63 results about "Average path length" patented technology

The invention belongs to a content data network technology, and discloses a data transmission method of a content-centric datacenter network. The method is characterized in that an NDN (Named Data Network) based content-centric routing and forwarding strategy is used; multicast characteristics and expandability of the CCDN (content-centric datacenter network) are guaranteed only by storing a complete PIT (Pending Interest Table) and part of an FIB (forwarding information base), by adopting a Hybrid Content and Location routing strategy and by taking limited storage resources and datacenter network topology of a datacenter switch into consideration; and different from an on-path caching strategy of the NDN, an off-path mechanism that a host provides data caching is adopted by taking large data volume characteristic of the datacenter network and host storage capacity of redundancy into consideration. According to multistage topological characteristics, of the datacenter network, such as Fat-Tree, the CCDN guarantees Distance-aware Content based Forwarding through FIB learning, average path length of data transmission is decreased, and network throughput is improved. By the aid of the adaptive forwarding strategy, the CCDN can still provide accurate and efficient data forwarding in network failure.

The invention provides an electroencephalogram signal characteristic extracting method. Network average route lengths and clustering coefficients are calculated through wavelet reconstruction, windowing horizontal visibility map complex network conversion and complex network analysis. The average route lengths and clustering coefficients composed of electroencephalogram signals are calculated to achieve characteristic analysis of electroencephalogram signals and chaotic time sequence signals of the electroencephalogram signals of different rhythms. The electroencephalogram signal characteristic extracting method has the advantages that one-dimensional chaotic time sequences are converted into complex networks, according to analysis of network characteristic parameters, fractal characters of the electroencephalogram signals are revealed, the complex non-linearity signals of the electroencephalogram signals are depicted from a brand new angle. The electroencephalogram signal characteristics can be applied to automatic diagnosis of mental disease and a characteristic identifying module of a brain-machine port system. The electroencephalogram signal characteristic extracting method can effectively distinguish the electroencephalogram signals of an epilepsia attach stage and an epilepsia non-attach stage, the equation p<0.1 is met after Mann-Whitney detection, and the electroencephalogram signal characteristic extracting method can be applied to epilepsia electroencephalogram automatic identification.

A method of placing nodes in an area that requires coverage, the method includes the step of creating a network topology such that the average path length is kept to a minimum number of hops at the time of placement of a new node, wherein a limit is placed on a number of neighbors at the time of placement of the new node, the number being a parameter that impacts the average path length, resiliency and capital investment. The new node is connected to at least one node in the network.

The invention provides a complex weighted traffic network key node identification method based on semi-local centrality, comprising the following steps: S1, constructing a complex weighted traffic network: constructing a traffic network by adopting an original method, and taking road sections as nodes and road sections as edges; generating a corresponding adjacent matrix; according to the road grade, obtaining a weighted adjacency matrix through the adjacency matrix; S2, processing the weighted adjacency matrix, and analyzing network characteristics: calculating degree distribution of nodes, calculating an average clustering coefficient, and calculating an average path length; analyzing network characteristics according to the degree distribution of the nodes, the average clustering coefficient and the average path length; S3, identifying key nodes by adopting a semi-local centrality algorithm; and S4, sorting the key nodes of the traffic network: sorting the nodes in a descending order according to the importance degree to obtain the key nodes in the traffic network. The road grade is used as the weight, and the semi-local centrality algorithm is adopted, so that the problems thatthe key node identification calculation complexity of the existing traffic network is high and the traffic network characteristics are not considered are solved.

The invention discloses an abnormal data detection method and system and a storage medium, and belongs to the technical field of the Internet. The method comprises the steps that a detection node obtains a pre-established isolated forest model, and the detection node obtains a to-be-detected data subset; the detection node calculates the average path length of each piece of to-be-detected data inthe isolated forest model; the detection node calculates the detection score of each piece of to-be-detected data according to the average path length of each piece of to-be-detected data in the isolated forest model; and the detection node determines a detection result of each piece of to-be-detected data according to the detection score of each piece of to-be-detected data. According to the invention, the to-be-detected data in the to-be-detected data set is distributed to different detection nodes for storage; compared with the prior art, the data storage amount of each detection node is reduced, and the isolated forest model is distributed to different detection nodes, so that abnormal data can be detected on different detection nodes in parallel, the detection speed is increased, andthe detection duration is shortened.

The invention discloses an isolation mechanism-based online photovoltaic hotspot fault detection method, and belongs to the field of photovoltaic power generation system fault detection. The method comprises the following steps of: acquiring a series of real-time operation data of photovoltaic panels in a photovoltaic array; establishing an isolation forest-based photovoltaic array hotspot fault estimation model by utilizing the characteristic that hotspot fault data is less and different; and finally traversing an average path length of each record in the forest so as to obtain a hotspot fault score. According to the method, online measurement can be carried out on the hotspot faults of any photovoltaic panel in the photovoltaic array.

Methods are provided for updating a digital road map in which road segments having segment lengths are stored. At least two vehicles are equipped with at least one respective onboard unit generating position readings. If a first set of position readings of a first onboard unit can be associated with a road segment, a first path length is determined based on the position readings of the first set. If a second set of position readings of a second onboard unit can be associated with the road segment, a second path length is determined based on the position readings of the second set. An average path length is calculated based on at least the first and second path lengths. If the average path length deviates from the stored segment length, the average path length is stored as an updated stored segment length of this road segment in the digital road map.

The present invention discloses a group degree based sorting method and model evolution method for important nodes on complex network. The method comprises: first, obtaining each order of group degree of each node on a complex network; then calculating each order of overall group degree of the complex network; normalizing each order of overall group degree; calculating a weight of each order of group degree according to a normalization result; finally, each node performing weighting on each order of group degree of the node according to the weight, wherein a result is an importance value of the node; and sorting each node according to the importance value. During model evolution, each time a new node is added, a connecting node of the new node is selected according to the importance values of existing nodes. According to the sorting method and model evolution method provided by the present invention, the importance value of the node is calculated based on the group degree, the obtained node importance sequence is better in line with the actual situation of the network, and the average path length and cluster coefficient property obtained through model evolution both have obvious advantages.

A method for characterizing the pore network structure of a coal rock comprises the following steps: obtaining three-dimensional data of a coal rock core pore structure by a CT technology, establishing a three-dimensional digital rock core, and establishing a three-dimensional pore network by using a central axis algorithm; simplifying the model data of the pore network, simplifying pores into nodes, simplifying pore passages into edges, marking the pores, and deriving pore communication information; and using a complex network to characterize the basic information of the network, comprising the total number of the nodes, the total number of the edges, node distribution, the average degree of the nodes, the average path length of the network, the network clustering coefficient, the networkmedia, the network density and the network robustness. Compared with traditional coal rock pore structure analysis methods, the method in the invention increases network property analysis, can analyze the seepage law of different pore network communication structures having the same porosity to achieve the purpose of improving the existing gas recovery rate at the microscopic level, and adopts acomplex network theory to quantitatively characterize the pore structure network parameters of the coal rock in order to accurately and comprehensively characterize the pore network communication property of the coal rock.

The invention provides a hyperspectral anomaly detection method based on discrimination forest subspace selection, and the method comprises the steps: randomly selecting a part of pixels from a hyperspectral image, constructing a subspace selection isolated binary tree, and constructing an isolated discrimination forest according to the subspace selection isolated binary tree; traversing the hyperspectral image through the constructed isolated discrimination forest, and calculating the average path length; and calculating an abnormal score value of each pixel to realize detection of an abnormal target. According to the invention, an isolated discrimination forest model is used to repeatedly learn and estimate the distribution rule of background classes and abnormal classes in the image inthe form of subsets. On the basis, an axis parallel subspace selection method is introduced, a wave band which is more beneficial to abnormal information discrimination is selected, the problem that abnormal information is buried due to wave band redundancy and too high dimension is avoided, abnormal score value information of the image is solved, and a final result of hyperspectral image abnormaltarget detection is obtained.

The invention discloses a multivariable causal-driven complex electromechanical system service security situation assessment method, and belongs to the field of complex electromechanical system service process state evaluation. The method comprises the following steps: firstly, extracting state monitoring data in a complex electromechanical system service process, preprocessing the state monitoring data, calculating a causal measure value between monitoring variables by using a GPDC method, and constructing a causal network topology model capable of reflecting a system state evolution process;extracting features of three dimensions of an average path length, a clustering coefficient and a network structure entropy based on the established causal network model; and reconstructing an abnormal state space of the system according to the extracted abnormal fluctuation information of the three features, normalizing abnormal over-limit indexes on the three dimensions, and effectively evaluating the service state of the complex electromechanical system by using the normalized abnormal indexes.

The invention discloses a composite bus subway network model construction method. Through comprehensively comparing an average clustering coefficient and an average path length of a composite network under non-weighted conditions and a weighted average path length of the network under weighted conditions, an optimal model parameter d can be acquired. The method comprises steps that firstly, a bus subway dual-layer network is constructed according to a dual-layer network model, connection between nodes in the network and connection between network layers of the dual-layer network are both realized. When a subway network is composed to a bus network, node mapping including node fusion and establishment and connection edge re-connection are both processed, and connection edge re-connection is influenced by the model parameter d. Parameters of the composite network under different d value conditions are compared, the optimal value of the d can be determined; buses and subways can be reasonably connected together through the optimal model parameter d, and passenger travel efficiency is improved.

The present invention relates to a method for constructing a bus-metro coupling network based on a space network. The method is characterized in that: a bus network and a metro network are connected by introducing a pedestrian way, the difference between the essence of the two kinds of transport carriers and difference of passengers flow rate caused by the habit of passengers who choose the transport carrier are taken into account; for the time-consuming issue most stressed by people to travel, the time spent on the car, walking time and transfer waiting time are included in the total time spent on the road to carry out the topology of a bus-metro coupling network based on the space network which is more in line with the actual situation; and in addition, a multi-layer network analysis method is also used to calculate the average path length and the average weighted path length of the network. The results show that the constructed network structure performance is better than the previous topology, so that compared with the previous modeling of single-layer and composite networks, the modeling manner of the method disclosed by the present invention can better reflect the real traffic system.

The invention discloses a power system transient stability simulation method based on network node numbering optimization. A sparse vector technology is widely applied to power system calculation. However, existing network node numbering methods applied to sparse vectors aim to achieve the purpose that the average forward and backward substitution calculation amount of all nodes is the smallest, and the characteristic that forward and backward substitution only needs to be conducted on active nodes during transient stability simulation is not taken into consideration. According to the power system transient stability simulation method, the characteristic that the sparse structure of an independent vector of a network algebraic equation and the sparser structure of a solution vector are identical and decided during transient stability simulation is fully considered, the network nodes are divided into active nodes and passive nodes, the out-degrees of the nodes, the number of active precursor nodes and the number of the precursor nodes are fully considered, the network nodes are numbered, under the condition that the number of newly increased elements of a factor table is small, the average path length of the active nodes is made the smallest, and no requirement for the path tree lengths of the passive nodes exists. By the adoption of the power system transient stability simulation method, the calculation amount of solving a differential algebra equation set during transient stability simulation of a power system can be remarkably reduced.

The invention discloses a container shipping network port importance evaluation method based on integrated centrality, which comprises the following steps: S1, obtaining port route information in thecontainer transportation process, and constructing a container shipping network database according to the port route information; S2, constructing a container shipping network model; S3, calculating the average path length, the aggregation coefficient, the degree and the distribution function thereof, and further analyzing the network type of the contain shipping network model; S4, calculating thescores of degree centrality, approaching centrality and intermediary centrality of each port in the container shipping network model, and calculating the comprehensive centrality scores of each portby the comprehensive centrality method, and then ranking and evaluating the importance of each port in the container shipping network model. The invention provides an evaluation mode of port importance, carries out importance analysis and research on ports in a container shipping network, and provides a reference basis for investment construction of a marine port and optimization of a shipping network.

The invention discloses a method for identifying vulnerable nodes of a grid structure, which includes: 1) reading in the node connection relationship data of the grid topology structure, and configuring the number k of pre-attacked or destroyed nodes; 2) calculating two nodes based on the network topology structure The shortest path between and the average path length L of the network; 3) Calculate the network efficiency E; 4) Delete the relevant path connected to the node and recalculate the network efficiency Ei of the point; 5) Calculate the network efficiency Ei of each node based on the network efficiency Ei 6) Based on the node vulnerability index, use the original input node connection relationship data to obtain the network topology diagram; 7) Remove the node with the largest node vulnerability value Vi, and loop through steps 2)-6) to obtain the second vulnerability node; 8) loop k times, and finally complete the identification of N-k vulnerability points. The vulnerability node identification method described above can dynamically identify the node vulnerability, taking into account the change of the node vulnerability under the changed network topology after the node is deleted.

The invention discloses a test method for physical design similarity of a circuit. The method comprises the following steps of: carrying out equivalence on components or modules in a circuit principle graph as nodes, carrying out equivalence on connection between components or modules, and building a complex network of circuit physical design; calculating the following four characteristic parameters of the complex network of circuit physical design: a node degree, node betweenness, a node cluster coefficient and a node average path length; and carrying out a Kolmogorov-Sirnov test of characteristic parameter accumulation distribution on a compared circuit physical design complex network, and obtaining overall similarity of each network characteristic parameter. By the adoption of the test method, according to complex network characteristics indicated by different circuit physical design, quantitative analysis on circuit physical design similarity is given out, and misjudgment brought by the affect of an subjective factor due to observation of a circuit design profile and the like in patent tort dispute of the circuit physical design is reduced.

The invention discloses a logistics unmanned aerial vehicle abnormal behavior intelligent identification method based on an isolated forest method. The logistics unmanned aerial vehicle abnormal behavior intelligent identification method comprises the following specific steps: 1, carrying out outlier calculation and observation on logistics unmanned aerial vehicle flight data; 2, constructing isolated trees according to the input data, combining the single isolated trees with the data features, and constructing a set of isolated forests; 3, calculating the average path length of each isolated tree and the expectation E (h (x)) of the path length, and finally solving the abnormal score of the sample through the E (h(x)); 4, dividing the abnormal data according to the calculation result of the abnormal score; 5, substituting the longitude, latitude, elevation angle, climbing speed and abnormal score data into the model, and evaluating the accuracy. The method has the advantages that intelligent learning and efficient detection of unmanned aerial vehicle abnormal behaviors are achieved, the pressure of high-speed development unmanned aerial vehicle operation on flight safety and public safety can be effectively relieved, and thereby a technical foundation is laid for development of the unmanned aerial vehicle logistics distribution industry.

The invention discloses a brain volume atrophy prediction method based on a brain function network. The brain volume atrophy prediction method comprises the following steps: 1, processing function image data, analyzing the function data and calculating a brain volume; 2, establishing a brain function network; 3, calculating the average path length and the aggregation coefficient of the brain network; and 4, establishing a prediction model of brain volume atrophy. Through the steps, the brain volume atrophy prediction method establishes the mapping relation between the topological characteristics of the brain function network and the brain atrophy information, and provides the brain volume atrophy prediction method based on the brain function network. The brain volume atrophy prediction method has early warning, systematicness and robustness, and the research result of the brain volume atrophy prediction method can provide powerful method support for future prediction of brain volume atrophy, and can achieve early prevention and treatment of brain atrophy.

The invention relates to a harmonic data anomaly detection method, terminal equipment and a storage medium. The method comprises the following steps: S1, collecting harmonic current data generated bythe equipment in a period of time sequence as a data set; S2, constructing a forest composed of a plurality of isolated trees according to the collected data set; S3, calculating the path length of each harmonic data in the data set; S4, calculating the average path length of all data in the data set; S5, calculating an abnormal score of each harmonic data; and S6, determining whether each harmonic data is abnormal or not according to the abnormal score of each harmonic data. The invention achieves the abnormality detection of the harmonic data through the isolated forest algorithm, reduces the expenditure of manpower and material resources, improves the economic benefits and service quality, improves the management level of a power grid, and also improves the safety operation capability of the power grid.

The invention provides a complex supply chain network node search model which is formed by the following steps that step 1, different unit entities of a supply network act as the nodes of the network,the links among the units act as the edges of the network, and the direction of the edges represents that the last node of the supply network points to the next node; and the weight of the edges represents the number of the child nodes included in the father node; step 2, when one node is additionally arranged and connection is established, the new node is connected to the top ranking number of original nodes so as to form a complex supply chain network in which the majority cascade nodes are governed by the minority gathering nodes; and step 3, the degree of the node is the number of the edges connected with the node in the complex supply chain network; and the average path length of the node is the number of the edges on the shortest path of the two nodes in the complex supply chain network, and the aggregation coefficient in the complex supply chain network is the parameter for measuring the node clustering condition in the complex supply chain network.

The invention relates to the field of artificial intelligence, and provides an abnormal data detection method and device, computer equipment and a storage medium. The method comprises the steps: acquiring the user driving behavior characteristics; screening out specified user driving behavior characteristics from the specified user driving behavior characteristic data, constructing a specified isolated tree based on a preset segmentation value, and generating a corresponding isolated forest; calculating the path length of the user driving behavior characteristic data from the root node of the isolated tree to each leaf node; calculating the average path length of all the user driving behavior characteristic data in the isolated forest; calculating an anomaly detection score of the driving behavior characteristic data of each user; and generating an anomaly detection result corresponding to each piece of user driving behavior feature data based on the anomaly detection score. The abnormal data can be quickly and accurately identified from all the user driving behavior characteristic data. The method can also be applied to the field of block chains, and the data such as the anomaly detection score can be stored on the block chain.