CN107168052A - A kind of MMC HVDC system control parameters optimization methods - Google Patents
A kind of MMC HVDC system control parameters optimization methods Download PDFInfo
- Publication number
- CN107168052A CN107168052A CN201710258224.6A CN201710258224A CN107168052A CN 107168052 A CN107168052 A CN 107168052A CN 201710258224 A CN201710258224 A CN 201710258224A CN 107168052 A CN107168052 A CN 107168052A
- Authority
- CN
- China
- Prior art keywords
- mrow
- particle
- msubsup
- mmc
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 68
- 238000005457 optimization Methods 0.000 title claims abstract description 28
- 239000002245 particle Substances 0.000 claims abstract description 113
- 230000003044 adaptive effect Effects 0.000 claims abstract description 36
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 23
- 230000006870 function Effects 0.000 claims abstract description 14
- 230000035772 mutation Effects 0.000 claims abstract description 14
- 238000004088 simulation Methods 0.000 claims abstract description 10
- 101100499229 Mus musculus Dhrsx gene Proteins 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims description 8
- 230000001133 acceleration Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000006978 adaptation Effects 0.000 claims description 3
- 238000013178 mathematical model Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 2
- 230000010354 integration Effects 0.000 claims description 2
- 238000012886 linear function Methods 0.000 claims description 2
- 230000007246 mechanism Effects 0.000 claims description 2
- 238000012804 iterative process Methods 0.000 claims 1
- 230000005055 memory storage Effects 0.000 abstract 3
- 230000021615 conjugation Effects 0.000 abstract 1
- NUFBIAUZAMHTSP-UHFFFAOYSA-N 3-(n-morpholino)-2-hydroxypropanesulfonic acid Chemical compound OS(=O)(=O)CC(O)CN1CCOCC1 NUFBIAUZAMHTSP-UHFFFAOYSA-N 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 238000010187 selection method Methods 0.000 description 3
- 238000011217 control strategy Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 201000004569 Blindness Diseases 0.000 description 1
- 208000011597 CGF1 Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000004039 social cognition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to electric power system model emulation and control field, more particularly to a kind of MMC HVDC system control parameters optimization methods, it is included on PSCAD and builds MMC HVDC simulation models to calculate adaptive value, the multi-objective particle swarm algorithm of operational development carries out optimizing to MMC HVDC control system PI parameters on MATLAB；Hierarchy optimization is carried out to multiple PI controller parameters.During algorithm iteration, the non-domination solution of acquisition is added to after external memory storage, mutation operation is carried out to the position of all particles in external memory storage and external memory storage is updated, a kind of leader's particle choosing method based on membership function is proposed.The present invention is added the diversity of particle using mutation operation, improves ability of searching optimum while basic multi-objective particle swarm algorithm feature is remained；Convergence is improved using leader's particle choosing method based on membership function, it is high with Practical Project conjugation.
Description
Technical Field
The invention belongs to the technical field of optimization of control parameters of power systems, and particularly relates to a method for optimizing control parameters of a modular multi-level converter type high-voltage direct-current transmission engineering system by using a multi-objective particle swarm algorithm.
Background
As a new voltage source type converter, a modular multilevel converter has been proposed so far to gain wide attention, and it uses a sub-module cascade mode to achieve the improvement of voltage class and the improvement of transmission capacity of the converter, and has many technical advantages due to its modular structure. So far, the modular multilevel converter type high-voltage direct-current transmission project, namely MMC-HVDC, has been successfully applied at home and abroad and is paid much attention.
The multi-module topology and multi-link control strategy needs to consider complex coordination control in simulation research and engineering practice, so that the performance requirement on the control system is very high. Proportional-integral (PI) controllers are widely used in practical engineering due to the advantages of fast adjustment, simple structure, easy understanding of parameter definition, easy implementation and the like, but the parameters are usually obtained by a trial-and-error method in the engineering, so that the method has certain blindness, and the workload and the efficiency are likely to be increased.
Disclosure of Invention
In order to solve the above problems, the present invention provides a method for optimizing control parameters of an MMC-HVDC system, which is characterized by comprising the following steps:
step 1, building an MMC-HVDC simulation model on PSCAD as a calculation model for parameter optimization to calculate an adaptive value;
step 2, compiling an improved multi-target particle swarm optimization algorithm on MATLAB, taking the control parameters of the MMC-HVDC system to be optimized as the positions of particles, and layering the control parameters to be optimized;
step 3, initializing algorithm parameters and particle information, and enabling the iteration number j to be 1;
step 4, entering a main loop, selecting leader particles by using a method combining a method based on congestion degree and a method based on membership function, updating the speed and the position of the particles, optimizing inner loop parameters, adding a non-dominated solution into an external memory, carrying out variation on the particles in the external memory and updating the external memory;
step 5, selecting leader particles, updating the particle speed and position, optimizing outer ring parameters, adding a non-dominated solution into an external memory, carrying out variation on the particles in the external memory and updating the external memory;
and 6, repeating the step 4 and the step 5 until the maximum iteration number is reached.
In the step 1, the MMC-HVDC simulation model controller adopts a vector control technology in direct current control, comprises inner loop current control and outer loop output control, converts three-phase alternating current quantity under an ABC coordinate system into direct current quantity under a DQ coordinate system and establishes a mathematical model of the MMC, thereby realizing DQ axis decoupling, simplifying the mathematical model of a current converter and being suitable for controlling the three-phase MMC. Two sets of PI controllers are respectively arranged in the inner loop current control and the outer loop output control.
The adaptive value is calculated by the control target of the converter through an equation (1), and the calculation method adopts an integral Time and Absolute error index of an Absolute error value multiplied by Time. In the formula yrefFor the control target reference value, y is the control target actual value, and the upper integration limit T is the dynamic process time.
In step 2, the improved multi-target particle swarm algorithm is based on the multi-target particle swarm algorithm proposed by Coello in 2004, an external memory and a self-adaptive grid mechanism are adopted to store a non-dominated solution, and leader particles are selected from the non-dominated solution to iteratively update the information of the particles. The position of the particle represents the control parameter and is updated according to equation (2) during the iteration.
Wherein v isidRepresents the flight velocity of the id particle, ω represents the inertial weight coefficient, c1And c2Denotes an acceleration factor, r is [0,1 ]]Uniformly distributed random numbers, pidRepresenting the position of the id particle, pbestRepresents the optimal position through which the id-th particle passes, gbestRepresenting the optimal position currently traversed by all particles.
The right side of the formula (2) is composed of three parts, wherein the first part is the speed before particle updating, has randomness, is beneficial to expanding a search space and exploring a new search area, and therefore has global search capability; the second part belongs to a self-cognition part and represents the thought of the particle; the third part belongs to the social cognition part and represents cooperation and information sharing among particles. The two extreme values can guide the particle position to quickly converge on the currently searched optimal region, and then the region is locally searched so as to obtain the optimal solution.
The layering is to divide the PI parameter to be optimized into an inner ring and an outer ring, wherein the inner ring control parameter layer is optimized first, and then the outer ring control parameter layer is optimized.
In step 3, the algorithm parameters comprise the number of particles, an inertia weight coefficient, an acceleration factor, the maximum iteration times, the capacity of an external memory, the number of grids, a grid expansion coefficient and the like; the particle information comprises dimension, speed, position and motion range thereof, adaptive value and the like.
In step 4, the leader particle selection method based on the crowdedness is a basic method of MOPSO, and the crowdedness of each grid is calculated firstly, a certain grid is selected by using a roulette method, and then a particle is randomly selected from the grids to serve as the leader particle. Assume the number of particles in each grid, giI represents the mesh number, and the probability of the mesh being selected is p 1/(g)iβ), i.e., the more crowded the particle, the lower the probability of selection.
The leader particle selection method based on the membership degree calculates the membership degree of an adaptive value of each non-dominated solution in an external memory to serve as an evaluation index for leader particle selection. In order to simplify the analysis and represent the analysis, a simple linear function is adopted as a membership function of the adaptive value, and the method specifically comprises the following steps:
firstly, find out the maximum and minimum values of each dimension, and mark as fimaxAnd fiminWherein i represents the ith dimension;
then, the adaptive value formed by the three-dimensional ITAE index is fuzzified,
wherein,an adapted value representing the ith non-dominated solution dimension,and the adaptive value is the corresponding adaptive value after fuzzification processing.
Through fuzzification processing, each dimension adaptive value is converted into a numerical value between 0 and 1, the larger numerical value represents that the dimension adaptive value is better, and the smaller numerical value represents that the dimension adaptive value is worse.
Finally, calculating membership function value L of each particlekFor simplicity of analysis, the fitness values of all dimensions are considered to be equally important, namely, the membership function value can be calculated by the formula (4), and one of the particles is selected as a leader particle by a roulette method.
Where n is the number of non-dominant solutions in the external memory.
The method combining the method based on the crowding degree and the method based on the membership function is set to select the leader particles by adopting the method based on the crowding degree in the first half of the iteration period of the algorithm, and the leader particles are selected by adopting the method based on the membership degree in the second half of the iteration period, so that the diversity and the global search capability of the particles are kept in the first half of the iteration period of the algorithm, and the rapid convergence is realized in the second half of the iteration period.
The optimized inner ring parameters are specifically as follows: and in each iteration process, the leader particle is selected twice, the particle speed and the position are updated twice, the adaptive values of the two times are calculated, and after the first update, the outer ring parameters of the leader particle are assigned to all the particles to try to optimize the inner ring parameters with better performance.
The mutation is to mutate the non-dominant solution in the external memory, and the position information x of the kth non-dominant solutionkThe following variation method was used:
first, a variation rate p is calculated according to the formula (5),
p＝(1-(j-1)/(Nloop-1))(1/m)(5)
in the formula: n is a radical ofloopM is the coefficient of variation for the maximum number of iterations.
Then, the variation interval is calculated, the interval is
[min(Vmin,xk-Δx),max(Vmax,xk+Δx)](6)
In the formula: vmaxAnd VminΔ x is calculated from equation (7) for the maximum and minimum values of the parameter optimization space.
Δx＝p×(Vmax-Vmin) (7)
Finally, the variation result X is calculated according to the formula (8)k。
Xk＝unifrnd(min(Vmin,xk-Δx),max(Vmax,xk+Δx)) (8)
The step of updating the external memory means that an adaptive value is calculated by using a result after mutation, and if the obtained adaptive value dominates the adaptive value before mutation, the original non-dominated solution in the memory is replaced by the mutation result, so that mutation operation is completed.
In step 5, the optimizing the outer ring parameters specifically comprises: and after the leader particle is selected for the second time and the particle speed and the particle position are updated, assigning the inner ring parameters of the leader particle to all the particles, and trying to optimize the outer ring parameters with better performance.
Advantageous effects
The invention provides an MMC-HVDC system control parameter optimization method, which utilizes variation operation to increase the diversity of particles and improve the global search capability while retaining the characteristics of a basic multi-target particle swarm algorithm; and the convergence of the algorithm is improved by using a leader particle selection method based on a membership function. The method has the advantages of good convergence, better dynamic performance of the obtained parameters, suitability for optimizing the control parameters of the MMC-HVDC system and high degree of combination with actual engineering.
Drawings
FIG. 1 is a diagram of a single-ended 101 level MMC-HVDC system architecture;
FIG. 2 is a flow chart of an improved MOPSO algorithm;
FIG. 3 is a graph of the response of the optimal parameter.
Detailed Description
The invention will be further elucidated with reference to the drawings and the specific examples, without however being limited to the examples described below.
The invention provides an MMC-HVDC system control parameter optimization method, which jointly calls PSCAD and MATLAB to optimize MMC-HVDC system control parameters.
Step 1, a single-ended 101-level MMC-HVDC system is set up in PSCAD to optimize PI parameters of a control system of the system, as shown in figure 1, a half-bridge detailed equivalent model and a nearest level approximation modulation method are adopted, and a constant active power and constant reactive power operation mode is adopted. The operating parameters are as follows: the active power set value is 100MW, the reactive power set value is 30Mvar, the number of each bridge arm submodule is 100, the capacitance value of the half-bridge submodule is 0.03F, the reactance value of the bridge arm is 0.007H, and the simulation time is 2 seconds. The control parameters to be optimized are 4 sets of PI parameters of an inner ring and an outer ring of a vector control strategy, and are divided into two layers of the inner ring and the outer ring. The adaptive value is an integral index of the error absolute value of the active power and the reactive power multiplied by time of the control target of the converter, and the value is calculated by PSCAD.
