CN109255163A - Circuit structure optimization method based on tracking coding and genetic algorithm - Google Patents
Circuit structure optimization method based on tracking coding and genetic algorithm Download PDFInfo
- Publication number
- CN109255163A CN109255163A CN201810968410.3A CN201810968410A CN109255163A CN 109255163 A CN109255163 A CN 109255163A CN 201810968410 A CN201810968410 A CN 201810968410A CN 109255163 A CN109255163 A CN 109255163A
- Authority
- CN
- China
- Prior art keywords
- circuit
- population
- coding
- genetic algorithm
- fitness
- 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
- 238000005457 optimization Methods 0.000 title claims abstract description 41
- 230000002068 genetic effect Effects 0.000 title claims abstract description 27
- 238000000034 method Methods 0.000 title claims abstract description 27
- 210000000349 chromosome Anatomy 0.000 claims abstract description 46
- 230000035772 mutation Effects 0.000 claims abstract description 22
- 238000012546 transfer Methods 0.000 claims description 21
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 238000013461 design Methods 0.000 abstract description 8
- 238000009795 derivation Methods 0.000 description 15
- 230000006870 function Effects 0.000 description 8
- 108090000623 proteins and genes Proteins 0.000 description 7
- 230000006978 adaptation Effects 0.000 description 5
- 239000004065 semiconductor Substances 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000010429 evolutionary process Effects 0.000 description 1
- 238000002986 genetic algorithm method Methods 0.000 description 1
- 238000010353 genetic engineering Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000013011 mating Effects 0.000 description 1
- 238000011017 operating method Methods 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Genetics & Genomics (AREA)
- Data Mining & Analysis (AREA)
- Physiology (AREA)
- Geometry (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Computer Hardware Design (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses the circuit structure optimization methods based on tracking coding and genetic algorithm the following steps are included: coding is tracked to device each in circuit and its parameter, to obtain population primary;All chromosomes in population primary are successively selected, are intersected and mutation operation, to obtain next-generation population;Emulation is decoded to next-generation population, to obtain the fitness of each chromosome, whether the fitness of whole chromosomes is all satisfied the fitness function of setting requirements, if so, exporting the population to be otherwise back to previous step as circuit optimization structure.Compared to traditional technology, step of the present invention is succinct, and design rationally, is easy to understand each device and its parameter in circuit, can carry out accurate optimization to circuit, is conducive to improve optimization accurate rate.
Description
Technical field
The present invention relates to circuit optimization fields, are based especially on the circuit structure optimization side of tracking coding and genetic algorithm
Method.
Background technique
With the continuous improvement of integrated circuit fabrication process level, the scale and complexity of circuit increasingly increase, and are based on
The circuit design method of Heuristics and design rule has more seemed painstaking.And evolving for using for reference from genetic algorithm is hard
Part invention, for we show a kind of completely new evolutionary circuit design methods, it is interior using evolutionary computation technique configuration circuit
Portion's structure is to obtain expected circuit function.
Currently, the circuit structure for this respect optimizes, it generally only considered and change shadow brought by genetic algorithm itself
It rings, is not clear how but to set circuit to be evolved, i.e., can not recognize the integrated circuit feature of circuit to be evolved, exist in this way
Also just careful specific device and its parameter into circuit can not be directed to when evolving based on genetic algorithm, it is clear that the knot obtained in this way
Fruit be it is inaccurate, hardly consistent with the target of engineer.
Summary of the invention
To solve the above-mentioned problems, the object of the present invention is to provide the circuit structure based on tracking coding and genetic algorithm is excellent
Change method is easy to understand each device and its parameter in circuit, in order to carry out the derivation of genetic algorithm respectively to it, favorably
Optimize accurate rate in improving.
In order to make up for the deficiencies of the prior art, the technical solution adopted by the present invention is that:
Circuit structure optimization method based on tracking coding and genetic algorithm, comprising the following steps:
S1, coding is tracked to device each in circuit and its parameter, to obtain population primary;
S2, all chromosomes in population primary are successively selected, are intersected and mutation operation, to obtain next-generation kind
Group;
S3, emulation is decoded to next-generation population, to obtain the fitness of each chromosome;Compare whole chromosomes
Fitness whether be all satisfied the fitness function of setting requirements, it is no if so, export the population using as circuit optimization structure
Then it is back to step S2.
Further, in the step S1, in circuit device and its parameter be tracked coding, to obtain primary kind
Group, comprising:
S11, setting device collection, circuit connection instruction set and device manifold simultaneously create global transfer point, the global shifting
Dynamic point is initialized as input terminal;
S12, it is concentrated at random from device and takes out device and take out instruction from circuit connection instruction set, combined the device and refer to
It enables, so that global transfer point be made to be changed according to instruction, device is added among circuit;
Whether device number meets the setting requirements of device manifold in S13, decision circuitry, if so, the overall situation is mobile
Point is connected to output end, to obtain population primary, is otherwise back to step S12.
Further, in the step S3, emulation is decoded to next-generation population, to obtain the adaptation of each chromosome
Degree, comprising: next-generation population is decoded, decoding gained information is written in netlist file, and call Hspice pairs
All chromosomes in next-generation population are emulated, to obtain the fitness of all chromosomes.
Further, the selection operation in the step S2 includes one in wheel disc selection, algorithm of tournament selection and sequencing selection
Kind is a variety of.
Further, the crossover operation in the step S2 includes single point crossing, two-point crossover, arithmetic crossover, linear crossing
One of operation and intersection based on direction are a variety of.
Further, the mutation operation in the step S2 includes uniform variation, Gaussian mutation, dynamic variation and is based on direction
One of variation or a variety of.
The beneficial effects of the present invention are: the initial device and its parameter of circuit to be evolved are obtained using tracking coding method,
Population primary is obtained, so as to execute the derivation of genetic algorithm for the device in each circuit respectively, that is, is got down
Generation population, since derivation range is big, there is no omission situations, therefore can reach optimization purpose as soon as possible, improve optimization essence
True rate;Also, finally by the fitness for judging each of which chromosome (i.e. each device) and artificial required fitness letter
Whether number matches, and to decide whether to re-execute the derivation of genetic algorithm, until meeting the requirements, then obtains required optimization circuit
Structure is circularly set guarantees that circuit among continuing to optimize, also can be improved optimization accurate rate in this way.Therefore, this hair
Bright step is succinct, and design rationally, is easy to understand each device and its parameter in circuit, can carry out accurate optimization to circuit, have
Optimize accurate rate conducive to improving.
Detailed description of the invention
Present pre-ferred embodiments are provided, with reference to the accompanying drawing with the embodiment that the present invention will be described in detail.
Fig. 1 is step flow chart of the invention;
Fig. 2 is the structural schematic diagram of circuit to be evolved of the invention;
Fig. 3 is the schematic illustration that MTN is operated in the present invention;
Fig. 4 is the encoding gene schematic diagram of device in the present invention;
Fig. 5 is simulation result diagram of the invention.
Specific embodiment
- Fig. 5 referring to Fig.1, the circuit structure optimization method of the invention based on tracking coding and genetic algorithm, including it is following
Step:
S1, coding is tracked to device each in circuit and its parameter, to obtain population primary;
S2, all chromosomes in population primary are successively selected, are intersected and mutation operation, to obtain next-generation kind
Group;
S3, emulation is decoded to next-generation population, to obtain the fitness of each chromosome;Compare whole chromosomes
Fitness whether be all satisfied the fitness function of setting requirements, it is no if so, export the population using as circuit optimization structure
Then it is back to step S2.
Specifically, the initial device and its parameter of circuit to be evolved are obtained using tracking coding method, that is, obtains primary kind
Group, so as to execute the derivation of genetic algorithm for the device in each circuit respectively, that is, gets next-generation population, due to
Derivation range is big, and there is no omission situations, therefore can reach optimization purpose as soon as possible, improves optimization accurate rate;Also, it is last
By judging whether the fitness of each of which chromosome (i.e. each device) matches with artificial required fitness function, to determine
The fixed derivation for whether re-executing genetic algorithm then obtains required optimization circuit structure until meeting the requirements, and circulation is set in this way
It sets and guarantees that circuit among continuing to optimize, also can be improved optimization accurate rate.Therefore, step of the present invention is succinct, design
Rationally, it is easy to understand each device and its parameter in circuit, accurate optimization can be carried out to circuit, be conducive to improve optimization accurately
Rate.
Further, in the step S1, in circuit device and its parameter be tracked coding, to obtain primary kind
Group, comprising:
S11, setting device collection, circuit connection instruction set and device manifold simultaneously create global transfer point, the global shifting
Dynamic point is initialized as input terminal;
S12, it is concentrated at random from device and takes out device and take out instruction from circuit connection instruction set, combined the device and refer to
It enables, so that global transfer point be made to be changed according to instruction, device is added among circuit;
Whether device number meets the setting requirements of device manifold in S13, decision circuitry, if so, the overall situation is mobile
Point is connected to output end, to obtain population primary, is otherwise back to step S12, it specifically can refer to Fig. 2, start node in figure
End then is initialized for global transfer point, end node is then output end.
Further, in the step S3, emulation is decoded to next-generation population, to obtain the adaptation of each chromosome
Degree, comprising: next-generation population is decoded, decoding gained information is written in netlist file, and call Hspice pairs
All chromosomes in next-generation population are emulated, to obtain the fitness of all chromosomes;Since Hspice is ability
Software commonly used by domain, therefore just repeat no more.
Further, the selection operation in the step S2 includes one in wheel disc selection, algorithm of tournament selection and sequencing selection
Kind is a variety of.
Further, the crossover operation in the step S2 includes single point crossing, two-point crossover, arithmetic crossover, linear crossing
One of operation and intersection based on direction are a variety of.
Further, the mutation operation in the step S2 includes uniform variation, Gaussian mutation, dynamic variation and is based on direction
One of variation or a variety of.
The following detailed description of the related content of the present invention:
Genetic algorithm originate from the 1960s, earliest by Michigan university of the U.S. professor Holland propose, it
It is a kind of a kind of novel Optimizing Search algorithm simulating biological heredity and evolutionary process in the natural environment and being formed, Zhi Houtong
The joint efforts for crossing numerous scholars continuously improve the theoretical invention of genetic algorithm, are widely applied to the mankind
The every field of development in science and technology.
The present invention to the effect that carries out cmos circuit topological structure and parameter based on genetic algorithm and tracking coding
Optimization intersects circuit, makes a variation, select etc. and operated that is, circuit code at the form of chromosome by means of genetic algorithm,
To obtain the circuit that performance indicator is met the requirements.
Relational language is described as follows:
Gene (gene): i.e. the information of device refers to type, nodal information and the parameter information of device in the present invention
Deng;
Chromosome (chromesome): including multiple genes, i.e., each chromosome contains whole devices an of circuit
Information;
Population (population): comprising all chromosome, that is, different circuits is contained;
In generation (generation): the population number in per generation, chromosome all may be different;
Fitness (fitness): refer to chromosome to the adaptation situation of environment, fitness here can pass through Hspice's
Performance parameter in output file .list file obtains after carrying out adaptation conversion;
Fitness function (fitness function): the mapping relations of chromosome fitness value are calculated, here if not
So value fitness value for being chromosome if needing the performance parameter to .lis file acquisition to convert;
Chromosome selection operation: selecting the population after initialization, such as wheel disc method, tournament method and ranking method
Deng being state of the art, just repeat no more;
Chiasma operation: crossover operation is carried out to the chromosome chosen, i.e., " mating between chromosome ", such as
Single point crossing, two-point crossover, arithmetic crossover, linear crossing and intersection based on direction etc., are state of the art, just
It repeats no more;
The mutation operation of chromosome: mutation operation is carried out to the chromosome that intersection obtains, i.e., " variation of gene ", variation can
It is state of the art, just no longer to be uniform crossover, Gaussian mutation, dynamic variation and variation based on direction etc.
