CN115954894A

CN115954894A - Energy storage primary frequency modulation self-adaptive comprehensive control method based on weight coefficient and optimized frequency modulation control

Info

Publication number: CN115954894A
Application number: CN202310008726.9A
Authority: CN
Inventors: 熊林云; 何亚兰; 朱银方; 郭世威; 何东林; 班昌宇; 汪清德; 宋瑞恺
Original assignee: Chongqing University
Current assignee: Chongqing University
Priority date: 2023-01-04
Filing date: 2023-01-04
Publication date: 2023-04-11

Abstract

The energy storage primary frequency modulation self-adaptive comprehensive control method based on the weight coefficient and the optimized frequency modulation control comprises the following steps of: 1) Monitoring the frequency of a power system and the charge state of an energy storage system in real time; 2) Dividing the state of charge of the energy storage system into a plurality of intervals; 3) Constructing an energy storage output model based on virtual inertia control and an energy storage output model based on droop control; 4) Constructing a charge-discharge coefficient model of the energy storage system during frequency modulation through variable coefficient droop control; 5) Constructing an optimal frequency modulation control charge-discharge coefficient model when the energy storage system participates in frequency modulation through variable coefficient droop control; 6) Constructing an adaptive comprehensive control model; 7) Calculating to obtain a virtual inertia control charge-discharge coefficient; 8) Constructing two energy storage output models for controlling the cooperative output; 9) Constructing different distribution coefficient models; 10 Substituting the obtained distribution coefficient into the energy storage output model in the step 8), and calculating to obtain the output of the energy storage system, thereby realizing the frequency modulation control of the energy storage system.

Description

Energy storage primary frequency modulation self-adaptive comprehensive control method based on weight coefficient and optimized frequency modulation control

Technical Field

The invention relates to the technical field of energy storage system participation primary frequency modulation, in particular to an energy storage primary frequency modulation self-adaptive comprehensive control method based on weight coefficients and optimized frequency modulation control.

Technical Field

With the change of global energy forms, new energy sources such as wind power, photovoltaic and the like are rapidly generated and gradually replace fossil fuels. Because new energy has characteristics of volatility and uncertainty, a serious challenge is brought to the safety of a power system with high penetration rate of the new energy, particularly the frequency safety of the system. Because the system frequency stability is established on the basis of the balance of active power of a power supply side and a load side, the output power of a power grid with high new energy occupation ratio such as wind power and photovoltaic is difficult to ensure to meet the requirement, and the problem of the system frequency stability is caused. As a new frequency modulation mode, the energy storage system has the advantages of high response speed, high control precision, bidirectional adjustment, accurate tracking and the like, and is widely researched in the aspect that energy storage participates in primary frequency modulation.

In view of the control strategy of the energy storage system participating in the primary frequency modulation of the power system, researchers at home and abroad carry out deep research on the control strategy, the frequency modulation model and other aspects. The scholars provide a regional power grid frequency modulation control strategy containing an energy storage system, and verify that the energy storage system can effectively improve the frequency dynamic characteristic. The virtual inertia control and the virtual droop control are used as main modes of the energy storage system participating in primary frequency modulation, and the virtual inertia control and the virtual droop control can effectively inhibit the frequency change rate of the system and effectively inhibit the maximum frequency deviation; the latter can reduce the steady state deviation of the system and ensure the frequency stability of the system. Meanwhile, virtual inertia control and droop control are adopted as the primary frequency modulation mode of the energy storage system, and the primary frequency modulation capacity of the energy storage system can be effectively improved. In addition, the virtual inertia control can be divided into positive virtual inertia and negative virtual inertia control according to the positive and negative frequency change rates, and the energy storage output is adjusted by combining droop control. In order to prolong the service life and improve the economic feasibility of the energy storage battery, the influence of the energy storage charge state on the charge-discharge coefficient is widely concerned.

In addition, due to the fact that the virtual inertia and droop control have different advantages on the energy storage frequency modulation, the problem that the energy storage system participates in primary frequency modulation is that the weight coefficients are reasonably distributed in two control modes at different frequency modulation stages. And correspondingly, the frequency modulation control strategy of the virtual inertia control and droop control adaptive factors with fuzzy control and the corresponding weight factor control strategy are distributed according to different stages of energy storage frequency modulation. And related researches optimize the weight coefficient distribution mode of positive virtual inertia, negative virtual inertia and droop control.

In the above control strategy, the control modes involved in the distribution of the weight coefficients of the virtual inertia control and the droop control of the energy storage system in different frequency modulation stages and the switching boundaries between the 2 control modes are different, but the self-adaptability of the setting of the relevant parameters of the disturbance with different amplitudes is not high, and the problems of energy storage power jump, frequency secondary disturbance and the like are difficult to avoid. At present, most of control modes for adjusting the energy storage charge and discharge coefficients by considering the energy storage charge state are similar, the energy storage charge state is controlled in a partitioning mode, the overcharge and overdischarge of an energy storage system can be effectively avoided, the energy storage charge and discharge coefficients under special conditions are not optimally designed, and the self-adaptability of energy storage with the maximum charge and discharge coefficient is low. And when the stored energy is output with the maximum charge-discharge coefficient, the corresponding charge state partition is too large, so that the energy storage frequency modulation possibly misses the optimal control opportunity, and the frequency response speed of the system is reduced.

Disclosure of Invention

The invention aims to provide an energy storage primary frequency modulation self-adaptive comprehensive control method based on weight coefficients and optimized frequency modulation control, which comprises the following steps of:

1) And monitoring the frequency of the power system and the state of charge of the energy storage system in real time.

2) And dividing the charge state of the energy storage system into a plurality of intervals according to the output standards in different capacity states.

3) And constructing an energy storage output model based on virtual inertia control and an energy storage output model based on droop control.

4) And constructing a charge-discharge coefficient model when the energy storage system participates in frequency modulation through variable coefficient droop control based on the charge state interval of the energy storage system.

5) And constructing an optimal frequency modulation control charge-discharge coefficient model when the energy storage system participates in frequency modulation through variable coefficient droop control based on the charge state interval of the energy storage system.

6) And integrating the charge-discharge coefficient of the variable coefficient droop control participating in frequency modulation with the optimized frequency modulation control charge-discharge coefficient, and constructing a self-adaptive comprehensive control model of the energy storage system participating in frequency modulation through the variable coefficient droop control.

7) And calculating to obtain a virtual inertia control charge-discharge coefficient based on the variable coefficient droop control charge-discharge coefficient.

8) And analyzing the virtual inertia control method and the droop control method, and constructing an energy storage output model of virtual inertia control and droop control cooperative output through the variable coefficient droop control charge-discharge coefficient and the virtual inertia control charge-discharge coefficient.

9) And aiming at different frequency change stages of the energy storage system participating in primary frequency modulation, different distribution coefficient models are constructed, and the distribution coefficient models are solved to obtain a weight coefficient a of virtual inertia control and a weight coefficient b of variable coefficient droop control.

10 Substituting the weight coefficient a of the virtual inertia control and the weight coefficient b of the variable coefficient droop control into the energy storage output model in the step 8), and calculating to obtain the output of the energy storage system, thereby realizing the frequency modulation control of the energy storage system.

Further, in step 1), the state of charge of the energy storage system includes a charging state and a discharging state.

When the power system generates positive frequency deviation, the energy storage system is in a charging state. When the power system generates negative frequency deviation, the energy storage system is in a discharging state.

Furthermore, the state of charge of the energy storage system is divided into regions according to the output standard under different capacity states, an S-shaped piecewise function based on the self-adaptive rule with the state of charge of the energy storage system as an independent variable and a charge-discharge coefficient as a dependent variable is adopted, and the divided regions of the state of charge of the energy storage system comprise (0, S) _min )、[S _min ，S _low )、[S _low ，S _high )、[S _high ，S _max )、[S _max 1), wherein S _min Is the minimum value of the state of charge partition. S. the _low A lower value for the state of charge partition. S. the _high Higher values for state of charge partitions. S _max The maximum value of the state of charge partition.

Further, the energy storage output model corresponding to the virtual inertia control is as follows:

where a is a weight coefficient of the virtual inertia.

Is the rate of change of frequency deviation. M _E Is the virtual inertia coefficient. Wherein the virtual inertia coefficient M _E Related to the state of charge of the energy storage system.

The energy storage output model corresponding to the variable coefficient droop control is as follows:

in the formula,. DELTA.P _E2 And controlling the energy storage output participating in frequency modulation for the variable coefficient droop. b is a weight coefficient for variable coefficient droop control, where a + b =1.Δ f is the frequency deviation. Δ f _db For the frequency deviation dead zone, when the frequency deviation Deltaf is smaller than the frequency deviation dead zone Deltaf _db When it is time, the system does not tune. K is _E The frequency modulation factor is controlled for droop.

In the formula, K _C And K _D The charge and discharge droop control coefficients are respectively.

Further, the energy storage system controls a charge-discharge coefficient model participating in frequency modulation through variable coefficient droop as follows:

in the formula: k is _max Is the maximum value of the virtual droop control coefficient. K _C1 And K _D1 The coefficients are the charge and discharge coefficients of the variable coefficient droop control strategy. And S represents the real-time monitored state of charge value of the energy storage system. And n is the adaptive coefficient of the curve.

Further, the model for optimizing the frequency modulation control charge-discharge coefficient is as follows:

in the formula, ω is a curve adjustment coefficient of the virtual inertia coefficient and the droop control coefficient. K _C2 And K _D2 The charging and discharging coefficients of the frequency control method are optimized respectively. S ₀ Represents a low intermediate value and has a value range of 0.1<S ₀ <0.45。S ₁ Represents a high intermediate value and has a value range of 0.55<S ₀ <0.9。

Further, the adaptive comprehensive control model is as follows:

K _C ＝K _C1 +K _C2 Δf＞0(8)

K _D ＝K _D1 +K _D2 Δf＜0(9)

integrating equation (4) to equation (9), the adaptive comprehensive control model is as follows:

further, the virtual inertia coefficient M of the virtual inertia control method _E The calculation formula of (c) is as follows:

M _E ＝λK _E (12)

in the formula: λ is the proportionality coefficient of droop control to virtual inertia.

Further, the energy storage output model of the virtual inertia control and droop control cooperative output is as follows:

where a is a weight coefficient of the virtual inertia. b is a weight coefficient for variable coefficient droop control. Δ f is the frequency deviation. Δ f _db In order to have a frequency deviation dead zone,

is the rate of change of frequency deviation. M _E Is the virtual inertia coefficient. K _E The frequency modulation factor is controlled for droop.

Further, the distribution coefficient analytical model is as follows:

when the frequency deviation delta f is not less than the frequency deviation dead zone delta f _db At a rate of change with frequency

The related virtual inertia control is mainly used, the droop control is used as an auxiliary, the energy storage output is correspondingly adjusted, and the distribution coefficient analytical model constructed by the method is as follows:

in the formula: and the adaptive coefficient of the distribution coefficient analysis model takes the vertical control below m as the main control and takes the virtual inertia control as the auxiliary control. a is the weight coefficient of the virtual inertia. b is a weight coefficient for variable coefficient droop control.

When the frequency deviation delta f reaches the set value delta f of the inertial response stage _set Later, the energy storage frequency modulation is mainly transited from virtual inertia control to droop control, and the distribution coefficient analytical model corresponding to the energy storage frequency modulation is as follows:

/>

in the formula: Δ f _low 、Δf _max Respectively representing the threshold value of the energy storage system participating in the primary frequency modulation and the maximum frequency deviation value of the primary frequency modulation of the energy storage system.

The technical effect of the present invention is undoubtedly that the present invention mainly uses virtual inertia frequency modulation at the initial stage of frequency change, and when a certain frequency setting value is reached, the present invention switches to adaptive frequency modulation control mainly using variable coefficient droop control. In addition, the energy storage charge state is comprehensively considered, and an S-shaped piecewise function based on the self-adaptive rule is constructed by taking the energy storage charge state as an independent variable and taking a charge-discharge coefficient as a dependent variable. The invention can effectively reduce the frequency fluctuation amplitude, avoid the frequency deterioration, accelerate the frequency recovery, and simultaneously smooth the energy storage output to avoid the energy shortage or out-of-limit of the energy storage charge state.

The invention provides a self-adaptive comprehensive control strategy which considers the characteristics of frequency change rate and frequency deviation of energy storage in different frequency modulation stages and further perfects the control of the maximum discharge rate and the maximum charge rate of the energy storage system when the state of charge is too large or too small, and is suitable for primary frequency modulation of a power grid.

According to the invention, corresponding control functions are constructed according to the energy storage state of charge in a partitioning manner, and an optimization control strategy is provided for the energy storage output under the two conditions of too high or too low energy storage state of charge, so that a self-adaptive control function is constructed, and thus, the frequency change can be controlled rapidly.

According to the method, different weight coefficients are configured according to different stages of energy storage frequency modulation, the frequency response speed of the system is improved mainly by virtual inertia control in the initial stage of frequency disturbance, and the deviation of an energy storage system is reduced; and (4) degrading the system frequency into a switching boundary, and constructing an output mode mainly based on droop control to control the system frequency until the system frequency reaches a steady-state frequency deviation.

Drawings

FIG. 1 is a regional power grid frequency modulation equivalent model of the energy storage system of the present invention participating in primary frequency modulation;

FIG. 2 illustrates the energy storage cell state of charge partitioning according to the present invention;

FIG. 3 is a graph of the virtual droop charge-discharge coefficient of the present invention;

FIG. 4 is a timing diagram illustrating an optimized frequency control operation according to the present invention;

FIG. 5 is a diagram of the optimized frequency control coefficients of the present invention;

FIG. 6 is a diagram illustrating the charging and discharging coefficients of the energy storage adaptive integrated frequency modulation according to the present invention;

FIG. 7 is a graph of the primary frequency modulation curve of the power system of the present invention, wherein FIG. 7 (a) shows a sudden load increase and FIG. 7 (b) shows a sudden load decrease;

FIG. 8 is a graph of the weighting coefficients for different inertia phases m according to the present invention;

FIG. 9 is a graph of the distribution coefficients at different m's of the primary frequency modulation stage of the present invention;

FIG. 10 is a flow chart of the adaptive complex frequency modulation control of the energy storage system of the present invention;

fig. 11 is a frequency deviation, an energy storage state of charge and an energy storage system output change curve under step load disturbance, fig. 11 (a) is a frequency deviation change curve under step load disturbance, fig. 11 (b) is an energy storage state of charge change curve under step load disturbance, and fig. 11 (c) is an energy storage system output change curve under step load disturbance;

fig. 12 is a frequency deviation, an energy storage state of charge and an energy storage system output change curve under continuous load disturbance, fig. 12 (a) is a frequency deviation change curve under continuous load disturbance, fig. 12 (b) is an energy storage state of charge change curve under continuous load disturbance, and fig. 12 (c) is an energy storage system output change curve under continuous load disturbance.

Detailed Description

The present invention is further illustrated by the following examples, but it should not be construed that the scope of the above-described subject matter is limited to the following examples. Various substitutions and alterations can be made without departing from the technical idea of the invention and the scope of the invention is covered by the present invention according to the common technical knowledge and the conventional means in the field.

Example 1:

referring to fig. 1 to 12, the energy storage primary frequency modulation adaptive integrated control method based on weight coefficient and optimized frequency modulation control includes the following steps:

6) And integrating the charge-discharge coefficient of the variable coefficient droop control participating in frequency modulation with the charge-discharge coefficient of the optimized frequency modulation control, and constructing a self-adaptive comprehensive control model of the energy storage system participating in frequency modulation through the variable coefficient droop control.

In the step 1), the charge state of the energy storage system comprises a charge state and a discharge state.

Dividing the state of charge of the energy storage system into regions according to the output standard under different capacity states, adopting an S-shaped piecewise function based on an adaptive rule with the state of charge of the energy storage system as an independent variable and a charge-discharge coefficient as a dependent variable, and dividing to obtain the energy storage system state of charge region including (0, S) _min )、[S _min ，S _low )、[S _low ，S _high )、[S _high ，S _max )、[S _max 1), wherein S _min Minimum value for state of charge partition; s _low A lower value for state of charge partition; s _high A higher value for state of charge partition; s _max The maximum value of the state of charge partition.

The energy storage output model corresponding to the virtual inertia control is as follows:

where a is a weight coefficient of the virtual inertia.

in the formula,. DELTA.P _E2 And controlling the energy storage output participating in frequency modulation for the variable coefficient droop. b is a weight coefficient for variable coefficient droop control, where a + b =1.Δ f is the frequency deviation. Δ f _db For the frequency deviation dead zone, when the frequency deviation delta f is smaller than the frequency deviation dead zone delta f _db The system does not tune. K _E The frequency modulation factor is controlled for droop.

In the formula, K _C And K _D The charge and discharge droop control coefficients are provided.

When the frequency deviation delta f is larger than 0, the stored energy is in a charging state, and the overcharging condition exists in the stored energy under the current condition, so that when the SOC of the stored energy is larger than or equal to S _max The time-storage charging coefficient is 0; and conversely, when the charge state of the energy storage is smaller, the maximum charge coefficient is set to control the energy storage to optimize the frequency modulation effect. Along with the increase of the energy storage charge state, the energy storage charge coefficient is continuously reduced, and the charge speed is reduced.

In addition, when the frequency deviation delta f is less than 0, the possibility of over-discharging of the energy storage energy exists, so that when the monitored SOC of the energy storage is less than or equal to S _min And when the discharge coefficient is set to zero, the stored energy stops discharging. On the contrary, when the state of charge of the energy storage is larger, the energy storage is output by the maximum coefficient and follows the state of charge of the energy storageThe energy storage output is reduced according to a self-adaptive rule.

The charge-discharge coefficient model of the energy storage system during the frequency modulation through the variable coefficient droop control is as follows:

in the formula: k is _max Is the maximum value of the virtual droop control coefficient. K is _C1 And K _D1 The coefficients of charge and discharge of the variable coefficient droop control strategy are respectively. And S represents the real-time monitored state of charge value of the energy storage system. And n is a self-adaptive coefficient of the curve, and the value of the self-adaptive coefficient determines the change trend of the curve.

The optimized frequency modulation control is to control the charging coefficient K on the basis of variable coefficient droop control _C S is less than or equal to the energy storage SOC _high Time and discharge coefficient K _D In the state of charge of energy storage SOC is more than or equal to S _low And time is further subdivided, and the model for optimizing the frequency modulation control charge-discharge coefficient is as follows:

in the formula, ω is a curve adjustment coefficient of the virtual inertia coefficient and the droop control coefficient, and can be adjusted according to the power capacity configuration of the energy storage system. K _C2 And K _D2 The charging and discharging coefficients of the frequency control method are optimized respectively. S. the ₀ Represents a low intermediate value, and has a value range of 0.1<S ₀ <0.45。S ₁ Represents a high intermediate value, and has a value range of 0.55<S ₀ <0.9。

Charging and discharging coefficients controlled according to virtual inertia and variable coefficient droop are respectively equal to or less than S at energy storage SOC _high 、SOC≥S _low For the output situation when Δ f > 0 and SOC ≦ S _high Δ f < 0 and SOC ≥ S _low The optimized frequency control method is constructed under two conditions:

when storing energy S _low ≤SOC≤S _high The state of charge of the stored energy is in an ideal state, and the charge-discharge coefficient is K _max (ii) a In addition, when the frequency deviation- Δ f _db ≤Δf≤Δf _db The time energy storage is in the frequency modulation dead zone. And dividing the residual region into 6 blocks by the boundary line between the upper limit and the lower limit of the frequency modulation dead zone and the ideal state of the energy storage charge state.

When Δ f < - Δ f _db And SOC is more than or equal to S _high ，SOC≥S _high Can be further divided into that the energy storage SOC is more than or equal to S _max And S _high ＜SOC＜S _max . When the energy storage SOC is more than or equal to S _max And when the system frequency is adjusted, the energy storage capacity is charged to the maximum output power. When storing energy S _high ＜SOC＜S _max In the process, the falling speed of the energy storage charge state is too high due to too fast discharge, and the discharge coefficient needs to be adaptively controlled for the energy storage charge state.

When Δ f > Δ f _db And SOC is less than or equal to S _low When the energy storage SOC is less than or equal to S _min The system frequency is regulated by storing energy with maximum power. When storing energy S _min ＜SOC＜S _low At times, the charge factor is set to a larger value and gradually decreases as the state of charge of the energy storage increases.

The adaptive comprehensive control model is as follows:

K _C ＝K _C1 +K _C2 Δf＞0 (8)

K _D ＝K _D1 +K _D2 Δf＜0 (9)

virtual inertia coefficient M of the virtual inertia control method _E The calculation formula of (a) is as follows:

M _E ＝λK _E (12)

The energy storage output model of the virtual inertia control and droop control cooperative output is as follows:

is the rate of change of frequency deviation. M _E Is the virtual inertia coefficient. K is _E The frequency modulation factor is controlled for droop.

The distribution coefficient analytical model is as follows:

The related virtual inertia control is mainly carried out, the droop control is assisted, the energy storage output is correspondingly adjusted, and the distribution coefficient analytical model constructed by the method is as follows:

in the formula: and the adaptive coefficients of the distribution coefficient analysis model with the vertical control below m as the main control and the virtual inertia control as the auxiliary control. a is a weight coefficient of the virtual inertia. b is a weight coefficient for variable coefficient droop control.

in the formula: Δ f _low 、Δf _max Respectively representing the threshold value of the energy storage system participating in primary frequency modulation and the maximum frequency deviation value of the energy storage system primary frequency modulation.

The method mainly comprises the steps of controlling virtual inertia frequency modulation in the initial stage of frequency change, and switching to adaptive frequency modulation control mainly comprising variable coefficient droop control when a certain frequency set value is reached. In addition, the energy storage charge state is comprehensively considered, and an S-shaped piecewise function based on the self-adaptive rule is constructed by taking the energy storage charge state as an independent variable and taking a charge-discharge coefficient as a dependent variable. The method can effectively reduce the frequency fluctuation amplitude, avoid the frequency deterioration, accelerate the frequency recovery, and simultaneously smooth the energy storage output to avoid the energy shortage or the out-of-limit of the energy storage charge state.

Example 2:

Example 3:

the energy storage primary frequency modulation self-adaptive comprehensive control method based on the weight coefficient and the optimized frequency modulation control mainly comprises the steps of embodiment 2, wherein in the step 1), the charge state of the energy storage system comprises a charge state and a discharge state.

Example 4:

the energy storage primary frequency modulation self-adaptive comprehensive control method based on weight coefficient and optimized frequency modulation control mainly comprises the following steps of embodiment 2, wherein the charge state of an energy storage system is divided into regions according to output standards under different capacity states, and an S-shaped piecewise function based on a self-adaptive rule with the charge state of the energy storage system as an independent variable and a charge-discharge coefficient as a dependent variable is adoptedAnd obtaining the state of charge zone of the energy storage system after division to comprise (0, S) _min )、[S _min ，S _low )、[S _low ，S _high )、[S _high ，S _max )、[S _max 1), wherein S _min Minimum value for state of charge partition; s _low A lower value for state of charge partition; s _high A higher value for state of charge partition; s _max The maximum value of the state of charge partition.

Example 5:

the energy storage primary frequency modulation self-adaptive comprehensive control method based on the weight coefficient and the optimized frequency modulation control mainly comprises the following steps of embodiment 2, wherein an energy storage output model corresponding to the virtual inertia control is as follows:

where a is a weight coefficient of the virtual inertia.

Is the rate of change of frequency deviation. M is a group of _E Is the virtual inertia coefficient. Wherein the virtual inertia coefficient M _E Related to the state of charge of the energy storage system.

in the formula,. DELTA.P _E2 And controlling the energy storage output participating in frequency modulation for the variable coefficient droop. b is a weight coefficient for variable coefficient droop control, where a + b =1.Δ f is the frequency deviation. Δ f _db For the frequency deviation dead zone, when the frequency deviation delta f is smaller than the frequency deviation dead zone delta f _db When it is time, the system does not tune. K _E The frequency modulation factor is controlled for droop.

When the frequency deviation delta f is larger than 0, the stored energy is in a charging state, and the overcharging condition exists in the stored energy under the current condition, so that when the SOC of the stored energy is larger than or equal to S _max The time-storage charging coefficient is 0; and conversely, when the charge state of the energy storage is smaller, the maximum charge coefficient is set to control the energy storage to optimize the frequency modulation effect. Along with the increase of the energy storage charge state, the energy storage charge coefficient is continuously reduced, and the charge speed is slowed down.

In addition, when the frequency deviation delta f is less than 0, the possibility of over-discharging of the energy storage energy exists, so that when the monitored SOC of the energy storage is less than or equal to S _min And when the discharge coefficient is set to zero, the stored energy stops discharging. On the contrary, when the energy storage charge state is larger, the energy storage is output by the maximum coefficient, and the energy storage output is reduced according to the self-adaptive rule along with the reduction of the energy storage charge state.

Example 6:

an energy storage primary frequency modulation adaptive comprehensive control method based on weight coefficients and optimized frequency modulation control mainly comprises the following steps of embodiment 2, wherein a charge-discharge coefficient model of an energy storage system during frequency modulation through variable coefficient droop control is as follows:

in the formula: k is _max Is the maximum value of the virtual droop control coefficient. K is _C1 And K _D1 The coefficients of charge and discharge of the variable coefficient droop control strategy are respectively. And S represents the real-time monitored state of charge value of the energy storage system. And n is the adaptive coefficient of the curve, and the value of the adaptive coefficient determines the change trend of the curve.

Example 7:

energy storage based on weight coefficient and optimized frequency modulation controlThe primary frequency modulation self-adaptive comprehensive control method mainly comprises the following steps of embodiment 2, wherein the optimized frequency modulation control is to charge the coefficient K on the basis of variable coefficient droop control _C S is less than or equal to the energy storage SOC _high Time and discharge coefficient K _D In the state of charge of energy storage SOC is more than or equal to S _low And further subdividing the time, wherein the model for optimizing the frequency modulation control charge-discharge coefficient is as follows:

in the formula, ω is a curve adjustment coefficient of the virtual inertia coefficient and the droop control coefficient, and can be adjusted according to the power capacity configuration of the energy storage system. K _C2 And K _D2 The charging and discharging coefficients of the frequency control method are optimized respectively. S ₀ Represents a low intermediate value and has a value range of 0.1<S ₀ <0.45。S ₁ Represents a high intermediate value and has a value range of 0.55<S ₀ <0.9。

Charging and discharging coefficients controlled according to virtual inertia and variable coefficient droop are respectively equal to or less than S at energy storage SOC _high 、SOC≥S _low For the output condition of time, for delta f > 0 and SOC ≦ S _high Δ f < 0 and SOC ≥ S _low The optimized frequency control method is constructed under two conditions:

when storing energy S _low ≤SOC≤S _high The state of charge of the stored energy is in an ideal state, and the charge-discharge coefficient is K _max (ii) a In addition, when the frequency deviation- Δ f _db ≤Δf≤Δf _db The time-stored energy is in a frequency modulation dead zone. And dividing the residual region into 6 blocks by the boundary line between the upper limit and the lower limit of the frequency modulation dead zone and the ideal state of the energy storage charge state.

Example 8:

the self-adaptive comprehensive control method of the energy storage primary frequency modulation based on the weight coefficient and the optimized frequency modulation control mainly comprises the following steps of embodiment 2, wherein the self-adaptive comprehensive control model is as follows:

K _C ＝K _C1 +K _C2 Δf＞0 (8)

K _D ＝K _D1 +K _D2 Δf＜0 (9)

example 9:

the self-adaptive comprehensive control method of the energy storage primary frequency modulation based on the weight coefficient and the optimized frequency modulation control mainly comprises the following steps of embodiment 2, wherein the virtual inertia coefficient M of the virtual inertia control method _E The calculation formula of (a) is as follows:

M _E ＝λK _E (12)

Example 10:

the energy storage primary frequency modulation self-adaptive comprehensive control method based on the weight coefficient and the optimized frequency modulation control mainly comprises the following steps of embodiment 2, wherein an energy storage output model of the virtual inertia control and droop control cooperative output is as follows:

Example 11:

the energy storage primary frequency modulation self-adaptive comprehensive control method based on the weight coefficient and the optimized frequency modulation control mainly comprises the following steps of embodiment 2, wherein an analytic model of the distribution coefficient is as follows:

in the formula: and the adaptive coefficients of the distribution coefficient analysis model with the vertical control below m as the main control and the virtual inertia control as the auxiliary control. a is the weight coefficient of the virtual inertia. b is a weight coefficient for variable coefficient droop control.

When the frequency deviation delta f reaches the set value delta f of the inertial response stage _set Later, the energy storage frequency modulation is controlled from virtual inertiaThe main transition is made to droop control, and the corresponding distribution coefficient analytical model is as follows:

Example 12:

1) Monitoring the frequency of the power system and the state of charge of the energy storage system in real time;

2) Constructing a system frequency control model of energy storage participating in primary frequency modulation;

3) Dividing the charge state of the energy storage system into five regions according to the output characteristics of the energy storage system in different capacity states;

4) Analyzing two control strategies of virtual inertia control and droop control of the energy storage system participating in primary frequency modulation of the power grid, and constructing an adaptive function model of the virtual inertia control and droop control cooperative output;

5) Analyzing the process that the energy storage system participates in primary frequency modulation, and constructing a charge-discharge coefficient model of a variable coefficient droop control strategy of the energy storage system frequency modulation;

6) Aiming at two conditions of sudden increase of load and sudden decrease of energy storage state of charge higher than a higher value and sudden decrease of load and lower than a lower value, optimal frequency control is provided, and the charge and discharge coefficients of the energy storage state of charge are improved;

7) Integrating the variable coefficient droop control coefficient and the optimized frequency control coefficient to obtain a self-adaptive comprehensive control strategy model;

8) Obtaining a charge-discharge coefficient controlled by virtual inertia through analyzing a droop control charge-discharge coefficient model;

9) Analyzing the advantages of virtual inertia control and droop control in energy storage primary frequency modulation, and constructing a corresponding distribution coefficient analysis model;

10 Different distribution coefficient models are constructed aiming at different frequency change stages of the energy storage system participating in primary frequency modulation, and switching critical points of different distribution analysis models are set.

Monitoring the frequency deviation of the power system and the charge state of the energy storage system in real time, and charging the energy storage system when the system has positive frequency deviation; on the contrary, when the system generates negative frequency deviation, the energy storage system discharges, and the specific charging and discharging depth is determined by the self-adaptive control function corresponding to the state of charge of the energy storage system.

In order to construct an S-shaped piecewise function based on an adaptive rule and taking the energy storage charge state as an independent variable and a charge-discharge coefficient as a dependent variable, the threshold of the energy storage system charge state partition is set as S _min 、S _low 、S _high And S _max Which represent the minimum, lower, higher, and maximum values of the energy storage state of charge partition, respectively.

Energy storage output magnitude delta P of virtual inertia control of one control strategy of primary frequency modulation of energy storage system _E1 And (3) related to the virtual inertia coefficient and the frequency deviation change rate, so that an energy storage output model corresponding to the energy storage control strategy is constructed:

wherein a is a weight coefficient of the virtual inertia;

is the frequency deviation change rate (frequency change rate for short); m _E Is the virtual inertia coefficient. Wherein the virtual inertia coefficient M _E Related to the state of charge of the energy storage system.

Energy storage output delta P of energy storage system participating in frequency modulation through variable coefficient droop control _E2 The expression of (a) is as follows:

in the formula, K _E Controlling the frequency modulation coefficient for droop; Δ f is the frequency deviation; Δ f _db For frequency deviation dead zone, when the frequency is less than Δ f _db The system does not modulate frequency; k _C And K _D Respectively are charge and discharge droop control coefficients; b is a weight coefficient for variable coefficient droop control, and a + b =1.

