Wind and light energy storage management method based on improved adaptive filtering and MPC

Abstract

The invention relates to the technical field of new energy power generation, in particular to a wind and light energy storage management method based on improved adaptive filtering and MPC, which comprises the following steps: building a wind and light energy storage grid-connected inverter system, and connecting wind power, photovoltaic and double hybrid energy storage modules in parallel with a direct current bus; an improved self-adaptive moving average filtering algorithm is provided, and data are processed in real time to realize local and global optimization of filtering stabilization; a double-hybrid energy storage energy management algorithm based on model predictive control (MPC) is designed, double-energy-storage compensation wind and light plan and actual power and load and plan power deviation are controlled respectively, energy storage SOC is restrained to be 0.2-0.8, and energy storage power exchange and power grid switching are controlled. The wind-solar power storage and supply grid-connected power stability, the wind power plan accuracy and the local load stability are effectively improved, the energy storage SOC out-of-limit frequency is reduced, the energy storage service life is prolonged, meanwhile, the system response speed is increased, and the overshoot is reduced.

Description

Technical Field

[0001] This invention relates to the field of new energy power generation technology, specifically a wind-solar-storage energy management method based on improved adaptive filtering and MPC. Background Technology

[0002] A stable energy supply provides a crucial guarantee for developing the national economy and improving people's living standards. However, the inherent undispatchability of wind and solar power weakens users' willingness to actively consume new energy sources. A grid-connected, localized power supply operation mode can, to some extent, improve the absorption rate of new energy sources and avoid grid instability caused by large-scale grid connection of new energy. Similar to grid-connected power generation, to maintain supply and demand balance in a grid-connected system, the power generation data must first be predictable, and the planned power data must be reliable and dispatchable. Secondly, the hardware facilities used for regulation must have flexible and controllable energy throughput to absorb positive and negative deviations in supply and demand. These two points rely on effective power generation planning, reliable energy management, and feasible control models.

[0003] The vast majority of existing wind power planning data comes directly from forecasts. Classic forecasting methods include combinations of backpropagation neural networks (BP neural networks), extreme learning machines (ELM), long short-term memory networks (LSTM), and improved particle swarm optimization (PSO) neural networks. However, the uncertain changes in natural climate conditions can lead to large fluctuations in predicted power, which is detrimental to supply and demand scheduling. To address this, current planning strategies involve smoothing forecast data and using energy storage systems to mitigate fluctuations in grid-connected renewable energy power. Commonly used techniques include improved wavelet packet techniques, moving average filtering, and adaptive moving average filtering. However, when considering adaptively adjusting the moving window to handle outliers, a large filtering window value can affect the algorithm's response speed or cause over-smoothing. The traditional adaptive moving average filtering algorithm selects its filtering window value based on the global optimum, which can lead to over-smoothing on local datasets. Furthermore, this algorithm is often used for offline data and cannot filter and smooth real-time data. Meanwhile, existing technologies lack a dual-hybrid energy storage management strategy that can balance the accuracy of wind and solar power generation plans with the stability of local load power supply. This makes it difficult to effectively solve the problem of the supply and demand balance of distributed generation affected by asynchronous changes in system wind speed and load, and also fails to effectively avoid the situation of hybrid energy storage exceeding the SOC limit, which is not conducive to extending the service life of energy storage equipment. Therefore, in view of the above situation, it is urgent to develop a wind and solar energy storage management method based on improved adaptive filtering and MPC to overcome the shortcomings in current practical applications. Summary of the Invention

[0004] The purpose of this invention is to provide a wind-solar-storage energy management method based on improved adaptive filtering and MPC, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A wind-solar-storage energy management method based on improved adaptive filtering and MPC includes the following steps:

[0007] (1) Build a wind-solar-storage grid-connected inverter system, and connect the wind power generation module, photovoltaic power generation module and dual hybrid energy storage module in parallel to the DC bus. The DC bus is connected to the power grid and local load respectively through converters;

[0008] (2) An improved adaptive moving average filtering algorithm is used to perform real-time filtering on wind and solar power generation data to achieve local and global optimization of data filtering and smoothing.

[0009] (3) Design a dual hybrid energy storage energy management algorithm based on model predictive control (MPC), which compensates for different types of power deviations through the dual hybrid energy storage modules, constrains the state of charge (SOC) of the dual hybrid energy storage modules within a preset range, and controls the power exchange and grid connection of the dual hybrid energy storage modules.

[0010] As a further aspect of the present invention: in step (1), the construction of the wind-solar-storage grid-connected inverter system specifically includes:

[0011] The wind power generation module is connected to the DC bus via an AC / DC converter, the photovoltaic power generation module is connected to the DC bus via a DC / DC converter, the dual hybrid energy storage module is connected in parallel to the DC bus via a DC / DC converter, the local load is directly connected to the DC bus, and the power grid is connected to the DC bus via an AC / DC converter.

[0012] As a further aspect of the present invention: the dual hybrid energy storage module includes a first hybrid energy storage unit and a second hybrid energy storage unit, both of which are composed of a battery and a supercapacitor.

[0013] As a further aspect of the present invention: In step (2), the improved adaptive moving average filtering algorithm processes the real-time collected wind and solar power generation data and load power consumption data by setting an adaptive variable filtering window value and a maximum filtering window value, so that the filtered grid-connected power meets the grid-connected power fluctuation rate standard under a preset time scale.

[0014] As a further aspect of the present invention, the implementation process of the improved adaptive moving average filtering algorithm includes:

[0015] If the grid-connected power corresponding to the current filter window value meets the volatility standard, then the filter window value is maintained or adjusted.

[0016] If the condition is not met, increase the filter window size;

[0017] If the current filter window value reaches the maximum filter window value, the grid-connected power is made to meet the volatility standard by constraining the difference between the grid-connected power at the current moment and the grid-connected power at the previous moment.

[0018] As a further aspect of the present invention: in step (3), the different types of power deviations include:

[0019] The first deviation power between the planned power of wind and solar power generation and the actual power of wind and solar power generation, and the second deviation power between the local load demand power and the planned power of wind and solar power generation;

[0020] The first hybrid energy storage unit is used to compensate for the first deviation power, and the second hybrid energy storage unit is used to compensate for the second deviation power.

[0021] As a further aspect of the present invention: the first deviation power and the second deviation power are both distributed to the battery and supercapacitor of the corresponding hybrid energy storage unit through a low-pass filter, the supercapacitor absorbs the high-frequency power component, and the battery regulates the low-frequency power component.

[0022] As a further aspect of the present invention: in step (3), the preset range is 0.2-0.8, and the charging and discharging states of the battery and supercapacitor are optimized by the dual hybrid energy storage energy management algorithm to avoid exceeding the SOC limit.

[0023] As a further aspect of the present invention: in step (3), the dual hybrid energy storage energy management algorithm further includes:

[0024] Based on the SOC of the dual hybrid energy storage module, the power exchange between the first hybrid energy storage unit and the second hybrid energy storage unit is controlled; when the SOC is about to exceed the limit or cannot be avoided by power exchange, the grid is controlled to switch in to compensate for the deviation power.

[0025] As a further aspect of the present invention: in step (3), the MPC is applied to the DC / DC converter control between the DC bus and the dual hybrid energy storage module.

[0026] Compared with the prior art, the beneficial effects of the present invention are:

[0027] 1. This invention constructs a wind / solar / storage grid-connected inverter power generation system, which connects the output power of the wind power generation system, the photovoltaic power generation system, and the dual hybrid energy storage module in parallel to the DC bus voltage, and then connects them to the grid through an AC / DC inverter. This structure is easy to control and has a simple structure.

[0028] 2. This invention proposes an improved adaptive moving average filtering algorithm, which sets an adaptive variable filtering window value to solve the problem of excessive smoothing in traditional filtering algorithms. At the same time, this algorithm can filter and smooth real-time data, thus improving the applicability of the algorithm.

[0029] 3. This invention proposes an energy management strategy for a dual hybrid energy storage system, which achieves a balance between power supply and demand for load demand while ensuring the reliability of planned wind and solar power generation. In addition, considering the state of state of state of state of dual hybrid energy storage, a dual hybrid energy storage exchange compensation power algorithm is designed to reduce the number of times the battery state of state of state of state of dual energy storage system exceeds the limit, increase the lifespan of energy storage, and improve the economic efficiency of the system.

[0030] 4. This invention applies model predictive control algorithms to the dual hybrid energy storage control strategy to achieve effective load tracking and reliable execution of plans. By predicting the behavior at future moments, it improves the response speed of the dual hybrid energy storage system and reduces system overshoot. Attached Figure Description

[0031] Figure 1 This is a structural diagram of the wind / solar / storage power supply system in an embodiment of the present invention.

[0032] Figure 2 This is a graph showing the prediction results of the wind power generation test set in an embodiment of the present invention.

[0033] Figure 3 The flowchart shows the improved adaptive moving average filtering algorithm in this embodiment of the invention.

[0034] Figure 4 This refers to the power fluctuation rate on a 1-minute timescale in this embodiment of the invention.

[0035] Figure 5 This refers to the power fluctuation rate over a 10-minute timescale in this embodiment of the invention.

[0036] Figure 6 This is a schematic diagram of wind power output compared to the improved adaptive moving average filtering algorithm and the traditional moving average filtering algorithm in an embodiment of the present invention.

[0037] Figure 7 This is a schematic diagram of a hybrid energy storage control model for a wind / storage power supply system in an embodiment of the present invention.

[0038] Figure 8 This is a flowchart of the dual-energy storage energy management and scheduling model in an embodiment of the present invention.

[0039] Figure 9 This is a graph showing the local load and grid-connected power within 24 hours in an embodiment of the present invention.

[0040] Figure 10 Dual Hybrid Energy Storage in this embodiment of the invention , , , Power distribution curve.

[0041] Figure 11 Dual Hybrid Energy Storage in this embodiment of the invention , , , The SOC curve.

[0042] Figure 12 The battery in this embodiment of the invention , Schematic diagram of power switching point and grid connection switching point. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0045] Please see Figures 1-12 This invention provides a wind-solar-storage energy management method based on improved adaptive filtering and MPC, aiming to solve the problems of asynchronous changes in system wind speed and load affecting the supply and demand balance of distributed generation, high local load power shortage rate, and insufficient stability of wind-solar-storage power supply system. The method includes the following:

[0046] I. Construction of Wind / Solar / Storage Grid-Connected Inverter Power Generation System

[0047] By constructing a wind / solar / storage grid-connected inverter power generation system, wind power modules, photovoltaic power modules, and dual hybrid energy storage modules are all connected in parallel to the DC bus, and then the power is supplied to the grid through a DC / AC inverter. This simplifies the energy flow path, facilitates subsequent accurate capture and compensation of power deviations, and provides hardware support for the overall energy management strategy.

[0048] Meanwhile, the local load is connected to the DC bus voltage. By controlling the dual hybrid energy storage module, the deviation between the planned power of wind and solar power generation and the power demand of the local load is compensated, thereby improving the stability of the grid-connected power of wind / solar / storage power supply.

[0049] The specific steps are as follows: Build a wind / solar / storage grid-connected inverter power generation system. The wind power generation system outputs wind power to the DC bus capacitor through an AC / DC converter module. The photovoltaic power generation system outputs photovoltaic power to the DC bus capacitor through a DC / DC converter module. The dual hybrid energy storage module is connected in parallel to the DC bus capacitor through a DC / DC converter. The local load is connected to the DC bus capacitor. The power grid is connected to the DC bus capacitor through an AC / DC converter.

[0050] like Figure 1 As shown, the wind / solar / storage power supply system includes a permanent magnet synchronous generator (PMSG), photovoltaic panels, supercapacitor SC1 and battery Bat1 in hybrid energy storage No. 1, supercapacitor SC2 and battery Bat2 in hybrid energy storage No. 2, a DC / DC converter, a DC / AC converter, a local load, and the power grid.

[0051] The PMSG operates in maximum power point tracking mode, generating electricity and transmitting it to the DC bus via an AC / DC converter. Due to the uncertainty and volatility of wind power generation, multiple forecasting algorithms are employed to predict wind power generation plans for the next day based on weather forecast data. An improved moving average filtering algorithm is used to reduce errors in the wind power plan reporting data. To ensure load balance, a hybrid energy storage system is used to compensate for power deviations between wind, solar, and energy storage systems.

[0052] Figure 1 middle This refers to the planned power output of wind power generation. This represents the actual power output of wind power generation. This refers to the planned power output of photovoltaic power generation. This represents the actual power output of photovoltaic power generation. For planned wind and solar power generation capacity, The actual power output of wind and solar power generation is equal to the power demand of local load. Grid-connected power .

[0053] Among them, the No. 1 hybrid energy storage target is to compensate for the planned power output of wind and solar power generation. Actual power generation of wind and solar power Deviation power between The hybrid energy storage includes the power of battery No. 1. State of charge Supercapacitor No. 1 power To ensure the accuracy of the plan; the given value model of the No. 1 hybrid energy storage system is shown in formula (1):

[0054] ;

[0055] Hybrid energy storage target No. 2: To compensate for local load demand. With wind and solar power planned power Deviation power between The hybrid energy storage includes the power of battery No. 2. State of charge Supercapacitor No. 2 power To ensure load power balance; the given value model of hybrid energy storage system No. 2 is shown in formula (2):

[0056] ;

[0057] Therefore, in the dual hybrid energy storage system, Hybrid Energy Storage No. 1 is used to ensure the reliability of the planned power output of wind and solar power generation; Hybrid Energy Storage No. 2 is used to ensure the balance between power supply and demand for load requirements; the deviation power setpoints are all distributed to the energy storage system through a low-pass filter; the energy management center can manage the setpoints according to the current working charge state of the dual hybrid energy storage system, providing flexible adjustment space for the subsequent implementation of the MPC energy management algorithm.

[0058] II. Implementation of the Improved Adaptive Moving Average Filtering Algorithm

[0059] In wind / solar / storage grid-connected inverter power generation systems, to address the local optima problem of the adaptive moving average filtering algorithm, an improved adaptive moving average filtering algorithm is proposed. This algorithm enables real-time processing of existing data while simultaneously filtering and smoothing out local and global optima. To achieve this goal, it is necessary to start with the selection of prediction algorithms and the definition of key indicators, and gradually build a complete filtering logic.

[0060] Traditional adaptive moving average filtering algorithms, in order to smooth out wind power fluctuations, first give an initial moving window, calculate the 1-minute and 10-minute fluctuation rates of wind power after filtering through the moving window. If both are satisfied, the moving window is reduced so that the filtered power can better track the wind power and reduce the burden on energy storage; if not both are satisfied, the moving window is increased so that the filtered power meets the grid-connected power fluctuation rate standard.

[0061] The traditional adaptive moving average filtering algorithm selects its filtering window value based on the global optimum, which can lead to over-smoothing on local datasets. Furthermore, it is often used for offline data and cannot filter and smooth real-time data.

[0062] To address the aforementioned shortcomings of the adaptive moving average filtering algorithm, this invention employs real-time data processing and a variable filtering window value to smooth out local and global optima. Simultaneously, a maximum moving average filtering window value is set. When this value is reached, the grid-connected power at the current moment is constrained by comparing it with the power at the previous moment, ensuring that its power fluctuation meets the national wind power grid-connected fluctuation standards for 1-minute and 10-minute periods. This invention improves system response speed by setting a maximum filtering window value, avoids excessively large filtering window values, and thus reduces computational load.

[0063] like Figure 2 As shown, the wind power output is predicted based on weather forecast data; this invention uses four prediction algorithms to predict the wind power output and compares the index results of wind power output under different prediction algorithms; including the maximum fluctuation rate and average fluctuation rate of wind power prediction, the particle swarm optimization BP neural network (PSO-BP) is selected as the prediction algorithm of this invention to obtain the basic data for subsequent calculations; wherein, formulas (3) and (4) are the calculation methods for average power fluctuation rate and maximum power fluctuation rate (used to quantify the degree of fluctuation of predicted power and provide an objective basis for algorithm selection):

[0064] ;

[0065] ;

[0066] In the formula: for The output power of the wind farm at any given time; for The output power of the wind farm at any given time; This represents the actual operating capacity of the wind farm. This represents the number of samples to be predicted.

[0067] The evaluation indicators for wind power and photovoltaic forecast data include mean absolute error (MAE) and root mean square error (RMSE). The specific formulas are (5) and (6):

[0068] ;

[0069] ;

[0070] In the formula: for The actual power during that period; for Projected output power for the period.

[0071] like Figure 3 As shown, the traditional adaptive moving average filtering algorithm selects its window value based on the global optimal condition, which leads to over-smoothing when processing local datasets. Furthermore, this algorithm can only be applied to offline data and cannot filter and smooth real-time data. Therefore, this invention proposes an improved adaptive moving average filtering algorithm for predicting wind power.

[0072] The volatility calculations at the 1-minute and 10-minute scales are shown in formula (7):

[0073] ;

[0074] in, for The output power of wind power and photovoltaic power at all times; Rated power for wind and solar power; Indicates the first time Volatility at different scales; .

[0075] This invention proposes an improved adaptive moving average filtering algorithm, which processes existing data in real time and achieves data filtering to smooth out local and global optima while meeting national wind power grid connection volatility standards. The specific steps are as follows:

[0076] Step 1: Set the current sampling window value for wind power and maximum window value and the maximum output power of wind power Proceed to step 2;

[0077] Step 2: Determine the current window value If the condition is met, proceed to step 3; otherwise, proceed to step 9.

[0078] Step 3: Calculate based on the current wind power and load power consumption. Grid-connected power at all times Proceed to step 4;

[0079] Step 4: Calculate whether the grid connection volatility and its corresponding indicators are satisfied under the 1-minute and 10-minute time scales according to formula (7); if satisfied, proceed to steps 5 and 6; otherwise, proceed to step 8.

[0080] Step 5: Output the current... The current filtering algorithm ends when the grid-connected power at the specified time and the fluctuation rates at the current 1-minute and 10-minute intervals are calculated.

[0081] Step 6: Set the current window value Proceed to step 7;

[0082] Step 7: End of this moment Then proceed to step 2 to begin the filtering algorithm for the next time step;

[0083] Step 8: Since the current window value filtering does not meet the volatility requirements, increase the current window value to... Then proceed to step 2;

[0084] Step 9: To improve system response speed and reduce computational load, set the maximum window size. When the current window value is greater than the maximum allowable filter window value, filter threshold processing is performed based on the national wind power grid connection standard, and then proceed to step 10.

[0085] Step 10: Determine the current situation Time and The power at time , if the power at the current time Greater than the power of the previous moment If the value is given, then the current power is taken as... If yes, proceed to step 4; otherwise, proceed to step 11.

[0086] Step 11: If the power at the current moment Less than the power of the previous moment If the value is given, then the current power is taken as... Then proceed to step 4.

[0087] like Figure 4 and Figure 5 As shown, the improved adaptive moving average filtering algorithm used can not only obtain the grid-connected power at the current moment in real time. The size of the window can also avoid the problem of excessively large window values ​​affecting the system's calculation speed; the volatility at each current moment is calculated, and if it meets the national standard, the current volatility and grid-connected power are directly output; if it does not meet the national grid-connection standard, the filter window is increased, and the filter is slid towards historical data, so a suitable window value can always be found. By setting the maximum filter window value, the system response speed can be improved and the amount of computation can be reduced.

[0088] like Figure 6 As shown, the AMAF algorithm can effectively reduce the fluctuation of predicted power.

[0089] III. Implementation of a Dual Hybrid Energy Storage Energy Management Algorithm Based on Model Predictive Control (MPC)

[0090] For the dual-hybrid energy storage control strategy, this invention designs a dual-hybrid energy storage energy management algorithm based on model predictive control. This algorithm separately controls the power deviation compensation targets of the first and second hybrid energy storage units, achieving accurate wind power plan reporting while ensuring the stability of local load power supply. Simultaneously, considering the SOC state of the dual hybrid energy storage units, it performs exchange compensation for energy deviations, reducing the number of SOC exceedances and improving the lifespan of the batteries and supercapacitors in the hybrid energy storage system. The specific steps are as follows:

[0091] First, this invention designs a dual hybrid energy storage energy management algorithm based on model predictive control, first considering the control objective of the dual hybrid energy storage system;

[0092] Among them, the No. 1 hybrid energy storage target is to compensate for the planned power output of wind and solar power generation. Actual power generation of wind and solar power Deviation power between To ensure the accuracy of the plan;

[0093] Hybrid energy storage target No. 2: To compensate for local load demand. With wind and solar power planned power Deviation power between This ensures a balanced power supply to the load.

[0094] Secondly, this invention manages the energy of hybrid energy storage systems 1 and 2. Based on the State of Charge (SOC) of the dual hybrid energy storage systems, it optimizes the charging and discharging states of the batteries and supercapacitors, ensuring that the SOC of both systems is constrained within 0.2 to 0.8, thus preventing batteries from exceeding their limits and increasing their lifespan. It also controls the power exchange point between batteries 1 and 2, allowing them to exchange compensating power to reduce the number of times the dual energy storage system exceeds its SOC limit. Finally, it controls the grid connection point in the wind / solar / storage grid-connected system, ensuring battery throughput power through grid connection. and It operates stably at all times.

[0095] Then, consider specific dual-hybrid energy storage control strategies, based on the actual power output of wind and solar power generation. With wind and solar power planned power Calculate the power compensation amount for the No. 1 hybrid energy storage deviation. and local load demand power With wind and solar power planned power Calculate the power compensation amount for the deviation of hybrid energy storage No. 2. .

[0096] at last, and The power setpoints of the two systems are filtered by a low-pass filter to distribute the hybrid energy storage power to the hybrid energy storage system; the energy management center manages the setpoints according to the state of charge (SOC) of the dual hybrid system; model predictive control is applied to the control of the DC / DC converter to improve the response speed of the hybrid energy storage system and reduce the system overshoot.

[0097] like Figure 7 As shown, the planned power of the wind power system Combined wind and solar power generation Deviation power between After passing through a low-pass filter (LPF), the reference power of the battery in the hybrid energy storage is generated. and supercapacitor reference power Then, the PI controller generates the corresponding current reference value. and By using model predictive control strategies, the drive signals of the DC / DC converter switching transistors are obtained, thereby controlling the energy throughput of hybrid energy storage, so as to achieve effective load tracking and reliable execution of scheduling plans.

[0098] like Figure 8 As shown, by implementing energy management for hybrid energy storage systems No. 1 and No. 2 in this system, hybrid energy storage unit No. 1 is used to compensate for planned wind and solar power generation. Actual power generation of wind and solar power Deviation power between To ensure the accuracy of wind and solar power generation plans; the No. 2 hybrid energy storage unit will be used to compensate for local load demand. With wind and solar power planned power Deviation power between To improve the local load power supply balance; among which The number of times power is scheduled within a day, taking a 1-minute time scale as an example. .

[0099] In addition, battery number 1 State of charge and No. 2 storage battery State of charge The specific steps are as follows:

[0100] Step 1: Set the initial state of charge of the hybrid energy storage to , , Time begins; Proceed to step 2;

[0101] Step 2: Read time , State of charge , ,calculate time and The amount of charge and discharge the battery needs and Size, according to , Based on the size, proceed to step 3;

[0102] Step 3: If satisfied and Proceed to step 4; if the conditions are met... and Proceed to step 6; if the conditions are met... and Proceed to step 8; otherwise, proceed to step 10.

[0103] Step 4: Determine After execution, is the condition satisfied? or If satisfied, then it will occur. or If the limit is exceeded, proceed to step 5; if the condition is not met, then... , Charge separately , ;renew Moment , Proceed to step 11;

[0104] Step 5: Turn on the grid connection switch. and The power deviation is compensated by the grid, retaining the previous time step. , The state of charge is determined, proceeding to step 11;

[0105] Step 6: Determine After execution, is the condition satisfied? or If not satisfied, then , Charge separately , ;renew Moment , Proceed to step 11; if satisfied, then , After swapping the charging capacities, proceed to step 7;

[0106] Step 7: Determine if the conditions are met after the swap. or If satisfied, then it will occur. or If the limit is exceeded, proceed to step 5; if the condition is not met, then... , Charge separately , ;renew Moment , Proceed to step 11;

[0107] Step 8: Determine After execution, is the condition satisfied? or If not satisfied, then , Charge separately , ;renew Moment , Proceed to step 11; if satisfied, then , After swapping the charging capacities, proceed to step 9;

[0108] Step 9: Determine if the conditions are met after the swap. or If satisfied, then it will occur. or If the limit is exceeded, proceed to step 5; if the condition is not met, then... , Charge separately , ;renew Moment , Proceed to step 11;

[0109] Step 10: Determine After execution, is the condition satisfied? or If satisfied, then it will occur. or If the limit is exceeded, proceed to step 5; if the condition is not met, then... , Charge separately , ;renew Moment , Proceed to step 11;

[0110] Step 11: Determine If the condition is met, proceed to step 2; otherwise, end the process.

[0111] like Figure 9As shown, the improved adaptive moving average filtering algorithm is incorporated into a model predictive control simulation verification platform for wind / solar / storage power supply load tracking using the Function module in Simulink, and the real-time scheduling operation results within one day are presented; from Figure 9 It can be seen that the wind power output power The actual power output of photovoltaic power generation changes in real time with wind speed. The power absorbed by the power grid is Local load consumes energy At midday, photovoltaic power generation compensated for wind power generation, achieving wind-solar complementary power generation and reducing the amount of electricity dispatched from the grid to local loads.

[0112] like Figure 10 As shown, under the hybrid energy storage dispatch method, the high-frequency part of the deviation power is absorbed by the supercapacitor, and the low-frequency power part is dispatched by the battery, so as to maintain the actual power generation of wind and solar power tracking the planned power, and the planned power generation of wind and solar power ensures the power demand of local load; when the hybrid energy storage exceeds the limit, the grid connection function of the system is activated, and the grid provides energy to the load.

[0113] like Figure 11 As shown, the SOC of both battery 1 and battery 2 is constrained to within 0.2 to 0.8, which prevents the batteries from exceeding the limit and thus increases their lifespan.

[0114] like Figure 12 As shown, the number of times the dual energy storage system exceeds its state of charge (SOC) limit is reduced by exchanging compensation power between battery 1 and battery 2, and the throughput power of the dual hybrid energy storage system is guaranteed by switching into the grid. , To ensure the stable operation of the wind / solar / storage grid-connected system at all times.

[0115] It should be noted that, in this invention, although the specification describes the embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A wind-solar-storage energy management method based on improved adaptive filtering and MPC, characterized in that, Includes the following steps: (1) Build a wind-solar-storage grid-connected inverter system, and connect the wind power generation module, photovoltaic power generation module and dual hybrid energy storage module in parallel to the DC bus. The DC bus is connected to the power grid and local load respectively through converters; (2) An improved adaptive moving average filtering algorithm is used to perform real-time filtering on wind and solar power generation data to achieve local and global optimization of data filtering and smoothing. (3) Design a dual hybrid energy storage energy management algorithm based on model predictive control (MPC), which compensates for different types of power deviations through the dual hybrid energy storage modules, constrains the state of charge (SOC) of the dual hybrid energy storage modules within a preset range, and controls the power exchange and grid connection of the dual hybrid energy storage modules.

2. The wind-solar-storage energy management method based on improved adaptive filtering and MPC according to claim 1, characterized in that, In step (1), the construction of the wind-solar-storage grid-connected inverter system specifically includes: The wind power generation module is connected to the DC bus via an AC / DC converter, the photovoltaic power generation module is connected to the DC bus via a DC / DC converter, the dual hybrid energy storage module is connected in parallel to the DC bus via a DC / DC converter, the local load is directly connected to the DC bus, and the power grid is connected to the DC bus via an AC / DC converter.

3. The wind-solar-storage energy management method based on improved adaptive filtering and MPC according to claim 1, characterized in that, The dual hybrid energy storage module includes a first hybrid energy storage unit and a second hybrid energy storage unit, both of which are composed of batteries and supercapacitors.

4. The wind-solar-storage energy management method based on improved adaptive filtering and MPC according to claim 1, characterized in that, In step (2), the improved adaptive moving average filtering algorithm processes the real-time collected wind and solar power generation data and load power consumption data by setting an adaptive variable filtering window value and a maximum filtering window value, so that the filtered grid-connected power meets the grid-connected power fluctuation rate standard under the preset time scale.

5. The wind-solar-storage energy management method based on improved adaptive filtering and MPC according to claim 4, characterized in that, The implementation process of the improved adaptive moving average filtering algorithm includes: If the grid-connected power corresponding to the current filter window value meets the volatility standard, then the filter window value is maintained or adjusted. If the condition is not met, increase the filter window size; If the current filter window value reaches the maximum filter window value, the grid-connected power is made to meet the volatility standard by constraining the difference between the grid-connected power at the current moment and the grid-connected power at the previous moment.

6. The wind-solar-storage energy management method based on improved adaptive filtering and MPC according to claim 3, characterized in that, In step (3), the different types of power deviations include: The first deviation power between the planned power of wind and solar power generation and the actual power of wind and solar power generation, and the second deviation power between the local load demand power and the planned power of wind and solar power generation; The first hybrid energy storage unit is used to compensate for the first deviation power, and the second hybrid energy storage unit is used to compensate for the second deviation power.

7. The wind-solar-storage energy management method based on improved adaptive filtering and MPC according to claim 6, characterized in that, The first and second deviation powers are both distributed to the batteries and supercapacitors of the corresponding hybrid energy storage units via low-pass filters. The supercapacitors absorb the high-frequency power components, while the batteries regulate the low-frequency power components.

8. The wind-solar-storage energy management method based on improved adaptive filtering and MPC according to claim 1, characterized in that, In step (3), the preset range is 0.2-0.

8. The dual hybrid energy storage energy management algorithm optimizes the charging and discharging state of the battery and supercapacitor to avoid exceeding the SOC limit.

9. The wind-solar-storage energy management method based on improved adaptive filtering and MPC according to claim 1, characterized in that, In step (3), the dual-hybrid energy storage energy management algorithm further includes: Based on the SOC of the dual hybrid energy storage module, the power exchange between the first hybrid energy storage unit and the second hybrid energy storage unit is controlled; when the SOC is about to exceed the limit or cannot be avoided by power exchange, the grid is controlled to switch in to compensate for the deviation power.

10. The wind-solar-storage energy management method based on improved adaptive filtering and MPC according to claim 1, characterized in that, In step (3), the MPC is applied to the DC / DC converter control between the DC bus and the dual hybrid energy storage module.

Images