CN116882304A - Closed bus temperature fault monitoring method based on multi-gradient descent bee colony algorithm - Google Patents
Closed bus temperature fault monitoring method based on multi-gradient descent bee colony algorithm Download PDFInfo
- Publication number
- CN116882304A CN116882304A CN202311152379.3A CN202311152379A CN116882304A CN 116882304 A CN116882304 A CN 116882304A CN 202311152379 A CN202311152379 A CN 202311152379A CN 116882304 A CN116882304 A CN 116882304A
- Authority
- CN
- China
- Prior art keywords
- temperature
- optimal
- current
- honey source
- gradient descent
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 98
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 60
- 238000012544 monitoring process Methods 0.000 title claims abstract description 39
- 235000012907 honey Nutrition 0.000 claims abstract description 121
- 230000008859 change Effects 0.000 claims abstract description 39
- 230000008569 process Effects 0.000 claims abstract description 38
- 238000005457 optimization Methods 0.000 claims abstract description 28
- 238000013178 mathematical model Methods 0.000 claims abstract description 20
- 230000007613 environmental effect Effects 0.000 claims abstract description 11
- 238000000605 extraction Methods 0.000 claims abstract description 7
- 238000012360 testing method Methods 0.000 claims abstract description 4
- 241000257303 Hymenoptera Species 0.000 claims description 67
- 230000007246 mechanism Effects 0.000 claims description 14
- 238000012216 screening Methods 0.000 claims description 12
- 238000004088 simulation Methods 0.000 claims description 9
- 230000007423 decrease Effects 0.000 claims description 4
- 229910000838 Al alloy Inorganic materials 0.000 claims description 3
- 239000003814 drug Substances 0.000 claims description 3
- 229940079593 drug Drugs 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 239000002994 raw material Substances 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 241000256844 Apis mellifera Species 0.000 description 58
- 239000013307 optical fiber Substances 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000009977 dual effect Effects 0.000 description 3
- 238000009529 body temperature measurement Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000009191 jumping Effects 0.000 description 2
- 238000012806 monitoring device Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 229910052782 aluminium Inorganic materials 0.000 description 1
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000020169 heat generation Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 239000011810 insulating material Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 238000011282 treatment Methods 0.000 description 1
- 230000001960 triggered effect Effects 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
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Biomedical Technology (AREA)
- Geometry (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of power equipment fault monitoring, and particularly relates to a closed bus temperature fault monitoring method based on a multi-gradient descent bee colony algorithm, which comprises the following steps of: s1, combining a multi-objective optimization process of a multi-gradient descent algorithm with an artificial bee colony algorithm, and establishing a multi-gradient descent bee colony algorithm model; s2, collecting an environment temperature and current test sample of the closed bus, performing feature extraction and multi-objective optimization, and establishing a temperature current multi-gradient descent swarm mathematical model according to the multi-gradient descent swarm algorithm model in S1; s3, searching an optimal honey source searching direction through the established temperature-current multi-gradient descending swarm mathematical model, and further obtaining an optimal solution of the environmental temperature and the current change rate of the enclosed bus, wherein the position of the optimal solution is most likely to be the temperature fault point of the enclosed bus. The invention effectively improves the efficiency and the precision of the temperature fault monitoring of the enclosed bus, shortens the time cost and improves the safety coefficient of the power system.
Description
Technical Field
The invention belongs to the technical field of power equipment fault monitoring, and particularly relates to a closed bus temperature fault monitoring method based on a multi-gradient descent bee colony algorithm.
Background
The enclosed bus is an important power transmission and distribution device in power equipment and is widely applied to energy transmission and distribution of a power system. The bus system is composed of a metal plate (steel plate or aluminum plate) as a protective shell, a conductive bar, an insulating material and related accessories. In the actual transmission process, each node of the bus generates a magnetic field under the action of current, so that a large amount of heat is generated, the heat generation depends on the magnitude of the bus current, and is influenced by the change of the ambient temperature. Therefore, the monitoring of the temperature change rate of each node of the enclosed bus is particularly important. The current change rate and the ambient temperature change rate become two main indexes for influencing the temperature change rate.
At present, the methods for monitoring the temperature faults of the enclosed bus mainly comprise the following steps:
and monitoring by an infrared temperature sensor. An infrared temperature sensor is a sensor capable of measuring the surface temperature of an object by capturing infrared energy radiated from the object. It uses the relationship between the infrared radiation and the temperature of the object surface to indirectly infer the temperature of the object by measuring the intensity of the infrared radiation. When using an infrared temperature sensor for closed busbar temperature monitoring, it should be ensured that there are no obstructions and sources of interference between the sensor and the object being measured, to ensure accurate temperature measurement and fault detection.
And monitoring the power transmitter. A power transmitter is a device for measuring and transmitting electrical signals. It can convert the temperature signal of the enclosed bus into an electrical signal for monitoring and processing. The power transmitter then amplifies, filters, and linearizes the converted electrical signal to ensure signal accuracy and stability. These treatments can be flexibly adjusted to the specific application requirements. If the temperature abnormality of the enclosed bus is detected, corresponding alarm or protection measures are triggered.
At present, the technology related to closed bus temperature fault monitoring generally lacks the problems of multi-target uniformity, single temperature fault monitoring mechanism, no optimal memory participation and the like, so that when different influencing factors act simultaneously, the conditions of information disorder, inaccurate temperature measurement, low monitoring efficiency and the like occur.
In the prior patent, CN110319951a discloses a temperature monitoring system and method for a closed bus, which is characterized in that an optical fiber line comprising a plurality of segments of temperature measuring optical fibers is arranged in the closed bus, and a light emitting and receiving device emits incident light to the optical fiber line; the temperature acquisition device determines temperature data of the enclosed bus according to an emergent spectrum after the incident light is reflected and transmitted by the temperature measuring optical fiber; the temperature monitoring device receives temperature data and realizes multi-monitoring-point and omnibearing temperature monitoring of the enclosed bus.
In this patent, the temperature detection and monitoring of the enclosed busbar are performed by the incident light and the optical fiber line including the temperature measuring optical fiber, and the temperature monitoring device, the temperature acquisition device, the light emitting and receiving device, the optical fiber line of the multi-section temperature measuring optical fiber, and the like are additionally arranged, so that the temperature detection device is easily affected by an interference source, the temperature fault monitoring mechanism is single, and the maintenance of equipment is required to be performed regularly.
In order to improve the efficiency and the precision of the closed bus temperature fault monitoring, the invention considers that the bee colony algorithm is applied to the closed bus temperature fault monitoring process, the artificial bee colony algorithm belongs to one of the swarm intelligent algorithms, and is inspired by the honey searching and collecting processes of bees.
However, the existing artificial bee colony algorithm is easy to fall into a local optimal solution under the condition of multiple targets, so that the artificial bee colony algorithm can be used after being improved.
Disclosure of Invention
According to the defects in the prior art, the invention provides the closed bus temperature fault monitoring method based on the multi-gradient descent bee colony algorithm, which effectively improves the efficiency and the precision of closed bus temperature fault monitoring, shortens the time cost and improves the safety coefficient of an electric power system.
In order to achieve the above purpose, the invention provides a closed bus temperature fault monitoring method based on a multi-gradient descent bee colony algorithm, which comprises the following steps:
s1, combining a multi-objective optimization process of a multi-gradient descent algorithm with an artificial bee colony algorithm, and establishing a multi-gradient descent bee colony algorithm model;
the optimal memory of the bee colony multi-element search mechanism can be realized, so that the defect of local optimal solution is overcome, and the aim of multi-objective parallel optimization is fulfilled.
S2, collecting an environment temperature and current test sample of the closed bus, performing feature extraction and multi-objective optimization, and establishing a temperature current multi-gradient descent swarm mathematical model according to the multi-gradient descent swarm algorithm model in S1;
s3, searching an optimal honey source searching direction through the established temperature-current multi-gradient descending swarm mathematical model, and further obtaining an optimal solution of the environmental temperature and the current change rate of the enclosed bus, wherein the position of the optimal solution is most likely to be the temperature fault point of the enclosed bus.
The multi-gradient descent bee colony algorithm model is based on an artificial bee colony algorithm, and on the basis that the problem of local optimal solution is easily caused due to single search mechanism of the scout bees without optimal memory participation, the honey searching mechanism of the scout bees is improved by establishing a multi-gradient model, and the speed of obtaining the optimal solution of the honey source is accelerated by using a multi-variable maximum gradient. The multi-gradient descent swarm algorithm model classifies swarm categories into three categories: the method comprises the steps of detecting bees, observing bees and collecting bees. In the algorithm, the bees are equivalent to execution units of optimization algorithms, form small groups through certain information exchange and interaction, continuously transmit information to each other, select jumping points and continuously update the positions of the bees.
In the step S1, the step of establishing a multi-gradient descent bee colony algorithm model is as follows:
s11, the swarm types of the multi-gradient descent swarm algorithm model are a reconnaissance bee, an observation bee and a bee collection, a solution space of a target problem is defined to be M dimensions, the number of the reconnaissance bee, the number of the observation bee and the number of the bee collection are N, W, Q respectively, and the reconnaissance bee randomly searches for new honey source information in each period so as to meet the following formula:
；
where N is the number of scout bees, n=1, 2,3, … …, N, M is the spatial dimension of the target problem solution, m=1, 2,3, … …, M,in order to newly generate a possible solution in m dimension when the number of scout bees is n,/i>In order to solve the original m-dimensional solution space when the number of the scout bees is n>Is defined as that when the number of the scout bees is n, the newly generated number falls within the interval [ -1,1 ] in m dimension]K is a defined space coefficient, k is not equal to n;
s12, after the reconnaissance bees feed back the honey source information to the observation bees, the observation bees select one honey source from the honey sources according to probability, and the honey source set obtained by the reconnaissance bees is defined asThe honey source is a local optimal solution because of a single search mechanism of the scout bees, and the selection probability is as follows:
；
in the method, in the process of the invention,the last level of bee collected for the w-th observation bee may solve +.>Is adapted to the value of->The probability of selection for the w-th observation bee;
s13, selecting bees according to observationSelecting probability and related honey source information, selecting optimal fitness value, namelyOptimal solution->，/>The honey source information of the previous period is set as +.>When honey source information->And (3) withUpdating the optimal honey source information when compared with the relative superiority, otherwise, discarding +.>The bee picking becomes a reconnaissance bee, search is continued in the current honey source neighborhood, and the search formula becomes:
；
in the method, in the process of the invention,in order to obtain the possible solution of m dimension newly generated when the number of bees is q, the weight of the bees is increased by->、/>When the number of the collected bees is q, the lower bound and the upper bound of m dimensions in the past and at present respectively, and r represents an interval [0,1]The random number on the table, k is not equal to q;
s14, optimizing a honey collecting mechanism of the scout bees according to the multi-gradient strategy to form optimal memory, and determining an optimal honey source searching directionThe honeybee can obtain the optimal honey source, so that a multi-target parameter honeybee-collecting optimization system is established:
；
in the method, in the process of the invention,for optimal honey source solution, the drug is prepared from the following raw materials>An ith input honey source information parameter (which can be understood as an output variable in practical application);
s15, setting the optimal honey source searching direction asAccording to a multiple gradient decrease, so that +.>The method meets the following conditions:
；
；
in the method, in the process of the invention,after searching honey sources for the scout bees, the honey source information distance of the neighborhood;
s16, in order to obtain an optimal honey source searching direction, and enable all searching directions to be towards the optimal honey source searching direction after each new honey source searching, updating information, and converting the formula in S15 into:
；
the search is iterated continuously, so that the Pareto optimal solution of the problem can be obtained:
；
s17, if an optimal honey source searching direction existsMake->At the same time satisfyAt this time +.>Is a multi-objective problemIs the optimal solution of (a); if the optimal honey source searching direction cannot be found +.>Then the Pareto optimal solution does not exist, at which point the last honey source is the optimal honey source.
The establishment of the multi-gradient descent bee colony algorithm model is that an optimal memory mechanism is established in honey source information acquisition through multi-gradient descent, and the process is represented by the following formula:
；
in the method, in the process of the invention,for bee colony data without optimal memory participation, < ->And (3) for the bee colony data in the participation of optimal memory, gamma is the input data of the optimal memory, and beta is the self-adaptive function value of the optimal honey source.
In the step S2, the step of establishing a temperature current multi-gradient descending bee colony mathematical model is as follows:
s21, collecting and collecting the environmental temperature contacted by the aluminum alloy provided with the enclosed bus, wherein the environmental temperature is as follows:
；
the current collection set is as follows:
；
the highest temperature of the sample data isThe minimum temperature is->Maximum current of->Minimum current is +.>；
S22, collecting temperature and current、/>Extracting features, screening information, and setting class parametersInitial center point +.>，/>，/>，/>Screening is performed on each sample data:
；
；
in the method, in the process of the invention,for the ith temperature parameter, +.>、/>For the parallel optimization parameters of temperature and current (i.e. the parallel optimization parameters of class 1 and class 2), and (2)>、/>、/>、/>The temperature is respectively the highest temperature and the lowest temperature of the ambient temperature and the bus current;
s23, after data screening through feature extraction, setting the extracted temperature set as follows:
；
the current set is:
；
establishing a temperature-current parameter honey collection optimizing system:
；
；
in the method, in the process of the invention,、/>for optimal temperature and current source solution, < >>、/>Inputting a temperature current information parameter for the ith input;
s24, maximizing、/>So that the minimum value of both is less than 0, the optimal temperature search direction +.>And optimal current search direction->Can make the two reach the purpose of descending at the same time, namely, the established optimal honey source searching direction +.>The process is as follows:
；
further, it is converted into:
；
by determining the optimal honey source searching direction when the temperature and current gradient change rate is maximumSuch that:
；
；
at this timeThe method is an optimal solution of the ambient temperature and the current change rate of the enclosed bus;
s25, optimal honey source searching directionAnd (3) carrying out normalization processing on the double-parameter gradient through a dual rule. If the temperature current gradient is not normalized, the smallest gradient element will be affected by the small norm gradient in the family, so there is a problem of whether the optimized gradient can be well balanced multi-parameter gradient descent. Setting a double-parameter search step size:
；
order the，/>，/>Are all within the defined r interval [0,1 ]]In (a) and (b); />Search step size for class i,/->Optimizing parameters for the j-th class of parallelism;
the temperature current parallel optimization parameters satisfy:
；
；
in the method, in the process of the invention,optimizing parameters for class i parallelism, +.>、/>Is->The temperature and current information parameters are input.
In the step S3, the step of searching the optimal solution is as follows:
determining optimal honey source searching according to the temperature current multi-gradient descending bee colony mathematical model established in the step S2Further determining the optimal solution of the ambient temperature and the current change rate of the enclosed bus>The model is as follows:
；
in the method, in the process of the invention,、/>、/>、/>temperature current optimal honey source solution>、/>Inputting the mean value and variance of the optimized parameters, +.>After searching honey sources for the scout bees, the information distance of the honey sources in the neighborhood,when the number of the scout bees is n, the dimension of the input parameter after optimization is +.>At this time, the newly generated temperature may be resolved,when the number of the scout bees is n, the dimension of the input parameter after optimization is +.>At this time, the newly generated current may be resolved,、/>、/>、/>for temperature and current in n spy bees, < >>Upper and lower boundaries of past and present dimensions;
through the mathematical model, the optimal honey source searching directionIs determined and the optimal solution of the ambient temperature and the current change rate of the enclosed bus is further obtained ∈>。
In summary, the temperature and current samples are collected for preliminary screening and optimization, and a related mathematical model is established according to the multi-gradient bee colony algorithm. The model determines the optimal honey source searching directionAnd establishing an optimal memory mechanism, continuously scaling all temperature and current optimization parameters to approximately the same range in a characteristic scaling mode, and finally updating all parameters together towards a direction with the highest descending speed so as to obtain the point with the largest temperature and current multi-gradient change rate, namely the optimal honey source. Therefore, the temperature-current multi-gradient descending swarm mathematical model established by the invention can effectively solve the problem of closed bus temperature faults, and is a set of feasible schemes.
In the step S3, simulation verification is carried out on the obtained optimal solution through MATLAB.
By means of MATLAB and the traditional monitoring method, the invention finds the optimal solution of the temperature and current multivariable mixed change rate through simulation, the change rate of the wall temperature of the closed bus is faster under the influence of the solution, the time cost required for reaching the upper limit of the safety temperature threshold is smaller, and the superiority of the model in the field of closed bus temperature fault monitoring is verified.
The algorithm of the present invention may be executed by an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the algorithm being implemented by the processor executing the program.
The invention has the beneficial effects that:
the invention improves the traditional artificial bee colony algorithm and is applied to the closed bus temperature fault monitoring, and the honey source information set is set by taking the closed bus current and the environment temperature as input parameters to carry out preliminary screening and optimization on data. Establishing a temperature-current multi-gradient descent swarm mathematical model, realizing parallel optimization of current and temperature double parameters, and searching for an optimal honey source searching direction meeting the conditionsTo determine the optimal solution when the rate of change of temperature and current mixing is at a maximum. Finally, by carrying out simulation and comparison with the traditional method, the temperature change rate of the wall of the closed bus is faster under the influence of the optimal solution, and the time cost required for reaching the upper limit of the safety temperature threshold is smaller, so that the position presenting the optimal solution with the mixed change rate is most likely to become the temperature fault point of the closed bus, thereby effectively solving the temperature fault problem of the closed bus and being a set of feasible schemes.
The simulation result can verify that the method effectively improves the efficiency and the precision of monitoring the temperature faults of the closed bus, shortens the time cost, enables the faults to be monitored at the fastest speed in a relatively short time, and improves the safety coefficient of the system.
Drawings
FIG. 1 is a schematic diagram of a temperature-current multi-gradient descent swarm algorithm processing model;
FIG. 2 is a schematic diagram of a honey collection mechanism of a bee colony;
FIGS. 3 and 4 are schematic diagrams of temperature distribution of the enclosed bus duct at different angles;
FIG. 5 is a graph of the wall temperature of a closed bus duct versus an influencing parameter;
FIG. 6 is a simulated plot of ambient temperature and bus current variation;
FIG. 7 is a dual vector gradient descent determination search directionSchematic of (2);
FIG. 8 is a graph of simulation of the effect of rate of change of mixing on rate of change of temperature obtained by different monitoring methods.
Detailed Description
Embodiments of the invention are further described below with reference to the accompanying drawings:
fig. 1 is a model of a temperature-current multi-gradient descent swarm algorithm established according to a principle flow. The model firstly collects the environmental temperature and the current sample of the enclosed bus and performs optimization screening, establishes a connection with the established multi-gradient bee colony algorithm model through a verification set and a control set, and searches the optimal honey source searching directionThe value of the rate of change of the mixture is further determined to determine the closed bus temperature fault location.
The closed bus temperature fault monitoring method based on the multi-gradient descent bee colony algorithm comprises the following steps:
s1, combining a multi-objective optimization process of a multi-gradient descent algorithm with an artificial bee colony algorithm, and establishing a multi-gradient descent bee colony algorithm model;
s2, collecting an environment temperature and current test sample of the closed bus, performing feature extraction and multi-objective optimization, and establishing a temperature current multi-gradient descent swarm mathematical model according to the multi-gradient descent swarm algorithm model in S1;
s3, searching an optimal honey source searching direction through the established temperature-current multi-gradient descending swarm mathematical model, and further obtaining an optimal solution of the environmental temperature and the current change rate of the enclosed bus, wherein the position of the optimal solution is most likely to be the temperature fault point of the enclosed bus.
The method has the advantages that after the optimal solution of the temperature and current multivariable mixed change rate is found through simulation by means of MATLAB and comparison simulation of a traditional monitoring method, the change rate of the wall temperature of the closed bus is faster under the influence of the solution, the time cost required for reaching the upper limit of a safe temperature threshold is lower, and the superiority of the model in the field of closed bus temperature fault monitoring is verified.
In S1, the steps of establishing a multi-gradient descent bee colony algorithm model are as follows:
fig. 2 shows a schematic diagram of the multi-gradient descent swarm algorithm model in processing information. Observing how the bees, the reconnaissance bees and the bee picking finish information acquisition and information processing exchange. In the algorithm, the bees are equivalent to execution units of optimization algorithms, form small groups through certain information exchange and interaction, continuously transmit information to each other, select jumping points and continuously update the positions of the bees.
S11, the swarm types of the multi-gradient descent swarm algorithm model are a reconnaissance bee, an observation bee and a bee collection, a solution space of a target problem is defined to be M dimensions, the number of the reconnaissance bee, the number of the observation bee and the number of the bee collection are N, W, Q respectively, and the reconnaissance bee randomly searches for new honey source information in each period so as to meet the following formula:
；
where N is the number of scout bees, n=1, 2,3, … …, N, M is the spatial dimension of the target problem solution, m=1, 2,3, … …, M,in order to newly generate a possible solution in m dimension when the number of scout bees is n,/i>In order to solve the original m-dimensional solution space when the number of the scout bees is n>Is defined as that when the number of the scout bees is n, the newly generated number falls within the interval [ -1,1 ] in m dimension]K is a defined space coefficient, k is not equal to n;
s12, after the reconnaissance bees feed back the honey source information to the observation bees, the observation bees select one honey source from the honey sources according to probability, and the honey source set obtained by the reconnaissance bees is defined asThe honey source is a local optimal solution because of a single search mechanism of the scout bees, and the selection probability is as follows:
；
in the method, in the process of the invention,the last level of bee collected for the w-th observation bee may solve +.>Is adapted to the value of->The probability of selection for the w-th observation bee;
s13, selecting the optimal fitness value according to the selection probability of the observed bees and the related honey source information, namelyOptimal solution->，/>The honey source information of the previous period is set as +.>When honey source information->And (3) withUpdating the optimal honey source information when compared with the relative superiority, otherwise, discarding +.>The bee picking becomes a reconnaissance bee, search is continued in the current honey source neighborhood, and the search formula becomes:
；
in the method, in the process of the invention,in order to obtain the possible solution of m dimension newly generated when the number of bees is q, the weight of the bees is increased by->、/>When the number of the collected bees is q, the lower bound and the upper bound of m dimensions in the past and at present respectively, and r represents an interval [0,1]The random number on the table, k is not equal to q;
s14, establishing a multi-objective parameter honey collection optimizing system:
；
in the method, in the process of the invention,for optimal honey source solution, the drug is prepared from the following raw materials>Inputting honey source information parameters for the ith input;
s15, setting the optimal honey source searching direction asAccording to a multiple gradient decrease, so that +.>The method meets the following conditions:
；
；
in the method, in the process of the invention,after searching honey sources for the scout bees, the honey source information distance of the neighborhood;
s16, in order to obtain an optimal honey source searching direction, and enable all searching directions to be towards the optimal honey source searching direction after each new honey source searching, updating information, and converting the formula in S15 into:
；
the search is iterated continuously, so that the Pareto optimal solution of the problem can be obtained:
；
s17, if an optimal honey source searching direction existsMake->At the same time satisfyAt this time +.>Is a multi-objective problemIs the optimal solution of (a); if the optimal honey source searching direction cannot be found +.>Then the Pareto optimal solution does not exist, at which point the last honey source is the optimal honey source.
The establishment of the multi-gradient descent bee colony algorithm model, namely, an optimal memory mechanism is established in honey source information acquisition through multi-gradient descent, and the process is represented by the following formula:
；
in the method, in the process of the invention,for bee colony data without optimal memory participation, < ->And (3) for the bee colony data in the participation of optimal memory, gamma is the input data of the optimal memory, and beta is the self-adaptive function value of the optimal honey source.
Fig. 3 and 4 show the internal and external temperature distribution diagrams of the enclosed bus duct at different angles. The temperature distribution diagram shows that the temperature of the enclosed bus gradually decreases from inside to outside by taking the bus as the axis.
Fig. 5 shows that the main parameters affecting the temperature change inside and outside the enclosed bus duct are ambient temperature, bus current. By the change trend relationship of the scatter diagram, it can be seen that the temperature inside and outside the groove and the two main parameters show an approximately linear increase relationship.
S2, establishing a temperature-current multi-gradient descending bee colony mathematical model, wherein the steps are as follows:
s21, collecting and collecting the environmental temperature contacted by the aluminum alloy provided with the enclosed bus, wherein the environmental temperature is as follows:
；
the current collection set is as follows:
；
the highest temperature of the sample data isThe minimum temperature is->Maximum current of->Minimum current is +.>；
S22, collecting temperature and current、/>Extracting features, screening information, and setting class parametersInitial center point +.>，/>，/>，/>Screening is performed on each sample data:
；
；
in the method, in the process of the invention,for the ith temperature parameter, +.>、/>Optimizing parameters for temperature and current parallelism, +.>、/>、、/>The temperature is respectively the highest temperature and the lowest temperature of the ambient temperature and the bus current;
s23, after data screening through feature extraction, setting the extracted temperature set as follows:
；
the current set is:
；
establishing a temperature-current parameter honey collection optimizing system:
；
；
in the method, in the process of the invention,、/>for optimal temperature and current source solution, < >>、/>Inputting a temperature current information parameter for the ith input;
regarding ambient temperature and bus current as two distinct vectorsGradient descent calculation is carried out on the quantity, and the optimal honey source searching direction is folded when the quantity and the quantity are positioned at different positionsThe manner of determination of (2) is also different, FIG. 7 shows that different angular vectors are directed to +.>Multi-gradient descent relationship and calculation method.
Two parameters which affect the temperature change of the enclosed bus duct are known to be the most important, and samples of the ambient temperature and the bus current are acquired through the honey collection principle. The amplitude of the analog parameter change follows, the curve of which is shown in fig. 6. The most important task at present is to find out the position when the mixing change rate of two parameters is maximum, namely the optimal solution of multi-gradient descent, according to the simulation curve.
S24, maximizing、/>So that the minimum value of both is less than 0, the optimal temperature search direction +.>And optimal current search direction->Can make the two reach the purpose of descending at the same time, namely, the established optimal honey source searching direction +.>The process is as follows:
；
further, it is converted into:
；
by determining the optimal honey source searching direction when the temperature and current gradient change rate is maximumSuch that:
；
；
at this timeThe method is an optimal solution of the ambient temperature and the current change rate of the enclosed bus;
s25, optimal honey source searching directionAnd (3) carrying out normalization processing on the double-parameter gradient through a dual rule. If the temperature current gradient is not normalized, the smallest gradient element will be affected by the small norm gradient in the family, so there is a problem of whether the optimized gradient can be well balanced multi-parameter gradient descent. Setting a double-parameter search step size:
；
order the，/>，/>Are all within the defined r interval [0,1 ]]In (a) and (b); />Search step size for class i,/->Optimizing parameters for the j-th class of parallelism;
the temperature current parallel optimization parameters satisfy:
；
；
in the method, in the process of the invention,optimizing parameters for class i parallelism, +.>、/>Is->The temperature and current information parameters are input.
FIG. 8 is a graph of simulated comparison of the optimal mixing rate of change obtained from a multi-gradient descent swarm algorithm versus solutions obtained from other conventional algorithms for varying the rate of change of the enclosed bus temperature. Here the upper limit threshold for the enclosed bus temperature is set to 900 ℃.
In S3, the step of searching the optimal solution is as follows:
determining optimal honey source searching according to the temperature current multi-gradient descending bee colony mathematical model established in the step S2Further determining the optimal solution of the ambient temperature and the current change rate of the enclosed bus>The model is as follows:
；
in the method, in the process of the invention,、/>、/>、/>temperature current optimal honey source solution>、/>Inputting the mean value and variance of the optimized parameters, +.>After searching honey sources for the scout bees, the information distance of the honey sources in the neighborhood,when the number of the scout bees is n, the dimension of the input parameter after optimization is +.>At this time, the newly generated temperature may be resolved,when the number of the scout bees is n, the dimension of the input parameter after optimization is +.>At this time, the newly generated current may be resolved,、/>、/>、/>for temperature and current in n spy bees, < >>Upper and lower boundaries of past and present dimensions;
through the mathematical model, the optimal honey source searching directionIs determined and the optimal solution of the ambient temperature and the current change rate of the enclosed bus is further obtained ∈>。/>
Claims (6)
1. The closed bus temperature fault monitoring method based on the multi-gradient descent bee colony algorithm is characterized by comprising the following steps of:
s1, combining a multi-objective optimization process of a multi-gradient descent algorithm with an artificial bee colony algorithm, and establishing a multi-gradient descent bee colony algorithm model;
s2, collecting an environment temperature and current test sample of the closed bus, performing feature extraction and multi-objective optimization, and establishing a temperature current multi-gradient descent swarm mathematical model according to the multi-gradient descent swarm algorithm model in S1;
s3, searching an optimal honey source searching direction through the established temperature-current multi-gradient descending swarm mathematical model, and further obtaining an optimal solution of the environmental temperature and the current change rate of the enclosed bus, wherein the position of the optimal solution is most likely to be the temperature fault point of the enclosed bus.
2. The closed bus temperature fault monitoring method based on the multi-gradient descent swarm algorithm according to claim 1, wherein the method is characterized by comprising the following steps of: in the step S1, the step of establishing a multi-gradient descent bee colony algorithm model is as follows:
s11, the swarm types of the multi-gradient descent swarm algorithm model are a reconnaissance bee, an observation bee and a bee collection, a solution space of a target problem is defined to be M dimensions, the number of the reconnaissance bee, the number of the observation bee and the number of the bee collection are N, W, Q respectively, and the reconnaissance bee randomly searches for new honey source information in each period so as to meet the following formula:
；
where N is the number of scout bees, n=1, 2,3, … …, N, M is the spatial dimension of the target problem solution, m=1, 2,3, … …, M,in order to newly generate a possible solution in m dimension when the number of scout bees is n,/i>In order to solve the original m-dimensional solution space when the number of the scout bees is n>Is defined as that when the number of the scout bees is n, the newly generated number falls within the interval [ -1,1 ] in m dimension]K is a defined space coefficient, k is not equal to n;
s12, after the reconnaissance bees feed back the honey source information to the observation bees, the observation bees select one honey source from the honey sources according to probability, and the honey source set obtained by the reconnaissance bees is defined asThe honey source is a local optimal solution because of a single search mechanism of the scout bees, and the selection probability is as follows:
；
in the method, in the process of the invention,the last level of bee collected for the w-th observation bee may solve +.>Is adapted to the value of->The probability of selection for the w-th observation bee;
s13, selecting the optimal fitness value according to the selection probability of the observed bees and the related honey source information, namelyOptimal solution->，/>The honey source information of the previous period is set as +.>When honey source information->And (3) withUpdating the optimal honey source information when compared with the relative superiority, otherwise, discarding +.>The bee picking becomes a reconnaissance bee, search is continued in the current honey source neighborhood, and the search formula becomes:
；
in the method, in the process of the invention,in order to obtain the possible solution of m dimension newly generated when the number of bees is q, the weight of the bees is increased by->、/>When the number of the collected bees is q, the lower bound and the upper bound of m dimensions in the past and at present respectively, and r represents an interval [0,1]The random number on the table, k is not equal to q;
s14, establishing a multi-objective parameter honey collection optimizing system:
；
in the method, in the process of the invention,for optimal honey source solution, the drug is prepared from the following raw materials>Inputting honey source information parameters for the ith input;
s15, setting the optimal honey source searching direction asAccording to a multiple gradient decrease, so that +.>The method meets the following conditions:
；
；
in the method, in the process of the invention,after searching honey sources for the scout bees, the honey source information distance of the neighborhood;
s16, in order to obtain an optimal honey source searching direction, and enable all searching directions to be towards the optimal honey source searching direction after each new honey source searching, updating information, and converting the formula in S15 into:
；
the search is iterated continuously, so that the Pareto optimal solution of the problem can be obtained:
；
s17, if an optimal honey source searching direction existsMake->At the same time satisfyAt this time +.>Is a multi-objective problemIs the optimal solution of (a); if the optimal honey source searching direction cannot be found +.>Then the Pareto optimal solution does not exist, at which point the last honey source is the optimal honey source.
3. The closed bus temperature fault monitoring method based on the multi-gradient descent swarm algorithm according to claim 2, wherein the method is characterized by comprising the following steps of: the establishment of the multi-gradient descent bee colony algorithm model is that an optimal memory mechanism is established in honey source information acquisition through multi-gradient descent, and the process is represented by the following formula:
；
in the method, in the process of the invention,for bee colony data without optimal memory participation, < ->And (3) for the bee colony data in the participation of optimal memory, gamma is the input data of the optimal memory, and beta is the self-adaptive function value of the optimal honey source.
4. The closed bus temperature fault monitoring method based on the multi-gradient descent swarm algorithm according to claim 2, wherein the method is characterized by comprising the following steps of: in the step S2, the step of establishing a temperature current multi-gradient descending bee colony mathematical model is as follows:
s21, collecting and collecting the environmental temperature contacted by the aluminum alloy provided with the enclosed bus, wherein the environmental temperature is as follows:
；
the current collection set is as follows:
；
the highest temperature of the sample data isThe minimum temperature is->Maximum current of->Minimum current is +.>；
S22, collecting temperature and current、/>Extracting features, screening information, and setting class parameter +.>Initial center point +.>，/>，/>，/>Screening is performed on each sample data:
；
；
in the method, in the process of the invention,for the ith temperature parameter, +.>、/>Optimizing parameters for temperature and current parallelism, +.>、/>、/>、The temperature is respectively the highest temperature and the lowest temperature of the ambient temperature and the bus current;
s23, after data screening through feature extraction, setting the extracted temperature set as follows:
；
the current set is:
；
establishing a temperature-current parameter honey collection optimizing system:
；
；
in the method, in the process of the invention,、/>for optimal temperature and current source solution, < >>、/>Inputting a temperature current information parameter for the ith input;
s24, maximizing、/>So that the minimum value of both is less than 0, the optimal temperature search direction +.>And optimal current search direction->Can make the two reach the purpose of descending at the same time, namely, the established optimal honey source searching direction +.>The process is as follows:
；
further, it is converted into:
；
by determining the optimal honey source searching direction when the temperature and current gradient change rate is maximumSuch that:
；
；
at this timeThe method is an optimal solution of the ambient temperature and the current change rate of the enclosed bus;
s25, setting a double-parameter search step length:
；
order the，/>，/>Are all within the defined r interval [0,1 ]]In (a) and (b); />Search step size for class i,/->Optimizing parameters for the j-th class of parallelism;
the temperature current parallel optimization parameters satisfy:
；
；
in the method, in the process of the invention,optimizing parameters for class i parallelism, +.>、/>Is->The temperature and current information parameters are input.
5. The closed bus temperature fault monitoring method based on the multi-gradient descent swarm algorithm according to claim 4, wherein the method is characterized by comprising the following steps of: in the step S3, the step of searching the optimal solution is as follows:
determining optimal honey source searching according to the temperature current multi-gradient descending bee colony mathematical model established in the step S2Further determining the optimal solution of the ambient temperature and the current change rate of the enclosed bus>The model is as follows:
；
in the method, in the process of the invention,、/>、/>、/>respectively the temperature and the currentOptimal Honey Source solution->、/>Inputting the mean value and variance of the optimized parameters, +.>After searching honey sources for the scout bees, the information distance of the honey sources in the neighborhood is +.>When the number of the scout bees is n, the dimension of the input parameter after optimization is +.>In the meantime, the newly generated temperature may be solved, < ->When the number of the scout bees is n, the dimension of the input parameter after optimization is +.>In the meantime, the newly generated current may solve, < ->、、/>、/>For temperature and current in n spy bees, < >>Upper and lower boundaries of past and present dimensions;
through the mathematical model, the optimal honey source searching directionIs determined and the optimal solution of the ambient temperature and the current change rate of the enclosed bus is further obtained ∈>。
6. The closed bus temperature fault monitoring method based on the multi-gradient descent swarm algorithm according to claim 1, wherein the method is characterized by comprising the following steps of: in the step S3, simulation verification is carried out on the obtained optimal solution through MATLAB.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311152379.3A CN116882304B (en) | 2023-09-08 | 2023-09-08 | Closed bus temperature fault monitoring method based on multi-gradient descent bee colony algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311152379.3A CN116882304B (en) | 2023-09-08 | 2023-09-08 | Closed bus temperature fault monitoring method based on multi-gradient descent bee colony algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116882304A true CN116882304A (en) | 2023-10-13 |
| CN116882304B CN116882304B (en) | 2023-12-05 |
Family
ID=88260933
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311152379.3A Active CN116882304B (en) | 2023-09-08 | 2023-09-08 | Closed bus temperature fault monitoring method based on multi-gradient descent bee colony algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116882304B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109272137A (en) * | 2018-07-13 | 2019-01-25 | 兰州交通大学 | A kind of scheduling of resource optimization method based on the global artificial bee colony algorithm of intersection |
| CN113537439A (en) * | 2020-04-17 | 2021-10-22 | 中国石油化工股份有限公司 | Improved artificial bee colony optimization algorithm |
| WO2022090461A1 (en) * | 2020-10-30 | 2022-05-05 | Clearbox Ai Solutions S.R.L. | A computer-implemented method for generating local explanations and error estimates in machine learning models, as well as a decision support system and computer program implementing said method |
-
2023
- 2023-09-08 CN CN202311152379.3A patent/CN116882304B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109272137A (en) * | 2018-07-13 | 2019-01-25 | 兰州交通大学 | A kind of scheduling of resource optimization method based on the global artificial bee colony algorithm of intersection |
| CN113537439A (en) * | 2020-04-17 | 2021-10-22 | 中国石油化工股份有限公司 | Improved artificial bee colony optimization algorithm |
| WO2022090461A1 (en) * | 2020-10-30 | 2022-05-05 | Clearbox Ai Solutions S.R.L. | A computer-implemented method for generating local explanations and error estimates in machine learning models, as well as a decision support system and computer program implementing said method |
Non-Patent Citations (1)
| Title |
|---|
| 邓建球;赵建忠;陈洪;丛林虎;: "ABC算法优化SVR的磨损故障预测模型", 兵工自动化, no. 10 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116882304B (en) | 2023-12-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110220602B (en) | Switch cabinet overheating fault identification method | |
| CN108956111B (en) | Abnormal state detection method and detection system for mechanical part | |
| CN110971677B (en) | Electric power internet of things terminal equipment side channel safety monitoring method based on countermeasure reinforcement learning | |
| CN111080074B (en) | System service security situation element obtaining method based on network multi-feature association | |
| CN104237235A (en) | Rapid detection method based on near-infrared imaging technology for food-borne pathogens | |
| CN116628616A (en) | Data processing method and system for high-power charging energy | |
| CN114374541A (en) | Abnormal network flow detector generation method based on reinforcement learning | |
| CN116882304B (en) | Closed bus temperature fault monitoring method based on multi-gradient descent bee colony algorithm | |
| CN118051083A (en) | Intelligent temperature control method and device for heating wire | |
| CN113536662B (en) | Electronic transformer error state prediction method based on firefly optimized LightGBM algorithm | |
| CN114444689A (en) | Transformer temperature detection and analysis method | |
| CN117918327A (en) | Weeding robot control method, weeding robot control device, weeding robot control medium and weeding robot | |
| CN107426141A (en) | Malicious code protection method, system and monitoring device | |
| CN117007883A (en) | Multipoint temperature testing system for photoelectric coupler | |
| CN116089865A (en) | Method for classifying and identifying OPGW abnormal vibration based on improved grid search random forest algorithm | |
| WO2020253485A1 (en) | Dbscan-based outer surface temperature measurement method for cracking furnace tube | |
| CN116398868A (en) | Wall temperature anomaly identification method and system based on virtual partition of thermal power station furnace tube | |
| Liu | A computer network intrusion detection technology based on improved neural network algorithm | |
| CN110514566B (en) | Novel smoke dust measuring device and method | |
| CN208606794U (en) | A kind of oil-immersed power transformer fault detection system | |
| CN107204741A (en) | A kind of method and apparatus for determining ambient parameter | |
| CN111879246A (en) | Tree diameter monitoring system and method based on fiber bragg grating sensor | |
| Mwasiagi et al. | The use of k-means and artificial neural network to classify cotton lint | |
| CN117870779B (en) | Remote wireless intelligent monitoring method and system for automation equipment | |
| Lu et al. | Development of the Abnormal Tension Pattern Recognition Module for Twisted Yarn Based on Deep Learning Edge Computing. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |