WO2014201849A1

WO2014201849A1 - Method for actively optimizing, adjusting and controlling distributed wind power plant provided with energy-storage power station

Info

Publication number: WO2014201849A1
Application number: PCT/CN2014/000576
Authority: WO
Inventors: 邢作霞; 王文杰; 杨俊友; 刘劲松; 李玉婷; 姜立兵; 崔嘉; 王海鑫
Original assignee: 国网辽宁省电力有限公司电力科学研究院; 沈阳工业大学
Priority date: 2013-06-18
Filing date: 2014-06-13
Publication date: 2014-12-24
Also published as: CN103441537A; CN103441537B

Abstract

A method for actively optimizing, adjusting and controlling a distributed wind power plant provided with an energy-storage power station, which relates to the constraint of the wind power grid access guide rule on the upper and lower limits of the active power output, setting a constraint condition by the maximum and minimum limit values of single power generation, and establishing the constraint condition of the active power output according to the single power generation characteristics of the wind power and energy-storage power supply. In accordance with the external conditions of the operation of the wind power plant, the operation is divided into a power rationed operation and non-power rationed operation, wherein the power rationed operation has the optimization objectives that the variable pitch control load and line loss are minimal and the frequency fluctuation is minimal; and the non-power rationed operation has the optimization objectives that the maximum wind energy capture rate and line loss are minimal and the frequency fluctuation is minimal, and uses a multi-objective layered optimization problem as an objective function. A particle swarm algorithm is used to obtain a solution, so that the wind power active power consumption which is output by the action of an energy-storage system minimizes the stable output, thereby achieving the effective joining of the energy-storage system with the existing scheduling operation mode, and achieving the best economic benefit at the same time.

Description

Active power optimization control method for distributed wind farms equipped with energy storage power stations

Technical field

The invention relates to a method for optimizing and controlling the active power of a distributed wind farm equipped with a energy storage power station, and belongs to the technical field of active power control of wind farm connected to the grid.

Background technique

As the country attaches great importance to renewable energy generation, China's wind power has become an important emerging industry that optimizes energy structure and promotes sustainable development. However, with the integration of large-scale wind power, the safety, reliability, power quality and grid scheduling of power systems are affected by the volatility and randomness of wind power generation. Centralized large power grids are strongly coupled and cannot track changes flexibly and reliably. The decentralized access wind power project refers to a wind power project that is located near the center of the power load and is not used for large-scale long-distance transmission of power, and the generated power is connected to the power grid and consumed locally. It can better solve some problems caused by the integration of centralized wind power generation. In this context, the state has proposed a policy to develop decentralized wind power.

Direct access to a low-voltage distribution network with a distributed wind farm equipped with a storage power station can reduce the loss and investment costs of the transmission network, improve the reliability of the distribution network, and ensure energy security to a large extent. However, with the increasing demand for power quality, energy saving and power grid safety and stability of wind farms, research on active power optimization and operation of decentralized wind farms equipped with energy storage power stations is to reduce grid losses and ensure grid stability. The most effective means of operation. By rationally configuring the active power supply (fan and energy storage power station) and reasonable control of active power loss, the frequency fluctuation of the power grid can be reduced, and the power loss of the power grid can be reduced, so that the power system can operate safely and stably.

A number of optimization methods have been proposed for the active control and optimization of wind farms. It is proposed that a single wind turbine can perform active power coordinated control strategy between wind turbines in high wind speed, medium wind speed and low wind speed stage to improve wind power generation margin. A rotor kinetic energy is proposed as an energy storage device. Wind farm control methods to reduce short-term power fluctuations and improve wind farm grid stability; comprehensively consider the units most closely related to output power fluctuations, and control their output to suppress fluctuations in output. These methods can achieve active optimization of the wind field.

However, there are few studies at home and abroad for the problem of decentralized active optimization with energy storage power stations. Moreover, the above optimization method does not take into account the multidimensional function relationship between the active power generated by the wind turbine and its own loss, and therefore cannot make the wind turbine complete as much as possible within the safe range of the frequency fluctuation of the grid connection point. Body loss is minimized.

Summary of the invention

Purpose of the invention

The invention provides a method for optimizing and controlling the active power of a distributed wind farm equipped with an energy storage power station, and the object thereof is to solve the problem that the overall loss of the wind power generator is too large in the safety range of the grid-connected frequency fluctuation in the prior art, and the wind is directed to the wind. The multi-objective function is established under different operating conditions of the field. While effectively controlling the frequency fluctuation, it can also reduce the fan pitch load and reduce the active loss of the wind farm.

Technical solutions

A method for controlling active power of a distributed wind farm equipped with an energy storage power station, characterized in that: the steps are as follows:

The first step is to measure the wind speed of the wind field, the fan speed, the active power of each unit, the active power of the substation side, the three-phase voltage, current, frequency, power factor and energy storage system of the energy storage system through the SCADA detection control and acquisition system. Power and grid point power quality data, and then send the data to the control center via a communication cable;

In the second step, the wind power fluctuation and the unit output constraint are set as follows:

The formula is the upper and lower limit constraint of the unit output; (2) is to consider the wind power power fluctuation range constraint of the energy storage system. In the time window of lmin or 30min, the variation range of the combined output power of the system must not be greater than the total rated output power of the wind farm. _{Rate /} ^ or ^, with ₃ . According to the grid guidance, the variation amplitude of the system's combined output power is the difference between the maximum value minus the minimum value;

In the third step, through the real-time measurement of the fan outlet voltage ^, the overhead line current /, the box-to-no load loss rate and the loss rate under the rated load 4, A, the total line loss A in the wind field is obtained, which is obtained as follows: .

P _L - 3 xl ² r L -\- c _{ x U -fc ₂ — -f <^ ₃ Where is the coefficient, the expression is as follows: c, = a 2 N

X—— xP^xr L

¹ N, ^ecc (4)

N x S

100 (5)

N x a x P„ where: " is the power output coefficient; N is the number of fans; For each fan rated capacity; I is the overhead line current value; r is the overhead line unit length resistance value; J is the length of each line of the overhead line; N, is the fan outlet cable number; is the box variable capacity; The resistance value per unit length; z is the length of the cable line four steps, calculate the other two optimization goals: the pitch aerodynamic drag coefficient, the frequency fluctuation of the grid point 4,

(7)

For the active output that needs to be adjusted; for the active output of the wind turbine unloading operation; f _b , for the set upper and lower limits of the frequency of the wind turbine that does not participate in the frequency modulation; f _a , ^ is the set frequency of the wind turbine Upper and lower limits; active output corresponding to points;

dP _ dP άβ da

(8) dC _d άβ da dC _d where: is the pitch angle, "is the inflow angle, both are measurable data;

The fifth step, according to the external conditions of the wind farm operation, is divided into power-limiting operation and non-electrical operation. In both cases, the active output of the wind farm is adjusted, and the optimization target obtained above is brought into the objective function; Power limitation: The optimization target is the frequency fluctuation Δ/sum of the grid connection point. When P _B W + (k) ≥ P _ref , P _r is the active output value of the given wind field, and the pitch control Δ^ is required. At this time, the optimization target needs to increase the pitch aerodynamic drag coefficient. The objective function is as follows :

Min _v = (9) h ₂ P ^A f (P _B (k) + _j p _l (k) ≤ p _re/ ) Unlimited conditions: When the wind speed is less than the rated wind speed, the maximum power tracking control is applied to the fan.

ΜΡΡΤ, optimize the target Δ / sum _; when the wind speed υ is less than or equal to the rated wind speed, the pitch control of the fan, the optimization target is C Δ / and ^

l ₂ P _L + / ₃ Δ/ {ν<υ _Γ , ΜΡΡΤ)

Min f = , (10)

|/ _ι ς _/ + +/ ₃ Δ/ (υ≥υ _Γ ,Αβ) where: ^/^ and ^^ are the weight coefficients for power-limited and unrestricted power, respectively. The constraints are:

(11)

P _B +∑P _t =P _D + P _L (12)

(11) Frequency fluctuation range constraint, 4 is the frequency offset limit; (12) Grid power supply balance constraint,

/ demand for total load;

In the sixth step, multi-objective active optimization is performed by using particle swarm optimization based vehicle path optimization algorithm.

For the above multi-objective optimization control problem, the particle swarm optimization algorithm is used for iterative search to find the optimal solution. The steps are as follows:

1: Input fan outlet voltage, power factor, main transformer parameters and actual input power, ratio of actual voltage to rated voltage, box to no-load loss rate and loss rate under rated load, pitch angle, inflow angle and other parameters. Calculate the total line loss, frequency fluctuation and drag coefficient in the wind field;

2 : Set the dimension, the maximum number of iterations, and the number of particles;

3 : Bring the result obtained in stepl into the formula (9) (10) to obtain the fitness value. Let /^ be equal to the position of the current particle p, ', ,

4: Initialize the position and velocity, calculate the position of the first individual optimal particle and set this ^(ί ) to the position of the global optimal particle currently found by the population.

5: If the current particle fitness value is less than the individual extreme value, the current individual extreme value is updated; ^ _{ϊ ;}

6 : If the current particle fitness value is less than the global extremum, update the current global extremum ^

7 : Update the velocity vector and position vector by equation (25) (26);

(') -X ^l) ) ₊ c ₂ r ₂ — ) !■ = 1,2,.·.," (25)

X ^c) ≤ Χ, Γ i = 2,...,n (26) where t is the current number of cycles; 4 is the particle weight coefficient; ω is the inertia weight; ν _ί is ( ₀

1) uniformly distributed random numbers; V _id , the position and velocity of the i-dimensional particles;

StepS: Calculate the fitness value with the updated velocity vector and position vector;

Step9: Repeat step5 to lj step7;

SteplO: Determine the number of iterations, and if it is satisfied, output the result; otherwise, return to Step7.

Advantages and effects

The invention proposes a multi-objective active power optimization method under the operating condition for the external operation characteristics of the distributed wind farm equipped with the energy storage power station, that is, the active power optimization control method of the distributed wind farm equipped with the energy storage power station: judging the wind farm Whether to operate under power-limited conditions, multi-objective optimization control based on wind speed conditions.

(1) The required data is measured by the SCADA system and then sent to the control center via a communication cable.

(2) Set wind power fluctuations and unit output constraints.

(3) Calculate the rated load loss of the box P„, the box becomes no-load loss., the cable line loss connected to the low-voltage side of the fan outlet, and the power loss of each line. The total line loss in the wind field is obtained after finishing ^ -

(4) Calculate the other two optimization objectives: pitch aerodynamic drag coefficient, frequency fluctuation of the grid point

Λ

(5) For the external conditions of wind farm operation, it is divided into power-limited operation and non-electrical operation. In both cases, the active output of the wind farm is adjusted. Multi-target control based on wind speed conditions, will be above The obtained optimization goal is brought into the objective function.

(6) Multi-objective active optimization using particle swarm optimization based vehicle path optimization algorithm.

The specific advantages and positive effects of the utility model are as follows: 1. According to the wind field operating conditions and the wind speed situation, a multi-objective function is established hierarchically, and the active power is controlled and controlled by adjusting variables such as line loss, frequency, and load. 2. This control method is based on the wind power penetration rate within the range set by the distribution network, and the stability of the system is improved on the premise of ensuring that the power fluctuation is within the safe operation range. 3. It has strong practicability and can be used for multi-target control of the entire distributed wind farm to adjust the active power to minimize the active loss of the entire wind field.

DRAWINGS

Fig.1 Typical topological structure diagram of energy storage and distributed wind power distributed into the power grid;

Figure 2 Flow chart of the wind farm active layering optimization control strategy;

Fig. 3 is a graph showing the variation of the pitch effect with the wind speed;

Figure 4 shows the resistance characteristic curve of the airfoil;

FIG. 5 is a schematic diagram of the principle of maximum power tracking optimization control in FIG. 2;

Figure 6 is a flow chart based on the particle swarm algorithm.

detailed description:

The present invention will be further described below in conjunction with the accompanying drawings.

As shown in Figure 1, the control object proposed by the present invention is: a wind turbine and an energy storage power station are connected to a 10 kV local substation loop, and then access to a public access point (PCC/1) through 10 kV/35 kV; Inside, there are m wind storage power stations of the same mode, which are delivered to different end-user loads after 35kV convergence; 35kV can be boosted and connected to the public large power grid.

The basic idea of the invention lies in: Optimizing the frequency fluctuation, the pitch load and the line loss of the power grid by optimizing the regulation, reducing the active loss of the power grid, improving the voltage quality, and using the electrical equipment to operate safely and reliably.

The active optimization problem considered in the present invention can be defined as follows: by various adjustment means, the objective function is optimized under the condition that the frequency constraint and the operation constraint are satisfied. It can be seen from the above that the active optimization problem is actually a typical constrained combinatorial optimization problem.

A method for active power optimization control of a distributed wind farm equipped with a storage power station, as shown in Figure 2, the steps are as follows:

The first step is to measure the wind speed of the wind field, the fan speed, and the machine through the SCADA detection control and acquisition system. Group active power, substation side active power, unit and network side three-phase voltage, frequency, power factor, energy storage system charge and discharge power, grid point power quality data, and then send these data to the control center through the communication cable.

In the second step, the set wind power fluctuation and the unit output constraint are:

1. The upper and lower limits of the unit output:

p, <p<p (1) where ^ is the minimum and maximum values of the unit's output.

2. Consider the power fluctuation range constraints of the energy storage system:

When the energy storage system responds to the charge and discharge power command value, it can be obtained:

P ₀ (k + \) = P _w (k) + P _B (k) (2)

P„W is the combined output power at the current time; i W is the wind power at the current time k; P k is the charge and discharge power of the energy storage system.

The energy stored by the energy storage system at the current time is:

E _B {k) = E _B {Q)-^∑P _B U) (3)

E _B (0) is the initial energy of the energy storage system.

Take P separately. (And is the state variable x A:) and 3⁄4 (&), P _w {k) is regarded as the external disturbance variable r ( , for the control input amount WW, then the state space model of the smoothing fluctuation control system is as follows:

(4) x ₂ (k + l) = x ₂ (k) - u(k)

(5)

Where: r(t) is the process output matrix. Max Y(k - ί) - min Y(k-i)≤ γρ

(6) max Y(ki)~ min Y(ki) _≤r ₃₀ p _rateii

, a total of M moments, where ;S: = 0, l, ..., A - l.

In any lmin time window, the variation amplitude of the system's combined output power (the difference between the maximum value minus the minimum value) must not be greater than the total rated output power of the wind farm / ^^; in any 30min time window, The variation of the system's combined output power must not be greater than the total rated output power of the wind farm, p _ra , _t£/ . , γ and ^ are specified by the grid guidelines.

In the third step, the collecting circuit of the wind farm is divided into a cable collecting line and an overhead line collecting line. Compared with a thermal power plant, the collecting line of the wind farm is more complicated and close to a small distribution network. When the power is transmitted between the wind turbine and the booster station through this small distribution network, there is a line loss that cannot be ignored, and the thermal power plant can ignore these losses during operation. Active power loss calculations for both cable and overhead collector circuits.

Calculate the rated load loss of the box Ρ„, the box becomes no-load loss Ρ., the cable line loss connected to the low-voltage side of the fan outlet ^, the power loss of each line ^. The total line loss in the wind field is obtained through finishing.

/L, for each line power loss:

P _loss = 3x 1 xi? = 3 x 7 xri ( 7) where: / is the line current; r is the resistance per unit length; £ is the length of each line,

Cable line loss for the fan outlet and the low-voltage side of the box:

Where: _α is the power output coefficient; is the rated capacity of each fan; N is the total number of fans; Μ is the number of cable outlets of the fan; ^ is the fan outlet voltage; r is the resistance value of the cable unit length; L _c is the cable length.

A is the box to change the no-load loss. -

Where: S is the box variable capacity; Α is the loss rate under no load <

Variable load loss for the box:

Where: 3⁄4 is the loss rate at rated load ( The total line loss in the wind farm is:

(11) After finishing, it is:

1 1

P, =3x1 xr xL + c, xi/, +c, he, ■ (12)

¹ Α β ₂ is a coefficient, and the expression is as follows:

Cj = α

N x S

100 (14)

N xa ² x „

c, =

100 x S, (15) Fourth, calculate the other two optimization goals: pitch aerodynamic drag coefficient, frequency fluctuation of the grid point 4 .

1. When the wind turbine is running above the rated wind speed, it is necessary to adjust the pitch angle to reduce the energy capture of the wind turbine, thereby adjusting the active power generation capability. During the pitching process, the wind turbine blade will withstand the pitch load caused by the aerodynamic force, and the momentum-leaf theory analysis will calculate the aerodynamic load of the wind turbine to establish a non-linear function relationship. The pitch load caused by aerodynamics can be expressed as:

Μ ₇ = ζάΜ _ζ = O.Spv cJ ² + C _d ² · 5C sin + θ) dr (16) where: C is the lift coefficient; C _rf is the drag coefficient; is the pitch angle; is the resultant force and tangential force clamp angle. It can be seen that the aerodynamically induced pitch load is related to the pitch angle, so changing the pitch angle can change the wind turbine aerodynamic load. However, at the same wind speed, the smaller the pitch angle, the greater the wind energy captured by the wind turbine, and the greater the load on the wind turbine. Therefore, the wind turbine load becomes a short board that restricts the safe and stable operation of the unit. In order to prevent excessive pitch load from damaging the life of the unit, the unit load must be evaluated when studying the pitch to minimize the load risk. The drag coefficient can largely represent the active loss caused by the pitch load.

C _d = (17) pxv _x xc dr Where p is the air density; _Vl is the airflow velocity; c is the blade chord length at the radius; is the resistance acting on the blade; is the leaf thickness. For the pitch angle, the relationship between ^ and wind speed is shown in Fig. 3; "For the inflow angle, the relationship is shown in Figure 4 dp da. It can be seen that the sinusoidal characteristic is exhibited within a certain range, so it is drawn up. For: dC

Sin ( - 45°) (18) da

So you can get it by (18) <

dC,

The relationship between the drag coefficients.

dP _ dP άβ da

(19) dC _d dfi da dC _d

2. The system frequency exceeds the set limit Λ</<Λ or _/;</</ JJ inch, the wind turbine adjusts the output, responds to the system frequency change, and has:

Where Δ _Ρ is the active output that needs to be adjusted; it is the active output under the unloading operation state of the wind turbine; f _h , is the frequency upper and lower limit of the set wind turbine that does not participate in the frequency modulation; f _a , is the set wind turbine Adjust the upper and lower limits of the frequency band; the active output corresponding to the point. Each quantity is a standard value, the power reference value is the active output determined by the wind speed, and the frequency reference value is the rated frequency 50 Hz.

In the fifth step, the external conditions for the operation of the wind farm are divided into a limited power operation and an unrestricted electric operation. Both of these are finally adjusted for the active output of the wind farm. Bring the optimization goal obtained above into the target letter

1. Power-limited situation: The optimization target is the frequency fluctuation of the grid-connected points 4~ and ^. When p) + ^( ≥p _rei s inch, pitch control Δ^ is required. To dispatch the active output value of the given wind field. At this time, the optimization target needs to increase the pitch. Aerodynamic drag coefficient G. The objective function is as follows: minf _x = (21)

i ( )< _re/ )

2. Unlimited conditions: When the wind speed u is less than the rated wind speed, the maximum power tracking control optimization target Δ/ sum is applied to the fan _; when the wind speed u is less than or equal to the rated wind speed, the pitch control of the fan is performed, and the optimization target is C _RF . Δ/ and.

l ₂ P _L + l ₃ Af {v<v _r , MPPT)

Min where: /, ΛΛ and / ₁₅ / ₂ , / ₃ are the weighting factors when the power is limited and the power is not limited.

The maximum power tracking control ΜΡ/Γ uses the optimal tip speed ratio method. When the wind speed changes, keep the tip speed ratio of the wind machine at the optimum value ^, ^ ^ is generally obtained by calculation or experiment, so the wind turbine has the highest utilization of wind energy at any wind speed. Figure 5 is a block diagram of its control principle. It takes the measured values of wind speed and wind turbine speed as the input signal of the control system, and calculates the actual tip speed ratio at this time, and then compares it with the optimal tip speed ratio. The error value is sent to the controller, which controls the output of the inverter to adjust the fan speed to ensure the tip speed ratio is optimal.

According to the above objective function (21) (22), the constraint is:

Frequency fluctuation range constraints:

(twenty three)

Grid power supply balance constraints:

(24) where Λ is the demand for total load ( Sixth, 3⁄4 uses multi-objective active optimization based on particle swarm optimization.

For the above-mentioned multi-standard optimization control 13⁄4 problem, use the particle fresh optimization algorithm to perform the generation search, and find the lowest solution, such as

Stepl: Input fan outlet power, power country, main variable ffi parameter and actual input power, ratio of actual electric I to rated voltage, box change no-load loss rate and loss rate under rated load, vegetable pitch angle, inflow angle "Parameters" Calculate the total line loss, frequency fluctuation and & force coefficient in the wind field

Ste _P 2: Set the dimension, the maximum number of iterations and the number of particles; ste _P 3: Bring the result obtained in stepl into the formula (21) (22), and obtain the fitness value /« _te , let / equal the current particle position

Step4: Initialize the position and the catch, calculate the position of the first individual optimal particle «, and set this ^p as the position pj of the global optimal particle found before the population;

Step5: If the value of the pre-particle fitness is less than the individual extremum, update the current individual child. ^ _;

Step6: If the pre-particle fitness value is less than the global extremum, update the global extreme value before 3⁄4

Step7: Update the velocity vector and position vector from equation (25) (26).

Where _t is the current number of cycles; 3⁄4, is the weight coefficient of the particle; is the inertia weight; , 3⁄4 is the (0, I) inner z-distribution random number; V, I is the i-dimensional particle brain position and velocity Step8: Calculate the fitness value with the updated velocity vector and position vector; Step9: Steps to step7;

StepiO: judge the number of iterations, and if it is satisfied, output the result; otherwise, return to Step

Claims

Claim

1. A control method for active power of a distributed wind farm equipped with a storage power station, characterized in that: the steps are as follows:

<P≤R i,.ax (1) max Y(ki)- win Y(ki)≤y _x p _rated

=0,1,...59

(1) is the upper and lower limits of the unit's output; (2) is to consider the wind power range fluctuation range of the energy storage system. In the time window of lmin or 30min, the variation of the system's combined output power must not exceed the total wind farm. Determine the output power of ^ or. , and ^ is regulated by the grid, the variation of the output power of the system is the difference between the maximum value minus the minimum value;

In the third step, the total output loss of the wind farm is obtained by real-time measurement of the fan outlet voltage ^, the overhead line current /, the box-to-no load loss rate and the loss rate under the rated load of 4, 3⁄4.

Where _Cp c ₂ , c ₃ are coefficients, and the expression is as follows -

2 N „ τ

c, - a x - xP x xL^

1 N, ^e (4)

NS _t

^C 2 =

100 (5)

N x ² x ₍

100 x 5, (6) where: "for the power output coefficient; N for the number of fans; ^ for each fan rated capacity; / for the overhead line current value; r for the overhead line unit length resistance value; for the overhead line each line Length The number of cable outlets of the machine; S is the capacity of the box; ^ is the resistance value of the length of the cable; A is the length of the cable;

The fourth step is to calculate the other two optimization goals: pitch aerodynamic drag coefficient c, frequency fluctuation of the grid point

For the active output that needs to be adjusted; for the active output of the wind turbine unloading operation; f _b , ./: is the frequency upper and lower limit of the set wind turbine that does not participate in the frequency modulation; f _a , / _rf is set The upper and lower limits of the frequency band of the wind turbine; the active output corresponding to the ^ point;

dP _ dP άβ da

( 8 ) dC _d άβ da dC _d where: is the pitch angle, "is the inflow angle, both are measurable data;

The fifth step, according to the external conditions of the wind farm operation, is divided into power-limited operation and non-electrical operation. In both cases, the active output of the wind farm is adjusted at the end, and the optimization target obtained above is brought into the target letter; Power limitation: The optimization target is the frequency fluctuation Δ/sum of the grid connection point. When Ρ _β (A) + (/c) ≥ P _Kf , P _nf is the active output value of the given wind field, and the pitch control Δ is required. At this time, the optimization target needs to increase the pitch aerodynamic drag coefficient C, and the objective function is as follows:

Unlimited conditions: When the wind speed is less than the rated wind speed, the maximum power tracking control is applied to the fan.

MPPT, optimize the target Δ / and P _; when the wind speed u is less than or equal to the rated wind speed u,., the pitch control of the fan, the optimization target is C _rf , Δ / and

;^^^ and ^ are the weighting factors for power limiting and no power, respectively. The constraint is -

Δ, ≤ Cll) P _B +∑P,=P _D + P _L (12)

(11) Frequency fluctuation IS is about east, 4 is the frequency offset 隠 value; (12) Grid power supply balance constraint, which is the demand of the load;

Sixth, the W-based particle path-based vehicle path optimization algorithm for multi-objective active optimization.

2. The distributed phoenix electric field active power tt control method equipped with an energy storage power station according to claim 1, wherein:

In view of the above multi-g standard optimization control problem, the particle swarm optimization ί method is used for iterative search to find the optimal solution. The steps are as follows:

0): Input fan power, power garden, main transformer parameters and actual input power, actual voltage and rated ft pressure ratio, box change no-load loss rate 靡 rated load loss rate, pitch angle, Calculate the total line loss, frequency fluctuation and drag coefficient in the wind field by parameters such as the inflow angle;

3 : Bring the result obtained in stepl into the formula (9) (10) to obtain the fitness value.

Let / equal the position of the front particle pj,:

4: Initialize the position and velocity, calculate the position _{P of the} first individual optimal particle, and set this ^W as the position of the global optimal particle currently found by the population / ^w ;

5: If the current particle fitness value is less than the individual pole tfi, update the individual plant value _ΛΑβ before _3⁄4;

6 If the current particle fitness value is less than the global extremum, update the current global extremum _3⁄4;

7 : Update the velocity vector and position vector by equation (25) (26);

ν,Γ ^] = l,X...,n (25) xj ^i+t) (26)

Where t is the current number of cycles; < , 3⁄4 is the particle weight coefficient; "is the inertia weight; 3⁄4, is a uniformly distributed random number in (0, 1); V _td , , is the position and velocity of the i-dimensional particle;

StepS: Calculate the fitness value using the updated velocity vector and position vector;

Step9: electricity 3⁄4 stepS to step7;

SteplO: Determine the number of iterations, and if it is satisfied, output the result; otherwise, return to Step7. '