CN102098687B - Multi-object optimized deployment method for industrial wireless sensor network - Google Patents
Multi-object optimized deployment method for industrial wireless sensor network Download PDFInfo
- Publication number
- CN102098687B CN102098687B CN201110049025.7A CN201110049025A CN102098687B CN 102098687 B CN102098687 B CN 102098687B CN 201110049025 A CN201110049025 A CN 201110049025A CN 102098687 B CN102098687 B CN 102098687B
- Authority
- CN
- China
- Prior art keywords
- harmony
- sensor node
- individual
- cluster head
- base station
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Mobile Radio Communication Systems (AREA)
- Arrangements For Transmission Of Measured Signals (AREA)
Abstract
The invention discloses a multi-object optimized deployment method for an industrial wireless sensor network, comprising the following steps: (1) dividing a monitoring region into three-dimensional grids, and arranging sensor nodes, cluster heads and base stations on cross points of the grids; (2) generating obstacle matrixes; (3) representing harmony individuals; (4) setting algorithm control parameters; (5) setting communication radiuses of the sensor and the cluster heads; (6) judging whether the communication between the sensor nodes and the cluster heads satisfies the conditions; (7) judging whether the hop number of communication between the sensor nodes and the base stations satisfies the conditions; (8) initializing harmony matrixes by using a heuristic strategy; (9) calculating an objective function value of each harmony; (10) finding out the optimized harmony; (11) generating a new harmony; (12) comparing the quality of the new harmony with the quality of the corresponding harmony in a harmony memory vault; (13) updating the optimized harmony; and (14) judging whether the stopping conditions are satisfied or not. The method can realize multi-object optimization for system reliability, and real-time performance, sensor node deployment cost and maintenance cost of the industrial wireless sensor network, and can satisfy the industrial actual requirements.
Description
Technical field
The present invention relates to industrial wireless sensor network and the big field of intelligence computation two, and in particular to a kind of method of industrial wireless sensor network multiobjective optimization deployment.
Background technology
Constantly maximize with industrial system, complicate, the scale of industrial control system constantly expands, it is installed, wiring cost is also continuously increased.Industrial transducer market share in 2001 is 11,000,000,000 dollars according to statistics, and its cost of installation and use（Mainly wiring cost）More than 100,000,000,000 dollars.Therefore, wireless sensor network（Wireless Sensor Networks, WSNs）The feature such as inexpensive, easy-to-use cause the extensive concern of industrial quarters, international well-known control system company, such as Emerson, Honeywell, General Electric have carried out industrial wireless sensor network（Industrial Wireless Sensor Networks, IWSNs）The research and development of technology.IWSNs is to be made up of the sensor node with wireless telecommunications and computing capability, be deployed in the self-organization distribution network intelligence system that industrial environment provides solution for certain commercial Application, is a hot spot technology of the industrial control field after field bus technique.IWSNs can reduce industry measurement and control system cost, improve industry measurement and control system application and reliability.Emerson company is in North America and European several field tests as a result, it was confirmed that using the reliability of Industrial Wireless data transfer more than 99%, and installation cost is lower than equal wired solution by 90%.For the huge prospect of Industrial Wireless, USDOE points out in " the futurity industry plan " of issue in 2004：This inexpensive TT＆C system based on Industrial Wireless is to realize to the overall energy consumption of the year two thousand twenty American industry to reduce the Main Means of 5% target, represents the developing direction of industrial automation system technology.
Based on generally comprising two class nodes in cluster structured IWSNs：Sensor node and cluster head, wherein sensor node are responsible for the collection of field data, and Data Integration is simultaneously sent to base station by the data that cluster head reception sensor node is sended over.Although IWSNs chief component is also wireless sensor node, but it is different from general non-industrial IWSNs, IWSN sensor node deployment is relevant with industrial environment, needs on the manually installed industrial equipment detected to needs, emphasizes the reliability detection to specified point.Compared to conventional wired networks, on the one hand, IWSNs is easier to be influenceed and broken down by factors such as network topology, environment, and failure is also inevitable in practical application.On the other hand, the communication modes of IWSNs multi-hops also cause system real time to decline, and show that the real-time of the data when leapfrog number is more than 6 will not ensure that in industrial experiment.IWSN, once its reliability and real-time, which cannot be guaranteed, will directly affect product quality, or even produce major accident, causes huge economic losses as the important component of control system.Therefore for large-scale industrial application, particularly closed-loop control application, IWSN reliability and real-time is most important.
In addition, sensor node is typically all powered using battery, and finite energy, the system in process industry, once node energy exhausts, it is necessary to change battery.If deployment design is improper, different nodes, cluster head battery loss difference it is excessive, unified change can cause unnecessary waste, and will then be caused continually to carry out battery altering according to actual electricity, and this obviously can increase the workload of network operation.
For the ease of managing, improving real-time, IWSN generally using cluster structured, make cluster inner sensor node only and this cluster cluster head is communicated, and the data that cluster head collection cluster interior nodes are sent simultaneously are sent to base station in a multi-hop fashion by remaining cluster head.In order to ensure possibility, it is necessary to be node configuring redundancy cluster head so that each node can communicate with two cluster heads（The using and the reserved）Can be quickly switched into standby cluster head when a cluster head breaks down makes system worked well, it is ensured that the reliability of system, simultaneously, system in commercial Application, system in particularly extensive process industry is, it is necessary to which detection variable is numerous, often up to points up to ten thousand, therefore, need Optimization deployment cluster head to reduce cost of investment, simplify network structure, be easy to manage and safeguard.The certain space of hold facility in actual industrial scene, there is also other barriers such as wall, these barriers can all influence the communication between node and cluster head, therefore the optimal deployment issues of IWSNs are actually to be related to a multi-objective optimization question with a variety of constraints, i.e.,,IWSNs is meeting, optimized under the constraint of system reliability and real-time, realizes the minimum of system constructing cost and maintenance cost.Large-scale industry wireless sensor network optimal deployment method is a class np hard problem, and traditional certainty optimization method can not realize the Efficient Solution to this class problem.In recent years, intelligent optimization algorithm embodies the advantage bigger than traditional deterministic optimization method when solving np hard problem, and wherein harmony is searched for（Harmony Search, HS）The excellent ability of searching optimum of algorithm and the extensive concern that researcher is caused the characteristics of simply easily realization.Harmonic search algorithm is a kind of new intelligent optimization algorithm proposed in recent years, HS basic thought：From the harmony data base randomly generated, candidate solution is produced by the thinking to harmony data base, random selection tone and tone regulation, then harmony data base is updated by contrasting the worst solution in candidate solution and data base.
But, basic HS algorithms use real number or discrete codes, it is mainly used in the optimization problem of continuous space and discrete space, for IWSNs layout optimizations problem, binary coding optimization problem, basic harmonic search algorithm tone regulation is operated degradation failure, therefore, basic HS algorithm is not suitable for binary coding optimization problem.
The content of the invention
The problem of it is an object of the invention to exist for prior art, a kind of method of industrial wireless sensor network multiobjective optimization deployment is provided, this method is optimized using heuristic binary system harmonic search algorithm to node deployment model, on the basis of system reliability and requirement of real-time is ensured, system maintenance cost and system constructing cost can be further reduced.
To reach above-mentioned purpose, the present invention uses following technical proposals：
The method of a kind of industrial wireless sensor network multiobjective optimization deployment, it is characterised in that this method is comprised the following steps that：
(1), monitored area is divided into according to industry spot real space, barrier size and location, wireless senser power, required precision first
Three-dimensional grid,R、S、TCorrespond to respectively on horizontal, vertical, ordinate and divide hop count, sensor node, cluster head and base station are all deployed on grid intersection respectively, according to industrial requirements, and sensor node is divided into two classes, and a class is general sensor node, and another kind of is key sensor node；According to actual process design requirement, have in industry spot real space
Individual sensor node, vacant
Cluster head and base station are disposed on individual mesh point；
(2), existed according to live physical device
Position dyspoiesis thing matrix in three-dimensional network coordinateB, barrier matrixBSize is, grid represents there is barrier on the grid for " 1 ", represents do not have barrier on the grid if grid is " 0 ",Think not communicating with one another between this two sensorses node if it there are barrier in the straight line path between any two points；
(3), the expression of harmony individual solution：
,
(4) each control parameter of heuristic binary system harmonic search algorithm is set, the control parameter of heuristic binary system harmonic search algorithm includes creation number of timesNI 、Harmony data base size
, harmony data base thinking probability
Probability is finely tuned with tone, and random initializtion harmony data baseHM；
(5), the communication radius of setting sensor node is
, the communication radius of cluster head is
, judge whether the distance between sensor node and cluster head are less than or equal to
If the distance between sensor node and cluster head are less than or equal to
, and when barrier is not present between both communication links, then it is assumed that sensor node is communicated with the cluster head, is loaded sensor node as one of the cluster head, otherwise it is assumed that sensor node can not be communicated with the cluster head；Judge whether the distance between two cluster heads are less than or equal to
If the distance between two cluster heads are less than or equal to
, and communication link between two cluster heads is when being not present barrier, then it is assumed that communicated with one another between two cluster heads；The total load of cluster head is cluster inner sensor nodes and the cluster head number sum communicated；
(6), judge whether that each general sensor node is at least communicated with 2 cluster heads, if each general sensor node can not be communicated at least with 2 cluster heads, then think that sensor node is unsatisfactory for system reliability requirement, then calculate all general sensor nodesPenalty value
Otherwise it is assumed that general sensor node meets system reliability requirement, the penalty value that general sensor is put successively is not calculated, judges whether key sensor node can communicate with more than or equal to 3 cluster heads, if can not be communicated with more than or equal to 3 cluster heads, all key sensor nodes are calculated
Penalty value
, otherwise it is assumed that key sensor node meets system reliability requirement, the penalty value of key node is not calculated；
(7), being set in monitored area hasN BASE Individual base station, is designated asN BASE , sensor nodeiIt is with hop count needed for nearest base station communication by cluster headH i ,In view of the real-time of industrial wireless sensor network, it is desirable to which the hop count that sensor node reaches base station is not more than Smax, judges the hop count of each sensor node and each base station communicationH i Whether satisfaction is less than maximum hop count Smax, if the hop count of each sensor node and each base station communicationH i It can not meet less than maximum hop count Smax, then it is assumed that sensor node is unsatisfactory for requirement of real-time, then calculate all the sensors node
Penalty value, otherwise it is assumed that sensor node meets requirement of real-time, the penalty value of sensor node is not calculated；
(8) harmony matrix, is initialized using heuristic strategies, random value is adjusted according to each sensor node distribution density, has tendency to produce harmony matrix, the size of the harmony matrix is designated as HMS；Simultaneously in order to ensure not losing possible solution, the harmony of one complete " 1 " is produced；
(9) each harmony, is calculatedxTarget function value, object function is：
Wherein, minf(x）Represent object function,
,
,
It is subfunction respectively
,
,
Weights,；Expression cluster head quantity,
The quantity of expression base station,
Represent the standard deviation of cluster head load；；
(10), best harmony is found in initialization neutralizes sound memory storehouseh g
Best harmony definition：The minimum harmony of target function value in harmony matrix；
(11) new harmony, is generated；
(12) new harmony, is compared
Harmony corresponding with harmony data base
Quality；If new harmonyHarmony more corresponding than in harmony data base
It is excellent, then with new harmony
Replace corresponding harmony in harmony data base, otherwise harmony matrix is constant；
（13）, update optimal harmonyh g ；
（14）, judge whether the harmony number that newly produces is less than HMS, if the harmony number newly produced is less than HMS, return to step（11）Continue to produce new harmony, if the harmony number newly produced is not less than HMS, and algorithm has reached maximum iteration, then stop iteration, generate optimal harmonyh g ,Otherwise return to step（11）Continue iteration.
Above-mentioned steps（11）Described in the new harmony of generation, its is specific as follows：
(11-1), with probabilityHMCRNew explanation is searched in harmony data base, with probability 1-HMCRSearched in the possibility domain of variable,It searches for formula：
Wherein, rand () represents the random number between 0 to 1,N ijRepresent the harmony individual newly produced, HMijI-th of individual of harmony matrix is represented,
In being searched in harmony data base, propose a kind of new individual search strategy of parallel correspondence, correspondence harmony data base generates multiple new explanations in new parallel individual search strategy an iteration, the strategy can effectively prevent that algorithm is precocious for large-scale industry wireless sensor network disposition optimization problem, improve the global optimization performance of algorithm, a kind of new heuristic strategies, heuristic strategies may be proposed in solution during domain search：It suppose there is
Individual sensor node, whereinkIndividual general sensor node, the cluster head number now needed is at least
Individual cluster head（In other words, wherein the number for producing bit " 1 " in new harmony is no less than and is
,）Therefore, adjustment randomly generates the probability of " 1 ", and the probability calculation formula of its " 1 " is：
Wherein, k is relaxed operator, can use 1.2-1.5, and the formula for randomly generating 0 or 1 is：
Wherein, NijThe harmony individual newly produced is represented, rand () represents the random number between 0 to 1,The heuristic strategies can effectively estimate solution, so as to improve the search efficiency and search quality of algorithm；
（11-2）Different from other application in binary-coded binary system harmonic search algorithm, operation operator is finely tuned using a kind of new tone based on didactic binary system harmonic search algorithm（In other words, if wherein new harmony come from harmony data base, global optimization tone fine setting is carried out to new harmony with probability P AR）,Its tone finely tunes operation operator calculating formula
Wherein,N ij The harmony individual newly produced is represented,
Harmony individual optimal in harmony matrix is represented,HM ij Represent i-th of individual in harmony matrix.
Above-mentioned steps（12）Described in the new harmony of comparison
Harmony corresponding with harmony data base
Quality, the more excellent entrance harmony data base of selection adaptive value is：
Wherein, HMiI-th of individual in harmony data base is represented,N i Represent the harmony individual newly produced..
A kind of method of industrial wireless sensor network multiobjective optimization deployment of the present invention compared with prior art, substantive distinguishing features and remarkable advantage is obviously protruded with following：
First, this method directly uses binary coding, and harmony individual is expressed as binary string, according to the numerical value of two close bits in individual vector determine its
What the correspondence position in three-dimensional grid was arranged is cluster head or base station, algorithm is realized simple, and it is very directly perceived, operand is small, speed is fast, and new harmony data base selection operation and tone fine setting operate to replace the origin operation in continuous space and discrete space in the prior art so that binary system harmonic search algorithm is applied to complicated binary coding optimization problem, proposing a kind of heuristic operation makes the efficiency of algorithm high, and optimization performance is good.
Secondly, this method ensures that each sensor node is at least communicated with 2 cluster heads when building IWSN, it is ensured that is rapidly switched to redundancy cluster head when the cluster head of work breaks down and continues to work, meets system redundancy demand, it is ensured that, optimize the reliability of system.Meanwhile, this method passes through the limitation to the required hop count that communicated between sensor node and base station, it is ensured that、The real-time of optimization system.This method is that system reliability, real-time, sensor node deployment cost and the maintenance cost of industrial wireless sensor network realize multiple-objection optimization, the position and quantity of cluster head and base station are optimized using newly proposing to search for based on didactic binary system harmony, solve the problems, such as single objective with constraints in the prior art, consider that reliability or cost are more comprehensive than simple, so as to really meet the actual demand in commercial Application.
Brief description of the drawings
Fig. 1 is the deployment schematic diagram in a kind of method of industrial wireless sensor network multiobjective optimization deployment of the present invention；
The flow chart for the method that Fig. 2 disposes for a kind of industrial wireless sensor network multiobjective optimization of the present invention.
Embodiment
The preferred embodiment of the present invention is described in further detail below in conjunction with the accompanying drawings.
Referring to Fig. 1, the method that a kind of industrial wireless sensor network multiobjective optimization of the invention is disposed, its operating procedure is as follows：
(1), monitored area is divided into according to industry spot real space, barrier size and location, wireless senser power, required precision first
Three-dimensional grid,R、S、TCorrespond to respectively on horizontal, vertical, ordinate and divide hop count, sensor node, cluster head and base station are all deployed on grid intersection respectively, according to industrial requirements, and sensor node is divided into two classes, and a class is general sensor node, and another kind of is key sensor node；According to actual process design requirement, have in industry spot real space
Individual sensor node, vacant
Optimization deployment cluster head and base station on individual mesh point, to ensure the reliability and real-time of industry measurement and control system, as shown in figure 1, in figure, DH represents cluster head, N represents sensor node, and B represents base station, RchRepresent the communication radius of cluster head, RnThe communication radius of sensor node is represented, the double-head arrow from sensor node to cluster head illustrates that the sensor node is communicated with cluster head, and the double-head arrow between cluster head and base station illustrates the intercommunication of cluster head and base station；
(2), existed according to live physical device
Position dyspoiesis thing matrix in three-dimensional network coordinateB, barrier matrix B size is
, grid represents there is barrier at this for " 1 ", represents do not have barrier on the grid if grid is " 0 ", thinks not communicating with one another between this two sensorses node if it there are barrier in the straight line path between any two points；
(3), the expression of harmony individual solution：
,
, it is vacant
Arrangement gathers head and base station on mesh point,
, wherein, n represents the grid sum for disposing cluster head and base station,
Represent thejIndividual abortive haul lattice point is vacant,
Represent thejSensor cluster head is disposed on individual abortive haul lattice point,
Represent thejBase station is disposed on individual abortive haul lattice point；
(4) each control parameter of heuristic binary system harmonic search algorithm, is set, the control parameter of heuristic binary system harmonic search algorithm includes creation number of timesNI 、Harmony data base size
, harmony data base thinking probability
Probability is finely tuned with tone
, and random initializtion harmony data baseHM；
(5), the communication radius of setting sensor node is
, the communication radius of cluster head is
, judge whether the distance between sensor node and cluster head are less than or equal toIf the distance between sensor node and cluster head are less than or equal to, and when barrier is not present between both communication links, then it is assumed that sensor node is communicated with the cluster head, is loaded sensor node as one of the cluster head, otherwise it is assumed that sensor node can not be communicated with the cluster head；Judge whether the distance between two cluster heads are less than or equal to
If the distance between two cluster heads are less than or equal to
, and communication link between two cluster heads is when being not present barrier, then it is assumed that communicated with one another between two cluster heads；The total load of cluster head is cluster inner sensor nodes and the cluster head number sum communicated；
（6）, judge whether that each general sensor node is at least communicated with 2 cluster heads, if each general sensor node can not be communicated at least with 2 cluster heads, then it is assumed that sensor node is unsatisfactory for system reliability requirement, then calculates all general sensor nodes
Penalty value
,
, whereinkp 1It is penalty coefficient；Judge whether key sensor node can communicate with more than or equal to 3 cluster heads, if can not be communicated with more than or equal to 3 cluster heads, calculate all key sensor nodes
Penalty value；
,kp 2 It is penalty coefficient；
(7), being set in monitored area hasN BASE Individual base station（N BASE Represent base station）, nodeiIt is with hop count needed for nearest base station communication by cluster headH i ,In view of the real-time of industrial wireless sensor network, it is desirable to which the hop count that node reaches base station is not more than Smax（Usual Smax should be less than 6 jumps）, judge the hop count of each sensor node and each base station communicationH i Whether satisfaction is less than maximum hop count Smax, if the hop count of each sensor node and each base station communicationH i It can not meet less than maximum hop count Smax, then it is assumed that sensor node is unsatisfactory for requirement of real-time, then calculate all nodes
Penalty value
, its calculating formula is：
, wherein,kp 3It is penalty coefficient, as shown in figure 1, result can be seen that each sensor node and at least be communicated with two cluster heads from figure, node is not more than Smax with the hop count needed for base station communication（Usual Smax should be less than 6）, meet the real-time and reliability requirement of system；
(8) harmony matrix, is initialized using heuristic strategies, random value is adjusted according to each sensor node distribution density, has tendency to produce harmony matrix, the size of the harmony matrix is designated as HMS；Simultaneously in order to ensure not losing possible solution, the harmony of one complete " 1 " is produced；
（9）, calculate each harmonyxTarget function value, object function is：
Wherein, minf(x）Represent object function,
,
,
It is subfunction respectively
,
,
Weights,
Expression cluster head quantity,
The quantity of expression base station,
Represent the standard deviation of cluster head load；Standard deviation
Calculating formula is：
(10), best harmony is found in initialization neutralizes sound memory storehouseh g
Best harmony definition：The minimum harmony of target function value in harmony matrix, is placed on minimum this refers to f (x) above；
(11) new harmony, is generated：
(11-1), with probabilityHMCRNew explanation is searched in harmony data base, with probability 1-HMCRSearched in the possibility domain of variable,It searches for formula：
Wherein, rand () represents the random number between 0 to 1,N ijRepresent the harmony individual newly produced, HM ij Represent i-th of individual of harmony matrix.
In being searched in harmony data base, propose a kind of new individual search strategy of parallel correspondence, correspondence harmony data base generates multiple new explanations in new parallel individual search strategy an iteration, the strategy can effectively prevent that algorithm is precocious for large-scale industry wireless sensor network disposition optimization problem, improve the global optimization performance of algorithm, a kind of new heuristic strategies, heuristic strategies may be proposed in solution during domain search：It suppose there is
Individual sensor node, whereinkIndividual general sensor node, the cluster head number now needed is at leastIndividual cluster head（In other words, wherein the number for producing bit " 1 " in new harmony is no less than and is
,）Therefore, adjustment randomly generates the probability of " 1 ", and the probability calculation formula of its " 1 " is：
Wherein, k is relaxed operator, can use 1.2-1.5, and the formula for randomly generating 0 or 1 is：
Wherein, NijThe harmony individual newly produced is represented, rand () represents the random number between 0 to 1,The heuristic strategies can effectively estimate solution, so as to improve the search efficiency and search quality of algorithm.
（11-2）Different from other application in binary-coded binary system harmonic search algorithm, operation operator is finely tuned using a kind of new tone based on didactic binary system harmonic search algorithm（In other words, if wherein new harmony come from harmony data base, global optimization tone fine setting is carried out to new harmony with probability P AR）,Its tone finely tunes operation operator calculating formula
Wherein,N ij The harmony individual newly produced is represented, 
Harmony individual optimal in harmony matrix is represented,HM ij Represent i-th of individual in harmony matrix.
(12) new harmony is compared
Harmony corresponding with harmony data base
Quality, the more excellent entrance harmony data base of selection adaptive value selection adaptive value is：
Wherein, HMiI-th of individual in harmony data base is represented,N i Represent the harmony individual newly produced；
（13）, update optimal harmonyh g ,
（14）, judge whether the harmony number that newly produces is less than HMS, if the harmony number newly produced is less than HMS, return to step（11）Continue to produce new harmony, if the harmony number newly produced is not less than HMS, and algorithm has reached maximum iteration, then stop iteration, generate optimal harmonyh g ,Otherwise return to step（11）Continue iteration.According to optimal harmonyh g The position and quantity of cluster head and base station in monitored area are determined, then according to sensor node and the communication radius of cluster head, realizes that industrial wireless sensor network multiobjective optimization is disposed.
A kind of method of industrial wireless sensor network multiobjective optimization deployment of the present invention is according to optimal harmonyh g Determine the position and quantity of cluster head and base station in monitored area, then the communication channels figure that industry spot region wireless sensor network coverage diagram, sensor node reach base station can be drawn according to the communication radius of sensor node and cluster head, the communication channels figure of base station is reached by region wireless sensor network coverage diagram coverage diagram and sensor node it can be seen that each sensor node is at least communicated with 2 cluster heads, the sensor node specified in network is at least communicated with 3 cluster heads, and each sensor node reaches the hop count Smax needed for base station（Usual Smax should be less than 6）, also at least 2 links lead to base station to each cluster head, therefore, and the present invention can ensure the communication reliability and real-time of system in sensor node layer with cluster head layer.By the rational deployment to leader cluster node and base station, balance the energy consumption of cluster head and reduce the hop count needed for being communicated between sensor node and base station, be conducive to extending the life cycle of network.In addition, the present invention can build network and maintenance cost with minimum based on heuristic binary system harmonic search algorithm into original, and ensure system redundancy demand, improve system reliability and real-time.
Claims (3)
1. a kind of method of industrial wireless sensor network multiobjective optimization deployment, it is characterised in that this method is comprised the following steps that：
(1), monitored area is divided into according to industry spot real space, barrier size and location, wireless senser power, required precision first
Three-dimensional grid,R、S、TCorrespond to respectively on horizontal, vertical, ordinate and divide hop count, sensor node, cluster head and base station are all deployed on grid intersection respectively, according to industrial requirements, and sensor node is divided into two classes, and a class is general sensor node, and another kind of is key sensor node；According to actual process design requirement, have in industry spot real spaceIndividual sensor node, vacant
Cluster head and base station are disposed on individual mesh point；
(2), existed according to live physical device
Position dyspoiesis thing matrix B in three-dimensional network coordinate, barrier matrixBSize is
, grid represents there is barrier on the grid for " 1 ", represents do not have barrier on the grid if grid is " 0 ",If there are barrier in the straight line path between sensor node, cluster head and base station any two points, then it is assumed that can not be communicated with one another between 2 points of this sensor node, cluster head and base station；
(3), the expression of harmony individual solution：
,
, it is vacant
Cluster head and base station are arranged on mesh point,
, wherein, n represents the grid sum for disposing cluster head and base station,
Represent thejIndividual abortive haul lattice point is vacant,OrRepresent thejSensor cluster head is disposed on individual abortive haul lattice point,
Represent thejBase station is disposed on individual abortive haul lattice point；
(4) each control parameter of heuristic binary system harmonic search algorithm, is set, the control parameter of heuristic binary system harmonic search algorithm includes creation number of timesNI 、Harmony data base size, harmony data base thinking probability and tone fine setting probability
, and random initializtion harmony data baseHM；
(5), the communication radius of setting sensor node is
, the communication radius of cluster head is
, judge whether the distance between sensor node and cluster head are less than or equal to
If the distance between sensor node and cluster head are less than or equal to
, and when barrier is not present between both communication links, then it is assumed that sensor node is communicated with the cluster head, is loaded sensor node as one of the cluster head, otherwise it is assumed that sensor node can not be communicated with the cluster head；Judge whether the distance between two cluster heads are less than or equal to
If the distance between two cluster heads are less than or equal to, and communication link between two cluster heads is when being not present barrier, then it is assumed that communicated with one another between two cluster heads；The total load of cluster head is cluster inner sensor nodes and the cluster head number sum communicated；
(6), judge whether that each general sensor node is at least communicated with 2 cluster heads, if each general sensor node can not be communicated at least with 2 cluster heads, then think that sensor node is unsatisfactory for system reliability requirement, then calculate all general sensor nodesPenalty value
,
, wherein
It is penalty coefficient, otherwise it is assumed that general sensor node meets system reliability requirement, the penalty value that general sensor is put successively is not calculated, judge whether key sensor node can communicate with more than or equal to 3 cluster heads, if can not be communicated with more than or equal to 3 cluster heads, all key sensor nodes are calculated
Penalty value,
,
It is penalty coefficient, otherwise it is assumed that key sensor node meets system reliability requirement, the penalty value of key node is not calculated；
(7), being set in monitored area has
Individual base station, is designated as
, sensor nodeiIt is with hop count needed for nearest base station communication by cluster head ,In view of the real-time of industrial wireless sensor network, it is desirable to which the hop count that sensor node reaches base station is not more than Smax, judges the hop count of each sensor node and each base station communication
Whether satisfaction is less than maximum hop count Smax, if the hop count of each sensor node and each base station communicationIt can not meet less than maximum hop count Smax, then it is assumed that sensor node is unsatisfactory for requirement of real-time, then calculate all the sensors node
Penalty value
,,It is penalty coefficient, otherwise it is assumed that sensor node meets requirement of real-time, the penalty value of sensor node is not calculated；
(8) harmony matrix, is initialized using heuristic strategies, random value is adjusted according to each sensor node distribution density, has tendency to produce harmony matrix, the size of the harmony matrix is designated as HMS；Simultaneously in order to ensure not losing possible solution, the harmony of one complete " 1 " is produced；
(9) each harmony, is calculatedxTarget function value, object function is：
Wherein, minf(x）Represent object function,
,
,
It is subfunction respectively
,
,Weights；
Expression cluster head quantity,
The quantity of expression base station,
Represent the standard deviation of cluster head load；
(10), best harmony is found in initialization neutralizes sound memory storehouse
Best harmony definition：The minimum harmony of target function value in harmony matrix；
(11) new harmony, is generated；
(12) new harmony, is compared
Harmony corresponding with harmony data base
Quality；If new harmony
Harmony more corresponding than in harmony data base
It is excellent, then with new harmonyReplace corresponding harmony in harmony data base, otherwise harmony matrix is constant；
（13）, update optimal harmony
；
（14）, judge whether the harmony number that newly produces is less than HMS, if the harmony number newly produced is less than HMS, return to step（11）Continue to produce new harmony, if the harmony number newly produced is not less than HMS, and algorithm has reached maximum iteration, then stop iteration, generate optimal harmony ,Otherwise return to step（11）Continue iteration.
2. a kind of method of industrial wireless sensor network multiobjective optimization deployment according to claim 1, it is characterised in that above-mentioned steps（11）Described in the new harmony of generation, its is specific as follows：
(11-1), with probabilityHMCRNew explanation is searched in harmony data base, with probability 1-HMCRSearched in the possibility domain of variable,It searches for formula：
Wherein, rand () represents the random number between 0 to 1,
The harmony individual newly produced is represented,
I-th of individual of harmony matrix is represented,
In being searched in harmony data base, propose a kind of new individual search strategy of parallel correspondence, correspondence harmony data base generates multiple new explanations in new parallel individual search strategy an iteration, the strategy can effectively prevent that algorithm is precocious for large-scale industry wireless sensor network disposition optimization problem, improve the global optimization performance of algorithm, a kind of new heuristic strategies, heuristic strategies may be proposed in solution during domain search：It suppose there is
Individual sensor node, whereinkIndividual general sensor node, the cluster head number now needed is at least
Therefore, adjustment randomly generates the probability of " 1 " to individual cluster head, and the probability calculation formula of its " 1 " is：
Wherein, kr is relaxed operator, can use 1.2-1.5, and the formula for randomly generating 0 or 1 is：
Wherein,
The harmony individual newly produced is represented, rand () represents the random number between 0 to 1,The heuristic strategies can effectively estimate solution, so as to improve the search efficiency and search quality of algorithm；
（11-2）Different from other application in binary-coded binary system harmonic search algorithm, operation operator is finely tuned using a kind of new tone based on didactic binary system harmonic search algorithm, its tone fine setting operation operator calculating formula is：
Wherein,
The harmony individual newly produced is represented,
Harmony individual optimal in harmony matrix is represented,
Represent i-th of individual in harmony matrix.
3. a kind of method of industrial wireless sensor network multiobjective optimization deployment according to claim 2, it is characterised in that above-mentioned steps（12）Described in the new harmony of comparisonHarmony corresponding with harmony data base
Quality, the more excellent entrance harmony data base of selection adaptive value is：
Wherein,
I-th of individual in harmony data base is represented,
For in harmony data base i-th individual adaptive value,
The harmony individual newly produced is represented,
For the adaptive value of the harmony individual newly produced.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110049025.7A CN102098687B (en) | 2011-03-02 | 2011-03-02 | Multi-object optimized deployment method for industrial wireless sensor network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110049025.7A CN102098687B (en) | 2011-03-02 | 2011-03-02 | Multi-object optimized deployment method for industrial wireless sensor network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102098687A CN102098687A (en) | 2011-06-15 |
| CN102098687B true CN102098687B (en) | 2014-01-15 |
Family
ID=44131508
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110049025.7A Expired - Fee Related CN102098687B (en) | 2011-03-02 | 2011-03-02 | Multi-object optimized deployment method for industrial wireless sensor network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102098687B (en) |
Families Citing this family (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102625324B (en) * | 2012-03-08 | 2014-10-01 | 上海大学 | Wireless optical fiber sensor network deployment method based on particle swarm optimization |
| CN102915355A (en) * | 2012-10-11 | 2013-02-06 | 李英明 | Multiprocessor task scheduling method based on harmony search and simulated annealing |
| CN102917388B (en) * | 2012-10-15 | 2015-03-25 | 中国联合网络通信集团有限公司 | Self-repairing method for base stations, cut-through management and control device, cluster head base station and communication system |
| CN103856952A (en) * | 2014-02-27 | 2014-06-11 | 上海大学 | Method for optimizing Pareto multiple target deployment of industrial wireless sensor network |
| CN103916927B (en) * | 2014-03-17 | 2017-06-13 | 华中科技大学 | A kind of wireless sensor network routing method based on improvement harmonic search algorithm |
| CN103997747B (en) * | 2014-05-14 | 2018-04-17 | 浪潮电子信息产业股份有限公司 | A kind of uniform dispositions method of spatial network node based on fictitious force |
| CN105163325B (en) * | 2015-09-25 | 2018-03-13 | 重庆工商大学 | Isomery directional sensor network dispositions method |
| CN105979540A (en) * | 2016-06-14 | 2016-09-28 | 昆明理工大学 | Real time positioning topology control method for three-dimensional wireless sensor network |
| CN106131862B (en) * | 2016-07-01 | 2019-08-16 | 厦门大学 | Optimization covering method based on multi-objective Evolutionary Algorithm in a kind of wireless sensor network |
| CN106550450B (en) * | 2016-10-11 | 2019-08-13 | 电子科技大学 | A kind of polyhedron grid covering searching method |
| CN107343283A (en) * | 2017-06-02 | 2017-11-10 | 电子科技大学 | A kind of three-dimensional static radio sensing network dispositions method based on genetic algorithm |
| CN107708133B (en) * | 2017-09-21 | 2021-02-26 | 浙江工业大学 | Mesh division-based underwater wireless sensor network multi-dimensional optimization node deployment method |
| CN108200549A (en) * | 2018-01-10 | 2018-06-22 | 深圳市鑫汇达机械设计有限公司 | Tea growth ambient intelligence monitoring system |
| CN109033520A (en) * | 2018-06-25 | 2018-12-18 | 北京交通大学 | A kind of system Multipurpose Optimal Method based on random enhancing harmony algorithm |
| CN111542070B (en) * | 2020-04-17 | 2023-03-14 | 上海海事大学 | Efficient multi-constraint deployment method for industrial wireless sensor network |
| CN114676539B (en) * | 2022-04-12 | 2022-10-14 | 山东华方智联科技股份有限公司 | NB-IoT-based multi-target deployment method for intelligent electric meter |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101221236A (en) * | 2008-02-02 | 2008-07-16 | 北京航空航天大学 | Node self-locating method based on sampling of wireless sensor network in three-dimensional space |
| CN101339240A (en) * | 2008-08-26 | 2009-01-07 | 中国人民解放军海军工程大学 | Wireless sensor network object tracking method based on double layer forecast mechanism |
| CN101383736A (en) * | 2008-10-15 | 2009-03-11 | 中国科学院上海微系统与信息技术研究所 | Optimizing method for wireless sensor network node laying oriented to area monitoring |
-
2011
- 2011-03-02 CN CN201110049025.7A patent/CN102098687B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101221236A (en) * | 2008-02-02 | 2008-07-16 | 北京航空航天大学 | Node self-locating method based on sampling of wireless sensor network in three-dimensional space |
| CN101339240A (en) * | 2008-08-26 | 2009-01-07 | 中国人民解放军海军工程大学 | Wireless sensor network object tracking method based on double layer forecast mechanism |
| CN101383736A (en) * | 2008-10-15 | 2009-03-11 | 中国科学院上海微系统与信息技术研究所 | Optimizing method for wireless sensor network node laying oriented to area monitoring |
Non-Patent Citations (4)
| Title |
|---|
| 无线传感器网络多目标跟踪数据融合;王换招、范琳、王海、李增智;《西安交通大学学报》;20061010;第40卷(第10期);全文 * |
| 无线传感器网络的多目标覆盖控制策略;梁天、周晖、谢静、王坤赤;《传感技术学报》;20100731;第23卷(第7期);全文 * |
| 梁天、周晖、谢静、王坤赤.无线传感器网络的多目标覆盖控制策略.《传感技术学报》.2010,第23卷(第7期),全文. |
| 王换招、范琳、王海、李增智.无线传感器网络多目标跟踪数据融合.《西安交通大学学报》.2006,第40卷(第10期),全文. |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102098687A (en) | 2011-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102098687B (en) | Multi-object optimized deployment method for industrial wireless sensor network | |
| CN102014398B (en) | Optimal deployment method of large-scale industrial wireless sensor network based on differential evolution algorithm | |
| Wang et al. | Hybrid multihop partition-based clustering routing protocol for WSNs | |
| Das et al. | Fuzzy based energy efficient multicast routing for ad-hoc network | |
| CN105246097A (en) | Lifetime optimization method of mobile Sink-based wireless sensor network | |
| CN102395146A (en) | Multiple-target monitoring oriented method for sensing topology construction in wireless sensor network | |
| CN103856952A (en) | Method for optimizing Pareto multiple target deployment of industrial wireless sensor network | |
| Zhou et al. | Genetic clustering route algorithm in WSN | |
| Cao et al. | Cluster heads election analysis for multi-hop wireless sensor networks based on weighted graph and particle swarm optimization | |
| CN102238561A (en) | Node deployment method for energy efficient hierarchical collaboration coverage model | |
| Yang et al. | A mobile sensors dispatch scheme based on improved SOM algorithm for coverage hole healing | |
| Elhabyan et al. | A full area coverage guaranteed, energy efficient network configuration strategy for 3d wireless sensor networks | |
| CN110322079A (en) | A kind of flow of power plan optimization method | |
| Abasi et al. | Fuzzy-clustering based data gathering in wireless sensor network | |
| Li et al. | An optimal design method for communication topology of wireless sensor networks to implement fully distributed optimal control in IoT-enabled smart buildings | |
| Shi et al. | Energy dependent divisible load theory for wireless sensor network workload allocation | |
| Wu et al. | Multiobjective optimization strategy of WSN coverage based on IPSO-IRCD | |
| Zhong et al. | Energy-efficient routing control algorithm in large-scale WSN for water environment monitoring with application to three gorges reservoir area | |
| Guo et al. | Design and analysis of a MST-based topology control scheme with PSO for wireless sensor networks | |
| CN112234599B (en) | Advanced dynamic self-adaptive partitioning method and system for multi-element complex urban power grid | |
| CN112116131B (en) | Multi-level optimization method for comprehensive energy system considering carbon emission | |
| CN109041159B (en) | Method for overcoming energy imbalance in wireless sensor network | |
| Yao et al. | Multihop clustering routing protocol based on improved coronavirus herd immunity optimizer and Q-learning in WSNs | |
| Hebal et al. | Solving energy routing problem in energy internet using a discrete artificial bee colony algorithm | |
| Akram et al. | On Layout Optimization of Wireless Sensor Network Using Meta-Heuristic Approach. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20140115 Termination date: 20210302 |