CN112505532A - Analog circuit single fault diagnosis method based on improved particle swarm optimization - Google Patents
Analog circuit single fault diagnosis method based on improved particle swarm optimization Download PDFInfo
- Publication number
- CN112505532A CN112505532A CN202011465725.XA CN202011465725A CN112505532A CN 112505532 A CN112505532 A CN 112505532A CN 202011465725 A CN202011465725 A CN 202011465725A CN 112505532 A CN112505532 A CN 112505532A
- Authority
- CN
- China
- Prior art keywords
- fault
- particle
- particles
- analog circuit
- representative
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/28—Testing of electronic circuits, e.g. by signal tracer
- G01R31/316—Testing of analog circuits
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Tests Of Electronic Circuits (AREA)
Abstract
The invention discloses an analog circuit single fault diagnosis method based on an improved particle swarm algorithm, which comprises the steps of firstly analyzing and obtaining a transmission function and fuzzy groups of an analog circuit, selecting a representative fault element for each fuzzy group, measuring and obtaining a fault voltage phasor when the analog circuit has a fault, taking a parameter vector of the fault element as a particle position, introducing cross variation in the particle swarm evolution process, calculating a fitness value, considering the Euclidean distance between an output voltage phasor and the fault voltage phasor and the number of the fault elements as 1, and taking the representative fault element with the parameter value in the global optimal position in the last generation of the group as a fault diagnosis result. The invention utilizes the improved particle swarm algorithm to find out the analog circuit transmission function parameter closest to the fault response, thereby finding out the fault source and improving the accuracy of fault diagnosis by improving the individual evolution process.
Description
Technical Field
The invention belongs to the technical field of analog circuit fault diagnosis, and particularly relates to an analog circuit single fault diagnosis method based on an improved particle swarm algorithm.
Background
With the rapid development of integrated circuits, digital and analog components are integrated on the same chip to improve product performance and reduce chip area and cost. It is reported that although the analog part only occupies 5% of the chip area, the failure diagnosis cost thereof occupies 95% of the total diagnosis cost, and the analog circuit failure diagnosis has been a bottleneck problem in the integrated circuit industry. Currently, in the field of analog circuit fault diagnosis, there are mainly a pre-test simulation (such as a fault dictionary method) and a post-test simulation method. Before-test emulation is to simulate possible faults of a circuit according to a circuit diagram, parameters and the like before test, store fault responses, and measure the fault responses by using the excitation adopted in the process of constructing a dictionary before the fault occurs in the circuit. And then searching the most similar response in the fault dictionary to find the fault source. The advantage of this method is that the fault diagnosis is fast, but the disadvantage is also obvious, namely, when constructing the dictionary, all faults need to be exhausted. In addition, the simulation element parameters are continuously changed, so the space complexity of the exhaustive method is high. In addition, the components of the analog circuit have tolerance characteristics, and many faults are soft faults (component parameter values exceed a tolerance range), and it is difficult to cover all faults by using a fault dictionary, so that the fault diagnosis accuracy is low.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides an analog circuit single fault diagnosis method based on an improved particle swarm algorithm.
In order to achieve the above object, the analog circuit single fault diagnosis method based on the improved particle swarm optimization of the present invention comprises the following steps:
s1: acquiring a transmission function of the analog circuit at a preset measuring point t, recording the number of elements in the analog circuit as C, and recording the nominal value of each element parameter asi is 1,2, …, C. Analyzing to obtain fuzzy groups for fault diagnosis of the output voltage of the analog circuit through the measuring point t, recording the number of the obtained fuzzy groups as N, selecting one element as a representative fault element in each fuzzy group, and recording the number of other non-representative fault elements as M;
s2: when in useWhen the analog circuit fails, the fault voltage phasor at the measuring point t is measured under the preset excitation signal
S3: with X ═ X1,…,xN,xN+1,…,xN+MAs the particle positions in the particle swarm, where the first N xnParameter values representing the nth representative faulty component, N being 1,2, …, N, followed by M xmThe parameter values representing the mth non-representative failed component, M ═ N +1, N +2, …, N + M. Initializing K particles to form an initial particle swarm P, randomly taking values of parameter values of N representative fault elements in each particle position, taking values of parameter values of M non-representative fault elements within a tolerance range, and recording the initial position of each particle asThen initializing each particle velocity
S4: respectively calculating the fitness value of each particle in the initial particle swarm P, wherein the calculation method of the fitness value comprises the following steps: respectively calculating particles according to transfer functionOutput voltage phasor at measurement point tThen calculating the output voltage phasor and the fault voltage phasorEuropean distance ofThen, parameters are determined according to the number of the representative fault elements with parameter values exceeding the tolerance range in the particle positions
Selecting the position of the particle with the minimum fitness value as an initial global optimal position Gbest, and enabling the initial position of each particleAs initial local optimum position Pbestk;
S5: initializing the iteration number i to 1;
s6: determining the inertia weight w of the iteration by the following formulai+1:
wi+1=wstart-(wstart-wend)×(i/I)2
Wherein, wstart、wendRespectively representing a start point value and an end point value of the inertia weight, wherein I represents a preset maximum iteration number;
Wherein, c1、c2Is a learning factor, r1、r2Is random between 0 and 1Counting;
updating the position of the particle by adopting the following formula to obtain the updated position
Then to the updated positionIn which the value of each element parameter is checked for boundaries, i.e. the position of the particlesThe parameter value of the non-representative fault element is limited within a tolerance range;
s8: updating the particle position according to the current iterationCalculating the fitness value of each particle;
s9: sorting the particles updated by the current iteration from small to large according to the fitness value, uniformly dividing the particles into D sections, and determining the value of D according to actual needs; recording the number of the particles in each section as Q, removing Q particles in the D section, selecting Q particles in each section from the 2 nd section to the D-1 th section to remove, and copying the Q particles in the 1 st section into two particle groups for supplementing;
s10: performing intersection and variation operation on the particle position vectors in the particle swarm obtained in the step S9, and ensuring that the parameter values of the non-representative fault elements are limited within a tolerance range;
s12: for each particleJudging whether the current fitness value is smaller than the local optimal position PbestkIf so, the local optimum position is setOtherwise, no operation is performed;
screening out the particles with the minimum fitness value in the particles subjected to the current cross variation operation, if the fitness value of the particles is smaller than the global optimal position Gbest, updating the global optimal position Gbest to the position of the particle, and otherwise, not performing any operation;
s13: judging whether an iteration termination condition is reached, if so, entering a step S14, otherwise, entering a step S15;
s14: returning to step S6 by making i + 1;
s15: and the representative fault element with the parameter value in the current global optimal position Gbest within the fault range is the fault diagnosis result.
The invention relates to an analog circuit single fault diagnosis method based on an improved particle swarm algorithm, which comprises the steps of firstly analyzing and obtaining a transmission function and fuzzy groups of an analog circuit, selecting a representative fault element for each fuzzy group, measuring and obtaining a fault voltage phasor when the analog circuit has a fault, taking a parameter vector of the fault element as a particle position, introducing cross variation in the particle swarm evolution process, considering Euclidean distance between an output voltage phasor and the fault voltage phasor and the number of the fault elements as 1 in calculating a fitness value, and taking the representative fault element with a parameter value in a fault range in a global optimal position in a last generation of group as a fault diagnosis result.
The invention utilizes the improved particle swarm algorithm to find out the analog circuit transmission function parameter closest to the fault response, thereby finding out the fault source and improving the accuracy of fault diagnosis by improving the individual evolution process.
Drawings
FIG. 1 is a flow chart of an embodiment of the single fault diagnosis method of the analog circuit based on the improved particle swarm optimization of the present invention;
FIG. 2 is a circuit diagram of a second order Thomas analog filter circuit in the present embodiment;
FIG. 3 shows R in this example3A global optimal fitness curve graph during fault;
FIG. 4 shows R in the present embodiment4Global optimal fitness curve graph in fault.
Detailed Description
The following description of the embodiments of the present invention is provided in order to better understand the present invention for those skilled in the art with reference to the accompanying drawings. It is to be expressly noted that in the following description, a detailed description of known functions and designs will be omitted when it may obscure the subject matter of the present invention.
To better explain the technical solution of the present invention, first, the technical idea of the present invention is briefly explained.
Assuming that the transfer function of the analog circuit at the preset measuring point is H (j, omega), wherein j represents an imaginary unit, omega represents angular frequency, X represents a component parameter vector, and X is [ X ═ X%1,x2,…,xC],xiThe parameter indicating the ith element, i ═ 1,2, …, C, and C indicate the number of elements in the analog circuit. The transfer function H (j, ω) can then be expressed as:
wherein the content of the first and second substances,which is representative of the phasor of the input voltage,representing the phasor of the output voltage at the measuring point, an,an-1,…,a0、bm,bm-1,…,b0Representing a function with the element parameter values as variables.
If the excitation signal frequency is constant, the transfer function H (j, ω) is determined only by the parameter vector X, so thatTo translate fault diagnosis into an optimization problem. When a circuit has a single fault, i.e. a certain faulty component c exceeds the tolerance rangeWhereinThe nominal value of the element parameter of the fault element c is represented, alpha represents a tolerance coefficient, and the value range of the tolerance coefficient is generally within the range of alpha and the element (0, 0.05)]. Measuring the actual fault voltage phasor of the circuit asThen the fault diagnosis is to find an optimal set of component parameters such that the following equation is taken to be the minimum:
wherein, E represents error, | | | | | represents two norms, namely Euclidean distance;calculated from the transfer function.
Since the present invention is directed to single fault diagnosis, then the fitness function f (x) of an individual can be defined:
f(X)=g(X)+γ(X) (3)
wherein:
finally, the fault diagnosis is converted into the following minimization problem:
minimize f(X)=g(X)+γ(X) (5)
based on the thought, the analog circuit single-fault diagnosis method based on the improved particle swarm optimization is provided. FIG. 1 is a flow chart of an embodiment of the method for diagnosing single fault of an analog circuit based on an improved particle swarm optimization. As shown in fig. 1, the method for diagnosing single fault of analog circuit based on improved particle swarm optimization of the present invention comprises the following specific steps:
s101: acquiring analog circuit information:
acquiring a transmission function of the analog circuit at a preset measuring point t, recording the number of elements in the analog circuit as C, and recording the nominal value of each element parameter asi-1, 2, …, C, determining the tolerance ranges of the respective elementsAnd analyzing to obtain fuzzy groups for fault diagnosis of the output voltage of the analog circuit through the measuring point t, recording the number of the obtained fuzzy groups as N, selecting one element as a representative fault element in each fuzzy group, and recording the number of other non-representative fault elements as M.
S102: determining the current output of the analog circuit:
when the analog circuit has a fault, measuring the fault voltage phasor at the measuring point t under the preset excitation signalIn order to make the output voltage under the fault state more accurate, the fault voltage phasor can be measured for multiple times under the same condition and then averaged, so that the fault voltage phasor can be obtained.
S103: initializing a particle swarm:
with X ═ X1,…,xN,xN+1,…,xN+MAs the particle positions in the particle swarm, where the first N xnParameter values representing the nth representative faulty component, N being 1,2, …, N, followed by M xmThe parameter values representing the mth non-representative failed component, M ═ N +1, N +2, …, N + M. Initializing K particles to form an initial particle swarm P, randomly taking values of parameter values of N representative fault elements in each particle position, taking values of parameter values of M non-representative fault elements within a tolerance range, and recording the initial position of each particle asThen initializing each particle velocityk=1,2,…,K。
S104: initializing an optimal position:
respectively calculating the fitness value of each particle in the initial particle swarm P, wherein the calculation method of the fitness value comprises the following steps: respectively calculating the output voltage phasor of the particle at the measuring point t according to the transfer functionThen calculating the output voltage phasor and the fault voltage phasorEuropean distance ofThen, parameters are determined according to the number of the representative fault elements with parameter values exceeding the tolerance range in the particle positions
In actual calculation, infinity is replaced with an absolute value.
The particle position with the smallest fitness value is selected as the initial global optimal position Gbest,initial position of each particleAs initial local optimum position Pbestk。
S105: the number of initialization iterations i is 1.
S106: updating the inertia weight:
determining the inertia weight w of the iteration by the following formulai+1:
wi+1=wstart-(wstart-wend)×(i/I)2 (8)
Wherein, wstart、wendRespectively representing the start value and the end value of the inertia weight, and I representing the preset maximum iteration number. In this embodiment, set wstart=0.9,wend=0.4。
S107: update the velocity and position of the particle:
Wherein, c1、c2Is a learning factor, r1、r2Is a random number between 0 and 1.
Updating the position of the particle by adopting the following formula to obtain the updated position
Then to the updated positionIn which the value of each element parameter is checked for boundaries, i.e. the position of the particlesThe parameter values of the non-representative faulty components are limited within a tolerance range.
S108: calculating a fitness value:
updating the particle position according to the current iterationThe fitness value of each particle is calculated.
S109: particle swarm optimization:
in order to improve the iteration efficiency, the invention adopts a layered elite retention strategy to select individuals, sorts the particles after current iteration updating from small to large according to the fitness value, and uniformly divides the particles into D sections, wherein the value of D is determined according to actual needs. Obviously, the 1 st segment is the most suitable particle and the D segment is the least suitable particle. And (3) recording the number of the particles in each section as Q, removing Q particles in the D section, selecting a total of Q particles from each section from the 2 nd section to the D-1 th section, removing Q particles in the 1 st section, copying two parts of the Q particles to supplement the particle swarm, and ensuring that the particle swarm scale is unchanged.
S110: cross mutation:
the intersection and variation operations on the particle position vectors in the particle swarm obtained in step S109 need to ensure that the parameter values of the non-representative fault elements are limited within the tolerance range.
S111: calculating a fitness value:
S112: updating the optimal position:
for each particleDetermine its current suitabilityWhether the strain value is less than the local optimum position PbestkIf so, the local optimum position is setOtherwise, no operation is performed.
And screening out the particles with the minimum fitness value in the particles subjected to the current cross variation operation, if the fitness value of the particles is smaller than the global optimal position Gbest, updating the global optimal position Gbest to the position of the particle, and otherwise, performing no operation.
S113: and judging whether an iteration termination condition is reached, namely whether the iteration number I is less than I or whether the global optimal position meets the precision condition e, if so, entering step S114, and if not, entering step S115.
S114: let i be i +1, return to step S106.
S115: determining a fault diagnosis result:
and the representative fault element with the parameter value in the current global optimal position Gbest within the fault range is the fault diagnosis result.
Examples
In order to better illustrate the technical scheme and the technical effect of the invention, a specific analog circuit is adopted to carry out experimental verification on the invention. Fig. 2 is a circuit diagram of a second-order thomas analog filter circuit in the present embodiment. As shown in fig. 2, the second-order thomas analog filter circuit in this embodiment includes 6 resistor elements, 2 capacitors and 3 amplifiers, and the nominal values of the parameters of the elements are as shown in fig. 2. In this example, V isoutAs a measurement point, the transfer function is shown as follows:
according to the symbol analysis method and the transmission function, the fuzzy group of the circuit is as follows: { R1},{R2},{R4,R5,R6,C2},{R3,C1}. With no distinction between faults of elements within fuzzy groupsFaults can theoretically be distinguished. In the present embodiment, the representative failure elements of the 4 fuzzy groups are R1,R2,R3,R4。
Randomly setting a fault, e.g. R31085 Ω, the other elements take values randomly within the tolerance range: r1=10388Ω、R2=9500.1Ω、R4=10191Ω、R5=9655Ω、R6=9878Ω、C1=0.0101μF、C2Obtaining the fault voltage phasor under the action of the excitation signal under the condition of 0.0104 mu F
The number of particles K is set to 100, the maximum number of iterations I is set to 1000, and the precision e is set to 0.000001. FIG. 3 shows R in this example3Global optimal fitness curve graph in fault. As shown in fig. 3, when the number of iterations is 158, the set accuracy is reached, and the loop is exited, where the global minimum fitness value is: 6.177e-05, the corresponding parameter values of each element in the global optimal position vector are: r1=10500Ω、R2=9505Ω、R3=1095Ω、R4=10496Ω、R5=10420Ω、R6=10095Ω、C1=0.0105μF、C20.0105 μ F. Obviously only the resistance R3And if the fault is beyond the tolerance range, the fault diagnosis is correct.
Setting fault R in the same way488891 Ω, the other elements take values randomly within the tolerance range: r1=10388Ω、R2=10050.1Ω、R3=10085Ω、R5=9655Ω、R6=9878Ω、C1=0.0101μF、C20.0104 μ F. FIG. 4 shows R in the present embodiment4Global optimal fitness curve graph in fault. As shown in fig. 4, when the number of iterations is 358, the set accuracy is reached, and the loop is exited, where the global minimum fitness value is: 7.065e-07, the corresponding global optimum position vector has the following parameter values of each element: r1=10281Ω、R2=9728Ω、R3=9747Ω、R4=93980Ω、R5=10151Ω、R6=9516Ω、C1=0.0105μF、C20.0105 μ F. Obviously only the resistance R4And if the fault is beyond the tolerance range, the fault diagnosis is correct.
And then setting 100 times of faults for each representative fault element, wherein each fault is a different fault value, and other fault-free elements are randomly set within a tolerance range, and counting the diagnosis accuracy of the invention.
Table 1 is a statistical table of the diagnosis accuracy of each representative faulty component in the present embodiment.
Component | R1 | R2 | R3 | R4 |
Accuracy of diagnosis | 95% | 90% | 100% | 100% |
TABLE 1
As shown in table 1, the diagnosis accuracy of each representative faulty component in this embodiment is more than 90%, and the average diagnosis accuracy is 96%, which can completely meet the application requirements.
Although illustrative embodiments of the present invention have been described above to facilitate the understanding of the present invention by those skilled in the art, it should be understood that the present invention is not limited to the scope of the embodiments, and various changes may be made apparent to those skilled in the art as long as they are within the spirit and scope of the present invention as defined and defined by the appended claims, and all matters of the invention which utilize the inventive concepts are protected.
Claims (1)
1. An analog circuit single fault diagnosis method based on an improved particle swarm algorithm is characterized by comprising the following steps:
s1: acquiring a transmission function of the analog circuit at a preset measuring point t, recording the number of elements in the analog circuit as C, and recording the nominal value of each element parameter asAnalyzing to obtain fuzzy groups for fault diagnosis of the output voltage of the analog circuit through the measuring point t, recording the number of the obtained fuzzy groups as N, selecting one element as a representative fault element in each fuzzy group, and recording the number of other non-representative fault elements as M;
s2: when the analog circuit has a fault, measuring the fault voltage phasor at the measuring point t under the preset excitation signal
S3: with X ═ X1,…,xN,xN+1,…,xN+MAs the particle positions in the particle swarm, where the first N xnParameter values representing the nth representative faulty component, N being 1,2, …, N, followed by M xmThe parameter values representing the mth non-representative failed component, M ═ N +1, N +2, …, N + M. Initializing K particles to form an initial particle swarm P, randomly taking values of parameter values of N representative fault elements in each particle position, taking values of parameter values of M non-representative fault elements within a tolerance range, and recording the initial position of each particle asThen initializing each particle velocity
S4: respectively calculating the fitness value of each particle in the initial particle swarm P, wherein the calculation method of the fitness value comprises the following steps: respectively calculating particles according to transfer functionOutput voltage phasor at measurement point tThen calculating the output voltage phasor and the fault voltage phasorEuropean distance ofThen, parameters are determined according to the number of the representative fault elements with parameter values exceeding the tolerance range in the particle positions
Selecting the particle position with the smallest fitness valueAs the initial global optimal position Gbest, the initial position of each particle is determinedAs initial local optimum position Pbestk;
S5: initializing the iteration number i to 1;
s6: determining the inertia weight w of the iteration by the following formulai+1:
wi+1=wstart-(wstart-wend)×(i/I)2
Wherein, wstart、wendRespectively representing a start point value and an end point value of the inertia weight, wherein I represents a preset maximum iteration number;
Wherein, c1、c2Is a learning factor, r1、r2Is a random number between 0 and 1;
updating the position of the particle by adopting the following formula to obtain the updated position
Then to the updated positionIn which each element parameter value is subjected to a boundary check, i.e. a particle bitDevice for placingThe parameter value of the non-representative fault element is limited within a tolerance range;
s8: updating the particle position according to the current iterationCalculating the fitness value of each particle;
s9: sorting the particles updated by the current iteration from small to large according to the fitness value, uniformly dividing the particles into D sections, and determining the value of D according to actual needs; recording the number of the particles in each section as Q, removing Q particles in the D section, selecting Q particles in each section from the 2 nd section to the D-1 th section to remove, and copying the Q particles in the 1 st section into two particle groups for supplementing;
s10: performing intersection and variation operation on the particle position vectors in the particle swarm obtained in the step S9, and ensuring that the parameter values of the non-representative fault elements are limited within a tolerance range;
s12: for each particleJudging whether the current fitness value is smaller than the local optimal position PbestkIf so, the local optimum position is setOtherwise, no operation is performed;
screening out the particles with the minimum fitness value in the particles subjected to the current cross variation operation, if the fitness value of the particles is smaller than the global optimal position Gbest, updating the global optimal position Gbest to the position of the particle, and otherwise, not performing any operation;
s13: judging whether an iteration termination condition is reached, if so, entering a step S14, otherwise, entering a step S15;
s14: returning to step S6 by making i + 1;
s15: and the representative fault element with the parameter value in the current global optimal position Gbest within the fault range is the fault diagnosis result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011465725.XA CN112505532A (en) | 2020-12-14 | 2020-12-14 | Analog circuit single fault diagnosis method based on improved particle swarm optimization |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011465725.XA CN112505532A (en) | 2020-12-14 | 2020-12-14 | Analog circuit single fault diagnosis method based on improved particle swarm optimization |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112505532A true CN112505532A (en) | 2021-03-16 |
Family
ID=74972862
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011465725.XA Pending CN112505532A (en) | 2020-12-14 | 2020-12-14 | Analog circuit single fault diagnosis method based on improved particle swarm optimization |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112505532A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114113891A (en) * | 2021-11-20 | 2022-03-01 | 国网四川省电力公司电力科学研究院 | Single-phase earth fault determination method of pulse neurolemma system based on distributed population |
CN114412883A (en) * | 2022-01-14 | 2022-04-29 | 西安建筑科技大学 | Hydraulic system control method, device and system and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101576604A (en) * | 2009-01-04 | 2009-11-11 | 湖南大学 | Method for diagnosing failures of analog circuit based on heterogeneous information fusion |
CN108957246A (en) * | 2018-09-27 | 2018-12-07 | 广东电网有限责任公司 | A kind of electrical power distribution network fault location method based on population |
CN110907810A (en) * | 2019-12-02 | 2020-03-24 | 电子科技大学 | Analog circuit single fault diagnosis method based on particle swarm algorithm |
-
2020
- 2020-12-14 CN CN202011465725.XA patent/CN112505532A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101576604A (en) * | 2009-01-04 | 2009-11-11 | 湖南大学 | Method for diagnosing failures of analog circuit based on heterogeneous information fusion |
CN108957246A (en) * | 2018-09-27 | 2018-12-07 | 广东电网有限责任公司 | A kind of electrical power distribution network fault location method based on population |
CN110907810A (en) * | 2019-12-02 | 2020-03-24 | 电子科技大学 | Analog circuit single fault diagnosis method based on particle swarm algorithm |
Non-Patent Citations (1)
Title |
---|
李妍: "基于改进极限学习机的轴承故障诊断方法研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114113891A (en) * | 2021-11-20 | 2022-03-01 | 国网四川省电力公司电力科学研究院 | Single-phase earth fault determination method of pulse neurolemma system based on distributed population |
CN114113891B (en) * | 2021-11-20 | 2023-04-25 | 国网四川省电力公司电力科学研究院 | Single-phase grounding fault judging method of pulse neural membrane system based on distributed population |
CN114412883A (en) * | 2022-01-14 | 2022-04-29 | 西安建筑科技大学 | Hydraulic system control method, device and system and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11875500B2 (en) | Failure diagnosis method for power transformer winding based on GSMallat-NIN-CNN network | |
CN112485652B (en) | Analog circuit single fault diagnosis method based on improved sine and cosine algorithm | |
CN110907810B (en) | Analog circuit single fault diagnosis method based on particle swarm algorithm | |
CN109581203B (en) | Post-test simulation fault diagnosis method based on genetic algorithm | |
CN111260063B (en) | Analog circuit fault positioning and parameter identification method based on genetic algorithm | |
CN109839583B (en) | Analog circuit multi-fault diagnosis method based on improved genetic algorithm | |
CN112505532A (en) | Analog circuit single fault diagnosis method based on improved particle swarm optimization | |
CN110308384B (en) | Analog circuit fault diagnosis method based on circle model and neural network | |
CN112505533A (en) | Analog circuit double-fault diagnosis method based on improved particle swarm optimization | |
CN110333078B (en) | Rolling bearing degradation state stage determination method | |
CN110673017B (en) | Analog circuit fault element parameter identification method based on genetic algorithm | |
EP0992807B1 (en) | Method and apparatus for board model correction | |
CN111007390B (en) | Analog circuit fault diagnosis model based on algebraic method | |
CN111308327B (en) | Analog circuit fault location and fault element parameter identification method | |
CN109557434B (en) | Partial discharge signal identification method under strong background noise based on composite dictionary sparse representation classification | |
CN109948267B (en) | Linear analog circuit fault diagnosis method based on circular model parameters | |
CN110308386B (en) | Analog circuit fault diagnosis method based on polynomial fitting and state monitoring | |
CN110598849A (en) | HMET scattering parameter extraction method and system based on neural network and storage medium | |
US20050137736A1 (en) | Fuzzy reasoning model for semiconductor process fault detection using wafer acceptance test data | |
CN112287628B (en) | Constraint optimization-based analog circuit fault parameter range determination method | |
CN111595433B (en) | Position determination method and system for vibration sensor of whole aircraft engine | |
CN108490344B (en) | Analog circuit fault diagnosis method based on training sample optimization | |
CN112485650B (en) | Analog circuit fault parameter range identification method based on PBI | |
CN112485651B (en) | Chebyshev-based analog circuit element fault parameter identification method | |
CN111950221B (en) | Method for determining fault parameter range of analog circuit based on genetic 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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210316 |