CN114064340A - Interactive fault diagnosis method suitable for multi-signal flow model - Google Patents

Interactive fault diagnosis method suitable for multi-signal flow model Download PDF

Info

Publication number
CN114064340A
CN114064340A CN202111367177.1A CN202111367177A CN114064340A CN 114064340 A CN114064340 A CN 114064340A CN 202111367177 A CN202111367177 A CN 202111367177A CN 114064340 A CN114064340 A CN 114064340A
Authority
CN
China
Prior art keywords
test
matrix
fault
module
lru
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
Application number
CN202111367177.1A
Other languages
Chinese (zh)
Inventor
王平
陈伟
付黄龙
娄康
贵忠东
罗杭建
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
704th Research Institute of CSIC
Original Assignee
704th Research Institute of CSIC
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 704th Research Institute of CSIC filed Critical 704th Research Institute of CSIC
Priority to CN202111367177.1A priority Critical patent/CN114064340A/en
Publication of CN114064340A publication Critical patent/CN114064340A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
    • G06F11/0706—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation the processing taking place on a specific hardware platform or in a specific software environment
    • G06F11/0736—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation the processing taking place on a specific hardware platform or in a specific software environment in functional embedded systems, i.e. in a data processing system designed as a combination of hardware and software dedicated to performing a certain function
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
    • G06F11/0766—Error or fault reporting or storing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
    • G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
    • G06F11/079—Root cause analysis, i.e. error or fault diagnosis

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Test And Diagnosis Of Digital Computers (AREA)

Abstract

The invention relates to an interactive fault diagnosis method suitable for a multi-signal flow model, which is based on the generation of a test-module adjacency matrix by the multi-signal flow model: converting the adjacent matrix to obtain a reachable matrix among the LRU modules; deducing an equipment fault module-test correlation fault diagnosis matrix from an adjacent matrix of the LRU module through the relevance of the test point where each test is located and the LRU module signal and the accessibility relation of the LRU module and the test point; compressing and storing the fault diagnosis matrix; when the fault diagnosis starts, extracting a serialized fault diagnosis matrix from a stored database to perform deserialization decompression, then simplifying the fault diagnosis matrix according to the existing fault symptoms of the current system and available field test tools, performing test recommendation by adopting a multi-step information heuristic algorithm based on fault probability information entropy and multi-weight priority, and performing interactive diagnosis of the fault. A superior platform is provided for the practical use of multiple signal flow model based fault diagnosis.

Description

Interactive fault diagnosis method suitable for multi-signal flow model
Technical Field
The invention relates to a ship control technology, in particular to an interactive fault diagnosis method suitable for a multi-signal flow model.
Background
The number of ships and the complexity of a control system are continuously improved, the controllable-pitch propeller serving as a main propulsion system of the ships and warships is extremely widely used, however, with the complexity of the controllable-pitch propeller system and the rapid increase of the application number, the defects of high maintenance cost and high difficulty are gradually shown, once the control system breaks down, the ships can have the risks of reef touch, grounding and collision, and the result of out-of-control in a complex water area is extremely serious. The system-level fault diagnosis research results for the electro-hydraulic coupling system at home and abroad are relatively few, the diagnosis mode based on the neural network, the support vector machine and the like cannot be suitable for systems with few diagnosis history samples such as the pitch-adjusting propellers and the like, the model-based multi-signal flow theory is a relatively mature testability modeling theory, and the ship pitch-adjusting propellers can have quick diagnosis capability in a short time by adopting the theory to carry out fault diagnosis, so that the research on a set of interactive fault diagnosis method suitable for the multi-signal flow model is very necessary and meaningful, and the application prospect is very wide.
Disclosure of Invention
Aiming at the problem that a multi-signal flow theory is applied to ship fault diagnosis, an interactive fault diagnosis method suitable for a multi-signal flow model is provided.
The technical scheme of the invention is as follows: an interactive fault diagnosis method suitable for a multi-signal flow model specifically comprises the following steps:
1) generating a test-module adjacency matrix based on the multi-signal flow model:
establishing a multi-signal flow model through historical data, and traversing the directional connection relation of the multi-signal flow model to obtain an adjacent matrix among the LRU modules, wherein the adjacent matrix is the adjacent relation between each LRU module and a test point of the system;
2) generating a reachable matrix: converting the adjacent matrix obtained in the step 1) to obtain reachable matrixes among the LRU modules, namely the reachability of the functional signals among the modules;
3) generating a fault diagnosis matrix: deducing an equipment fault module-test correlation fault diagnosis matrix from an adjacent matrix of the LRU module through the relevance of the test point where each test is located and the LRU module signal and the accessibility relation of the LRU module and the test point;
4) after the fault diagnosis matrix is obtained, performing serialized persistent storage on the fault diagnosis matrix according to the test column, and performing compressed storage;
5) starting diagnosis: when the fault diagnosis starts, the method extracts a serialized fault diagnosis matrix from the database stored in the step 4) to perform deserialization decompression, then simplifies the fault diagnosis matrix according to the existing fault symptoms and available field test tools of the current system, adopts a multi-step information heuristic algorithm based on fault probability information entropy and multi-weight priority to perform test recommendation, performs interactive diagnosis on the fault according to the test points and the test method recommended by the generated diagnosis strategy, and finally determines the fault position.
Furthermore, the multi-signal flow model represents the signal flow direction and the composition and interconnection relationship of each component unit by a layered directed graph, and represents the correlation among the system composition, the function, the fault and the test by defining the correlation among the signals and the component units, the test and the signals.
Further, the derivation process of step 3) is as follows:
3.1) taking each LRU module as a row vector, taking each basic test as a column vector to construct a fault module-test correlation matrix, and setting all matrix element values to be 0 during initialization;
3.2) assigning the correlation matrix in the following way: firstly, signal lists contained in each LRU module and basic test in a matrix are obtained, then, the test of each column is sequentially obtained, a test point to which the test belongs is inquired, the reachability relation between the test point and the module is obtained from the reachability matrix, and if the test point and the module can reach and the intersection of the signal lists of the test point and the module is not empty, the element value of a correlation matrix of the test and the module is set to be 1.
Further, the equipment fault module-test correlation matrix represents the correlation between each test result of the system and the LRU module, and when the test result is normal, the LRU modules related to the test in the correlation matrix are characterized to be normal in function, that is, the element value is 1; when the test result is abnormal, the functions of the LRU modules related to the test in the characterization correlation matrix can be abnormal, each time the test is carried out, the matrix rows which are possibly abnormal are left according to the test result, the matrix rows with normal functions are deleted, the number of the matrix rows is continuously reduced until the matrix rows cannot be reduced, and the LRU modules with the remaining row vectors are possible fault modules.
Further, the adoption of a multi-step information heuristic algorithm based on the fault probability information entropy and the multi-weight priority: the method comprises the steps of carrying out normalization processing on fault probabilities of all possible fault modules of the current system, using the fault probabilities as weights of module fault information to bring the weights into an information entropy, selecting a test with the maximum reduction value of the system information entropy to detect, carrying out weight addition on the test priority, the test cost, the test time and the information entropy to calculate the total recommendation probability, setting and updating the preferences of the test priority, the test cost, the test time and the information entropy weight at any time in a diagnosis process, wherein the test preferences comprise test cost priority, test time priority, test priority and default priority strategies.
The invention has the beneficial effects that: the interactive fault diagnosis method is suitable for the multi-signal flow model, improves the diagnosis efficiency based on the multi-signal flow fault diagnosis, provides the real-time selection capability of diagnosis preference in the diagnosis process, and provides a high-quality platform for the practical use of the multi-signal flow model based fault diagnosis.
Drawings
FIG. 1 is a flow chart of the present invention for interactive fault diagnosis for a multiple signal flow model;
FIG. 2 is a flowchart illustrating the method of generating a diagnostic strategy according to the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
Fig. 1 shows an interactive fault diagnosis flowchart suitable for a multiple-signal flow model, which specifically includes the following steps:
1. firstly, generating a test-module adjacency matrix based on a multi-signal flow model: the multi-signal flow model represents the signal flow direction and the composition and interconnection relationship of each composition Unit (fault mode) by a layered directed graph, represents a model of the correlation between system composition, function, fault and test by defining the correlation between signals (function) and composition units (fault mode) and test and signals, establishes the multi-signal flow model by historical data, and can acquire the adjacent matrix among each LRU (Line Replaceable Unit) module by traversing the depth of the directional connection relationship of the multi-signal flow model, and the following formula is shown as follows:
Figure BDA0003361083210000041
where s is the total number of system LURs and test points, miAll modules and test points (such as power supply, power supply output port test point, relay, etc.) G of system LRUs×sFor a system LRU module, a contiguous matrix between test points, gijA value of 0 or 1, a value of 0 indicates no adjacency between modules or between modules and the test, and a value of 1 indicates an adjacency.
2. Generating a reachable matrix: the reachable matrix among the LRU modules can be obtained by converting the adjacent matrix, namely the reachability of the function signals among the modules, and the conversion formula and the reachable matrix are as follows:
tmpG＝G+G2+G3+...+Gs (2)
Figure BDA0003361083210000042
Figure BDA0003361083210000043
the expression is a conversion formula of the reachable matrix and the adjacent matrix, the intermediate matrix tmpG can be derived from the formula (2), and the reachable matrix rechG can be obtained by performing binarization conversion on each element value in the intermediate matrix tmpG by the piecewise function of the formula (3).
3. Generating a fault diagnosis matrix: by the relevance of the test point where each test is located and the LRU module signal and the accessibility relation of the LRU module and the test point, an equipment fault module-test relevance fault diagnosis matrix can be derived from the adjacency matrix of the LRU module, and the specific derivation process is as follows:
3.1, with each LRU module as a row vector and each basic test as a column vector, the following fault module-test correlation matrix D is constructedm×mWherein m is the number of LRU modules in the system, liFor the system LRU module, n is the number of system base tests, tiFor system foundation tests (such as voltage detection of power output port, power internal resistance detection, etc.), dijTo equip the LRU fault module-test dependency information, with values of either 0 or 1, the matrix element values are all set to 0 at initialization.
Figure BDA0003361083210000051
And 3.2, assigning the correlation matrix in the following way: firstly, signal lists contained in each LRU module and basic test in a matrix are obtained, then, the test of each column is sequentially obtained, a test point to which the test belongs is inquired, the reachability relation between the test point and the module is obtained from the reachability matrix, and if the test point and the module can reach and the intersection of the signal lists of the test point and the module is not empty, the element value of a correlation matrix of the test and the module is set to be 1.
4. After the fault diagnosis matrix is obtained, performing serialized persistent storage on the fault diagnosis matrix according to the test column, and performing compressed storage, wherein the specific serialized compression mode is as follows: converting each 4 bits of the matrix binary value into a 16-bit unit, complementing the last four bits with 0 if the last four bits are not enough, and storing the data in a database in a character string form.
The equipment fault module-test correlation matrix characterizes the correlation between each test result and the LRU module of the system, when the test result is normal, the LRU modules related to the test (namely, the element value is 1) in the correlation matrix are characterized to be normal in function, and when the test result is abnormal, the LRU modules related to the test in the correlation matrix are characterized to be abnormal in function. For the fault diagnosis of the controllable pitch propeller, every time a test is carried out, a possibly abnormal matrix row is left according to a test result, the matrix row with a normal function is deleted, so that the number of the matrix rows is continuously reduced until the matrix rows cannot be reduced, at the moment, the LRU module with the remaining row vectors is a possible fault module, and the fault diagnosis can be directly carried out.
The process of fault diagnosis from the fault diagnosis correlation matrix can be equivalent to set division problem processing, and can prove that the construction problem of the optimized decision tree belongs to NP-complete problem, and a heuristic search method based on information entropy can be adopted to obtain a better solution.
5. Starting diagnosis: when the fault diagnosis starts, the method extracts a serialized fault diagnosis matrix from the database stored in the step 4 to perform deserialization decompression, then simplifies the fault diagnosis matrix according to the existing fault symptoms of the current system and available field test tools, then executes a diagnosis strategy method generation process shown in fig. 2, performs test recommendation by adopting a multi-step information heuristic algorithm based on fault probability information entropy and multi-weight priority, performs interactive diagnosis on the fault according to the test points and the test method recommended by the generated diagnosis strategy, and finally determines the fault position.
The multi-step information heuristic diagnosis strategy generation method based on the fault probability information entropy and the multi-weight priority is composed of a single-step information heuristic search method based on the fault probability information entropy, a multi-step information heuristic search based on the single-step information heuristic and a multi-weight priority method, and the specific methods and the execution sequence of each part are as follows:
1) the single-step information heuristic search method based on the fault probability information entropy comprises the following steps:
shannon proposed that the information is a removal of uncertainty. The more uncertainty removed, the greater the amount of information obtained. The information entropy can also be used as a measure of the complexity of the system, and if the more complex the system is, the more the different situations occur, the larger the information entropy is. In the diagnosis of the pitch propeller system, the system is simplified from the information acquisition of the individual test results, so that the uncertainty of a fault device is reduced. For the device of the controllable pitch propeller system, the probability of possible faults of the modules in the system is related to the fault probability of the modules, so the algorithm normalizes the fault probabilities of all the possible fault modules of the current system, takes the normalized fault probabilities as the weight of the fault information of the modules and brings the weighted fault information into an information entropy formula, and then selects the test with the maximum reduction value of the system information entropy for detection.
Information entropy Single-step information heuristic search method based on Fault probability information entropy, in which the row elements (i.e., LRU modules) remaining in the current correlation matrix are referred to as fuzzy sets x, and test t is performed to maximize the Fault probability entropy for a single test, which is reduced to a test selection criterionkIs selected, if this test maximizes the entropy of the probability of failure of the individual test minus:
IG(x,tj)＝-{p(xjp)log2p(xjp)+p(xjp)log2(xjf)} (6)
k＝arg maxj{IG(x,tj)} (7)
wherein xjpAnd xjfIs the fuzzy set x is performing the test tjTwo result sets, p (x), divided laterjp) And p (x)jf) Normalized fault summary for two result-focused modulesSum of rates, IG (x, t)j) Namely, the entropy subtraction value of the fuzzy group of the test pair system, and k is the test serial number with the maximum entropy subtraction value of the fuzzy group of the test pair system.
2) A multi-step information heuristic search method based on single-step information inspiration comprises the following steps:
the single-step information heuristic search method based on the fault probability information entropy judges the quality of a test result only through a single-layer result, the result is not good for the predictability of the subsequent steps, and the subsequent test can influence the final diagnosis result, so that the single-step information heuristic search method based on the single-step information heuristic can be used for greatly optimizing the test recommendation result. The multi-step information heuristic search method based on single-step information inspiration comprises the following steps:
step 1, selecting a test t from available test seti，tiAnd splitting the current fault set x into two subsets according to the passing or the failure of the output result.
And 2, for each subset, selecting an available optimal test based on the single-step information heuristic search, and splitting the subset according to the passing and the failure of an output result.
And 3, repeating the step 2 for each subset recursion until the depth of the partial diagnostic tree reaches a certain depth step or the fault set is not subdividable (no fuzzy group). And calculating the fault probability information entropy reduction of the single step average of the partial diagnosis tree.
And 4, repeating the step 2 and the step 3 for each selected test in the step 1, and obtaining the single-step average fault probability information entropy subtraction of all tests in the current test set.
And 5: and selecting the test with the maximum entropy of the single-step average fault probability information as the optimal test for the current fuzzy set.
Figure BDA0003361083210000071
Figure BDA0003361083210000072
3) The multi-weight priority method comprises the following steps:
in the fault diagnosis process, the field environment is changeable, the step optimal diagnosis tree obtained on the basis of the module fault probability is not necessarily an optimal solution, and other influence factors also comprise the factors such as test time, test cost and the like (for example, the high-position oil tank of the controllable pitch propeller is very difficult to test and has higher cost, the measurement is not recommended, a system access pressure gauge needs to be closed when the valve port pressure of the reversing valve is measured, the test cost is higher and the like), so the step 5 in the multi-step heuristic information search method is further optimized, the selection function is changed into the multi-weight priority function, and the function method comprises the following steps:
Figure BDA0003361083210000081
wherein f is1For test information entropy minus the weight value in the selection, f2Weight value in selection for testing time cost, f3To test the weight value spent in the selection, f4To test the weight value of the subjective priority in the selection, tj(time) is test tjThe time cost required for execution, tj(cost) is test tjCost required for execution, tj(Priority) is test tjAnd finally, the optimal test in the current fault mode set is selected as the current recommended test to execute the fault diagnosis process shown in the figure 1.
The information entropy calculation of the algorithm is carried out substitution calculation according to the normalized fault probability of each isolation module, and the total recommendation probability is calculated by carrying out weight addition on the test priority, the test cost, the test time and the information entropy, the preference of the test priority, the test cost, the test time and the information entropy weight can be set and updated at any time in the diagnosis process, and the test preference comprises a test cost priority strategy, a test time priority strategy, a test priority strategy and a default priority strategy.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims (5)

1. An interactive fault diagnosis method suitable for a multi-signal flow model is characterized by specifically comprising the following steps:
1) generating a test-module adjacency matrix based on the multi-signal flow model:
establishing a multi-signal flow model through historical data, and traversing the directional connection relation of the multi-signal flow model to obtain an adjacent matrix among the LRU modules, wherein the adjacent matrix is the adjacent relation between each LRU module and a test point of the system;
2) generating a reachable matrix: converting the adjacent matrix obtained in the step 1) to obtain reachable matrixes among the LRU modules, namely the reachability of the functional signals among the modules;
3) generating a fault diagnosis matrix: deducing an equipment fault module-test correlation fault diagnosis matrix from an adjacent matrix of the LRU module through the relevance of the test point where each test is located and the LRU module signal and the accessibility relation of the LRU module and the test point;
4) after the fault diagnosis matrix is obtained, performing serialized persistent storage on the fault diagnosis matrix according to the test column, and performing compressed storage;
5) starting diagnosis: when the fault diagnosis starts, the method extracts a serialized fault diagnosis matrix from the database stored in the step 4) to perform deserialization decompression, then simplifies the fault diagnosis matrix according to the existing fault symptoms and available field test tools of the current system, adopts a multi-step information heuristic algorithm based on fault probability information entropy and multi-weight priority to perform test recommendation, performs interactive diagnosis on the fault according to the test points and the test method recommended by the generated diagnosis strategy, and finally determines the fault position.
2. The interactive fault diagnosis method for multiple signal flow models according to claim 1, wherein the multiple signal flow models represent the signal flow direction and the composition and interconnection relationship of each component unit by a layered directed graph, and represent a model of the correlation between the system composition, function, fault and test by defining the correlation between the signal and the component unit, test and signal.
3. The interactive fault diagnosis method for multiple signal flow models according to claim 1 or 2, wherein the derivation process of step 3) is as follows:
3.1) taking each LRU module as a row vector, taking each basic test as a column vector to construct a fault module-test correlation matrix, and setting all matrix element values to be 0 during initialization;
3.2) assigning the correlation matrix in the following way: firstly, signal lists contained in each LRU module and basic test in a matrix are obtained, then, the test of each column is sequentially obtained, a test point to which the test belongs is inquired, the reachability relation between the test point and the module is obtained from the reachability matrix, and if the test point and the module can reach and the intersection of the signal lists of the test point and the module is not empty, the element value of a correlation matrix of the test and the module is set to be 1.
4. The interactive fault diagnosis method for multiple signal flow models according to claim 3, wherein the equipment fault module-test correlation matrix characterizes the correlation between each test result and the LRU module of the system, and when the test result is normal, the LRU modules related to the test in the correlation matrix are characterized to be normal in function, i.e. the element value is 1; when the test result is abnormal, the functions of the LRU modules related to the test in the characterization correlation matrix can be abnormal, each time the test is carried out, the matrix rows which are possibly abnormal are left according to the test result, the matrix rows with normal functions are deleted, the number of the matrix rows is continuously reduced until the matrix rows cannot be reduced, and the LRU modules with the remaining row vectors are possible fault modules.
5. The interactive fault diagnosis method suitable for the multi-signal flow model according to claim 3, characterized in that a multi-step information heuristic algorithm based on fault probability information entropy and multi-weight priority is adopted: the method comprises the steps of carrying out normalization processing on fault probabilities of all possible fault modules of the current system, using the fault probabilities as weights of module fault information to bring the weights into an information entropy, selecting a test with the maximum reduction value of the system information entropy to detect, carrying out weight addition on the test priority, the test cost, the test time and the information entropy to calculate the total recommendation probability, setting and updating the preferences of the test priority, the test cost, the test time and the information entropy weight at any time in a diagnosis process, wherein the test preferences comprise test cost priority, test time priority, test priority and default priority strategies.
CN202111367177.1A 2021-11-18 2021-11-18 Interactive fault diagnosis method suitable for multi-signal flow model Pending CN114064340A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111367177.1A CN114064340A (en) 2021-11-18 2021-11-18 Interactive fault diagnosis method suitable for multi-signal flow model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111367177.1A CN114064340A (en) 2021-11-18 2021-11-18 Interactive fault diagnosis method suitable for multi-signal flow model

Publications (1)

Publication Number Publication Date
CN114064340A true CN114064340A (en) 2022-02-18

Family

ID=80278409

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111367177.1A Pending CN114064340A (en) 2021-11-18 2021-11-18 Interactive fault diagnosis method suitable for multi-signal flow model

Country Status (1)

Country Link
CN (1) CN114064340A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114553273A (en) * 2022-02-25 2022-05-27 广州大学 Effective search method for large-scale MIMO optimal signal detection

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114553273A (en) * 2022-02-25 2022-05-27 广州大学 Effective search method for large-scale MIMO optimal signal detection
CN114553273B (en) * 2022-02-25 2023-05-23 广州大学 Efficient searching method for large-scale MIMO optimal signal detection

Similar Documents

Publication Publication Date Title
CN108520080B (en) Ship diesel generator fault prediction and health state online evaluation system and method
CN109948833A (en) A kind of Hydropower Unit degradation trend prediction technique based on shot and long term memory network
CN110046409B (en) ResNet-based steam turbine component health state evaluation method
WO2005001757A1 (en) Information processor, state judging unit and diagnostic unit, information processing method, state judging method and diagnosing method
CN110994604A (en) Electric power system transient stability evaluation method based on LSTM-DNN model
CN111340248A (en) Transformer fault diagnosis method and system based on intelligent integration algorithm
CN109709483A (en) A kind of variable-pitch system of wind turbine generator method for diagnosing faults
CN114064340A (en) Interactive fault diagnosis method suitable for multi-signal flow model
CN110132627A (en) A kind of method for diagnosing faults of propeller
CN108431712A (en) Control system for operating carrier
CN112461543B (en) Rotary machine fault diagnosis method based on multi-classification support vector data description
JPH11212637A (en) Method and device for preventive maintenance
CN113459867A (en) Electric vehicle charging process fault early warning method based on adaptive deep confidence network
CN111400964B (en) Fault occurrence time prediction method and device
CN113850305A (en) Generator fault diagnosis and health state assessment method and electronic equipment
CN112802011A (en) Fan blade defect detection method based on VGG-BLS
CN105976127B (en) Risk assessment method and device for comprehensive electric propulsion system
CN117096871A (en) Wind power probability density prediction method based on space-time distribution
CN117491872A (en) Reconfigurable battery module fault multistage diagnosis method
Huan et al. River dissolved oxygen prediction based on random forest and LSTM
CN116577843A (en) Wind power prediction method for distributed sodar
CN115292820A (en) Method for predicting residual service life of urban rail train bearing
CN113884903B (en) Battery fault diagnosis method based on multi-layer perceptron neural network
CN115795360A (en) Cable fault detection method based on artificial neural network
CN115585103A (en) Double-fed fan variable pitch fault diagnosis method and system

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