Energy storage power station energy management control method based on fluctuation stabilization

Abstract

The invention discloses an energy storage power station energy management control method based on fluctuation stabilization. The method comprises the steps that power grid operation data, meteorological data related to power grid power changes and energy storage power station state data are monitored and recorded in real time and preprocessed; constructing a hybrid prediction algorithm, and predicting a power grid power change trend in a future period of time by learning historical power grid operation data and historical meteorological data in a past period of time and combining meteorological forecast data; according to the power grid operation data, the meteorological data and the energy storage power station state data which are monitored in real time and the prediction result, a corresponding stabilizing strategy is formulated, and corresponding control operation is executed; according to the power grid operation data and the energy storage power station state data which are monitored in real time, the execution effect of the control operation is evaluated, whether an expected stabilizing target is achieved or not is judged, and if the expected stabilizing target is not achieved, the stabilizing strategy is adjusted through feedback operation. The scheme can effectively improve the stability of the power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of energy management of energy storage power stations, and particularly to an energy management control method for an energy storage power station based on suppressing fluctuations. Background Art

[0002] With the large-scale access of renewable energy, the problem of power fluctuations in the power grid has become increasingly prominent, seriously affecting the stable operation of the power grid. As an effective means of suppressing fluctuations, the energy management control method of an energy storage power station is of great significance for improving the stability of the power grid. However, the existing energy management control schemes for energy storage power stations only include two parts: data monitoring and collection, and strategy formulation and execution. First, sensors and data acquisition systems are used to monitor in real time parameters such as the voltage, current, and frequency of the power grid, and at the same time, information such as the state of charge, charge and discharge rate of the energy storage power station is collected; then, according to the detection results and preset control rules, the charge and discharge strategies of the energy storage power station are formulated and the formulated charge and discharge strategies are executed to adjust the output power of the energy storage power station.

[0003] There are many technical problems in the existing energy management control schemes for energy storage power stations. For example, there is a lack of comprehensive analysis ability for multi-source data (such as meteorological data, load data, etc.) closely related to the power generation of renewable energy, and it is impossible to accurately capture the complex laws of power changes, resulting in lack of flexibility in strategy formulation. The control rules are often based on experience and are difficult to adapt to the rapid changes in the operating state of the power grid. In addition, the existing feedback mechanisms are usually relatively simple, unable to detect and correct deviations in the process of strategy execution in a timely manner, and difficult to meet the high requirements for power grid stability. Therefore, it is necessary to develop more advanced and intelligent energy management control methods to address these challenges. Summary of the Invention

[0004] A series of simplified concepts are introduced in the Summary of the Invention section, which will be further elaborated in the Detailed Description section. The Summary of the Invention section of the present invention does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the protection scope of the claimed technical solution.

[0005] In order to at least partially solve the above problems, the present invention provides an energy management control method for an energy storage power station based on suppressing fluctuations, including:

[0006] S1. Monitor and record in real time the power grid operation data, meteorological data related to the power grid power change, and energy storage power station status data, and preprocess the above data. The power grid operation data includes real-time power, voltage, current, power grid frequency, and line load. The meteorological data includes wind speed, wind direction, and air temperature. The energy storage power station status data includes the state of charge, charge and discharge rate, and battery temperature of the energy storage power station;

[0007] S2. Build a hybrid prediction algorithm to predict the power change trend of the power grid in a future period by learning the historical power grid operation data and historical meteorological data over a past period and combining with meteorological forecast data;

[0008] S3. According to the real-time monitored power grid operation data, meteorological data, energy storage power station status data and the prediction results obtained in the above step S2, formulate corresponding smoothing strategies and execute corresponding control operations;

[0009] S4. According to the real-time monitored power grid operation data and energy storage power station status data, evaluate the execution effect of the control operation, judge whether the expected smoothing target is achieved. If the expected smoothing target is not achieved, adjust the smoothing strategy through feedback operations.

[0010] Furthermore, the step S1 includes normalizing the monitored power grid operation data and meteorological data; the step S2 includes denormalizing the calculation results obtained by the hybrid prediction algorithm to obtain the predicted power value.

[0011] Furthermore, the step S2 includes:

[0012] S21. Predict the power change trend of the power grid in a future period through the ARIMA model to obtain the first predicted power value;

[0013] S22. Predict the power change trend of the power grid in a future period through the LSTM model to obtain the second predicted power value;

[0014] S23. Based on the above first predicted power value, second predicted power value and weight coefficient, establish an ARIMA-LIST hybrid prediction algorithm formula to obtain the hybrid predicted power value.

[0015] Furthermore, the step S21 includes: based on the power mean value μ of a past period extracted from historical data, the power value P hist (t) at a certain past moment and the meteorological data value W(t - 1), the power value P hist (t - 1) at a specific moment before the certain past moment, and the meteorological forecast data W(t), establish an ARIMA model.

[0016] Furthermore, the meteorological data value W(t - 1) at a certain past moment and the meteorological forecast data W(t) are both obtained by fusing wind speed data and wind direction data.

[0017] Furthermore, the step S22 includes: training the LIST network through historical power data, and constructing the output layer of the LIST network into a linear regression model including the power value P hist (t) at a certain past moment.

[0018] Further, it further includes step S24: comparing the actually measured power output value with the hybrid predicted power value in real time, and adjusting the weight coefficient according to the comparison result.

[0019] Further, the power smoothing strategy in step S3 includes an active power adjustment strategy and a reactive power control strategy.

[0020] Further, step S4 includes: detecting the deviation between the actually measured data value and the smoothing target value and performing PID closed-loop control, and adjusting the control parameters based on the deviation value.

[0021] Further, step S3 further includes: adjusting the charge and discharge power of the energy storage power station according to the actually measured state data of the energy storage power station, so that the SOC value is maintained within a set reasonable fluctuation range.

[0022] This application fully considers various factors affecting the power change of the power grid. On the basis of real-time monitoring of basic power grid data, it also collects meteorological data and line load information related to the power change of the power grid, and constructs an innovative hybrid prediction algorithm, and combines weather forecast information to improve the prediction model; and then dynamically formulates and executes a smoothing strategy according to real-time monitoring and prediction data, and at the same time evaluates the execution effect and adjusts the smoothing strategy through feedback operations. Therefore, it can more comprehensively and accurately master the law of power grid power change and correct the result in time, thus effectively improving the stability of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The following drawings of the present invention are used as a part of the present invention to understand the present invention. The embodiments of the present invention are shown in the drawings and their descriptions are used to explain the device and principle of the present invention. In the drawings,

[0024] Figure 1 is a flowchart of the energy management control method of the energy storage power station based on suppressing fluctuations according to an embodiment of the present invention;

[0025] Figure 2 is a schematic structural diagram of the energy management system of the energy storage power station based on suppressing fluctuations according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] In the following description, a large number of specific details are given to provide a more thorough understanding of the present invention. However, it is obvious to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, some technical features well known to the art are not described to avoid confusion with the present invention.

[0027] To thoroughly understand the present invention, detailed structures will be presented in the following description to illustrate the present invention. Apparently, the implementation of the present invention is not limited to the specific details familiar to those skilled in the art of this technology field. The preferred embodiments of the present invention are described in detail below. However, in addition to these detailed descriptions, the present invention may also have other embodiments and should not be construed as limited to the embodiments presented here.

[0028] It should be understood that the purpose of the terms used herein is only to describe specific embodiments and is not a limitation of the present invention. The singular forms "a", "an" and "the" are also intended to include the plural forms unless the context clearly indicates otherwise. When the terms "comprising" and / or "including" are used in this specification, they specify the presence of the described features, wholes, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their combinations. The terms "upper", "lower", "front", "rear", "left", "right" and similar expressions used in the present invention are only for the purpose of illustration and are not limitations.

