CN109217702A - A kind of MMC parameter optimization method based on adaptive glowworm swarm algorithm - Google Patents
A kind of MMC parameter optimization method based on adaptive glowworm swarm algorithm Download PDFInfo
- Publication number
- CN109217702A CN109217702A CN201710544780.XA CN201710544780A CN109217702A CN 109217702 A CN109217702 A CN 109217702A CN 201710544780 A CN201710544780 A CN 201710544780A CN 109217702 A CN109217702 A CN 109217702A
- Authority
- CN
- China
- Prior art keywords
- firefly
- parameter
- mmc
- adaptive
- value
- 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.)
- Pending
Links
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
- H02M7/00—Conversion of ac power input into dc power output; Conversion of dc power input into ac power output
- H02M7/42—Conversion of dc power input into ac power output without possibility of reversal
- H02M7/44—Conversion of dc power input into ac power output without possibility of reversal by static converters
- H02M7/48—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode
- H02M7/483—Converters with outputs that each can have more than two voltages levels
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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
- H02M7/00—Conversion of ac power input into dc power output; Conversion of dc power input into ac power output
- H02M7/42—Conversion of dc power input into ac power output without possibility of reversal
- H02M7/44—Conversion of dc power input into ac power output without possibility of reversal by static converters
- H02M7/48—Conversion of dc power input into ac power output without possibility of reversal by static converters using discharge tubes with control electrode or semiconductor devices with control electrode
- H02M7/483—Converters with outputs that each can have more than two voltages levels
- H02M7/4835—Converters with outputs that each can have more than two voltages levels comprising two or more cells, each including a switchable capacitor, the capacitors having a nominal charge voltage which corresponds to a given fraction of the input voltage, and the capacitors being selectively connected in series to determine the instantaneous output voltage
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Feedback Control In General (AREA)
Abstract
The invention patent discloses a kind of MMC parameter optimization method based on adaptive glowworm swarm algorithm, and the optimization that can carry out control parameter to the modularization multi-level converter (MMC) using phase-shifting carrier wave modulation calculates.The technical program key step includes: the adaptive glowworm swarm algorithm parameter of initialization, controller parameter is expressed as firefly position, there is shown the matrix of firefly population；Initialize the position distribution of firefly；Simulation model is run, target function value, the Attraction Degree of each firefly are calculated；Firefly searches other fireflies according to Attraction Degree size, is updated according to location formula to position；Judge whether to meet iteration termination condition, if meeting the controller parameter after as optimizing result output.If conditions are not met, then updating brightness and Attraction Degree again, searched, until meeting condition.The simulation result explanation of embodiment can significantly improve precision of fuzzy controller after this method optimizes, and optimize the current output waveform of MMC.
Description
Technical field
The invention patent relates to the power electronics fields in HVDC transmission system, are mainly used in flexible direct current
The modularization multi-level converter part for link of transmitting electricity, specifically designs a kind of controller of modularization multi-level converter (MMC)
Parameter optimization method.
Background technique
Voltage-source type HVDC transmission system (VSC-HVDC) is widely used in electric system transmission of electricity link, and voltage source
Inverter is its core component.German scholar Lesnicar and Marquardt propose modularization multi-level converter (MMC) with
Its unique high modularization structure becomes one kind superior in VSC-HVDC with advantages such as good fault ride-through capacities
Topological structure.MMC has modular topological structure, opens the variation of submodule number purpose using upper and lower bridge arm and realizes voltage and function
The change of rate grade, output voltage waveforms can save the friendship of large capacity smoothly close to ideal sinusoidal waveform in this way on hardware
Filter is flowed, cost is saved.But power dissipation is stored in the capacitor of each submodule by MMC, therefore submodule at runtime
The equalization problem of capacitance voltage is particularly critical.
For the equalization problem of submodule capacitor voltage, there is scholar to propose to answer phase-shifting carrier wave modulation strategy (CPS-SPWM)
For in MMC, according to two kinds of principles of the submodule equipartition of energy and electric voltage equalization, design to be suitable for the voltage balancing control plan of MMC
Slightly.But the big multi-parameter of the strategy relies primarily on artificial experience setting, related commissioning heavy workload.In order to solve this problem,
The present invention proposes the MMC Optimization about control parameter based on adaptive glowworm swarm algorithm in phase-shifting carrier wave modulation strategy theoretical basis
Method.
Summary of the invention
Present invention aims at carry out parameter optimization for using the MMC controller of phase-shifting carrier wave modulator approach, propose one
Parameter optimization method of the kind based on adaptive glowworm swarm algorithm, makes the parameter of controller reach best configuration, mentions high control precision
And control effect, inverter output current wave are improved.
To achieve the above object, the technical solution adopted is that:
(1) parameters such as population number, the number of iterations, the initial step length of adaptive glowworm swarm algorithm are initialized, are arranged in population
Firefly number is N, is W by the controller parameter of required optimization, then the position vector of each firefly has W control parameter
Composition, i.e., the array of one two dimension D=W, the firefly population can be expressed as the matrix of N* (D+2).Initial parameter choose according to
It is chosen according to requirement of engineering in the range that each parameter generally allows for.
(2) position distribution of each firefly is initialized according to location formula, and initializes controller ginseng to be optimized
Number.
(3) operation simulation model calculates the target function value of each firefly according to objective function Equation.
(4) each firefly searches other fireflies according to Attraction Degree size in algorithm, is greater than itself to brightness
Individual, moved to it, and position is updated according to location formula.
(5) judge whether to meet iteration termination condition, if meeting result output is optimal solution, after being optimized
Controller parameter.If conditions are not met, then updating brightness and Attraction Degree by third step again, it is iterated search again, until meeting
Condition.
The invention patent has following gain effect:
The MMC parameter optimization method based on adaptive glowworm swarm algorithm that the invention patent is proposed, Algorithm Convergence is strong,
It is possible to prevente effectively from Premature Convergence and the case where falling into local optimum, for the controller of modularization multi-level converter (MMC)
After carrying out parameter optimization, output current wave is improved, loop current suppression effect enhancing, for the equalization stable of submodule voltage
Play positive effect.
Detailed description of the invention
Fig. 1 MMC topology diagram
The flow chart of parameter optimization in Fig. 2 present invention
MMC control block diagram in Fig. 3 embodiment of the present invention
Optimization Simulation result figure in Fig. 4 embodiment of the present invention
Specific embodiment
The MMC parameter optimization method based on adaptive glowworm swarm algorithm is carried out specifically with example with reference to the accompanying drawing
It is bright.
As shown in Figure 1 in MMC topological diagram, modular multilevel (MMC) has 6 bridge arms, and each bridge arm is by many submodules
It cascades.This example uses 11 level MMC HVDC transmission systems, so each bridge arm has 10 submodules, upper and lower bridge
Each 5, arm.11 level MMC control systems are built in Matlab/simulink.AC system voltage is 220V, bridge arm inductance
3mH is taken, nominal DC side voltage is 250V, transformer capacity 1200MVA.Submodule capacitor is 1.3mF.
From the figure 3, it may be seen that the control of MMC capacitance voltage is divided into two parts, capacitor voltage balance and bridge arm loop current suppression.Capacitor
Be balance of voltage mesh guarantee submodule on bridge arm voltage-tracing its with reference to threshold voltage, pass through each submodule electricity of real-time monitoring
Pressure value ucWith nominal reference threshold voltage ucref, and determine according to the direction of bridge arm current the working condition of the submodule.When bridge arm electricity
Stream is timing, if ucLess than nominal reference voltage ucref, circulator should from DC side obtain energy to capacitor charging, adjust at this time
Voltage u processedcIt * is positive value, module capacitance will constantly charge and voltage increases；Work as uc> ucrefWhen, modulation voltage ucIt * is negative value, this
When the submodule charging time reduce, capacitance voltage amplitude incrementss reduce.
The purpose of bridge arm loop current suppression is to limit the circulation between bridge arm in a certain range, reduces it to bridge arm current
Influence.By the double-closed-loop control of external voltage outer ring and current inner loop, make the variation of circulation tracking circulation reference value, by ring
Flow control is within tolerance interval.Outer voltage uses pi regulator, and control phase element average voltage level tracks nominal reference
Voltage signal, and the PI controller of current inner loop is used to control the variation of circulation tracking circulation reference value, output is used as capacitor
The regulated quantity u of voltage balance controlzj*。
It is as follows that outer voltage controls formula
Current inner loop control formula is as follows
In above-mentioned formulaucjiFor the capacitance voltage of i-th of submodule of jth phase.
Finally, the amplitude of phase-shifting carrier wave modulating wave is
Upper bridge arm modulating wave
Lower bridge arm modulating wave
This example optimal target is exactly that two controller parameters in the control of bridge arm circulation are optimized.It is arranged first
Firefly population number is 50, and firefly population number is 50, and the number of iterations is 20 times, and dimension 4, step-length initial value takes 0.02, is attracted
Degree initial value takes 0.5.There are four Optimal Parameters needed for controller, respectively Proportional coefficient K p1, Kp2 and integral parameter Ki1, Ki2.
For set firefly, specific algorithm Optimization Steps are as follows
(1) each firefly i is by a vector xiIt indicates, wherein m is the number for needing optimal control parameter
The value of step factor α influences the distance that firefly is moved in search space, and bigger step factor value has
Conducive to remote search.And the value of optical absorption intensity γ influences Attraction Degree with the change degree of distance, under normal circumstances
Value range is between 0 to 10.The selection of the two coefficients influences convergence and final result.Firefly in the present invention
Fireworm algorithm adaptivity just embodies herein, and α and γ participate in whole search process, and the addition of auto-adaptive parameter will make algorithm
Convergence greatly increase
(2) initial position of firefly is obtained by formula
Wherein, rand is to obey equally distributed random number on 0 to 1.
(3) brightness of each firefly is obtained by formula,
Ii=f (xm)
F (x) is generally corresponding target function value.Objective function is chosen for energy reaction system regulation quality in the present invention
Time multiplies objective function of the integral of absolute value of error ITAE as optimizing.Its expression formula is
E (t) is to miss absolute value of the difference；T is time definite value, and general value is larger to allow system to enter stabilization.Due to control
The purpose of the parameter optimization of device processed is that input value is allowed to track given reference value, i.e., PI controller error input e (t) is minimum, and firefly
What fireworm algorithm optimizing was asked is maximum value.So objective function should be rewritten as
(4) the Attraction Degree formula between firefly i and j is as follows
βij=(βMax, i, j-βMin, i, j)exp(-γmrM, n 2)+βMin, i, j
Wherein, rI, jFor the distance between i and j
γ is light intensity absorption coefficient, depends on the dynamic range of search space.
(5) when firefly i is mobile to the firefly j stronger than oneself brightness, its position will be updated by location formula.
xi=xi+βij×(xj-xi)+α×(rand-1/2)
Wherein α is step factor, and value is 0 to 1.Rand is to obey equally distributed random number on [0,1].Pass through addition
α × (rand-1/2) disturbance term, increases search range, avoids falling into local optimum.
The core of glowworm swarm algorithm is exactly to constantly update brightness and Attraction Degree, allows firefly in iteration moving process, most
After concentrate on the maximum position of brightness, as optimal solution.
Optimization method process of the invention is as shown in Fig. 2, the firefly population in the present embodiment is expressed in matrix as
N is population number, i.e. it is 6 in this example that 50, D, which is dimension,.After initializing each firefly position, each firefly
Position is the parameter setting of controller, by the way that simulation model is run multiple times, makes the brightness (i.e. target function value) of firefly no
It is disconnected to update, location updating iteration is carried out by algorithm firefly, it is finally most of to gather near a certain position, represented by the position
Parameter be optimal solution, can realize the optimization of MMC parameter.Since adaptive glowworm swarm algorithm can be optimized with adjust automatically
In parameter, the case where local optimum can occur to avoid algorithm Premature Convergence, optimizing effect is better than general optimization algorithm.
Optimization method of the invention can be generalized to the parameter optimization of other controllers, have certain extensibility.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, although referring to aforementioned reality
Applying example, invention is explained in detail, for those skilled in the art, still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features.It is all in essence of the invention
Within mind and principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (4)
1. a kind of MMC parameter optimization method based on adaptive glowworm swarm algorithm comprising the steps of:
1) parameters such as population number, the number of iterations, the initial step length of adaptive glowworm swarm algorithm are initialized, firefly in population is set
Number is N, is W by the controller parameter of required optimization, then the position vector of each firefly is made of W control parameter,
The array of i.e. one two dimension D=W, the firefly population can be expressed as the matrix of N* (D+2).Initial parameter basis for selecting engineering
The range generally allowed in each parameter is needed to choose.
2) position distribution of each firefly is initialized according to location formula.The location information of each firefly is to excellent
Each parameter value of the controller of change.
3) operation simulation model calculates the corresponding target function value of each firefly, as adaptively according to objective function Equation
Firefly brightness in algorithm.
4) each firefly searches other fireflies according to Attraction Degree size in algorithm, and of itself is greater than to brightness
Body is moved to it, and is updated according to location formula to position.
5) judge whether to meet iteration termination condition, if meeting result output is optimal solution, the control after being optimized
Device parameter.If conditions are not met, then updating brightness and Attraction Degree by third step again, it is iterated search again, until meeting condition.
2. the MMC parameter optimization method according to claim 1 based on adaptive glowworm swarm algorithm, which is characterized in that step
It is rapid 1) in foundation for adaptive firefly kind mass matrix, each firefly i is by a vector xiIt indicates, wherein m is to need
Want the number of optimal control parameter
The value of step factor α influences the distance that firefly is moved in search space, and the value of optical absorption intensity γ influences
To Attraction Degree with the change degree of distance, value range is between 0 to 10 under normal circumstances.Also α and γ are regard as Optimal Parameters
Whole search process is participated in, the addition of auto-adaptive parameter will be such that convergence greatly increases.
3. the MMC parameter optimization method according to claim 1 based on adaptive glowworm swarm algorithm, which is characterized in that step
It is rapid 3) in, by run simulation model obtain target function value.Objective function is chosen for the time of energy reaction system regulation quality
Multiply objective function of the integral of absolute value of error ITAE as optimizing.Its expression formula is
The purpose of the parameter optimization of controller is that input value is allowed to track given reference value, i.e. PI controller error input e (t) most
It is small, and what glowworm swarm algorithm optimizing was asked is maximum value.So objective function should be rewritten as
Optimization object function, which is arranged, is
Q (t)=c1f1(t)+c2f2(t)
4. the MMC parameter optimization method according to claim 1 based on adaptive glowworm swarm algorithm, which is characterized in that step
It is rapid 5) in, emulation controller, calculating target function are run by continuous iteration, brightness as firefly simultaneously updates every
The brightness of one firefly, it is mobile by algorithm firefly, it is final most of when meeting termination condition when reaching the number of iterations
Gather near a certain position, parameter represented by the position is optimal solution, can realize the optimization of MMC parameter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710544780.XA CN109217702A (en) | 2017-07-06 | 2017-07-06 | A kind of MMC parameter optimization method based on adaptive glowworm swarm algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710544780.XA CN109217702A (en) | 2017-07-06 | 2017-07-06 | A kind of MMC parameter optimization method based on adaptive glowworm swarm algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109217702A true CN109217702A (en) | 2019-01-15 |
Family
ID=64992795
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710544780.XA Pending CN109217702A (en) | 2017-07-06 | 2017-07-06 | A kind of MMC parameter optimization method based on adaptive glowworm swarm algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109217702A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112764785A (en) * | 2020-12-24 | 2021-05-07 | 江苏云涌电子科技股份有限公司 | Method for automatically upgrading multi-stage controller |
| CN113541170A (en) * | 2021-06-16 | 2021-10-22 | 武汉理工大学 | Fuel cell emergency power supply grid-connected inversion control method and system |
-
2017
- 2017-07-06 CN CN201710544780.XA patent/CN109217702A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112764785A (en) * | 2020-12-24 | 2021-05-07 | 江苏云涌电子科技股份有限公司 | Method for automatically upgrading multi-stage controller |
| CN113541170A (en) * | 2021-06-16 | 2021-10-22 | 武汉理工大学 | Fuel cell emergency power supply grid-connected inversion control method and system |
| CN113541170B (en) * | 2021-06-16 | 2023-11-24 | 武汉理工大学 | Grid-connected inversion control method and system for emergency power supply of fuel cell |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Meng et al. | Dynamic consensus algorithm based distributed global efficiency optimization of a droop controlled DC microgrid | |
| CN110048606B (en) | DC-DC boost converter dynamic sliding mode voltage control method based on interval two-type self-adaptive fuzzy neural network | |
| CN108365784A (en) | Based on the control method for brushless direct current motor for improving PSO-BP neural networks | |
| Piltan et al. | Novel artificial control of nonlinear uncertain system: design a novel modified PSO SISO Lyapunov based fuzzy sliding mode algorithm | |
| Lu et al. | The PID Controller Based on the Artificial Neural Network and the Differential Evolution Algorithm. | |
| Yeganeh et al. | Intelligent secondary control of islanded AC microgrids: A brain emotional learning-based approach | |
| Khorashadizadeh et al. | Voltage tracking control of DC-DC boost converter using brain emotional learning | |
| CN106959636B (en) | Intelligence adapts to the modularized distribution type driving control system and method for support surface | |
| CN109217702A (en) | A kind of MMC parameter optimization method based on adaptive glowworm swarm algorithm | |
| CN109980666A (en) | A kind of adaptive disturbance compensation passive control method of microgrid mixed energy storage system | |
| CN105226711A (en) | A kind of control method of grid-connected inverter based on fuzzy control | |
| Massenio et al. | Data-driven sparsity-promoting optimal control of power buffers in DC microgrids | |
| Ko et al. | Analysis, design and implementation of fuzzy logic controlled quasi-resonant zero-current switching switched-capacitor bidirectional converter | |
| Huangfu et al. | Learning-based optimal large-signal stabilization for DC/DC boost converters feeding CPLs via deep reinforcement learning | |
| Oshnoei et al. | Data-driven coordinated control of AVR and PSS in power systems: A deep reinforcement learning method | |
| Dhivya et al. | Neural network controller for boost converter | |
| CN113420863A (en) | Energy storage bidirectional converter control parameter setting method based on Harris eagle algorithm | |
| CN113541170A (en) | Fuel cell emergency power supply grid-connected inversion control method and system | |
| Ye et al. | Deep Deterministic Policy Gradient Algorithm Based Reinforcement Learning Controller for Single-Inductor Multiple-Output DC-DC Converter | |
| CN115987086A (en) | Single-switch DC-DC converter on-line control method based on neural network | |
| Mohammadzadeh et al. | Application of mixture of experts in machine learning-based controlling of DC-DC power electronics converter | |
| CN110518625A (en) | A kind of gird-connected inverter DC component suppressing method of variable learning rate BP-PID control | |
| Pandey et al. | PSS tuning with firefly driven knowledge domain-a smart control concept | |
| CN114637204A (en) | Virtual synchronous machine control parameter setting method based on mixed H2/H infinity performance index | |
| Al-Mosawi et al. | An Adaptive Parallel Fuzzy And Proportional Integral Controller (APFPIC) For The Contractor Pneumatic Muscle Actuator Position Control |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| DD01 | Delivery of document by public notice | ||
| DD01 | Delivery of document by public notice |
Addressee: Jiao Xiaopeng Document name: Notification of before Expiration of Request of Examination as to Substance |
|
| DD01 | Delivery of document by public notice | ||
| DD01 | Delivery of document by public notice |
Addressee: Jiao Xiaopeng Document name: Deemed as a notice of withdrawal |
|
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190115 |