And 2, writing an improved multi-target particle swarm algorithm program in MATLAB, wherein the flow of the multi-target particle swarm algorithm is shown in figure 2.
Step 3, initializing algorithm parameters and particle information, setting the population size to be 50, the capacity of an external memory to be 50, the adaptation value dimension to be 2, the maximum iteration frequency to be 50, the inertia weight coefficient omega to be 0.7, and the acceleration factor c1＝c2The particle velocity, position, adaptation value, etc. are initialized, and the mesh is initialized, 1.5. Let the iteration number j equal to 1.
And 4, selecting leader particles by using a method based on the crowding degree in the first 25 times of iteration process, selecting leader particles by using a method based on a membership function in the last 25 times of iteration process, updating the speed and the position of the particles, assigning the outer ring parameters of the leader particles to all the particles, further performing simulation calculation on adaptive values, adding non-inferior solutions into an external memory, and finally performing variation on the particles in the external memory and updating the external memory.
And 5, after the optimization of the inner ring parameters is finished, selecting the leader particles again, updating the particle speed and the positions, assigning the inner ring parameters of the leader particles to all the updated particles, carrying out simulation calculation on adaptive values, adding non-inferior solutions into an external memory, and finally carrying out variation on the particles in the external memory and updating the external memory.
And 6, repeating the step 4 and the step 5 until the maximum iteration number is reached, and exiting the program.
In order to verify the effectiveness of the method, the result obtained by the method is compared with the result obtained by the basic MOPSO, the parameters of the two optimization methods are compared in the table 1, the optimization effects of the two methods are compared in the table 2, and the table shows that smaller adaptive values can be obtained by adopting the method, so that the effectiveness of the method is verified.
TABLE 1 comparison of parameters before and after optimization
TABLE 2 comparison of before and after optimization
In order to verify the effect of the optimal solution of the PI parameter obtained by the method, the result obtained by adopting the basic MOPSO and the method is verified by adopting PSCAD simulation respectively. The ac voltage drops to 0.8pu at 4s setting and the response curves for active and reactive power are shown in fig. 3. The result obtained by the method of the invention can ensure that the response obtains smaller overshoot and adjustment time, and the performance is obviously improved.
Claims (10)
1. A method for optimizing control parameters of an MMC-HVDC system, the method comprising:
step 1, building an MMC-HVDC simulation model on PSCAD as a calculation model for parameter optimization to calculate an adaptive value;
step 2, an improved multi-target particle swarm optimization algorithm is adopted on MATLAB, the control parameters of the MMC-HVDC system to be optimized are positions of particles, and the control parameters to be optimized are layered;
step 3, initializing algorithm parameters and particle information, and setting the maximum iteration times, wherein the iteration time j is 1;
step 4, entering a main cycle, wherein j is j +1, combining a method based on congestion degree and a method based on membership function to select leader particles, updating the speed and the position of the particles, optimizing inner-loop parameters, adding a non-dominated solution into an external memory, carrying out variation on the particles in the external memory and updating the external memory;
step 5, selecting leader particles, updating the particle speed and position, optimizing outer ring parameters, adding a non-dominated solution into an external memory, carrying out variation on the particles in the external memory and updating the external memory;
and 6, repeating the step 4 and the step 5 until the maximum iteration number is reached.
2. The MMC-HVDC system control parameter optimization method of claim 1, wherein the control system of the MMC-HVDC simulation model in step 1 adopts a vector control technique in direct current control, comprising inner loop current control and outer loop output control, which converts three-phase alternating current quantity under an ABC coordinate system into direct current quantity under a DQ coordinate system and establishes a mathematical model of MMC, and each of the inner loop current control and the outer loop output control has two sets of PI controllers.
3. The MMC-HVDC system control parameter optimization method of claim 1, wherein the adaptation value in step 1 is calculated by the control objective of the converter by using the integral ITAE index of the absolute value of the error multiplied by time:
<mrow> <mi>f</mi> <mo>=</mo> <msubsup> <mo>&Integral;</mo> <mn>0</mn> <mi>T</mi> </msubsup> <mi>t</mi> <mo>&times;</mo> <mo>|</mo> <msub> <mi>y</mi> <mrow> <mi>r</mi> <mi>e</mi> <mi>f</mi> </mrow> </msub> <mo>-</mo> <mi>y</mi> <mo>|</mo> <mi>d</mi> <mi>t</mi> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </mrow>
in the formula yrefFor the control target reference value, y is the control target actual value, and the upper integration limit T is the dynamic process time.
4. The MMC-HVDC system control parameter optimization method of claim 1, wherein, the improved multi-target particle swarm algorithm in step 2 is based on a multi-target particle swarm algorithm, which adopts an external memory and an adaptive grid mechanism to store a non-dominated solution, from which leader particles are selected to iteratively update the information of the particles; the position of the particle represents the control parameter, and the iterative process is updated according to equation (2):
<mrow> <mfenced open = "{" close = ""> <mtable> <mtr> <mtd> <mrow> <msubsup> <mi>v</mi> <mrow> <mi>i</mi> <mi>d</mi> </mrow> <mi>j</mi> </msubsup> <mo>=</mo> <mi>&omega;</mi> <mo>&times;</mo> <msubsup> <mi>v</mi> <mrow> <mi>i</mi> <mi>d</mi> </mrow> <mrow> <mi>j</mi> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mo>+</mo> <msub> <mi>c</mi> <mn>1</mn> </msub> <mo>&times;</mo> <mi>r</mi> <mo>&times;</mo> <mrow> <mo>(</mo> <mrow> <msubsup> <mi>p</mi> <mrow> <mi>b</mi> <mi>e</mi> <mi>s</mi> <mi>t</mi> </mrow> <mrow> <mi>j</mi> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mo>-</mo> <msubsup> <mi>p</mi> <mrow> <mi>i</mi> <mi>d</mi> </mrow> <mrow> <mi>j</mi> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> </mrow> <mo>)</mo> </mrow> <mo>+</mo> <msub> <mi>c</mi> <mn>2</mn> </msub> <mo>&times;</mo> <mi>r</mi> <mo>&times;</mo> <mrow> <mo>(</mo> <mrow> <msubsup> <mi>g</mi> <mrow> <mi>b</mi> <mi>e</mi> <mi>s</mi> <mi>t</mi> </mrow> <mrow> <mi>j</mi> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mo>-</mo> <msubsup> <mi>p</mi> <mrow> <mi>i</mi> <mi>d</mi> </mrow> <mrow> <mi>j</mi> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> </mrow> <mo>)</mo> </mrow> </mrow> </mtd> </mtr> <mtr> <mtd> <mrow> <msubsup> <mi>p</mi> <mrow> <mi>i</mi> <mi>d</mi> </mrow> <mi>j</mi> </msubsup> <mo>=</mo> <msubsup> <mi>p</mi> <mrow> <mi>i</mi> <mi>d</mi> </mrow> <mrow> <mi>j</mi> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mo>+</mo> <msubsup> <mi>v</mi> <mrow> <mi>i</mi> <mi>d</mi> </mrow> <mi>j</mi> </msubsup> </mrow> </mtd> </mtr> </mtable> </mfenced> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </mrow>
wherein v isidRepresents the flight velocity of the id particle, ω represents the inertial weight coefficient, c1And c2Denotes an acceleration factor, r is [0,1 ]]Uniformly distributed random numbers, pidRepresenting the position of the id particle, pbestRepresents the optimal position through which the id-th particle passes, gbestRepresenting the optimal position currently traversed by all particles.
5. The MMC-HVDC system control parameter optimization method of claim 1, wherein layering in step 2 means that PI parameters to be optimized are divided into an inner loop layer and an outer loop layer, and the inner loop control parameter layer is optimized first, and then the outer loop control parameter layer is optimized.
6. The MMC-HVDC system control parameter optimization method of claim 1, wherein the algorithm parameters in step 3 include particle number, inertia weight coefficient, acceleration factor, maximum iteration number, external memory capacity, grid number, grid expansion coefficient; the particle information comprises dimension, speed, position, motion range and adaptive value.
7. The MMC-HVDC system control parameter optimization method of claim 1, wherein the congestion degree-based method in step 4 is to first calculate the congestion distance of each gridSelecting a certain grid by using a roulette method, and randomly selecting a particle from the grid as a leader particle; let the number of particles in each grid be giI represents the mesh number, and the probability of the mesh being selected is p 1/(g)i{ circumflex over) } β), i.e., the more crowded the particle, the lower the probability of selection;
the method based on the membership degree is characterized in that the membership degree of an adaptive value of each non-dominated solution in an external memory is calculated and is used as an evaluation index for leader particle selection; a simple linear function is adopted as a membership function of the adaptive value, and the method specifically comprises the following steps:
first, find the maximum value f of each dimension's fitness valueimaxAnd minimum value fiminWherein i represents the ith dimension;
then, fuzzification processing is carried out on the adaptive value formed by the three-dimensional ITAE index according to the formula (3);
<mrow> <msubsup> <mi>l</mi> <mi>i</mi> <mi>k</mi> </msubsup> <mo>=</mo> <mfrac> <mrow> <msub> <mi>f</mi> <mrow> <mi>i</mi> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> <mo>-</mo> <msubsup> <mi>x</mi> <mi>i</mi> <mi>k</mi> </msubsup> </mrow> <mrow> <msub> <mi>f</mi> <mrow> <mi>i</mi> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>f</mi> <mrow> <mi>i</mi> <mi>m</mi> <mi>i</mi> <mi>n</mi> </mrow> </msub> </mrow> </mfrac> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>3</mn> <mo>)</mo> </mrow> </mrow>
wherein,an adapted value representing the ith non-dominated solution dimension,the adaptive value after the corresponding fuzzification processing is obtained; through fuzzification processing, each dimension adaptive value is converted into a numerical value between 0 and 1, the larger the numerical value is, the better the dimensional adaptive value is, and the smaller the numerical value is, the worse the dimensional adaptive value is;
finally, calculating membership function value L of each particlekCalculating a membership function value through a formula (4), and selecting one particle as a leader particle by using a roulette method;
<mrow> <msub> <mi>L</mi> <mi>k</mi> </msub> <mo>=</mo> <mfrac> <mrow> <munderover> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>3</mn> </munderover> <msubsup> <mi>l</mi> <mi>i</mi> <mi>k</mi> </msubsup> </mrow> <mrow> <munderover> <mo>&Sigma;</mo> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>n</mi> </munderover> <munderover> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mn>3</mn> </munderover> <msubsup> <mi>l</mi> <mi>i</mi> <mi>k</mi> </msubsup> </mrow> </mfrac> <mo>-</mo> <mo>-</mo> <mo>-</mo> <mrow> <mo>(</mo> <mn>4</mn> <mo>)</mo> </mrow> </mrow>
wherein n is the number of non-dominant solutions in the external memory;
the process of selecting the leader particle by combining the method based on the crowding degree and the method based on the membership function specifically comprises the following steps:
the leader particle is selected by adopting a method based on the crowding degree in the first half of the iteration period of the algorithm, and the leader particle is selected by adopting a method based on the membership degree in the second half of the iteration period, so that the diversity and the global search capability of the particles are kept in the first half of the iteration period of the algorithm, and the rapid convergence is realized in the second half of the iteration period.
8. The MMC-HVDC system control parameter optimization method of claim 1, wherein the inner loop parameter optimization in step 4 specifically comprises: and in each iteration process, the leader particle is selected twice, the particle speed and the particle position are updated twice, the adaptive values of the two times are calculated, after the first update, the outer ring parameters of the leader particle are assigned to all the particles so as to calculate the adaptive values, and the inner ring parameters with better performance are tried to be optimized.
9. The MMC-HVDC system control parameter optimization method of claim 1, wherein the mutation in step 4 is a mutation to a non-dominant solution in the external memory, and the position information x of the kth non-dominant solutionkThe specific process of mutation is
First, a variation rate p is calculated according to the formula (5),
p＝(1-(j-1)/(Nloop-1))(1/m)(5)
in the formula: n is a radical ofloopM is the coefficient of variation for the maximum number of iterations;
then, the variation interval is calculated, the interval is
[min(Vmin,xk-Δx),max(Vmax,xk+Δx)](6)
In the formula, VmaxAnd VminΔ x is calculated from equation (7) for the maximum and minimum values of the parameter optimization space;
Δx＝p×(Vmax-Vmin) (7)
finally, the variation result X is calculated according to the formula (8)k；
Xk＝unifrnd(min(Vmin,xk-Δx),max(Vmax,xk+Δx)) (8)
The step of updating the external memory means that an adaptive value is calculated by using a result after mutation, and if the adaptive value obtained dominates the adaptive value before mutation, the original non-dominated solution in the memory is replaced by the mutation result, so that mutation operation is completed.
10. The MMC-HVDC system control parameter optimization method of claim 1, wherein the step 5 of optimizing the outer loop parameters specifically comprises: and after the leader particle is selected for the second time and the particle speed and the particle position are updated, assigning the inner ring parameters of the leader particle to all the particles to further calculate adaptive values, and trying to optimize the outer ring parameters with better performance.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710258224.6A CN107168052B (en) | 2017-04-19 | 2017-04-19 | MMC-HVDC system control parameter optimization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710258224.6A CN107168052B (en) | 2017-04-19 | 2017-04-19 | MMC-HVDC system control parameter optimization method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107168052A true CN107168052A (en) | 2017-09-15 |
| CN107168052B CN107168052B (en) | 2020-01-03 |
Family
ID=59813933
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710258224.6A Active CN107168052B (en) | 2017-04-19 | 2017-04-19 | MMC-HVDC system control parameter optimization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107168052B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020162664A1 (en) * | 2019-02-08 | 2020-08-13 | 효성중공업 주식회사 | Sub-module for conversion of direct current power and method for updating program for sub-module |
| CN113919210A (en) * | 2021-09-22 | 2022-01-11 | 国网江苏省电力有限公司电力科学研究院 | PI parameter optimization method and device of VSC-HVDC system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008086824A1 (en) * | 2007-01-18 | 2008-07-24 | Telefonaktiebolaget Lm Ericsson (Publ) | Codec list transfer comprising a dummy codec in a call path including a tfo leg |
| CN102405575A (en) * | 2009-01-27 | 2012-04-04 | Abb研究有限公司 | Controlling a high-voltage direct-current (hvdc) link |
| CN103840695A (en) * | 2014-02-27 | 2014-06-04 | 河海大学 | Identification method for control parameters of photovoltaic grid-connected inverter |
| CN105426954A (en) * | 2015-08-20 | 2016-03-23 | 武汉科技大学 | Particle swarm optimization method based on multi-strategy synergistic function |
| CN106026736A (en) * | 2016-05-13 | 2016-10-12 | 电子科技大学 | Modular multilevel converter hierarchical control method |
| CN106229964A (en) * | 2016-07-22 | 2016-12-14 | 南京工程学院 | A kind of based on the electrical power distribution network fault location method improving binary particle swarm algorithm |
-
2017
- 2017-04-19 CN CN201710258224.6A patent/CN107168052B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008086824A1 (en) * | 2007-01-18 | 2008-07-24 | Telefonaktiebolaget Lm Ericsson (Publ) | Codec list transfer comprising a dummy codec in a call path including a tfo leg |
| CN102405575A (en) * | 2009-01-27 | 2012-04-04 | Abb研究有限公司 | Controlling a high-voltage direct-current (hvdc) link |
| CN103840695A (en) * | 2014-02-27 | 2014-06-04 | 河海大学 | Identification method for control parameters of photovoltaic grid-connected inverter |
| CN105426954A (en) * | 2015-08-20 | 2016-03-23 | 武汉科技大学 | Particle swarm optimization method based on multi-strategy synergistic function |
| CN106026736A (en) * | 2016-05-13 | 2016-10-12 | 电子科技大学 | Modular multilevel converter hierarchical control method |
| CN106229964A (en) * | 2016-07-22 | 2016-12-14 | 南京工程学院 | A kind of based on the electrical power distribution network fault location method improving binary particle swarm algorithm |
Non-Patent Citations (3)
| Title |
|---|
| 凌海风，等: "《装备保障智能优化决策方法与应用》", 31 May 2015, 北京：国防工业出版社 * |
| 施展，等: "基于QPSO和拥挤距离排序的多目标量子粒子群优化算法", 《控制与决策》 * |
| 洪国巍: "模块化多电平换流器控制策略改进和控制参数优化", 《万方数据HTTP://D.WANFANGDATA.COM.CN/THESIS/Y3115020》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020162664A1 (en) * | 2019-02-08 | 2020-08-13 | 효성중공업 주식회사 | Sub-module for conversion of direct current power and method for updating program for sub-module |
| US11909328B2 (en) | 2019-02-08 | 2024-02-20 | Hyosung Heavy Industries Corporation | Submodule for conversion of direct current power and method for updating program for submodule |
| CN113919210A (en) * | 2021-09-22 | 2022-01-11 | 国网江苏省电力有限公司电力科学研究院 | PI parameter optimization method and device of VSC-HVDC system |
| CN113919210B (en) * | 2021-09-22 | 2024-07-02 | 国网江苏省电力有限公司电力科学研究院 | PI parameter optimization method and device of VSC-HVDC system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107168052B (en) | 2020-01-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109945881B (en) | Mobile robot path planning method based on ant colony algorithm | |
| CN110347151B (en) | Robot path planning method fused with Bezier optimization genetic algorithm | |
| CN115333143A (en) | Deep learning multi-agent micro-grid cooperative control method based on double neural networks | |
| CN103955864B (en) | Based on the electric system multiple target differentiation planing method for improving harmonic search algorithm | |
| Lu et al. | The PID Controller Based on the Artificial Neural Network and the Differential Evolution Algorithm. | |
| CN110888317A (en) | PID controller parameter intelligent optimization method | |
| Yu et al. | Stochastic optimal generation command dispatch based on improved hierarchical reinforcement learning approach | |
| CN111523749B (en) | Intelligent identification method for hydroelectric generating set model | |
| Li et al. | A hybrid assembly sequence planning approach based on discrete particle swarm optimization and evolutionary direction operation | |
| Sayed et al. | Gain tuning PI controllers for boiler turbine unit using a new hybrid jump PSO | |
| CN112636368B (en) | Automatic power generation control method for multi-source multi-region interconnected power system | |
| CN108413963A (en) | Bar-type machine people's paths planning method based on self study ant group algorithm | |
| Debnath et al. | Particle swarm optimization based adaptive strategy for tuning of fuzzy logic controller | |
| CN112966445B (en) | Reservoir flood control optimal scheduling method based on reinforcement learning model FQI | |
| CN107168052B (en) | MMC-HVDC system control parameter optimization method | |
| Liao et al. | AGV path planning model based on reinforcement learning | |
| CN116169698A (en) | Distributed energy storage optimal configuration method and system for stable new energy consumption | |
| CN116231679A (en) | Self-adaptive virtual synchronous machine control method based on deep reinforcement learning | |
| CN116706917A (en) | Intelligent park collaborative regulation and control method and system based on rapid alternating direction multiplier method | |
| CN117419739B (en) | Path planning optimization method for coal conveying system inspection robot | |
| Mai-Phuong et al. | Balancing a practical inverted pendulum model employing novel meta-heuristic optimization-based fuzzy logic controllers | |
| Huang et al. | Mobile robot path planning based on improved genetic algorithm | |
| Xiangde et al. | Global dynamic path planning algorithm based on harmony search algorithm and artificial potential field method | |
| Zuo et al. | Fast nonlinear model predictive control parallel design using QPSO and its applications on trajectory tracking of autonomous vehicles | |
| Sun | Path Planning of Mobile Robot Based on Improved Ant Colony Algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |