Pulse load-containing microgrid operation control method and device

Abstract

The embodiment of the invention discloses a pulse load-containing micro-grid operation control method and device, and the scheme can comprise the steps: building a daily operation cost target function of a to-be-analyzed wind-solar-storage island micro-grid, the operation cost in the daily operation cost objective function comprises the operation cost of an energy storage system, the operation cost of a wind driven generator and the operation cost of photovoltaic power generation; modeling pulse load related constraints including pulse load characteristics, system dynamic frequency response characteristics, wind and light storage micro-grid dynamic frequency safety constraints and wind and light storage frequency control constraints; constructing an operation constraint model of the to-be-analyzed wind and light storage island microgrid, wherein the operation constraint model comprises power balance, Disflow power flow constraint, energy storage system constraint, photovoltaic constraint and fan constraint; and solving the operation constraint model by using a particle swarm algorithm to obtain an operation scheduling decision scheme of the to-be-analyzed wind and light storage island microgrid.

Description

Technical Field

[0001] The present application relates to the technical field of power system optimization and dispatching, and in particular to a method and device for controlling the operation of a microgrid containing pulse loads. Background Art

[0002] With the increasingly severe global energy situation and growing awareness of environmental protection, energy restructuring has become a key measure for countries around the world to address energy and environmental challenges. Against this backdrop, renewable energy sources such as wind and solar power, driven by their clean and sustainable nature, have seen large-scale development and utilization. Furthermore, microgrid systems, comprised of distributed energy resources, have gained widespread application in the energy sector due to their significant advantages, including low cost, environmental friendliness, and flexibility. In particular, microgrid systems can effectively utilize abundant renewable energy resources in remote areas, such as islands, mountainous areas, and other areas far from the main power grid, meeting local electricity needs and alleviating energy supply pressures.

[0003] Economic dispatch is a crucial task in the operation and management of isolated microgrids. Its core goal is to determine the optimal dispatch plan for the units and minimize power generation costs while strictly meeting various operational constraints. Effective economic dispatch not only significantly improves energy utilization efficiency and reduces microgrid operating costs, but also further enhances the economic and stability of microgrid operations, which is of great significance to the sustainable development of microgrids.

[0004] However, with the large-scale integration of renewable energy sources such as photovoltaics and wind turbines in isolated microgrids, and the growing demand for frequency response to various pulsed loads, the economic operation of isolated microgrids faces unprecedented challenges. For one thing, the power generated by photovoltaics and wind turbines is highly random and volatile, significantly affected by natural factors such as weather and seasons. For example, instantaneous changes in sunlight intensity can cause sharp fluctuations in photovoltaic power generation, while unstable wind speeds can make wind turbine power generation unpredictable. This unstable power generation characteristic poses significant challenges to power balance control in microgrids and increases the complexity of microgrid operation and scheduling.

[0005] On the other hand, pulse loads have the unique characteristics of high peak power and low average power. When pulse loads are connected to a microgrid, they generate strong current surges, causing significant fluctuations in system frequency, increasing system energy consumption, and potentially even triggering power outage protection and microgrid frequency collapse, seriously threatening the safe and stable operation of the microgrid. For example, the instantaneous startup and shutdown of some industrial equipment can generate pulse loads, significantly impacting the stability of the microgrid.

[0006] Furthermore, existing operational scheduling methods for isolated microgrids have numerous shortcomings when it comes to addressing the randomness of renewable energy and the impact of pulse loads. Traditional scheduling methods often fail to accurately predict renewable energy generation and effectively address sudden power fluctuations caused by pulse loads, resulting in high microgrid operating costs and low resource utilization. Therefore, there is an urgent need for an innovative operational scheduling decision-making method that fully considers the primary frequency regulation capabilities of wind, solar, and energy storage, effectively addresses the impact of pulse loads, optimizes microgrid operational scheduling, and improves the economic and safety of microgrid operations. Summary of the Invention

[0007] The embodiments of this specification provide a method and device for controlling the operation of a microgrid containing pulse loads to solve at least one of the technical problems mentioned above.

[0008] To solve the above technical problems, the embodiments of this specification are implemented as follows:

[0009] According to a first aspect of an embodiment of the present invention, there is provided a method for controlling operation of a microgrid containing pulse loads, comprising:

[0010] Construct the wind power generation characteristic model, photovoltaic power generation characteristic model and energy storage model in the wind-solar-storage island microgrid to be analyzed;

[0011] Establishing a daily operating cost objective function for the wind-solar-storage island microgrid to be analyzed, wherein the operating costs in the daily operating cost objective function include the operating costs of the energy storage system, the wind turbine generator, and the photovoltaic power generation;

[0012] Modeling of pulse load-related constraints, including pulse load characteristics, system dynamic frequency response characteristics, wind-solar-storage microgrid dynamic frequency security constraints, and wind-solar-storage microgrid frequency control constraints;

[0013] Constructing an operation constraint model for the wind-solar-storage island microgrid to be analyzed, including power balance, disflow constraints, energy storage system constraints, photovoltaic constraints, and wind turbine constraints;

[0014] The particle swarm algorithm is used to solve the operation constraint model to obtain the operation scheduling decision plan of the wind-solar-storage island microgrid to be analyzed.

[0015] In some optional implementations, the wind speed in the wind power generation characteristic model satisfies the Weibull distribution, and the probability distribution function f(v) of the wind speed is expressed as:

[0016]

[0017] Among them, the symbol c represents the scale parameter of the Weibull distribution function, the symbol k represents the shape parameter of the Weibull distribution function, and the symbol v represents the wind speed in the actual environment;

[0018] The wind speed power characteristic curve model of the wind turbine WT in the wind power generation characteristic model is expressed as follows:

[0019]

[0020] Among them, the symbol P WT Indicates the output power of the wind turbine, symbol P′ r Indicates the rated power of the wind turbine, symbol v ci Indicates the cut-in wind speed of the wind turbine, symbol v′ r Indicates the rated wind speed of the wind turbine, symbol v′ co Indicates the cut-out wind speed of the wind turbine, and the symbols a′, b′, c′ and d′ represent different wind speed parameters respectively;

[0021] The output power model of the photovoltaic cell in the photovoltaic power generation characteristic model is expressed by the following formula:

[0022]

[0023] Among them, the symbol P′ pv Represents the output active power of the photovoltaic cell, symbol R′ pv Indicates the photovoltaic output power under standard test conditions, symbol q′ pv Indicates the derating factor of photovoltaic power, symbol I T Indicates the actual solar radiation intensity, symbol I STC Indicates the solar radiation intensity under standard test conditions, symbol α p Indicates the temperature coefficient of the photovoltaic panel, symbol T c Indicates the photovoltaic cell temperature at the current time step, symbol T stc Indicates the photovoltaic cell temperature under standard test;

[0024] Among them, the photovoltaic cell temperature T c It is expressed as follows:

[0025]

[0026] The energy storage state of charge model in the energy storage model is as follows:

[0027] P s.max =μE s.max

[0028] Among them, the symbol P s.max Indicates the upper limit of energy storage charging and discharging power, symbol Es.max Indicates the maximum capacity of the energy storage system, and the symbol μ represents the fixed proportional coefficient between the upper limit of energy storage power and capacity;

[0029] The energy storage state of charge model is as follows:

[0030]

[0031] Among them, the symbol SOC(t) represents the state of charge of the energy storage system at time t, the symbol SOC(t-1) represents the state of charge of the energy storage at the previous time step, and the symbol E s.max Indicates the maximum capacity during energy storage dispatch; symbol Indicates the charging and discharging power of the battery at time t, symbol represents the discharge power of the battery at time t, the symbol η represents the charge and discharge efficiency of the energy storage unit, and the symbol N t represents the scheduling period, and the symbol Δt represents the time step difference.

[0032] In some optional implementations, the mathematical formula of the daily operating cost objective function is as follows:

[0033] minF

[0034] The symbol F represents the daily operating cost of the island microgrid during the entire economic dispatch process, which is expressed as follows:

[0035]

[0036] Among them, the symbol C WT (t) represents the operating cost of the wind turbine at time t, symbol C PV (t) represents the photovoltaic power generation operating cost at time t, symbol C S (t) represents the operating cost of the energy storage system at time t, symbol N t Indicates the total scheduling cycle time;

[0037] Energy storage system operating cost C at time t S The model of (t) is expressed as follows:

[0038]

[0039] Among them, the symbol K S Indicates the converted unit charge and discharge cost, symbol and They represent the charging / discharging power input / output of the AC side of the energy storage inverter during period t, and the symbol η represents the charging and discharging efficiency of the energy storage unit;

[0040] Wind turbine operating cost C at time t WT (t) is expressed as follows:

[0041] C WT (t) = k W P WT (t)Δt

[0042] Among them, the symbol k W Indicates the operating cost coefficient of the fan, symbol P WT (t) represents the fan output at time t;

[0043] The photovoltaic power generation operating cost C at time t PV (t) is expressed as follows:

[0044] C PV (t) = k P P PV (t)Δt

[0045] Among them, the symbol k P Indicates the photovoltaic operation cost coefficient, symbol P PV (t) represents the photovoltaic output at time t.

[0046] In some optional implementations, the specific steps of modeling the pulse load related constraints are as follows:

[0047] Model the pulse load characteristics, including: the pulse load impact duration ratio is Among them, the symbol t i Indicates the duration of the impulse signal, symbol t n Indicates the non-impact time, and the symbol T indicates the entire working cycle of the load; the power characteristic of the pulse load is simplified to a single pulse impact, which is expressed by the following formula: P av =DP tp ; Among them, the symbol P avg Indicates the average power of the load, P tp Indicates the peak power of the load, and symbol D indicates the duty cycle of a single pulse in the scheduling period. Figure 3 As shown, Figure 3 The figure shows the pulse load power changing with time. In this figure, the horizontal axis t represents time and the vertical axis P represents power. The figure shows the change of pulse load power at different time points. tp Represents the power value of the pulse load during the pulse period. During the pulse period, the load power remains at this high level. avis the average power, represented by a dotted line, which reflects the average power level of the pulse load in one cycle. DT represents the duration of a single pulse, that is, the duration from the start to the end of the pulse. T represents the period of the pulse load, that is, the time interval between the start of two adjacent pulses, such as from 0 to T is a period, and T to 2T is the next period. t1 is the time when the pulse ends in each period. This characteristic of the pulse load has a great impact on the operation of the isolated microgrid containing wind, solar and storage. Therefore, in the technical solution of this application, it is necessary to model the pulse load characteristics and consider the impact of its instantaneous high power. By setting the dynamic frequency security constraints and wind, solar and storage frequency control constraints of the wind, solar and storage microgrid, energy storage, wind turbines, photovoltaics and other equipment can quickly adjust the output when the pulse load occurs, maintain the power balance and frequency stability of the microgrid, and ensure the safe and economical operation of the microgrid.

[0048] Modeling of the system's dynamic frequency response characteristics includes: The frequency response of the wind-solar-storage island microgrid under pulse load shock is expressed as follows:

[0049]

[0050] Where Δf n represents the frequency deviation of the nth time step in the frequency response process; Δτ represents the time step of the frequency response model; ΔP s,n represents the output response of the energy storage system at the nth time step; ΔP WT,n Indicates the output response of wind power generation at the nth time step, symbol ΔP pv,n represents the output response of photovoltaic power generation at the nth time step; the symbol H represents the microgrid inertia, the symbol D represents the microgrid damping; the symbol λ represents the power impact caused by the pulse load, and the symbol S N Indicates the rated capacity of the wind, solar and storage island microgrid;

[0051] The dynamic frequency security constraints of wind, solar and storage microgrids include:

[0052] Limiting the frequency deviation of the microgrid:

[0053]

[0054] Limit the frequency change rate of the microgrid:

[0055]

[0056] Limit the initial frequency change rate of the microgrid under pulse load:

[0057] 2HRoCoF min ≤λ / S N ≤2HRoCoF max

[0058] Quasi-steady-state frequency deviation constraint:

[0059]

[0060] In the above formula, the symbol Δf min Indicates the minimum allowable frequency deviation, symbol Δf max Indicates the maximum allowable frequency deviation; symbol RoCoF min Indicates the minimum allowable frequency change rate, symbol RoCoF max Indicates the maximum allowable frequency change rate; symbol Δf qss It represents the quasi-steady-state frequency deviation, which is expressed as the frequency response of the last time step in the scheduling period; the symbol Indicates the minimum value of the quasi-steady-state frequency deviation, symbol Indicates the maximum value of quasi-steady-state frequency deviation;

[0061] Wind, solar and energy storage frequency control constraints include:

[0062] The power regulation of energy storage during frequency response is expressed as:

[0063]

[0064] ΔP s,n =ΔP s,d,n +ΔP s,i,n

[0065] Where ΔP s,d,n Indicates the frequency modulation response of the energy storage system under droop control; ΔP s,i,n Indicates the frequency modulation response of the energy storage system under virtual inertia control; Represents the time constant of the energy storage ESS; Indicates the time constant of the filter; symbol H ess Represents the inertia of the energy storage system, symbol R ess represents the droop constant of stored energy;

[0066] When frequency instability occurs, droop control is used to provide frequency regulation backup for primary frequency regulation. The power regulation of wind turbines and photovoltaics during the frequency response process is expressed as:

[0067]

[0068] Where ΔP PV,n and ΔP WT,n They represent the primary frequency regulation response of photovoltaic and wind power under load reduction consideration; R PV and R WT Represent the droop coefficients of photovoltaic and wind power respectively; T PV Represents the photovoltaic time constant; T WT represents the time constant of wind power;

[0069] Primary frequency response quantity constraint:

[0070]

[0071] Among them, the symbol and symbol Respectively represent the minimum and maximum values ​​of the primary frequency modulation response of energy storage; symbols and symbol Represents the maximum output of photovoltaic and wind power at the maximum power point respectively; symbol ΔP PV,n , symbol ΔP WT,n and symbol ΔP s,n They represent photovoltaic, wind power and energy storage primary frequency regulation and backup respectively.

[0072] In some optional implementations, the following formula is used in the operation constraint model of the wind-solar-storage island microgrid to be analyzed to ensure the power balance of the wind-solar-storage island microgrid during the scheduling process:

[0073] P PV (t)+P WT (t)+P s (t) = P L (t)

[0074] Where, P PV (t), P PV (t), P s (t) are the active power outputs of wind turbine, photovoltaic and energy storage system at time t; P L (t) is the total load at time t;

[0075] The Disflow model is used to constrain the microgrid flow:

[0076]

[0077] Where E represents the set of lines in the microgrid, and the symbol (i, j) represents the line from node i to node j; P i and Q i are the active and reactive power of node i respectively; P Li and Q Li represent the active load and reactive load at node i respectively; P ij , Q ij 、l ij 、r ij and x ij Represent the active power, reactive power, square of current amplitude, resistance and inductive reactance on line (i, j) respectively; U i represents the square of the voltage amplitude at node i;

[0078] Microgrid node allowable voltage deviation constraints:

[0079] U min ≤U i ≤U max

[0080] Where U min and U min are the lower and upper limits of the allowed voltage offset respectively;

[0081] Energy storage system constraints include charge and discharge power constraints and energy storage initial and final state of charge constraints. The charge and discharge power constraints are expressed as follows:

[0082]

[0083] Where, The maximum charge and discharge power allowed by the energy storage system is mainly limited by the capacity of the energy storage grid-connected inverter device; and Respectively represent the energy storage discharge and charging power at time t; U S (t) represents the charge and discharge state of energy storage, U S (t)∈{0,1}, when the value is 1, it indicates discharge, and when the value is 0, it indicates charge;

[0084] The energy storage charge state constraint at the beginning and end of the day is used to ensure that the energy storage capacity is equal at the beginning and end of the dispatch, and is expressed as the following formula:

[0085]

[0086] To prevent the energy storage system from overcharging or over-discharging, the energy storage state of charge constraint:

[0087] SOC min ≤SOC(t)≤SOC max

[0088] Where, SOC max and SOC min The maximum / minimum remaining capacity allowed during the energy storage dispatch process;

[0089] The output constraints of photovoltaic and wind turbines in each time step are as follows:

[0090]

[0091] Where, and are the predicted output values ​​of photovoltaic and wind turbine at time t respectively.

[0092] According to a second aspect of an embodiment of the present invention, there is provided a microgrid operation control device including a pulse load, comprising:

[0093] The power generation and energy storage model construction module is used to construct the wind power generation characteristic model, photovoltaic power generation characteristic model and energy storage model in the wind-solar-storage island microgrid to be analyzed;

[0094] A daily operating cost construction module is used to establish a daily operating cost objective function of the wind-solar-storage island microgrid to be analyzed, where the operating costs in the daily operating cost objective function include the operating costs of the energy storage system, the wind turbine operating costs, and the photovoltaic power generation operating costs;

[0095] The pulse load constraint modeling module is used to model pulse load-related constraints, including pulse load characteristics, system dynamic frequency response characteristics, wind-solar-storage microgrid dynamic frequency security constraints, and wind-solar-storage microgrid frequency control constraints;

[0096] A microgrid operation constraint module is used to construct an operation constraint model of the wind-solar-storage island microgrid to be analyzed, including power balance, disflow constraints, energy storage system constraints, photovoltaic constraints, and wind turbine constraints;

[0097] The microgrid dispatch decision solving module is used to solve the operation constraint model using a particle swarm algorithm to obtain an operation dispatch decision solution for the wind, solar and storage island microgrid to be analyzed.

[0098] One embodiment of this specification can achieve at least the following beneficial effects:

[0099] 1. This technical solution constructs a daily operating cost objective function for an isolated wind, solar, and energy storage microgrid, comprehensively considering the operating costs of the energy storage system, wind turbines, and photovoltaic power generation. By quantifying and modeling these costs, a clear target is provided for optimization. In actual operation, different power sources have different power generation costs at different times. The objective function can coordinate the output of each power source to minimize overall operating costs while meeting load demand.

[0100] 2. The technical solution of this application adopts a particle swarm algorithm solution model. This particle swarm algorithm determines the optimal output of wind turbines, photovoltaics, and energy storage systems by continuously iteratively searching for the optimal solution while satisfying various constraints. Within a certain scheduling cycle, the algorithm can reasonably allocate the power generation of wind turbines and photovoltaics based on conditions such as wind speed and light intensity. When wind and solar power generation is sufficient, the energy storage system is charged to store excess electricity; when wind and solar power generation is insufficient or there is a pulse load impact, the energy storage system discharges to supplement electricity. Such optimized scheduling makes the output of each power source more reasonable, avoids unnecessary power generation costs, and thus reduces the daily operating costs of the isolated island microgrid.

[0101] 3. This technical solution models the characteristics of pulse loads, clarifying their high peak power and low average power characteristics and their impact on microgrid stability. By defining the pulse load impact duration ratio and simplifying the power characteristics, we gain a deeper understanding of the operating characteristics of pulse loads. Based on this, we establish a system dynamic frequency response model to analyze the output responses of energy storage systems, wind power generation, and photovoltaic power generation when pulse load power suddenly changes, as well as their impact on microgrid frequency deviation.

[0102] 4. The technical solution of this application has formulated dynamic frequency security constraints and wind, solar and storage frequency control constraints for the wind, solar and storage microgrid. By limiting the frequency deviation, frequency change rate and initial frequency change rate of the microgrid, it ensures that the frequency of the microgrid is stable under pulse load impact, avoiding power supply protection action and microgrid frequency collapse. When the frequency is unstable, wind power and photovoltaic power use droop control to provide frequency regulation backup, and the energy storage system also performs corresponding power regulation. These measures work together to enable the microgrid to quickly adjust the output of each power source under pulse load impact, maintain system frequency stability, and improve the system's ability to cope with pulse load impact.

[0103] 5. The technical solution of this application establishes an operation constraint model for a wind, solar and energy storage microgrid, covering power balance and disflow constraints, energy storage system-related constraints, photovoltaic-related constraints, wind turbine-related constraints, etc. The power balance constraint ensures the power balance of the microgrid during the scheduling process, and the flow constraint ensures the reasonable transmission of electricity in the microgrid. The charge and discharge power constraints and charge state constraints of the energy storage system prevent overcharging and over-discharging of the energy storage, ensuring the safe and reliable operation of the energy storage system; the photovoltaic and wind turbine output constraints are limited according to their power generation characteristics and predicted values ​​to ensure the rationality of power generation. These constraints cooperate with each other to ensure the reliable operation of the microgrid from multiple aspects. In actual operation, no matter how the wind, solar and energy storage power sources output, these constraints must be met. This not only ensures the safety of the microgrid equipment, but also ensures the reliability of the load power supply, so that the microgrid can operate stably and reliably. BRIEF DESCRIPTION OF THE DRAWINGS

[0104] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0105] Figure 1 This is a flow chart of a microgrid operation control method containing pulse loads provided by the present invention;

[0106] Figure 21 is a network structure diagram of a wind-solar-storage island microgrid system in a microgrid operation control method containing pulse loads provided by the present invention;

[0107] Figure 3 A load power characteristic diagram of a microgrid operation control method containing pulse loads provided by the present invention;

[0108] Figure 4 A flow chart of a multi-objective particle swarm algorithm in a microgrid operation control method containing pulse loads provided by the present invention;

[0109] Figure 5 The present invention provides Figure 1 A structural diagram of a microgrid operation control device containing pulse loads. DETAILED DESCRIPTION

[0110] To make the purpose, technical solutions, and advantages of one or more embodiments of this specification more clear, the technical solutions of one or more embodiments of this specification will be clearly and completely described below in conjunction with the specific embodiments of this specification and the corresponding drawings. Obviously, the embodiments described are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of one or more embodiments of this specification.

[0111] It should be understood that although the terms first, second, third, etc. may be used in this application document to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other.

[0112] Next, a method provided in the embodiment of the specification will be described in detail with reference to the accompanying drawings. Figure 1 This is a flow chart of a microgrid operation control method containing pulse loads provided by the present invention; Figure 2 1 is a network structure diagram of a wind-solar-storage island microgrid system in a microgrid operation control method containing pulse loads provided by the present invention; Figure 3 A load power characteristic diagram of a microgrid operation control method containing pulse loads provided by the present invention; Figure 4 This is a flow chart of a multi-objective particle swarm algorithm in a microgrid operation control method containing pulse loads provided by the present invention.

[0113] A wind, solar, and storage island microgrid is a small power system that operates independently of the main grid. It consists of wind power (wind), photovoltaic power (photovoltaic), and energy storage devices (storage). It possesses self-control, protection, and management capabilities, enabling the production, transmission, distribution, and use of electricity within a specific area. It consists of a wind power generation system, a photovoltaic power generation system, and an energy storage system. The wind power generation system converts wind energy into electricity through wind turbines, while the photovoltaic power generation system converts solar energy into electricity using photovoltaic cells. The energy storage system typically uses electrochemical energy storage technology, such as lead-acid batteries. The energy storage system plays a key role in energy regulation and balancing in a wind, solar, and storage island microgrid. When wind and photovoltaic power generation generate excess electricity, the energy storage system stores the excess energy. During periods of insufficient power generation or peak load demand, the stored energy is released to maintain power balance in the microgrid.

[0114] Pulse load is a special type of power load. When it is connected to a microgrid or other power system, it exhibits a specific power variation characteristic, which has a significant impact on the operation of the system. The characteristics of pulse load include: (1) Power mutation: Pulse load has the significant characteristics of high peak power and low average power. During operation, its power will change dramatically in a short period of time, jumping from a low average power level to a very high peak power, and then quickly dropping. For example, some large electric welding machines in industrial production require extremely high power to melt metal and form a strong arc at the moment of welding, but the power demand is low during the non-welding period; some electromagnetic pulse equipment will generate short and powerful power pulses when working. This power mutation characteristic is in sharp contrast to traditional stable loads (such as lighting equipment, constant power motors, etc.). The power demand of traditional loads is relatively stable and changes relatively slowly. (2) Impact and intermittency: The operation of pulse load is intermittent, and its high power pulses appear discontinuously, showing the characteristics of impact. It may have multiple power pulses in a period of time, with a certain time interval between the pulses, which makes it difficult for the system to continuously and stably supply it with power. These intermittent surges can cause a series of problems in the power system, such as fluctuations in system voltage and frequency, which can adversely affect system stability and power quality. For example, in a microgrid containing pulse loads, each pulse load startup causes a momentary surge in system voltage and frequency, potentially causing brief voltage drops or frequency fluctuations in other electrical devices, impacting their normal operation.

[0115] Due to the sudden change in power of the pulse load, the power balance of the microgrid will be instantly broken, which will cause frequency fluctuations. This impact is more obvious when the pulse load is connected to an isolated microgrid with a smaller capacity. According to the system dynamic frequency response characteristic model, the power impact brought by the pulse load will cause the frequency of the microgrid to change, which may cause the frequency deviation to exceed the allowable range, and even cause frequency instability, such as frequency oscillation or frequency collapse. For example, in a wind, solar and energy storage isolated microgrid, if a high-power pulse load is suddenly connected, and the power generation equipment and energy storage system in the microgrid cannot respond in time and provide sufficient power support, the frequency of the microgrid will drop rapidly, affecting the stable operation of the entire microgrid and causing damage to other sensitive loads (such as electronic equipment, precision instruments, etc.).

[0116] When introducing the technical solution of this application, first Figure 2 Explain the contents of Figure 2 This is a schematic diagram of the topological structure of a wind-solar-storage island microgrid with pulse loads. Figure 2 The figure shows the power supply, energy storage, and load components. In the power supply component, wind turbines (blue blocks, node 2) and photovoltaics (red blocks, node 3) generate electricity according to their respective characteristic models. Wind turbine power is affected by wind speed, which follows a Weibull distribution and is calculated using the wind speed-power characteristic curve formula. Photovoltaic cell output power is related to light intensity and temperature and is calculated using corresponding formulas. These renewable energy sources exhibit intermittent and fluctuating characteristics. Energy storage (green blocks, node 1) provides regulation and buffering. It stores energy during periods of excess renewable energy generation and releases it during periods of insufficient generation or pulsed loads. Its operation follows the energy storage model, including constraints such as charge and discharge power limits and state of charge. In the load component, pulsed loads (yellow blocks, node 6) exhibit large instantaneous power fluctuations, which can impact the microgrid's frequency. Based on the pulsed load characteristics and the system's dynamic frequency response, dynamic frequency security constraints and wind-solar-storage frequency control constraints for the wind-solar-storage microgrid are established to mitigate these impacts and maintain system stability.

[0117] During operation and scheduling, the entire microgrid uses daily operating costs as the objective function, comprehensively considers operating constraint models such as power balance and energy storage charging and discharging, and optimizes the output of wind turbines, photovoltaics, and energy storage. While meeting the needs of pulse loads and other requirements, it improves the economy and safety of microgrid operation.

[0118] The technical solution of the present application provides a method for controlling the operation of a microgrid containing pulse loads, which may include:

[0119] Step S1: Construct a wind power generation characteristic model, a photovoltaic power generation characteristic model, and an energy storage model in the wind-solar-storage island microgrid to be analyzed.

[0120] Step S2: Establishing a daily operating cost objective function for the wind-solar-storage island microgrid to be analyzed, where the operating costs in the daily operating cost objective function include the operating costs of the energy storage system, the wind turbine generator, and the photovoltaic power generation;

[0121] Step S3: Model pulse load related constraints, including pulse load characteristics, system dynamic frequency response characteristics, wind-solar-storage microgrid dynamic frequency security constraints, and wind-solar-storage frequency control constraints.

[0122] Step S4: Construct an operation constraint model of the wind-solar-storage island microgrid to be analyzed, including power balance, disflow constraints, energy storage system constraints, photovoltaic constraints and wind turbine constraints.

[0123] Step S5: using a particle swarm algorithm to solve the operation constraint model and obtain an operation scheduling decision plan for the wind-solar-storage island microgrid to be analyzed.

[0124] The following is a detailed explanation of this solution. In the wind power generation characteristic model, the wind speed satisfies the Weibull distribution, and the probability distribution function of the wind speed f(v) can be expressed as:

[0125]

[0126] Among them, the symbol c represents the scale parameter of the Weibull distribution function, the symbol k represents the shape parameter of the Weibull distribution function, and the symbol v represents the wind speed in the actual environment;

[0127] The wind speed power characteristic curve model of the wind turbine WT in the wind power generation characteristic model is expressed as follows:

[0128]

[0129] Among them, the symbol P′ WT Indicates the output power of the wind turbine, symbol P′ r Indicates the rated power of the wind turbine, symbol v′ ci Indicates the cut-in wind speed of the wind turbine, symbol v′ r Indicates the rated wind speed of the wind turbine, symbol v′ co Indicates the cut-out wind speed of the wind turbine, and the symbols a′, b′, c′ and d′ represent different wind speed parameters respectively;

[0130] Among them, the symbol P′ WTIt represents the output power of the wind turbine, usually in kilowatts (kW) or megawatts (MW). It reflects the ability of the wind turbine to transmit electricity to the grid under different wind speed conditions. Its value is directly related to the contribution of the wind turbine to the overall power supply in the microgrid. When the wind speed is low and does not reach the cut-in wind speed, the wind turbine cannot generate electricity. WT = 0; As the wind speed enters the power generation range, the input power is calculated according to the corresponding formula and changes with the wind speed. Symbol P′ r Indicates the rated power of the wind turbine, the unit is the same as P' WT Similarly, this is the maximum power that the wind turbine can stably output under the design conditions, reflecting the upper limit of the wind turbine's power generation capacity. When the wind speed reaches the rated wind speed v′ r When the wind turbine is in a specific range, it outputs at rated power, and the wind turbine is in the best power generation state. Different models of wind turbines have different rated powers. Large wind turbines can reach several megawatts, while small wind turbines range from tens of kilowatts to hundreds of kilowatts. Symbol v′ ci Indicates the cut-in wind speed of the wind turbine, in meters per second (m / s). It is the minimum wind speed at which the wind turbine can start generating electricity. When the actual wind speed is lower than this, the wind turbine blades cannot obtain sufficient driving force and the wind turbine does not generate electricity. P′ WT = 0. The cut-in wind speed is closely related to the structural design and blade characteristics of the wind turbine. Different types of wind turbines have different cut-in wind speeds, generally around 3-5m / s. Small wind turbines may have a lower cut-in wind speed to adapt to low wind speed environments; large wind turbines may have a slightly higher cut-in wind speed to ensure power generation efficiency and equipment stability. Symbol v′ r Indicates the rated wind speed of the wind turbine, in meters per second (m / s). When the wind speed reaches the rated wind speed, the wind turbine reaches the rated power output state. Symbol v′ co Indicates the cut-out wind speed of the wind turbine, also in meters per second (m / s). When the wind speed is higher than the cut-out wind speed, the wind turbine will stop running to protect the safety of the wind turbine equipment. At this time, P' WT = 0. The cut-out wind speed setting depends on factors such as the mechanical strength and electrical performance of the fan, and is generally around 20-25m / s. When encountering strong winds and the wind speed exceeds the cut-out wind speed, the fan control system will automatically shut down to prevent the fan blades, generator and other components from being damaged due to excessive mechanical stress and electrical load. ci ≤v≤v′ r In the range, the wind turbine output power increases linearly with the increase of wind speed; in v r <v<v′ coWithin this range, the wind turbine maintains rated power output. Rated wind speed is an important indicator for measuring wind turbine performance and determining its operating status. In microgrid planning, it is necessary to select wind turbines with rated wind speeds that match local wind speed resources to improve power generation efficiency and economic benefits. The symbols a′, b′, c′, and d′ represent different wind speed parameters.

[0131] The output power model of the photovoltaic cell in the photovoltaic power generation characteristic model is expressed by the following formula:

[0132]

[0133] Among them, the symbol P′ pv Represents the output active power of the photovoltaic cell, symbol R′ pv Indicates the photovoltaic output power under standard test conditions, symbol q′ pv Indicates the derating factor of photovoltaic power, symbol I T Indicates the actual solar radiation intensity, symbol I STC Indicates the solar radiation intensity under standard test conditions, symbol α p Indicates the temperature coefficient of the photovoltaic panel, symbol T c Indicates the photovoltaic cell temperature at the current time step, symbol T stc Indicates the photovoltaic cell temperature under standard testing.

[0134] Among them, the photovoltaic cell temperature T c It is expressed as follows:

[0135] T c =T stc +0.134·(1+0.031T stc )·(1-0.042v)I T

[0136] The energy storage state of charge model in the energy storage model is as follows:

[0137] P s.max =μE s.max

[0138] Among them, the symbol P s.max Indicates the upper limit of energy storage charging and discharging power, symbol E s.max Indicates the maximum capacity of the energy storage system, and the symbol μ represents the fixed proportional coefficient between the upper limit of energy storage power and capacity;

[0139] The energy storage state of charge model is as follows:

[0140]

[0141] Among them, the symbol SOC(t) represents the state of charge of the energy storage system at time t, the symbol SOC(t-1) represents the state of charge of the energy storage at the previous time step, and the symbol E s.max Indicates the maximum capacity during energy storage dispatch; symbol Indicates the charging and discharging power of the battery at time t, symbol represents the discharge power of the battery at time t, the symbol η represents the charge and discharge efficiency of the energy storage unit, and the symbol N t represents the scheduling period, and the symbol Δt represents the time step difference.

[0142] In an optional embodiment, the mathematical formula of the daily operating cost objective function may be as follows:

[0143] minF

[0144] The symbol F represents the daily operating cost of the island microgrid during the entire economic dispatch process, which is expressed as follows:

[0145]

[0146] Among them, the symbol C WT (t) represents the operating cost of the wind turbine at time t, symbol C PV (t) represents the photovoltaic power generation operating cost at time t, symbol C S (t) represents the operating cost of the energy storage system at time t, symbol N t Indicates the total scheduling cycle time;

[0147] Energy storage system operating cost C at time t S The model of (t) is expressed as follows:

[0148]

[0149] Among them, the symbol K S Indicates the converted unit charge and discharge cost, symbol and They represent the charging / discharging power input / output of the AC side of the energy storage inverter during period t, and the symbol η represents the charging and discharging efficiency of the energy storage unit;

[0150] Wind turbine operating cost C at time t WT (t) is expressed as follows:

[0151] C WT (t) = k W P WT (t)Δt

[0152] Among them, the symbol k W Indicates the operating cost coefficient of the fan, symbol P WT (t) represents the fan output at time t;

[0153] The photovoltaic power generation operating cost C at time t PV (t) is expressed as follows:

[0154] C PV (t) = k P P PV (t)Δt

[0155] Among them, the symbol k P Indicates the photovoltaic operation cost coefficient, symbol P PV (t) represents the photovoltaic output at time t.

[0156] In an optional embodiment and technical solution, the specific steps of modeling the pulse load related constraints may be as follows:

[0157] Model the pulse load characteristics, including: the pulse load impact duration ratio is The pulse load impact duration ratio is an indicator used to quantify the working characteristics of the impact load, where the symbol t i Indicates the duration of the impulse signal, symbol t n It represents the non-impact time, and the symbol T represents the entire working cycle of the load.

[0158] Take a working cycle T = 10s, impact time t i Taking a pulse load with a load current of 1 = 2 seconds as an example, IC = 0.2. The larger the IC value, the longer the pulse load impact duration relative to the duty cycle, and the more persistent the impact on microgrid stability. From a power perspective, according to P = UI, a long-term impact will cause the microgrid to experience power fluctuations for a long time. Assuming the microgrid voltage is generally stable, power fluctuations lead to unstable current I, which in turn affects the normal operation of microgrid equipment. This indicator quantifies the impact characteristics and provides a basis for subsequent scheduling strategy formulation.

[0159] The power characteristics of the pulse load are simplified to a single pulse impact and are expressed as follows: P av =DP tp ; Among them, the symbol P avg Indicates the average power of the load, P tp It represents the peak power of the load, and the symbol D represents the duty cycle of a single pulse in the scheduling period.

[0160] Modeling the system's dynamic frequency response characteristics can specifically include: The frequency response of the wind-solar-storage island microgrid under pulse load impact is expressed as follows:

[0161]

[0162] Where Δf nrepresents the frequency deviation of the nth time step in the frequency response process; Δτ represents the time step of the frequency response model; ΔP s,n represents the output response of the energy storage system at the nth time step; ΔP WT,n Indicates the output response of wind power generation at the nth time step, symbol ΔP pv,n represents the output response of photovoltaic power generation at the nth time step; the symbol H represents the microgrid inertia, the symbol D represents the microgrid damping; the symbol λ represents the power impact caused by the pulse load, and the symbol S N Indicates the rated capacity of the wind, solar and storage island microgrid;

[0163] The dynamic frequency security constraints of wind, solar and storage microgrids include:

[0164] Limiting the frequency deviation of the microgrid:

[0165]

[0166] Limit the frequency change rate of the microgrid:

[0167]

[0168] Limit the initial frequency change rate of the microgrid under pulse load:

[0169] 2HRoCoF min ≤λ / S N ≤2HRoCoF max

[0170] Quasi-steady-state frequency deviation constraint:

[0171]

[0172] In the above formula, the symbol Δf min Indicates the minimum allowable frequency deviation, symbol Δf max Indicates the maximum allowable frequency deviation; symbol RoCoF min Indicates the minimum allowable frequency change rate, symbol RoCoF max Indicates the maximum allowable frequency change rate; symbol Δf qss It represents the quasi-steady-state frequency deviation, which is expressed as the frequency response of the last time step in the scheduling period; the symbol Indicates the minimum value of the quasi-steady-state frequency deviation, symbol Indicates the maximum value of quasi-steady-state frequency deviation;

[0173] Wind, solar and energy storage frequency control constraints include:

[0174] The power regulation of energy storage during frequency response is expressed as:

[0175]

[0176] ΔP s,n =ΔP s,d,n +ΔP s,i,n

[0177] Where ΔP s,d,n Indicates the frequency modulation response of the energy storage system under droop control; ΔP s,i,n Indicates the frequency modulation response of the energy storage system under virtual inertia control; Represents the time constant of the energy storage ESS; Indicates the time constant of the filter; symbol H ess Represents the inertia of the energy storage system, symbol R ess represents the droop constant of stored energy;

[0178] When frequency instability occurs, droop control is used to provide frequency regulation backup for primary frequency regulation. The power regulation of wind turbines and photovoltaics during the frequency response process is expressed as:

[0179]

[0180] Where ΔP PV,n and ΔP WT,n They represent the primary frequency regulation response of photovoltaic and wind power under load reduction consideration; R PV and R WT Represent the droop coefficients of photovoltaic and wind power respectively; T PV Represents the photovoltaic time constant; T WT represents the time constant of wind power;

[0181] Primary frequency response quantity constraint:

[0182]

[0183] Among them, the symbol and symbol Respectively represent the minimum and maximum values ​​of the primary frequency modulation response of energy storage; symbols and symbol Represents the maximum output of photovoltaic and wind power at the maximum power point respectively; symbol ΔP PV,n , symbol ΔP WT,n and symbol ΔP s,n They represent photovoltaic, wind power and energy storage primary frequency regulation and backup respectively.

[0184] In an optional embodiment and technical solution, the following formula can be used in the operation constraint model of the wind-solar-storage island microgrid to be analyzed to ensure the power balance of the wind-solar-storage island microgrid during the scheduling process:

[0185] P PV (t)+P WT (t)+P s(t) = P L (t)

[0186] Where, P PV (t), P PV (t), P s (t) are the active power outputs of wind turbine, photovoltaic and energy storage system at time t; P L (t) is the total load at time t. This indicates that the sum of the active power output of the wind turbine, photovoltaic, and energy storage systems at time t must be equal to the total load at that moment. From the perspective of energy conservation, if the power is unbalanced, when the generated power is less than the load power, it will cause the voltage and frequency to drop, affecting the normal operation of the equipment, and in severe cases, may cause a power outage. When the generated power is greater than the load power, the excess electricity has nowhere to be consumed, which not only wastes energy but may also increase the system voltage and damage the equipment. This constraint ensures that the microgrid is always in a power balance state by adjusting the output of wind turbines, photovoltaics, and energy storage in real time, maintaining stable system operation.

[0187] The Disflow model is used to constrain the microgrid flow:

[0188]

[0189] Where E represents the set of lines in the microgrid, and the symbol (i, j) represents the line from node i to node j; P i and Q i are the active and reactive power of node i respectively; P Li and Q Li represent the active load and reactive load at node i respectively; P ij , Q ij 、l ij 、r ij and x ij Represent the active power, reactive power, square of current amplitude, resistance and inductive reactance on line (i, j) respectively; U i Represents the square of the voltage amplitude at node i.

[0190] Taking active power as an example, if the constraint is not met on a certain line (m,n), This means that the active power flowing into node m is greater than the sum of the outgoing power and the node load, causing power accumulation at the node and potentially increasing the node voltage. Excessive voltage can affect the insulation performance of devices connected to the node, shortening their service life or even damaging them. Conversely, if the incoming power is less than the sum of the outgoing power and the load, the node voltage will decrease, affecting the normal operation of the equipment. This power flow constraint allows for precise control of power distribution among nodes and lines in the microgrid, preventing problems such as line overloads and abnormal node voltages, and ensuring the security and stability of power transmission.

[0191] Microgrid node allowable voltage deviation constraints:

[0192] U min ≤U i ≤U max

[0193] Where U min and U min They are the lower and upper limits of the allowed voltage deviation respectively. Electrical equipment has a rated operating voltage range. Assuming that the rated voltage of a device is V rated , if the node voltage V i Less than V min , according to P = UI (for resistive load, it can be further derived as ), when the voltage is reduced, the actual power P of the equipment will decrease, which may cause the equipment to fail to start normally or reduce working efficiency. For example, if the voltage is too low, the speed of the motor will drop or even stall, causing the motor to overheat and burn. i Greater than V min Excessive voltage can subject the device insulation layer to excessive electric field strength, accelerating insulation aging and increasing the risk of insulation breakdown, which can also damage the device. This constraint ensures that the voltage fluctuations at each node remain within a reasonable range, protecting electrical equipment and ensuring the normal operation of the microgrid.

[0194] Energy storage system constraints include charge and discharge power constraints and energy storage initial and final state of charge constraints. The charge and discharge power constraints are expressed as follows:

[0195]

[0196] Where, The maximum charge and discharge power allowed by the energy storage system is mainly limited by the capacity of the energy storage grid-connected inverter device; and Respectively represent the energy storage discharge and charging power at time t; U S (t) represents the charge and discharge state of energy storage, U S (t)∈{0,1}, where a value of 1 indicates discharging, and a value of 0 indicates charging. Excessive current (corresponding to excessive power) generates excessive heat, exceeding the device's heat dissipation capacity, causing the device to overheat, damaging electronic components and affecting the normal operation of the energy storage system. This constraint protects the energy storage grid-connected inverter device by limiting the charge and discharge power, extending its service life.

[0197] The state of charge of the energy storage at the beginning and end of the day is constrained to SOC start =SOC end , if the energy storage system has different charge states at the beginning and end during a scheduling cycle, assuming that SOC end <SOC start, which means that the energy consumed by the energy storage during this cycle has not been replenished. In the long run, the available capacity of the energy storage system will gradually decrease and will not be able to meet the energy storage needs of the microgrid. On the contrary, if SOC end >SOC start , which may cause the energy storage system to be overcharged at the beginning of the subsequent scheduling cycle, affecting its performance and lifespan. This constraint ensures that the energy storage capacity is equal at the beginning and end of the scheduling cycle, allowing the energy storage system to maintain a stable operating state across different scheduling cycles, improving the service life and operating efficiency of the energy storage equipment, and ensuring that the energy storage system can continuously provide energy support for the microgrid.

[0198] The energy storage charge state constraint at the beginning and end of the day is used to ensure that the energy storage capacity is equal at the beginning and end of the dispatch, and is expressed as the following formula:

[0199]

[0200] To prevent the energy storage system from overcharging or over-discharging, the energy storage state of charge constraint:

[0201] SOC min ≤SOC(t)≤SOC max

[0202] Where, SOC max and SOC min The maximum / minimum remaining capacity allowed by the energy storage during the dispatch process. Taking overcharge as an example, when SOC(t)>SOC max When the battery is overcharged, side reactions will occur inside the battery. For example, overcharging of lithium batteries may cause lithium dendrites to grow, pierce the diaphragm, and cause battery short circuit, seriously affecting battery performance and safety, and even causing fire and other dangers. When over-discharged, that is, SOC(t) <SOC min , which can damage the battery electrode material structure, reduce the battery's reversible capacity, and shorten the battery life. This constraint ensures that the energy storage system operates within a safe state of charge range, protects the energy storage batteries, and maintains the reliability and stability of the energy storage system.

[0203] The output constraints of photovoltaic and wind turbines in each time step are as follows:

[0204]

[0205] Where, and are the predicted output values ​​of photovoltaic and wind turbine at time t respectively.

[0206] Taking photovoltaic as an example, assuming that the photovoltaic output forecast value P PV,forecase (t) is obtained by a specific algorithm based on factors such as meteorological data and photovoltaic cell characteristics. PV (t)>P PV,forecase(t), it may be that the prediction model is inaccurate or the photovoltaic equipment is abnormal, which will cause the microgrid power scheduling to deviate and affect the system stability. PV If (t) < 0, the PV system may be faulty and unable to generate power normally. This constraint limits the actual output range based on the output forecast, aligning the power generation of PV and wind turbines with the microgrid's load requirements and system operating characteristics. This improves the efficiency of renewable energy utilization and facilitates reasonable power scheduling in the microgrid, ensuring stable system operation.

[0207] In an optional embodiment and technical solution, the use of a particle swarm algorithm to solve the operation constraint model to obtain the operation scheduling decision plan of the wind-solar-storage island microgrid to be analyzed may specifically include:

[0208] Data initialization: Input the microgrid system composition and structural parameters, model-related parameters, where the model-related parameters include day-ahead load forecast data, photovoltaic and wind turbine output forecast values, energy storage operation parameters and related costs; at the same time, initialize the particle population, which includes particle position, speed, individual extreme value pbest, global extreme value gbest, and each particle in the population corresponds to a scheduling plan within a scheduling cycle, which covers the wind turbine output P WT , Photovoltaic output P PV , Energy storage system output P s , Photovoltaic load shedding and frequency regulation standby output ΔP PV,n , fan load reduction frequency regulation standby output ΔP WT,n , Energy storage frequency regulation standby output ΔP s,n ;

[0209] Fitness setting: The daily operating cost objective function value of the wind-solar-storage island microgrid with pulse load and the primary frequency regulation capability of wind-solar-storage is used as the individual fitness value, and the particle individual is input into the simulation model as the system variable. The variables that violate the power balance and Disflow flow constraints, energy storage system related constraints, photovoltaic related constraints, wind turbine related constraints, etc. are corrected; through the formula v i (t+1)=ω·v i (t)+c1·rand1(t)·(pbest i -x i (t))+c2·rand(t)·(gbest-x i (t)) Update the particle velocity by the formula x i (t+1)=x i (t)+v i (t+1) Update the particle position to obtain the offspring population, where Among them, v i (t) is the velocity of the particle, x i(t) is the position of the particle, pbest i is the best position found by each particle so far, gbest is the best position found by all particles in the entire group, rand1(t) and rand2(t) are random numbers between (0, 1), c1 and c2 are learning factors, ω max is the initial inertia weight, ω min is the inertia weight when iterating to the maximum algebra, inter max is the maximum number of iterations, and inter is the current number of iterations;

[0210] Determine the individual extreme value pbest: Use the initialized pbest as the initial individual extreme value of the particle. If the current particle can dominate pbest, then the current particle is taken as the individual extreme value of pbest; if they cannot be compared, calculate the number of other particles dominated by the two in the group, and the particle that dominates more is taken as the individual extreme value pbest. This step may specifically include:

[0211] Step 1: Initialize pbest, including: at the beginning of the particle swarm algorithm, each particle is assigned an initial individual extreme value pbest. This initial pbest is determined based on the scheduling scheme represented by the particle's initial position. For the wind, solar, and storage island microgrid operation and scheduling decision-making method with pulse loads that takes into account the primary frequency regulation capability of wind, solar, and storage, the position of each particle represents a possible scheduling scheme, which includes the values ​​of a series of variables such as wind turbine output, photovoltaic output, energy storage system output, photovoltaic load shedding and frequency regulation standby output, wind turbine load shedding and frequency regulation standby output, and energy storage frequency regulation standby output within the scheduling cycle. At this time, pbest is determined based on the objective function value corresponding to the initial position of the particle (such as comprehensive indicators such as total operating cost and frequency deviation).

[0212] Step 2: Compare the current particle with pbest, including:

[0213] Dominance relationships are determined, including calculating the objective function value of the current particle during each iteration. For the dispatch decision-making problem of an isolated wind-solar-storage microgrid with pulsed loads and taking into account the primary frequency regulation capability of wind, solar, and storage, the objective function may be a complex function that comprehensively considers multiple factors, such as the microgrid's daily operating costs, frequency stability, and power balance. The objective function value of the current particle is compared with the currently stored objective function value of pbest to determine whether a dominance relationship exists.

[0214] The definition of the dominance relationship is as follows: Assume that the objective function value of the current particle is F current , the objective function value of pbest is F pbest , for a minimization problem, if F current Less than or equal to F on all targetspbest , and on at least one target F current Less than F pbest , then the current particle is said to dominate pbest.

[0215] For example, assuming that the objective function includes two objectives: daily operating cost C and frequency deviation Δf, the objective function value F of the current particle current =(C current , Δf current ) and the objective function value F of pbest pbest =(C pbest , Δf pbest ), if C current ≤C pbest And Δf current ≤Δf pbest , and at least one inequality is strictly true (i.e. C current <C pbest or Δf current <Δf pbest ), then the current particle dominates pbest.

[0216] Step 3: Update the pbest situation, including:

[0217] Direct Domination Update: If the current particle dominates pbest, the schedule represented by the current particle outperforms the previously stored schedule represented by pbest in terms of overall performance. In this case, the current particle's position is updated to the new pbest. This is because the current particle has found a better solution that performs better on at least one objective and is comparable to the previous best solution on other objectives. Therefore, the individual extremum should be updated so that subsequent search processes can use this better solution as a reference for optimization.

[0218] Indirect Domination Update: If it is not possible to directly determine whether the current particle dominates pbest, that is, the two have advantages and disadvantages in different objectives, then it is necessary to calculate the number of other particles they dominate in the group. This is because when direct comparison cannot determine the superiority or inferiority, comparing their relative advantages in the entire group can more comprehensively evaluate their superiority or inferiority.

[0219] For each particle, calculate the number of other particles it dominates in the entire swarm. Specifically, for the current particle and pbest, traverse all other particles in the swarm. For another particle k in the swarm, use the dominance relationship determination method described above to determine whether the current particle dominates particle k, and whether pbest dominates particle k. Count the number of particles dominated by the current particle and pbest, respectively.

[0220] For example, the current particle dominates n in the populationcurrent particles, and pbest dominates n pbest particles. If n current >n pbest , indicating that the current particle is relatively better in the group because it can provide better scheduling solutions than other particles in more cases, so the current particle is updated to the pbest individual extreme value.

[0221] In this scheme, the main purpose of determining the individual extreme value pbest is to allow each particle to remember its best position (optimal solution) during the search process. By continuously updating pbest, particles can search in the direction of their own previously found optimal solution. At the same time, combined with the global optimal value gbest of the entire group, the search has both local and global search capabilities.

[0222] In complex multi-objective optimization scenarios, such as scheduling decisions for isolated wind, solar, and energy storage microgrids, a single comparison may overlook scheduling solutions that offer advantages across different objectives. By considering both dominance relationships and the number of dominances within the swarm, the quality of each particle's solution can be more comprehensively evaluated, avoiding the bias inherent in judging from a single metric. This allows the algorithm to identify optimal scheduling solutions within a complex solution space. This approach ensures that each particle has its own exploration direction, based on its own previously found optimal solutions. Incorporating global information (via gbest) allows the particle swarm algorithm to maintain both diversity (different particles have different pbests) and converge toward a more optimal solution space during the search process, improving the algorithm's search efficiency and the likelihood of finding the optimal solution. For wind, solar, and energy storage microgrid problems involving multiple constraints and objectives, this approach effectively balances different performance metrics, ultimately identifying scheduling solutions that excel in multiple aspects, such as cost, frequency stability, and power balance.

[0223] In summary, through this method of determining the individual extreme value pbest, the particle swarm algorithm can better search in complex optimization problems with multiple objectives and multiple constraints, avoid falling into local optimality, and find better operation scheduling decision-making solutions, while taking into account the balance between different objectives and the improvement of comprehensive performance.

[0224] Determine the global optimal value gbest: sort the population in layers, store the optimal non-dominated Pareto solution in an external archive set, remove non-Pareto solutions, and determine whether the external archive set exceeds the specified capacity. If it exceeds, select m particles based on the crowding distance; use the Pareto optimal solution saved in the external archive set, and use the roulette wheel method to select gbest from the external set based on the crowding distance of the optimal solution;

[0225] Iterative solution: Return to the fitness setting step and repeat the iteration until the preset termination condition is met. Under the relevant constraints, the wind-solar-storage island microgrid operation and scheduling decision plan considering pulse load is output. The plan includes wind turbine output, photovoltaic output, energy storage system output, photovoltaic load shedding and frequency regulation standby output, wind turbine load shedding and frequency regulation standby output, and energy storage frequency regulation standby output.

[0226] Step 5-2, fitness setting. The daily total operating cost F of the objective function in the model is used as the individual fitness value, and the individual particles are input into the simulation model as system variables, and the variables that violate the constraints are corrected. Using the individual fitness as the input value of the optimization model, the particle speed and position are updated by the following formula to obtain the offspring population:

[0227] V i+1 =ωV i +c1rand()(pbest i -X i )+c2rand()(gbest i -X i )

[0228] X i+1 =X i +V i+1

[0229] This formula shows that the velocity of particle i in the next iteration consists of three parts, namely the first part ωV i The inertia part is based on the current velocity, which ensures that the particle has a certain amount of inertia so that it can continue to move in the current direction. It is the cognitive part, which guides the particle to the optimal solution pbest it has found. i Close, reflecting the self-learning ability of particles. i -X i ) is the social part, which guides particles to approach the best solution in the group, reflecting the social learning ability of particles, enabling particles to share information and jointly find the optimal solution. Position update formula X i+1 =X i +V i+1 The position x of particle i is updated according to the updated velocity i (t+1). The updated position reflects the updated scheduling plan. The new position represents a new scheduling plan. For example, the new wind turbine output, photovoltaic output, energy storage system charge and discharge power and other parameters will change in the hope of finding a better scheduling plan.

[0230] Symbol V i Indicates the speed of the particle; symbol X i Indicates the position of the particle; pbesti is the best position found so far for each particle; gbest i is the best position found by all particles in the entire group; rand() is a random number between (0,1); c1 and c2 are learning factors, usually positive numbers, which determine the particle's movement toward the individual extreme value. The random number rand() is used to introduce randomness into the search process to avoid falling into the local optimal solution. ω is the dynamic weight value of the particle swarm, and its value is as follows:

[0231]

[0232] Among them, the symbol ω max represents the initial inertia weight; ω min Inertia weight when iterating to the maximum algebra; max is the maximum number of iterations; inter is the current number of iterations. At the beginning of the algorithm, ω is large, giving the particles a greater velocity and enabling them to have a stronger global search capability and explore a wider solution space. As the number of iterations increases, ω gradually decreases, the particle velocity decreases, the search range narrows, and the focus is placed on local search, which is conducive to accurately finding the optimal solution.

[0233] Step 5-3: Determine the individual extreme value pbest. Use the initialized pbest as the particle's initial individual probability value. If the current particle can dominate pbest, then the current particle is considered the individual extreme value of pbest. If they cannot be compared, calculate the number of other particles dominated by the two particles in the group. The particle with the most dominance is considered the individual extreme value pbest.

[0234] Step 5-4: Determine the global optimal value gbest. Perform stratified sorting on the population and store the optimal non-dominated Pareto solution in an external archive. Remove non-Pareto solutions and determine whether the external archive exceeds the specified capacity. If so, select m particles based on the crowding distance. Using the Pareto optimal solution stored in the external archive, use the roulette wheel method to select gbest from the external set based on the crowding distance of the optimal solution.

[0235] The following are some of the steps involved in the explanation:

[0236] (1) Explanation of hierarchical sorting and identification of non-dominated solutions, including:

[0237] Hierarchical sorting: In multi-objective optimization problems, all particles in a population are sorted hierarchically. Hierarchical sorting is based on dominance relationships. For two particles A and B in the particle swarm algorithm, A is said to dominate B if A is not inferior to B in all objective functions and is superior to B in at least one objective function. Based on this dominance relationship, the particles in the population are divided into different levels. The first level contains all particles that are not dominated by any other particle. These particles are called non-dominated solutions and form the Pareto front. The second level contains particles that are dominated only by particles in the first level. This process continues, and the entire population is divided into different levels.

[0238] Storing Pareto optimal solutions: The identified optimal non-dominated solutions (i.e., particles in the first layer, i.e., Pareto optimal solutions) are stored in an external archive set. These Pareto optimal solutions are "equally good" in the sense of multi-objective optimization, because they achieve a balance between different objectives, and it is impossible to optimize a certain objective without compromising other objectives. For example, when considering the operation and scheduling of wind, solar, and storage island microgrids, there may be multiple scheduling schemes at the same time. They have different trade-offs between multiple objectives such as cost, frequency regulation capability, and power balance, but none of them can improve one aspect without making other aspects worse. These schemes are Pareto optimal solutions.

[0239] (2) External archive collection capacity management, including:

[0240] Capacity judgment: When storing Pareto optimal solutions, it is necessary to determine whether the external archive collection exceeds the specified capacity. This is to avoid storing too many optimal solutions, which would lead to waste of computing resources and degradation of algorithm performance.

[0241] Crowding distance calculation and particle selection: If the number of particles in the external archive collection exceeds the specified capacity, it needs to be streamlined. First, the crowding distance of each particle is calculated. The crowding distance is an indicator used to measure the density of particles around a particle. It is calculated as follows:

[0242] For each objective function, the particles stored in the external archive collection are sorted according to the value of the objective function.

[0243] For particles on the boundary (i.e., particles with the minimum and maximum values ​​of the objective function), their crowding distance is set to infinity because they are on the boundary and are important for maintaining the diversity of the population.

[0244] For other particles, the crowding distance is the sum of the absolute values ​​of the differences between the two adjacent particles on the objective function. This ensures that the crowding distance comprehensively considers the distribution of particles in multiple objective function dimensions and avoids excessive particle density in a certain area.

[0245] m particles are selected based on the crowding distance. A larger crowding distance indicates a smaller number of particles surrounding the particle, making it more isolated and representative in the solution space of the multi-objective optimization problem. Therefore, it is more worthy of retention. This ensures greater diversity in the stored Pareto optimal solutions and prevents the algorithm from prematurely converging to a local optimal solution while ignoring other possible excellent solutions.

[0246] (3) Use the roulette wheel method to select gbest, the specific contents include:

[0247] Let's first explain the principles of roulette. Roulette is a probability-based selection method. After determining m particles in an external archive set, the probability of each particle being selected is calculated based on their crowding distance. The probability of each particle being selected is proportional to its crowding distance; the larger the crowding distance, the greater the probability of being selected. The calculation is as follows:

[0248] First, the crowding distances of all particles are summed up to get the total crowding distance D.

[0249] For each particle i, the probability of being selected is P i is the crowding distance d of the particle i Divide by the total crowding distance D, that is

[0250] Gbest selection: Based on the calculated probability, a particle is randomly selected as the global optimal value (gbest) using a roulette wheel. This can be implemented by generating a random number r in the range [0, 1], then accumulating particles according to their probabilities. When the accumulated probability exceeds r, the corresponding particle is selected as the gbest. This random selection method ensures that particles with larger crowding distances have a higher probability of being selected. It also introduces a degree of randomness into the search process, preventing the algorithm from falling into local optimal solutions and allowing it to explore a wider space of excellent solutions, which is more conducive to finding the true global optimal solution.

[0251] This approach not only finds multiple Pareto-optimal solutions to the current multi-objective optimization problem and stores them in an external archive, but also selects the global optimal value (gbest) from these optimal solutions using the crowding distance and roulette wheel method. By comprehensively considering the quality and diversity of the solutions, the final gbest selected is more representative and exploratory, helping to find more optimal scheduling decisions for wind, solar, and energy storage island microgrids. When considering multiple objectives such as cost, frequency stability, and power balance, this approach avoids single-minded optimization and instead seeks a solution with the best overall performance. This approach also considers the balance and diversity among different solutions, ensuring that appropriate scheduling decisions can be found under diverse conditions and requirements. This approach is particularly important in complex multi-objective optimization problems, as there is often no single, absolute optimal solution. Instead, a set of excellent solutions that balance different objectives must be found. From these solutions, a representative gbest that is relatively optimal in the current search phase is selected to guide the algorithm towards a more optimal solution.

[0252] Step 5-5, return to step 5-2, repeat the iteration until the termination condition is met, and output the wind-solar-storage island microgrid operation scheduling decision plan considering pulse load under the relevant constraints, including the wind turbine output P WT , Photovoltaic output P PV , Energy storage system output P s , Photovoltaic load shedding and frequency regulation standby output ΔP PV,n , fan load reduction frequency regulation standby output ΔP WT,n , Energy storage frequency regulation standby output ΔP s,n wait.

[0253] The technical solution of this application constructs a daily operating cost objective function for an isolated wind, solar, and energy storage microgrid, comprehensively considering the operating costs of the energy storage system, wind turbines, and photovoltaic power generation. By quantifying and modeling these costs, a clear target is provided for optimization. In actual operation, different power sources have different power generation costs at different times. The objective function can coordinate the output of each power source to minimize the overall operating cost while meeting load demand. The technical solution of this application adopts a particle swarm algorithm solution model. This particle swarm algorithm continuously iteratively searches for the optimal solution and determines the optimal output of the wind turbine, photovoltaic, and energy storage systems while meeting various constraints. Within a certain scheduling cycle, the algorithm can reasonably allocate the power generated by the wind turbine and photovoltaic based on conditions such as wind speed and sunlight. When wind and solar power generation is sufficient, the energy storage system charges and stores excess energy. When wind and solar power generation is insufficient or there is a pulse load shock, the energy storage system discharges to supplement energy. This optimized scheduling makes the output of each power source more reasonable, avoids unnecessary power generation costs, and thus reduces the daily operating cost of the isolated island microgrid. The technical solution of this application models the characteristics of pulse loads, clarifying their high peak power and low average power, and their impact on microgrid stability. By defining the pulse load impact duration ratio and simplifying the power characteristics, a deeper understanding of the operating characteristics of pulse loads is achieved. Based on this, a system dynamic frequency response characteristic model is established to analyze the output responses of the energy storage system, wind power generation, and photovoltaic power generation when the pulse load power suddenly changes, as well as their impact on the microgrid's frequency deviation. The technical solution of this application establishes dynamic frequency security constraints and wind, solar, and storage frequency control constraints for the wind, solar, and storage microgrid. By limiting the microgrid's frequency deviation, frequency change rate, and initial frequency change rate, this ensures the microgrid's frequency stability under pulse load impacts, avoiding power supply protection activation and microgrid frequency collapse. When the frequency is unstable, wind power and photovoltaic power use droop control to provide frequency regulation backup, and the energy storage system also performs corresponding power regulation. These measures work together to enable the microgrid to quickly adjust the output of each power source under pulse load impacts, maintain system frequency stability, and enhance the system's ability to cope with pulse load impacts. The technical solution of this application establishes an operation constraint model for a wind, solar and energy storage microgrid, covering power balance and disflow constraints, energy storage system-related constraints, photovoltaic-related constraints, wind turbine-related constraints, etc. The power balance constraint ensures the power balance of the microgrid during the scheduling process, and the flow constraint ensures the reasonable transmission of electricity in the microgrid. The charge and discharge power constraints and charge state constraints of the energy storage system prevent overcharging and over-discharging of the energy storage, ensuring the safe and reliable operation of the energy storage system; the photovoltaic and wind turbine output constraints are limited according to their power generation characteristics and predicted values ​​to ensure the rationality of power generation. These constraints cooperate with each other to ensure the reliable operation of the microgrid from multiple aspects. In actual operation, no matter how the wind, solar and energy storage power sources output, these constraints must be met. This not only ensures the safety of the microgrid equipment, but also ensures the reliability of the load power supply, so that the microgrid can operate stably and reliably.

[0254] At the same time, if Figure 5 As shown, the technical solution of the present application provides a microgrid operation control device containing pulse loads, which may include:

[0255] The power generation and energy storage model construction module 502 is used to construct a wind power generation characteristic model, a photovoltaic power generation characteristic model and an energy storage model in the wind, solar and energy storage island microgrid to be analyzed.

[0256] The daily operating cost construction module 504 is used to establish the daily operating cost objective function of the wind-solar-storage island microgrid to be analyzed. The operating costs in the daily operating cost objective function include the operating costs of the energy storage system, the wind turbine operating costs, and the photovoltaic power generation operating costs.

[0257] The pulse load constraint modeling module 506 is used to model pulse load related constraints, including pulse load characteristics, system dynamic frequency response characteristics, wind-solar-storage microgrid dynamic frequency security constraints, and wind-solar-storage frequency control constraints.

[0258] The microgrid operation constraint module 508 is used to construct an operation constraint model of the wind-solar-storage island microgrid to be analyzed, including power balance, disflow constraints, energy storage system constraints, photovoltaic constraints and wind turbine constraints.

[0259] The microgrid dispatch decision solving module 510 is used to solve the operation constraint model using a particle swarm algorithm to obtain an operation dispatch decision solution for the wind-solar-storage island microgrid to be analyzed.

[0260] Those skilled in the art will appreciate that the accompanying drawings are merely schematic diagrams of an embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0261] Those skilled in the art will appreciate that the modules in the apparatuses of the embodiments may be distributed in the apparatuses of the embodiments as described in the embodiments, or may be located in one or more apparatuses different from the embodiments with corresponding changes. The modules in the above embodiments may be combined into one module or further divided into multiple sub-modules.

[0262] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A microgrid operation control method containing pulse loads, characterized in that: The method comprises the following steps: Construct the wind power generation characteristic model, photovoltaic power generation characteristic model and energy storage model in the wind-solar-storage island microgrid to be analyzed; Establishing a daily operating cost objective function for the wind-solar-storage island microgrid to be analyzed, wherein the operating costs in the daily operating cost objective function include the operating costs of the energy storage system, the wind turbine generator, and the photovoltaic power generation; Modeling of pulse load-related constraints, including pulse load characteristics, system dynamic frequency response characteristics, wind-solar-storage microgrid dynamic frequency security constraints, and wind-solar-storage microgrid frequency control constraints; Constructing an operation constraint model for the wind-solar-storage island microgrid to be analyzed, including power balance, disflow constraints, energy storage system constraints, photovoltaic constraints, and wind turbine constraints; The particle swarm algorithm is used to solve the operation constraint model to obtain the operation scheduling decision plan of the wind-solar-storage island microgrid to be analyzed.

2. The microgrid operation control method with pulse load according to claim 1, characterized in that: The wind speed in the wind power generation characteristic model satisfies the Weibull distribution, and the probability distribution function of the wind speed f(v) is expressed as: Among them, the symbol c represents the scale parameter of the Weibull distribution function, the symbol k represents the shape parameter of the Weibull distribution function, and the symbol v represents the wind speed in the actual environment; The wind speed power characteristic curve model of the wind turbine WT in the wind power generation characteristic model is expressed as follows: Among them, the symbol P′ WT Indicates the output power of the wind turbine, symbol P′ r Indicates the rated power of the wind turbine, symbol v′ ci Indicates the cut-in wind speed of the wind turbine, symbol v′ r Indicates the rated wind speed of the wind turbine, symbol v′ co Indicates the cut-out wind speed of the wind turbine, and the symbols a′, b′, c′ and d′ represent different wind speed parameters respectively; The output power model of the photovoltaic cell in the photovoltaic power generation characteristic model is expressed by the following formula: Among them, the symbol P′ pv Represents the output active power of the photovoltaic cell, symbol R′ pv Indicates the photovoltaic output power under standard test conditions, symbol q′ pv Indicates the derating factor of photovoltaic power, symbol I T Indicates the actual solar radiation intensity, symbol I STC Indicates the solar radiation intensity under standard test conditions, symbol α p Indicates the temperature coefficient of the photovoltaic panel, symbol T c Indicates the photovoltaic cell temperature at the current time step, symbol T stc Indicates the photovoltaic cell temperature under standard test; Among them, the photovoltaic cell temperature T c It is expressed as follows: T c =T stc +0.134·(1+0.031T stc )·(1-0.042v)I T The energy storage state of charge model in the energy storage model is as follows: P s.max =μE s.max Among them, the symbol P s.max Indicates the upper limit of energy storage charging and discharging power, symbol E s.max Indicates the maximum capacity of the energy storage system, and the symbol μ represents the fixed proportional coefficient between the upper limit of energy storage power and capacity; The energy storage state of charge model is as follows: Among them, the symbol SOC(t) represents the state of charge of the energy storage system at time t, the symbol SOC(t-1) represents the state of charge of the energy storage at the previous time step, and the symbol E s.max Indicates the maximum capacity during energy storage dispatch; symbol Indicates the charging and discharging power of the battery at time t, symbol represents the discharge power of the battery at time t, the symbol η represents the charge and discharge efficiency of the energy storage unit, and the symbol N t represents the scheduling period, and the symbol Δt represents the time step difference.

3. The microgrid operation control method with pulse load according to claim 1, characterized in that: The mathematical formula of the daily operating cost objective function is as follows: minF The symbol F represents the daily operating cost of the island microgrid during the entire economic dispatch process, which is expressed as follows: Among them, the symbol C WT (t) represents the operating cost of the wind turbine at time t, symbol C PV (t) represents the photovoltaic power generation operating cost at time t, symbol C S (t) represents the operating cost of the energy storage system at time t, symbol N t Indicates the total scheduling cycle time; Energy storage system operating cost C at time t S The model of (t) is expressed as follows: Among them, the symbol K S Indicates the converted unit charge and discharge cost, symbol and They represent the charging / discharging power input / output of the AC side of the energy storage inverter during period t, and the symbol η represents the charging and discharging efficiency of the energy storage unit; Wind turbine operating cost C at time t WT (t) is expressed as follows: C WT (t)=k W P WT (t)Δt Among them, the symbol k W Indicates the operating cost coefficient of the fan, symbol P WT (t) represents the fan output at time t; The photovoltaic power generation operating cost C at time t PV (t) is expressed as follows: C PV (t)=k P P PV (t)Δt Among them, the symbol k P Indicates the photovoltaic operation cost coefficient, symbol P PV (t) represents the photovoltaic output at time t.

4. The microgrid operation control method with pulse load according to claim 1, characterized in that: The specific steps of modeling the pulse load related constraints are as follows: Model the pulse load characteristics, including: the pulse load impact duration ratio is Among them, the symbol t i Indicates the duration of the impulse signal, symbol t n Indicates the non-impact time, and the symbol T indicates the entire working cycle of the load; the power characteristic of the pulse load is simplified to a single pulse impact, which is expressed by the following formula: P av =DP tp ; Among them, the symbol P avg Indicates the average power of the load, P tp It represents the peak power of the load, and the symbol D represents the duty cycle of a single pulse in the scheduling period; Modeling of the system's dynamic frequency response characteristics includes: The frequency response of the wind-solar-storage island microgrid under pulse load shock is expressed as follows: Where Δf n represents the frequency deviation of the nth time step in the frequency response process; Δτ represents the time step of the frequency response model; ΔP s,n represents the output response of the energy storage system at the nth time step; ΔP WT,n Indicates the output response of wind power generation at the nth time step, symbol ΔP pv,n represents the output response of photovoltaic power generation at the nth time step; the symbol H represents the microgrid inertia, the symbol D represents the microgrid damping; the symbol λ represents the power impact caused by the pulse load, and the symbol S N Indicates the rated capacity of the wind, solar and storage island microgrid; The dynamic frequency security constraints of wind, solar and storage microgrids include: Limiting the frequency deviation of the microgrid: Limit the frequency change rate of the microgrid: Limit the initial frequency change rate of the microgrid under pulse load: 2HRoCoF min ≤λ / S N ≤2HRoCoF max Quasi-steady-state frequency deviation constraint: In the above formula, the symbol Δf min Indicates the minimum allowable frequency deviation, symbol Δf max Indicates the maximum allowable frequency deviation; symbol RoCoF min Indicates the minimum allowable frequency change rate, symbol RoCoF max Indicates the maximum allowable frequency change rate; symbol Δf qss It represents the quasi-steady-state frequency deviation, which is expressed as the frequency response of the last time step in the scheduling period; the symbol Indicates the minimum value of the quasi-steady-state frequency deviation, symbol Indicates the maximum value of quasi-steady-state frequency deviation; Wind, solar and energy storage frequency control constraints include: The power regulation of energy storage during frequency response is expressed as: ΔP s,n =ΔP s,d,n +ΔP s,i,n Where ΔP s,d,n Indicates the frequency modulation response of the energy storage system under droop control; ΔP s,i,n Indicates the frequency modulation response of the energy storage system under virtual inertia control; Represents the time constant of the energy storage ESS; Indicates the time constant of the filter; symbol H ess Represents the inertia of the energy storage system, symbol R ess represents the droop constant of stored energy; When frequency instability occurs, droop control is used to provide frequency regulation backup for primary frequency regulation. The power regulation of wind turbines and photovoltaics during the frequency response process is expressed as: Where ΔP PV,n and ΔP WT,n They represent the primary frequency regulation response of photovoltaic and wind power under load reduction consideration; R PV and R WT Represent the droop coefficients of photovoltaic and wind power respectively; T PV Represents the photovoltaic time constant; T WT represents the time constant of wind power; Primary frequency response quantity limitation constraint: Among them, the symbol and symbol Respectively represent the minimum and maximum values ​​of the primary frequency modulation response of energy storage; symbols and symbol Represents the maximum output of photovoltaic and wind power at the maximum power point respectively; symbol ΔP PV,n , symbol ΔP WT,n and symbol ΔP s,n They represent photovoltaic, wind power and energy storage primary frequency regulation and backup respectively.

5. The microgrid operation control method with pulse load according to claim 1, characterized in that: The following formula is used in the operation constraint model of the wind-solar-storage island microgrid to be analyzed to ensure the power balance of the wind-solar-storage island microgrid during the scheduling process: P PV (t)+P WT (t)+P s (t)=P L (t) Where, P PV (t), P PV (t), P s (t) are the active power outputs of wind turbine, photovoltaic and energy storage system at time t; P L (t) is the total load at time t; The Disflow model is used to constrain the microgrid flow: Where E represents the set of lines in the microgrid, and the symbol (i, j) represents the line from node i to node j; P i and Q i are the active and reactive power of node i respectively; P Li and Q Li represent the active load and reactive load at node i respectively; P ij , Q ij 、l ij 、r ij and x ij Represent the active power, reactive power, square of current amplitude, resistance and inductive reactance on line (i, j) respectively; U i represents the square of the voltage amplitude at node i; Microgrid node allowable voltage deviation constraints: IN min ≤U i ≤U max Where U min and U min are the lower and upper limits of the allowed voltage offset respectively; Energy storage system constraints include charge and discharge power constraints and energy storage initial and final state of charge constraints. The charge and discharge power constraints are expressed as follows: Where, The maximum charge and discharge power allowed by the energy storage system is mainly limited by the capacity of the energy storage grid-connected inverter device; and Respectively represent the energy storage discharge and charging power at time t; U S (t) represents the charge and discharge state of energy storage, U S (t)∈{0,1}, when the value is 1, it indicates discharge, and when the value is 0, it indicates charge; The energy storage charge state constraint at the beginning and end of the day is used to ensure that the energy storage capacity is equal at the beginning and end of the dispatch, and is expressed as the following formula: To prevent the energy storage system from overcharging or over-discharging, the energy storage state of charge constraint: SOC min ≤SOC(t)≤SOC max Where, SOC max and SOC min The maximum / minimum remaining capacity allowed during the energy storage dispatch process; The output constraints of photovoltaic and wind turbines in each time step are as follows: Where, and are the predicted output values ​​of photovoltaic and wind turbine at time t respectively.

6. The microgrid operation control method with pulse load according to claim 1, characterized in that: The particle swarm algorithm is used to solve the operation constraint model to obtain the operation scheduling decision plan of the wind-solar-storage island microgrid to be analyzed, which specifically includes: Data initialization: Input microgrid system composition and structural parameters, and model-related parameters, including day-ahead load forecast data, photovoltaic and wind turbine output forecast values, energy storage operating parameters, and related costs. At the same time, initialize the particle population, which includes particle position, velocity, individual extreme value pbest, and global extreme value gbest. Each particle in the population corresponds to a scheduling plan within a scheduling cycle, which covers wind turbine output, photovoltaic output, energy storage system output, photovoltaic load shedding and frequency regulation standby output, wind turbine load shedding and frequency regulation standby output, and energy storage frequency regulation standby output. Fitness setting: The daily operating cost objective function value of the wind-solar-storage island microgrid with pulse load and the primary frequency regulation capability of wind-solar-storage is used as the individual fitness value, and the particle individual is input into the simulation model as the system variable. The variables that violate the power balance and Disflow flow constraints, energy storage system related constraints, photovoltaic related constraints, wind turbine related constraints, etc. are corrected; through the formula v i (t+1)=ω·v i (t)+c1·rand1(t)·(pbest i -x i (t))+c2·rand(t)·(gbest-x i (t)) Update the particle velocity by the formula x i (t+1)=x i (t)+v i (t+1) Update the particle position to obtain the offspring population, where Among them, v i (t) is the velocity of the particle, x i (t) is the position of the particle, pbest i is the best position found by each particle so far, gbest is the best position found by all particles in the entire group, rand1(t) and rand2(t) are random numbers between (0, 1), c1 and c2 are learning factors, ω max is the initial inertia weight, ω min is the inertia weight when iterating to the maximum algebra, inter max is the maximum number of iterations, and inter is the current number of iterations; Determine the individual extreme value pbest: use the initialized pbest as the initial individual extreme value of the particle. If the current particle can dominate pbest, then the current particle is taken as the individual extreme value of pbest. If they cannot be compared, calculate the number of other particles dominated by the two in the group, and the particle that dominates more is taken as the individual extreme value pbest. Determine the global optimal value gbest: sort the population in layers, store the optimal non-dominated Pareto solution in an external archive set, remove non-Pareto solutions, and determine whether the external archive set exceeds the specified capacity. If it exceeds, select m particles based on the crowding distance; use the Pareto optimal solution saved in the external archive set, and use the roulette wheel method to select gbest from the external set based on the crowding distance of the optimal solution; Iterative solution: Return to the fitness setting step and repeat the iteration until the preset termination condition is met. Under the relevant constraints, the wind-solar-storage island microgrid operation and scheduling decision plan considering pulse load is output. The plan includes wind turbine output, photovoltaic output, energy storage system output, photovoltaic load shedding and frequency regulation standby output, wind turbine load shedding and frequency regulation standby output, and energy storage frequency regulation standby output.

7. A microgrid operation control device containing pulse loads, characterized in that: The device comprises the following modules: The power generation and energy storage model construction module is used to construct the wind power generation characteristic model, photovoltaic power generation characteristic model and energy storage model in the wind-solar-storage island microgrid to be analyzed; A daily operating cost construction module is used to establish a daily operating cost objective function of the wind-solar-storage island microgrid to be analyzed, where the operating costs in the daily operating cost objective function include the operating costs of the energy storage system, the wind turbine operating costs, and the photovoltaic power generation operating costs; The pulse load constraint modeling module is used to model pulse load-related constraints, including pulse load characteristics, system dynamic frequency response characteristics, wind-solar-storage microgrid dynamic frequency security constraints, and wind-solar-storage microgrid frequency control constraints; A microgrid operation constraint module is used to construct an operation constraint model of the wind-solar-storage island microgrid to be analyzed, including power balance, disflow constraints, energy storage system constraints, photovoltaic constraints, and wind turbine constraints; The microgrid dispatch decision solving module is used to solve the operation constraint model using a particle swarm algorithm to obtain an operation dispatch decision solution for the wind, solar and storage island microgrid to be analyzed.

Images