CN110034596B - Multi-base-station charging method based on SOM neural network in WRSNs - Google Patents
Multi-base-station charging method based on SOM neural network in WRSNs Download PDFInfo
- Publication number
- CN110034596B CN110034596B CN201910282982.0A CN201910282982A CN110034596B CN 110034596 B CN110034596 B CN 110034596B CN 201910282982 A CN201910282982 A CN 201910282982A CN 110034596 B CN110034596 B CN 110034596B
- Authority
- CN
- China
- Prior art keywords
- charging
- path
- nodes
- primary mobile
- neural network
- 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.)
- Active
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 title claims abstract description 20
- 238000005265 energy consumption Methods 0.000 claims abstract description 14
- 230000002068 genetic effect Effects 0.000 claims abstract description 11
- 238000012549 training Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 210000002569 neuron Anatomy 0.000 description 15
- 230000006870 function Effects 0.000 description 8
- 238000013507 mapping Methods 0.000 description 6
- 108090000623 proteins and genes Proteins 0.000 description 6
- 238000012545 processing Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000035772 mutation Effects 0.000 description 3
- 230000000737 periodic effect Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000007423 decrease Effects 0.000 description 2
- 210000004205 output neuron Anatomy 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 210000000349 chromosome Anatomy 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- H02J7/0003—
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/18—Network planning tools
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W84/00—Network topologies
- H04W84/18—Self-organising networks, e.g. ad-hoc networks or sensor networks
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/70—Energy storage systems for electromobility, e.g. batteries
Abstract
The invention discloses a multi-base station charging method based on an SOM neural network in WRSNs, which comprises the following steps: classifying the wireless chargeable sensor network by utilizing an SOM neural network; dividing the second level of the network into areas with the same number of sensor nodes according to the energy consumption of the sensor and the battery capacity of the charger, and allocating a first-level mobile charger; and optimally planning the traversal path of the primary mobile charger by using a genetic algorithm and selecting a resident point on the path. Planning an optimal path by utilizing a genetic algorithm according to the residence point, and arranging a secondary mobile charger and a base station on the path; in each charging period, the primary mobile charger charges the sensor node, and after all the primary mobile chargers are charged, the secondary mobile charger starts to charge the primary mobile charger. The invention has the following beneficial effects: the algorithm is low in complexity, the sensor classes can be accurately divided, the full-coverage charging of the sensor network is realized, and the mobile energy loss and the charging loss are reduced.
Description
Technical Field
The invention belongs to the technical field of wireless sensor networks, and particularly relates to a multi-base-station charging method based on an SOM neural network in high-density low-power-consumption WSNs.
Background
In the industrial Internet of things, the sensor network is widely applied. However, sensors are added to rotating equipment to monitor the operating conditions of the equipment, large coal mine environmental monitoring, etc., all of which are not possible using wired connections. Therefore, the wireless sensor network is developed vigorously, is often connected with a low-power-consumption wide area network, and has the characteristics of large size and high density. The key for promoting the development of the wireless sensor network in the industrial Internet of things is to solve the problem of limited energy of sensor nodes.
Wireless charging sensor networks have gained attention in recent research, including planning mobile charging vehicle paths, making charging decisions, and designing mobile charging vehicle coordination. In some studies, mobile chargers periodically move through a wireless sensor network and charge sensors within their range. Research has mostly planned mobile charger paths to maximize the time the mobile charger stays at a Base Station (BS). Some studies have also proposed on-demand mobile charging strategies to maximize the efficiency of mobile charger operation. In addition, these solutions all use a single base station equipped with one or more mobile chargers to charge the sensors. However, the number of wireless sensor network nodes in the context of industrial internet of things is large, and they may not cover enough sensors, especially for energy-limited. In addition, the length of a moving path of the mobile charging vehicle is not considered in most of the research schemes, and a large amount of moving energy loss is caused by an overlong moving path in a large wireless sensor network under the background of the industrial internet of things.
In a recent study, in 2017, "scalable mobile switching policy for permanent operation in large-scale wireless rechargeable sensor networks" proposed that a network be divided into multiple regions, with nodes having similar energy consumption in each region. To improve charging efficiency, the authors propose an adaptive charging algorithm in which the variation of the mobile charger is based on the energy consumption of each zone. However, in the background of the industrial internet of things, the density of the wireless sensor network is high, and accurate division is difficult to realize by the algorithm region division method.
In 2015, Sheng Zhang et al, "capacitive Mobile Charging" suggested that Mobile chargers could not only charge sensors, but also charge each other. The authors propose a theoretical cooperative charging scheme named PushWait to maximize charging efficiency. However, the algorithm involves frequent charging between charging vehicles, and each charging between charging vehicles causes a charging loss.
In summary, although there have been great advances in wirelessly rechargeable networks. However, since the wireless chargeable network in the background of the industrial internet of things has the characteristics of large size and high density, there are still some problems to be researched in a targeted way:
(1) the large wireless chargeable sensor network has the problem that all sensors are difficult to charge and cover;
(2) the problem of large amount of mobile energy loss is caused by overlong mobile paths in a large wireless sensor network;
(3) the problem that accurate division is difficult to realize in the area division of the high-density wireless sensor network;
(4) a large wireless sensor network relates to the problem of large charging loss caused by frequent charging between charging vehicles.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and provides a multi-base station charging method based on an SOM neural network in low-power WRSNs in the background of an industrial Internet of things.
In order to solve the problems in the prior art, the invention discloses a multi-base-station charging method based on an SOM neural network in WRSNs with high density and low power consumption, which comprises the following steps:
(1) classifying the wireless chargeable sensor network into classes by utilizing an SOM neural network according to the characteristics of energy consumption, residual energy, position and the like of the sensor nodes;
(2) before the charging operation begins, dividing the ith level of the network into k with the number of sensor nodes being approximately equal according to the energy consumption of the sensor and the battery capacity of the mobile charging vehicle i Each area is provided with a primary mobile charger, so that the nodes can be fully charged;
(3) and performing optimal path planning on the traversal path of the primary mobile charger by using a genetic algorithm. Selecting residence points on the optimal traversal path of the primary mobile charging vehicle, planning the shortest path according to the residence points by using a genetic algorithm again, and configuring a secondary mobile charger and setting a base station on the optimal path; the primary mobile charging vehicle is sent out from the residence point, traverses all sensor nodes in the area where the primary mobile charging vehicle is located, charges the sensor nodes, and stays at the residence point after the charging is finished;
(4) and in each charging period, the primary mobile charger charges the sensor nodes, and after all the primary mobile charger nodes are charged, the secondary mobile charger starts to traverse the resident points to charge the primary mobile charger.
Further, the SOM neural network training data set in step (1) may be obtained as follows: and sending a broadcast packet to the whole wireless sensor network, and collecting the energy consumption rate of all nodes of the whole wireless sensor network, the residual energy of the nodes and the abscissa and the ordinate corresponding to the positions of the nodes. Selecting nodes (such as nodes on a key path) with typical hop count and node residual energy characteristics, determining the categories of the nodes, and combining energy consumption rates, residual energy and sink distance corresponding to the nodes to prepare SOM neural network learning data.
Further, the number k of regions in the step (2) i The calculation is as follows:
wherein N is i The representation indicates the number of i-th sensors, s i The number of the i-th sensors which can be charged by each charging vehicle is represented, so that almost exhausted charging energy sources are guaranteed after the charging of the primary mobile charger is finished, and the charging efficiency is improved.
Further, the staying point in the step (3) is a staying point of the primary charging moving vehicle. In order to facilitate the secondary mobile charging vehicle to charge the primary mobile charging vehicle, when the primary mobile charging vehicle charges the nodes of the served area in a traversing manner, the starting point and the end point of the traversing path should coincide, and the coincident point is the residence point of the primary mobile charging vehicle.
Further, the rule for selecting the parking point of the primary mobile charging vehicle is as follows: and selecting the node closest to the sink in each region as an initial position, and then optimizing by a triangle rule to obtain the position of the residence point.
Further, the equipping rule of the secondary mobile charging vehicle in the step (3) is as follows: according to the number of the primary mobile charging cars and the number of the secondary mobile charging cars capable of charging the primary mobile charging cars, the obtained shortest path is divided into several sections with the number close to equal to that of the primary mobile charging cars, so that the primary mobile charging cars can be fully charged.
Further, the setting up rule of the base station in step (3) is as follows: and a charging base station is arranged at the intersection of any two paths of the secondary mobile charging trolleys, and the charging base station is used for completely charging the arriving secondary mobile charging trolleys once in one period.
Further, the length of the charging period in the step (4) is far shorter than the life of the sensor. When a charging cycle begins, the secondary mobile charging vehicle charges in the base station at one end of the charging path. After the primary mobile charging vehicle returns to the residence point after traversing the nodes, the secondary mobile charging vehicle charges the primary mobile charging vehicle along the charging path and finally reaches the base station at the other end of the charging path. Similarly, in the next charging cycle, the secondary mobile charging vehicle starts from the base station at the other end of the charging path and returns to the other end along the charging path.
The invention has the following beneficial effects:
1. charging coverage of all sensor nodes of the large-scale wireless sensor network under the background of the industrial Internet of things can be realized through cooperative cooperation of multiple base stations and multiple chargers;
2. the shortest moving path of the charger is planned, and part of the charger stays at a residence point without returning to a base station, so that the loss of moving energy is reduced;
3. the accurate and reasonable division of the high-density wireless sensor network is realized through the accurate classification of the sensor nodes by the SOM neural network;
4. the charging loss is reduced by reducing frequent charging among charging vehicles;
5. the charging strategy adopts a genetic algorithm to obtain an approximate optimal solution, and the complexity of the algorithm is reduced.
Drawings
FIG. 1 is a schematic diagram illustrating classification in step (1);
FIG. 2 is a schematic view of the divided regions in step (2);
fig. 3 is a schematic diagram illustrating the planning of the secondary mobile charging vehicle traverse path and the setup of the charging base station in step (3);
FIG. 4 is a schematic diagram of the SOM neural network in step (1) according to the preferred embodiment;
FIG. 5 is a schematic diagram of the principle of triangle rule.
Reference numerals are as follows:
2-1, planning a path of the primary mobile charging vehicle;
3-1, a primary mobile charging vehicle parking point; 3-2 charging the base station.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
A multi-base-station charging method based on an SOM neural network in high-density low-power-consumption WRSNs under the background of an industrial Internet of things comprises the following steps:
(1) classifying the wireless chargeable sensor network into categories by utilizing an SOM neural network according to the characteristics of energy consumption, residual energy, position and the like of the sensor nodes, as shown in FIG. 1;
(2) before the charging operation is started, dividing the ith level of the network into k with the number of sensor nodes close to equal according to the energy consumption of the sensor and the battery capacity of the mobile charging vehicle i Each area is provided with a primary mobile charger, so that the nodes can be fully charged, as shown in FIG. 2;
(3) and performing optimal path planning on the traversal path of the primary mobile charger by using a genetic algorithm. Selecting residence points on the optimal traversal path of the primary mobile charging vehicle, planning the shortest path according to the residence points by using a genetic algorithm again, and configuring a secondary mobile charger and setting a base station on the optimal path, as shown in fig. 3; the primary mobile charging vehicle is sent out from the residence point, traverses all sensor nodes in the area and charges, and stays at the residence point after the charging is finished;
(4) and in each charging period, the primary mobile charger charges the sensor nodes, and after all the primary mobile charger nodes are charged, the secondary mobile charger starts to traverse the residence point to charge the primary mobile charger.
Preferably, as shown in fig. 4, a typical SOM neural network is selected in step (1), and the topology structure thereof is composed of an input layer and a mapping layer. The input layer is composed of m neurons, a b neurons of the mapping layer form a two-dimensional planar array, and the neurons of the input layer are fully connected with the neurons of the competition layer. The network model of the SOM neural network consists of the following 4 parts:
an array of processing units: for receiving event inputs and forming a "discriminant function" on these signals;
and (3) a comparison selection mechanism: a processing unit for comparing the 'discriminant functions' and selecting one having the largest function output value;
local interconnection: for simultaneously energizing selected processing elements and their adjacent processing elements;
the self-adaptive process: for modifying the parameters of the energized processing element to increase its output value corresponding to a particular input "discriminant function".
And sending a broadcast packet to the whole wireless sensor network, and collecting the energy consumption rate of all nodes of the whole wireless sensor network, the residual energy of the nodes and the abscissa and the ordinate corresponding to the positions of the nodes. Selecting nodes (such as nodes on a critical path) with typical hop count and node residual energy characteristics, determining the categories of the nodes, and combining energy consumption rates, residual energy and a sink distance corresponding to the nodes to prepare SOM neural network learning data (obviously, an input vector of a training set is three-dimensional, and m is 3).
More specifically, according to the obtained SOM neural network learning data set, a Kohonen self-organizing feature mapping algorithm is selected to find out the similarity between input data, and the input with high similarity is configured nearby on the network. The learning algorithm comprises the following steps:
1. network initialization: and assigning an initial value to the weight value connected between the input layer and the mapping layer, wherein the assignment process of the initial value is random. That is, the assigned weights are all weights for the full connection of the input layer neurons to the contention layer neurons, and the assigned weights are smaller. The set of j adjacent neurons of the output neuron is S j Let S stand out j (t) represents the set of "adjacent neurons" at time t and decreases with increasing time.
2. Input of the input vector: the input vector is X ═ X 1 ,x 2 ,x 3 ,...x m ) T Input layer neurons receive input.
3. And calculating Euclidean distance according to the input vector and the weight vector of the mapping layer, wherein the Euclidean distance is shown as the following formula:
wherein d is j Is the distance between the jth neuron of the competition layer and the input vector, w ij Is the weight between the i neuron of the input layer and the j neuron of the mapping layer. Screening to obtain the neuron with the shortest Euclidean distance as the winning neuron, i.e. determining a unit k so that any j has
4. Learning the weight value: updating the weights of the output neuron and the adjacent neuron as shown in the following formula:
Δw ij ＝w ij (t+1)-w ij (t)＝η(t)(x i (t)-w ij (t))
wherein the function η (t) ranges from 0 to 1 and gradually decreases to 0 with time. The following may be taken:
5. computing the output o k
o k ＝f(min||X-W j ||)。
6. Judging whether the learning requirement is met: if the requirement is met, the learning is finished; otherwise, returning to the step (2) to carry out the next round of learning.
As a preferable mode, k in the step (2) i Is calculated asThe following:
wherein, N i The representation indicates the number of i-th sensors, s i The number of the i-th sensors which can be charged by each charging vehicle is represented, so that almost exhausted charging energy sources are ensured after the charging of the primary mobile charger is finished, and the charging efficiency is improved.
Preferably, the genetic algorithm in step (3) is modeled as follows:
step1 encoding strategy
The invention adopts a decimal coding strategy. The sequence obtained by coding and the individuals in the population are ensured to be in one-to-one correspondence. I.e. using a random decimal sequence W 1 W 2 ...W n (n is the number of nodes or dwell points traversed by the charging vehicle) represents the chromosomes of an individual in the population, where 0<W i <1，W 1 ＝0，W n ＝1。
Step2 initial population
Using modified circle method to find a better initial population, i.e. for the initial circle
C＝W 1 ...W u-1 W u W u+1 ...W v-1 W v W v+1 ...W n
2≤u≤v≤n
The order between u and v is transformed, at which time the new path is:
W 1 ...W u-1 W v W v-1 ...W u+1 W u W v+1 ...W n
order to
If and only if Δ <0, modify the old path with the new path.
Step3 fitness function
The fitness function is the path length through all nodes, i.e.:
step4 crossover operation
For two selected parent individuals, the t-th gene is randomly selected as a cross point, and the filial generation codes obtained after cross operation are s 1 And s 2 ，s 1 The gene of (a) consisting of 1 The first t genes and f 2 The last n-t genes of (a) are composed of 2 The same applies to the genes of (1).
The method of the cross operation has a plurality of choices, and a good cross mode should be selected as far as possible to ensure that the offspring can inherit the excellent characteristics of the parent, and the cross operation also contains mutation operation.
Step5 mutation operation
Mutation is also a means to achieve population diversity and is also a guarantee for global optimization. Randomly taking three integers for the selected variant individuals according to a given variant rate to satisfy the requirement
1<u<v<w<n
The gene segment between u and v is inserted behind w.
Step6 selection
And a deterministic selection strategy is adopted, namely, the evolution with the minimum objective function value is selected to the next generation, so that the excellent characteristics of the parent generation can be guaranteed to be preserved.
Preferably, the dwell point in step (3) is the dwell point of the primary mobile moving vehicle. In order to facilitate the secondary mobile charging vehicle to charge the primary mobile charging vehicle, when the primary mobile charging vehicle charges the nodes of the served area in a traversing manner, the starting point and the end point of the traversing path should coincide, and the coinciding point at this time is the residence point of the primary mobile charging vehicle.
Preferably, the residence point selection rule of the primary mobile charging vehicle is as follows: and selecting the node closest to the sink in each region as an initial position, planning a path through a genetic algorithm according to the initial position of the dwell point, and optimizing through a triangle rule to obtain the position of the dwell point.
More specifically, as shown in fig. 5, the triangle rule is as follows: the current residence point, the adjacent residence point B and the adjacent residence points A and C form a triangle, and nodes which are positioned in the triangle and in the area of the residence point B meet the following two rules:
1. the lower the height, the shorter the AB '+ B' C path;
2.
when the AB '+ B' C path is shortest.
Preferably, the equipping rule of the secondary mobile charging vehicle in the step (3) is as follows: according to the number of the primary mobile charging vehicles and the number of the secondary mobile charging vehicles capable of charging the primary mobile charging vehicles, the obtained shortest path is divided into several sections with the number close to equal to that of the primary mobile charging vehicles, so that the primary mobile charging vehicles can be fully charged.
Preferably, the base station setting rule in step (3) is as follows: and a charging base station is arranged at the intersection of any two paths of the secondary mobile charging vehicles, and the charging base station is used for completely charging the arriving secondary mobile charging vehicles once in a period.
Preferably, the charging period in step (4) is much shorter than the sensor lifetime. When a charging cycle begins, the secondary mobile charging vehicle charges in the base station at one end of the charging path. After the primary mobile charging vehicle returns to the residence point after traversing the nodes, the secondary mobile charging vehicle charges the primary mobile charging vehicle along the charging path and finally reaches the base station at the other end of the charging path. Similarly, in the next charging cycle, the secondary mobile charging vehicle starts from the base station at the other end of the charging path and returns to the other end along the charging path.
As a preferred scheme, the first deployment process of the primary mobile charging vehicle, the secondary mobile charging vehicle and the base station is as follows: and the sink sends a broadcast packet to the whole wireless sensor network to prepare SOM neural network learning data. And then, the sink classifies the wireless sensor network and divides the area by using the trained SOM neural network, and sends the path of the traversing sensor in the area corresponding to the first-stage mobile charging vehicle, the position information of the corresponding residence point and the periodic charging information. And then, the sink sends the position information and the periodic charging information of the path corresponding to the secondary mobile charging vehicle and the base stations at the two ends. Subsequently, the sink sends location information for setting up a plurality of base stations to an administrator. And finally, after the base station is set up, the primary mobile charging car reaches a residence point and the secondary mobile charging car reaches the corresponding base station, and the periodic charging is started.
Claims (5)
1. A multi-base-station charging method based on an SOM neural network in WRSNs is characterized in that: the method comprises the following steps:
(1) classifying the wireless chargeable sensor network by utilizing an SOM neural network;
the training data set of the SOM neural network in the step (1) can be obtained by: sending a broadcast packet to the whole wireless sensor network, and collecting the energy consumption rate of all nodes of the whole wireless sensor network, the residual energy of the nodes, and the abscissa and the ordinate corresponding to the positions of the nodes; selecting nodes with typical hop count and node residual energy characteristics, determining the categories of the nodes, and combining energy consumption rates, residual energy and sink distances corresponding to the nodes to prepare SOM neural network learning data;
(2) before the charging operation begins, dividing the ith level of the network into k with the number of sensor nodes being approximately equal according to the energy consumption of the sensor and the battery capacity of the mobile charging vehicle i Each area is provided with a primary mobile charger, so that the nodes can be fully charged;
(3) performing optimal path planning on a traversal path of the primary mobile charger by using a genetic algorithm; selecting a residence point on the optimal traversal path of the primary mobile charging vehicle, planning a shortest path according to the residence point by using a genetic algorithm again, and configuring a secondary mobile charger and establishing a base station on the optimal path; the primary mobile charging vehicle is sent out from the residence point, traverses all sensor nodes in the area and charges, and stays at the residence point after the charging is finished;
the residence point in the step (3) is the residence point of the primary mobile charging vehicle, so that the primary mobile charging vehicle can be charged by the secondary mobile charging vehicle, when the primary mobile charging vehicle is used for traversing and charging the nodes in the served area, the starting point and the end point of the traversing path should be overlapped, and the overlapped point is the residence point of the primary mobile charging vehicle;
the secondary mobile charging vehicle in the step (3) is provided with the following rules: dividing the obtained shortest path into a plurality of sections with the number close to equal to that of the primary mobile charging vehicles according to the number of the primary mobile charging vehicles and the number of the secondary mobile charging vehicles capable of charging the primary mobile charging vehicles, thereby ensuring that the primary mobile charging vehicles can be fully charged;
(4) in each charging period, the primary mobile charger charges the sensor nodes, and after all the primary mobile charger nodes are charged, the secondary mobile charger starts to traverse the resident points to charge the primary mobile charger.
2. The SOM neural network-based multi-base-station charging method in WRSNs according to claim 1, wherein the number of regions k in the step (2) i The calculation is as follows:
wherein, N i Indicating the number of class i sensors, s i The number of the i-th sensors which can be charged by each charging vehicle is represented, so that almost exhausted charging energy sources are guaranteed after the charging of the primary mobile charger is finished, and the charging efficiency is improved.
3. The method of claim 1, wherein the SOM neural network based multi-base station charging method in WRSNs comprises the following dwell point selection rules: and selecting the node closest to the sink in each region as an initial position, and then optimizing by a triangle rule to obtain the position of the residence point.
4. The SOM neural network-based multi-base station charging method in WRSNs according to claim 1, wherein the rules for setting up the base stations in said step (3) are as follows: and a charging base station is arranged at the intersection of any two secondary mobile charging vehicles, and the charging base station is used for completely charging the arriving secondary mobile charging vehicles once in a period.
5. The method for charging multiple base stations based on the SOM neural network in WRSNs according to claim 1, wherein the charging period in the step (4) is much shorter than the life of the sensor; when a charging period begins, the secondary mobile charging vehicle charges in a base station at one end of a charging path; after the primary mobile charging vehicle returns to the residence point after traversing the nodes, the secondary mobile charging vehicle charges the primary mobile charging vehicle along the charging path and finally reaches the base station at the other end of the charging path; in the next charging cycle, the secondary mobile charging vehicle starts from the base station at the other end of the charging path, and returns to the other end along the charging path.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910282982.0A CN110034596B (en) | 2019-04-10 | 2019-04-10 | Multi-base-station charging method based on SOM neural network in WRSNs |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910282982.0A CN110034596B (en) | 2019-04-10 | 2019-04-10 | Multi-base-station charging method based on SOM neural network in WRSNs |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110034596A CN110034596A (en) | 2019-07-19 |
| CN110034596B true CN110034596B (en) | 2022-08-19 |
Family
ID=67237858
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910282982.0A Active CN110034596B (en) | 2019-04-10 | 2019-04-10 | Multi-base-station charging method based on SOM neural network in WRSNs |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110034596B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110470301B (en) * | 2019-08-13 | 2020-12-11 | 上海交通大学 | Unmanned aerial vehicle path planning method under multi-dynamic task target point |
| CN111126712B (en) * | 2019-12-30 | 2023-09-01 | 长安大学 | Parking charging transfer system planning method for commuting corridor |
| CN112327838B (en) * | 2020-10-29 | 2023-03-31 | 哈尔滨工程大学 | Multi-unmanned surface vessel multi-task allocation method based on improved self-mapping algorithm |
| CN115723610A (en) * | 2021-08-26 | 2023-03-03 | 博泰车联网科技(上海)股份有限公司 | Vehicle charging planning method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105979488A (en) * | 2016-07-26 | 2016-09-28 | 河海大学常州校区 | Collaborative charging algorithm based on region partition in wireless sensor network |
| CN107835499A (en) * | 2017-10-26 | 2018-03-23 | 河海大学常州校区 | A kind of mobile charging method based on sub-clustering and energy relay in WSNs |
| CN108448731A (en) * | 2018-03-07 | 2018-08-24 | 南京邮电大学 | A kind of collaborative wireless sense network energy compensation process and its wireless sense network |
-
2019
- 2019-04-10 CN CN201910282982.0A patent/CN110034596B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105979488A (en) * | 2016-07-26 | 2016-09-28 | 河海大学常州校区 | Collaborative charging algorithm based on region partition in wireless sensor network |
| CN107835499A (en) * | 2017-10-26 | 2018-03-23 | 河海大学常州校区 | A kind of mobile charging method based on sub-clustering and energy relay in WSNs |
| CN108448731A (en) * | 2018-03-07 | 2018-08-24 | 南京邮电大学 | A kind of collaborative wireless sense network energy compensation process and its wireless sense network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110034596A (en) | 2019-07-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110034596B (en) | Multi-base-station charging method based on SOM neural network in WRSNs | |
| Qian et al. | Enhanced coordinated operations of electric power and transportation networks via EV charging services | |
| CN109583665B (en) | Unmanned aerial vehicle charging task scheduling method in wireless sensor network | |
| CN109495945A (en) | A kind of collaboration charging method based on cluster sub-clustering in WSNs | |
| Rostami et al. | Wireless sensor network clustering using particles swarm optimization for reducing energy consumption | |
| CN110276517A (en) | A kind of electric automobile charging station site selecting method based on MOPSO algorithm | |
| Raj et al. | A survey on cluster head selection and cluster formation methods in wireless sensor networks | |
| CN104883388A (en) | Car networking road-side unit deployment method based on genetic algorithm | |
| Banerjee et al. | An integrated application of IoT‐based WSN in the field of indian agriculture system using hybrid optimization technique and machine learning | |
| CN109871619B (en) | Static charging pile deployment method based on grid division | |
| Li et al. | A many-objective optimization charging scheme for wireless rechargeable sensor networks via mobile charging vehicles | |
| Zhou et al. | Intelligent path planning strategy for electric vehicles combined with urban electrified transportation network and power grid | |
| CN110048483B (en) | Multi-base-station cooperative charging method based on SOM neural network in high-power-consumption WRSNs | |
| CN114827931A (en) | WSN energy efficiency optimization routing method based on multi-agent reinforcement learning | |
| CN107295081B (en) | Combined routing optimization energy supplementing device and energy supplementing method thereof | |
| Prakash et al. | Mixed Linear Programming for Charging Vehicle Scheduling in Large-Scale Rechargeable WSNs | |
| Wagh et al. | Optimal route selection for vehicular adhoc networks using lion algorithm | |
| Wang et al. | Charging path optimization for wireless rechargeable sensor network | |
| CN113365222B (en) | Mobile sensor intelligent track design method based on sustainable data acquisition | |
| Shehab et al. | Self-organizing single-hop wireless sensor network using a genetic algorithm: Longer lifetimes and maximal throughputs | |
| Zhang et al. | An immune chaotic adaptive evolutionary algorithm for energy-efficient clustering management in LPWSN | |
| CN112243281A (en) | Bayesian game routing design method based on rechargeable wireless sensor network | |
| CN110134121A (en) | A kind of optimum path planning method of mobile charger | |
| CN113727275B (en) | Unmanned aerial vehicle-assisted wireless sensor network node charging selection method | |
| Yi et al. | Energy balancing and path plan strategy for rechargeable underwater sensor network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |