A grid frequency-based control method and device for gravity energy storage

Abstract

The present application relates to the field of gravity energy storage control, and particularly relates to a gravity energy storage control method and device based on power grid frequency. The method comprises the following steps: collecting power grid frequency data based on a power grid access point; performing short-period frequency prediction on the power grid frequency data to output a power grid frequency prediction sequence of a future time window; estimating power supply and demand state according to the power grid frequency prediction sequence, calculating a potential energy storage window and a corresponding power grid surplus power value; performing stable window screening according to the potential energy storage window and the corresponding power grid surplus power value, marking an effective energy storage window, and extracting a corresponding effective surplus power value; obtaining operation state information of a gravity energy storage device, performing residual potential energy calculation based on the operation state information, and outputting maximum absorbable energy storage power. The present application realizes reduction of peak shaving cost and improvement of gravity energy storage device utilization rate.

Description

Technical Field

[0001] This invention relates to the field of gravity energy storage control, and more particularly to a gravity energy storage control method and apparatus based on grid frequency. Background Technology

[0002] Gravity energy storage, as an energy storage method based on mechanical potential energy conversion, has the characteristics of large capacity, long life and strong environmental adaptability, and shows good application prospects in the field of power system peak shaving and frequency regulation. It uses electrical energy to drive the lifting of a heavy object, converting electrical energy into gravitational potential energy for storage, and releases the energy by lowering the heavy object when the grid needs it, realizing bidirectional conversion of electrical energy and power regulation.

[0003] In practical engineering applications, gravity energy storage systems face a critical challenge when operating within the power grid: accurately identifying available energy storage periods based on grid frequency variations and achieving efficient matching of power and mechanical boosting processes within a limited time window. Due to the strong randomness, short-term abrupt changes, and multi-source disturbance characteristics of grid frequency fluctuations, traditional methods based on fixed thresholds or static scheduling strategies struggle to accurately characterize the dynamic relationship between frequency changes and power surplus. This can easily lead to problems such as delayed energy storage startup, insufficient power utilization, or frequent start-ups and shutdowns of the mechanical system, thus affecting overall energy storage efficiency and system stability. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a gravity energy storage control method and device based on grid frequency, thereby resolving at least one of the aforementioned technical problems.

[0005] To achieve the above objectives, the present invention provides a gravity energy storage control method based on grid frequency, comprising the following steps: Step S1: Collect power grid frequency data based on the power grid access point; perform short-period frequency prediction on the power grid frequency data, and output the power grid frequency prediction sequence for future time windows; Step S2: Estimate the power supply and demand status based on the power grid frequency prediction sequence, and calculate the potential energy storage window and the corresponding power grid surplus value; Step S3: Based on the potential energy storage windows and the corresponding grid surplus power values, perform stable window screening, mark effective energy storage windows, and extract the corresponding effective surplus power values; Step S4: Obtain the operating status information of the gravity energy storage device, calculate the remaining potential energy based on the operating status information, and output the maximum absorbable energy storage power; Step S5: Calculate the potential energy conversion for lifting heavy objects based on the maximum absorbable energy storage power and the effective surplus power value, and output the number and height of the lifted heavy objects; Step S6: Drive the lifting operation according to the number and height of the heavy objects to be lifted, so as to complete the gravity energy storage control process.

[0006] In this invention, step S1 specifically involves the following steps: High-precision frequency measurement units are deployed at the power grid access point to collect power grid frequency data; The power grid frequency data is continuously sampled and its features are analyzed to extract frequency feature parameters; Perform time-series statistics on the frequency characteristic parameters over multiple periods to construct a first frequency sequence; Anomaly point identification and high-frequency filtering are performed on the power grid frequency fluctuation sequence to output a second frequency sequence; The frequency change trend of the second frequency sequence is analyzed, and short-period frequency prediction is performed to output the power grid frequency prediction sequence for future time windows.

[0007] In this invention, step S2 specifically involves the following steps: Set the rated reference frequency; calculate the frequency deviation of the rated reference frequency based on the power grid frequency prediction sequence to obtain frequency deviation indices at multiple time points; Calculate the frequency change rate index based on the power grid frequency prediction sequence; Based on the frequency deviation index and frequency change rate index, the power supply and demand status is estimated to obtain the power supply and demand status of the power grid at multiple time points. Based on the power supply and demand status of the power grid, the time periods in which the power generation side is greater than the power absorption side of the load side are identified and marked as multiple potential energy storage windows; Calculate the grid surplus power value of the potential energy storage window.

[0008] In this invention, step S3 specifically involves the following steps: Calculate the duration of the potential energy storage window; The power fluctuation intensity is calculated based on the power grid surplus power value, and the power fluctuation intensity parameters are extracted. Based on the duration and power fluctuation intensity parameters, a stable window is selected, an effective energy storage window is marked, and the corresponding effective surplus power value is extracted.

[0009] In this invention, step S4 specifically involves the following steps: Acquire the operating status information of the gravity energy storage device, the operating status information including the remaining liftable height parameter and the mass of a single heavy object; Based on the remaining liftable height parameter and the mass of a single object, the potential energy is calculated to obtain the remaining potential energy storage capacity. Obtain the rated power parameters of the lifting mechanism; calculate the energy storage power based on the rated power parameters and the remaining potential energy storage capacity, and output the maximum absorbable energy storage power.

[0010] In this invention, step S5 specifically involves the following steps: Based on the matching analysis between the effective surplus power value and the maximum absorbable energy storage power, when the maximum absorbable energy storage power is greater than the effective surplus power value, the effective surplus power value is taken as the available energy storage power. When the maximum absorbable energy storage power is not greater than the effective surplus power value, the maximum absorbable energy storage power is taken as the available energy storage power. The potential energy of lifting the weights sequentially is calculated based on the available energy storage power and the mass of a single weight, and the number and height of the weights lifted are output.

[0011] In this invention, step S6 specifically involves the following steps: The total lifting time is calculated based on the number and height of the lifted objects to obtain the energy storage operation time. Based on the operation time and effective energy storage window, the safety constraint calculation of the lifting rate is performed to generate the average lifting speed of the heavy object. Based on statistics of the average lifting speed of the lifting mechanism, the number of heavy objects lifted, and the height, a set of energy storage control parameters is output. The lifting operation of heavy objects is driven by the energy storage control parameter set to complete the gravity energy storage control process.

[0012] In this invention, a gravity energy storage control device based on grid frequency is provided for executing the gravity energy storage control method based on grid frequency as described above, comprising: The acquisition module is used to acquire power grid frequency data based on the power grid access point; perform short-period frequency prediction on the power grid frequency data, and output a power grid frequency prediction sequence for future time windows; The supply and demand status assessment module is used to estimate the power supply and demand status based on the power grid frequency prediction sequence, and to calculate the potential energy storage window and the corresponding power grid surplus power value. The window filtering module is used to filter stable windows based on the potential energy storage windows and the corresponding grid surplus power values, mark effective energy storage windows, and extract the corresponding effective surplus power values. The potential energy calculation module is used to obtain the operating status information of the gravity energy storage device, calculate the remaining potential energy based on the operating status information, and output the maximum absorbable energy storage power. The potential energy conversion module is used to calculate the potential energy conversion of lifting heavy objects based on the maximum absorbable energy storage power and the effective surplus power value, and outputs the number and height of the lifted heavy objects. The energy storage control module is used to drive the lifting operation of the heavy objects according to the number and height of the heavy objects to be lifted, so as to complete the gravity energy storage control process.

[0013] The specific benefits of this invention are as follows: By acquiring high-precision data on the power grid frequency and combining it with short-cycle prediction, the frequency change trend within future time windows can be obtained in advance, enabling the system to have forward-looking scheduling capabilities. This avoids the delayed response caused by relying solely on real-time frequency, improving the sensing speed and prediction accuracy of the gravity energy storage system to power grid fluctuations. Based on the frequency prediction results, power supply and demand status estimation can transform abstract frequency changes into quantifiable power surplus and deficit information, thereby achieving an engineering expression of the power grid status. This allows the energy storage system to accurately identify potential energy storage opportunity windows, improving the targeting of scheduling. By performing stability screening on potential energy storage windows and extracting effective surplus power, short-term fluctuations or unstable power ranges can be effectively eliminated, avoiding frequent start-ups and shutdowns of energy storage devices, improving the continuity of the energy storage process and the stability of system operation, and reducing the risk of ineffective charging. Calculating the remaining potential energy and maximum absorbable energy storage power based on the equipment's operating status ensures that energy storage control matches the physical capabilities of the equipment, avoiding overload operation or space waste, and improving the safety and energy utilization rate of the gravity energy storage system. By matching the maximum absorbable energy storage capacity with the effective surplus power, the dynamic optimization allocation of the number and height of lifted objects is achieved, making the conversion of electrical energy into gravitational potential energy more refined and controllable, and improving energy storage efficiency and power utilization balance. Driving the lifting operation based on lifting parameters enables closed-loop implementation of energy storage control commands, ensuring the stable conversion of surplus power from the grid into mechanical potential energy storage, while simultaneously improving the system's dynamic response to frequency fluctuations and execution reliability. Attached Figure Description

[0014] Figure 1 This is a schematic flowchart of the steps of a gravity energy storage control method based on grid frequency according to the present invention. Figure 2 This is a detailed flowchart illustrating the implementation steps of step S1. Figure 3 This is a flowchart illustrating the detailed implementation steps of step S2. Detailed Implementation

[0015] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0016] This application provides a gravity energy storage control method and apparatus based on grid frequency. The execution entities of the gravity energy storage control method and apparatus based on grid frequency include, but are not limited to, mechanical equipment, data processing platforms, cloud server nodes, network upload devices, etc., which can be considered as general computing nodes in this application. The data processing platform includes, but is not limited to, at least one of an audio / image management system, an information management system, and a cloud data management system.

[0017] In this embodiment, see Figure 1 The diagram below illustrates the steps of a gravity energy storage control method based on grid frequency according to the present invention. In this example, the steps of the gravity energy storage control method based on grid frequency include: Step S1: Collect power grid frequency data based on the power grid access point; perform short-period frequency prediction on the power grid frequency data, and output the power grid frequency prediction sequence for future time windows; Step S2: Estimate the power supply and demand status based on the power grid frequency prediction sequence, and calculate the potential energy storage window and the corresponding power grid surplus value; Step S3: Based on the potential energy storage windows and the corresponding grid surplus power values, perform stable window screening, mark effective energy storage windows, and extract the corresponding effective surplus power values; Step S4: Obtain the operating status information of the gravity energy storage device, calculate the remaining potential energy based on the operating status information, and output the maximum absorbable energy storage power; Step S5: Calculate the potential energy conversion for lifting heavy objects based on the maximum absorbable energy storage power and the effective surplus power value, and output the number and height of the lifted heavy objects; Step S6: Drive the lifting operation according to the number and height of the heavy objects to be lifted, so as to complete the gravity energy storage control process.

[0018] In this embodiment, a high-precision frequency measurement unit is deployed at the power grid access point to continuously collect power grid frequencies. The measurement unit adopts a synchronous phasor measurement or phase-locked tracking structure, with the frequency measurement resolution maintained between 0.001 Hz and 0.0001 Hz. The sampling frequency is set to 200 Hz to 1000 Hz, and the time synchronization error is controlled within 1 millisecond to ensure that the power grid frequency data has high time consistency and high dynamic response capability. The collected power grid frequency data is segmented according to a sliding time window of 1 to 10 seconds, and the window overlap ratio is set to 50% to 80%. Feature parameters such as the frequency mean, frequency maximum, frequency minimum, frequency standard deviation, and frequency change rate are extracted within each time window. The frequency fluctuation amplitude is composed of the difference between the maximum and minimum frequencies within the same window. The multi-window features are arranged in chronological order to form a standard frequency sequence. Based on this, the standard frequency sequence is subjected to short-period trend extrapolation processing, with the time prediction range set to 10 seconds to 300 seconds. The frequency change rate and fluctuation amplitude are jointly analyzed to form a future power grid frequency prediction sequence.

[0019] Based on the power grid frequency prediction sequence, frequency deviation analysis is performed at each prediction time point. The frequency deviation is calculated point by point using the rated reference frequency of 50 Hz as the reference standard. At the same time, the power grid power change trend is judged by combining the frequency change rate. The frequency rise and the change rate are positive, which indicates that the power surplus trend on the generation side is strengthening, and the frequency fall and the change rate is negative, which indicates that the demand on the load side is increasing. On this basis, the frequency deviation and the change rate are jointly mapped to the power grid power supply and demand status sequence, which is used to characterize the power surplus or power deficit of the power grid in the future time window. When the power supply and demand status shows that the output power of the generation side is continuously greater than the power absorbed by the load side, the corresponding continuous time interval is marked as a potential energy storage window, and the surplus power in each potential energy storage window is estimated. In the estimation process, the surplus power level is derived based on the intensity and duration of the frequency deviation, forming a corresponding power grid surplus power value sequence.

[0020] Stability screening analysis is performed on each window, calculating the duration of each potential energy storage window. The duration is obtained by the time difference between the start and end of the window, and time consistency verification is performed on all windows, with the time resolution maintained between 0.1 seconds and 1 second. Simultaneously, the fluctuation intensity of the surplus power sequence within the window is calculated, and the power fluctuation intensity parameter is formed by statistically analyzing the maximum, minimum, and standard deviation of the power, which is used to characterize the power stability within the window. During the screening process, a minimum threshold for duration is set between 30 and 120 seconds, and a joint judgment is made based on the power fluctuation intensity. When the window duration meets the threshold and the power fluctuation intensity is lower than 15% to 40% of the system's rated power, it is determined to be an effective energy storage window. Finally, the set of effective energy storage windows is output, and the average surplus power value after stability correction within the corresponding window is extracted as the effective surplus power value.

[0021] The system acquires the operating status information of the gravity energy storage device, including the remaining liftable height parameter and the mass parameter of a single object. The mass of the object is maintained between 1000 kg and 20000 kg, and the remaining liftable height is measured by a position encoder with a measurement accuracy controlled between 1 cm and 5 cm. Based on the remaining liftable height and the mass of the single object, the remaining potential energy is calculated, and a unified potential energy model is performed using gravitational acceleration. A mechanical efficiency correction coefficient between 0.85 and 0.95 is introduced to account for transmission and structural losses. Based on this, the rated power parameter of the lifting mechanism is obtained, preferably between 100 kW and 5000 kW, and corrected using an efficiency reduction coefficient. Through a dual-constraint analysis of the remaining potential energy storage capacity and the lifting mechanism power capacity, the smaller constraint value is taken as the maximum absorbable energy storage power, thus obtaining the upper limit of the system's ability to safely absorb surplus power from the grid under the current operating state.

[0022] A matching analysis is performed based on the maximum absorbable energy storage power and the effective surplus power. When the maximum absorbable energy storage power is greater than the effective surplus power, the effective surplus power is used as the available energy storage power. When the maximum absorbable energy storage power is not greater than the effective surplus power, the maximum absorbable energy storage power is used as the available energy storage power. A safety margin coefficient between 0.85 and 0.95 is introduced for correction. On this basis, the available energy storage power is converted into gravitational potential energy demand. Combined with the mass range of a single heavy object (1000 kg to 20000 kg) and the remaining liftable height parameter, a joint calculation is performed to determine the potential energy increment that can be lifted per unit time. The lifting process of the heavy object is discretized according to the potential energy increment, and the number of heavy objects to be lifted and the corresponding lifting height allocation are calculated. The lifting height of the heavy object is divided into discrete levels from 0.5 meters to 5 meters, thus forming a multi-object segmented lifting strategy, and outputting a set of parameters for the number of heavy objects lifted and their corresponding heights.

[0023] The gravity energy storage system is controlled and driven based on a set of parameters including the number and height of the lifted objects. These parameters are converted into a sequence of commands for the lifting mechanism, which then executes the lifting operation sequentially. During execution, motor speed, current load, and object position feedback are collected in real time with a sampling period of 0.05 to 0.5 seconds. Simultaneously, the deviation between the actual lifting speed and the planned average speed is monitored. When the speed deviation exceeds 10%, dynamic power correction is performed to maintain a stable energy storage process. On the grid side, frequency change trends are monitored synchronously. When the frequency decreases or surplus power diminishes, the lifting speed is automatically reduced or part of the lifting operation is paused, achieving dynamic coordinated control between grid frequency status and energy storage actions. Finally, the entire lifting process is completed within the effective energy storage window, achieving a stable conversion and storage of surplus grid power into gravitational potential energy, and forming a complete gravity energy storage control output.

[0024] In this embodiment, see Figure 2 The diagram below illustrates the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: High-precision frequency measurement units are deployed at the power grid access point to collect power grid frequency data; The power grid frequency data is continuously sampled and its features are analyzed to extract frequency feature parameters; Perform time-series statistics on the frequency characteristic parameters over multiple periods to construct a first frequency sequence; Anomaly point identification and high-frequency filtering are performed on the power grid frequency fluctuation sequence to output a second frequency sequence; The frequency change trend of the second frequency sequence is analyzed, and short-period frequency prediction is performed to output the power grid frequency prediction sequence for future time windows.

[0025] In this embodiment, a high-precision frequency measurement unit is deployed at the power grid access point to continuously acquire the power grid frequency. The measurement unit employs synchronous phasor measurement or a high-precision phase-locked loop frequency tracking structure, with its frequency measurement resolution preferably controlled within the range of 0.001 Hz to 0.0001 Hz, and the sampling frequency set to 200 Hz to 1000 Hz. The time synchronization error is controlled within 1 millisecond to ensure time consistency between different sampling points. The acquired raw power grid frequency data is segmented into sliding time windows of 1 to 10 seconds, with an overlap ratio of 50% to 80%. Within each time window, basic characteristic parameters such as the mean, maximum, minimum, standard deviation, and rate of change of frequency are extracted. The frequency fluctuation amplitude is obtained from the difference between the maximum and minimum frequencies within the same window. Simultaneously, the dominant frequency fluctuation period is identified using spectral analysis, with the optimal range for identification being 0.5 to 30 seconds, used to characterize the dominant fluctuation rhythm of the power grid frequency. The multidimensional frequency features extracted from each time window are arranged chronologically to construct a first frequency sequence. This sequence is then normalized to ensure its numerical range is stably mapped to the positive and negative one interval, thereby reducing data scale differences under different operating conditions. Based on this first frequency sequence, a frequency wave is constructed using a sliding window method. The dynamic sequence is calculated by setting the sliding window length to 5 to 30 seconds and using an exponential weighting method to fuse the frequency characteristics at different time points. The frequency mean contribution weight is preferably 0.5, the frequency change rate weight is preferably 0.3, and the frequency fluctuation amplitude weight is preferably 0.2, thus forming a comprehensive frequency fluctuation index. To address potential communication noise, electromagnetic interference, and instantaneous abnormal jumps during the acquisition process, a combination of median filtering and wavelet denoising is used for signal purification. The median filtering window is set to 3 to 7 points, the wavelet denoising uses the db4 wavelet function, the decomposition level is set to 3 to 5 levels, and an adaptive threshold method is used to suppress high-frequency noise while retaining low-frequency trend components. After processing, a second frequency sequence is generated, and local trend analysis is performed on this sequence. The frequency change trend slope parameter is calculated using the least squares fitting method, with the trend calculation window set to 10 to 60 seconds to characterize the strength of the rising or falling trend of the power grid frequency.

[0026] A short-period frequency prediction model is constructed based on the second frequency sequence. First-order differencing is applied to the sequence to mitigate non-stationarity, and the stability of the sequence is verified using a stationarity test. If the test results do not meet the stability condition, further second-order differencing or moving average smoothing is performed. The processed frequency sequence is used as input data to construct a multivariate prediction model. Input features include frequency values, frequency change rates, and frequency fluctuation amplitudes. The historical data backtracking time window is set to 60 to 600 seconds, and the prediction time window is set to 10 to 300 seconds. A deep learning prediction model is used for short-period prediction. The deep learning model can adopt a recurrent neural network structure, with a hidden layer size of 64 to 128 neural units and a time step of 10 to 20 sampling periods. An error correction mechanism is introduced during the prediction process. By statistically analyzing the average level and fluctuation range of historical prediction deviations, the prediction results are corrected and compensated, enabling the prediction output to have confidence interval representation capabilities. The confidence range is preferably controlled between ±0.01 Hz and ±0.05 Hz, thus outputting the power grid frequency prediction sequence within the future time window.

[0027] In this embodiment, see Figure 3 The diagram below illustrates the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Set the rated reference frequency; calculate the frequency deviation of the rated reference frequency based on the power grid frequency prediction sequence to obtain frequency deviation indices at multiple time points; Calculate the frequency change rate index based on the power grid frequency prediction sequence; Based on the frequency deviation index and frequency change rate index, the power supply and demand status is estimated to obtain the power supply and demand status of the power grid at multiple time points. Based on the power supply and demand status of the power grid, the time periods in which the power generation side is greater than the power absorption side of the load side are identified and marked as multiple potential energy storage windows; Calculate the grid surplus power value of the potential energy storage window.

[0028] In this embodiment, the rated reference frequency of the power grid is set as a unified reference standard. In a conventional AC power system, the rated reference frequency is preferably set to 50 Hz, and fine-tuning is allowed in different power grid operating areas. The adjustment range is generally controlled between 49.8 Hz and 50.2 Hz to adapt to different operating conditions. After obtaining the power grid frequency prediction sequence, the frequency value at each time point in the prediction sequence is compared with the rated reference frequency point by point to form the frequency deviation index of the corresponding time sequence. The frequency deviation is used to characterize the degree of power supply and demand imbalance of the power grid at different time points in the future. When the deviation is positive, it indicates that the power grid frequency is higher than the reference value, indicating that the power generation side has a relative surplus. When the deviation is negative, it indicates that the power grid frequency is lower than the reference value, indicating that the demand on the load side is high. The time step of each time step in the prediction sequence is uniformly time-calibrated. The time interval is preferably set between 0.1 seconds and 5 seconds to ensure the continuity and traceability of the frequency deviation change. The frequency deviation sequence is then smoothed by sliding. The smoothing window is preferably 3 to 9 sampling points to reduce the impact of prediction error on deviation judgment. Finally, the frequency deviation index sequence corresponding to multiple time points is output.

[0029] After obtaining the power grid frequency prediction sequence, a time-dimensional change analysis is performed on the sequence. By calculating the frequency difference between adjacent time points and combining it with the time interval, a frequency change rate index is obtained. This index reflects the acceleration or deceleration trend of power grid frequency changes. In actual calculations, the time interval is preferably set to 0.1 seconds to 5 seconds to adapt to the dynamic response requirements of different scheduling cycles. Simultaneously, the frequency change rate is divided into intervals: slight change interval, moderate change interval, and drastic change interval. Slight change corresponds to a frequency change rate below 0.01 Hz per second, and moderate change is 0.01 to 0.05 Hz. The frequency deviation is measured at 0.05 Hz per second, with drastic changes exceeding 0.05 Hz per second, to enhance the ability to identify the dynamic state of the power grid. Based on this, the frequency deviation index and the frequency change rate index are jointly analyzed to construct a power supply and demand estimation rule for the power grid. When both the frequency deviation and the frequency change rate are positive, it is determined to be a state of increasing power surplus. When the frequency deviation is positive but the change rate is decreasing, it is determined to be a state of decreasing power surplus. When the frequency deviation is negative and the change rate continues to decrease, it is determined to be a state of expanding power deficit. Through this joint determination method, a multi-dimensional estimation of the future power supply and demand state of the power grid is achieved, and a power supply and demand state sequence corresponding to multiple time points is output.

[0030] Based on the power supply and demand state sequence of the power grid, a continuity analysis of the state types at each time point is performed. Time intervals in which continuous power surplus states occur are identified as potential energy storage windows. During the identification process, a minimum energy storage window duration threshold is set, preferably between 30 and 300 seconds, to avoid misjudging short-term fluctuations as effective energy storage opportunities. At the same time, the continuity of the power surplus state is evaluated for stability. By counting the number of power supply and demand state changes within the window, if the number of state switching times per unit time is less than a preset threshold, such as no more than 2 times per 10 seconds, the window is considered to have high stability. Based on this, time intervals that meet the duration requirements and stability standards are marked as multiple potential energy storage windows. The power surplus level within each energy storage window is calculated, including average surplus power, peak surplus power, and power fluctuation amplitude. The average surplus power is preferably used for subsequent energy storage capacity calculation, the peak surplus power is used for equipment safety verification, and the power fluctuation amplitude is used to assess the difficulty of energy storage regulation, thus forming a complete set of energy storage window characteristics.

[0031] For the identified multiple potential energy storage windows, integral energy statistics are performed on the power data within each window. By accumulating the surplus power at each time point within the window, the total grid surplus power value for that time period is obtained. In practical engineering implementation, the time sampling interval is preferably set to 0.1 to 2 seconds to ensure the accuracy and real-time performance of energy calculation. At the same time, upper and lower limit constraints are applied to the power data to avoid energy calculation deviations caused by abnormal prediction values. The upper limit of power is set according to the grid operating capacity and generally does not exceed 120% of the system's rated load. In the calculation process, a window stability correction coefficient is further introduced. When the power fluctuation within the energy storage window is large, the surplus power value is reduced. The reduction ratio is preferably set between 0.7 and 0.95 to reflect the actual available energy storage capacity. Finally, the grid surplus power value corresponding to each potential energy storage window is output, forming a one-to-one correspondence between energy storage windows and surplus power.

[0032] In this embodiment, step S3 includes the following steps: Calculate the duration of the potential energy storage window; The power fluctuation intensity is calculated based on the power grid surplus power value, and the power fluctuation intensity parameters are extracted. Based on the duration and power fluctuation intensity parameters, a stable window is selected, an effective energy storage window is marked, and the corresponding effective surplus power value is extracted.

[0033] In this embodiment, the duration of the window is determined by parsing the timestamp information of the start and end times of the window. The timestamps are derived from the unified time reference system of the power grid frequency prediction sequence, and the time resolution is preferably set to 0.1 seconds to 1 second to ensure the accuracy of the window duration calculation. During the calculation, each potential energy storage window is represented as a set of continuous time segments, and the continuity of the time segments is checked. When the interval between adjacent time segments is less than 1 second, they are considered as the same continuous window; otherwise, the window is segmented to avoid misjudgment of window breakage due to prediction jitter. In specific implementation, the duration parameter of each potential energy storage window is calculated. This parameter is equal to the window end time minus the window start time, and the duration of all windows is uniformly normalized to map them to a standardized range of 0 to 1, which facilitates subsequent multi-parameter fusion analysis. At the same time, a minimum effective duration threshold is set, preferably between 30 seconds and 120 seconds, to filter out invalid energy storage opportunities with extremely short durations, thereby ensuring that the identified windows have engineering feasibility and energy storage system response reachability.

[0034] The power sequence is resampled at equal time intervals, with the sampling interval preferably set to 0.1 to 2 seconds to ensure that the details of power changes are fully captured. Statistical fluctuation characteristics of the power data within the window are calculated, including the maximum power value, minimum power value, power mean, and power standard deviation. The power fluctuation intensity is characterized by the difference between the maximum and minimum values ​​combined with the standard deviation to reflect the severity of power changes. In specific engineering implementations, a sliding sub-window mechanism is introduced to segment the power changes within the sub-window for analysis. The sub-window length is preferably set to 5 to 20 seconds, and the average absolute deviation of the power changes within each sub-window is calculated to form a set of local fluctuation characteristics. Furthermore, the full-window statistical fluctuation and local fluctuation characteristics are weighted and fused, with the global fluctuation weight preferably set to 0.6 and the local fluctuation weight preferably set to 0.4, thereby generating a unified power fluctuation intensity parameter. This parameter is used to characterize the stability level and dispatchability difficulty of the grid power within the energy storage window.

[0035] Based on the duration and power fluctuation intensity parameters of potential energy storage windows, a multi-condition joint screening analysis is performed on each window. During the screening process, a duration constraint is set, requiring the window duration to be no less than a minimum threshold of 30 to 120 seconds, while not exceeding the maximum dispatchable time range of 600 seconds. This avoids situations where excessively short windows cannot complete energy storage operations or excessively long windows increase prediction uncertainty. Secondly, the stability of the power fluctuation intensity parameter is assessed, and a fluctuation intensity threshold range is set, preferably between 15% and 40% below the system's rated power. When the power fluctuation intensity exceeds this range, the window is considered invalid. If stable energy storage conditions are not met, a joint scoring mechanism is used to process the duration and power fluctuation intensity. By positively weighting the duration and negatively penalizing the power fluctuation intensity, a stability score value is obtained for each window, and the windows are sorted and filtered according to the score values. Finally, the windows with scores higher than a preset threshold are selected as effective energy storage windows, and the corresponding effective surplus power value is extracted for each effective energy storage window. This effective surplus power value is the average power value within the window after stability correction, which is used for subsequent power matching and heavy object lifting control calculations of the gravity energy storage system, thereby completing the engineering-level screening and optimization output of energy storage windows.

[0036] In this embodiment, step S4 includes the following steps: Acquire the operating status information of the gravity energy storage device, the operating status information including the remaining liftable height parameter and the mass of a single heavy object; Based on the remaining liftable height parameter and the mass of a single object, the potential energy is calculated to obtain the remaining potential energy storage capacity. Obtain the rated power parameters of the lifting mechanism; calculate the energy storage power based on the rated power parameters and the remaining potential energy storage capacity, and output the maximum absorbable energy storage power.

[0037] In this embodiment, based on the real-time monitoring architecture of the gravity energy storage system, multi-source acquisition and fusion processing of equipment operation status information is performed. The operation status information includes at least the remaining liftable height parameter and the mass parameter of a single heavy object, while also expanding the acquisition to include the current mechanical position status of the lifting structure, the tension status of the steel cable, and the operational safety margin information. The remaining liftable height parameter is obtained through a position encoder or laser ranging unit, with the measurement accuracy preferably controlled within the range of 1 cm to 5 cm. The mass parameter of a single heavy object is updated through factory calibration combined with an online load correction coefficient, and the mass range is preferably between 1000 kg and 20000 kg depending on the system configuration. The remaining liftable height is segmented and analyzed, dividing it into multiple equivalent lifting unit height intervals, with each interval preferably ranging from 0.5 m to 5 m. Safety constraint markings are applied to different intervals to avoid structural risks when approaching the limit position. At the same time, the consistency verification of the mass of a single heavy object is performed. If there is a multi-object system, the equivalent single mass or equivalent mass distribution coefficient is calculated.

[0038] After obtaining the parameters of the remaining liftable height and the mass of a single object, the remaining potential energy storage capacity of the gravity energy storage system is calculated and analyzed. Potential energy modeling is performed based on standard gravitational acceleration parameters, with the gravitational acceleration preferably set to 9.81 m / s², maintaining a fixed baseline value in engineering calculations. The remaining liftable height is used as the potential energy conversion space, and it is combined with the mass of a single object to obtain the maximum potential energy capacity that the system can store in the current state. The height parameter is dynamically corrected according to the actual mechanical structure, for example, considering the influence of guide rail friction loss and structural deformation, introducing a structural correction coefficient between 0.95 and 1.0. The remaining height is calculated layer by layer, and the potential energy contribution of different height intervals is segmented and superimposed to obtain a continuous potential energy storage capacity curve. An upper limit constraint is applied to this curve to ensure that it does not exceed the safety limit of the mechanical structure. Finally, the remaining potential energy storage capacity parameter is output, which characterizes the theoretical upper limit of the current gravity energy storage system that can continue to absorb and store electrical energy without changing the structure.

[0039] The rated power parameters of the lifting mechanism are obtained. These parameters are jointly determined by the nameplate parameters of the drive motor and the rated output capacity of the frequency converter control system. The power range is preferably set between 100 kW and 5000 kW, depending on the scale of the energy storage system. Simultaneously, the rated power is corrected for operating conditions, including temperature correction coefficient, load correction coefficient, and efficiency reduction coefficient. The overall efficiency is preferably set between 0.85 and 0.95. A joint constraint analysis is performed on the remaining potential energy storage capacity and the rated power parameters of the lifting mechanism. Power boundary limits are imposed based on the maximum input energy per unit time to avoid exceeding the drive system's carrying capacity. Next, considering the remaining potential energy space, it is determined whether there are two constraint states under the current structural conditions: power-limited or space-limited. When the system is in a power-limited state, the rated power is used as the maximum absorption boundary; when the system is in a space-limited state, the remaining potential energy storage capacity is used as the dominant constraint. Finally, the maximum absorbable energy storage power is determined by the principle of taking the smaller of the two constraints, and this power is output as the maximum upper limit value that the gravity energy storage system can safely absorb from the grid surplus power under the current operating conditions.

[0040] In this embodiment, the specific steps of step S5 are as follows: Based on the matching analysis between the effective surplus power value and the maximum absorbable energy storage power, when the maximum absorbable energy storage power is greater than the effective surplus power value, the effective surplus power value is taken as the available energy storage power. When the maximum absorbable energy storage power is not greater than the effective surplus power value, the maximum absorbable energy storage power is taken as the available energy storage power. The potential energy of lifting the weights sequentially is calculated based on the available energy storage power and the mass of a single weight, and the number and height of the weights lifted are output.

[0041] In this embodiment, based on the effective surplus power value and the maximum absorbable energy storage power obtained in the previous steps, a matching analysis is performed on the available power capacity of the gravity energy storage system within the current scheduling cycle. The two types of power parameters are normalized to a unified time scale, with the time resolution preferably set to 0.1 to 1 second to ensure that the power matching process is synchronized with grid frequency changes. The effective surplus power value is used as the upper limit of grid-side supply capacity, and the maximum absorbable energy storage power is used as the upper limit of equipment-side absorption capacity. By comparing the two time-by-time, it is determined whether the system is in a power redundancy state or a power-limited state. When the maximum absorbable energy storage power... When the maximum absorbable energy storage power is greater than the effective surplus power value, it indicates that the energy storage device has the ability to fully absorb the surplus power of the grid. In this case, the effective surplus power value is directly defined as the usable energy storage power and used as the reference for actual control input power. When the maximum absorbable energy storage power is not greater than the effective surplus power value, it indicates that the grid supply capacity exceeds the device's absorption capacity. In this case, the maximum absorbable energy storage power is used as the usable energy storage power to avoid overload operation of the device. A safety margin coefficient is further introduced during the matching process, preferably set between 0.85 and 0.95, to reduce boundary operation risks, thereby outputting the final stable and executable usable energy storage power parameters.

[0042] Using the mass of a single heavy object as the basic calculation unit, the preferred mass parameter range is 1000 kg to 20000 kg, and the standard value of gravitational acceleration of 9.81 m / s² is used as a fixed calculation benchmark. Based on the available energy storage power and the mass of a single heavy object, the theoretical lifting capacity achievable per unit time is calculated, and a mechanical efficiency correction coefficient is introduced, preferably set between 0.85 and 0.95, to reflect transmission mechanism losses and friction losses. The lifting process is segmented, dividing the total lifting process into multiple equivalent lifting cycles, each preferably lasting 5 to 30 seconds. Within each cycle, calculations are performed... It can increase the height increment and accumulate the height increment to form the total lifting height; during the calculation process, the lifting process of each heavy object is scheduled and controlled sequentially. When a single heavy object reaches the current available lifting height limit, the lifting stops and switches to the next heavy object unit, thus forming a multi-object rotation lifting strategy; finally, the number of heavy objects lifted and the corresponding lifting height allocation scheme are output. The number of heavy objects lifted is determined by the available energy storage power and the energy demand for lifting a single heavy object, while the height allocation is constrained by the remaining liftable space and the power duration, thereby realizing the stable conversion of surplus power of the grid into gravitational potential energy and distributed storage.

[0043] In this embodiment, the specific steps of step S6 are as follows: The total lifting time is calculated based on the number and height of the lifted objects to obtain the energy storage operation time. Based on the operation time and effective energy storage window, the safety constraint calculation of the lifting rate is performed to generate the average lifting speed of the heavy object. Based on statistics of the average lifting speed of the lifting mechanism, the number of heavy objects lifted, and the height, a set of energy storage control parameters is output. The lifting operation of heavy objects is driven by the energy storage control parameter set to complete the gravity energy storage control process.

[0044] In this embodiment, the lifting process of each heavy object unit is regarded as an independent but serially executable energy conversion sub-process. The corresponding single lifting time is calculated for each sub-process. The lifting time is determined by the available energy storage power, the mass of a single heavy object, the lifting height, and the mechanical efficiency. The mechanical efficiency is preferably set between 0.85 and 0.95 to reflect the influence of transmission loss and friction loss. All heavy object lifting tasks are arranged sequentially or grouped in parallel. The scheduling is optimized according to the number of lifting mechanisms and the load capacity. The maximum number of heavy object units that can be lifted in a single parallel operation is set to 1 to 10 to avoid instantaneous power surges in the system. The lifting times of all sub-lifting tasks are accumulated and a window constraint is introduced to ensure that the total lifting time does not exceed 90% of the duration of the effective energy storage window, thereby preventing the energy storage process from crossing the power surplus range. Finally, the energy storage operation time parameter is obtained, which is used to characterize the overall time scale required to complete the storage of all heavy object potential energy under the current grid surplus conditions.

[0045] The average lifting speed of the foundation is obtained by proportionally calculating the total lifting height and energy storage operation time. This speed is then corrected by incorporating mechanical structure safety constraints, including the maximum load-bearing rate of the steel cable, the response time of the deceleration mechanism, and the thermal stability time of the motor. The comprehensive safety factor is preferably set between 0.8 and 0.95. Simultaneously, speed balancing is applied to the lifting process of different heavy objects to prevent excessive acceleration of local heavy objects, which could lead to system vibration or load imbalance. In the case of parallel lifting of multiple objects, the overall system speed is uniformly limited to ensure that the average speed of all execution units remains consistent or fluctuates within the allowable deviation range, preferably within ±5%. Finally, the average lifting speed parameter is output, which serves as the core speed benchmark for subsequent control execution, ensuring the stable operation of the gravity energy storage process within a safe range.

[0046] A three-dimensional control parameter system is constructed by using the number of heavy objects as the parallel control dimension, the lifting height as the spatial energy dimension, and the average speed as the time adjustment dimension. The lifting path of each heavy object unit is standardized, and different height ranges are converted into equivalent lifting levels. Each level is preferably between 0.5 meters and 5 meters to facilitate discretization scheduling by the control system. The speed parameter is segmented and mapped, dividing the average speed into low-speed, medium-speed, and high-speed ranges. The low-speed range is preferably less than 0.2 meters per second, the medium-speed range is 0.2 to 0.8 meters per second, and the high-speed range is 0.8 to 2 meters per second to adapt to different power fluctuation scenarios in the power grid. Based on this, the number of heavy objects, lifting height, and speed parameters are jointly encoded to form a structured energy storage control parameter set. This parameter set includes the start time, stop time, lifting speed range, and target lifting height of each heavy object unit, which serves as a unified set of input commands for subsequent execution control.

[0047] The lifting mechanism control unit parses the parameter set into a sequence of motor control commands and executes the lifting action of the heavy object sequentially according to the time order. During the execution, real-time feedback information on motor speed, current load, steel cable tension, and heavy object position is collected. The sampling period is preferably set to 0.05 seconds to 0.5 seconds to ensure control accuracy and response speed. At the same time, the deviation between the actual lifting speed and the target average speed is monitored. When the speed deviation exceeds a preset threshold, such as 10%, a dynamic correction mechanism is automatically triggered to adjust the motor output power to maintain a stable lifting process. On the grid side, frequency changes are monitored synchronously. If a frequency drop or a decrease in power surplus is detected, the lifting speed is reduced or part of the heavy object lifting task is suspended, thereby achieving coordinated control of grid status and energy storage action. Finally, the entire heavy object lifting process is completed within the effective energy storage window, realizing the continuous absorption of surplus power from the grid and the conversion of gravitational potential energy, and completing the complete gravity energy storage control process.

[0048] In this embodiment, a gravity energy storage control device based on grid frequency is provided for executing the gravity energy storage control method based on grid frequency as described above, including: The acquisition module is used to acquire power grid frequency data based on the power grid access point; perform short-period frequency prediction on the power grid frequency data, and output a power grid frequency prediction sequence for future time windows; The supply and demand status assessment module is used to estimate the power supply and demand status based on the power grid frequency prediction sequence, and to calculate the potential energy storage window and the corresponding power grid surplus power value. The window filtering module is used to filter stable windows based on the potential energy storage windows and the corresponding grid surplus power values, mark effective energy storage windows, and extract the corresponding effective surplus power values. The potential energy calculation module is used to obtain the operating status information of the gravity energy storage device, calculate the remaining potential energy based on the operating status information, and output the maximum absorbable energy storage power. The potential energy conversion module is used to calculate the potential energy conversion of lifting heavy objects based on the maximum absorbable energy storage power and the effective surplus power value, and outputs the number and height of the lifted heavy objects. The energy storage control module is used to drive the lifting operation of the heavy objects according to the number and height of the heavy objects to be lifted, so as to complete the gravity energy storage control process.

[0049] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0050] In the embodiments provided in this application, it should be understood that the disclosed apparatus / devices and methods can be implemented in other ways. For example, the apparatus / device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0051] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0052] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0053] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein are implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A gravity energy storage control method based on grid frequency, characterized in that, Includes the following steps: Step S1: Collect power grid frequency data based on the power grid access point; perform short-cycle frequency prediction on the power grid frequency data, and output the power grid frequency prediction sequence for future time windows; Step S2: Estimate the power supply and demand status based on the power grid frequency prediction sequence, and calculate the potential energy storage window and the corresponding power grid surplus value; Step S3: Based on the potential energy storage windows and the corresponding grid surplus power values, perform stable window screening, mark effective energy storage windows, and extract the corresponding effective surplus power values; Step S4: Obtain the operating status information of the gravity energy storage device, calculate the remaining potential energy based on the operating status information, and output the maximum absorbable energy storage power; Step S5: Calculate the potential energy conversion for lifting heavy objects based on the maximum absorbable energy storage power and the effective surplus power value, and output the number and height of the lifted heavy objects; Step S6: Drive the lifting operation according to the number and height of the loads to complete the gravity energy storage control process.

2. The gravity energy storage control method based on grid frequency according to claim 1, characterized in that, The specific steps of step S1 are as follows: High-precision frequency measurement units are deployed at the power grid access point to collect power grid frequency data; The power grid frequency data is continuously sampled and its features are analyzed to extract frequency feature parameters; Perform time-series statistics on the frequency characteristic parameters over multiple periods to construct a first frequency sequence; Anomaly point identification and high-frequency filtering are performed on the power grid frequency fluctuation sequence to output a second frequency sequence; The frequency change trend of the second frequency sequence is analyzed, and short-period frequency prediction is performed to output the power grid frequency prediction sequence for future time windows.

3. The gravity energy storage control method based on grid frequency according to claim 2, characterized in that, The frequency characteristic parameters include frequency fluctuation amplitude, frequency fluctuation period, and frequency change rate.

4. The gravity energy storage control method based on grid frequency according to claim 3, characterized in that, The specific steps of step S2 are as follows: Set the rated reference frequency; calculate the frequency deviation of the rated reference frequency based on the power grid frequency prediction sequence to obtain frequency deviation indices at multiple time points; Calculate the frequency change rate index based on the power grid frequency prediction sequence; Based on the frequency deviation index and frequency change rate index, the power supply and demand status is estimated to obtain the power supply and demand status of the power grid at multiple time points. Based on the power supply and demand status of the power grid, the time periods in which the power generation side is greater than the power absorption side of the load side are identified and marked as multiple potential energy storage windows; Calculate the grid surplus power value of the potential energy storage window.

5. The gravity energy storage control method based on grid frequency according to claim 4, characterized in that, Step S3 is as follows: Calculate the duration of the potential energy storage window; The power fluctuation intensity is calculated based on the power grid surplus power value, and the power fluctuation intensity parameters are extracted. Based on the duration and power fluctuation intensity parameters, a stable window is selected, an effective energy storage window is marked, and the corresponding effective surplus power value is extracted.

6. The gravity energy storage control method based on grid frequency according to claim 5, characterized in that, The specific steps of step S4 are as follows: Acquire the operating status information of the gravity energy storage device, the operating status information including the remaining liftable height parameter and the mass of a single heavy object; Based on the remaining liftable height parameter and the mass of a single object, the potential energy is calculated to obtain the remaining potential energy storage capacity. Obtain the rated power parameters of the lifting mechanism; calculate the energy storage power based on the rated power parameters and the remaining potential energy storage capacity, and output the maximum absorbable energy storage power.

7. The gravity energy storage control method based on grid frequency according to claim 6, characterized in that, The specific steps of step S5 are as follows: Based on the matching analysis between the effective surplus power value and the maximum absorbable energy storage power, when the maximum absorbable energy storage power is greater than the effective surplus power value, the effective surplus power value is taken as the available energy storage power. When the maximum absorbable energy storage power is not greater than the effective surplus power value, the maximum absorbable energy storage power is taken as the available energy storage power. The potential energy of lifting the weights sequentially is calculated based on the available energy storage power and the mass of a single weight, and the number and height of the weights lifted are output.

8. The gravity energy storage control method based on grid frequency according to claim 7, characterized in that, The specific steps of step S6 are as follows: The total lifting time is calculated based on the number and height of the lifted objects to obtain the energy storage operation time. Based on the operation time and effective energy storage window, the safety constraint calculation of the lifting rate is performed to generate the average lifting speed of the heavy object. Based on statistics of the average lifting speed of the lifting mechanism, the number of heavy objects lifted, and the height, a set of energy storage control parameters is output. The lifting operation of heavy objects is driven by the energy storage control parameter set to complete the gravity energy storage control process.

9. A gravity energy storage control device based on grid frequency, characterized in that, For executing the gravity energy storage control method based on grid frequency as described in claim 1, including: The acquisition module is used to acquire power grid frequency data based on the power grid access point; perform short-period frequency prediction on the power grid frequency data, and output a power grid frequency prediction sequence for future time windows; The supply and demand status assessment module is used to estimate the power supply and demand status based on the power grid frequency prediction sequence, and to calculate the potential energy storage window and the corresponding power grid surplus power value. The window filtering module is used to filter stable windows based on the potential energy storage windows and the corresponding grid surplus power values, mark effective energy storage windows, and extract the corresponding effective surplus power values. The potential energy calculation module is used to obtain the operating status information of the gravity energy storage device, calculate the remaining potential energy based on the operating status information, and output the maximum absorbable energy storage power. The potential energy conversion module is used to calculate the potential energy conversion of lifting heavy objects based on the maximum absorbable energy storage power and the effective surplus power value, and outputs the number and height of the lifted heavy objects. The energy storage control module is used to drive the lifting operation of the heavy objects according to the number and height of the heavy objects to be lifted, so as to complete the gravity energy storage control process.

Images