The expression of the virtual droop charge-discharge coefficient with the energy storage charge state as the independent variable for the charge-discharge droop control coefficient in the formula (3) is shown as the following formula:

in the formula: k _max Is the maximum value of the virtual droop control coefficient; k _C1 And K _D1 The charge and discharge coefficients of the variable coefficient droop control strategy are respectively; s represents an energy storage state-of-charge value monitored by the system in real time; and n is the adaptive coefficient of the curve, and the value of the adaptive coefficient determines the change trend of the curve.

The optimized frequency control method can construct an expression of a charge-discharge coefficient according to the energy storage charge state, and comprises the following steps:

in the formula, omega is a curve adjustment coefficient of the virtual inertia coefficient and the droop control coefficient, and can be correspondingly adjusted according to the power capacity configuration of the energy storage system; k _C2 And K _D2 The coefficients of charge and discharge of the variable coefficient droop control strategy are respectively; n is consistent with the value of the adaptive coefficient of the variable coefficient droop control; s ₀ Represents a low intermediate value, and usually has a value in the range of 0.1 < S ₀ Less than 0.45; s ₁ Representing high intermediate values, typically in the range of 0.55 < S ₀ Is less than 0.9.

In the comprehensive frequency modulation stage of energy storage, the control method of the energy storage in the virtual inertia frequency modulation stage and the variable coefficient droop control stage is required to be combined, and the energy storage output delta P _E The expression is as follows:

the charge-discharge coefficient comprises two parts, namely the charge-discharge coefficient components according to formulas (9) to (10), wherein the charge-discharge coefficient components comprise a variable coefficient droop control coefficient and an optimized frequency modulation control coefficient.

K _C ＝K _C1 +K _C2 Δf＞0 (9)

K _D ＝K _D1 +K _D2 Δf＜0 (10)

The optimized frequency modulation control is to control the charging coefficient K on the basis of variable coefficient droop control _C S is less than or equal to the energy storage SOC _high Time and discharge coefficient K _D In the state of charge of energy storage SOC is more than or equal to S _low The charge and discharge are further subdivided, and finally the charge and discharge are integrated to obtain the results in the formulas (11) and (12) which are the charge and discharge constructed by the inventionAnd (3) a piecewise function control strategy with the coefficient changing along with the state of charge of the energy storage.

The calculation method of the virtual inertia charge-discharge coefficient is similar to the calculation method of the droop control coefficient, and can be expressed as the following equation:

M _E ＝λK _E (13)

In the initial stage of system frequency fluctuation in the inertial response stage, the energy storage output is adjusted correspondingly by taking virtual inertial control related to the frequency change rate d Δ fdt as a main control and taking droop control as an auxiliary control, so that a weight coefficient analysis model is constructed:

in the formula: and m is a self-adaptive coefficient, and the value of m determines the change trend of the weight coefficient curve.

The frequency deviation reaches the set value delta f of the inertial response stage _set And then, the energy storage frequency modulation is mainly transited from virtual inertia control to droop control, and the corresponding weight coefficient control function is as follows:

in the formula: Δ f _low 、Δf _max Respectively representing the threshold value of the energy storage participating in the primary frequency modulation and the maximum frequency deviation value of the energy storage primary frequency modulation; m is the same as the adaptive coefficient in equation (14).

Example 13:

1) And constructing a system model of the energy storage system participating in primary frequency modulation control, and researching an energy storage frequency modulation control strategy.

2) And performing primary frequency modulation self-adaptive comprehensive control on the battery energy storage system considering the charge state.

The method for carrying out primary frequency modulation self-adaptive comprehensive control on the battery energy storage system considering the charge state comprises the following steps: the method mainly comprises the steps of controlling virtual inertia frequency modulation in the initial stage of frequency change, and switching to adaptive frequency modulation control mainly comprising variable coefficient droop control when a certain frequency set value is reached. In addition, the energy storage charge state is comprehensively considered, and an S-shaped piecewise function based on the self-adaptive rule with the energy storage charge state as an independent variable and a charge-discharge coefficient as a dependent variable is constructed. The method can effectively reduce the frequency fluctuation amplitude, avoid the frequency deterioration, accelerate the frequency recovery, and simultaneously smooth the energy storage output to avoid the energy shortage or the out-of-limit of the energy storage charge state.

The method for constructing the charge-discharge coefficient model of the primary frequency modulation self-adaptive comprehensive control method of the battery energy storage system considering the charge state comprises the following steps of:

2.1 By virtual inertia coefficient M) _E And rate of change of frequency deviation

Constructing an energy storage output model corresponding to the virtual inertia control strategy:

wherein a is a weight coefficient of the virtual inertia;

2.2 By droop control factor K) _E And constructing an energy storage output model corresponding to the variable coefficient droop control strategy with the frequency deviation delta f:

in the formula, K _E Is a droop control coefficient; Δ f is the frequency deviation; Δ f _db For frequency deviation dead zone, when the frequency is less than Δ f _db The system does not modulate frequency; b is a weight coefficient for variable coefficient droop control, and a + b =1. Droop control coefficient K _E The method is divided into two parts according to different signs of frequency deviation:

in the formula, K _C And K _D The charge and discharge droop coefficients are provided, respectively.

Wherein, when the frequency deviation delta f is more than 0, the stored energy is in a charging state, and the stored energy has an overcharge condition under the current condition, so that when the SOC of the stored energy is more than or equal to S _max The time-storage charging coefficient is 0; and conversely, when the charge state of the energy storage is smaller, the maximum charge coefficient is set to control the energy storage to optimize the frequency modulation effect. Along with the increase of the energy storage charge state, the energy storage charge coefficient is continuously reduced, and the charge speed is reduced.

In the formula: k _max Is the maximum value of the virtual droop control coefficient; k _C1 The charging coefficient of the variable coefficient droop control strategy is S, which represents the energy storage charge state value monitored by the system in real time; and n is the adaptive coefficient of the curve.

In addition, when the frequency deviation delta f is less than 0, the possibility of over-discharging of the energy storage energy exists, so that when the monitored SOC of the energy storage is less than or equal to S _min And when the discharge coefficient is set to zero, the stored energy stops discharging. On the contrary, when the energy storage charge state is larger, the energy storage is in the maximum systemAnd outputting the data, and reducing the energy storage output according to a self-adaptive rule along with the reduction of the energy storage charge state. The expression of the virtual droop charge-discharge coefficient taking the energy storage charge state as an independent variable is shown as the following formula:

in the formula, K _D1 Is the discharge coefficient of the variable coefficient droop control strategy; s and n are the same as those in the formula (4).

2.3 Charge and discharge coefficients controlled according to the virtual inertia and the droop of the variable coefficient are respectively less than or equal to S at the SOC of the stored energy _high 、SOC≥S _low For the output situation when Δ f > 0 and SOC ≦ S _high Δ f < 0 and SOC ≥ S _low The optimized frequency control method is constructed under two conditions:

when storing energy S _low ≤SOC≤S _high The state of charge of the stored energy is in an ideal state, and the charge-discharge coefficient is K _max (ii) a In addition, when the frequency deviation- Δ f _db ≤Δf≤Δf _db The time-stored energy is in a frequency modulation dead zone. And dividing the rest area into 6 blocks by the boundary line between the upper limit and the lower limit of the frequency modulation dead zone and the ideal state of the energy storage charge state.

When Δ f < - Δ f _db And SOC is more than or equal to S _high ，SOC≥S _high Can be further divided into that the energy storage SOC is more than or equal to S _max And S _high ＜SOC＜S _max . When the energy storage SOC is more than or equal to S _max And when the system frequency is adjusted, the energy storage capacity is charged to the maximum output power. When storing energy S _high ＜SOC＜S _max In the process, the too fast discharge will result in the too fast drop speed of the energy storage charge state, and the self-adaptive control of the discharge coefficient to the energy storage charge state is needed. An expression of the discharge coefficient is constructed from the energy storage state of charge as follows:

in the formula, omega is a curve adjustment coefficient of the virtual inertia coefficient and the droop control coefficient, and can be adjusted according to the power of the energy storage systemThe capacity configuration is correspondingly adjusted; k _D2 Optimizing the discharge coefficient of the frequency control strategy; the adaptive coefficient n is the same as the formula (4); s ₀ Represents a low intermediate value, and usually has a value in the range of 0.1 < S ₀ Less than 0.45; s ₁ Representing high intermediate values, typically in the range of 0.55 < S ₀ Is less than 0.9.

When Δ f > Δ f _db And SOC is less than or equal to S _low When the energy storage SOC is less than or equal to S _min The system frequency is regulated by storing energy with maximum power. When storing energy S _min ＜SOC＜S _low When the charging coefficient is set to a large value, the charging coefficient gradually decreases with the increase of the energy storage state of charge.

Constructing an expression of the charge coefficient according to the energy storage state of charge as follows:

in the formula, K _C2 Is the charge factor that optimizes the frequency control strategy, the other factors being the same as in equation (6).

2.4 Control method for combining energy storage in virtual inertia frequency modulation stage and variable coefficient droop control stage, energy storage output delta P _E The expression is as follows:

/>

the charge and discharge coefficient comprises a variable coefficient droop control coefficient and an optimized frequency modulation control coefficient, namely:

K _C ＝K _C1 +K _C2 Δf＞0 (9)

K _D ＝K _D1 +K _D2 Δf＜0(10)

in the formula, K _C Is the charge factor; k _D Is the discharge coefficient.

The piecewise function control strategy of the charge-discharge coefficient changing along with the energy storage charge state constructed by the invention is as follows:

2.5 The calculation method of the virtual inertia charge-discharge coefficient is similar to the calculation method of the droop control coefficient, and the virtual inertia coefficient needs to be subjected to parameter correction and can be expressed as the following equation:

M _E ＝λK _E (13)

3) The method comprises the following steps of performing primary frequency modulation self-adaptive comprehensive control on a battery energy storage system considering the state of charge, wherein a self-adaptive comprehensive control weight coefficient distribution analytic model comprises the following steps:

3.1 In the initial stage of system frequency fluctuation, virtual inertia control related to the frequency change rate d Δ f/dt is mainly used, droop control is used for assisting in adjusting energy storage output to construct a weight coefficient analysis model as follows:

3.2 Frequency deviation reaches the inertial response stage setpoint Δ f _set And later, the energy storage frequency modulation is mainly transited from virtual inertia control to droop control. Constructing an energy storage output model taking variable coefficient droop control as a main part and virtual inertia control as an auxiliary part, wherein a weight coefficient control function corresponding to the energy storage output model is as follows:

in the formula,. DELTA.f _low 、Δf _max Respectively representing threshold value of energy storage participating in primary frequency modulation and energy storage primary frequency modulationA maximum frequency offset value.

And constructing corresponding control functions according to the energy storage state of charge partitions, proposing an optimization control strategy aiming at the energy storage output under the two conditions of overhigh or overlow energy storage state of charge, and constructing a self-adaptive control function, thereby being capable of quickly controlling the frequency change.

Different weight coefficients are configured according to different stages of energy storage frequency modulation, the frequency response speed of the system is improved mainly by virtual inertia control in the initial stage of frequency disturbance, and the deviation of an energy storage system is reduced; and (4) degrading the system frequency into a switching boundary, and constructing an output mode mainly based on droop control to control the system frequency until the system frequency reaches a steady-state frequency deviation.

Example 14:

referring to fig. 1 to 9, an adaptive integrated control strategy suitable for primary frequency modulation of a power grid considering the state of charge of energy storage includes the following steps:

1) The energy storage frequency modulation control strategy in the figure 1 is designed by taking the variable coefficient droop control of the charge state of the energy storage system into consideration and comprehensively considering the frequency modulation advantages of the droop control and the virtual inertia control. The equivalent model of the frequency modulation of the regional power grid with the energy storage system participating in primary frequency modulation is shown in figure 1. In the figure: s is a Laplace operator; k _G The droop coefficient of the traditional frequency modulation unit is obtained; delta P _L (s) is the load disturbance amount; delta F(s) is regional power grid frequency deviation; delta P _G (s) the output of the traditional frequency modulation unit; delta P _E (s) is the energy storage system output; m is a power grid inertia time constant; d is a load damping coefficient; f _HP Gaining a reheater of the steam turbine; t is a unit of _E Is the energy storage response time constant; t is _g 、T _RH And T _CH The time constant of the speed regulator, the time constant of the reheater and the time constant of the steam turbine of the traditional frequency modulation unit are respectively. E _rated For storing batteriesThe rated capacity of the system; s _in Is the initial state of charge of the energy storage system.

2) In order to determine the output standard during the energy storage frequency modulation, the energy storage needs to be divided into regions according to different charge states, and the division result is shown in fig. 2. Wherein S _min 、S _low 、S _high 、S _max The threshold values are respectively set to be 0.1, 0.45, 0.55 and 0.9.

3) The method for constructing the charging and discharging coefficient model of the primary frequency modulation self-adaptive comprehensive control method of the battery energy storage system considering the charge state comprises the following steps of:

3.1 The magnitude of the virtual inertia frequency modulation control energy storage output is related to the virtual inertia coefficient and the frequency deviation change rate, so that an energy storage output model corresponding to an energy storage control strategy is constructed:

wherein a is a weight coefficient of the virtual inertia;

is the frequency deviation change rate (frequency change rate for short); m _E Is the virtual inertia coefficient.

3.2 The response expression of the energy storage system participating in frequency modulation through variable coefficient droop control is as follows:

in the formula, K _E Is a droop control coefficient; Δ f is the frequency deviation; Δ f _db For frequency deviation dead zone, when the frequency is less than Δ f _db The system does not modulate frequency; k is _C And K _D Are respectively provided withIs the charge and discharge droop coefficient; b is a weight coefficient for variable coefficient droop control, and a + b =1.

The expression of the virtual droop charge-discharge coefficient taking the energy storage charge state as an independent variable is shown as the following formula:

in the formula: k _max Is the maximum value of the virtual droop control coefficient; s represents an energy storage state of charge value monitored by the system in real time; n is the adaptive coefficient of the curve, the value of which determines the variation trend of the curve, and the variation curve of the virtual droop control coefficient at different n is shown in fig. 3. It can be seen from fig. 3 that the curve changes steeply when the adaptive coefficient n is large; in addition, when the value of the adaptive coefficient n is small, the charge-discharge droop coefficient is small along with the energy storage charge state transformation.

3.3 Charge and discharge coefficients controlled according to the virtual inertia and the droop of the variable coefficient respectively have an energy storage SOC less than or equal to S _high 、SOC≥S _low For the output situation when Δ f > 0 and SOC ≦ S _high Δ f < 0 and SOC ≧ S _low The optimized frequency control method is constructed under two conditions:

when storing energy S _low ≤SOC≤S _high When the energy storage charge state is in an ideal state, the charge-discharge coefficient is K _max (ii) a In addition, when the frequency deviation- Δ f _db ≤Δf≤Δf _db The time-stored energy is in a frequency modulation dead zone. The remaining region is divided into 6 by the boundary line between the upper and lower limits of the frequency modulation dead zone and the ideal state of the energy storage charge state, and 4 regions when the energy storage works with the maximum charge-discharge coefficient are deeply analyzed, as shown in fig. 4.

When Δ f < - Δ f _db And SOC is more than or equal to S _high ，SOC≥S _high Can be further divided into that the energy storage SOC is more than or equal to S _max And S _high ＜SOC＜S _max . When the energy storage SOC is more than or equal to S _max And when the system frequency is adjusted, the energy storage capacity is charged to the maximum output power. When storing energy S _high ＜SOC＜S _max In the process, the falling speed of the energy storage charge state is too high due to too fast discharge, and the discharge coefficient needs to be adaptively controlled for the energy storage charge state. An expression of the discharge coefficient is constructed from the energy storage state of charge as follows:

in the formula, omega is a curve adjustment coefficient of the virtual inertia coefficient and the droop control coefficient, and can be correspondingly adjusted according to the power capacity configuration of the energy storage system; k is _D2 Optimizing the discharge coefficient of the frequency control strategy; the adaptive coefficient n is the same as the formula (4); s. the ₀ Representing low intermediate values, typically in the range of 0.1 < S ₀ Less than 0.45; s. the ₁ Representing high intermediate values, typically in the range of 0.55 < S ₀ Is less than 0.9.

When Δ f > Δ f _db And SOC is less than or equal to S _low When the energy storage SOC is less than or equal to S _min And regulating the system frequency by storing energy with the maximum power. When storing energy S _min ＜SOC＜S _low When the charging coefficient is set to a large value, the charging coefficient gradually decreases with the increase of the energy storage state of charge.

in the formula, K _C1 Is the charge factor that optimizes the frequency control strategy, the other factors being the same as in equation (6). Wherein S is ₀ And S ₁ The value of (2) will directly influence the variation trend of the curve, so S is the ratio ₀ And S ₁ 6 of each sample was analyzed, and the corresponding curve is shown in FIG. 5. S. the ₀ Too small of a value, S ₁ The large value can lead to poor self-adaptability of charge and discharge coefficients and is not beneficial to energy storage frequency modulation; otherwise, S ₀ Over-value, S ₁ If the value is too small, the energy storage output is too large, which may cause the energy storage state of charge to fall out of the ideal state quickly.

And the other 2 areas are all subjected to variable coefficient droop control.

3.4 Control method for energy storage in virtual inertia frequency modulation stage and variable coefficient droop control stage, energy storage output delta P _E The expression is as follows:

K _C ＝K _C1 +K _C2 Δf＞0 (9)

K _D ＝K _D1 +K _D2 Δf＜0 (10)

in the formula, K _C Is the charge factor; k is _D Is the discharge coefficient. The charge-discharge coefficient curve of the adaptive integrated frequency modulation control is shown in fig. 6.

3.5 ) the calculation method of the virtual inertia charge-discharge coefficient is similar to the calculation method of the droop control coefficient, and the virtual inertia coefficient needs to be subjected to parameter correction, which can be expressed as the following equation:

M _E ＝λK _E (13)

4) And performing primary frequency modulation self-adaptive comprehensive control on the battery energy storage system considering the charge state.

When the system frequency is disturbed, the primary frequency modulation process of the power system is shown in fig. 7, and the primary frequency modulation equation is shown in equation (9). In the case of a sudden load increase, during the period from the occurrence of the disturbance to the peak of the frequency deviation, the frequency deviation Δ f is negative and | Δ f | gradually increases, and the frequency change rate d Δ f/dt is negative and | d Δ f/dt | gradually decreases from a larger value to 0. The fast suppression of the frequency fluctuation should be the main factor in this stage, and the value of the virtual inertia weight coefficient a should be set to be large and gradually reduced with the suppression of the frequency fluctuation. After the frequency deviation recovers from the peak value, the frequency deviation delta f is still negative and is gradually reduced to be | delta f |, the frequency change rate d delta f/dt is positive, the frequency change rate is gradually increased from 0 and is finally reduced to zero again, the virtual inertia frequency modulation force should be weakened in the process, and the droop control frequency modulation is mainly used.

The self-adaptive comprehensive control weight coefficient distribution analytical model comprises the following steps:

4.1 In the initial stage of system frequency fluctuation, virtual inertia control related to the frequency change rate d Δ f/dt is mainly used, droop control is used for assisting in adjusting energy storage output to construct a weight coefficient analysis model as follows:

in the formula: m is an adaptive coefficient, and the value of m determines the variation trend of the weight coefficient curve, as shown in fig. 8. When the value of the exponential function corresponding to a in equation (14) is 0.5 or more, the value of a is determined by the exponential function. However, as the frequency deviation increases, the value of a is always equal to 0.5 when the value of the exponential function is less than 0.5.b is constructed similarly to a, when the system has frequency disturbance, the value of b increases exponentially from 0 to equal to 0.5, and when the calculated result is greater than 0.5, b =0.5. Regarding the adaptive coefficient m, when the coefficient is too small, the weight coefficient is less changed with the system frequency deviation; when the coefficient m is too large, the weight coefficient is too sensitive to the variation of the system frequency deviation.

4.2 Frequency deviation reaches the inertial response stage setpoint Δ f _set Later, energy storage frequency modulationAnd (4) transitioning from the virtual inertia control to the droop control. Constructing an energy storage output model taking variable coefficient droop control as a main part and virtual inertia control as an auxiliary part, wherein the weight coefficient control function corresponding to the energy storage output model is as follows:

in the formula,. DELTA.f _low 、Δf _max Respectively representing the threshold value of the energy storage participation primary frequency modulation and the maximum frequency deviation value of the energy storage primary frequency modulation. The value of m determines the variation trend of the weight coefficient, and 4 different values of m are taken for comparison, as shown in fig. 9. When m is too small, the frequency deviation interval corresponding to a = b =0.5 is too large; and when m is too large, the sensitivity of the weight coefficient to the frequency deviation is too high, and the capacity of the energy storage power supply cannot be optimally utilized.

5) A primary frequency modulation self-adaptive comprehensive control method for a battery energy storage system considering a charge state is disclosed.

An adaptive integrated control framework that takes into account the energy storage state of charge is constructed as shown in fig. 10. When the frequency deviation delta f of the system is less than or equal to | delta f _db When the energy storage SOC is more than or equal to 0.9 or less than or equal to 0.1, the energy storage system does not modulate frequency; conversely, the frequency deviation Δ f > | Δ f _db And the energy storage SOC is more than S _min (sudden increase in load) or SOC < S _max (sudden load decrease), the stored energy calculates the battery output according to equation (8). Taking the sudden load increase as an example, the control step of the frequency modulation of the energy storage system comprises the following steps:

5.1 At the initial stage of frequency disturbance, the rate of change of frequency d Δ f/dt is < 0. When the system frequency deviation Δ f > | Δ f _db And the energy storage SOC is more than S _min And then, the frequency modulation is carried out in an inertia response stage with virtual inertia control as a main part and droop control as an auxiliary part. The virtual inertia control weight coefficient a is decreased from 1, the droop control weight coefficient b is increased from 0, and the adaptive function is as shown in equation (14). Calculating droop control coefficient K of formula (12) according to energy storage charge state _D And calculating a virtual inertia coefficient M by the equation (13) _E And then calculating the energy storage output by combining the formula (8).

5.2 When | Δ f | = | Δ f _set L, rate of change of frequency d Δ f/dt> 0, transition control strategy: droop control is used as a main control, and virtual inertia control is used as an auxiliary control. And (3) obtaining the weight coefficients of the two control modes according to the formula (15), wherein the droop control weight coefficient b is more than or equal to 0.5, the virtual inertia control weight coefficient a is less than or equal to 0.5, and calculating the energy storage output according to the formula (8).

5.3 When the system frequency deviation is smaller than a certain steady-state frequency deviation value, the energy storage system participates in the primary frequency modulation process and ends.

The calculation conditions of the specific examples are illustrated below:

the simulation model shown in the figure 1 is built in an MATLAB/Simulink simulation environment, wherein the rated capacity of the thermal power generating unit is set to be 1000MW, the rated power of a power grid is set to be 50Hz, the rated parameter of an energy storage system is set to be 10MW/1MW & h, and a dead zone delta f of primary frequency modulation of the thermal power generating unit is set _db The frequency modulation dead zone is +/-0.033 Hz, and the frequency modulation dead zone of the energy storage system is 60% (0.0004 pu) of the frequency modulation of the thermal power generating unit. And performing per unit on other simulation parameters according to the rated capacity of the thermal power generating unit and the rated frequency of the power grid as a reference, wherein the simulation parameters are shown in table 1.

TABLE 1 simulation System parameters

When the system generates 0.02pu step load disturbance at the moment of 2S, the simulation duration is set to be 120S, and the initial energy storage S is carried out _in Set to a larger value of 0.85. The frequency deviation change curve, the energy storage charge state change curve and the energy storage system output curve under the step load disturbance are shown in FIG. 11, and the maximum value delta f of the frequency deviation under the step load disturbance is _m Deviation from steady state frequency Δ f _s And (5) evaluating, wherein the strategy evaluation indexes under the step load disturbance are shown in table 2. Maximum frequency deviation Δ f of the invention _m 57.30% less than the non-stored energy and steady state frequency deviation delta f _s The reduction is 55.70% of the non-stored energy. Maximum value of frequency deviation Δ f _m Compared with the control without the optimized frequency modulation, the control method improves the control by 16.67 percent, and the steady-state frequency deviation delta f _s The reduction is 14.29%. Energy storage output delta P _E Compared with the control strategy without the optimized frequency modulation, the method has the advantage that the improvement is 24.75%.

In conclusion, the strategy of the invention is compared with a non-optimization control strategy and a non-energy storage method, and the effectiveness of frequency response speed and frequency modulation output control is known, so that the superiority of the strategy frequency modulation of the invention is verified.

TABLE 2 strategic evaluation index (per unit value) under step load disturbance

For continuous load disturbance, the root mean square value delta f of frequency deviation _rms In order to make the evaluation index more reflect the energy storage frequency modulation effect, the expression is as follows:

in the formula: Δ f _i Is the system frequency deviation at time i; n is the total number of samples. Δ f _rms The smaller the value of (A) is, the better the energy storage frequency modulation effect is.

Adding load disturbance signals of (-0.025pu, 0.025pu) into the simulation model, and adding S _in =0.15, the simulation duration is set to 6min. The frequency deviation change curve, the energy storage state of charge change curve and the energy storage system output curve of the three control modes under continuous load disturbance are compared and shown in figure 12, and the strategy evaluation index under continuous load disturbance is shown in table 3. When S is _in Δ f for the inventive strategy when =0.15 _rms Frequency modulation delta f without energy storage _rms The reduction is 71.16%, which is 13.71% lower than that of the control without the optimized frequency modulation.

In conclusion, when the system has continuous load disturbance, the strategy of the invention can effectively reduce the frequency fluctuation of the power grid, and can effectively avoid the overcharge and overdischarge of the system when the charge state of the energy storage is too large or too small, thereby having better frequency modulation effect.

TABLE 3 policy evaluation index (per unit value) under continuous load disturbance

/>

Claims

1. The energy storage primary frequency modulation self-adaptive comprehensive control method based on the weight coefficient and the optimized frequency modulation control is characterized by comprising the following steps of:

2) Dividing the charge state of the energy storage system into a plurality of intervals according to output standards under different capacity states;

3) Constructing an energy storage output model based on virtual inertia control and an energy storage output model based on droop control;

5) Based on the charge state interval of the energy storage system, constructing an optimal frequency modulation control charge-discharge coefficient model when the energy storage system participates in frequency modulation through variable coefficient droop control;

6) Integrating the charge-discharge coefficient of the variable coefficient droop control participating in frequency modulation with the optimized frequency modulation control charge-discharge coefficient, and constructing a self-adaptive comprehensive control model of the energy storage system participating in frequency modulation through the variable coefficient droop control;

7) Based on the variable coefficient droop control charge-discharge coefficient, calculating to obtain a virtual inertia control charge-discharge coefficient;

9) Aiming at different frequency change stages of the energy storage system participating in primary frequency modulation, different distribution coefficient models are constructed, and the distribution coefficient models are resolved to obtain a weight coefficient a of virtual inertia control and a weight coefficient b of variable coefficient droop control;

2. The energy storage primary frequency modulation adaptive integrated control method based on the weight coefficient and the optimized frequency modulation control is characterized in that in the step 1), the charge state of the energy storage system comprises a charge state and a discharge state;

when the power system generates positive frequency deviation, the energy storage system is in a charging state; when the power system generates negative frequency deviation, the energy storage system is in a discharging state.

3. The energy storage primary frequency modulation adaptive integrated control method based on the weight coefficient and the optimized frequency modulation control as claimed in claim 1, wherein the energy storage system state of charge is divided into regions according to the output standard under different capacity states, an adaptive rule-based S-type piecewise function with the energy storage system state of charge as an independent variable and the charge-discharge coefficient as a dependent variable is adopted, and the divided regions of the energy storage system state of charge comprise (0, S) _min )、[S _min ，S _low )、[S _low ，S _high )、[S _high ，S _max )、[S _max 1), wherein S _min Minimum value for state of charge partition; s _low A lower value for state of charge partition; s _high A higher value for state of charge partition; s _max The maximum value of the state of charge partition.

4. The energy storage primary frequency modulation adaptive integrated control method based on the weight coefficient and the optimized frequency modulation control as claimed in claim 1, wherein the energy storage output model corresponding to the virtual inertia control is as follows:

wherein a is a weight coefficient of the virtual inertia;

is the rate of change of frequency deviation; m is a group of _E Is the virtual inertia coefficient; wherein the virtual inertia coefficient M _E Related to the state of charge of the energy storage system;

in the formula,. DELTA.P _E2 Controlling the energy storage output participating in frequency modulation for the variable coefficient droop; b is a weight coefficient for variable coefficient droop control, where a + b =1; Δ f is the frequency deviation; Δ f _db For the frequency deviation dead zone, when the frequency deviation Deltaf is smaller than the frequency deviation dead zone Deltaf _db When the system is not frequency-modulated; k _E Controlling the frequency modulation coefficient for droop;

5. The energy storage primary frequency modulation adaptive comprehensive control method based on the weight coefficient and the optimized frequency modulation control as claimed in claim 1, wherein a charge-discharge coefficient model of the energy storage system participating in frequency modulation through variable coefficient droop control is as follows:

in the formula: k _max Is the maximum value of the virtual droop control coefficient; k _C1 And K _D1 The charge and discharge coefficients of the variable coefficient droop control strategy are respectively; s represents the real-time monitored state of charge value of the energy storage system; and n is the adaptive coefficient of the curve.

6. The energy storage primary frequency modulation adaptive comprehensive control method based on the weight coefficient and the optimized frequency modulation control as claimed in claim 1, wherein the model of the optimized frequency modulation control charge-discharge coefficient is as follows:

in the formula, omega is a curve adjustment coefficient of the virtual inertia coefficient and the droop control coefficient; k _C2 And K _D2 Respectively optimizing the charging and discharging coefficients of the frequency control method; s ₀ Represents a low intermediate value and has a value range of 0.1<S ₀ <0.45；S ₁ Represents a high intermediate value, and has a value range of 0.55<S ₀ <0.9。

7. The adaptive integrated control method for energy storage primary frequency modulation based on weight coefficient and optimized frequency modulation control according to claim 1, wherein the adaptive integrated control model is as follows:

K _c ＝K _cl +K _c2 Δf>0 (8)

K _D ＝K _D1 +K _D2 Δf＜0 (9)

8. the adaptive integrated control method for energy storage primary frequency modulation based on weight coefficient and optimized frequency modulation control according to claim 1, wherein the virtual inertia coefficient M of the virtual inertia control method _E The calculation formula of (a) is as follows:

M _E ＝λK _E (12)

9. The energy storage primary frequency modulation adaptive integrated control method based on the weight coefficient and the optimized frequency modulation control as claimed in claim 1, wherein the energy storage output model of the virtual inertia control and droop control cooperative output is as follows:

wherein a is a weight coefficient of the virtual inertia; b is a weight coefficient of variable coefficient droop control; Δ f is the frequency deviation; Δ f _db In order to have a frequency deviation dead zone,

is the rate of change of frequency deviation; m _E Is the virtual inertia coefficient; k _E The frequency modulation factor is controlled for droop.

10. The energy storage primary frequency modulation adaptive integrated control method based on the weight coefficient and the optimized frequency modulation control as claimed in claim 1, wherein the distribution coefficient analytical model is as follows:

when the frequency deviation delta f is not less than the frequency deviation dead zone delta f _db At a frequency equal toRate of change

in the formula: the adaptive coefficient of the distribution coefficient analysis model with the vertical control below m as the main control and the virtual inertia control as the auxiliary control; a is a weight coefficient of the virtual inertia; b is a weight coefficient of variable coefficient droop control;