[0029] The ordinal numbers such as "first" and "second" cited in the present invention are only identifiers and do not have any other meanings, such as a specific order, etc. Moreover, for example, the term "first component" does not imply the existence of a "second component" by itself, and the term "second component" does not imply the existence of a "first component" by itself.

[0030] Hereinafter, the specific embodiments of the present invention will be described in more detail with reference to the accompanying drawings. These drawings show representative embodiments of the present invention and do not limit the present invention.

[0031] The present invention provides an energy management control method for an energy storage power station based on suppressing fluctuations, and this method is particularly applicable to the energy management of a wind farm energy storage power station. As Figure 1 shown, this method includes:

[0032] Step S1, real-time monitoring and recording of grid operation data, meteorological data related to grid power changes, and energy storage power station status data, and preprocessing of the above data. The grid operation data includes real-time power, voltage, current, grid frequency, and line load. The meteorological data includes wind speed, wind direction, and air temperature. The energy storage power station status data includes the state of charge, charge and discharge rate, and battery temperature of the energy storage power station.

[0033] In this embodiment, a power transmitter, a voltage sensor, and a current sensor are deployed at key grid nodes to collect real-time grid operation data such as the power, voltage, and current of the grid; a wind speed sensor, a wind direction sensor, and a temperature sensor are deployed at multiple locations in the wind power station to collect meteorological data such as the wind speed, wind direction, and temperature of the wind power station in real time; the collected data is transmitted to the energy management system of the energy storage power station through a high-speed communication network, and the energy management system performs preprocessing on the received data including filtering and calibration to ensure the accuracy and reliability of the data.

[0034] In specific applications, it is necessary to perform data normalization on the received data. Since historical power data and meteorological data may have different dimensions and magnitudes, directly performing data analysis may cause some data features to be overamplified or ignored. Therefore, in order to achieve the accuracy of the monitoring results, it is necessary to perform data normalization. The implementation scheme is to use min-max normalization to map the data to the [0,1] interval. For any data point x, its normalized value x norm can be calculated using the following formula:

[0035]

[0036] where x min and x max are the minimum and maximum values in the data respectively.

[0037] For example: Suppose a set of historical power data P hist (t) and the corresponding wind speed data W(t) have been collected. For the sake of simplicity in explanation, only the processing process of a single data point is considered. For example, assume that the historical power data at a certain moment is P hist (t) = 150 MW, and the corresponding wind speed data is W(t) = 10 m / s. The data normalization process is as follows:

[0038] Suppose the minimum value of the historical power data is P min = 100 MW, and the maximum value is P max = 200 MW.

[0039] The minimum value of the wind speed data is W min = 5 m / s, and the maximum value is W max = 15 m / s.

[0040] Then the normalized power data is:

[0041] The normalized wind speed data is:

[0042] Similarly, the same normalization process is performed on the wind direction. Taking the direction parallel to the fan blade as the minimum value with A = 0 degrees and the direction perpendicular to the fan blade as the maximum value with A = 90 degrees. Assuming A(t) = 45 degrees, the normalized wind direction data is as follows: 0.5. At the same time, the normalized wind speed and wind direction data are fused to obtain the comprehensive exogenous variable data: W t = ω·W norm (t)+(1 - ω)·A norm (t), where ω can be calculated by analyzing the historical power data of the wind power generation equipment and the corresponding relationship between the wind speed and wind direction, and by using the maximum likelihood estimation optimization algorithm. Here, ω = 0.5 is taken. Through the above formula calculation, the exogenous variable data: W t = 0.5.

[0043] Step S2: Construct a hybrid prediction algorithm. By learning the historical power grid operation data and historical meteorological data over a period of time, combined with meteorological forecast data, predict the power change trend of the power grid over a period of time in the future.

[0044] The core of this step is to construct an ARIMA-LSTM hybrid prediction model algorithm to predict the power change trend over a period of time in the future. Specifically, it can include:

[0045] S21: Predict the power change trend of the power grid over a period of time in the future through the ARIMA model to obtain the first predicted power value.

[0046] Specifically, construct the ARIMA model formula:

[0047]

[0048] where, P ARIMA (t) is the first predicted power value obtained from the ARIMA model,

[0049] μ is the mean of the historical data,

[0050] and θ j are the autoregressive coefficient and moving average coefficient of the ARIMA model respectively, p and q are the autoregressive order and moving average order respectively,

[0051] ∈ t is the error term,

[0052] P hist (t) is the normalized power data,

[0053] W t is the exogenous variable, that is, the above-mentioned normalized meteorological data fused with wind speed data and wind direction data,

[0054] β k is the coefficient of the exogenous variable.

[0055] In the above ARIMA model formula, the model order can be determined by the autocorrelation function (ACF) and partial autocorrelation function (PACF) plots to obtain the orders p and q of the ARIMA model; the parameters of the ARIMA model are estimated using the maximum likelihood estimation optimization algorithm and θ j ; during model testing, the fitting effect of the model is tested through the residual analysis method.

[0056] S22. Predict the power change trend of the power grid over a period of time in the future through the LSTM model to obtain the second predicted power value.

[0057] By training the LSTM network, convert the historical power data into a sequence format suitable for input to the LSTM network, design the structure of the LSTM network, including the number of neurons in the input layer, hidden layer, and output layer, etc. Use historical data to train the LSTM network, and optimize the network parameters through the backpropagation algorithm. The output layer of the LSTM network is designed as a linear regression model, and its simplified calculation formula is:

[0058] P LSTM (t) = P hist (t) + w out ·h t-1 + b out .

[0059] Among them, P LSTM (t) is the second predicted power value obtained by the LSTM model,

[0060] h t-1 is the hidden state of the previous moment of the LSTM network,

[0061] w out and b out are the weight and bias of the output layer respectively.

[0062] S23. Based on the above first predicted power value, second predicted power value, and weight coefficient, establish the ARIMA-LIST hybrid prediction algorithm formula to obtain the hybrid predicted power value.

[0063] The ARIMA-LSTM hybrid prediction algorithm formula is:

[0064] P pred (t) = α·P ARIMA (t) + (1 - α)·P LSTM (t)

[0065] Among them, α is the weight coefficient, which can be adjusted according to the subsequent monitoring execution situation. Specifically, the actual power output value obtained by real-time monitoring is compared with the ARIMA-LSTM hybrid predicted power value P pred (t + 1). When P pred (t + 1) is greater than the actual power output value, the weight coefficient α is increased; when P pred (t + 1) is less than the actual power output value, the weight coefficient α is decreased; when P pred (t + 1) is equal to the actual power output value, the weight coefficient α remains unchanged.

[0066] For example: Suppose the parameters of the ARIMA model have been obtained by fitting historical data: μ = 0.5, θ1 = 0.4, β1 = 0.5, β2 = -0.3, and the error term ∈ t is not considered in this calculation (i.e., assumed to be 0). Now, to predict the power value at the next moment, the relevant values are substituted into the ARIMA model calculation formula:

[0067]

[0068] Substitute specific values: Let the power value P hist (t) = 0.6 and the meteorological data value W(t - 1) = 0.4. The power value P hist (t - 1) = 0.55 at a specific moment before this moment (these values are obtained from historical data, where W(t - 1) is obtained by fusing historical wind speed and wind direction data), W(t) = 0.5 (this value is obtained from the wind speed and wind volume in the meteorological forecast data on the Internet and is converted after normalization), and assume ∈ t-1 = 0. At this time, the calculation is as follows:

[0069] P ARIMA (t + 1) = 0.5 + 0.5×(0.6 - 0.5) + (-0.3)×(0.55 - 0.5) + 0.4×0 + 0.5

[0070] ×(0.5 - 0.5) + (-0.3)×(0.4 - 0.5) = 0.565

[0071] That is, through the ARIMA model calculation formula, the predicted power value at the next moment is P ARIMA (t + 1) = 0.565.

[0072] Suppose the parameters of the LSTM network calculation formula have been obtained by fitting historical data. Among them: P hist (t) = 0.6, h t-1 = 0.55 (this is the value after normalization), w out= 0.8, b out = -0.5, substituting the relevant values into the LSTM network calculation formula:

[0073] P LSTM (t + 1)= 0.6 + 0.55×0.8 - 0.5 = 0.54.

[0074] That is, through the LSTM network calculation formula, the power value at the next moment is predicted as P LSTM (t + 1)= 0.54.

[0075] In the ARIMA-LSTM hybrid prediction algorithm formula, after learning and calculation, the preset weight coefficient α = 0.6 is initially determined. The power value at the next moment obtained through the ARIMA model and the LSTM network calculation formula is calculated as follows:

[0076] P pred (t + 1)= 0.6·0.565+(1 - 0.6)·0.54 = 0.555

[0077] The predicted value P pred (t + 1)= 0.555 is the value after normalization. It is necessary to convert it back to the original dimension. The denormalization formula is:

[0078] P pred (t)= P norm (t)×(P max - P min )+ P min

[0079] Substitute the normalized predicted value:

[0080] P pred (t)= 0.555×(200 - 100)+100 = 155.5 MW

[0081] Therefore, the predicted power value at the next moment is 155.5 MW.

[0082] That is, through the RIMA-LSTM hybrid prediction algorithm formula, the power value at the next moment is predicted as P pred (t + 1)= 155.5 MW. After the energy management system of the energy storage power station receives this predicted value, it can make corresponding adjustments in advance according to the current operating state and scheduling plan of the power grid, such as adjusting the output of other power sources or the charge and discharge strategy of the energy storage power station, to suppress the upcoming power fluctuations, that is, the following step S3.

[0083] Step S3: According to the real-time monitored power grid operation data, meteorological data, energy storage power station status data, and the prediction results obtained in the above step S2, formulate corresponding suppression strategies and execute corresponding control operations.

[0084] In this embodiment, the flattening strategy includes the active power adjustment strategy AGC and the reactive power control strategy AVC for the energy storage power station.

[0085] The active power adjustment strategy includes the following steps: 1) Set the target power P target , which is determined according to the current demand of the power grid and the predicted future power change; 2) Calculate the available maximum / minimum active power output range according to the SOC value of the energy storage power station; 3) Adjust the active power output of the energy storage power station to make it close to or reach the target power P target . The following calculation formula can be used:

[0086] P target =P current +ΔP forecast

[0087] where P current is the active power of the current power grid,

[0088] ΔP forecast is the active power that needs to be adjusted according to the prediction.

[0089] The reactive power control strategy plan includes the following steps: 1) The system monitors key parameters such as the voltage and frequency of the power grid in real time; 2) According to the actual situation of the power grid, the system calculates the amount of reactive power ΔQ that needs to be adjusted; 3) Adjust the output of the reactive power compensation device (such as SVG, SVC, etc.) of the energy storage power station to achieve the purpose of adjusting the reactive power. The following calculation formula can be used:

[0090] ΔQ = K v ·(target voltage - actual voltage) + K f ·(target frequency - actual frequency)

[0091] For example: In the afternoon, due to a large amount of wind power generation being connected to the grid, the power of a certain power grid is excessive and the grid frequency rises. Assume that the relevant data of the power grid at this time is: the current active power of the power grid P current = 500 MW, it is predicted that the wind power generation will increase by 200 MW in the next hour, but the load demand only increases by 50 MW. Therefore, the power output needs to be reduced by 150 MW. The current SOC value of the energy storage power station is 75%, and the maximum discharge power is 200 MW. The applied formula is: P target = 500 MW - 150 MW = 350 MW, and the system adjusts the active power output of the energy storage power station to 350 MW to absorb the excess power.

[0092] During the peak period of a certain power grid, due to a sudden increase in load, the voltage drops. Assume that the relevant data of the power grid at this time is: the actual voltage is U actual = 198 kV, and the target voltage is U target= 200 kV; the actual frequency is f actual = 50.1 Hz, and the target frequency is f target = 50 Hz. Set the adjustment coefficient K v = 10 Mvar / kV, K f = 10 Mvar / Hz. The applied formula is: ΔQ = 10×(200 - 198) + 10×(50 - 50.1) = 20 Mvar - 1 Mvar = 19 Mvar. The system adjusts the output of the reactive power compensation device of the energy storage power station to increase the reactive power output by 19 Mvar to improve the grid voltage.

[0093] In some embodiments of this embodiment, the active power / AGC and reactive power / AVC controls can be divided into four modes: dispatching control, manual setting, plan tracking, and exit. Among them, the dispatching mode has the highest priority and is the default mode. In the default mode, the system will perform automatic control and adjustment based on the above specific algorithm to suppress the fluctuations of the power grid. In addition, the operator can also adjust the mode according to the power supply plan at the power grid control center, switching to the manual setting, plan tracking, and exit modes. At this time, the real-time monitoring and prediction module can still operate automatically, and at the same time, the real-time monitoring data and prediction data of the power grid are displayed on the display instrument or screen at the control center, so as to provide comprehensive data reference for the operator's control.

[0094] Step S4: According to the real-time monitored power grid operation data and the energy storage power station status data, evaluate the execution effect of the control operation, and judge whether the expected suppression target is achieved. If the expected suppression target is not achieved, adjust the suppression strategy through feedback operation.

[0095] In this embodiment, the deviation between the real-time monitored data value and the suppression target value is detected and the strategy is adjusted through the feedback link, which may specifically include: 1) deviation detection, the system compares the actual effect with the expected target in real time; 2) calculate the deviation value; 3) dynamically adjust the control parameters according to the magnitude and direction of the deviation value, such as the proportional gain coefficient K p and the integral gain coefficient K i ; 4) recalculate the new control instruction using the adjusted control parameters. To describe the strategy adjustment process more accurately, the following PID (Proportional-Integral-Differential) closed-loop control formula is introduced:

[0096]

[0097] where, P new is the adjusted control output;

[0098] P old is the previous control output;

[0099] K p and Ki , K d are the proportional, integral, and derivative gain coefficients, respectively;

[0100] ∫(P Δ )dt and

[0101] In the example described above, the system calculates that the active power output of the energy storage power station needs to be adjusted to 350 MW to suppress the power surplus problem of the power grid. When the strategy is executed, the system sends a command to the energy storage power station to adjust its active power output to 350 MW. The system monitors the output power of the energy storage power station and parameters such as the voltage and frequency of the power grid in real time. Assuming the actual output power is 345 MW, there is a deviation of 5 MW from the target value; then parameter adjustment is carried out: assuming that initially K p = 0.1, K i = 0.01, K d = 0.05, according to the deviation value, the system dynamically adjusts these coefficients. For example, in order to eliminate the deviation faster, the value of K p can be appropriately increased and adjusted to K p = 0.15, and the new output power is recalculated using the PID control formula:

[0102]

[0103] Assuming that the influence of the integral term and the derivative term is small, it can be simplified to:

[0104] P new ≈ 345 MW + 0.15 × 5 MW = 345.75 MW

[0105] According to the calculation result, the system adjusts the output power of the energy storage power station to 345.75 MW. In the scenario where the power grid fluctuates frequently, PID control can provide more accurate and smooth control effects. By adjusting the control parameters in real time, the system can quickly respond to the changes of the power grid and effectively eliminate the deviation. In the above example, by dynamically adjusting coefficients such as K p , the system approaches the target value faster, improving the efficiency and accuracy of the suppression strategy. This strategy adjustment method is crucial for ensuring the stable operation of the power grid, especially when dealing with emergencies or large fluctuations in load.

[0106] The above methods are implemented through each module of the energy storage power station energy management system, specifically including the real-time monitoring and prediction module, the suppression strategy formulation module, and the strategy execution and feedback module. As Figure 2 shown, the energy storage power station capacity management system also includes a function expansion module, and the function expansion module specifically implements the following solutions:

[0107] I. SOC automatic maintenance function

[0108] Intelligent algorithm: Adopt intelligent optimization algorithms (such as genetic algorithms, particle swarm optimization), and optimize the charge and discharge strategies of the energy storage power station according to historical data and real-time monitoring data.

[0109] SOC range setting: Set a reasonable SOC fluctuation range, such as 20%-80%, to ensure the safe operation and lifespan of the battery.

[0110] Charge and discharge control: According to the optimized strategy, the system automatically adjusts the charge and discharge power of the energy storage power station to keep the SOC value within the set reasonable range.

[0111] During the automatic SOC maintenance, let the current SOC value be SOC current , and the target SOC range is [SOC min , SOC max . The charge and discharge rate limit is ±ΔP max . The optimization objective function can be expressed as:

[0112] Minimize|SOC target -SOC current |

[0113] Constraint conditions:

[0114] SOC min ≤SOC target ≤SOC max

[0115] -ΔP max ≤ΔP≤ΔP max

[0116] where ΔP is the adjusted charge and discharge power.

[0117] Example: Assume the current SOC value is 30%, and the target range is 40%-70%. After optimization by the intelligent algorithm, the system decides to charge at a power of 10 MW to increase the SOC value to within the target range.

[0118] II. Emergency power control function

[0119] Emergency event detection: Real-time monitor the grid status and detect emergency events, such as large-scale power outages, line faults, etc.

[0120] Maximum power scheduling: In the emergency power control mode, the system schedules the energy storage power station to perform emergency charging / discharging / power outage operations at the maximum power.

[0121] Safety protection: When performing emergency operations, the system takes necessary safety protection measures, such as preventing the battery from overcharging / overdischarging to protect electrical equipment.

[0122] In an emergency, the system needs to quickly dispatch the energy storage power station to discharge at the maximum power P max for discharging.

[0123] The discharge time (T) can be estimated by the following formula:

[0124]

[0125] where E battery is the total energy capacity of the battery.

[0126] For example: Suppose in a power grid emergency, the system detects a sharp drop in the power grid frequency and needs to immediately discharge to stabilize the power grid. The current SOC value is 60%, the total energy capacity of the battery is 2000 kWh, and the maximum discharge power is 500 kW. Calculated according to the formula:

[0127]

[0128] The system immediately dispatches the energy storage power station to discharge at the maximum power of 500 kW, which is expected to last for 2.4 hours, thus effectively relieving the power grid pressure and ensuring the safe and stable operation of the power grid.

[0129] The above SOC automatic maintenance function extends the battery life and reduces the maintenance cost through intelligent optimization algorithms. The emergency power control function provides strong power grid support capabilities at critical moments, ensuring the safe and stable operation of the power grid during emergencies. These two functions together enhance the importance and application value of the energy storage power station in modern power systems.

[0130] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner can depend on the specific application and design constraints of the technical solution. The technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the embodiments of this disclosure.

[0131] The steps of the methods or algorithms described in combination with the embodiments disclosed in this article can be directly implemented by hardware, software modules executed by a processor, or a combination of the two. The software modules can be placed in a random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the technical field.

[0132] In the embodiments disclosed herein, the disclosed methods, products (including but not limited to devices, equipment, etc.) can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units can be merely a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Additionally, the couplings or direct couplings or communication connections shown or discussed among each other can be through some interfaces, and the indirect couplings or communication connections of the devices or units can be in electrical, mechanical or other forms.

[0133] Unless otherwise defined, the technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the technical field of the present invention. The terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention. Terms such as "part" and "component" that appear herein can represent either a single part or a combination of multiple parts. Terms such as "mounted" and "set" that appear herein can represent either a component being directly attached to another component or a component being attached to another component through an intermediate member. Features described in one embodiment herein can be applied to another embodiment alone or in combination with other features, unless the feature is not applicable or otherwise stated in that other embodiment.

[0134] The present invention has been illustrated by the above embodiments, but it should be understood that the above embodiments are only for purposes of illustration and example, and are not intended to limit the present invention to the scope of the described embodiments. In addition, those skilled in the art can understand that the present invention is not limited to the above embodiments, and more variations and modifications can be made according to the teachings of the present invention, and these variations and modifications all fall within the scope claimed by the present invention. The protection scope of the present invention is defined by the appended claims and their equivalent scope.

Claims

1. An energy management and control method for an energy storage power station based on smoothing fluctuations, characterized in that: include: S1. Real-time monitoring and recording of grid operation data, meteorological data related to grid power changes, and energy storage power station status data, and pre-processing of the above data. The grid operation data includes real-time power, voltage, current, grid frequency, and line load. The meteorological data includes wind speed, wind direction, and temperature. The energy storage power station status data includes the state of charge, charge and discharge rate, and battery temperature of the energy storage power station. S2. Construct a hybrid prediction algorithm to predict the power change trend of the power grid in the future by learning the historical power grid operation data and historical meteorological data in the past period of time and combining it with meteorological forecast data; S3, formulate corresponding stabilization strategies and perform corresponding control operations according to the real-time monitored grid operation data, meteorological data, energy storage power station status data and the prediction results obtained in the above step S2; S4. Based on the real-time monitored grid operation data and energy storage power station status data, evaluate the execution effect of the control operation and determine whether the expected smoothing target is achieved. If the expected smoothing target is not achieved, adjust the smoothing strategy through feedback operation.

2. The method according to claim 1, characterized in that The step S1 includes: normalizing the monitored power grid operation data and meteorological data; the step S2 includes: denormalizing the calculation results obtained by the hybrid prediction algorithm to obtain the predicted power value.

3. The method according to claim 1, characterized in that The step S2 comprises: S21, predicting the power change trend of the power grid in the future period by using the ARIMA model to obtain a first predicted power value; S22, predicting the power change trend of the power grid in the future period of time through the LSTM model to obtain a second predicted power value; S23. Based on the first predicted power value, the second predicted power value and the weight coefficient, an ARIMA-LIST hybrid prediction algorithm formula is established to obtain a hybrid predicted power value.

4. The method according to claim 3, characterized in that The step S21 includes: based on the power mean μ for a period of time extracted from the historical data, the power value P at a certain moment in the past hist (t) and the meteorological data value W(t-1), the power value P at a specific time before a certain time in the past hist (t-1), and weather forecast data W(t), to establish the ARIMA model.

5. The method according to claim 4, characterized in that The meteorological data value W(t-1) at a certain moment in the past and the meteorological forecast data W(t) are both obtained by fusing the wind speed data and wind direction data.

6. The method according to claim 3, characterized in that The step S22 includes: training the LIST network through historical power data, constructing the output layer of the LIST network to include the power value P at a certain moment in the past hist (t) Linear regression model.

7. The method according to claim 3, characterized in that The method further includes step S24: comparing the actual power output value monitored in real time with the hybrid predicted power value, and adjusting the weight coefficient according to the comparison result.

8. The method according to claim 1, characterized in that The stabilization strategy of step S3 includes an active power adjustment strategy and a reactive power control strategy.

9. The method according to claim 1, characterized in that: The step S4 includes: detecting the deviation between the real-time monitored data value and the smoothing target value and performing PID closed-loop control, and adjusting the control parameters based on the deviation value.

10. The method according to any one of claims 1 to 9, characterized in that: The step S3 also includes: adjusting the charging and discharging power of the energy storage power station according to the real-time monitored energy storage power station status data, so that the SOC value is maintained within a set reasonable fluctuation range.

Images