CN101739025A - Immunity genetic algorithm and DSP failure diagnostic system based thereon - Google Patents

Immunity genetic algorithm and DSP failure diagnostic system based thereon Download PDF

Info

Publication number
CN101739025A
CN101739025A CN200910228954A CN200910228954A CN101739025A CN 101739025 A CN101739025 A CN 101739025A CN 200910228954 A CN200910228954 A CN 200910228954A CN 200910228954 A CN200910228954 A CN 200910228954A CN 101739025 A CN101739025 A CN 101739025A
Authority
CN
China
Prior art keywords
dsp
probability
genetic algorithm
fault
algorithm
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
CN200910228954A
Other languages
Chinese (zh)
Inventor
周雪松
田密
马幼捷
宋代春
权博
李圣明
程德树
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.)
Tianjin University of Technology
Original Assignee
Tianjin University of Technology
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 Tianjin University of Technology filed Critical Tianjin University of Technology
Priority to CN200910228954A priority Critical patent/CN101739025A/en
Publication of CN101739025A publication Critical patent/CN101739025A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention discloses a DSP failure diagnostic system based on an immunity genetic algorithm, which consists of an FTU, a RTU, a DSP and an industrial controlling upper computer of a failure signal output display unit. The immunity genetic algorithm comprises the steps of: preparing a parameter code; setting an initial group; designing an adaptability function; setting a matching set; setting a controlling parameter; and solving iteration. The application of DSP failure diagnosis based on the algorithm comprises the steps of: taking the DSP as a core control process system; taking a discrete signal as an input signal of the DSP core control process system when the FTU and the RTU monitor the variable quantity of failure power frequency; and transferring a result which is obtained by the algorithm into a failure message in a practical system through the DSP core control process system to be showed on a monitoring upper computer. The invention has the advantages of: (1) solving the problem of the failure diagnosis of an electrical power system when a protection message is imperfect; and (2) inducing the application of an immune system-matching set to improve the whole searching capability of the algorithm.