It repeats;
It is circuit evolving process code into chromosome that tracking, which encodes the main thought of this coding mode,.In cataloged procedure
In, there is a global transfer point, global transfer point changes with the variation of instruction, and the device that each instruction cooperation is selected can
It is added into a new circuit devcie, the device number contained by the circuit then terminates when meeting the requirements.Global transfer point is initialized as 0
It is initialized as input terminal, is connected to the output after the completion of circuit evolving, after then being decoded according to coding rule to chromosome,
The connecting node of each device can be obtained.
Secondly, the circuit connection instruction set of circuit code process includes: MTN, CTP, CTV, CTG, CTO in the present invention,
CTB。
MTN (move-to-new): referring to Fig. 3, a new device is added, input terminal is global transfer point point, output
End is (global transfer point+1), that is, is inserted into a device, and global transfer point is moved backward one.
CTP (cast-to-previous): one device of addition, input terminal are global transfer point point, and output end is
(step-length of global transfer point);
CTV (cast-to-Vdd): a node is connected to VDD/VCC;
CTG (cast-to-Ground): a node is grounded;
CTO (cast-to-Output): a node is connected to the output end of entire circuit;
CTB (cast-to-Bias): a node is connect bias current.
Referring to Fig. 4, the meaning that each section of gene is as follows:
Ins:instruction is instructed, and Ins is some instruction selected at random in circuit connection instruction set;
Type: i.e. part category.According to pre-set device collection, it may be possible to the both ends such as capacitor, inductance, resistance mouthpart
Part, it is also possible to this four port devices of metal-oxide-semiconductor.
Node1-Node4: when part category is two end device, it is only necessary to which two nodal informations are set as Node1
Input, Node2 are set as exporting, and Node3, Node4 are set as sky;
When part category four port devices this for metal-oxide-semiconductor, node1, node2, node3 and node4 respectively represent leakage
Pole, grid, source electrode and substrate.It drains NMOS, default sources extremely incoming (device input) for outcoming (device
Part output end);It drains PMOS, default gate incoming for outcoming.Incoming, outcoming are not
It is fixed, it can be changed by genetic manipulation (such as variation).For the non-incoming and outcoming node of metal-oxide-semiconductor, adopt
Its connection type can be confirmed with CTP, CTV CTG, CTO, CTB operational order.
Para1 and Para2 expression parameter information, for two end device such as resistance, Para1 represents resistance sizes, Para2
It is set as empty;For metal-oxide-semiconductor, Para1 represents ditch road length, and Para2 represents ditch road width.Parameter needs are given birth at random in given range
At each parameter has maximum value max, minimum value min, then parameter value are as follows:
parai=(max-min) * λ
Wherein, the random number that λ is 0 to 1.
In an encoding process, a device is increased using an instruction every time, it can be complete when device number is met the requirements
A complete chromosome is obtained at coding.
Wherein a kind of specific implementation of selection operation, crossover operation and mutation operation is as follows:
(1) selection operation: being operated using algorithm of tournament selection, is taken out certain amount individual from population every time, is then selected
Wherein best one enters progeny population, the operation is repeated, until new population scale reaches original population scale.Specifically
Operating procedure it is as follows:
It determines the individual amount (indicating to account for the percentage of individual number in population) selected every time, is typically chosen 2;
Several individuals (each selected probability of individual is identical) are randomly choosed from population to constitute group, according to each individual
Fitness value, selection wherein the best individual of fitness value enter progeny population;
Second step is repeated, until obtained individual constitutes population of new generation.
(2) crossover operation: selection single point crossing randomly chooses two individuals in parent population, determines crosspoint at random, from
It respectively takes one section of gene to form a new individual in two parent individualities, completes crossover operation.
(3) mutation operation: when carrying out mutation operation, the characteristics of combined circuit, variation is divided into three kinds, is respectively: MOS
Variation, the variation of CTP instruction step-length and the variation of parameter of pipe Inport And Outport Node.It is specific because are as follows:
The variation of metal-oxide-semiconductor Inport And Outport Node: for four port devices, because being selected there are four port in initialization
Two ports as input/output port, but input/output port be it is randomly selected, do not ensure that its correctness, institute
Input/output port is adjusted to need mutation operation.
The variation of CTP instruction step-length: the step-length of CTP instruction, which is determined, adds the device between which two node device
The node of addition is different, and obtained circuit performance is also different, so needing mutation operation to adjust the step-length of CTP instruction.
The variation of parameter: the variation of parameter uses random number mutation operator, assigns one newly to parameter by the operation of initialization
Random value.
Finally, it is preferable that referring to Fig. 5, the fit_value in figure refers to that fitness, genenration refer to derivative generation
Quantity, abbreviation algebra;In the present invention, effect of optimization is verified using five pipe operational amplifiers, using optimization of the present invention
One five pipe operational amplifier, using gain as fitness value, fitness value is continued to optimize with the increase of algebra.With it is artificial
Five pipe amplifiers of design are compared, and the amplifier gain of engineer is 48.5, and with five pipe amplifiers of optimization of the present invention, 280
In generation, has nearby run out of 54.2 gain, it can be seen that gain is optimized, i.e., effect of optimization is obvious.
Circuit structure based on tracking coding and genetic algorithm optimizes device, comprising:
Coding module, for being tracked coding to device each in circuit and its parameter, to obtain population primary;
Derivation module, for successively being selected all chromosomes in population primary, being intersected and mutation operation, to obtain
Obtain next-generation population;
Identification module, for being decoded emulation to next-generation population, to obtain the fitness of each chromosome;Compare
Whether the fitness of whole chromosomes is all satisfied the fitness function of setting requirements, if so, exporting the population using as circuit
Optimize structure, is otherwise back to derivation module.
Specifically, the initial device and its parameter of circuit to be evolved are obtained using tracking coding method, that is, obtains primary kind
Group, so as to execute the derivation of genetic algorithm for the device in each circuit respectively, that is, gets next-generation population, due to
Derivation range is big, and there is no omission situations, therefore can reach optimization purpose as soon as possible, improves optimization accurate rate;Also, it is last
By judging whether the fitness of each of which chromosome (i.e. each device) matches with artificial required fitness function, to determine
The fixed derivation for whether re-executing genetic algorithm then obtains required optimization circuit structure until meeting the requirements, and circulation is set in this way
It sets and guarantees that circuit among continuing to optimize, also can be improved optimization accurate rate.Therefore, step of the present invention is succinct, design
Rationally, it is easy to understand each device and its parameter in circuit, accurate optimization can be carried out to circuit, be conducive to improve optimization accurately
Rate.
Further, the coding module, comprising:
Grouping module, for device collection, circuit connection instruction set and device manifold to be arranged and creates global transfer point, institute
It states global transfer point and is initialized as input terminal;
Adding module instructs, combination for concentrating to take out device and take out from circuit connection instruction set from device at random
Device is added among circuit by the device and instruction so that global transfer point be made to be changed according to instruction;
Whether judgment module meets the setting requirements of device manifold for device number in decision circuitry, if so, will
Global transfer point is connected to output end, to obtain population primary, is otherwise back to adding module.
Further, the identification module is decoded emulation to next-generation population, to obtain the adaptation of each chromosome
Degree, comprising: next-generation population is decoded, decoding gained information is written in netlist file, and call Hspice pairs
All chromosomes in next-generation population are emulated, to obtain the fitness of all chromosomes.
Further, the selection operation in the derivation module includes in wheel disc selection, algorithm of tournament selection and sequencing selection
It is one or more.
Further, the crossover operation in the derivation module includes single point crossing, two-point crossover, arithmetic crossover, linear friendship
One of fork operation and the intersection based on direction are a variety of.
Further, the mutation operation in the derivation module includes uniform variation, Gaussian mutation, dynamic variation and is based on side
To one of variation or a variety of.
Presently preferred embodiments of the present invention and basic principle is discussed in detail in the above content, but the invention is not limited to
Above embodiment, those skilled in the art should be recognized that also have on the premise of without prejudice to spirit of the invention it is various
Equivalent variations and replacement, these equivalent variations and replacement all fall within the protetion scope of the claimed invention.
Claims (6)
1. the circuit structure optimization method based on tracking coding and genetic algorithm, which comprises the following steps:
S1, coding is tracked to device each in circuit and its parameter, to obtain population primary;
S2, all chromosomes in population primary are successively selected, are intersected and mutation operation, to obtain next-generation population;
S3, emulation is decoded to next-generation population, to obtain the fitness of each chromosome;Compare the suitable of whole chromosomes
Whether response is all satisfied the fitness function of setting requirements, if so, exporting the population otherwise to return as circuit optimization structure
It is back to step S2.
2. the circuit structure optimization method according to claim 1 based on tracking coding and genetic algorithm, which is characterized in that
In the step S1, in circuit device and its parameter be tracked coding, to obtain population primary, comprising:
S11, setting device collection, circuit connection instruction set and device manifold simultaneously create global transfer point, the overall situation transfer point
It is initialized as input terminal;
S12, it is concentrated at random from device and takes out device and take out instruction from circuit connection instruction set, combine the device and instruction,
To make global transfer point be changed according to instruction, device is added among circuit;
Whether device number meets the setting requirements of device manifold in S13, decision circuitry, if so, global transfer point is connected
It is connected to output end, to obtain population primary, is otherwise back to step S12.
3. the circuit structure optimization method according to claim 1 based on tracking coding and genetic algorithm, which is characterized in that
In the step S3, emulation is decoded to next-generation population, to obtain the fitness of each chromosome, comprising: to next
It is decoded, decoding gained information is written in netlist file, and call Hspice in next-generation population for population
All chromosomes are emulated, to obtain the fitness of all chromosomes.
4. the circuit structure optimization method according to claim 1 based on tracking coding and genetic algorithm, which is characterized in that
Selection operation in the step S2 includes one of wheel disc selection, algorithm of tournament selection and sequencing selection or a variety of.
5. the circuit structure optimization method according to claim 1 based on tracking coding and genetic algorithm, which is characterized in that
Crossover operation in the step S2 include single point crossing, two-point crossover, arithmetic crossover, linear crossing operation and based on direction
One of intersection is a variety of.
6. the circuit structure optimization method according to claim 1 based on tracking coding and genetic algorithm, which is characterized in that
Mutation operation in the step S2 includes one of uniform variation, Gaussian mutation, dynamic variation and variation based on direction
Or it is a variety of.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810968410.3A CN109255163A (en) | 2018-08-23 | 2018-08-23 | Circuit structure optimization method based on tracking coding and genetic algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810968410.3A CN109255163A (en) | 2018-08-23 | 2018-08-23 | Circuit structure optimization method based on tracking coding and genetic algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109255163A true CN109255163A (en) | 2019-01-22 |
Family
ID=65050407
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810968410.3A Pending CN109255163A (en) | 2018-08-23 | 2018-08-23 | Circuit structure optimization method based on tracking coding and genetic algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109255163A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110032902A (en) * | 2019-03-12 | 2019-07-19 | 佛山市顺德区中山大学研究院 | A kind of reader collision-proof method and its device based on partheno genetic algorithm |
CN110119558A (en) * | 2019-04-30 | 2019-08-13 | 广州大学 | A kind of circuit parameter optimization method based on difference optimization algorithm |
CN110147884A (en) * | 2019-04-02 | 2019-08-20 | 佛山市顺德区中山大学研究院 | A kind of efficient circuits topological structure evolvement method and its system |
CN112182974A (en) * | 2020-10-09 | 2021-01-05 | 中国人民解放军陆军工程大学 | Neuron circuit evolution design method |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105279555A (en) * | 2015-10-28 | 2016-01-27 | 清华大学 | Self-adaptive learning neural network implementation method based on evolutionary algorithm |
-
2018
- 2018-08-23 CN CN201810968410.3A patent/CN109255163A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105279555A (en) * | 2015-10-28 | 2016-01-27 | 清华大学 | Self-adaptive learning neural network implementation method based on evolutionary algorithm |
Non-Patent Citations (4)
Title |
---|
JASON D. LOHN,ET AL: "《Automated Analog Circuit Synthesis Using a Linear Representation》", 《INTERNATIONAL CONFERENCE ON EVOLVABLE SYSTEMS》 * |
李志: "《模拟电路负相关演化方法及设计》", 《万方数据知识服务平台》 * |
李胜旺,等: "《改进遗传算法在数字电路中的应用》", 《信息通信》 * |
许颖: "《基于改进型遗传算法的数字电路优化》", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110032902A (en) * | 2019-03-12 | 2019-07-19 | 佛山市顺德区中山大学研究院 | A kind of reader collision-proof method and its device based on partheno genetic algorithm |
CN110032902B (en) * | 2019-03-12 | 2022-04-15 | 佛山市顺德区中山大学研究院 | Reader anti-collision method and device based on single parent genetic algorithm |
CN110147884A (en) * | 2019-04-02 | 2019-08-20 | 佛山市顺德区中山大学研究院 | A kind of efficient circuits topological structure evolvement method and its system |
CN110119558A (en) * | 2019-04-30 | 2019-08-13 | 广州大学 | A kind of circuit parameter optimization method based on difference optimization algorithm |
CN110119558B (en) * | 2019-04-30 | 2023-11-14 | 广州大学 | Circuit parameter optimization method based on differential optimization algorithm |
CN112182974A (en) * | 2020-10-09 | 2021-01-05 | 中国人民解放军陆军工程大学 | Neuron circuit evolution design method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109255163A (en) | Circuit structure optimization method based on tracking coding and genetic algorithm | |
CN103838820B (en) | Evolutionary multi-objective optimization community detection method based on affinity propagation | |
CN107122843A (en) | A kind of traveling salesman problem method for solving based on improved adaptive GA-IAGA | |
CN111551825B (en) | Self-adaptive power distribution network fault positioning method based on fault current path | |
CN103793467B (en) | Method for optimizing real-time query on big data on basis of hyper-graphs and dynamic programming | |
CN112468401B (en) | Network-on-chip routing communication method for brain-like processor and network-on-chip | |
CN105976048A (en) | Power transmission network extension planning method based on improved artificial bee colony algorithm | |
CN107391549A (en) | News based on artificial intelligence recalls method, apparatus, equipment and storage medium | |
CN103428804A (en) | Method for searching mapping scheme between tasks and nodes of network-on-chip (NoC) and network code position | |
CN109190278A (en) | A kind of sort method of the turbine rotor movable vane piece based on the search of Monte Carlo tree | |
CN105654187A (en) | Grid binary tree method of control system midpoint locating method | |
CN104866903A (en) | Most beautiful path navigation algorithm based on genetic algorithm | |
CN106919504B (en) | Test data evolution generation method based on GA algorithm | |
CN116955846B (en) | Cascade information propagation prediction method integrating theme characteristics and cross attention | |
CN101859315A (en) | Heuristic solving method for maximizing influence of social network | |
CN102768735B (en) | Network community partitioning method based on immune clone multi-objective optimization | |
CN108694439A (en) | A kind of topological construction method of Bayesian network | |
Xu et al. | An improved immune evolutionary algorithm for multimodal function optimization | |
CN104615679A (en) | Multi-agent data mining method based on artificial immunity network | |
CN116485057A (en) | Method for solving path planning by self-adaptive genetic algorithm | |
CN103426127A (en) | Urban power network planning method based on pseudo-crossover taboo hybrid genetic algorithm | |
CN116307328A (en) | Greedy solving method for travel business problem | |
CN112148030B (en) | Underwater glider path planning method based on heuristic algorithm | |
CN106383863A (en) | Isomorphic sub-graph query optimization method | |
CN114564039A (en) | Flight path planning method based on deep Q network and fast search random tree 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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190122 |
|
RJ01 | Rejection of invention patent application after publication |