CN105932860B - A kind of power converter double-loop control strategy based on particle swarm algorithm - Google Patents
A kind of power converter double-loop control strategy based on particle swarm algorithm Download PDFInfo
- Publication number
- CN105932860B CN105932860B CN201610408203.3A CN201610408203A CN105932860B CN 105932860 B CN105932860 B CN 105932860B CN 201610408203 A CN201610408203 A CN 201610408203A CN 105932860 B CN105932860 B CN 105932860B
- Authority
- CN
- China
- Prior art keywords
- particle
- value
- control
- global optimum
- power converter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000002245 particle Substances 0.000 title claims abstract description 212
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 52
- 238000011217 control strategy Methods 0.000 title claims abstract description 29
- 238000012544 monitoring process Methods 0.000 claims abstract description 11
- 238000000034 method Methods 0.000 claims description 37
- 238000005457 optimization Methods 0.000 claims description 19
- 238000012804 iterative process Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 11
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 230000000694 effects Effects 0.000 claims description 7
- 230000003044 adaptive effect Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000009466 transformation Effects 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 3
- 230000005611 electricity Effects 0.000 claims description 2
- 238000005070 sampling Methods 0.000 claims description 2
- 238000000518 rheometry Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 5
- 230000001276 controlling effect Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 5
- 230000006698 induction Effects 0.000 description 5
- 238000004088 simulation Methods 0.000 description 4
- 230000007812 deficiency Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000033228 biological regulation Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 230000001105 regulatory effect Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000012937 correction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010616 electrical installation Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000004146 energy storage Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02M—APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF
- H02M1/00—Details of apparatus for conversion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02M—APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF
- H02M1/00—Details of apparatus for conversion
- H02M1/0003—Details of control, feedback or regulation circuits
- H02M1/0025—Arrangements for modifying reference values, feedback values or error values in the control loop of a converter
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The present invention relates to a kind of power converter double-loop control strategy based on particle swarm algorithm, belongs to operation of power networks and control technology field.Its feature comprises the steps of: the double-loop control strategy for 1) implementing outer voltage and current inner loop to power converter；2) the control parameter adjustable range of the controlling unit of on-line tuning is determined；3) monitoring control target value；If 4) control target exceeds allowed band, control parameter is updated using particle swarm algorithm；5) if control target keeps control parameter constant without departing from allowed band.The present invention has height reliability and adaptivity, the power control that power converter can be realized under the disturbance of power grid, has important practical significance to the operational reliability and controllability that improve power converter.
Description
Technical field
The present invention relates to a kind of power converter control strategy, in particular to a kind of power conversion based on particle swarm algorithm
Device double-loop control strategy belongs to low voltage electric network operation and control technology field.
Background technique
AC-to DC, direct current to energy between exchange, AC to AC or DC to DC may be implemented in power converter
The transformation of amount has a wide range of applications in electric system.With new energy in low voltage electric network using more and more, wind, light,
Energy storage etc. is connected to the grid, the operation and control of micro-capacitance sensor, power converter device is all be unable to do without, for rectifying, inversion, panoramic limit
Power control, reactive compensation etc..Currently, power converter device generallys use voltage source converter, using what can be turned off
High power device.How to carry out control to these current convertor devices is the key that it is applied in power grid.
The conventional control methods of power converter are that the double-closed-loop control that power control is outer ring, current control is inner ring is calculated
The active and idle decoupling control of power converter may be implemented in method, this method.Electric system will appear in the process of running compared with
More interference and uncertain factor.For example, motor start-up, will cause grid voltage sags, route occurs short trouble, can produce
Raw biggish short circuit current.Power grid disturb under the conditions of, conventional PI controller does not have good control effect,
The selection of control parameter is given by experience, does not adapt to the variation of operation of power networks state, does not have adaptivity, control effect
Demand for control is unable to satisfy under the conditions of certain operations of power networks.
For the deficiency of conventional electric power current transformer control strategy, some new control methods start for power converter
Control.Have than more typical:
1) the power converter control method of fuzzy control is used.The outer loop control of power converter is carried out mould by this method
Control target value is carried out Multi sectional division by fuzzy rule, realizes control target measured value and control by fuzzy rule by gelatinization
Nonlinear Mapping relationship between parameter processed, to improve the adaptability of control system at different operating conditions.
2) the power converter control method of neural network is used.This method is by collecting a large amount of control parameter and control
Associated data between target, Lai Xunlian neuroid, so that the incidence relation between control parameter and control effect is formed,
By the feasible control parameter of neural network prediction, to realize the control strategy of power converter.
3) using the power converter control method of Modern Nonlinear control.Such method makes full use of Modern Nonlinear control
System is theoretical, by carrying out mathematical modeling to current transformer, to set up Controlling model, obtains corresponding control rule on this basis
Rule.Such as use the variable-structure control of exact linearization method, Nonlinear Decoupling control etc..
The above-mentioned control technology for power converter, biases toward theoretical research more, is difficult to realize in engineering, above-mentioned control
Method processed is also difficult to functionization in a short time.
Summary of the invention
The purpose of the present invention is being based on smart random optimization algorithm, a kind of power converter based on particle swarm algorithm is proposed
Double-loop control strategy, the present invention have preferable reliability and adaptivity, can realize that electric power becomes under the disturbance of power grid
The power control for flowing device has important practical significance to the operational reliability and controllability that improve power converter.
It is an object of the invention to be based on existing power converter control technology, propose a kind of based on particle swarm algorithm
Power converter double-loop control strategy provides technical basis and practical method for the reliable control of power converter, improves
Application of the electrical installation based on power conversion technology in power grid.
The technical solution adopted by the present invention is that:
A kind of power converter double-loop control strategy based on particle swarm algorithm adopts voltage source type electric power current transformer
With the double-loop control strategy of outer voltage and current inner loop；Using particle swarm optimization algorithm, online adaptive adjustment control
The parameter of link realizes optimal control of the power converter under the various disturbances of power grid.It is characterized by comprising following steps:
Step 1) implements the double-loop control strategy of outer voltage and current inner loop to power converter；
Step 2) determines the control parameter adjustable range of the controlling unit of on-line tuning；
Step 3) monitoring control target value；
If step 4), which controls target, exceeds allowed band, control parameter is updated using particle swarm algorithm；
(1) setting the number of iterations initial value is zero；
(2) population calculating parameter, position, speed, aceleration pulse and inertial factor including particle are initialized；
(3) optimal value and global optimum of its initial position are determined for each particle；
(4) individual optimal value and the global optimum in its iterative process are determined for each particle；
(5) more new particle speed next time and position；
(6) optimal value and global optimum of population are updated；
(7) the number of iterations increases by 1, if reaching the number of iterations upper limit, carries out step (8), is otherwise transferred to step (4)；
(8) using particle group optimizing calculated result as power converter control parameter value.
If step 5) controls target without departing from allowed band, keep control parameter constant.
The voltage source type electric power current transformer, by device (such as insulated gate bipolar transistor with turn-off capacity
(IGBT)) power converter formed, has the function of rectification, inversion, AC conversion or DC conversion；
The power control of the double-loop control strategy of the outer voltage and current inner loop, power converter is closed using double
Ring control, outer ring respond active and reactive control target, and inner ring uses current decoupled control strategy, real by rotating coordinate transformation
The decoupling control of existing watt current and reactive current；
The parameter of the described online adaptive adjustment controlling unit, in power converter power outer loop control, than
Particle swarm optimization algorithm optimizing, online dynamic adjustment proportional plus integral control parameter, to improve control is added in example integration control ring
Adaptability of the device processed to extraneous uncertain noises factor；
The monitoring controls target value, for voltage source type electric power current transformer, can simultaneously to two control targets into
Row monitoring control, control target can be active power, reactive power, DC voltage or alternating voltage；
Described determines individual optimal value and global optimum in its iterative process for each particle, excellent in population
In the iterative process for changing algorithm, the value of the corresponding all positions of each particle is judged using fitness function, takes its minimum
Value is the optimal value of the particle；The value that all positions of all particles are judged using fitness function, takes minimum value therein
For global optimum；
More new particle speed next time and position, for some particle P, in i+1 time iteration, grain
The position and speed of son can be calculated with the result of i-th iteration, and calculation expression is as follows.
In formula, K represents the position of particle；V represents the speed of particle；C1 and C2 represents the aceleration pulse of particle；ω is represented
The inertial factor of particle；Pbest is the optimal value of particle；Gbest is the global optimum of particle；Rand (1), rand (2) are
It is the random number between (0,1)；Aceleration pulse c1、c2Value be all 2.
Above formula represents the process of particle optimizing, by the formula can continuous more new particle position and speed, so as to
Access the individual optimal value of particle and the global optimum of all particles.
The optimal value and global optimum of the update population solve particle next using fitness function
Fitness value when position, the particle optimal value and global optimum that the fitness value solved at this time and last iteration are obtained
It is compared, obtains particle optimal value.
Compared with prior art, the invention has the advantages that
1. can be improved power converter based on the power converter double-loop control strategy of particle swarm algorithm and resist outside
The ability of boundary interference and uncertain factor, improves its reliability.By online Correction and Control parameter, and asked by intelligent optimization
Optimal control parameter is taken, to realize the intelligent control of power converter.
2. this method is easy to implement.This method is carried out on the basis of the double-loop control strategy of original power converter
It is improved, do not change original controlling unit, but passes through the monitoring to control target, online amendment outer loop control link
Proportion integral modulus there is preferable reliability, economy and operational efficiency so that a whole set of control program is easy to implement.
3. this method is convenient for commercial development.With the grid-connected increase of new energy, power converter will be answered in low voltage electric network
With more and more extensive, the present invention has preferable commercial exploitation prospects and application value.
Detailed description of the invention
Fig. 1 is the power converter double-loop control strategy step block diagram based on particle swarm algorithm；
Fig. 2 is voltage source converter double-closed-loop control block diagram；
Fig. 3 is particle group optimizing PI control system figure
Fig. 4 is the control parameter on-line optimization step block diagram based on particle swarm algorithm；
Fig. 5 is that control simulation result compares figure.
Specific embodiment
Further details of the technical solution of the present invention with emulation experiment with reference to the accompanying drawing.
As shown in figure 1, figure 2, figure 3, figure 4 and figure 5, a kind of power converter double-closed-loop control plan based on particle swarm algorithm
Slightly
The following steps are included:
Step 1) implements the double-loop control strategy of outer voltage and current inner loop to power converter；
Step 2) determines the control parameter adjustable range of the controlling unit of on-line tuning；
Step 3) monitoring control target value；
If step 4), which controls target, exceeds allowed band, control parameter is updated using particle swarm algorithm；
(1) setting the number of iterations initial value is zero；
(2) population calculating parameter, position, speed, aceleration pulse and inertial factor including particle are initialized；
(3) optimal value and global optimum of its initial position are determined for each particle；
(4) individual optimal value and the global optimum in its iterative process are determined for each particle；
(5) more new particle speed next time and position；
(6) optimal value and global optimum of population are updated；
(7) the number of iterations increases by 1, if reaching the number of iterations upper limit, carries out step (8), is otherwise transferred to step (4)；
(8) using particle group optimizing calculated result as power converter control parameter value.
If step 5) controls target without departing from allowed band, keep control parameter constant.
The voltage source type electric power current transformer, by device (such as insulated gate bipolar transistor with turn-off capacity
(IGBT)) power converter formed, has the function of rectification, inversion, AC conversion or DC conversion；
The power control of the double-loop control strategy of the outer voltage and current inner loop, power converter is closed using double
Ring control, outer ring respond active and reactive control target, and inner ring uses current decoupled control strategy, real by rotating coordinate transformation
The decoupling control of existing watt current and reactive current；
The parameter of the described online adaptive adjustment controlling unit, in power converter power outer loop control, than
Particle swarm optimization algorithm optimizing, online dynamic adjustment proportional plus integral control parameter, to improve control is added in example integration control ring
Adaptability of the device processed to extraneous uncertain noises factor；
The monitoring controls target value, for voltage source type electric power current transformer, can simultaneously to two control targets into
Row monitoring control, control target can be active power, reactive power, DC voltage or alternating voltage；
Described determines individual optimal value and global optimum in its iterative process for each particle, excellent in population
In the iterative process for changing algorithm, the value of the corresponding all positions of each particle is judged using fitness function, takes its minimum
Value is the optimal value of the particle；The value that all positions of all particles are judged using fitness function, takes minimum value therein
For global optimum；
More new particle speed next time and position, for some particle P, in i+1 time iteration, grain
The position and speed of son can be calculated with the result of i-th iteration, and calculation expression is as follows.
In formula, K represents the position of particle；V represents the speed of particle；C1 and C2 represents the aceleration pulse of particle；ω is represented
The inertial factor of particle；Pbest is the optimal value of particle；Gbest is the global optimum of particle；Rand (1), rand (2) are
It is the random number between (0,1)；Aceleration pulse c1、c2Value be all 2.
Above formula represents the process of particle optimizing, by the formula can continuous more new particle position and speed, so as to
Access the individual optimal value of particle and the global optimum of all particles.
The optimal value and global optimum of the update population solve particle next using fitness function
Fitness value when position, the particle optimal value and global optimum that the fitness value solved at this time and last iteration are obtained
It is compared, obtains particle optimal value.
Fig. 1 is a kind of key step of the power converter double-loop control strategy based on particle swarm algorithm of the present invention, packet
Include the double-loop control strategy for implementing outer voltage and current inner loop to power converter；Determine the controlling unit of on-line tuning
Control parameter adjustable range；Monitoring control target value；If controlling target exceeds allowed band, is updated and controlled using particle swarm algorithm
Parameter；If controlling target without departing from allowed band, keep control parameter constant.The present invention is with height reliability and adaptively
Property, the power control of power converter can be realized under the disturbance of power grid, to the operational reliability for improving power converter and
Controllability has important practical significance.
Fig. 2 is voltage source converter double-closed-loop control block diagram.
Voltage source converter uses vector control mode, and vector controlled uses the two close cycles knot of current inner loop and target outer loop
Structure.Rotating coordinate system and three-phase power grid voltage synchronous rotary, and d axis is overlapped with power grid A phase voltage vector, d axis component is at this time
For active current, and the q axis component of electric current is then reactive current component.
Under dq coordinate system, the d axis component and q axis component of electric current are all related to voltage, and there are serious coupled relations.Cause
This introduces decoupling algorithm in current inner loop control, and design inner ring current decoupled control device structure is as shown in Figure 2:
Outer ring is power control loop, and inner ring is current regulator.Active a reference value PrefIt is passed through with the difference of active actual value P
Cross a reference value i after PI is adjusted as electric current d axisdref, idle a reference value QrefIt is adjusted with the difference of idle actual value Q by PI
A reference value i as electric current q axis afterwardsqref.Inner ring current differential is by PI link later using inner ring decoupling control and feedforward
Voltage compensation obtains d axis and q axis is worn by standard value ucdAnd ucq, it is sent into PWM modulator and modulates, obtains control signal, realize electricity
The control target of Source Con-verters.
Fig. 3 is particle group optimizing PI control system figure
The performance of PI control ring directly depends on parameter kp、kIValue it is whether reasonable, thus optimize PI control ring parameter tool
There is important meaning.Currently, the parameter of PI control ring mainly by being manually adjusted, not only debug numerous by this method when emulation
It is trivial, and it is different surely obtain optimal value, therefore PI control ring is adjusted using particle swarm algorithm herein, controlling unit
As shown in Figure 3.
Shown in Fig. 3, proportionality coefficient kpEffect be to speed up the response speed of system, improve degree of regulation, value selection
The improper situation for being easy to appear overshoot or precision deficiency.Integral coefficient kIEffect be elimination system steady-state error, equally,
Its value selects the improper problem for being also easy to produce overshoot or degree of regulation deficiency.Thus, the mesh of particle group optimizing PI control ring
Be to find kp、kIOptimal value so that regulating effect is best.
Fig. 4 is the control parameter on-line optimization step block diagram based on particle swarm algorithm；
The process that particle group optimizing PI control ring is realized mainly has the following steps:
(1) the value range i.e. k for the PI control ring parameter for keeping system stable is determinedp、kIValue.
(2) variation of power converter outer loop control target is detected.If control target value changes greatly, particle group optimizing
PI control ring plays a role, adjust automatically kp、kIValue, the PI control ring parameter for making it obtain can satisfy requirement.
(3) using particle swarm optimization algorithm update PI controller parameter, detailed process is as follows.
(a) position of particle is initializedSpeedAceleration pulse c1、c2And inertial factor ω.
Particle swarm optimization algorithm determines the initial positions of all particles, initial speed first in the 0th beginning iteration at this time
Degree, aceleration pulse and inertial factor.The initial position of particle isBe according to system stablize when PI control ring parameter in
Machine selection, particle initial velocityIt is randomly selected in (- 1,1).Aceleration pulse c1、c2It is adjustment experience and society
The weight of meeting experience effect played in its movement, for general problem, its value is all taken as 2.Inertial factor ω calculates population
The convergence of method plays great role, can change the ability of global optimizing and local optimal searching, usual situation by adjusting its value
Under when iteration starts by ω setting it is bigger, later as the number of iterations constantly reduces, in order to simplify operation, this paper value
It is 0.6.Gained PI parameter value had both been met in this way using 10 particles and iteration 20 times in particle swarm optimization algorithm herein
Accuracy meets calculating speed again, can preferably realize the control performance of power converter.
(b) the optimal value pbest and global optimum gbest of its initial position are determined for each particle.
In the parameter k of adjusting PI control ringp、kIWhen, need to evaluate the performance of its dynamic using fitness function.In order to make
Network voltage can quickly restore, using when particle swarm algorithm using control target value integral of absolute value of error as target letter
Number is fitness function, for controlling target value and be voltage.Its expression formula is as shown in Equation 2:
In formula (2), u (t) is voltage at tie point；Δ t is the sampling period.
In view of least absolute error integral can make the dynamic response of system have shorter adjustment time and smaller
Overshoot, therefore, in particle swarm algorithm choose voltage integral of absolute value of error as fitness function.It is fixed when initialization
Optimal value pbest and global optimum the gbest (minimum for meeting fitness function of all particles after initialization of adopted particle
Value) it is respectively as follows:
(c) individual optimal value pbest and the global optimum gbest in its iterative process are determined for each particle.
In the iterative process of particle swarm optimization algorithm, judge that each particle is corresponding all using fitness function
The value of position, taking its minimum value is the optimal value of the particle, as shown in Equation 4.
The value that all positions of all particles are judged using fitness function, taking minimum value therein is global optimum
Value, as shown in Equation 5.
(d) in i+1 time iteration, by formula (6) more new particle speed next time and position.
For particle P, in i+1 time iteration, the position and speed of particle can be carried out with the result of i-th iteration
It calculates, calculation expression is as follows.
In formula (4.29), rand (1), rand (2) are the random numbers between (0,1)；Aceleration pulse c1、c2Value be all
2。
Formula (6) represents the process of particle optimizing, by the formula can continuous more new particle position and speed, so as to
Access the individual optimal value of particle and the global optimum of all particles.For the optimizing of more intuitive analysis particle
Journey, by formula 6 it can be concluded that each particle is iterated to calculate by two extreme values of tracking in an iterative process, to obtain particle
Global optimum in body extreme value and all particles.
For particle P, in i+1 time iteration, position is obtained by formula (7).
In formula (7), Δ KP1For i-th iteration positionIt is directed toward particle individual optimal value KP(pbest)Velocity vector；ΔKP2
For i-th iteration positionIt is directed toward particle global optimum K(gbest)Velocity vector；ΔKP3For with i-th iteration speed's
The identical velocity vector in direction.
By formula (7) it can be concluded that the speed that particle moves during i-th iterationThe reference value mobile as particle makes
Particle P is obtained in i+1 time iterative process to particle optimal value KP(pbest)And global optimum K(gbest)It is close.
(e) the optimal value K of more new particle PP(pbest)With global optimum K(gbest)。
After i-th iteration, the optimal value K of current particle PP(pbest)And global optimum K(gbest)Fitness value
Respectively f (KP(pbest))、f(K(gbest)).Particle P is solved using fitness function to existFitness value when positionParticle optimal value and global optimum that the fitness value solved at this time is obtained with last iteration are compared
Compared with.As shown in formula (8) and (9).
If formula (8) is set up,It will be chosen as the new individual optimal value K of particleP(pbest), otherwise particle optimal value is protected
It holds constant；If formula (9) is set up,It will be chosen as the new global optimum K of particle(gbest), otherwise particle global optimum
It remains unchanged.
(f) step (d) and step (e) are repeated 20 times i.e. iteration 20 times.
(g) the parameter value k in PI control ring is determined after iterationp、kIFor global optimum K at this time(gbest)。
Fig. 5 is that control simulation result compares figure.
In order to verify the validity of particle swarm algorithm optimization PI control ring parameter, be connected on 6kV bus based on voltage
For the static passive compensation device of source type current transformer, simulating, verifying is carried out.It is negative that one induction machine is set near 6kV bus
It carries, using electric motor starting as grid disturbance.In order to reduce grid disturbance pair voltage dip influence, introduce particle group optimizing
Static passive compensation device is incorporated at 6kV bus under conditions of PI algorithm, considers that conventional pair, uncompensated device, addition is closed respectively
The power converter of ring control strategy and the power converter for introducing particle swarm algorithm influence network voltage in the case of three kinds
Improve situation and carries out simulation analysis.
It can be obtained from the figure that when 3s 6kV busbar voltage decline respectively it is larger, and voltage restore it is slower, it is seen that induction conductivity pair
Voltage dip is influenced than more serious, and voltage drop significantly reduces after being incorporated to reactive power compensator, voltage recovery time also obtain compared with
Good improvement.Wherein particle group optimizing control is best to the compensation effect of voltage, the time required to voltage restores and shortest.
Compare figure it is found that the control strategy of particle group optimizing PI control ring improves voltage level of power grid aspect better than traditional
PI control, the reactive compensation grid-connected initial stage is slightly fluctuated, by that can stablize rising after a certain period of time and restore normal level,
In addition, particle group optimizing PI control ring can make network voltage extensive faster in time for conventional PI control ring
Maintenance level is arrived again.Since conventional PI control parameter is fixed, when accessing the induction conductivity of different capabilities, control effect is obtained not
Improve to preferable, but particle group optimizing PI control ring can adjust PI parameter according to the induction conductivity of different capabilities, make its tool
There is good regulating effect, to guarantee that network voltage can quickly and effectively be restored to normal level when induction conductivity starting
And keep stable.
The above simulation result, demonstrates effectiveness of the invention.
Claims (7)
1. a kind of power converter double-closed-loop control method based on particle swarm algorithm, refers to voltage source type electric power current transformer,
Using the double-loop control strategy of outer voltage and current inner loop；Using particle swarm optimization algorithm, online adaptive adjustment control
The parameter of link processed realizes optimal control of the power converter under the various disturbances of power grid；It is characterized by comprising following steps
It is rapid:
Step 1) implements the double-loop control strategy of outer voltage and current inner loop to power converter；
Step 2) determines the control parameter adjustable range of the controlling unit of on-line tuning；
Step 3) monitoring control target value；
If step 4), which controls target, exceeds allowed band, control parameter is updated using particle swarm algorithm；
(1) setting the number of iterations initial value is zero；
(2) population calculating parameter, position, speed, aceleration pulse and inertial factor including particle are initialized；
(3) optimal value and global optimum of its initial position are determined for each particle；
(4) individual optimal value and the global optimum in its iterative process are determined for each particle；
(5) more new particle speed next time and position；
(6) optimal value and global optimum of population are updated；
(7) the number of iterations increases by 1, if reaching the number of iterations upper limit, carries out step (8), is otherwise transferred to step (4)；
(8) using particle group optimizing calculated result as power converter control parameter value；
Individual optimal value and global optimum in its iterative process are determined for each particle, in changing for particle swarm optimization algorithm
During generation, the value of the corresponding all positions of each particle is judged using fitness function, taking its minimum value is the particle
Optimal value；The value that all positions of all particles are judged using fitness function, taking minimum value therein is global optimum
Value；
The process that particle group optimizing PI control ring is realized mainly has the following steps:
1. determining the value range i.e. k for the PI control ring parameter for keeping system stablep、kIValue；
2. the variation of power converter outer loop control target is detected, if control target value changes greatly, particle group optimizing PI control
Ring processed plays a role, adjust automatically kp、kIValue, the PI control ring parameter for making it obtain can satisfy requirement；
3. updating PI controller parameter using particle swarm optimization algorithm, detailed process is as follows:
(a) position of particle is initializedSpeedAceleration pulse c1、c2And inertial factor ω；
Particle swarm optimization algorithm in the 0th beginning iteration, determines the initial position of all particles at this time, initial velocity, adds first
Fast constant and inertial factor；The initial position of particle isIt is to be randomly choosed in PI control ring parameter when being stablized according to system
, particle initial velocityIt is randomly selected in (- 1,1)；Aceleration pulse c1、c2It is adjustment experience and social experience
The weight of the effect played in its movement, value are all taken as 2；Inertial factor ω plays a role the convergence of particle swarm algorithm,
The ability for changing global optimizing and local optimal searching by adjusting its value, bigger, the Zhi Housui that ω is arranged when iteration starts
The number of iterations constantly reduce, in order to simplify operation, value 0.6；10 particles are used in particle swarm optimization algorithm and are changed
In generation 20 times, had not only met the accuracy of gained PI parameter value in this way but also had met calculating speed, and can be realized power converter
Control performance；
(b) the optimal value pbest and global optimum gbest of its initial position are determined for each particle；
In the parameter k of adjusting PI control ringp、kIWhen, need to evaluate the performance of its dynamic using fitness function；In order to make power grid
Voltage can quickly restore, using when particle swarm algorithm using control target value integral of absolute value of error as objective function
Fitness function, for controlling target value and be voltage；Shown in its expression formula such as formula (2):
In formula (2), f is the target voltage of control；u*For network voltage；U (t) is voltage at tie point；Δ t is the sampling period；
The integral of absolute value of error of voltage is chosen in particle swarm algorithm as fitness function；When initialization, particle is defined most
Figure of merit pbest and global optimum gbest, the minimum value for meeting fitness function of all particles, is respectively as follows: after initialization
In formula (3), pbest is the optimal value of particle；Gbest is global optimum；For control ring parameter kpInitialization grain
Sub- position；For control ring parameter kIInitialization particle position；
(c) individual optimal value pbest and the global optimum gbest in its iterative process are determined for each particle；
In the iterative process of particle swarm optimization algorithm, the corresponding all positions of each particle are judged using fitness function
Value, take its minimum value be the particle optimal value, as shown in Equation 4；
In formula (4), KP(pbest)For the particle optimal value of particle；For control ring parameter kpParticle optimal value；For control
Ring parameter k processedIParticle optimal value；
The value that all positions of all particles are judged using fitness function, taking minimum value therein is global optimum, such as
Shown in formula 5；
In formula (5), K(gbest)For global optimum；For control ring parameter kpGlobal optimum；For control ring parameter
kIGlobal optimum；
(d) in i+1 time iteration, by formula (6) more new particle speed next time and position；
For particle P, in i+1 time iteration, the result of the position and speed i-th iteration of particle is calculated, meter
Operator expression formula is as follows；
In formula (6), rand (1), rand (2) are the random numbers between (0,1)；Aceleration pulse c1、c2Value be all 2；
Formula (6) represents the process of particle optimizing, by the position and speed of the continuous more new particle of the formula, so as to obtain grain
The individual optimal value of son and the global optimum of all particles；For the searching process of more intuitive analysis particle, by formula 6
Show that each particle is iterated to calculate by two extreme values of tracking in an iterative process, to obtain particle individual extreme value and institute
There is the global optimum in particle；
For particle P, in i+1 time iteration, position is obtained by formula (7)；
In formula (7), Δ KP1For i-th iteration positionIt is directed toward particle individual optimal value KP(pbest)Velocity vector；ΔKP2It is
I iterative positionIt is directed toward particle global optimum K(gbest)Velocity vector；ΔKP3For with i-th iteration speedDirection
Identical velocity vector；
The speed of particle movement during i-th iteration is obtained by formula (7)The reference value mobile as particle makes particle P exist
To particle optimal value K in i+1 time iterative processP(pbest)And global optimum K(gbest)It is close；
(e) the optimal value K of more new particle PP(pbest)With global optimum K(gbest)；
After i-th iteration, the optimal value K of current particle PP(pbest)And global optimum K(gbest)Fitness value difference
For f (KP(pbest))、f(K(gbest))；Particle P is solved using fitness function to existFitness value when positionIt will
The particle optimal value and global optimum that the fitness value solved at this time and last iteration obtain are compared；Such as formula (8)
(9) shown in；
If formula (8) is set up,It will be chosen as the new individual optimal value K of particleP(pbest), otherwise particle optimal value is kept not
Become；If formula (9) is set up,It will be chosen as the new global optimum K of particle(gbest), otherwise particle global optimum is kept
It is constant；
(f) step (d) and step (e) are repeated 20 times i.e. iteration 20 times；
(g) the parameter value k in PI control ring is determined after iterationp、kIFor global optimum K at this time(gbest)；
If step 5) controls target without departing from allowed band, keep control parameter constant.
2. a kind of power converter double-closed-loop control method based on particle swarm algorithm according to claim 1, feature
Be: voltage source type electric power current transformer, the power converter being made of the device with turn-off capacity have rectification, inversion, friendship
Rheology stream or DC conversion function.
3. a kind of power converter double-closed-loop control method based on particle swarm algorithm according to claim 1, feature
Be: the power control of the double-loop control strategy of outer voltage and current inner loop, power converter uses double-closed-loop control, outside
Ring responds active and reactive control target, and inner ring realizes active electricity by rotating coordinate transformation using current decoupled control strategy
The decoupling control of stream and reactive current.
4. a kind of power converter double-closed-loop control method based on particle swarm algorithm according to claim 1, feature
It is: the parameter of online adaptive adjustment controlling unit, in power converter power outer loop control, in proportional plus integral control
Particle swarm optimization algorithm optimizing, online dynamic adjustment proportional plus integral control parameter, to improve controller to the external world is added in ring
The adaptability of uncertain noises factor.
5. a kind of power converter double-closed-loop control method based on particle swarm algorithm according to claim 1, feature
Be: monitoring control target value is monitored control for voltage source type electric power current transformer, while to two control targets, controls
Target processed is active power, reactive power, DC voltage or alternating voltage.
6. a kind of power converter double-closed-loop control method based on particle swarm algorithm according to claim 1, feature
It is: more new particle speed next time and position, for some particle P, in i+1 time iteration, the position of particle
It is calculated with the result of speed i-th iteration, calculation expression is as follows:
In formula, K represents the position of particle；V represents the speed of particle；C1And C2Represent the aceleration pulse of particle；ω represents particle
Inertial factor；Pbest is the optimal value of particle；Gbest is the global optimum of particle；Rand (1), rand (2) are (0,1)
Between random number；Aceleration pulse c1、c2Value be all 2；
Above formula represents the process of particle optimizing, by the position and speed of the continuous more new particle of the formula, so as to obtain grain
The individual optimal value of son and the global optimum of all particles.
7. a kind of power converter double-closed-loop control method based on particle swarm algorithm according to claim 1, feature
It is: updates the optimal value and global optimum of population, solves particle at next position using fitness function
Fitness value compares particle optimal value and global optimum that the fitness value solved at this time is obtained with last iteration
Compared with obtaining particle optimal value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610408203.3A CN105932860B (en) | 2016-06-08 | 2016-06-08 | A kind of power converter double-loop control strategy based on particle swarm algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610408203.3A CN105932860B (en) | 2016-06-08 | 2016-06-08 | A kind of power converter double-loop control strategy based on particle swarm algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105932860A CN105932860A (en) | 2016-09-07 |
| CN105932860B true CN105932860B (en) | 2019-05-03 |
Family
ID=56832833
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610408203.3A Active CN105932860B (en) | 2016-06-08 | 2016-06-08 | A kind of power converter double-loop control strategy based on particle swarm algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105932860B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104950680A (en) * | 2015-06-18 | 2015-09-30 | 温州大学 | Shore power controller parameter setting optimization method |
| CN104965981A (en) * | 2015-06-18 | 2015-10-07 | 温州大学 | Multi-parameter optimization method of reversible conversion shore power network side controller |
| CN105631518A (en) * | 2015-12-23 | 2016-06-01 | 西安理工大学 | Multi-parameter multi-object chaotic particle swarm parameter optimization method |
-
2016
- 2016-06-08 CN CN201610408203.3A patent/CN105932860B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104950680A (en) * | 2015-06-18 | 2015-09-30 | 温州大学 | Shore power controller parameter setting optimization method |
| CN104965981A (en) * | 2015-06-18 | 2015-10-07 | 温州大学 | Multi-parameter optimization method of reversible conversion shore power network side controller |
| CN105631518A (en) * | 2015-12-23 | 2016-06-01 | 西安理工大学 | Multi-parameter multi-object chaotic particle swarm parameter optimization method |
Non-Patent Citations (1)
| Title |
|---|
| 《含STATCOM直驱式风力发电系统电压稳定控制研究》;唐东成;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20150415;全文 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105932860A (en) | 2016-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109642536B (en) | Control or processing system and method | |
| CN103999315B (en) | Coordinated control method and controller thereof for the electromotor and SVC that improve power plant active power volume of production | |
| CN108649780A (en) | A kind of LCL filter parameter optimization method considering light current inverter stability off the net | |
| CN110492803A (en) | Permanent magnet direct-driving aerogenerator parameter identification method based on MDPSO | |
| CN106712552B (en) | A kind of aviation more electric engin VIENNA rectifier control method | |
| CN110206686A (en) | A kind of adaptive maximum power tracking and controlling method for wind power generating set | |
| CN112688307B (en) | AC/DC hybrid micro-grid controller and control method | |
| Zamzoum et al. | Active and reactive power control of wind turbine based on doubly fed induction generator using adaptive sliding mode approach | |
| Abo-Khalil et al. | Sensorless control for DFIG wind turbines based on support vector regression | |
| CN113541146B (en) | Power flow calculation optimization method of power system considering distributed power supply | |
| WO2022121446A1 (en) | Control system, reactive voltage control method and device, medium, and calculation device | |
| CN108879721B (en) | Frequency control method based on wind power consumption | |
| CN106953361B (en) | Exchange micro-grid operational control method and device | |
| CN109687468A (en) | A kind of micro-capacitance sensor voltage distributed control method based on electric power spring | |
| CN117458534A (en) | Novel liquid flow energy storage peak regulation and frequency modulation method and device | |
| CN107591848A (en) | A kind of droop control method and system | |
| CN105932860B (en) | A kind of power converter double-loop control strategy based on particle swarm algorithm | |
| CN104795836B (en) | A kind of single-phase photovoltaic grid-connected generating detecting system and its non-linear current control method | |
| CN113922384B (en) | Distributed reactive voltage optimization coordination control method for wind farm | |
| CN114024340B (en) | Offshore wind power plant control system and model predictive control method | |
| CN110880778A (en) | Improved nonlinear droop control method for multi-terminal flexible direct-current power transmission system | |
| CN115498616A (en) | Method for quickly identifying high-frequency oscillation leading link of flexible direct-current transmission system | |
| Hburi et al. | Maximum power point tracking control for wind turbines with battery storage system | |
| Venkatesh et al. | A Novel Approach to Optimal Design of PI Controller of Doubly Fed Induction Generator using Particle Swarm Optimization | |
| Dehghani et al. | Dynamic behavior control of induction motor with STATCOM |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |