CN-121980918-A - LCC-MMC hybrid converter topology structure optimization method

CN121980918ACN 121980918 ACN121980918 ACN 121980918ACN-121980918-A

Abstract

The invention discloses a topology structure optimization method of an LCC-MMC hybrid converter. The method comprises the steps of firstly establishing a multidimensional optimization model containing core variables and defining constraint boundaries based on converter topology construction, element parameters and operation conditions, reconstructing four-objective weighted superposition functions such as sub-module volume, total loss and the like, determining dynamic weights through fusion of a hierarchical analysis method and an entropy weight method, then solving optimal parameter combinations by adopting a multi-objective particle swarm algorithm containing adaptive inertia weight and other improvement strategies, and finally verifying scheme results from three aspects through simulation and scale model machine tests. The method realizes collaborative optimization of subjective requirements and objective data by establishing a multidimensional optimization variable model, constructing a weighted superposition optimization function and adopting a method of fusing a analytic hierarchy process and an entropy weight process to determine dynamic weights, thereby remarkably improving the compactness, economy and stability of the converter and providing reliable support for engineering design and operation optimization.

Inventors

LIU FEI
YIN JUN
QI ZHIRONG
PANG YU
ZHAO XUE
QUAN HUIJUAN
ZHANG JUN
PENG FEI
LU WANGSHENG
XIAO QIAN
JIN YU
WANG YUANLING
YU HAOLIN
XU ZHE
Cao Huanxi
LIU LIANTAO
FAN RUIMING
LI JITAI
CHENG WENJUN
TIAN XU
WANG SHIBIN
CHE YANYING

Assignees

国网青海省电力公司经济技术研究院
国网青海省电力公司
天津大学

Dates

Publication Date: 20260505
Application Date: 20251226

Claims (10)

1. The topological structure optimization method of the LCC-MMC hybrid converter is characterized by comprising the following steps of: S1, establishing a multidimensional optimization variable model and defining a voltage constraint boundary and a current constraint boundary based on converter topology, element parameters and operation conditions; s2, constructing a four-item-label weighted optimization function according to a multidimensional optimization variable model, a voltage constraint boundary and a current constraint boundary, and determining dynamic weights by adopting an analytic hierarchy process and an entropy weight process; S3, solving an optimal topological structure parameter combination by improving a multi-objective particle swarm optimization algorithm according to the dynamic weight and combining the topological constraint and the objective function; And S4, verifying the optimal topological structure parameter combination through simulation test and experiment, and evaluating an optimization result from three aspects to provide basis for the engineering design and operation optimization of the converter.
2. The method for optimizing the topology of an LCC-MMC hybrid converter according to claim 1, wherein step S1 comprises: the converter topology comprises an LCC converter bridge, an MMC converter valve bank, a submodule unit, a connecting reactor and a filter device; The element parameters comprise submodule capacitance values, IGBT rated current and voltage, reactor inductance values and converter transformer transformation ratios; The operation working conditions comprise rated transmission power, voltage class, load fluctuation range and fault scene probability; The core variables of the multi-dimensional optimization variable model comprise submodule type selection, submodule serial quantity, MMC bridge arm reactor inductance value, LCC converter bridge trigger angle and submodule capacitance value; The constraint boundaries comprise voltage constraint, current constraint, space constraint, thermal constraint and fault constraint, wherein the voltage constraint is that the voltage fluctuation range of an output line of the converter is less than or equal to +/-5% of rated voltage, the current constraint is that the working current of elements is less than or equal to 1.2 times of rated current, the space constraint is that the total installation volume of sub-modules is less than or equal to a preset threshold, the thermal constraint is that the highest working temperature of the elements is less than or equal to 125 ℃, and the converter still can maintain greater than or equal to 80% of rated power output when the fault constraint is single-module fault.
3. The method for optimizing the topology of an LCC-MMC hybrid converter according to claim 1, wherein step S2 comprises: S21, optimizing objective function The expression of (2) is: Wherein, the In order to comprehensively optimize the target value, As a weight of the volume of the sub-module, As a weight for the total loss of the sub-modules, As a weight for sub-module failure and repair costs, Is the weight of the temperature difference of the submodule and meets the following conditions And 0< <1,0< <1,0< <1,0< <1; In the form of a sub-module volume, As a result of the total loss, In order to be able to do this with respect to failure and maintenance costs, Is the temperature difference; S22, determining subjective weights by using a analytic hierarchy process, namely constructing a hierarchical structure of a target layer and a criterion layer, constructing a judgment matrix by expert scoring, and calculating the subjective weights of the sub-module volumes of all targets Subjective weighting of total loss of submodules Subjective weighting of sub-module failure and repair costs Subjective weighting of submodule temperature differences ; S23, constructing a decision matrix based on m groups of historical operation data or simulation data Standardized processing to obtain standard decision matrix : Wherein, the Is that First, the Line 1 The data of the column is stored, Is that First, the Line 1 The data of the column is stored, Is that First, the The data of the column is stored, As a function of the minimum value of the function, As a function of the maximum value, For the number of rows, Is a column ordinal number; Calculating the entropy of each target information By means of Obtaining objective weight : Wherein, the Is a weight ordinal number; S24, dynamic weight fusion to obtain final weight : Wherein, the Is a weight balance coefficient.
4. A method for optimizing the topology of an LCC-MMC hybrid converter according to claim 3, characterized in that each objective function of step S21 is defined as follows: wherein N is the number of sub-modules, In the case of a single capacitor volume, For a single IGBT module volume, Is a single heat dissipation structure volume; Wherein, the For the conduction loss of the converter, In the event of a switching loss, In order to be a loss of the reactor, Is transformer loss; And The calculation formula is as follows: Wherein, the For the on-resistance of the component, In order for the operating current to be sufficient, For the number of switching times of the IGBT, Single switching loss; Wherein, the In order to be a cost of the failure, Is maintenance cost; Wherein, the For the highest temperature of the inner submodule of the bridge arm, The lowest temperature of the bridge arm inner submodule is obtained.
5. The method for optimizing the topology of an LCC-MMC hybrid converter of claim 3, The value range of (2) is 0.3-0 ≤0.7。
6. The method for optimizing the topology of an LCC-MMC hybrid converter according to claim 1, wherein step S3 comprises: s31, abstracting each group of topological structure parameter combinations into a particle, wherein the dimension of the particle completely corresponds to the quantity of the optimized variables, and randomly generating an initial particle population based on a preset constraint boundary; S32, substituting the topological parameters corresponding to each particle into a weighted superposition optimization objective function to calculate an initial fitness value, presetting constraint conditions, and judging the feasibility of the particles; S33, dynamically adjusting the speed and the position of each particle according to three types of optimal solutions, wherein an individual optimal solution is an adaptive optimal solution found by the particle in historical iteration, a global optimal solution is an optimal solution of the whole particle population in the current iteration, and a neighborhood optimal solution is an optimal solution of other particles in the peripheral range of the particle; S34, setting a double convergence condition, wherein the iteration times reach a preset maximum value, and the adaptation value variation of the global optimal solution is smaller than a threshold value in 10 successive iterations, and stopping iteration when any condition is met; And S35, after iteration is ended, extracting all non-dominant solutions from the elite storage pool, and screening out optimal topological structure parameter combinations considering core requirements and other target performances from the non-dominant solution sets in combination with the priority requirements actually set by engineering.
7. The method for optimizing the topological structure of the LCC-MMC hybrid converter according to claim 1, wherein the simulation test in the step S4 is based on PSCAD to build a simulation model of the LCC-MMC hybrid converter, input optimal topological parameters and simulate three typical working conditions of rated power, abrupt load change and single-module fault; topology rationality, namely, the compactness of the sub-module layout is more than or equal to 0.85, and the uniformity of element stress distribution is less than or equal to 1.2; Compared with the prior optimization, the method has the advantages that the submodule volume reduction rate is more than or equal to 15%, the total loss reduction rate is more than or equal to 10%, the fault and maintenance cost reduction rate is more than or equal to 8%, and the temperature difference reduction rate is more than or equal to 20%; The running stability is that the ripple rate of the direct current is less than or equal to 2 percent, the trigger delay time of the converter is less than or equal to 5ms, and the voltage recovery time under the fault working condition is less than or equal to 0.1s; The test verifies that a 1:10 scale sampler is adopted to test target parameters and stability indexes in actual operation, and if the error of simulation and test results is less than or equal to 5%, optimization is confirmed.
8. An LCC-MMC hybrid converter topology optimization device, characterized by comprising: The optimization variable modeling module is used for establishing a multi-dimensional optimization variable model and defining a voltage constraint boundary and a current constraint boundary based on the converter topology, the element parameters and the operation conditions; The objective function and weight construction module is used for constructing a four-item-scale weighted optimization function according to the multidimensional optimization variable model, the voltage constraint boundary and the current constraint boundary, and determining dynamic weights by adopting a hierarchical analysis method and an entropy weight method; The optimization solving module is used for solving the optimal topological structure parameter combination by improving a multi-objective particle swarm optimization algorithm according to the dynamic weight and combining the topological constraint and the objective function; the result evaluation module is used for verifying the optimal topological structure parameter combination through simulation test and experiment, evaluating the optimization result from three aspects and providing basis for the engineering design and operation optimization of the converter.
9. An electronic device comprising a processor, a communication interface, a memory and a communication bus, characterized in that the processor, when executing a computer program, implements the steps of a method for optimizing the topology of an LCC-MMC hybrid converter according to any of claims 1-7.
10. A non-transitory computer readable storage medium, having stored thereon a computer program, characterized in that the computer program, when executed by a processor, implements the steps of a LCC-MMC hybrid converter topology optimization method according to any of claims 1 to 7.

Description

LCC-MMC hybrid converter topology structure optimization method Technical Field The invention relates to the technical field of power electronics, in particular to a topology structure optimization method of an LCC-MMC hybrid converter. Background Along with the wide application of the hvdc transmission technology in the cross-region energy transmission and the interconnection of power grids, the LCC-MMC hybrid converter becomes one of the core devices of the hvdc transmission system by virtue of the high-capacity and low-cost advantages of the LCC (Line Commutated Converter, grid commutated converter) converter and the low-harmonic and flexible control characteristics of the MMC (Modular Multilevel Converter, modularized multi-level converter) converter. However, the topology design of the existing LCC-MMC hybrid converter still has a plurality of technical pain points, which restrict the economical efficiency and reliability of the engineering application: 1. the topology optimization target is single, and the coupling relation among the volume of the sub-module, the total loss, the fault maintenance cost and the local temperature difference is ignored; 2. The existing method has the defects of lack of rationality in weight distribution, mostly adopts fixed subjective weight or single objective weight, and has poor adaptability in optimization results; 3. the traditional multi-objective optimization has the problems of low convergence speed and the like, and is difficult to quickly find the global optimal topology parameter combination under the complex constraint condition. Therefore, development of a topological structure optimization method of an LCC-MMC hybrid converter with multi-objective collaborative optimization, scientific weight distribution, high algorithm efficiency and sufficient verification is needed, the defects of the prior art are overcome, and the comprehensive performance and engineering applicability of the converter are improved. Disclosure of Invention The present invention is directed to solving at least one of the technical problems existing in the related art. Therefore, the invention provides an LCC-MMC hybrid converter topology structure optimization method, which verifies the validity of an optimization result from multiple dimensions and provides a basis for the engineering design and operation optimization of a converter. The invention provides a topology structure optimization method of an LCC-MMC hybrid converter, which comprises the following steps: S1, establishing a multidimensional optimization variable model and defining a voltage constraint boundary and a current constraint boundary based on converter topology, element parameters and operation conditions; s2, constructing a four-item-label weighted optimization function according to a multidimensional optimization variable model, a voltage constraint boundary and a current constraint boundary, and determining dynamic weights by adopting an analytic hierarchy process and an entropy weight process; S3, solving an optimal topological structure parameter combination by improving a multi-objective particle swarm optimization algorithm according to the dynamic weight and combining the topological constraint and the objective function; And S4, verifying the optimal topological structure parameter combination through simulation test and experiment, and evaluating an optimization result from three aspects to provide basis for the engineering design and operation optimization of the converter. According to the method for optimizing the topological structure of the LCC-MMC hybrid converter, the step S1 comprises the following steps: The LCC-MMC topology structure comprises an LCC converter bridge, an MMC converter valve bank, a submodule unit, a connecting reactor and a filter device, wherein element parameters comprise submodule capacitance values, IGBT rated current and voltage, reactor inductance values and converter transformer transformation ratios, and the operating conditions comprise rated transmission power, voltage class, load fluctuation range and fault scene probability; The core variables of the multi-dimensional optimization variable model comprise submodule type selection, submodule serial quantity, MMC bridge arm reactor inductance value, LCC converter bridge trigger angle and submodule capacitance value; The constraint boundaries comprise voltage constraint, current constraint, space constraint, thermal constraint and fault constraint, wherein the voltage constraint is that the voltage fluctuation range of an output line of the converter is less than or equal to +/-5% of rated voltage, the current constraint is that the working current of elements is less than or equal to 1.2 times of rated current, the space constraint is that the total installation volume of sub-modules is less than or equal to a preset threshold, the thermal constraint is that the highest working temperature of the elements is less than or