Description

A kind of immune genetic algorithm reaches the DSP fault diagnosis system based on this algorithm
(1) technical field:
The invention belongs to the power system failure diagnostic technical field, particularly a kind of immune genetic algorithm reaches the DSP fault diagnosis system based on this algorithm.
(2) background technology:
Reliability is one of basic demand of electric system.But because weather, influence of various factors such as artificial, the fault of electric system is inevitable.When electric system is broken down, just require operations staff's normal operation of failure judgement reason, excision fault element and recovery system rapidly, so that reduce damage, guarantee to power reliably to user security to power equipment.
When electric system was broken down, tripping, malfunction, warning information may take place and may go out active, distortion in transmission course in protection and isolating switch; Fault element has a plurality of.These make that all physical fault is complicated, and the individual is difficult to make at short notice right judgement.Fault diagnosis system can help the yardman to find out fault element rapidly, is the prerequisite that restores electricity fast.
(3) summary of the invention
The object of the present invention is to provide a kind of immune genetic algorithm and based on the DSP fault diagnosis system of this algorithm, it makes full use of the characteristic of immune genetic algorithm, the problem of fault diagnosis is changed into 0,1 integer programming problem, make full use of the senior decision making function of computing machine and become the current digital signal processor spare (DSP) of hot spot technology newly and do real-time processing, realization is to the diagnosis of the current power system failure, make it quick identification fault zone and element, analysis causes the reason of fault, and recovery system normally moves.
Technical scheme of the present invention: a kind of immune genetic algorithm reaches the DSP fault diagnosis system based on this algorithm, it is characterized in that this system is made up of fault-signal detecting unit feed line automatization terminal (FTU) and remote-terminal unit (RTU), fault-signal processing unit digital signal processor (DSP), fault-signal output display unit industry control host computer; The output terminal of the output terminal of said fault-signal detecting unit feed line automatization terminal (FTU) and remote-terminal unit (RTU) all is connected with fault-signal processing unit digital signal processor (DSP), and the output terminal of fault-signal processing unit digital signal processor (DSP) connects fault-signal output display unit industry control host computer.
A kind of immune genetic algorithm of above-mentioned DSP fault diagnosis system, so it is characterized in that constituting by following steps:
(1) parameter coding in the making algorithm;
(2) setting of initial population;
(3) design of fitness function;
(4) setting of set of matches;
(5) setting of controlled variable;
(6) iterative.
Said parameter coding adopts binary coding method in the above-mentioned steps (1), after system breaks down, the FTU or the RTU that are installed on each block switch and interconnection switch place can monitor fault power frequency variable, be uploaded to the main station failure signal with its setting valve after relatively, and this signal is 0,1 discrete binary signal.
The setting of said initial population is that genetic algorithm is operated the evolution that colony carries out in the above-mentioned steps (2), needs to prepare the initial population data of the initial search point of expression.The generation of initial population is to be benchmark with reflection system failure signal, and system can produce initial population at random near this state.
The design of said fitness function is also referred to as the evaluation to hereditary ideal adaptation degree in the above-mentioned steps (3), is meant in genetic algorithm, determines that with the size of ideal adaptation degree this individuality is genetic to the probability in the colony of future generation.Objective function of the present invention is:
F ( S ) = W - f Mp × Σ i = 1 n i Σ l = 1 n c | Mp l - Mp l , i * ( S ) | - f Fp × Σ j = 1 n j Σ l = 1 n c | Fp l - Fp l , j * ( S ) | -
f Sp × Σ k = 1 n k Σ l = 1 n c | Sp l - Sp l , k * ( S ) | - f Rp × Σ h = 1 n h Rp h *
In the formula:
S: the element state vector-be 0 just often, during fault 1 (n dimension, n is a parts number);
n c: the number of isolating switch;
n i: the main protection number;
n j: the first back-up protection number;
n k: the second back-up protection number;
n h: the reclosing number;
Mp l: l isolating switch is at the main protection state of action period;
Mp L, i *: l the isolating switch function definite according to i main protection operating principle;
Fp l: l isolating switch is at first back-up protection state of action period;
Fp L, j *: according to j l the isolating switch number that the first back-up protection operating principle is definite;
Sp l: l isolating switch is at second back-up protection state of action period;
Sp L, k *: according to k l the isolating switch number that the second back-up protection operating principle is definite;
f MP: the main protection contribution factor;
f Fp: the first back-up protection contribution factor;
f Sp: the second back-up protection contribution factor;
f Rp *: the reclosing contribution factor;
R Ph *: the reclosing expectation state;
W (gets W=10 for any given very big positive number in the formula 6, be used to guarantee that F (S) perseverance just is). the task of algorithm is exactly the maximal value of asking under this fitness function.
In the above-mentioned steps (4) setting of said set of matches be meant immune genetic algorithm in the colony in twos the set of matches that produces of competition operate, thereby the Local Search that in set of matches, carries out be begin to search for by per generation before more new settings intersection, variation probability to its intersect, making a variation produces new population.
The setting of said controlled variable is meant in the above-mentioned steps (5) needs definite parameter value in the genetic algorithm, comprise the long l of string, group size n, crossover probability p c, the variation Probability p mThe string length of parameter is that group size was set at for 100 generations by the decision of the number of system break road device among the present invention, and crossover probability p cWith the variation Probability p mWe set before per generation produces set of matches and upgrade, and promptly crossover probability and variation probability are variable.From the calculating mechanism of genetic algorithm as can be known, in iteration early stage, crossover probability should be bigger, and making a variation probability should be smaller, to guarantee steadily carrying out of computation process.In the iteration later stage, the string of Xie Qunzhong tends towards stability, may converge on locally optimal solution, the effect that intersect this moment reduces, and the probability of its generation can reduce, and the probability of variation should givenly get more greatly, jump out locally optimal solution so that have an opportunity, enter new search volume.The crossover probability that the present invention adopts with the formula that the increase of probability with iterations that make a variation changes is:
Figure G2009102289547D00031
In the formula:
T: repeatedly be with number of times;
P c 0: the initial value of crossover probability;
P c t: crossover probability the t time repeatedly with the time value;
P m 0: the initial value of variation probability;
P m t: variation probability the t time repeatedly with the time value;
Mgen: maximum allow repeatedly be with number of times;
Crossover probability is linear decrease with the increase of iterations, and the variation probability is linear increment with the increase of iterations, and in the present invention, in the iteration early stage of the first generation, crossover probability and variation probability are taken as 0.9 and 0.001 respectively.
Said iterative is meant in the above-mentioned steps (6): the output of immune genetic algorithm net result is the iterative by big twice on the whole.For the first time be that the set of matches that is produced by the system of selection of competing in twos in per generation is carried out the n/2 iteration in step, per step iteration is carried out cross and variation according to the controlled variable of setting to it to the individual in twos of picked at random from set of matches, separates for two that will newly produce then and puts into Xie Qun of future generation; Be after finishing for the second time, judge whether final generation,, change follow-on iterative over to if be less than final generation when the first step iteration of former generation.At first the crossover probability and the variation probability in this generation are set, and produced set of matches, at this moment repeat the previous generation's a iterative process by the selection way of intersecting in twos.Jump out circulation and output final structure and fitness value thereof through algorithm behind the second iteration of setting algebraically.
A kind of application of the DSP fault diagnosis based on this algorithm, it is characterized in that adopting DSP as the core control processing system, when monitoring fault power frequency variable by feed line automatization terminal in the system (FTU) and remote-terminal unit (RTU), the discrete signal that will obtain after relatively with its setting valve is as the input signal of DSP core control system, this signal is also as the input signal that is embedded in immune genetic algorithm wherein, the failure message that the final net result that algorithm is obtained by DSP core control processing system is converted in the real system is reflected to above the monitoring host computer, and correlation engineering personnel can carry out maintenance platoon to fault according to the information that system reflected and remove.
The immune genetic algorithm of the said DSP fault diagnosis of the present invention principle of work: at first the fault-signal in the system is carried out parameter coding, parameter coding is that the application genetic algorithm is the matter of utmost importance that will solve, a committed step when also being the design genetic algorithm.Coding method is except having determined individual chromosome spread pattern, coding/decoding method when he has also determined the individual phenotype that changes to solution space from the search volume to the genotype, coding method also have influence on the operational method of genetic operators such as the intersection of back and mutation operator.
The setting of initial population is that genetic algorithm is operated the evolution that colony carries out, and need prepare the initial population data that some represent initial search point to it.General initial population be set with two kinds of methods: (1) manages to hold the distribution range of optimum solution take up space in whole problem space according to the intrinsic knowledge of problem, the setting initial population.(2) elder generation generates the individuality of some at random, therefrom chooses best individuality and is added in the initial population.The continuous iteration of this process is up to reaching predetermined scale.
The design of fitness function also is called the evaluation to hereditary ideal adaptation degree, is meant in genetic algorithm, determines that with the size of ideal adaptation degree this individuality is genetic to the probability in the colony of future generation.Individual fitness is genetic to follow-on probability with this individuality and is directly proportional.By the ability assessment to hereditary ideal adaptation environment, as the foundation of selection operation, it is to be formed by the objective function conversion.Unique requirement is that its result is a nonnegative value to fitness function.The change of scale of fitness is certain mapping transformation to the objective function codomain, can overcome the prematurity convergence and roam phenomenon at random.
The introducing of set of matches is for immune thought is incorporated in the genetic algorithm, thereby constitutes the genetic Optimization Algorithm based on immunity.This algorithm synthesis the advantage of immune system and genetic algorithm.Genetic algorithm is a kind of massive parallelism global search algorithm of finding the solution problem in essence, and immune system is then utilized the heuristic Local Search that instructs of particular problem, and can finely tune in solution space.
The setting of controlled variable is meant: need in the genetic algorithm to determine some parameter values, mainly contain the long l of string, group size n, crossover probability p c, the variation Probability p mDeng, very big to the performance of genetic algorithms influence.At present parameter is according to circumstances adjusted to change and study often, and the general parameter area of determining is: n=20～200, p c=0.5～1.0, p m=0～0.05.
DSP is meant as the core control processing system and adopts DSP as accepting the external fault signal and these input signals being carried out analyzing and processing that handle the reason of back output fault, thereby can make maintenance personal's maintenance system fault as early as possible, recovery system normally moves.
Superiority of the present invention and technique effect are: the analytic model that 1. proposes to describe with unconfined Zero-one integer programming problem power system failure diagnostic, this model has solved the power system failure diagnostic problem when protection information is imperfect to a certain extent, combine with the recognition methods of power supply interrupted district after fault is calmed down, can onlinely be applied to the actual electrical system; 2. constantly in whole iterative process with traditional genetic algorithm crossover probability and variation probability different be, it is linear decrease with the increase of iterations that the present invention makes crossover probability, the variation probability is linear increment with the increase of iterations, jump out locally optimal solution so that have an opportunity, enter new search volume; 3. introduce the application of immune system-set of matches, thereby in the search procedure in per generation, can instruct Local Search, in solution space, finely tune, thereby improved the whole search capability of algorithm during evolution; 4. we utilize DSP control processing system and computing machine data computation and data-handling capacity at a high speed in actual applications, can improve the reliability of this fault diagnosis system greatly, have wide market application prospect; 5. will apply to based on the genetic algorithm of immunity in the practical power systems, handle a large amount of Operation of Electric Systems information in real time, the zone that identification is broken down, further tracing trouble element, thereby increase substantially the reliability and the economy of power supply, and, guaranteed the stable operation of relevant power equipment, circuit from having reduced the loss that short circuit malfunction brings to the full extent; 6. fitness function has been considered the various situations of electric system comprehensively; not only considered the influence of reclosing; but also according to main protection, first back-up protection and first back-up protection Different Effects refinement in addition for fault; different protections has been carried out weighting by contribution factor, tallied with the actual situation more.
(4) description of drawings
Fig. 1 is that the related a kind of immune genetic algorithm of the present invention reaches the overall work structural representation based on the DSP fault diagnosis system of this algorithm;
Fig. 2 is the related a kind of immune genetic algorithm handling failure signal workflow diagram of the present invention.
(5) embodiment
Embodiment: a kind of immune genetic algorithm reaches the DSP fault diagnosis system based on this algorithm, (see figure 1) is characterized in that this system is made up of fault-signal detecting unit feed line automatization terminal (FTU) and remote-terminal unit (RTU), fault-signal processing unit digital signal processor (DSP), fault-signal output display unit industry control host computer; The output terminal of the output terminal of said fault-signal detecting unit feed line automatization terminal (FTU) and remote-terminal unit (RTU) all is connected with fault-signal processing unit digital signal processor (DSP), and the output terminal of fault-signal processing unit digital signal processor (DSP) connects fault-signal output display unit industry control host computer.
A kind of immune genetic algorithm (see figure 2) of above-mentioned DSP fault diagnosis system, so it is characterized in that constituting by following steps:
(1) parameter coding in the making algorithm;
(2) setting of initial population;
(3) design of fitness function;
(4) setting of set of matches;
(5) setting of controlled variable;
(6) iterative.
Said parameter coding adopts binary coding method in the above-mentioned steps (1), after system breaks down, the FTU or the RTU that are installed on each block switch and interconnection switch place can monitor fault power frequency variable, be uploaded to the main station failure signal with its setting valve after relatively, and this signal is 0,1 discrete binary signal.
The setting of said initial population is that genetic algorithm is operated the evolution that colony carries out in the above-mentioned steps (2), needs to prepare the initial population data of the initial search point of expression.The generation of initial population is to be benchmark with reflection system failure signal, and system can produce initial population at random near this state.
The design of said fitness function is also referred to as the evaluation to hereditary ideal adaptation degree in the above-mentioned steps (3), is meant in genetic algorithm, determines that with the size of ideal adaptation degree this individuality is genetic to the probability in the colony of future generation.Objective function of the present invention is:
F ( S ) = W - f Mp × Σ i = 1 n i Σ l = 1 n c | Mp l - Mp l , i * ( S ) | - f Fp × Σ j = 1 n j Σ l = 1 n c | Fp l - Fp l , j * ( S ) | -
f Sp × Σ k = 1 n k Σ l = 1 n c | Sp l - Sp l , k * ( S ) | - f Rp × Σ h = 1 n h Rp h *
In the formula:
S: the element state vector-be 0 just often, during fault 1 (n dimension, n is a parts number);
n c: the number of isolating switch;
n i: the main protection number;
n j: the first back-up protection number;
n k: the second back-up protection number;
n h: the reclosing number;
Mp l: l isolating switch is at the main protection state of action period;
Mp L, i *: l the isolating switch function definite according to i main protection operating principle;
Fp l: l isolating switch is at first back-up protection state of action period;
Fp L, j *: according to j l the isolating switch number that the first back-up protection operating principle is definite;
Sp l: l isolating switch is at second back-up protection state of action period;
Sp L, k *: according to k l the isolating switch number that the second back-up protection operating principle is definite;
f MP: the main protection contribution factor;
f Fp: the first back-up protection contribution factor;
f Sp: the second back-up protection contribution factor;
f Rp *: the reclosing contribution factor;
R Ph *: the reclosing expectation state;
W (gets W=10 for any given very big positive number in the formula 6, be used to guarantee that F (S) perseverance just is). the task of algorithm is exactly the maximal value of asking under this fitness function.
In the above-mentioned steps (4) setting of said set of matches be meant immune genetic algorithm in the colony in twos the set of matches that produces of competition operate, thereby the Local Search that in set of matches, carries out be begin to search for by per generation before more new settings intersection, variation probability to its intersect, making a variation produces new population.
The setting of said controlled variable is meant in the above-mentioned steps (5) needs definite parameter value in the genetic algorithm, comprise the long l of string, group size n, crossover probability p c, the variation Probability p mThe string length of parameter is that group size was set at for 100 generations by the decision of the number of system break road device among the present invention, and crossover probability p cWith the variation Probability p mWe set before per generation produces set of matches and upgrade, and promptly crossover probability and variation probability are variable.From the calculating mechanism of genetic algorithm as can be known, in iteration early stage, crossover probability should be bigger, and making a variation probability should be smaller, to guarantee steadily carrying out of computation process.In the iteration later stage, the string of Xie Qunzhong tends towards stability, may converge on locally optimal solution, the effect that intersect this moment reduces, and the probability of its generation can reduce, and the probability of variation should givenly get more greatly, jump out locally optimal solution so that have an opportunity, enter new search volume.The crossover probability that the present invention adopts with the formula that the increase of probability with iterations that make a variation changes is:
Figure G2009102289547D00081
Figure G2009102289547D00082
In the formula:
T: repeatedly be with number of times;
P c 0: the initial value of crossover probability;
P c t: crossover probability the t time repeatedly with the time value;
P m 0: the initial value of variation probability;
P m t: variation probability the t time repeatedly with the time value;
Mgen: maximum allow repeatedly be with number of times;
Crossover probability is linear decrease with the increase of iterations, and the variation probability is linear increment with the increase of iterations, and in the present invention, in the iteration early stage of the first generation, crossover probability and variation probability are taken as 0.9 and 0.001 respectively.
Said iterative is meant in the above-mentioned steps (6): the output of immune genetic algorithm net result is the iterative by big twice on the whole.For the first time be that the set of matches that is produced by the system of selection of competing in twos in per generation is carried out the n/2 iteration in step, per step iteration is carried out cross and variation according to the controlled variable of setting to it to the individual in twos of picked at random from set of matches, separates for two that will newly produce then and puts into Xie Qun of future generation; Be after finishing for the second time, judge whether final generation,, change follow-on iterative over to if be less than final generation when the first step iteration of former generation.At first the crossover probability and the variation probability in this generation are set, and produced set of matches, at this moment repeat the previous generation's a iterative process by the selection way of intersecting in twos.Jump out circulation and output final structure and fitness value thereof through algorithm behind the second iteration of setting algebraically.
A kind of application of the DSP fault diagnosis based on this algorithm, it is characterized in that adopting DSP as the core control processing system, when monitoring fault power frequency variable by feed line automatization terminal in the system (FTU) and remote-terminal unit (RTU), the discrete signal that will obtain after relatively with its setting valve is as the input signal of DSP core control system, this signal is also as the input signal that is embedded in immune genetic algorithm wherein, the failure message that the final net result that algorithm is obtained by DSP core control processing system is converted in the real system is reflected to above the monitoring host computer, and correlation engineering personnel can carry out maintenance platoon to fault according to the information that system reflected and remove.

Claims (9)

1. an immune genetic algorithm and based on the DSP fault diagnosis system of this algorithm is characterized in that this system exports display unit industry control host computer by fault-signal detecting unit feed line automatization terminal (FTU) and remote-terminal unit (RTU), fault-signal processing unit digital signal processor (DSP), fault-signal and forms; The output terminal of the output terminal of said fault-signal detecting unit feed line automatization terminal (FTU) and remote-terminal unit (RTU) all is connected with fault-signal processing unit digital signal processor (DSP), and the output terminal of fault-signal processing unit digital signal processor (DSP) connects fault-signal output display unit industry control host computer.
2. the immune genetic algorithm of a DSP fault diagnosis system, so it is characterized in that constituting by following steps:
(1) parameter coding in the making algorithm;
(2) setting of initial population;
(3) design of fitness function;
(4) setting of set of matches;
(5) setting of controlled variable;
(6) iterative.
3. according to the immune genetic algorithm of the said a kind of DSP fault diagnosis system of claim 2, it is characterized in that being that the parameter coding in the said step (1) adopts binary coding method, after system breaks down, the FTU or the RTU that are installed on each block switch and interconnection switch place can monitor fault power frequency variable, be uploaded to the main station failure signal with its setting valve after relatively, and this signal is 0,1 discrete binary signal.
4. according to the immune genetic algorithm of the said a kind of DSP fault diagnosis system of claim 2, it is characterized in that being that the setting of the initial population in the said step (2) is that genetic algorithm is operated the evolution that colony carries out, need to prepare the initial population data of the initial search point of expression.The generation of initial population is to be benchmark with reflection system failure signal, and system can produce initial population at random near this state.
5. according to the immune genetic algorithm of the said a kind of DSP fault diagnosis system of claim 2, the design that it is characterized in that being the fitness function in the said step (3) is also referred to as the evaluation to hereditary ideal adaptation degree, be meant in genetic algorithm, determine that with the size of ideal adaptation degree this individuality is genetic to the probability in the colony of future generation.Objective function of the present invention is:
F ( S ) = W - f Mp × Σ i = 1 n i Σ l = 1 n c | Mp l - Mp l , i * ( S ) | - f Fp × Σ j = 1 n j Σ l = 1 n c | Fp l - Fp l , j * ( S ) | -
f Sp × Σ k = 1 n k Σ l = 1 n c | Sp l - Sp l , k * ( S ) | - f Rp × Σ h = 1 n h Rp h *
In the formula:
S: element state vector one is 0 just often, is 1 (n dimension, n is a parts number) during fault;
n c: the number of isolating switch;
n i: the main protection number;
n j: the first back-up protection number;
n k: the second back-up protection number;
n h: the reclosing number;
Mp l: l isolating switch is at the main protection state of action period;
Mp L, i *: l the isolating switch function definite according to i main protection operating principle;
Fp l: l isolating switch is at first back-up protection state of action period;
Fp L, j *: according to j l the isolating switch number that the first back-up protection operating principle is definite;
Sp l: l isolating switch is at second back-up protection state of action period;
Sp L, k *: according to k l the isolating switch number that the second back-up protection operating principle is definite;
f MP: the main protection contribution factor;
f Fp: the first back-up protection contribution factor;
f Sp: the second back-up protection contribution factor;
f Rp *: the reclosing contribution factor;
R Ph *: the reclosing expectation state;
W (gets W=10 for any given very big positive number in the formula 6, be used to guarantee that F (S) perseverance just is). the task of algorithm is exactly the maximal value of asking under this fitness function.
6. according to the immune genetic algorithm of the said a kind of DSP fault diagnosis system of claim 2, the setting that it is characterized in that being the set of matches in the said step (4) is meant that immune genetic algorithm operates the set of matches that competition in twos in the colony produces, thus the Local Search that in set of matches, carries out be begin to search for by per generation before more new settings intersection, variation probability to its intersect, making a variation produces new population.
7. according to the immune genetic algorithm of the said a kind of DSP fault diagnosis system of claim 2, it is characterized in that the setting that is the controlled variable in the said step (5) is meant the parameter value that need determine in the genetic algorithm, comprises the long l of string, group size n, crossover probability p c, the variation Probability p mThe string length of parameter is that group size was set at for 100 generations by the decision of the number of system break road device among the present invention, and crossover probability p cWith the variation Probability p mWe set before per generation produces set of matches and upgrade, and promptly crossover probability and variation probability are variable.From the calculating mechanism of genetic algorithm as can be known, in iteration early stage, crossover probability should be bigger, and making a variation probability should be smaller, to guarantee steadily carrying out of computation process.In the iteration later stage, the string of Xie Qunzhong tends towards stability, may converge on locally optimal solution, the effect that intersect this moment reduces, and the probability of its generation can reduce, and the probability of variation should givenly get more greatly, jump out locally optimal solution so that have an opportunity, enter new search volume.The crossover probability that the present invention adopts with the formula that the increase of probability with iterations that make a variation changes is:
Figure F2009102289547C00031
Figure F2009102289547C00032
In the formula:
T: repeatedly be with number of times;
P c 0: the initial value of crossover probability;
P c t: crossover probability the t time repeatedly with the time value;
P m 0: the initial value of variation probability;
P m t: variation probability the t time repeatedly with the time value;
Mgen: maximum allow repeatedly be with number of times;
Crossover probability is linear decrease with the increase of iterations, and the variation probability is linear increment with the increase of iterations, and in the present invention, in the iteration early stage of the first generation, crossover probability and variation probability are taken as 0.9 and 0.001 respectively.
8. according to the immune genetic algorithm of the said a kind of DSP fault diagnosis system of claim 2, it is characterized in that being that the iterative in the said step (6) is meant: the output of immune genetic algorithm net result is the iterative by big twice on the whole.For the first time be that the set of matches that is produced by the system of selection of competing in twos in per generation is carried out the n/2 iteration in step, per step iteration is carried out cross and variation according to the controlled variable of setting to it to the individual in twos of picked at random from set of matches, separates for two that will newly produce then and puts into Xie Qun of future generation; Be after finishing for the second time, judge whether final generation,, change follow-on iterative over to if be less than final generation when the first step iteration of former generation.At first the crossover probability and the variation probability in this generation are set, and produced set of matches, at this moment repeat the previous generation's a iterative process by the selection way of intersecting in twos.Jump out circulation and output final structure and fitness value thereof through algorithm behind the second iteration of setting algebraically.
9. application based on the DSP fault diagnosis of this algorithm, it is characterized in that adopting DSP as the core control processing system, when monitoring fault power frequency variable by feed line automatization terminal in the system (FTU) and remote-terminal unit (RTU), the discrete signal that will obtain after relatively with its setting valve is as the input signal of DSP core control system, this signal is also as the input signal that is embedded in immune genetic algorithm wherein, the failure message that the final net result that algorithm is obtained by DSP core control processing system is converted in the real system is reflected to above the monitoring host computer, and correlation engineering personnel can carry out maintenance platoon to fault according to the information that system reflected and remove.
CN200910228954A 2009-12-03 2009-12-03 Immunity genetic algorithm and DSP failure diagnostic system based thereon Pending CN101739025A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN200910228954A CN101739025A (en) 2009-12-03 2009-12-03 Immunity genetic algorithm and DSP failure diagnostic system based thereon

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN200910228954A CN101739025A (en) 2009-12-03 2009-12-03 Immunity genetic algorithm and DSP failure diagnostic system based thereon

Publications (1)

Publication Number Publication Date
CN101739025A true CN101739025A (en) 2010-06-16

Family

ID=42462593

Family Applications (1)

Application Number Title Priority Date Filing Date
CN200910228954A Pending CN101739025A (en) 2009-12-03 2009-12-03 Immunity genetic algorithm and DSP failure diagnostic system based thereon

Country Status (1)

Country Link
CN (1) CN101739025A (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102435893A (en) * 2011-11-04 2012-05-02 国电南京自动化股份有限公司 Oil-immersed transformer fault diagnosis method based on self-adaptive genetic algorithm
CN102645615A (en) * 2012-04-26 2012-08-22 中国人民解放军海军工程大学 Marine electric power system fault diagnosis method based on quantum genetic algorithm
CN102801351A (en) * 2012-08-20 2012-11-28 武汉大学 Optimal space vector PWM control method for three-phase inverter based on immune algorithm
CN105634821A (en) * 2016-01-13 2016-06-01 上海金智晟东电力科技有限公司 FA execution condition evaluation method and system based on mirror image message
CN105843072A (en) * 2016-03-22 2016-08-10 华南理工大学 Wastewater treatment energy conservation optimization method based on improved local search immune genetic algorithm
CN106803101A (en) * 2016-12-30 2017-06-06 北京交通大学 Odometer method for diagnosing faults based on HMM
CN108734349A (en) * 2018-05-15 2018-11-02 国网山东省电力公司菏泽供电公司 Distributed generation resource addressing constant volume optimization method based on improved adaptive GA-IAGA and system
CN109214090A (en) * 2018-09-07 2019-01-15 哈尔滨工业大学 Digital microcurrent-controlled failure of chip restorative procedure based on improved adaptive GA-IAGA
CN110085027A (en) * 2019-03-28 2019-08-02 中国公路工程咨询集团有限公司 A kind of decomposition method of large-scale road network group traffic flow guidance task
CN111681607A (en) * 2020-08-17 2020-09-18 武汉精立电子技术有限公司 Gamma adjusting method and system based on genetic algorithm
CN112172991A (en) * 2020-09-18 2021-01-05 苏州三六零智能安全科技有限公司 Balance car fault positioning method, equipment, storage medium and device
CN113740650A (en) * 2021-09-06 2021-12-03 集美大学 Ship power system fault detection method, terminal device and storage medium

Cited By (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102435893B (en) * 2011-11-04 2013-10-09 国电南京自动化股份有限公司 Oil-immersed transformer fault diagnosis method based on self-adaptive genetic algorithm
CN102435893A (en) * 2011-11-04 2012-05-02 国电南京自动化股份有限公司 Oil-immersed transformer fault diagnosis method based on self-adaptive genetic algorithm
CN102645615A (en) * 2012-04-26 2012-08-22 中国人民解放军海军工程大学 Marine electric power system fault diagnosis method based on quantum genetic algorithm
CN102645615B (en) * 2012-04-26 2014-04-02 中国人民解放军海军工程大学 Marine electric power system fault diagnosis method based on quantum genetic algorithm
CN102801351A (en) * 2012-08-20 2012-11-28 武汉大学 Optimal space vector PWM control method for three-phase inverter based on immune algorithm
CN102801351B (en) * 2012-08-20 2014-12-31 武汉大学 Optimal space vector PWM control method for three-phase inverter based on immune algorithm
CN105634821A (en) * 2016-01-13 2016-06-01 上海金智晟东电力科技有限公司 FA execution condition evaluation method and system based on mirror image message
CN105634821B (en) * 2016-01-13 2017-07-04 上海金智晟东电力科技有限公司 A kind of FA implementation status evaluation method and system based on mirror image message
CN105843072B (en) * 2016-03-22 2019-01-18 华南理工大学 Sewage treatment energy conservation optimizing method based on the immune genetic algorithm for improving local search
CN105843072A (en) * 2016-03-22 2016-08-10 华南理工大学 Wastewater treatment energy conservation optimization method based on improved local search immune genetic algorithm
CN106803101A (en) * 2016-12-30 2017-06-06 北京交通大学 Odometer method for diagnosing faults based on HMM
CN106803101B (en) * 2016-12-30 2019-11-22 北京交通大学 Odometer method for diagnosing faults based on Hidden Markov Model
CN108734349A (en) * 2018-05-15 2018-11-02 国网山东省电力公司菏泽供电公司 Distributed generation resource addressing constant volume optimization method based on improved adaptive GA-IAGA and system
CN108734349B (en) * 2018-05-15 2020-11-17 国网山东省电力公司菏泽供电公司 Improved genetic algorithm-based distributed power supply location and volume optimization method and system
CN109214090A (en) * 2018-09-07 2019-01-15 哈尔滨工业大学 Digital microcurrent-controlled failure of chip restorative procedure based on improved adaptive GA-IAGA
CN109214090B (en) * 2018-09-07 2022-08-30 哈尔滨工业大学 Digital microfluidic chip fault repairing method based on improved genetic algorithm
CN110085027A (en) * 2019-03-28 2019-08-02 中国公路工程咨询集团有限公司 A kind of decomposition method of large-scale road network group traffic flow guidance task
CN111681607A (en) * 2020-08-17 2020-09-18 武汉精立电子技术有限公司 Gamma adjusting method and system based on genetic algorithm
CN112172991A (en) * 2020-09-18 2021-01-05 苏州三六零智能安全科技有限公司 Balance car fault positioning method, equipment, storage medium and device
CN113740650A (en) * 2021-09-06 2021-12-03 集美大学 Ship power system fault detection method, terminal device and storage medium
CN113740650B (en) * 2021-09-06 2023-09-19 集美大学 Ship electric power system fault detection method, terminal equipment and storage medium

Similar Documents

Publication Publication Date Title
CN101739025A (en) Immunity genetic algorithm and DSP failure diagnostic system based thereon
Hu et al. Short term electric load forecasting model and its verification for process industrial enterprises based on hybrid GA-PSO-BPNN algorithm—A case study of papermaking process
CN103995215B (en) A kind of smart power grid fault diagnostic method based on multi-level feedback adjustment
CN201666923U (en) Device for processing DSP failure diagnosis of electric power system by using immune genetic algorithm
Tao et al. Joint optimization of wind turbine micrositing and cabling in an offshore wind farm
CN112884008B (en) Prediction evaluation method and device for running state of power information acquisition system
CN106405319A (en) Rough set electric power system fault diagnosis method based on heuristic information
Gao et al. Application of AI techniques in monitoring and operation of power systems
Jalilpoor et al. Resilient energy management incorporating energy storage system and network reconfiguration: A framework of cyber‐physical system
CN111917111B (en) Method, system, equipment and storage medium for online evaluation of distributed photovoltaic power supply acceptance capacity of power distribution network
CN106097137A (en) A kind of anti-electricity-theft monitoring system of differentiation based on power information acquisition system
CN110994594B (en) Cross-zone standby configuration method and device of power system
CN112327098A (en) Power distribution network fault section positioning method based on low-voltage distribution network comprehensive monitoring unit
CN113850017A (en) System-level fault analysis system and method based on power flow change map
CN117557047A (en) Power distribution equipment operation and maintenance optimization method and system based on deep reinforcement learning
CN202330618U (en) Fault location device for distribution network
Teng et al. On-line fault diagnosis technology and application based on deep learning of fault characteristic of power grid
CN110991519A (en) Intelligent switch state analysis and adjustment method and system
Ge et al. Improved harris hawks optimization for configuration of PV intelligent edge terminals
Guan et al. Enhancing security and resilience of bulk power systems via multisource big data learning
Pan et al. Random vector functional link network optimized by jaya algorithm for transient stability assessment of power systems
Bhattacharya et al. Deep fault analysis and subset selection in solar power grids
CN113191522A (en) Self-healing control system and method for intelligent power distribution network
Haidar et al. An intelligent load shedding scheme using neural networks and neuro-fuzzy
CN106451565A (en) Perturbation theory-based online analysis method of power plant optimization transmission

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C02 Deemed withdrawal of patent application after publication (patent law 2001)
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20100616