A hierarchical coordinated frequency regulation control method for wind-storage-load in microgrids

Abstract

A microgrid wind-storage-load layered coordinated frequency modulation control method, the method comprising the following steps: S1, based on the operating characteristics of each resource in the microgrid system, establish a wind-storage-load layered coordinated frequency modulation control architecture; S2, respectively establish frequency modulation control models for gas turbines, doubly-fed wind turbines, energy storage systems and controllable loads, and establish an overall frequency response model of the microgrid system; S3, establish a prediction model for the frequency response characteristic index of the microgrid system, and use the maximum frequency deviation as the frequency stratification index, propose a wind-storage-load layered coordinated frequency modulation control strategy based on the prediction model, and compensate the power absorbed by the wind turbine through the energy storage system during the wind turbine speed recovery stage. The present invention can control the wind turbine, energy storage system and controllable load to participate in frequency modulation in advance, thereby improving the system frequency stability.

Description

Technical Field

[0001] The present invention belongs to the field of microgrid frequency control, and in particular relates to a microgrid wind-storage-load layered coordinated frequency modulation control method. Background Art

[0002] Microgrids, small-scale power systems integrating distributed renewable energy generation, are rapidly developing. The increasing proportion of renewable energy sources such as wind power in microgrids is making them cleaner and more environmentally friendly. However, when wind power is connected to the grid via power electronics, its rotor speed is decoupled from the grid frequency, making it unable to directly provide frequency support. This reduces the microgrid's inertia, frequency response capacity, and ability to withstand frequency fluctuations, posing a serious challenge to frequency stability in the microgrid.

[0003] To improve the frequency stability of microgrids, domestic and foreign scholars have proposed applying additional frequency control to wind turbines to give them a certain degree of frequency regulation capability. Existing technologies use virtual inertia control for doubly-fed wind turbines. This can create a large virtual inertia on the wind turbine rotor side, providing inertia support for the system by releasing rotor kinetic energy. This has a fast response speed, but an excessive drop in rotor speed may cause a secondary drop in system frequency. To improve the primary frequency regulation capability of wind turbines, virtual inertia control is usually combined with overspeed load reduction control and pitch angle control. This control allows the wind turbine to retain a certain amount of frequency regulation reserve capacity, but at the expense of wind energy, making it impossible to maximize wind energy utilization.

[0004] In recent years, the rapid development of energy storage technology has brought new approaches to frequency regulation in microgrids. Energy storage, with its advantages of fast response and flexible regulation, can assist wind power in providing frequency support. Furthermore, some loads in microgrids, such as air conditioners, electric water heaters, and electric vehicles, possess energy storage characteristics and can provide power support to the microgrid through demand response technology. Existing technologies use energy storage to assist wind turbine rotor recovery, thus avoiding secondary frequency drops. Existing technologies have proposed coordinated control strategies for wind-storage systems that take wind speed into account, using fuzzy logic control to achieve rational allocation of frequency regulation power. Existing technologies have established a cluster control model for air conditioning loads that takes electrical characteristics into account, using frequency droop control to smooth microgrid power fluctuations and improve system frequency stability. To address the reduced grid inertia caused by electric vehicle integration, existing technologies have proposed a virtual synchronous generator control strategy for electric vehicles that takes user charging needs into account. This strategy enables primary and secondary frequency regulation of electric vehicles, reducing grid frequency fluctuations while meeting user needs. These studies have primarily focused on coordinated control of wind-storage systems or frequency regulation of controllable loads, with limited research on how to coordinate wind power, energy storage, and controllable loads in frequency regulation. At the same time, the above studies all use real-time control, which cannot provide early warning of the severity of the system frequency drop. When the load mutation is large, the frequency regulation resources in the system may not be able to respond in time, causing serious frequency drops. Summary of the Invention

[0005] The purpose of the present invention is to overcome the defects and problems of poor frequency regulation effect in the prior art and to provide a microgrid wind-storage-load layered collaborative frequency regulation control method with good frequency regulation effect.

[0006] To achieve the above objectives, the technical solution of the present invention is: a microgrid wind-storage-load layered coordinated frequency modulation control method, the method comprising the following steps:

[0007] S1. Based on the operating characteristics of each resource in the microgrid system, a wind-storage-load layered coordinated frequency regulation control architecture is established;

[0008] S2. Establish frequency regulation control models for the gas turbine, doubly-fed wind turbine, energy storage system, and controllable load, and establish an overall frequency response model for the microgrid system;

[0009] S3. Establish a prediction model for the frequency response characteristic indicators of the microgrid system, and use the maximum frequency deviation as the frequency stratification indicator. Propose a wind-storage-load stratified collaborative frequency regulation control strategy based on the prediction model. During the wind turbine speed recovery phase, the energy storage system is used to compensate for the power absorbed by the wind turbine.

[0010] In step S1, the operating characteristics of the gas turbine, wind turbine, energy storage system, and controllable load in the microgrid system are analyzed, and a wind-storage-load layered collaborative frequency regulation control architecture is established. The frequency regulation is divided into four layers: the first layer is gas turbine frequency regulation; the second layer is wind turbine frequency regulation; the third layer is energy storage system frequency regulation; and the fourth layer is controllable load frequency regulation.

[0011] In step S2, the gas turbine frequency regulation control model includes an inertial response model and a primary frequency regulation model;

[0012] The inertial response model is:

[0013]

[0014] Where ΔP Gm is the change in the mechanical torque of the gas turbine, ΔP Ge is the change of electromagnetic torque, H m is the inertia time constant of the gas turbine, Δf is the system frequency change, t is time, D m is the damping coefficient of the gas turbine;

[0015] The primary frequency modulation model is:

[0016]

[0017] Where K m is the power-frequency characteristic coefficient of the gas turbine, T mis the response time constant of the gas turbine, s is a complex variable, and Δf(s) is the frequency change in the complex frequency domain.

[0018] In step S2, the fan frequency regulation control model includes a virtual inertia control model and an overspeed load reduction control model;

[0019] The virtual inertia control model is:

[0020]

[0021] Where ΔP W is the power regulation of the doubly fed wind turbine, K w is the virtual inertia coefficient, T w is the response time constant of the doubly fed wind turbine, s is a complex variable, and Δf(s) is the frequency variation in the complex frequency domain;

[0022] The overspeed load reduction control model is:

[0023]

[0024] Where, P Wm is the mechanical power output of the wind turbine, C p is the wind energy conversion coefficient, ρ a is the air density, S w is the wind turbine blade area, V ω is the wind speed;

[0025] Using overspeed load reduction control, the wind turbine power-speed curve under different wind speeds meets the following requirements:

[0026] P opt =P max ·η

[0027] Where, P max is the maximum power, P opt is the overspeed load reduction power, η is the load reduction power ratio;

[0028] In the overspeed load reduction mode, the double-fed wind turbine has a margin of ΔP opt The power involved in frequency modulation:

[0029] ΔP opt =P max (1-η)

[0030] Where ΔP opt The reserved frequency regulation power for the doubly fed wind turbine.

[0031] In step S2, the frequency regulation control model of the energy storage system is:

[0032]

[0033] Where λsoc is the state of charge of the energy storage system, λ soc0 is the initial state of charge of the energy storage system, P E is the output power of the energy storage system, E N is the rated capacity of the energy storage system;

[0034] Using droop control to provide power support for microgrids:

[0035]

[0036] Where ΔP E is the power regulation of the energy storage system, K E is the droop coefficient of the energy storage system, T E is the response time constant of the energy storage system, S is a complex variable, and Δf(s) is the frequency change in the complex frequency domain.

[0037] In step S2, the controllable load frequency regulation control model is:

[0038]

[0039] Where ΔP L is the power regulation of the controllable load, K L is the controllable load droop coefficient, T L is the response time constant of the controllable load, S is a complex variable, and Δf(s) is the frequency change in the complex frequency domain.

[0040] In step S2, the overall frequency response model of the microgrid system is:

[0041]

[0042] Where Δf is the system frequency change, ΔP0 is the system load disturbance, H s is the system inertial response transfer function, G T , G w , G E , G L They are gas turbine, doubly fed wind turbine, energy storage system, and controllable load frequency response transfer function, u w 、u E 、u L are the state variables of the double-fed wind turbine, energy storage system, and controllable load frequency regulation, respectively. m is the inertia time constant of the gas turbine, D m is the damping coefficient of the gas turbine, K m is the power-frequency characteristic coefficient of the gas turbine, T m is the response time constant of the gas turbine, K w is the virtual inertia coefficient, T w is the response time constant of the doubly fed wind turbine, KE is the droop coefficient of the energy storage system, T E is the response time constant of the energy storage system, K L is the controllable load droop coefficient, T L is the response time constant of the controllable load, and S is a complex variable.

[0043] In step S3, the characteristic indicators of the system frequency response include the maximum frequency change rate, the maximum frequency deviation and the steady-state frequency deviation;

[0044] When the load is disturbed, the maximum frequency change rate of the system is obtained by the initial value theorem for:

[0045]

[0046] The system steady-state frequency deviation Δf is obtained from the final value theorem ∞ for:

[0047]

[0048] The frequency modulation control model of each resource in the system is equivalent to an equivalent frequency modulation control model with the same characteristics. The transfer function G of the equivalent frequency modulation control model is eq for:

[0049]

[0050] Where A eq 、B Eq 、C eq are all characteristic parameters of the equivalent model; t mid It is half of the time-weighted average value of the frequency modulation steady state of each resource; are the frequency regulation steady-state times of the gas turbine, doubly-fed wind turbine, energy storage system, and controllable load respectively;

[0051] The time series expression Δf(t) of the frequency deviation of the system equivalent frequency response model is:

[0052]

[0053] c=6H s C eq (a 2 -b 2 )+2a(2H s +A eq +C eq D eq )

[0054] +B eq +D eq

[0055] d=12abHs C eq +2b(2H s +A eq +C eq D eq )

[0056] g=(aC eq +1)c+bdC eq ;h=bcC eq -(aC eq +1)d

[0057] Where, M, N, a, b, c, d, h, g are all frequency timing expression parameters; D eq is the system equivalent damping coefficient;

[0058] The maximum frequency deviation of the system at time t tir and the maximum frequency deviation Δf max for:

[0059]

[0060] In step S3, when the system frequency changes, the gas turbine responds in real time;

[0061] When the system maximum frequency deviation is predicted to exceed the stratification index Δf max1 When , the wind turbine adopts virtual inertia control to reduce the system frequency change;

[0062] When the system maximum frequency deviation is predicted to exceed the stratification index Δf max2 When , the energy storage system adopts droop control to reduce the system frequency deviation;

[0063] When the system maximum frequency deviation is predicted to exceed the stratification index Δf max3 When , the controllable load adopts droop control, which reduces the system frequency deviation by increasing or decreasing the load power.

[0064] In step S3, the energy storage system's power is regulated to compensate for the energy absorbed by the wind turbine until the wind turbine returns to its initial operating state. Subsequently, the energy storage system begins to smoothly exit frequency modulation, and ultimately all frequency modulation power is provided by the gas turbine. After detecting that the system frequency change rate is zero, the frequency modulation control equation of the energy storage system is modified to:

[0065]

[0066] Where ΔP′ E is the power regulation of the energy storage system after correction, and Δf(s) is the frequency change in the complex frequency domain.

[0067] Compared with the prior art, the present invention has the following beneficial effects:

[0068] The present invention provides a microgrid wind-storage-load layered coordinated frequency regulation control method, which fully considers the frequency regulation capability of each resource in the microgrid system. Through the coordinated frequency regulation control of the gas turbine, wind power, energy storage system and controllable load, the frequency stability of the system is improved. At the same time, in order to improve the frequency regulation capability of the wind turbine, overspeed load reduction control is adopted for the wind turbine, and an overall frequency response model of the microgrid system is established. A calculation method for characteristic indicators of the system frequency response is proposed, including the maximum frequency change rate, the maximum frequency deviation and the steady-state frequency deviation. The maximum frequency deviation is used as the wind, storage and load frequency regulation layered indicator. The maximum frequency deviation of the system is predicted by detecting the initial frequency change rate of the system, and the maximum frequency deviation is compared with each layered indicator to determine whether each resource participates in frequency regulation, thereby realizing model predictive control of wind-storage-load frequency regulation, controlling the corresponding resources to participate in frequency regulation in advance, effectively reducing the system frequency deviation and shortening the stabilization time of the system frequency. In addition, during the wind turbine speed recovery process, the power of the energy storage system is regulated to compensate for the power absorbed by the wind turbine, which can reduce system frequency fluctuations, accelerate frequency recovery and avoid secondary frequency drops. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 This is a flow chart of a microgrid wind-storage-load layered coordinated frequency regulation control method of the present invention.

[0070] Figure 2 It is a structural diagram of the microgrid system in the present invention.

[0071] Figure 3 This is the wind-storage-load layered coordinated frequency modulation control architecture of the present invention.

[0072] Figure 4 It is a schematic diagram of the power-speed curve of the doubly-fed wind turbine in the present invention.

[0073] Figure 5 It is a schematic diagram of the frequency regulation process of the doubly fed wind turbine under overspeed control in the present invention.

[0074] Figure 6 It is the overall frequency response model of the system in the present invention.

[0075] Figure 7 This is the system frequency change process when the load suddenly increases in the present invention.

[0076] Figure 8 This is a flow chart of wind-storage-load coordinated control based on model prediction in the present invention.

[0077] Figure 9 This is a diagram of the system frequency response results when the load suddenly increases by 0.5MW in the present invention.

[0078] Figure 10This is a diagram of the system frequency response results when the load suddenly increases by 1.5MW in the present invention.

[0079] Figure 11 This is a diagram of the system frequency response results when the load suddenly increases by 2.5MW in the present invention.

[0080] Figure 12 This is a comparison chart of the effects of the hierarchical control strategy when the load suddenly increases by 0.5MW in the present invention.

[0081] Figure 13 This is a comparison chart of the effects of the hierarchical control strategy when the load suddenly increases by 1.5MW in the present invention.

[0082] Figure 14 This is a comparison chart of the effects of the hierarchical control strategy when the load suddenly increases by 2.5MW in the present invention.

[0083] Figure 15 This is a diagram of the system frequency response results before and after energy storage compensation in the present invention.

[0084] Figure 16 This is a power variation diagram of the doubly-fed wind turbine in the present invention.

[0085] Figure 17 This is a diagram showing changes in energy storage power in the present invention.

[0086] Figure 18 This is a diagram of the speed change of the doubly fed wind turbine in the present invention. DETAILED DESCRIPTION

[0087] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0088] The basic concept of the present invention is as follows: (1) The operating characteristics of resources such as gas turbines, wind power, energy storage systems and controllable loads in the microgrid are analyzed, and a wind-storage-load layered collaborative frequency regulation control architecture is constructed. The frequency regulation is divided into four layers: the first layer is the gas turbine frequency regulation, when the system frequency changes, the gas turbine responds in real time; the second layer is the wind turbine frequency regulation, when the maximum system frequency deviation is predicted to exceed Δf max1 When the wind turbine adopts virtual inertia control to reduce the system frequency change; the third layer is the energy storage system frequency regulation. When the maximum frequency deviation of the system is predicted to exceed Δf max2 When the maximum frequency deviation of the system exceeds Δf, the energy storage system adopts droop control to reduce the system frequency deviation; the fourth layer is controllable load frequency regulation. max3When the load is controlled, the controllable load adopts droop control, and the system frequency deviation is reduced by increasing or decreasing the load power. (2) Based on the established wind-storage-load layered collaborative frequency regulation control architecture, the frequency regulation control models of gas turbines, wind turbines, energy storage and controllable loads are established respectively, and then the overall frequency response model of the microgrid system is established. (3) Based on the established overall frequency response model of the microgrid system, the calculation method of the system's maximum frequency change rate, maximum frequency deviation and steady-state frequency deviation is proposed; and the maximum frequency deviation is used as the wind, storage and load frequency regulation layered index to predict the system's maximum frequency deviation, and compare it with each layered index to determine whether each resource participates in frequency regulation, thereby realizing the model predictive control of wind-storage-load frequency regulation.

[0089] The microgrid system structure including wind power, energy storage and controllable load is as follows: Figure 2 As shown in the figure, wind turbines are represented by doubly-fed wind turbines (DFIGs), which are connected to the microgrid via voltage-source PWM converters. Energy storage systems are represented by electrochemical energy storage, which are connected to the microgrid via DC / DC and DC / AC converters. System loads are divided into controllable and uncontrollable loads, which are connected to the microgrid via step-up transformers.

[0090] In this microgrid system, gas turbines, as the main frequency regulation resource, need to participate in frequency regulation at all times. Wind turbines have a large installed capacity and a high penetration rate. According to the specification "Technical Regulations for Wind Farm Access to the Power System", grid-connected wind turbines must have inertia support and primary frequency regulation capabilities, so wind turbines have priority in frequency regulation. However, the operating cost of the energy storage system is relatively high, and its frequency regulation priority is relatively low. The regulation of controllable loads will affect the user's electricity demand. In order to ensure electricity reliability, controllable loads will only participate in frequency regulation when the frequency regulation capacity of gas turbines, wind turbines and energy storage is difficult to guarantee the safety requirements of the system frequency. Therefore, in order to make full use of the frequency regulation capabilities of resources such as wind power, energy storage and controllable loads in the system, and at the same time consider the economy of the regulation of each resource, the present invention adopts a hierarchical collaborative control architecture, that is, different frequency limits are set for the control methods of each frequency regulation resource, and the working status of each resource is collaboratively controlled by detecting or predicting the maximum deviation of the system frequency, such as Figure 3 shown.

[0091] See also Figure 1 A microgrid wind-storage-load layered coordinated frequency modulation control method, the method comprising the following steps:

[0092] S1. Based on the operating characteristics of each resource in the microgrid system, a wind-storage-load layered coordinated frequency regulation control architecture is established;

[0093] The operating characteristics of gas turbines, wind turbines, energy storage systems and controllable loads in the microgrid system are analyzed, and a wind-storage-load layered collaborative frequency regulation control architecture is established. The frequency regulation is divided into four layers: the first layer is gas turbine frequency regulation; the second layer is wind turbine frequency regulation; the third layer is energy storage system frequency regulation; and the fourth layer is controllable load frequency regulation.

[0094] S2. Establish frequency regulation control models for the gas turbine, doubly-fed wind turbine, energy storage system, and controllable load, and establish an overall frequency response model for the microgrid system;

[0095] The gas turbine frequency regulation control model includes the inertial response model and the primary frequency regulation model;

[0096] (1) Inertial response model

[0097] The inertia of the power system reflects the system's ability to resist frequency changes, which is mainly provided by the gas turbine. During normal operation, the mechanical torque of the gas turbine is balanced with the electromagnetic torque and friction damping. However, when a fault or disturbance occurs in the system, the electromagnetic torque of the gas turbine suddenly changes, and its rotor kinetic energy is used to balance the load change. Its rotor motion state satisfies the rotor motion equation:

[0098]

[0099] Where ΔP Gm is the change in the mechanical torque of the gas turbine, ΔP Ge is the change of electromagnetic torque, H m is the inertia time constant of the gas turbine, Δf is the system frequency change, t is time, D m is the damping coefficient of the gas turbine;

[0100] (2) Primary frequency modulation model

[0101] The gas turbine has the ability to perform primary frequency regulation, which is achieved through its speed control system. The speed control system detects the difference between the rotor speed and the rated speed in real time, and drives the turbine valve through the integrated amplifier and PID control link to achieve control of the input mechanical power.

[0102] The complex frequency domain expression of the primary frequency regulation of the gas turbine is:

[0103]

[0104] Where K m is the power-frequency characteristic coefficient of the gas turbine, T m is the response time constant of the gas turbine, S is a complex variable, and Δf(s) is the frequency change in the complex frequency domain.

[0105] The fan frequency regulation control model includes a virtual inertia control model and an overspeed load reduction control model;

[0106] (1) Virtual inertia control

[0107] Doubly-fed wind turbine rotors contain a large amount of rotational kinetic energy, capable of providing inertia for the power grid. However, since doubly-fed wind turbines are connected to the grid via power electronic converters, their rotors are decoupled from the grid and cannot respond to changes in system frequency. Therefore, auxiliary control strategies are needed to release the rotor's kinetic energy and provide inertia support for the grid.

[0108] Referring to the inertia control principle of traditional thermal power units, virtual inertia control is adopted for the doubly fed wind turbine. That is, within the safe range of rotor speed (0.67~1.33pu), the wind turbine output power is controlled in proportion to the grid frequency change rate to slow down the frequency change speed. The control equation is:

[0109]

[0110] Where ΔP W is the power regulation of the doubly fed wind turbine, K w is the virtual inertia coefficient, T w is the response time constant of the doubly fed wind turbine, s is a complex variable, and Δf(s) is the frequency variation in the complex frequency domain;

[0111] (2) Overspeed load reduction control

[0112] According to aerodynamics knowledge, the mechanical kinetic energy output by the wind turbine is shown as follows:

[0113]

[0114] Where, P Wm is the mechanical power output of the wind turbine; C p is the wind energy conversion coefficient, which is related to the wind turbine pitch angle, wind speed, and wind turbine speed; ρ a is the air density, S w is the wind turbine blade area, V ω is the wind speed;

[0115] In order to make full use of wind energy, doubly fed wind turbines usually adopt maximum power point tracking (MPPT) control, that is, under a certain wind speed and fixed pitch angle, the wind turbine output power is maximized by adjusting the wind turbine speed; however, when the doubly fed wind turbine participates in frequency modulation, it needs to consume its rotor kinetic energy to provide power support for the power grid. If the doubly fed wind turbine still operates at the maximum power point, it will consume a lot of rotor kinetic energy. Due to the limit of the rotor safety speed, the frequency modulation capacity provided by the doubly fed wind turbine is relatively small. At the same time, during the recovery process of the rotor speed, it will absorb a large amount of energy from the power grid, causing a serious secondary drop in frequency; therefore, the doubly fed wind turbine needs to reserve a certain amount of frequency modulation reserve power, and it can usually be operated on both sides of the maximum power point, that is, overspeed control or deceleration control; and research has found that deceleration control will cause static instability problems, so the present invention adopts overspeed load reduction control, see Figure 4 , the wind turbine power-speed curve under different wind speeds satisfies:

[0116] P opt =P max ·η

[0117] Where, P max is the maximum power, P opt is the overspeed load reduction power, η is the load reduction power ratio;

[0118] In the overspeed load reduction mode, the double-fed wind turbine has a margin of ΔP opt The power of the quartz crystal participates in frequency modulation, improving its frequency modulation capability:

[0119] ΔP opt =P max (1-η)

[0120] Where ΔP opt The reserved frequency regulation power for the doubly-fed wind turbine;

[0121] Under the overspeed load reduction control mode, the change curves of electromagnetic power and mechanical power during the frequency regulation of the doubly fed wind turbine are as follows: Figure 5As shown; when the system is operating normally, the doubly fed wind turbine operates stably at point a. When a grid fault occurs and the load suddenly increases, the system frequency decreases. Under the control of virtual inertia, the electromagnetic power output by the doubly fed wind turbine will increase to point b, while its mechanical power is still at point a, which is less than the electromagnetic power. The wind turbine begins to release rotor kinetic energy and the speed gradually decreases; in the process of system frequency decrease, under the regulation of various frequency regulation resources, the system frequency change rate gradually decreases, the electromagnetic power of the doubly fed wind turbine will gradually decrease along curve bc, the rotor speed will also decrease accordingly, the mechanical power will first increase and then decrease along the ac curve, and the system frequency reaches the lowest point at point c; thereafter, the frequency begins to rise, and under the control of virtual inertia, the electromagnetic power of the doubly fed wind turbine will decrease to point d, which is less than the mechanical power, and the rotor speed begins to recover; in the process of speed recovery, the electromagnetic power of the doubly fed wind turbine will gradually increase along the da curve, and the mechanical power will return along the ca curve, and finally stabilize at the initial operating point a;

[0122] In summary, overspeed control utilizes the surplus mechanical power of the doubly fed wind turbine during the frequency regulation process. While providing the same frequency regulation capacity, the rotor speed variation range of the doubly fed wind turbine using overspeed control is smaller than that of MPPT control. At the same time, during the speed recovery process, the surplus mechanical power also speeds up the recovery of the wind turbine speed.

[0123] Electrochemical energy storage has the advantages of fast response, high energy density, and flexible power regulation. It can quickly charge and discharge in the event of a grid fault to provide effective frequency support for the system. Currently, research on electrochemical energy storage is relatively mature and widely used. The energy storage system is connected to the grid through a bidirectional converter, and the power is decoupled from the grid. When the energy storage system participates in frequency regulation, its charge and discharge power can be controlled by controlling its grid-connected converter to meet the frequency regulation requirements of the grid. The state of charge is usually used to represent the energy stored in the energy storage system. Its operating characteristics are:

[0124]

[0125] Where λ soc is the state of charge of the energy storage system, λ soc0 is the initial state of charge of the energy storage system, P E is the output power of the energy storage system, E N is the rated capacity of the energy storage system;

[0126] In order to give full play to the frequency regulation capability of the energy storage system, droop control is used to provide power support for the microgrid:

[0127]

[0128] Where ΔP E is the power regulation of the energy storage system, K E is the droop coefficient of the energy storage system, TE is the response time constant of the energy storage system, s is a complex variable, and Δf(s) is the frequency change in the complex frequency domain.

[0129] There are a large number of user-side loads, which can be divided into important loads, transferable loads and interruptible loads according to their importance. Among them, important loads have high requirements for power reliability and power outages are not allowed within a specific time, while transferable loads and interruptible loads can participate in the regulation of the power grid within a certain range. Typical loads include air conditioners, electric water heaters, electric vehicles, etc. With the support of demand response technology, these loads can regulate their own power through direct load control, provide power support for the power grid after a fault, and promote the recovery of system frequency.

[0130] The controllable loads in a microgrid are numerous and dispersed. If decentralized control is adopted, the control cost will be greatly increased. Therefore, load aggregators can be used to aggregate the controllable loads into a whole, so that they can participate in the frequency control of the power grid in a unified manner and provide certain compensation to users. Droop control is usually used to provide power support for the power grid, namely:

[0131]

[0132] Where ΔP L is the power regulation of the controllable load, K L is the controllable load droop coefficient, T L is the response time constant of the controllable load, s is a complex variable, and Δf(s) is the frequency change in the complex frequency domain;

[0133] See also Figure 6 , the overall frequency response model of the microgrid system is:

[0134]

[0135] Where Δf is the system frequency change, ΔP0 is the system load disturbance, H s is the system inertial response transfer function, G T , Gw, G E , G L are the frequency response transfer functions of gas turbine, doubly fed wind turbine, energy storage system and controllable load respectively; u w 、u E 、u L are the state variables of the double-fed wind turbine, energy storage system, and controllable load frequency regulation, which are 1 when participating in frequency regulation and 0 when not participating in frequency regulation; H m is the inertia time constant of the gas turbine, D m is the damping coefficient of the gas turbine, K m is the power-frequency characteristic coefficient of the gas turbine, T m is the response time constant of the gas turbine, K wis the virtual inertia coefficient, T w is the response time constant of the doubly fed wind turbine, K E is the droop coefficient of the energy storage system, T E is the response time constant of the energy storage system, K L is the controllable load droop coefficient, T L is the response time constant of the controllable load, and S is a complex variable.

[0136] Through the coordinated frequency regulation of wind power, energy storage and controllable loads, the transient frequency stability of the microgrid can be effectively improved and the frequency deviation of the system can be reduced. However, on a long time scale, gas turbines are still needed to maintain the power balance of the system.

[0137] S3. Establish a prediction model for the frequency response characteristic indicators of the microgrid system, use the maximum frequency deviation as the frequency stratification indicator, and propose a wind-storage-load stratified coordinated frequency regulation control strategy based on the prediction model. During the wind turbine speed recovery phase, the energy storage system compensates for the power absorbed by the wind turbine.

[0138] When a microgrid fault causes a sudden increase in load, the system frequency changes under the influence of various frequency regulation resources as follows: Figure 7 shown; according to Figure 7 The characteristic indicators of system frequency response include maximum frequency change rate, maximum frequency deviation and steady-state frequency deviation; among them, steady-state frequency deviation reflects the stability of system frequency in long time scale under the action of single frequency modulation, while maximum frequency change rate and maximum frequency deviation reflect the short-term stability of system frequency;

[0139] When the load is disturbed, the maximum frequency change rate of the system is obtained by the initial value theorem for:

[0140]

[0141] According to the final value theorem, the system steady-state frequency deviation Δf∞ is:

[0142]

[0143] To find the analytical expression for the maximum frequency deviation, it is necessary to find the time-domain expression for the frequency deviation and then differentiate the time-domain expression with respect to time to find the maximum frequency deviation moment and the maximum frequency deviation. However, the system is a high-order system, and it is difficult to find the maximum frequency deviation moment analytically after converting it into a time-domain expression. Therefore, the system needs to be reduced in order. The transfer functions of the gas turbine, doubly fed wind turbine, energy storage system, and controllable load primary frequency regulation have the same form and can be expressed in the following unified form:

[0144]

[0145] In the formula, A, B, and C are characteristic parameters;

[0146] The frequency modulation control model of each resource in the system is equivalent to an equivalent frequency modulation control model with the same characteristics. The transfer function G of the equivalent frequency modulation control model is eq for:

[0147]

[0148] Where A eq 、B eq 、C eq are all characteristic parameters of the equivalent model; t mid It is half of the time-weighted average value of the frequency modulation steady state of each resource; are the frequency regulation steady-state times of the gas turbine, doubly-fed wind turbine, energy storage system, and controllable load respectively;

[0149] The time series expression Δf(t) of the frequency deviation of the system equivalent frequency response model is:

[0150]

[0151] c=6H s C eq (a 2 -b 2 )+2a(2H s +A eq +C eq D eq )

[0152] +B eq +D eq

[0153] d=12abH s C eq +2b(2H s +A eq +C eq D eq )

[0154] g=(aC eq +1)c+bdC eq ;h=bcC eq -(aC eq +1)d

[0155] Where, M, N, a, b, c, d, h, g are all frequency timing expression parameters; D eq is the system equivalent damping coefficient;

[0156] The maximum frequency deviation of the system at time t tir and the maximum frequency deviation Δf max for:

[0157]

[0158] When the load disturbance is constant, the maximum frequency change rate of the system is only related to the inertia of the gas turbine. As many gas turbines as possible should be connected to the power grid to increase the system inertia. The steady-state frequency deviation of the system is related to the damping coefficient of the gas turbine and the droop coefficient of its governor, the droop coefficient of the energy storage system, and the droop coefficient of the controllable load. Therefore, putting the energy storage system and the controllable load into frequency regulation and increasing their droop coefficient can effectively reduce the steady-state frequency deviation of the system. The maximum frequency deviation of the system is related to the operating parameters and control parameters of each frequency regulation resource. Increasing the inertia of the gas turbine, putting each resource into frequency regulation, and increasing the frequency regulation coefficient of each resource can all reduce the maximum frequency deviation of the system.

[0159] In summary, the maximum frequency change rate, maximum frequency deviation, and steady-state frequency deviation are indicators of system frequency safety and stability. After a fault occurs, the frequency regulation function of various resources within the microgrid system can limit the maximum frequency change rate, maximum frequency deviation, and steady-state frequency deviation to a safe range, thereby meeting the microgrid frequency safety and stability requirements.

[0160] When a fault or load disturbance occurs in a microgrid, the original power balance of the system is broken and the frequency begins to change. Under the action of various frequency regulation resources in the microgrid, the system power reaches a new balance and the frequency reaches a new stable value. Among them, each frequency regulation resource mainly realizes frequency regulation through inertia control and droop control. Inertia control is related to the system frequency change rate, which can damp the system frequency change and suppress frequency fluctuations. Droop control is related to the system frequency deviation, which can reduce the frequency deviation and accelerate the recovery of the system frequency. The frequency stability of the microgrid can be effectively improved by coordinating inertia control and droop control. When a fault occurs in the microgrid or the load is switched on and off, if all wind, storage, load and other resources participate in frequency regulation, it will be detrimental to the system economy. If the system frequency regulation capacity is insufficient, it will be difficult to meet the frequency stability requirements. Therefore, the above-mentioned wind-storage-load layered collaborative frequency regulation control strategy is set based on the frequency safety and stability index.

[0161] From the above analysis, it can be seen that the maximum frequency change rate of the system is only related to the inertia of the gas turbine. Usually, the inertia of the gas turbine in the system is fixed and can meet the frequency stability requirements. At the same time, the steady-state frequency deviation of the system reflects the frequency stability under long-term scales. The present invention mainly focuses on the frequency stability problem under transient conditions. Therefore, the present invention only considers the maximum frequency deviation of the system and uses it as a stratified index for wind, storage, and load frequency regulation. During the frequency regulation process, if the system frequency deviation is detected to exceed the stratified index and then the corresponding resources are controlled to participate in the frequency regulation, the frequency regulation capacity of each resource cannot be fully utilized, which may cause the frequency to dynamically exceed the limit. Therefore, it is necessary to predict the maximum frequency deviation of the system in advance so as to control each resource to participate in the frequency regulation in advance. In the actual operation of the microgrid, it is usually difficult to measure the power shortage of the system after a fault in a timely manner. The initial frequency change rate of the system can be measured in real time by the wide-area measurement system. Therefore, by detecting the initial frequency change rate of the system, the maximum frequency deviation of the system can be predicted and compared with each stratified index to determine whether each resource participates in the frequency regulation, thereby realizing the model predictive control of wind-storage-load frequency regulation. The wind-storage-load collaborative control process based on model prediction is as follows: Figure 8 shown.

[0162] When the system frequency changes, the gas turbine responds in real time;

[0163] When the system maximum frequency deviation is predicted to exceed the stratification index Δf max1 When , the wind turbine adopts virtual inertia control to reduce the system frequency change;

[0164] When the system maximum frequency deviation is predicted to exceed the stratification index Δf max2 When , the energy storage system adopts droop control to reduce the system frequency deviation;

[0165] When the system maximum frequency deviation is predicted to exceed the stratification index Δf max3 When , the controllable load adopts droop control, which reduces the system frequency deviation by increasing or decreasing the load power;

[0166] Depend on Figure 5 It can be seen that when the doubly fed wind turbine participates in frequency regulation, the wind turbine speed gradually recovers during the frequency recovery process, and it also absorbs energy from the power grid, which may cause a secondary drop in frequency, which is not conducive to system frequency recovery. Therefore, the system needs to provide additional power support during the wind turbine speed recovery process. In the wind-storage-load system, since the energy storage system has a fast adjustment speed and can provide a large amount of power support in a short time, the energy storage system power is adjusted to compensate for the energy absorbed by the wind turbine until the wind turbine returns to its initial operating state. Subsequently, the energy storage system begins to smoothly exit frequency regulation, and finally all the frequency regulation power is provided by the gas turbine. After detecting that the system frequency change rate is 0, the frequency regulation control equation of the energy storage system is corrected to:

[0167]

[0168] Where ΔP′ E is the power regulation of the energy storage system after correction, and Δf(s) is the frequency change in the complex frequency domain.

[0169] In order to verify the effectiveness of the above-mentioned microgrid wind-storage-load layered coordinated frequency modulation control method, the present invention built the following on Matlab / Simuink: Figure 2 The microgrid simulation model shown in Figure 1 is as follows; the gas turbine capacity is 9MW; the doubly fed wind turbine capacity is 6MW, and the wind turbine load reduction power ratio is 0.9; the energy storage system capacity is set to 20% of the wind power capacity, which is 1.2MW·h; the load size is 10MW, and the controllable load accounts for 20%; the frequency regulation model parameters of each resource are shown in Table 1; the hierarchical index Δf is set max1 , Δf max2 , Δf max3 0.25Hz, 0.5Hz, and 0.75Hz respectively.

[0170] Table 1 Frequency modulation model parameters of each resource

[0171] Model parameters Value / s Model parameters Numerical <![CDATA[H m ]]> 4 <![CDATA[D m ]]> 1 <![CDATA[T m ]]> 3.8 <![CDATA[K m ]]> 20 <![CDATA[T w ]]> 0.1 <![CDATA[K w ]]> 40 <![CDATA[T E ]]> 0.3 <![CDATA[K E ]]> 50 <![CDATA[T L ]]> 1 <![CDATA[K L ]]> 45

[0172] 1. System frequency response results under different load disturbances

[0173] For different maximum frequency drop values, the frequency modulation resources invested by the system are also different. In order to analyze the frequency modulation effect of each resource, the present invention sets three scenarios in which the system load suddenly increases by 0.5MW, 1.5MW and 2.5MW in 5s. The maximum frequency deviation of the system is predicted based on the measured system initial frequency change rate. The prediction results are shown in Table 2. Under the three scenarios, the frequency response curves of the system are as follows Figures 9 to 11 shown.

[0174] Table 2 Output of thermal power units

[0175]

[0176] As shown in Table 2, when the load suddenly increases by 0.5MW, if only the gas turbine participates in frequency regulation, the predicted maximum frequency deviation of the system is 0.394Hz, which is greater than the stratification index Δf max1 , so wind power is needed to participate in frequency regulation. Figure 9It can be seen that before wind power participated in frequency regulation, the system frequency dropped significantly, with the maximum deviation reaching 0.395Hz, which is basically consistent with the predicted value. After wind power participated in frequency regulation, under the virtual inertia control of the wind turbine, the maximum frequency deviation of the system was reduced to 0.221Hz, and the frequency drop rate was slowed down, improving the frequency stability of the system. When the load suddenly increased by 1.5MW, if only gas turbines and wind power participated in frequency regulation, the predicted maximum frequency deviation was 0.666Hz, requiring energy storage to participate in frequency regulation. Figure 10 It can be seen that after energy storage participates in frequency regulation, the system's maximum frequency deviation is reduced to 0.485Hz. At the same time, the system's steady-state frequency deviation is reduced, and the frequency stabilization time is reduced to 20s. When the load suddenly increases by 2.5MW, if the gas turbine, wind power, and energy storage participate in frequency regulation, the predicted maximum frequency deviation is 0.814Hz, requiring controllable load to participate in frequency regulation. Figure 11 It can be seen that after the controllable load participates in frequency modulation, the maximum frequency deviation of the system is reduced to 0.682Hz, and the steady-state frequency deviation is also reduced.

[0177] In summary, the system maximum frequency deviation prediction model established by the present invention has high accuracy and can predict the system maximum frequency deviation, thereby controlling the corresponding resources to participate in frequency modulation in advance, reducing frequency drops, and improving system frequency stability.

[0178] 2. Comparison with traditional hierarchical control strategy

[0179] To further verify the effectiveness of the hierarchical control strategy based on model prediction, the present invention compares this strategy with the traditional hierarchical control strategy. The simulation results are as follows: Figures 12 to 14 Among them, the traditional hierarchical control strategy is that each frequency modulation resource in the system automatically participates in frequency modulation after detecting that the frequency deviation exceeds the threshold in real time.

[0180] Depend on Figures 12 to 14 It can be seen that under the model prediction-based hierarchical control strategy, the system frequency deviation is significantly reduced compared to the traditional hierarchical control strategy, the frequency change rate is reduced, and the frequency stabilization time is also shortened. Taking a 0.5MW load surge as an example, when the load surge is 0.5MW, under the traditional hierarchical control strategy, the wind turbine will not participate in frequency regulation until the frequency deviation exceeds 0.25Hz. However, the control strategy proposed in this invention can control the wind turbine to participate in frequency regulation in advance, reducing the system frequency deviation.

[0181] Therefore, the hierarchical control strategy based on model prediction can control the corresponding resources to participate in frequency modulation in advance, avoid excessive frequency drops in the system, and improve the frequency stability of the system.

[0182] 3. Verification of the effectiveness of fan speed recovery control strategy

[0183] In order to verify the effect of energy storage power compensation in the wind turbine speed recovery stage, the present invention compares the frequency response results of the system before and after energy storage compensation when the load suddenly increases by 2.5MW. Figure 15 shown. Figures 16 to 18 The power changes of the wind turbine and energy storage as well as the speed changes of the wind turbine are shown respectively after the energy storage compensates the power.

[0184] Depend on Figures 15 to 17 As can be seen, after the system frequency reaches its lowest point and enters the recovery phase, the wind turbines absorb energy from the system, hindering frequency recovery. The system frequency stabilizes at 20 seconds. After energy storage compensation, the system frequency stabilization time is shortened to 15 seconds, frequency fluctuations are reduced, and recovery speed is increased.

[0185] Depend on Figure 16 and Figure 18 It can be seen that when the wind turbine operates in overspeed load reduction mode, the wind turbine speed decreases, and the mechanical power of the wind turbine increases, providing a certain amount of frequency modulation power and preventing the wind turbine speed from dropping too much. At the same time, during the speed recovery process, the higher mechanical power also increases the rotor's acceleration torque, speeding up the rotor's recovery and providing energy for the rotor speed recovery, reducing the energy absorbed by the wind turbine from the system.

[0186] In summary, the microgrid wind-storage-load layered collaborative frequency regulation control method proposed in the present invention can control wind turbines, energy storage and controllable loads to participate in frequency regulation in advance, thereby improving the frequency stability of the microgrid.

Claims

1. A microgrid wind-storage-load layered coordinated frequency modulation control method, characterized in that: The method comprises the following steps: S1. Based on the operating characteristics of various resources in the microgrid system, a wind-storage-load layered collaborative frequency regulation control architecture is established; S2. Establish frequency regulation control models for gas turbines, doubly-fed wind turbines, energy storage systems, and controllable loads, and establish an overall frequency response model for the microgrid system: In the formula, Δf is the system frequency change, ΔP0 is the system load disturbance, H s is the system inertial response transfer function, G T , G w , G E , G L are the gas turbine, doubly-fed fan, energy storage system, and controllable load frequency response transfer function, u w 、u E 、u L are the double-fed wind turbine, energy storage system, and controllable load frequency regulation state variables, respectively. m is the inertia time constant of the gas turbine, D m is the damping coefficient of the gas turbine, K m is the power-frequency characteristic coefficient of the gas turbine, T m is the response time constant of the gas turbine, K w is the virtual inertia coefficient, T w is the response time constant of the doubly fed wind turbine, K E is the droop coefficient of the energy storage system, T E is the response time constant of the energy storage system, K L is the controllable load droop coefficient, T L is the response time constant of the controllable load, S is a complex variable; S3. Establish a prediction model for the frequency response characteristic index of the microgrid system, and use the maximum frequency deviation as the frequency stratification index. Propose a wind-storage-load stratified coordinated frequency regulation control strategy based on the prediction model. In the wind turbine speed recovery stage, the energy storage system is used to compensate for the power absorbed by the wind turbine. The characteristic indicators of system frequency response include maximum frequency change rate, maximum frequency deviation and steady-state frequency deviation; When the load is disturbed, the maximum frequency change rate of the system is obtained by the initial value theorem for: The system steady-state frequency deviation Δf is obtained by the final value theorem ∞ for: The frequency modulation control model of each resource in the system is equivalent to an equivalent frequency modulation control model with the same characteristics. The transfer function G of the equivalent frequency modulation control model is eq for: In the formula, A eq , B eq , C eq are the characteristic parameters of the equivalent model; t mid It is half of the steady-state time-weighted average value of each resource frequency modulation; They are the frequency regulation steady-state time of gas turbine, doubly-fed fan, energy storage system and controllable load respectively; The time series expression of the frequency deviation of the system equivalent frequency response model Δf(t) is: c=6H s C eq (a 2 -b 2 )+2a(2H s +A eq +C eq D eq )+B eq +D eq d=12abH s C eq +2b(2H s +A eq +C eq D eq ) g=(aC eq +1)c+bdC eq h=bcC eq -(a.C eq +1) d Where, M, N, a, b, c, d, h, g are all frequency timing expression parameters; D eq is the system equivalent damping coefficient; The maximum frequency deviation of the system is at time t tir and the maximum frequency deviation Δf max for:

2. A microgrid wind-storage-load layered coordinated frequency modulation control method according to claim 1, characterized in that: In step S1, the operating characteristics of the gas turbine, wind turbine, energy storage system and controllable load in the microgrid system are analyzed, and a wind-storage-load layered collaborative frequency regulation control architecture is established, and the frequency regulation is divided into four layers: the first layer is the gas turbine frequency regulation; the second layer is the wind turbine frequency regulation; the third layer is the energy storage system frequency regulation; and the fourth layer is the controllable load frequency regulation.

3. A microgrid wind-storage-load layered coordinated frequency modulation control method according to claim 1, characterized in that: In step S2, the gas turbine frequency regulation control model includes an inertial response model and a primary frequency regulation model; The inertial response model is: Where ΔP Gm is the change in mechanical torque of the gas turbine, ΔP Ge is the change of electromagnetic torque, H m is the inertia time constant of the gas turbine, Δf is the system frequency change, t is time, D m is the damping coefficient of the gas turbine; The primary frequency modulation model is: In the formula, K m is the power-frequency characteristic coefficient of the gas turbine, T m is the response time constant of the gas turbine, s is a complex variable, and Δf(s) is the frequency change in the complex frequency domain.

4. A microgrid wind-storage-load layered coordinated frequency modulation control method according to claim 1, characterized in that: In step S2, the fan frequency regulation control model includes a virtual inertia control model and an overspeed load reduction control model; The virtual inertia control model is: Where ΔP W is the power regulation of the doubly fed wind turbine, K w is the virtual inertia coefficient, T w is the response time constant of the doubly fed wind turbine, S is a complex variable, Δf(s) is the frequency change in the complex frequency domain; The overspeed load reduction control model is: Where P Wm is the mechanical power output of the wind turbine, C p is the wind energy conversion coefficient, ρ a is the air density, S w is the wind turbine blade area, V ω is the wind speed; Using overspeed load reduction control, the fan power-speed curve under different wind speeds meets the following requirements: P.S opt JP max ·η Where P max is the maximum power, P opt is the overspeed load reduction power, η is the load reduction power ratio; In overspeed load reduction mode, the double-fed wind turbine has a margin of ΔP opt The power involved in frequency modulation: ΔP opt =P max (1st) In the formula, ΔP opt The reserved frequency regulation power for the doubly-fed wind turbine.

5. A microgrid wind-storage-load layered coordinated frequency modulation control method according to claim 1, characterized in that: In step S2, the frequency regulation control model of the energy storage system is: In the formula, λ soc is the state of charge of the energy storage system, λ soc0 is the initial charge state of the energy storage system, P E is the output power of the energy storage system, E N is the rated capacity of the energy storage system; Using droop control to provide power support for microgrids: In the formula, ΔP E is the power regulation of the energy storage system, K E is the droop coefficient of the energy storage system, T E is the response time constant of the energy storage system, s is a complex variable, and Δf(s) is the frequency change in the complex frequency domain.

6. A microgrid wind-storage-load layered coordinated frequency modulation control method according to claim 1, characterized in that: In step S2, the controllable load frequency regulation control model is: Where ΔP L is the power regulation of the controllable load, K L is the controllable load droop coefficient, T L is the response time constant of the controllable load, s is a complex variable, and Δf(s) is the frequency change in the complex frequency domain.

7. A microgrid wind-storage-load layered coordinated frequency modulation control method according to claim 1, characterized in that: In step S3, when the system frequency changes, the gas turbine responds in real time; When the system maximum frequency deviation is predicted to exceed the stratification index Δf max 1 When the wind turbine is in a state of inertia, it uses virtual inertia control to reduce the frequency change of the system; When the system maximum frequency deviation is predicted to exceed the stratification index Δf max 2 When , the energy storage system adopts droop control to reduce the system frequency deviation; When the system maximum frequency deviation is predicted to exceed the stratification index Δf max 3 When the load is controlled, droop control is adopted to reduce the system frequency deviation by increasing or decreasing the load power.

8. A microgrid wind-storage-load layered coordinated frequency modulation control method according to claim 7, characterized in that: In step S3, the energy absorbed by the fan is compensated by regulating the power of the energy storage system until the fan returns to the initial operating state. Then, the energy storage system begins to smoothly exit the frequency modulation, and the frequency modulation power is finally provided entirely by the gas turbine. After detecting that the system frequency change rate is 0, the frequency modulation control equation of the energy storage system is corrected to: Where ΔP′ E is the power regulation of the energy storage system after correction, and Δf(s) is the frequency change in the complex frequency domain.

Images