Method for constructing and optimizing energy transmission path in wireless power transmission network

Abstract

The invention relates to energy supply of load nodes in a wireless power transmission network, in particular to a method for constructing and optimizing an energy transmission path in the wireless power transmission network. The invention discloses a method for constructing and optimizing an energy transmission path in a wireless power transmission network. The method comprises the steps of constructing a sub-network based on a DBSCAN clustering algorithm, constructing an inter-cluster energy transmission path based on a Prim algorithm and optimizing an intra-cluster energy transmission path based on a breadth-first search algorithm. The method comprises the following steps: firstly, dividing a whole wireless power transmission network into a plurality of clusters by adopting a density-based clustering algorithm DBSCAN, constructing a shortest energy transmission path from a cluster head to each load node in the cluster through a breadth-first search algorithm, and constructing a shortest path between the clusters on the basis of a Prim minimum spanning tree algorithm; therefore, the optimal energy transmission of the whole network is realized.

Description

Technical Field

[0001] The present invention relates to energy supply of load nodes in a wireless power transmission network, and in particular to a method for constructing and optimizing an energy transmission path in a wireless power transmission network. Background Art

[0002] Wireless power transmission network (WPTN) consists of nodes and power transmission routes, which can supply power to load nodes. In wireless power transmission network, due to the complex cross-coupling between nodes and the random and dynamic distribution of load nodes, it is very difficult to meet the stable energy supply of each load node and ensure the optimal energy transmission efficiency of the system as a whole. Summary of the invention

[0003] The present invention aims at solving the problem that the energy supply efficiency of load nodes is low due to cross-coupling and random distribution between nodes in a wireless power transmission network, and provides a method for constructing and optimizing an energy transmission path in a wireless power transmission network.

[0004] The present invention is implemented by adopting the following technical solutions: a method for constructing and optimizing energy transmission paths in a wireless power transmission network, including constructing subnetworks based on a DBSCAN clustering algorithm, constructing inter-cluster energy transmission paths based on a Prim algorithm, and optimizing intra-cluster energy transmission paths based on a breadth-first search algorithm;

[0005] The implementation steps for constructing a subnetwork based on the DBSCAN clustering algorithm are as follows:

[0006] Step 1: Define a set D containing the coordinates of all load nodes in the wireless power transmission network = {P1, P2, P3, ..., P i ,…,P j ,…,P n}, neighborhood radius Eps and neighborhood density threshold MinPts;

[0007] Step 2: Select any unprocessed load node P from set D. i Starting from, with Eps as the radius, traverse and search for all load nodes in the network. In this process, it is necessary to calculate the distance between two load nodes to determine whether the two are directly density-reachable and form a cluster;

[0008] Step 3: If the number of load nodes covered within the Eps range is not less than the pre-set neighborhood density threshold MinPts, then these load nodes are merged into the neighborhood density threshold Pts. i In a cluster C where is the core point;

[0009] Step 4: If a load node P j, if the number of load nodes covered by its neighborhood range does not reach the neighborhood density threshold MinPts, the load node will be temporarily treated as a noise point;

[0010] Step 5: Visit other unprocessed load nodes in turn to form several core points, noise points and clusters, and merge and expand the load nodes and clusters that meet direct density reachability or density reachability;

[0011] Step 6: Repeat steps 2 to 5 until no new load nodes are merged into any cluster or marked as noise points, and then terminate the algorithm;

[0012] Step 7: Output clustering results;

[0013] The implementation steps of constructing the inter-cluster energy transfer path based on Prim's algorithm are as follows:

[0014] Step 1: Define the shortest distance between every two clusters constructed based on the DBSCAN clustering algorithm as the weight, and mark the coordinates of the nodes that constitute the distance;

[0015] Step 2: Select any cluster and add it to the minimum spanning tree;

[0016] Step 3: Determine whether the minimum spanning tree contains all clusters. If not, find the edge with the smallest weight from other adjacent clusters and add this edge and the cluster connected to this edge to the minimum spanning tree.

[0017] Step 4: Repeat step 3 until all clusters are included in the minimum spanning tree;

[0018] The implementation steps of cluster energy transfer path optimization based on breadth-first search algorithm are as follows:

[0019] Step 1: Define the node position coordinate matrix, output tree, traversed node coordinate matrix and untraversed node coordinate matrix in each cluster constructed based on DBSCAN clustering algorithm;

[0020] Step 2: Grow one level of output tree starting from the initial node;

[0021] Step 3: Add a node in the tree to the explored node coordinate matrix Explored_node_matrix, and add other nodes in the same level tree to the unexplored node coordinate matrix Unexplored_node_matrix, and then construct the energy transmission path between the nodes of the two levels of trees according to the transmission law of energy flow;

[0022] Step 4: Move the nodes from the unexplored_node_matrix to the explored_node_matrix until all nodes in the same level of the tree have been moved, return to step 2, and grow the output tree by one level;

[0023] Step 5: Continue to construct the energy transfer path between the two-level trees until all nodes are explored;

[0024] Step 6: Compare the number of nodes in different paths in the traversed node coordinate matrix Explored_node_matrix, and define the path with the least number of nodes as the optimal path.

[0025] The above-mentioned method for constructing and optimizing the energy transmission path in the wireless power transmission network uses the modified Minkowski distance to calculate the distance between two load nodes. The two nodes in the wireless power transmission network are P i (a pi1 ,a pi2 ) and P j (a pj1 ,a pj2 ), then the Minkowski distance D between two nodes M It is expressed as: The modified Minkowski distance between two nodes can be expressed as: D MTD =1++2×||a pi ,a pj || ∞ +P(||a pi ,a pj ||1)Q(||a pi ,a pj ||1)less(a pi -a pj ), where: In the above formula, P is the relative position correction factor, Q is the starting position correction factor, less is the logic function, when k = 1, it means calculating the Manhattan distance, and when k = infinity, it means calculating the Chebyshev distance.

[0026] The above-mentioned method for constructing and optimizing the energy transmission path in the wireless power transmission network, the shortest distance D between clusters in the Prim algorithm CMTD for A and B are any two clusters constructed by the DBSCAN clustering algorithm, N (A) and N (B) The shortest distance D between any two nodes in cluster A and cluster B is CMTD It is the shortest distance among all the nodes between two clusters.

[0027] The above-mentioned method for constructing and optimizing the energy transmission path in the wireless power transmission network is particularly suitable for an array-type wireless power transmission network. The array-type wireless power transmission network is a wireless power transmission network in which the nodes in the wireless power transmission network are arranged in an array structure in advance, and the transmitting coil and the receiving coil of every two adjacent nodes in the array are opposite, that is, the receiving coils between adjacent nodes are staggered, and the distance between them is relatively far, so the mutual inductance between them is very small, even close to 0, satisfying the constraint condition of node decoupling; after the array-type wireless power transmission network is formed, the method of the present invention is applied to construct and optimize the energy transmission path.

[0028] The present invention firstly adopts a density-based spatial clustering algorithm (Density-Based Spatial Clustering of Applications with Noise, DBSCAN) to divide the entire wireless power transmission network into multiple clusters (sub-networks), and constructs the shortest energy transmission path from the cluster head to each load node in the cluster through a breadth-first search algorithm, and then constructs the shortest path between each cluster based on the Prim minimum spanning tree algorithm, thereby achieving optimal energy transmission of the entire network. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 Flowchart of the application of DBSCAN clustering algorithm in WPTN.

[0030] Figure 2 Schematic diagram of clustering results based on DBSCAN clustering algorithm.

[0031] Figure 3 Schematic diagram of the shortest energy transfer path between clusters.

[0032] Figure 4 Schematic diagram of constructing energy transmission paths in wireless power transmission networks based on breadth-first search algorithm.

[0033] Figure 5 Schematic diagram of the construction process of the optimal energy transmission path within the cluster.

[0034] Figure 6 Schematic diagram of the optimal energy transmission path within a sub-cluster.

[0035] Figure 7 Schematic diagram of a 3×3 array wireless power transmission network. DETAILED DESCRIPTION

[0036] In the WPT system, ensuring efficient energy transmission of the system is a major goal of wireless power transmission technology research. In the wireless power transmission network, the network domain contains a large number of nodes and energy transmission paths. In different energy transmission paths, the transmission efficiency of the system varies greatly. Therefore, the present invention optimizes the energy transmission path of the system and selects the most efficient path to transmit energy from the power node to each load node to ensure efficient energy transmission of the system as a whole. As we all know, the fewer the number of relay nodes, the lower the loss and the higher the efficiency of the system. Therefore, the present invention will optimize the transmission path and transmission efficiency of the system with the goal of finding the least number of relay nodes. It can be seen that the optimization goal of the energy transmission path in the present invention can be described as: 1) meet the power requirements of all load nodes; 2) the number of relay nodes in the path is minimal. Construction of subnetwork based on DBSCAN clustering algorithm

[0037] As the scale of WPTN expands and the number of load nodes increases, directly constructing energy transmission paths between multiple load nodes is not only inefficient, but also very complicated. Therefore, the present invention uses the DBSCAN clustering algorithm to divide the entire network into multiple sub-networks to improve the timeliness of energy transmission path construction in WPTN.

[0038] The DBSCAN clustering algorithm is a typical density-based algorithm. Given N data points {P1, P2, P3, ..., P i ,…,P j ,…,P n}, the DBSCAN algorithm can be used to divide data points into multiple clusters and mark noise. According to the two parameters Eps and MinPts, all data points can be divided into core points, density-reachable points, and noise. If p i It is a core point, which forms a cluster with all density-reachable points. Each cluster has at least one core point, and boundary points can also be part of a cluster. The DBSCAN clustering algorithm does not need to pre-set the number of clusters, and it can effectively identify clusters with uniform density. The core idea of ​​the DBSCAN clustering algorithm can be described as: starting from a specified core point, gradually diffusing to the density-reachable range, and finally obtaining a maximum area containing core points and boundary points, and any two points in the area are connected by density.

[0039] Figure 1 The workflow of DBSCAN clustering algorithm to build sub-network in WPTN is shown. For the convenience of research, corresponding to the sample set D consisting of multiple load nodes in WPTN, several core definitions of DBSCAN clustering algorithm are given below:

[0040] (1) Neighborhood radius Eps: The area determined by taking a given object as the center and Eps as the radius is the neighborhood of the object;

[0041] (2) Neighborhood density threshold MinPts: the minimum number of objects in a neighborhood determined by taking a given object as the center and Eps as the radius;

[0042] (3) Core point: For a given object p i (p i ∈D), if p i If the Eps neighborhood of p contains at least MinPts objects, then p i As the core point;

[0043] (4) Boundary point: For a given object p i (p i ∈D), if p i In the neighborhood of a core point, but the number of points in its neighborhood Eps is less than MinPts, then p i is the boundary point;

[0044] (5) Noise point: object p i does not belong to any cluster in the data set, then p i It is a noise point;

[0045] (6) Direct density reachability: In a given set of objects, if object p i is a core point, P j In p i In the Eps neighborhood, it is called p i To P j Direct density is reachable;

[0046] (7) Density can be reached: If P is satisfied i+1 YesP i With respect to Eps and MinPts, the density is directly reachable, so P1 to P n Density can reach;

[0047] (8) Density connection: If in the same object set, p i and P j are all reachable with respect to Eps and MinPts density, then the object p i and object P j are density-connected.

[0048] Algorithm implementation steps:

[0049] Step 1: Define a set D containing the coordinates of all load nodes in the wireless power transmission network = {P1,

[0050] P2, P3, ..., P i ,…,P j ,…,P n}, neighborhood radius Eps and neighborhood density threshold MinPts;

[0051] Step 2: Select any unprocessed load node P from set D. i Starting from, with Eps as the radius, traverse and search for all load nodes in the network. In this process, it is necessary to calculate the distance between two load nodes to determine whether the two are directly density-reachable and form a cluster;

[0052] Step 3: If the number of load nodes covered within the Eps range is not less than the pre-set neighborhood density threshold MinPts, then these load nodes are merged into the neighborhood density threshold Pts. i In a cluster C where is the core point;

[0053] Step 4: If a load node P j , if the number of load nodes covered by its neighborhood range does not reach the neighborhood density threshold MinPts, the load node will be temporarily treated as a noise point;

[0054] Step 5: Visit other unprocessed load nodes in turn to form several core points, noise points and clusters, and merge and expand the load nodes and clusters that meet direct density reachability or density reachability;

[0055] Step 6: Repeat steps 2 to 5 until no new load nodes are merged into any cluster or marked as noise points, and then terminate the algorithm;

[0056] Step 7: Output clustering results.

[0057] From some core definitions and workflows of the DBSCAN clustering algorithm, it is known that in the DBSCAN clustering algorithm, it is necessary to determine whether the nodes are density-reachable by calculating the shortest distance between the load nodes. In previous literature, the distance between the nodes is usually determined by calculating the Euclidean distance, Manhattan distance or Chebyshev distance between the nodes. However, under the energy transmission rules designed by the present invention, the distance calculation methods used in previous literature will no longer be applicable. Therefore, the present invention will calculate the shortest distance between the load nodes based on the modified Minkowski distance calculation method.

[0058] Assume that the coordinates of the two nodes in WPTN are: i (a pi1 ,a pi2 ) and P j (a pj1 ,a pj2 ), then the Minkowski distance (Minkowski Distance, D M ) can be expressed as:

[0059]

[0060] The modified Minkowski distance between two nodes can be expressed as:

[0061] D MTD =1+2×||a pi ,a pj || ∞ +P(||a pi ,a pj ||1)Q(||a pi ,a pj ||1)less(a pi -a pj )

[0062] in:

[0063]

[0064] In the above formula, P is the relative position correction factor, Q is the starting position correction factor, less is the logic function, when k = 1, it means calculating the Manhattan distance, and when k = infinity, it means calculating the Chebyshev distance.

[0065] Construction of inter-cluster energy transfer path based on Prim algorithm

[0066] Based on the above clustering results, the present invention calculates the shortest distance between load nodes of clusters and defines it as weight, and constructs the shortest energy transmission path between clusters based on Prim algorithm. Prim algorithm is an algorithm in graph theory, which can construct a minimum spanning tree in a weighted graph, and the minimum spanning tree constructed by this algorithm not only includes all nodes in the graph, but also the sum of weights of all its edges is minimized. Through the application of Prim algorithm in WPTN, a path can be constructed in a weighted graph composed of various subnetworks to traverse all subnetworks, and the sum of weights of all paths can be minimized, so as to achieve the purpose of minimizing the sum of distances between clusters.

[0067] First, take clusters A and B as examples to define the shortest distance between clusters (Cluster Minimum Transfer Distance, D CMTD )for

[0068]

[0069] N (A) and N (B) The shortest distance D between any two nodes in cluster A and cluster B is CMTD It can be defined as the shortest of the shortest distances between all nodes between two clusters.

[0070] By calculating the shortest distance D between each cluster CMTD , and mark it in Figure 3 Then, based on the Prim algorithm, the inter-cluster energy transmission path shown by the red arrow line in the figure is constructed. The implementation steps of the Prim algorithm are as follows:

[0071] Step 1: Define the shortest distance between every two clusters as the weight and mark the coordinates of the nodes that constitute the distance;

[0072] Step 2: Select any cluster and add it to the minimum spanning tree;

[0073] Step 3: Determine whether the minimum spanning tree contains all clusters. If not, find the edge with the smallest weight from other adjacent clusters and add this edge and the cluster connected to this edge to the minimum spanning tree.

[0074] Step 4: Repeat step 3 until all clusters are included in the minimum spanning tree.

[0075] Breadth-first search algorithm process

[0076] like Figure 4 As shown in (a), in a sub-cluster, N is set in the array structure. 11 is the starting node, N 33 and N 54 is the load node. According to the requirements of optimal energy transmission, energy is transferred from N 11 Transmit to N 33 and N 54 , and the number of relay nodes in the path is the least, which is also defined as the optimal path optimization flow chart as shown in Figure 5 As shown in the figure, the optimization result of the optimal energy transmission path is as follows Figure 6 shown.

[0077] Algorithm implementation steps:

[0078] Step 1: Define the node position coordinate matrix (Node_position_matrix), output tree (Output_tree), explored node coordinate matrix (Explored_node_matrix) and unexplored node coordinate matrix (Unexplored_node_matrix);

[0079] Step 2: Grow one level of output tree starting from the initial node;

[0080] Step 3: Add a node in the tree to the explored node coordinate matrix Explored_node_matrix, and add other nodes in the same level tree to the unexplored node coordinate matrix Unexplored_node_matrix, and then construct the energy transmission path between the nodes of the two levels of trees according to the transmission law of energy flow;

[0081] Step 4: Move the nodes from the unexplored_node_matrix to the explored_node_matrix until all nodes in the same level of the tree have been moved, return to step 2, and grow the output tree by one level;

[0082] Step 5: Continue to construct the energy transfer path between the two-level trees until all nodes are explored;

[0083] Step 6: Compare the number of nodes in different paths in the traversed node coordinate matrix Explored_node_matrix, and define the path with the least number of nodes as the optimal path.

[0084] The method of the present invention is more suitable for array-type wireless power transmission networks. The array-type wireless power transmission network is a network in which the nodes in the wireless power transmission network are arranged in an array structure in advance, and the transmitting coils and receiving coils of every two adjacent nodes in the array are opposite. A 3×3 array-type wireless power transmission network is as follows: Figure 7 As shown in (a), it can be seen that under this node arrangement and coil direction selection mode, the receiving coils between adjacent nodes are not only naturally perpendicular to each other, but also far apart, so the mutual inductance between them is very small, even close to 0, satisfying the constraint conditions of node decoupling.

[0085] The present invention proposes a method for constructing and optimizing an energy transmission path in a WPTN. In order to improve the timeliness of the energy transmission path construction, the present invention applies the DBSCAN clustering algorithm to divide the entire network into multiple sub-networks. On this basis, the Prim algorithm and the breadth-first search algorithm are used to construct the optimal energy transmission paths between clusters and within clusters. In addition, under complex cross-coupling conditions, each node can maintain the characteristics of constant current output.

Claims

1. A method for constructing and optimizing an energy transmission path in a wireless power transmission network, characterized in that: Including building sub-networks based on DBSCAN clustering algorithm, building inter-cluster energy transfer paths based on Prim algorithm, and optimizing intra-cluster energy transfer paths based on breadth-first search algorithm; The implementation steps for constructing a subnetwork based on the DBSCAN clustering algorithm are as follows: Step 1: Define a set D containing the coordinates of all load nodes in the wireless power transmission network = {P1, P2, P3, ..., P i ,…,P j ,…,P n }, neighborhood radius Eps and neighborhood density threshold MinPts; Step 2: Select any unprocessed load node P from set D. i Starting from, with Eps as the radius, traverse and search for all load nodes in the network. In this process, it is necessary to calculate the distance between two load nodes to determine whether the two are directly density-reachable and form a cluster; Step 3: If the number of load nodes covered within the Eps range is not less than the pre-set neighborhood density threshold MinPts, then these load nodes are merged into the neighborhood density threshold Pts. i In a cluster C where is the core point; Step 4: If a load node P j , if the number of load nodes covered by its neighborhood range does not reach the neighborhood density threshold MinPts, the load node will be temporarily treated as a noise point; Step 5: Visit other unprocessed load nodes in turn to form several core points, noise points and clusters, and merge and expand the load nodes and clusters that meet direct density reachability or density reachability; Step 6: Repeat steps 2 to 5 until no new load nodes are merged into any cluster or marked as noise points, and then terminate the algorithm; Step 7: Output clustering results; The implementation steps of constructing the inter-cluster energy transfer path based on Prim's algorithm are as follows: Step 1: Define the shortest distance between every two clusters constructed based on the DBSCAN clustering algorithm as the weight, and mark the coordinates of the nodes that constitute the distance; Step 2: Select any cluster and add it to the minimum spanning tree; Step 3: Determine whether the minimum spanning tree contains all clusters. If not, find the edge with the smallest weight from other adjacent clusters and add this edge and the cluster connected to this edge to the minimum spanning tree. Step 4: Repeat step 3 until all clusters are included in the minimum spanning tree; The implementation steps of cluster energy transfer path optimization based on breadth-first search algorithm are as follows: Step 1: Define the node position coordinate matrix, output tree, traversed node coordinate matrix and untraversed node coordinate matrix in each cluster constructed based on DBSCAN clustering algorithm; Step 2: Grow one level of output tree starting from the initial node; Step 3: Add a node in the tree to the explored node coordinate matrix Explored_node_matrix, and add other nodes in the same level tree to the unexplored node coordinate matrix Unexplored_node_matrix, and then construct the energy transmission path between the nodes of the two levels of trees according to the transmission law of energy flow; Step 4: Move the nodes from the unexplored_node_matrix to the explored_node_matrix until all nodes in the same level of the tree have been moved, return to step 2, and grow the output tree by one level; Step 5: Continue to construct the energy transfer path between the two-level trees until all nodes are explored; Step 6: Compare the number of nodes in different paths in the traversed node coordinate matrix Explored_node_matrix, and define the path with the least number of nodes as the optimal path.

2. The method for constructing and optimizing an energy transmission path in a wireless power transmission network according to claim 1, characterized in that: The distance between two load nodes is calculated by using the modified Minkowski distance. The two nodes in the wireless power transmission network are P i (a pi1 , a pi2 ) and P j (a pj1 , a pj2 ), then the Minkowski distance DM between two nodes is expressed as: The modified Minkowski distance between two nodes can be expressed as: D MTD =1+2×||a pi , a pj || ∞ +P(||a pi , a pj ||1)Q(||a pi , a pj ||1)less(a pi -a pj ), where: In the above formula, P is the relative position correction factor, Q is the starting position correction factor, less is the logic function, when k = 1, it means calculating the Manhattan distance, and when k = infinity, it means calculating the Chebyshev distance.

3. The method for constructing and optimizing an energy transmission path in a wireless power transmission network according to claim 1 or 2, characterized in that: The shortest distance D between clusters in Prim's algorithm CMTD for A and B are any two clusters constructed by the DBSCAN clustering algorithm, N (A) and N (B) The shortest distance D between any two nodes in cluster A and cluster B is CMTD It is the shortest distance among all the nodes between two clusters.

4. The method for constructing and optimizing an energy transmission path in a wireless power transmission network according to claim 1 or 2, characterized in that: The method is particularly suitable for an array-type wireless power transmission network, in which nodes in the wireless power transmission network are arranged in an array structure in advance, and the transmitting coils and receiving coils of every two adjacent nodes in the array are opposite.

Images