm and collaborative strategy are called according to the mapping relationship to achieve coordinated control of the flywheel and battery. For example, when it is detected that the flywheel speed in the real-time control data subset is close to its minimum safe speed, the dynamic adjustment algorithm will trigger the flywheel's charging state switching logic, stopping its discharge and starting charging. At the same time, the collaborative strategy will adjust the battery's discharge power to compensate for the power gap caused by the cessation of flywheel discharge, ensuring that the power supply to the load is not affected.

[0037] The core of this embodiment lies in determining the dynamic control boundary by analyzing the historical load fluctuation data of the current operating mode, the energy storage response data of adjacent modes, and the preset mode division threshold. It then extracts a subset of real-time control data within a preset response time window, and establishes a mapping relationship between the dynamic adjustment algorithm and the relevant parameters of the coordination strategy within this subset, forming the core control logic content, thereby providing a foundation for efficient control of the hybrid energy storage system. This process fully considers the system's real-time operating status and historical data, enabling precise control of the hybrid energy storage system under different operating modes, ensuring stable and reliable operation of the UPS system under various load conditions.

[0038] Example 2: When determining the working mode of the hybrid energy storage system, when it is detected that the load demand power exceeds the preset power threshold, it is determined to be a high-load main mode and the flywheel energy storage priority response mechanism is activated. The setting of the preset power threshold needs to comprehensively consider the design capacity of the UPS system, the characteristics of the load, and the performance of the energy storage equipment. For example, for a UPS system with a design capacity of 1000kW, the preset power threshold may be set to 800kW depending on the type and importance of the load. When it is detected that the load demand power exceeds 800kW, the system determines to enter the high-load main mode. At this time, since the flywheel energy storage has the characteristics of fast response speed and can quickly provide high-power support, its response mechanism is activated first. After activating the flywheel energy storage priority response mechanism, the system will quickly adjust the operating state of the flywheel, so that it changes from the energy storage state to the discharge state to meet the sudden increase in power demand of the load.

[0039] When the remaining battery capacity is detected to be below a preset capacity threshold, the flywheel speed margin and power compensation capability are classified based on a first dynamic threshold algorithm. Data for each compensation level is obtained and its response characteristic vector is calculated. The preset capacity threshold is set to ensure safe operation and service life of the battery, for example, it may be set at 20% of the battery's rated capacity. When the remaining battery capacity falls below this threshold, it indicates that the battery's energy storage capacity has reached its lower limit and requires power compensation from other energy storage devices. The first dynamic threshold algorithm evaluates the flywheel's speed margin—the excess power capacity it can provide when increasing from its current speed to its maximum allowable speed—based on factors such as the flywheel's current speed, maximum allowable speed, and speed change rate. The algorithm also considers factors such as the flywheel's power compensation efficiency and response time to comprehensively assess its power compensation capability. Based on the evaluation results, the flywheel's power compensation capability is classified into different compensation levels, for example, high, medium, and low. For each compensation level, a response characteristic vector is calculated, which contains characteristic parameters such as the flywheel's response time, power output stability, and efficiency at that compensation level.

[0040] The energy storage system's power supply reliability level is determined based on real-time grid status. This includes factors such as voltage stability, frequency stability, and the presence of harmonic interference. By monitoring the grid voltage in real time and analyzing its fluctuation range and frequency, voltage stability is assessed. For example, under normal circumstances, the grid voltage should fluctuate within a range of 220V ±5%. If the fluctuation range exceeds this range, it indicates poor voltage stability. The Chinese standard for grid frequency is 50Hz. The deviation between the actual frequency and the standard frequency is monitored to assess frequency stability. Harmonic interference can affect power quality. By monitoring the harmonic content in the grid, the presence and extent of harmonic interference can be determined. Taking these factors into consideration, the energy storage system's power supply reliability level is classified into different levels, such as reliable, basically reliable, and unreliable.

[0041] When the match between the response eigenvector and the preset compensation level exceeds a set threshold, hybrid compensation mode is selected; otherwise, conventional power supply mode is selected. The threshold is determined based on system design requirements and actual operating experience, for example, 80%. The calculated response eigenvector is compared with the preset standard eigenvectors for each compensation level to calculate their similarity or match. If the match exceeds 80%, the flywheel is effectively providing power compensation, and hybrid compensation mode is selected. In hybrid compensation mode, the flywheel and battery work together to provide power support to the load. The flywheel is primarily responsible for quickly responding to sudden load changes, providing short-term high-power support, while the battery provides continuous power output to meet the load's stable requirements. If the match between the response eigenvector and the preset compensation level is less than 80%, conventional power supply mode is selected. In conventional power supply mode, the system operates as normal, with the energy storage device providing only small amounts of power regulation when necessary to maintain stable system operation.

[0042] In practical applications, the above operating mode determination process needs to be flexibly adjusted to accommodate different load characteristics and grid conditions. For example, for critical loads with extremely high power supply reliability requirements, such as life-support equipment in hospitals and servers in data centers, the threshold setting may be appropriately lowered when determining the operating mode, allowing hybrid compensation mode to be entered earlier to ensure load power supply security. For non-critical loads with relatively low power supply reliability requirements, such as general lighting equipment and air conditioners, the threshold setting can be appropriately increased to reduce unnecessary losses in energy storage equipment.

[0043] Furthermore, when the system is in high-load primary mode, the flywheel and battery status parameters must be closely monitored. For example, the flywheel's speed must not exceed its maximum allowable speed, otherwise it may damage the flywheel; the battery's charge and discharge currents must also be controlled within a safe range to avoid excessive charge and discharge that may affect its lifespan. After activating the flywheel energy storage priority response mechanism, the system monitors the flywheel's speed and power output in real time. When the flywheel speed approaches the maximum allowable speed, the system automatically adjusts its power output or switches to another operating mode to ensure safe operation of the flywheel.

[0044] Monitoring the remaining battery capacity is also crucial. In addition to determining the remaining capacity based on a preset capacity threshold, factors such as the battery's charge and discharge efficiency and self-discharge rate must also be considered. For example, in high-temperature environments, the battery's self-discharge rate increases, reducing the actual available capacity. In this case, the preset capacity threshold may need to be appropriately raised to ensure safe battery operation. Furthermore, during the battery's charge and discharge process, parameters such as temperature and voltage must be monitored to ensure they are within normal operating ranges.

[0045] When determining the power supply reliability level of an energy storage system, the grid's resilience must also be considered. If the grid is experiencing only temporary fluctuations and is expected to return to normal quickly, the energy storage system can adopt a more conservative power supply strategy to extend the service life of the energy storage equipment. Conversely, if the grid failure is more severe and recovery takes longer, the energy storage system may need to adopt a more proactive power supply strategy to ensure stable power supply to the load for a longer period of time.

[0046] In hybrid compensation mode, the coordinated operation of the flywheel and battery requires precise control. The system must dynamically adjust the power distribution ratio between the flywheel and battery based on the real-time load demand. For example, when the load power demand suddenly increases, the flywheel will rapidly increase its power output to meet the short-term load demand, while the battery will gradually increase its power output to share some of the load. Once the load power demand stabilizes, the system adjusts the power distribution between the flywheel and battery based on their status, allowing the flywheel to gradually restore energy storage while the battery continues to provide stable power output.

[0047] Even in conventional power supply mode, the system requires regular maintenance and management of energy storage equipment. For example, batteries must be regularly charged and discharged to maintain performance, and flywheel speed monitoring and balancing must be performed to ensure stable operation. Furthermore, the health of the energy storage equipment must be monitored to identify potential faults and take appropriate measures to address them.

[0048] Example 3: Hybrid energy storage control covers environmental monitoring data and equipment health data, and operating modes include primary power supply mode, backup power supply mode, and transition switching mode. When generating a hybrid energy storage control strategy set, the environmental monitoring data needs to be classified and processed to extract operating condition characteristics. The specific steps are as follows: First, real-time environmental monitoring data is collected. This data includes environmental parameters such as temperature, humidity, air pressure, dust concentration, and vibration frequency. For example, in a data center computer room scenario, the temperature may be maintained at 20-25°C and the humidity controlled at 40%-60%. In outdoor scenarios, these parameters may vary drastically with the season and weather. The collected data is transmitted in real time to the UPS system's control module via a sensor network, forming a set of environmental data samples.

[0049] Next, the first classification algorithm is used to extract operating condition features from the current environmental monitoring data, obtaining data for each environmental condition category. Taking the K-means clustering algorithm as an example, the first classification algorithm calculates the Euclidean distance between environmental data samples. The Euclidean distance measures the similarity between samples in a multidimensional space; closer distances indicate more similar sample characteristics. For example, the Euclidean distance between sample A (temperature 22°C, humidity 50%, air pressure 101 kPa) and sample B (temperature 23°C, humidity 48%, air pressure 100 kPa) is smaller, indicating similar environmental conditions. Cluster centers are then initialized. The number of cluster centers can be determined based on historical data experience or system presets. For example, if the environmental conditions are categorized as "low temperature and low humidity," "medium temperature and medium humidity," and "high temperature and high humidity," three cluster centers are initialized.

[0050] Samples are divided into nearest-neighbor cluster centers through iterative updates. In each iteration, each sample is assigned to the category of the cluster center with the smallest Euclidean distance to it. The cluster center positions are then recalculated based on the new categories until the cluster centers no longer change significantly or the preset number of iterations is reached. For example, after the first iteration, some samples are assigned to cluster center 1 (representing low-temperature and low-humidity conditions). As the iterations proceed, the position of cluster center 1 is continuously adjusted based on the average value of all samples in that category, ultimately forming a stable category division.

[0051] The final classification of environmental conditions is determined based on the preset number of clusters or the silhouette coefficient threshold. If the number of clusters is set to three, the algorithm will ultimately classify the environmental data samples into three categories. If a silhouette coefficient threshold (such as 0.5) is used, the clustering effect is considered satisfactory when the average silhouette coefficient of all samples reaches or exceeds this threshold. The iteration stops and the classification results are finalized. Each category corresponds to an environmental condition, such as Category 1 for "low temperature and dryness," Category 2 for "normal temperature and humidity," and Category 3 for "high temperature and high humidity."

[0052] For the current environmental operating condition category data, feature screening is performed based on the weight of each environmental parameter's impact on energy storage efficiency and the rate of change to generate a set of key influencing factors. First, the weight of each environmental parameter's impact on energy storage efficiency must be determined. This weight can be set through statistical analysis of historical data or expert experience. For example, the weight of temperature on flywheel energy storage efficiency may be 0.4, humidity 0.3, air pressure 0.2, and vibration frequency 0.1; while for battery energy storage efficiency, the weight of temperature impact may be higher, reaching 0.5, humidity 0.3, and other parameters 0.2. The rate of change reflects the magnitude of change of environmental parameters per unit time. For example, a temperature change of 2°C per hour is a low rate, and a temperature change of 5°C per hour is a high rate.

[0053] By calculating the product of the impact weight and the rate of change of each environmental parameter, parameters with higher product values are selected as key influencing factors. For example, if the temperature change rate in a certain environmental operating condition category is 3°C per hour, the impact weight is 0.4, and the product is 0.12; if the humidity change rate is 1% per hour, the impact weight is 0.3, and the product is 0.03, then the temperature parameter will be included in the set of key influencing factors. In addition, the correlation between parameters must be considered to avoid repeated screening. For example, if temperature and humidity may have a positive correlation, the parameter with a more significant impact on energy storage efficiency should be retained.

[0054] Environmental condition labels are determined based on the correlation between the distribution characteristics of a set of key influencing factors and energy storage response indicators. Energy storage response indicators include flywheel speed fluctuation amplitude, battery charge and discharge efficiency change rate, and equipment failure rate. For example, in an environmental condition where the key influencing factors are high temperature (temperature > 30°C) and high humidity (humidity > 70%), historical data statistics show that the flywheel speed fluctuation amplitude increases by 15% compared to normal temperature and humidity conditions, battery charge and discharge efficiency decreases by 8%, and the equipment failure rate increases by 5%. Therefore, the environmental condition label can be determined as "high temperature, high humidity, and high risk condition." For another example, when the key influencing factors are low temperature (temperature < 5°C) and low humidity (humidity < 30%), the flywheel energy storage response delay increases by 10%, but the battery self-discharge rate decreases by 3%. Therefore, the label can be determined as "low temperature, dryness, and low efficiency condition."

[0055] In primary power supply mode, backup power supply mode, and transition switching mode, the environmental condition label serves as an important basis for generating control strategies. For example, when the system is in primary power supply mode and the environmental condition label is "high temperature, high humidity, and high risk conditions," the flywheel's operating parameters must be adjusted in conjunction with the flywheel's dynamic response curve. The flywheel's dynamic response curve records data such as the flywheel's speed response time and power output stability under different temperature and humidity conditions. Based on the curve characteristics corresponding to the label, optimization strategies such as reducing the flywheel's rated power output and increasing the cooling system's operating frequency may be adopted to prevent flywheel performance degradation or failure due to high temperature environments.

[0056] When in backup power mode and the environmental operating condition is labeled "Low Temperature, Dryness, and Low Efficiency," a battery protection strategy is generated based on the battery charge-discharge characteristic curve and life decay model. The battery charge-discharge characteristic curve shows that in low-temperature environments, the battery's internal resistance increases, charge and discharge efficiency decreases, and excessive discharge may cause plate sulfation. The life decay model predicts the battery's remaining life based on factors such as temperature and charge and discharge depth. In this case, strategies such as increasing the charge voltage threshold and limiting the discharge depth (for example, limiting the discharge depth to less than 50%) may be implemented to minimize the impact of low temperatures on battery life.

[0057] In transitional switching mode, environmental condition tags influence the stability of the switching strategy. For example, under "high vibration conditions" (determined by vibration frequency as the key influencing factor), historical switching delay data may show that switching times are 20% longer than under normal conditions, and system stability indicators (such as voltage fluctuations) exceed allowable ranges. Therefore, when generating a seamless switching strategy, it is necessary to increase the buffer time in the pre-switching phase and strengthen the status monitoring of the flywheel and battery to ensure uninterrupted load power supply during mode switching in vibrating environments.

[0058] Furthermore, equipment health data (such as flywheel bearing wear and battery plate aging) must be analyzed in conjunction with environmental condition tags. For example, if the flywheel bearing wear rate is accelerated compared to normal operating conditions under "high dust conditions," this indicates that dust may be causing poor bearing lubrication. In this case, the control strategy should increase the maintenance frequency of the bearing lubrication system or automatically activate dust filtration devices when dust concentration exceeds a threshold.

[0059] Example 4: After determining the working mode, the system needs to generate targeted control strategies based on the characteristics and requirements of different modes, as follows: When in the main power supply mode, the environmental condition label needs to be matched with the flywheel dynamic response curve to generate a flywheel energy storage optimization strategy. The environmental condition label is a comprehensive description of the working condition obtained by analyzing environmental parameters such as temperature, humidity, and air pressure, such as "high temperature and high humidity conditions" and "low temperature and low pressure conditions". The flywheel dynamic response curve is obtained in advance through experiments and historical data accumulation, recording key parameters such as the speed change, power output efficiency, and energy loss of the flywheel under different environmental conditions. For example, in a high temperature environment, the mechanical loss of the flywheel will increase, and its speed drop rate may be faster than in a normal temperature environment. The corresponding dynamic response curve will show a trend of power output attenuation.

[0060] During the matching process, the system first obtains the current environmental condition label, such as "high temperature and high humidity conditions", and then retrieves the corresponding curve from the flywheel dynamic response curve database. According to the changes in flywheel performance reflected by the curve, the control parameters of the flywheel are adjusted. For example, under high temperature and high humidity conditions, to avoid failure of the flywheel due to overheating, the system may reduce the maximum allowable output power of the flywheel and extend its charging cycle to reduce the loss caused by long-term high-load operation. In addition, the flywheel's cooling system operation strategy may also be adjusted, such as increasing the speed of the cooling fan or increasing the flow of coolant to maintain the flywheel operating temperature within a safe range. Through this matching and parameter adjustment, the flywheel energy storage system is optimized to ensure that it can stably and efficiently meet the load requirements in the main power supply mode.

[0061] When in backup power mode, a battery protection strategy is generated based on the battery charge-discharge characteristic curve and life decay model. The battery charge-discharge characteristic curve describes how battery parameters such as voltage, current, and capacity change during the charge and discharge process. For example, during discharge, as capacity decreases, the battery terminal voltage gradually decreases. Once the voltage drops to a certain threshold, continued discharge will cause irreversible damage to the battery. The life decay model comprehensively considers the impact of factors such as charge and discharge depth, number of cycles, and operating temperature on battery life, and uses a mathematical model to predict the battery's remaining useful life.

[0062] In backup power mode, the system's primary goal is to ensure stable power supply to the load while maximizing the battery's service life. Therefore, based on the battery's charge and discharge characteristic curves, the system sets appropriate charge and discharge cutoff voltages and current limits. For example, when the battery voltage drops to the preset discharge cutoff voltage, the system automatically disconnects the discharge circuit to prevent over-discharge. Simultaneously, incorporating a lifespan degradation model, the system optimizes the charge and discharge strategy, such as adopting a "shallow charge and shallow discharge" strategy to control the charge and discharge depth within a certain range to minimize plate wear. Furthermore, the charging voltage is adjusted based on ambient temperature. For example, in low-temperature environments, the charging voltage is appropriately increased to compensate for the decreased charging efficiency caused by the increased internal resistance of the battery, while also preventing the battery from remaining in a long-term low-power state due to insufficient charging. These protection strategies effectively slow the battery's lifespan degradation, improving its reliability and service life.

[0063] When in transition switching mode, a seamless switching strategy is generated based on historical switching delay data and system stability requirements. Transition switching mode refers to the process by which a UPS system switches between different operating modes (such as primary and backup power modes, or backup and mains power modes). Historical switching delay data records the time required for each previous mode switch, including detection signal transmission time, control logic processing time, and switching device actuation time. System stability requirements include indicators such as the load voltage fluctuation range and switching interruption time during the switching process. For example, load voltage fluctuations during the switching process must not exceed ±10% of the rated voltage, and the interruption time must not exceed 10 milliseconds.

[0064] When generating a seamless handover strategy, the system first analyzes historical handover delay data to identify the main factors affecting handover time, such as lengthy processing steps in certain control logic and slow switching device actuation. To address these factors, the system optimizes the control process, for example by pre-determining handover conditions to reduce real-time computation and selecting high-speed switching devices to shorten actuation time. Furthermore, a buffering mechanism is designed during the handover process based on system stability requirements. For example, before handover, a backup energy storage device (such as a flywheel or battery) is pre-activated, entering hot standby mode to ensure immediate takeover in the event of a primary power outage. During the handover process, dynamic voltage adjustment technology monitors the load voltage in real time and adjusts the output power of the energy storage device to compensate for potential voltage fluctuations during the handover process. Furthermore, a contingency mechanism is implemented for handover failures. For example, if an anomaly is detected during the handover process, a backup switching path or emergency power supply solution is automatically triggered to ensure continuous load power supply.

[0065] In practice, control strategies for different operating modes are not independent but require coordination. For example, the flywheel energy storage optimization strategy generated in the primary power supply mode may affect the battery's performance in the backup power supply mode. When the flywheel reduces its output power under high-temperature conditions, the battery may need to bear more load power. In this case, the battery protection strategy in the backup power supply mode needs to be adjusted accordingly, such as relaxing the limit on the battery discharge current while shortening its continuous discharge time to prevent overuse.

[0066] Furthermore, changes in environmental conditions can also affect the effectiveness of control strategies in each mode. For example, in primary power supply mode, if the ambient conditions suddenly change from normal temperature and humidity to high temperature and humidity, the flywheel's dynamic response curve will change, and the corresponding optimization strategy will need to be adjusted in real time. At this point, the system will re-acquire the environmental condition label, update the flywheel's control parameters, and communicate this change to the control logic of the backup power supply mode and transition switching mode, ensuring that the strategies in each mode can coordinately adapt to environmental changes.

[0067] Equipment health data also plays a crucial role in developing control strategies. For example, if the battery plate aging threshold is detected, the system will implement stricter protective measures, such as further reducing the depth of discharge and increasing the charging frequency, even in backup power mode. Furthermore, if the flywheel bearing is severely worn during transitional switching, the switching strategy will increase the frequency of flywheel speed monitoring and extend the stable operation time after switching to prevent switching failures due to mechanical component failures.

[0068] Example 5: After generating the hybrid energy storage control strategy set, the system needs to dynamically optimize the control parameters, operating modes, and strategy priorities based on real-time data, and execute preset switching rules and database management operations. The specific implementation methods are as follows: 1. Control parameter update and working mode recalculation The control parameters of each working mode are updated based on the load fluctuation data collected in real time. Real-time load fluctuation data includes the instantaneous value of the load power, the rate of change (such as the power change per second), etc. For example, when it is detected that the load power suddenly increases from 500kW to 700kW within 1 second, the system determines that the load is in a state of violent fluctuation and needs to update the power output upper limit parameter of the flywheel in the high-load main mode from 600kW to 750kW to match the real-time demand. The logic of updating parameters is based on the historical statistical laws of load fluctuations. For example, the mean and variance of the load power in the past 5 minutes are calculated through a sliding window algorithm. If the current instantaneous value exceeds the mean ±2 times the variance range, the parameter update process is triggered.

[0069] The adaptive operating mode of the energy storage device is recalculated based on the updated control parameters. The adaptive operating mode of the energy storage device requires a comprehensive evaluation of multiple data dimensions, including load power, flywheel speed, and remaining battery capacity. For example, if the flywheel power output upper limit is increased after the update, and the current load power is 750kW (exceeding 93.75% of the preset power threshold of 800kW), but the remaining battery capacity is 25% (higher than the preset capacity threshold of 20%), the system will determine that the high-load primary mode is still currently adapted through logical judgment (e.g., maintaining the high-load primary mode when the load power ≥ the threshold and the battery capacity ≥ the threshold). If the remaining battery capacity simultaneously drops to 18%, then combined with the first dynamic threshold algorithm's evaluation of the flywheel speed margin, the hybrid compensation mode may be re-determined.

[0070] 2. Dynamic Optimization of Control Strategy Set Execution Priority Based on the recalculated operating mode, the execution priority of the control strategy set is dynamically adjusted. The control strategy set includes multiple sub-strategies, such as flywheel charging and discharging control, battery protection, and mode switching logic. Priority is determined by preset rules (such as "safety protection strategy > real-time response strategy > efficiency optimization strategy") and operating condition weights. For example, when the operating mode switches from conventional power supply mode to hybrid compensation mode, the priority of the "flywheel speed abnormality protection strategy" is increased from third to first because it is directly related to the safety of the energy storage equipment; the priority of the "battery balanced charging strategy" is reduced from second to fourth because the current operating conditions prioritize power output over long-term battery maintenance.

[0071] The mathematical expression of priority adjustment is: in: After adjustment The priority value of each sub-strategy (the smaller the value, the higher the priority); Before adjustment The initial priority value of each sub-strategy; Adjust the step size for the priority triggered by the working condition (for example, when switching working mode Indicates that the priority is increased by 2 levels); For the Working conditions strategy The weight coefficient of (the value range is 0.5~1.5, determined by the correlation between the strategy and the working conditions).

[0072] 3. Implementation of Preset Switching Rules ① Dynamic adjustment of sampling frequency The sampling frequency of the control strategy is dynamically adjusted based on the confidence level of the historical operating data of the energy storage device. The confidence level is evaluated by the standard deviation of the historical data, and the formula is: in: is the standard deviation of historical data; For the Historical data points (such as flywheel speed, battery voltage); is the mean of historical data; is the number of data points.

[0073] when When the data confidence level is less than the preset threshold (e.g., the standard deviation of the flywheel speed is ≤50r / min), the sampling frequency is reduced from 100Hz to 50Hz to reduce the computational load. When the threshold is exceeded, the sampling frequency is increased to 200 Hz to capture high-frequency fluctuations.

[0074] ② Backup control module trigger mechanism The backup control module is automatically triggered when a flywheel speed deviation (the difference between actual and target speed) or battery voltage anomaly (e.g., voltage below 85% or above 115% of rated value) exceeds a set tolerance. These tolerances are pre-set based on the equipment's safe operating range, for example, a flywheel speed deviation tolerance of ±200 rpm and a battery voltage tolerance of ±10% of rated voltage. The backup control module, comprised of hardware circuitry or software threads independent of the primary control logic, immediately takes over control upon triggering, executing operations such as emergency shutdown and bypass switching.

[0075] ③Control signal calibration mechanism Establish a control signal calibration mechanism to perform error compensation calibration on the control loop based on the test values of a reference voltage source (such as a standard 10V voltage source) and a standard load (such as a 100Ω precision resistor) before each mode switch. The calibration process includes: Input the reference voltage source signal to the control loop, measure the output voltage, and calculate the error value , Used to reflect the deviation of the control loop in the voltage output link, is the actual detected control loop output voltage value, It is the standard voltage input used for calibration; Connect a standard load, measure the loop current, and calculate the error value ( ), is the current error value, which is used to reflect the deviation of the control loop in the current output link. is the actual detected current value in the control loop, Theoretical current value calculated according to Ohm's law; according to and Generate compensation coefficients and make real-time corrections to the control signal.

[0076] 4. Hybrid Energy Storage Control Database Management After generating the overall control plan, a hybrid energy storage control database is established to store parameter configuration lists, strategy lists, real-time control data, and operating mode records. Real-time control data includes load power, real-time flywheel speed, and battery terminal voltage and current, collected every 10ms. Operating mode records include the mode switch time (accurate to the millisecond), switch trigger conditions (e.g., "load power exceeding 800kW triggers high-load primary mode"), and mode duration. Each operating mode record must be associated with a snapshot of real-time control data for the corresponding time period. The database uses a relational structure. For example, the "Parameter Configuration Table" contains fields such as the operating mode ID, parameter name, current value, and update time; the "Real-time Control Data Table" records data points such as load power, flywheel speed, and battery voltage every second.

[0077] Historical data in the database is regularly validated and redundant data is removed. Validation is implemented using logical rules, such as checking whether the load power is negative or whether the flywheel speed exceeds the maximum allowable value, to eliminate abnormal data points. Redundant data removal uses a time window mechanism, such as retaining the last three months of real-time data and deleting older historical records to free up storage space.

[0078] A time series analysis algorithm is used to longitudinally compare the control effects of energy storage systems and evaluate the long-term stability of control strategies. Time series analysis includes trend analysis (e.g., changes in average flywheel response time over time) and periodic analysis (e.g., daily / weekly fluctuations in load power). By comparing control parameter settings with system response data over different time periods, it identifies areas of strategy performance degradation or room for improvement, providing a basis for subsequent strategy iterations.

[0079] V. Synergistic Effects of Multiple Mechanisms Each of the aforementioned links does not operate independently, but rather forms a closed-loop control system through data interaction. For example, historical switching delay data stored in the database provides a basis for optimizing seamless switching strategies; adjustments to the real-time sampling frequency affect the accuracy of load fluctuation data collection, which in turn affects the control parameter update logic; and the results of control signal calibration directly affect the monitoring accuracy of flywheel speed deviation and battery voltage, thereby triggering or inhibiting the activation of the backup control module. Through this collaborative mechanism, the system can continuously optimize control performance in dynamically changing operating conditions, ensuring the reliability and efficiency of the hybrid energy storage system.

[0080] 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," "includes," 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.

[0081] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A UPS control method for a hybrid flywheel energy storage and battery, characterized in that: The method comprises: Based on the load characteristics of the UPS system, the state parameters of the energy storage equipment, and the energy storage performance constraints, the hybrid energy storage control process is planned to determine the required dynamic adjustment algorithm and coordinated control strategy. The dynamic adjustment algorithm includes the charge and discharge state switching logic, and the coordinated control strategy includes the power distribution rules between the flywheel and the battery. Determine the operating mode of the hybrid energy storage system according to the real-time load demand of the UPS system and the state parameters of the energy storage device; Based on the working mode, dynamic adjustment algorithm and coordinated control strategy, a hybrid energy storage control strategy set is generated, and the execution order of the control strategy set is determined according to a priority sorting principle; Generate a parameter configuration list for each mode in each mode according to a preset control cycle, the working mode and the energy storage performance constraint condition, and generate a strategy solution list for each mode according to a preset switching rule, the working mode and the energy storage performance constraint condition; Based on the execution order of the control strategy set, the parameter configuration list, the strategy scheme list and the hybrid energy storage control content are associated and matched to generate an overall control scheme for the UPS system, and real-time control data and working mode information are synchronously recorded.

2. The flywheel energy storage and battery hybrid UPS control method according to claim 1, characterized in that: Based on the load characteristics of the UPS system, the state parameters of the energy storage device, and the energy storage performance constraints, the hybrid energy storage control process is planned, including: For the current operating mode, determine the dynamic control boundary and preset response time window of the mode based on the load fluctuation history data of the mode, the energy storage response data of the adjacent modes, and the preset mode division threshold; extracting a subset of real-time control data of the mode according to the dynamic control boundary and the preset response time window; The mapping relationship between the dynamic adjustment algorithm and the coordination strategy related parameters in the real-time control data subset is determined as the core content of the control logic of the mode.

3. The UPS control method of flywheel energy storage and battery hybrid according to claim 1, characterized in that: Determining the operating mode of the hybrid energy storage system according to the real-time load demand of the UPS system and the state parameters of the energy storage device includes: When it is detected that the load demand power exceeds the preset power threshold, it is determined to be in high-load main mode and the flywheel energy storage priority response mechanism is activated; When it is detected that the remaining capacity of the battery is lower than a preset capacity threshold, the flywheel speed margin and the power compensation capability are classified based on a first dynamic threshold algorithm, data of each compensation level is obtained and its response characteristic vector is calculated; Determine the power supply reliability level of the energy storage system based on the real-time grid status; When the matching degree between the response characteristic vector and the preset compensation level exceeds a set threshold, it is determined to be a mixed compensation mode; Otherwise, it is determined to be in normal power supply mode.

4. The flywheel energy storage and battery hybrid UPS control method according to claim 3, characterized in that: The hybrid energy storage control content also includes environmental monitoring data and equipment health data. The operating mode includes a primary power supply mode, a backup power supply mode, and a transition switching mode. Generating a hybrid energy storage control strategy set includes: The first classification algorithm is used to extract the working condition characteristics of the current environmental monitoring data to obtain the data of each environmental working condition category; For the current environmental condition category data, feature screening is performed based on the influence weight and change rate of each environmental parameter on energy storage efficiency to generate a set of key influencing factors; The environmental operating condition label is determined according to the correlation between the distribution characteristics of the key influencing factor set and the energy storage response index.

5. The flywheel energy storage and battery hybrid UPS control method according to claim 4, characterized in that: After determining the working mode, the method further includes: When in the main power supply mode, the environmental condition label is matched with the flywheel dynamic response curve to generate a flywheel energy storage optimization strategy; When in backup power supply mode, a battery protection strategy is generated based on the battery charge and discharge characteristic curve and life attenuation model; When in transition switching mode, a seamless switching strategy is generated based on historical switching delay data and system stability requirements.

6. The flywheel energy storage and battery hybrid UPS control method according to claim 5, characterized in that: After generating the hybrid energy storage control strategy set, it also includes: Update the control parameters of each working mode based on the real-time collected load fluctuation data; Recalculate the adaptive working mode of the energy storage device according to the updated control parameters; The execution priority of the control strategy set is dynamically optimized based on the recalculated operating mode.

7. The flywheel energy storage and battery hybrid UPS control method according to claim 1, characterized in that: The preset switching rules include: Dynamically adjust the sampling frequency of the control strategy based on the confidence level of the energy storage device's historical operating data; When the flywheel speed deviation or battery voltage abnormality exceeding the set tolerance is detected, the backup control module is automatically triggered.

8. The flywheel energy storage and battery hybrid UPS control method according to claim 7, characterized in that: The preset switching rules also include: A control signal calibration mechanism is established to perform error compensation calibration on the control loop based on the test values of the reference voltage source and standard load before each mode switching.

9. The flywheel energy storage and battery hybrid UPS control method according to claim 1, characterized in that: After generating the overall control plan, it also includes: Establishing a hybrid energy storage control database to store the parameter configuration list, strategy solution list, real-time control data and working mode records; Regularly verify the validity of historical data in the database and remove redundant data; Based on the time series analysis algorithm, the control effects of the energy storage system are compared longitudinally to evaluate the long-term stability of the control strategy.

10. The UPS control method of flywheel energy storage and battery hybrid according to claim 4, characterized in that: The first classification algorithm is a K-means clustering algorithm, and the specific steps of classifying the environmental monitoring data include: Calculate the Euclidean distance between environmental data samples and initialize the cluster center; Divide the samples into the nearest neighbor cluster centers through iterative updates; The final environmental condition classification result is determined based on the preset cluster number or silhouette coefficient threshold.

Images