Microgrid energy optimization configuration method under hierarchical droop control strategy of direct current microgrid

Abstract

This invention discloses a method for optimizing energy allocation in a DC microgrid under a hierarchical droop control strategy, belonging to the field of power system operation and control technology. The method includes: constructing a rolling optimization model for the DC microgrid with the goal of minimizing system operating costs, and establishing minute-level optimal economic operating benchmarks for each unit in the microgrid in real time; constructing a hierarchical droop control strategy, determining the participation status and response mode of energy storage, charging piles, and the power grid in different regions by dividing the operating areas corresponding to the degree of bus voltage deviation; establishing an integrated scheduling-control coordination mechanism, quantifying the reserve demand of the underlying dynamic droop control for transient power, and embedding this reserve demand as a robust constraint into the upper-level rolling optimization model. This invention solves the problem of the separation between economic objectives and voltage safety objectives in traditional microgrid energy management, providing an effective method for achieving safe, economical, and grid-friendly operation of high-proportion renewable energy DC microgrids.

Description

Technical Field

[0001] This invention belongs to the field of power system operation and control technology, and particularly relates to a method for optimizing energy allocation of a DC microgrid under a hierarchical droop control strategy. Background Technology

[0002] With the increasing penetration of impactful and fluctuating resources such as distributed photovoltaic power and electric vehicle charging stations in DC microgrids, the uncertainty and power volatility of the system have significantly increased. Traditional microgrid energy management systems mainly adopt day-ahead or intraday economic dispatch methods based on prediction, with the core objective of optimizing operating costs, often ignoring the physical impact of minute-level power imbalances on bus voltage stability.

[0003] In actual operation, instantaneous load startup, rapid changes in renewable energy output, or prediction errors can all cause real-time power imbalances, impacting the DC bus voltage. Existing master-slave control strategies rely excessively on a single master converter for voltage support; when faced with massive load surges exceeding its regulation capacity, they are highly susceptible to voltage collapse, resulting in poor system robustness. While traditional fixed-coefficient droop control can achieve multi-power source coordinated voltage stabilization, it lacks flexibility and cannot differentiate between disturbance magnitudes, leading to frequent operation of energy storage and other devices involved in droop control. This increases unnecessary operating losses and accelerates equipment aging. Furthermore, both strategies may frequently transmit internal power fluctuations to the public grid, affecting power quality and grid stability.

[0004] Therefore, how to deeply integrate the economic dispatch at the upper level with the voltage control at the lower level, and design an advanced control strategy that can ensure high economic efficiency, effectively resist impact loads, improve voltage safety robustness, and actively smooth internal fluctuations and reduce impact on the power grid, has become a key technical problem that urgently needs to be solved to restrict the safe, economical and stable operation of DC microgrids. Summary of the Invention

[0005] To address the aforementioned shortcomings in existing technologies, this invention provides a microgrid energy optimization configuration method under a hierarchical droop control strategy for DC microgrids. This method not only ensures voltage stability within the DC microgrid but also maximizes the rapid smoothing of new energy fluctuations and local absorption of load impacts, thereby significantly reducing power exchange fluctuations with the main grid and improving the performance of the DC microgrid as a grid-friendly unit.

[0006] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows: a microgrid energy optimization configuration method under a hierarchical droop control strategy for DC microgrids, comprising the following steps:

[0007] S1: With the goal of minimizing system operating costs, a rolling optimization model for DC microgrids is constructed to determine the optimal economic operating benchmark for each unit in the microgrid at the minute level in real time.

[0008] S2: Construct a hierarchical droop control strategy. By dividing the operating areas corresponding to the degree of bus voltage deviation, determine the participation status and response mode of energy storage, charging piles and power grid in different areas, and realize time-sharing and hierarchical coordinated response to power disturbances.

[0009] S3: Establish an integrated scheduling-control coordination mechanism, quantify the reserve requirements of the underlying dynamic droop control for transient power, and embed this reserve requirement as a robust constraint into the upper-level rolling optimization model to ensure the physical feasibility of the economic operating benchmark and voltage safety.

[0010] Furthermore, the DC microgrid rolling optimization model adopts a model predictive control framework, and its objective function is:

[0011] ;

[0012] ;

[0013] ;

[0014] ;

[0015] in, This represents the total operating cost within the predicted time domain. This represents the total number of time periods within the prediction time domain. Indicates the duration of each time period. This represents the cost of electricity purchased per unit of time. Indicates the operating cost of photovoltaic systems. This indicates the cost of operating energy storage. Indicates time period, This represents the cost of purchasing electricity per unit of electricity. Indicates the power transmitted by the power grid. This indicates the cost per unit of electricity generated by photovoltaic power generation. Indicates photovoltaic power. This represents the cost per unit of power generated or absorbed by energy storage. , These represent the energy storage charging and discharging power, respectively.

[0016] The constraints include: power balance constraints, grid interaction power constraints, photovoltaic output constraints, energy storage operation constraints, energy storage charging and discharging constraints, air conditioning constraints, and charging pile constraints.

[0017] Furthermore, the layered droop control strategy includes the following steps:

[0018] S21: Set the voltage dead zone and divide the operating state of the bus voltage into a stable zone and an offset zone:

[0019] When the percentage of bus voltage deviation is less than or equal to the voltage dead zone, the operating state is in the stable zone;

[0020] When the percentage of bus voltage deviation is greater than the voltage dead zone, the operating state is in the offset zone;

[0021] S22: Determine the participation status and response methods of energy storage, charging piles, and the power grid in different regions:

[0022] When the operating state is in the stable region:

[0023] Only energy storage and charging piles participate in droop control, and the effective droop coefficient of the power grid is set to infinity; the actual output of each unit of energy storage and charging piles... for:

[0024] ;

[0025] in, This represents the power reference value issued by the upper-level scheduler. This indicates the corrective power generated by the droop control;

[0026] ;

[0027] ;

[0028] in, and These represent the corrected power for energy storage and charging piles, respectively. and These represent the droop coefficients for energy storage and charging piles, respectively. Indicates the rated voltage of the DC bus. Indicates the actual voltage of the DC bus;

[0029] When the running state is in the offset region:

[0030] The power grid, energy storage, and charging stations all participate in droop control, and the power grid's correction power is:

[0031] ;

[0032] in, Indicates the corrected power of the power grid. This represents the droop factor of the power grid;

[0033] The droop coefficients of the power grid, energy storage, and charging piles are adaptively adjusted based on the bus voltage deviation.

[0034] ;

[0035] ;

[0036] in, This represents the dynamically calculated adaptive droop coefficient. This indicates the foundation sagging coefficient of the equipment. Represents the adaptive gain coefficient. This represents the percentage of bus voltage deviation.

[0037] Furthermore, when the operating state is in the offset zone and certain conditions are met, the energy storage will temporarily withdraw from droop control and hand over its voltage stabilization task to the grid and charging piles:

[0038] Real-time recording of energy storage units in the past time window Actual output within And calculate its average output magnitude. :

[0039] ;

[0040] in, For integration variables;

[0041] In discrete-time systems, the above equation is approximated as:

[0042] ;

[0043] in, Indicates the discrete sampling time. This indicates the number of sampling points included in the time window. This represents the index for summing the sample points within the sliding window;

[0044] When the operating state is in the offset zone, and the average output of energy storage is... If the preset threshold is exceeded, the exit mechanism is triggered, and the energy storage droop coefficient is adjusted. If immediately set to infinity, it will no longer respond to voltage changes;

[0045] When the actual voltage of the DC bus Once it returns to and stabilizes in the stable region, the energy storage exit state is automatically lifted, and its droop factor is restored to the base droop factor, ready to respond to the next voltage fluctuation.

[0046] Furthermore, the integrated scheduling-control mechanism reserves a reserve margin for devices participating in droop control during the optimized scheduling phase:

[0047] Dynamic reserve capacity of energy storage Defined as:

[0048] ;

[0049] in, This indicates the preset threshold for the average output of energy storage;

[0050] In each optimization period of the DC microgrid rolling optimization model, the optimization variables for energy storage charging and discharging power are... and The modified power upper and lower limit constraints must be met:

[0051] ;

[0052] in, , These represent the maximum charging and discharging power of the energy storage, respectively. , These represent 0-1 variables representing the charging and discharging states of energy storage, respectively.

[0053] The beneficial effects of this invention are:

[0054] (1) This invention introduces a hierarchical droop control strategy, transforming the traditional single, rigid voltage regulation mode into a multi-resource, flexible, collaborative autonomous mode, significantly improving the voltage stability and operational robustness of the DC microgrid. When facing unforeseen major disturbances such as sudden load changes and fluctuations in new energy output, this strategy can quickly mobilize multiple resources, including energy storage, adjustable loads, and the power grid, to share the power impact. Compared to the traditional master-slave control method, which is prone to voltage collapse after the master controller reaches its power limit, this invention can effectively avoid system instability, ensure the microgrid's survivability under extreme conditions, and greatly enhance the system's safety and reliability.

[0055] (2) This invention, by proactively embedding the stability requirements of the lower-level control into the upper-level economic optimization in the form of reserved dynamic backup in the scheduling layer model, achieves a high degree of synergy and balance between system operation safety and economy. This method pursues the optimal total daily operating cost of the system while ensuring the safety of the lower-level voltage. Compared with traditional droop control, which sacrifices some economy due to frequent energy storage operations, or master-slave control, which ignores potential safety risks in pursuit of economy, this invention achieves a better balance between the two, ensuring both voltage quality and maximizing the overall operational efficiency of the DC microgrid.

[0056] (3) This invention reduces the frequency and magnitude of DC microgrids seeking power support from the main grid due to internal disturbances by prioritizing the use of energy storage and charging piles to smooth out fluctuations within the voltage dead zone, thereby enhancing the microgrid's friendly interaction with the main grid. This not only reduces the impact on the grid but also improves the microgrid's independent operation and self-regulation capabilities, making it a more stable and controllable "friendly" unit during grid-connected operation, providing strong support for the stable operation of the grid under future high-proportion renewable energy access. In addition, the dynamic energy storage exit mechanism avoids long-term deep discharge of energy storage, effectively reducing equipment operating losses and extending its service life. Attached Figure Description

[0057] Figure 1 This is a flowchart of the microgrid energy optimization configuration method under the DC microgrid hierarchical droop control strategy of the present invention.

[0058] Figure 2 This is a diagram of the DC microgrid structure of the present invention.

[0059] Figure 3 This is a graph showing the 24-hour power prediction values ​​for photovoltaic, air conditioning, and conventional loads according to the present invention.

[0060] Figure 4 This invention provides a comparison curve of bus voltage stability under different control strategies when a DC microgrid encounters a sudden load impact, as shown in the embodiment of the invention.

[0061] Figure 5 A power comparison curve of grid transformer, energy storage and charging pile under different control strategies provided in the embodiments of the present invention.

[0062] Figure 6 A comparison chart of the energy storage SOC change trajectory under different control strategies provided in the embodiments of the present invention.

[0063] Figure 7 This is a daily operating curve of each device in a microgrid using the layered droop control strategy of this invention.

[0064] Figure 8 This is a daily operating curve of each device in a microgrid using the traditional master-slave control strategy of this invention.

[0065] Figure 9 This is a daily operating curve of each device in a microgrid using the traditional droop control strategy of this invention. Detailed Implementation

[0066] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0067] like Figure 1 As shown, a method for optimizing microgrid energy configuration under a hierarchical droop control strategy for DC microgrids includes the following steps:

[0068] S1: With the goal of minimizing system operating costs, a rolling optimization model for DC microgrids is constructed to determine the optimal economic operating benchmark for each unit in the microgrid at the minute level in real time.

[0069] This step, as the core of the energy management layer, aims to plan the most economically optimal power scheduling strategy for a longer timescale (24 hours) while satisfying all equipment physical constraints and operational tasks. The model employs a Model Predictive Control (MPC) framework, executing in a rolling optimization manner every 30 minutes to revise the future scheduling plan based on the latest system status and forecast information.

[0070] First, the system collects input information, including but not limited to: predicted photovoltaic output for future periods, predicted conventional DC load, total daily electric vehicle charging load, and time-of-use electricity price information from the external power grid. Second, a mathematical model is constructed with the objective of minimizing the system's total daily operating cost. The objective function is specifically defined as the sum of all electricity purchase costs, photovoltaic operation and maintenance costs, and energy storage operation and maintenance costs over the entire prediction time domain. Its mathematical expression can be defined as:

[0071] ;

[0072] ;

[0073] ;

[0074] ;

[0075] in, This represents the total operating cost within the predicted time domain. This represents the total number of time periods within the prediction time domain. Indicates the duration of each time period. This represents the cost of electricity purchased per unit of time. Indicates the operating cost of photovoltaic systems. This indicates the cost of operating energy storage. Indicates time period, This represents the cost of purchasing electricity per unit of electricity. Indicates the power transmitted by the power grid. This indicates the cost per unit of electricity generated by photovoltaic power generation. Indicates photovoltaic power. This represents the cost per unit of power generated or absorbed by energy storage. , These represent the energy storage charging and discharging power, respectively.

[0076] A series of strict constraints must be satisfied during the optimization process, including:

[0077] Power balance constraints:

[0078] ;

[0079] In the formula, , and These represent the power of the charging pile, the air conditioner, and the conventional load, respectively.

[0080] Power grid interaction constraints:

[0081] ;

[0082] In the formula, This indicates the maximum power purchase capacity of the power grid transformer;

[0083] Photovoltaic output constraints:

[0084] ;

[0085] In the formula, Indicates the maximum power of photovoltaic power;

[0086] Energy storage operation constraints:

[0087] ;

[0088] ;

[0089] In the formula, , Indicates the charging and discharging power of energy storage. , Indicates the maximum charging and discharging power of the energy storage; These represent 0-1 variables representing the charging and discharging states of energy storage, respectively.

[0090] This is a crucial collaborative step in this invention. When setting the upper limit of the charging and discharging power of the energy storage unit, its physical maximum value is not used; instead, a portion of reserve capacity is reserved. This reserve capacity does not participate in economic dispatch but is specifically used for droop control in step S2, providing power support to cope with sudden disturbances. During each optimization period of the MPC model, the energy storage charging and discharging power optimization variables must satisfy the modified upper and lower power limit constraints.

[0091] Energy storage charge and discharge limits:

[0092] ;

[0093] In the formula, Indicates energy storage The state of charge during a given period, i.e., the ratio of remaining energy storage capacity to total capacity. , These represent the maximum and minimum allowable states of charge for energy storage, respectively. Indicates the rated capacity of energy storage. , This indicates the charging and discharging efficiency of the energy storage cell.

[0094] Air conditioning constraints:

[0095] ;

[0096] In the formula, This indicates the maximum power of the air conditioner;

[0097] The relationship between air conditioner output and outdoor temperature:

[0098] ;

[0099] In the formula, This indicates the preset target indoor temperature. This represents the predicted outdoor ambient temperature. This indicates the cooling / heating power coefficient in summer and the heating power coefficient in winter. Air conditioner power varies with ambient temperature, allowing for different optimization strategies based on different ambient temperatures.

[0100] Charging station constraints:

[0101] ;

[0102] In the formula, , These represent the maximum charging and discharging power of the charging pile, respectively.

[0103] Discharge from charging stations not considered:

[0104] ;

[0105] ;

[0106] In the formula, This indicates the total amount of charging tasks that need to be completed throughout the day.

[0107] Finally, the above model is solved using the optimization solver gurobi. The model output is the optimal power plan for each future time period. This step only extracts the planned power value for the first time period, i.e., the 30 minutes starting from the current moment, as the economic operating benchmark for grid transformers, energy storage, and charging piles, denoted as . And then send it down to the physical control layer.

[0108] S2: Construct a hierarchical droop control strategy. By dividing the operating areas corresponding to the degree of bus voltage deviation, determine the participation status and response mode of energy storage, charging piles and power grid in different areas, and realize time-sharing and hierarchical coordinated response to power disturbances.

[0109] This step proposes a hierarchical droop control strategy that takes into account voltage dead zone, dynamic activation, and adaptive gain. By distinguishing the degree of voltage deviation, it achieves time-sharing and hierarchical coordinated response to power disturbances. Energy storage and charging piles prioritize handling minor fluctuations, while the power grid and energy storage work together to cope with major shocks, ensuring rapid and precise stability of the bus voltage.

[0110] This step, as the core of the physical stabilization layer, is responsible for handling real-time power imbalances caused by prediction errors and unforeseen load surges, maintaining the stability of the DC bus voltage. It receives the economic operating baseline from S1 and performs power fine-tuning based on real-time voltage deviations. This strategy comprises the following two core mechanisms:

[0111] 1) Graded response mechanism considering voltage dead zone:

[0112] To prevent critical equipment from responding frequently to minor, harmless voltage fluctuations, a voltage dead zone is set. Bus voltage The operating status is divided into two regions. Percentage of bus voltage deviation:

[0113] ;

[0114] In the formula, Indicates the rated voltage of the DC bus. This indicates the actual voltage of the DC bus.

[0115] Stable zone (within the dead zone): when At this time, the system is under slight disturbance. Only fast-responding energy storage units and charging piles with power regulation capabilities perform droop control to smooth out minor power fluctuations. The grid-side droop controller is inactive, and its effective droop coefficient... Setting the value to infinity avoids the frequent transmission of minor disturbances within the microgrid to the main grid, thus improving grid friendliness. Actual output of each unit. Determined by the following formula:

[0116] ;

[0117] in, This represents the power reference value issued by the upper-level scheduler. This indicates the corrective power generated by the droop control;

[0118] ;

[0119] ;

[0120] in, and These represent the corrected power for energy storage and charging piles, respectively. and These represent the droop coefficients for energy storage and charging piles, respectively. The negative sign indicates that the charging pile, acting as a load, responds in the opposite direction to the energy storage, which acts as a power source. Indicates the rated voltage of the DC bus. Indicates the actual voltage of the DC bus;

[0121] Offset zone (outside the dead zone): when When the system is in the offset region, it indicates that the system has encountered a significant power disturbance, requiring the activation of a stronger response mechanism. At this time, the droop controller on the grid side is immediately activated, working together with energy storage units and charging piles to participate in voltage regulation, forming a multi-unit collaborative support structure to ensure that the voltage can be quickly pulled back to the stable range. The grid correction power is calculated as follows:

[0122] ;

[0123] in, Indicates the corrected power of the power grid. This represents the droop factor of the power grid;

[0124] In a DC microgrid, the droop coefficients of the power grid, energy storage, and charging piles are adaptively adjusted based on the bus voltage deviation.

[0125] ;

[0126] ;

[0127] in, This represents the dynamically calculated adaptive droop coefficient. This indicates the foundation sagging coefficient of the equipment. This represents the adaptive gain coefficient, used to control the drastic change in the droop coefficient. This represents the percentage of bus voltage deviation.

[0128] The physical meaning of this formula is that when the voltage deviation is small, the droop coefficient is large, and the adjustment effect is relatively mild; when the voltage deviation increases, the droop coefficient will dynamically decrease, which means that the power response of the device to a unit voltage change will be more drastic and rapid. This adaptive mechanism enables the system to recover with the most appropriate "force" when facing disturbances of different levels, achieving a balance between stability, speed, and smoothness.

[0129] 2) Dynamic decommissioning mechanism for energy storage units

[0130] When the operating state is in the offset zone and certain conditions are met, the energy storage will temporarily withdraw from droop control and hand over its voltage stabilization task to the grid and charging piles:

[0131] Real-time recording of energy storage units in the past time window Actual output within And calculate its average output magnitude. :

[0132] ;

[0133] in, For integration variables;

[0134] In discrete-time systems, the above equation is approximated as:

[0135] ;

[0136] in, Indicates the discrete sampling time. This indicates the number of sampling points included in the time window. This represents the index for summing the sample points within the sliding window;

[0137] When the operating state is in the offset zone, and the average output of energy storage is... If the preset threshold is exceeded, the exit mechanism is triggered, and the energy storage droop coefficient is adjusted. If immediately set to infinity, it will no longer respond to voltage changes;

[0138] When the actual voltage of the DC bus Once it returns to and stabilizes in the stable region, the energy storage exit state is automatically lifted, and its droop factor is restored to the base droop factor, ready to respond to the next voltage fluctuation.

[0139] To prevent energy storage units from experiencing lifespan loss and excessive SOC drift due to continuous high-frequency, high-power voltage regulation, this invention designs a dynamic activation logic. When the bus voltage is outside the dead zone, and the energy storage unit continuously operates at a rate exceeding a preset threshold... After adjusting the average power for a period of time, the control system will determine that there is a persistent significant disturbance. At this time, the system will temporarily force the energy storage unit out of droop control, that is, reduce its droop coefficient. Set to infinity, only the upper-level power reference value will be applied. The regulation task is entirely delegated to the grid, which has stronger support capabilities. When the bus voltage recovers to within the dead zone, it indicates that the major disturbance has been eliminated, and the energy storage unit will automatically reset its state and re-enter standby mode, ready to respond to the next fluctuation. This mechanism realizes an intelligent switching of "energy storage leading the way during normal disturbances, and the grid providing backup during major disturbances".

[0140] S3: Establish an integrated scheduling-control coordination mechanism, quantify the reserve requirements of the underlying dynamic droop control for transient power, and embed this reserve requirement as a robust constraint into the upper-level rolling optimization model to ensure the physical feasibility of the economic operating benchmark and voltage safety.

[0141] The goal of this step is to feed back the underlying control physical behavior described in step S2 to the scheduling model in step S1 in a simplified form, thereby achieving a balance between top-level economics and bottom-level feasibility. The core of the collaborative mechanism is to reserve necessary reserve margins for devices participating in droop control (especially energy storage) during the optimized scheduling phase.

[0142] Dynamic reserve capacity Definition: Based on the requirements of the underlying droop control, define the power capacity that the energy storage unit must reserve for voltage regulation. This value should cover most of the power imbalance caused by prediction errors, and its magnitude is based on the exit threshold in S2. To set:

[0143] ;

[0144] in, This indicates the preset threshold for the average output of energy storage;

[0145] In each optimization period of the DC microgrid rolling optimization model, the optimization variables for energy storage charging and discharging power are... and The modified power upper and lower limit constraints must be met:

[0146] ;

[0147] in, , These represent the maximum charging and discharging power of the energy storage, respectively. , These represent 0-1 variables representing the charging and discharging states of energy storage, respectively.

[0148] Within each scheduling instruction execution cycle, minute-level simulations are used to accurately simulate the actual physical response of the microgrid. The core of this simulation is to solve for the quasi-steady-state equilibrium point of the system under given power reference values ​​and actual load disturbances.

[0149] Power balance equation: At any given moment, taking into account the dynamic responses of all components involved in the droop control unit, the total power of the system must be balanced, as shown in the following formula:

[0150] ;

[0151] In the formula, These represent the actual photovoltaic power, air conditioning power, and conventional load power at that moment, respectively. These represent the power reference values ​​for the power grid, energy storage, and charging piles issued by the upper-level optimized scheduling, respectively.

[0152] This mechanism ensures that while maintaining voltage stability, the system's average operating point over a long time scale is always anchored on the economically optimal trajectory, achieving a decoupling and unification of economy and stability.

[0153] This invention does not separate upper-level economic optimization from lower-level physical control, but rather achieves deep integration and synergy between the two through two key interfaces. On one hand, there is "feedforward" synergy from the optimization layer to the control layer: the economic operating benchmark calculated by S1... The output is fed forward to S2, and the actual output of each unit in S2 does not fluctuate around zero, but is finely adjusted around its own economic benchmark.

[0154] On the other hand, there is a "feedback" collaboration from the control layer to the optimization layer: S2's requirement for physical stability is fed back to the optimization model of S1 in the form of constraints. The "droop reserve constraint" mentioned in S1 is precisely the resource requirement that S2 proposes to the upper layer. When formulating economic plans, the optimization layer has already anticipated and satisfied the physical layer's requirement for reserve capacity, thus avoiding the risk of sacrificing system safety margins due to excessive pursuit of economy. This closed-loop collaboration ensures the feasibility of upper-level decisions and the effectiveness of lower-level controls.

[0155] In one embodiment of the present invention, the method of the present invention is compared with two benchmark methods by establishing a minute-level quasi-steady-state simulation environment for DC microgrids to verify its superiority.

[0156] To verify the effectiveness and superiority of the method described in this invention, a DC microgrid simulation model was built based on the MATLAB R2023a software platform, including grid transformers, photovoltaics, energy storage, charging piles, air conditioning, and conventional loads. The specific structure is as follows: Figure 2 As shown in Table 1, the equipment capacity and parameter settings are set in Table 1. The rolling optimization time interval for the optimization layer is 30 minutes, the total optimization time is 24 hours a day, and the physical simulation step size is 1 minute. The core parameters for the layered droop control of this invention are shown in Table 2.

[0157] Table 1 Parameters of DC Microgrid System

[0158]

[0159] Table 2 Parameters for Layered Sag Control

[0160]

[0161] This embodiment selects a typical summer day as the operating scenario, where the predicted power output of photovoltaic power, air conditioning, and conventional loads for a 48.5-hour period are as follows: Figure 3 As shown, the air conditioning load is calculated based on the above model. To simulate the randomness in the real world, ±8% random noise is superimposed on the predicted values ​​of photovoltaics and load (excluding charging piles) in the physical simulation layer.

[0162] The time-of-use electricity price is based on the local power grid sales price policy: 0.2795 yuan / kWh during off-peak hours (0:00-8:00), 0.7354 yuan / kWh during normal hours (8:00-10:00, 12:00-14:00, 19:00-24:00), and 1.2502 yuan / kWh during peak hours (10:00-12:00, 14:00-19:00).

[0163] Extreme disturbance event injection: To rigorously test the robustness of various control strategies under extreme conditions, at the 10th hour (600th minute) of the simulation, a sudden DC load surge of 300kW with a constant power for 10 minutes was artificially injected into the system. This event simulated high-risk scenarios such as the sudden start-up of large equipment during peak load periods or the simultaneous connection of multiple high-power charging piles.

[0164] The "layered droop control strategy" described in this invention is rigorously compared with the following two widely used control strategies in the industry, and ±3% of the rated voltage is set as the safe voltage fluctuation range:

[0165] Comparative Example 1: Traditional Master-Slave Control Strategy. Under this strategy, the power grid acts as the sole master control unit, responsible for balancing all power deficits or surpluses within the microgrid. Slave units such as energy storage and charging stations operate strictly according to the economic dispatch instructions issued by the optimization layer and do not participate in real-time voltage regulation. Because only the power grid participates in maintaining the bus voltage constant at its rated value... Normally, the bus voltage is a straight line, but in extreme cases, when the load disturbance exceeds the grid's regulation capacity, the voltage collapses instantly.

[0166] Comparative Example 2: Traditional droop control strategy. Under this strategy, the power grid, energy storage, and charging piles all participate in droop control to maintain voltage stability. All droop coefficients are fixed values ​​(0.5V / kW for the power grid, 0.1V / kW for energy storage, and 0.8V / kW for charging piles), and there are no advanced regulation mechanisms such as voltage dead zone, dynamic activation, or adaptive gain.

[0167] Through quantitative comparison and analysis of simulation results, and in conjunction with the appendix Figure 4 To be continued Figure 9 The technical advantages of this invention are thus clearly demonstrated.

[0168] I. Significant advantages in ensuring bus voltage safety

[0169] See Figure 4 The figure shows the dynamic response curves of the bus voltage under the three strategies when a 300kW load impact is applied at the 10th hour (horizontal axis 10h).

[0170] In the figure, the horizontal axis represents time in hours (h), ranging from 0 to 24; the vertical axis represents DC bus voltage in volts (V). The red curve represents the voltage trajectory of Comparative Example 1 (traditional master-slave control). As shown in the figure, when the surge occurs, the curve drops sharply vertically, with the voltage value instantly falling from the rated 750V to more than 3% of the voltage safety lower limit, indicating that the system has experienced voltage collapse.

[0171] The green curve represents the voltage trajectory of Comparative Example 2 (traditional droop control). At the impact point, the curve shows a significant dip, with its lowest point reaching 721V, below the voltage safety lower limit of 727.5V, indicating insufficient stability margin.

[0172] The blue curve represents the voltage trajectory of this invention. Under the same impact, this curve shows only a small and smooth dip, with the lowest voltage point maintained at 742.3V, slightly exceeding the 1% lower limit of the voltage dead zone, and the recovery speed is significantly faster than that of the green curve.

[0173] pass Figure 4 As can be seen from the direct comparison, the method of the present invention can suppress voltage fluctuations to a very small range when dealing with extreme disturbances, demonstrating excellent voltage robustness.

[0174] II. Excellent performance in balancing the economic efficiency of the system's operation throughout the day

[0175] Table 3 shows a comparison of the total operating costs of the three control strategies over a 24-hour simulation period. The cost of Comparative Example 1 is 7444.10 yuan, which can be considered the benchmark for optimal economic efficiency. Compared to Comparative Example 1, the cost of this invention is increased by 0.85%. This demonstrates that while achieving significant safety gains through measures such as reserving backups, the impact on system economy is kept within a negligible range, successfully achieving a high degree of synergy between safety and economy.

[0176] Table 3 Comparison of operating costs of three control strategies

[0177]

[0178] In addition, the energy storage SOC trajectory and the daily operating curves of each device in the microgrid under the three control strategies are as follows: Figures 6-9 As shown.

[0179] III. Outstanding Contributions to Enhancing Grid Friendliness

[0180] See Figure 5 This figure compares the power output of grid transformers, energy storage, and charging piles under three different strategies. The comparison of grid power purchases shows that:

[0181] The power curve of Comparative Example 1 shows frequent and violent fluctuations, filled with a large number of high-frequency spikes, indicating that the power grid needs to respond to all power imbalances within the microgrid at all times.

[0182] The fluctuation range of the grid power curve in Comparative Example 2 has slowed down, but the overall curve is still relatively rough, especially during periods of large changes in photovoltaic output;

[0183] The grid power curve of this invention exhibits the smoothest performance, appearing as a gentle step shape for most of the time. This clearly demonstrates that, due to the existence of the voltage dead zone mechanism, a large number of minute internal power disturbances are absorbed locally by energy storage units and are not transmitted to the grid.

[0184] Figure 5 The comparison powerfully demonstrates that the present invention can effectively isolate internal fluctuations, reduce the impact on the upstream power grid, and significantly improve the grid friendliness of the microgrid.

[0185] In summary, through rigorous simulation comparisons under typical summer days with superimposed extreme load impact scenarios, the microgrid energy optimization configuration method proposed in this invention, based on a "source-storage-charging" collaborative DC microgrid hierarchical droop control strategy, demonstrates significant technical advantages over traditional master-slave control and traditional droop control in both key performance dimensions: voltage stability and grid friendliness.

[0186] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of the invention.

Claims

1. A method for optimizing microgrid energy allocation under a hierarchical droop control strategy for DC microgrids, characterized in that, Includes the following steps: S1: With the goal of minimizing system operating costs, a rolling optimization model for DC microgrids is constructed to determine the optimal economic operating benchmark for each unit in the microgrid at the minute level in real time. S2: Construct a hierarchical droop control strategy. By dividing the operating areas corresponding to the degree of bus voltage deviation, determine the participation status and response mode of energy storage, charging piles and power grid in different areas, and realize time-sharing and hierarchical coordinated response to power disturbances. The layered droop control strategy includes the following steps: S21: Set the voltage dead zone and divide the operating state of the bus voltage into a stable zone and an offset zone: When the percentage of bus voltage deviation is less than or equal to the voltage dead zone, the operating state is in the stable zone; When the percentage of bus voltage deviation is greater than the voltage dead zone, the operating state is in the offset zone; S22: Determine the participation status and response methods of energy storage, charging piles, and the power grid in different regions: When the operating state is in the stable region: Only energy storage and charging piles participate in droop control, and the effective droop coefficient of the power grid is set to infinity; the actual output of each unit of energy storage and charging piles... for: ； in, This represents the power reference value issued by the upper-level scheduler. This indicates the corrective power generated by the droop control; ； ； in, and These represent the corrected power for energy storage and charging piles, respectively. and These represent the droop coefficients for energy storage and charging piles, respectively. Indicates the rated voltage of the DC bus. Indicates the actual voltage of the DC bus; When the running state is in the offset region: The power grid, energy storage, and charging stations all participate in droop control, and the power grid's correction power is: ； in, Indicates the corrected power of the power grid. This represents the droop factor of the power grid; The droop coefficients of the power grid, energy storage, and charging piles are adaptively adjusted based on the bus voltage deviation. ； ； in, This represents the dynamically calculated adaptive droop coefficient. This indicates the foundation sagging coefficient of the equipment. Represents the adaptive gain coefficient. Indicates the percentage of bus voltage deviation; S3: Establish an integrated scheduling-control coordination mechanism, quantify the reserve margin of the underlying dynamic droop control for transient power, and embed this reserve margin as a robust constraint into the upper-level rolling optimization model to ensure the physical feasibility of the economic operating benchmark and voltage safety.

2. The microgrid energy optimization configuration method under the hierarchical droop control strategy of DC microgrid according to claim 1, characterized in that, The DC microgrid rolling optimization model adopts a model predictive control framework, and its objective function is: ； ； ； ； in, This represents the total operating cost within the predicted time domain. This represents the total number of time periods within the prediction time domain. Indicates the duration of each time period. express Electricity purchase cost for a given period of time, express The operating cost of photovoltaic power during a given period express Energy storage operating costs during different time periods Indicates the cost of purchasing electricity. Indicates the operating cost of photovoltaic systems. This indicates the cost of operating energy storage. Indicates time period, This represents the cost of purchasing electricity per unit of electricity. Indicates the power transmitted by the power grid. This indicates the cost per unit of electricity generated by photovoltaic power generation. Indicates photovoltaic power. This represents the cost per unit of power generated or absorbed by energy storage. , These represent the energy storage charging and discharging power, respectively. The constraints include: power balance constraints, grid interaction power constraints, photovoltaic output constraints, energy storage operation constraints, energy storage charging and discharging constraints, air conditioning constraints, and charging pile constraints.

3. The microgrid energy optimization configuration method under the hierarchical droop control strategy of DC microgrid according to claim 1, characterized in that, When the operating state is in the offset zone and certain conditions are met, the energy storage will temporarily withdraw from droop control and hand over its voltage stabilization task to the grid and charging piles: Real-time recording of energy storage units in the past time window Actual output within And calculate its average output magnitude. : ； in, For integration variables; In discrete-time systems, the above equation is approximated as: ； in, Indicates the discrete sampling time. This indicates the number of sampling points included in the time window. This represents the index for summing the sample points within the sliding window; When the operating state is in the offset zone, and the average output of energy storage is... If the preset threshold is exceeded, the exit mechanism is triggered, and the energy storage droop coefficient is adjusted. If immediately set to infinity, it will no longer respond to voltage changes; When the actual voltage of the DC bus Once it returns to and stabilizes in the stable region, the energy storage exit state is automatically lifted, and its droop factor is restored to the base droop factor, ready to respond to the next voltage fluctuation.

4. The microgrid energy optimization configuration method under the hierarchical droop control strategy of DC microgrid according to claim 1, characterized in that, The integrated scheduling-control mechanism reserves a backup margin for equipment participating in droop control during the optimization scheduling phase: Energy storage reserve margin Defined as: ； in, This indicates the preset threshold for the average output of energy storage; In each optimization period of the DC microgrid rolling optimization model, the optimization variables for energy storage charging and discharging power are... and The modified power upper and lower limit constraints must be met: ； in, , These represent the maximum charging and discharging power of the energy storage, respectively. , These represent 0-1 variables representing the charging and discharging states of energy storage, respectively.

Images