CN-122027455-A - Decision scheduling method, storage medium and equipment based on Internet of things

CN122027455ACN 122027455 ACN122027455 ACN 122027455ACN-122027455-A

Abstract

The invention discloses a decision scheduling method, a storage medium and equipment based on the Internet of things, which comprise the steps of collecting attribute information of an equipment layer, a network layer and a service layer in the Internet of things, determining the relation among the equipment, the network and the service layer, constructing an Internet of things knowledge graph, carrying out health assessment on the real-time state of the Internet of things according to the collected attribute information, if the Internet of things is healthy, operating according to an original decision scheduling strategy, otherwise, carrying out fault equipment tracing by combining a retrieval enhancement technology with the constructed Internet of things knowledge graph, determining the standby equipment with optimal performance based on a genetic algorithm, switching the fault equipment into the optimal standby fault equipment, and updating the decision scheduling strategy. The invention can rapidly and accurately trace the source of the fault equipment in the Internet of things, and improves the efficiency and the precision of decision scheduling.

Inventors

WU DAOHONG
ZHANG XU
WANG PUPU

Assignees

中国电信股份有限公司
广东亿迅科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260213

Claims (10)

1. The decision scheduling method based on the Internet of things is characterized by comprising the following steps of: step S1, acquiring attribute information of a device layer, a network layer and a service layer in the Internet of things, determining the relationship among the device, the network and the service, and constructing an Internet of things knowledge graph; Step S2, carrying out health assessment on the real-time state of the Internet of things according to the attribute information acquired in the step S1, if the Internet of things is healthy, operating the Internet of things according to an original decision scheduling strategy, and if the Internet of things is healthy, executing the step S3; S3, performing fault equipment tracing by utilizing a retrieval enhancement technology and combining the constructed Internet of things knowledge graph; And S4, determining the spare equipment with the optimal performance by the spare equipment of the fault equipment based on a genetic algorithm, switching the fault equipment into the optimal spare fault equipment, and updating a decision scheduling strategy.
2. The decision scheduling method based on the Internet of things, which is characterized in that the attribute information of the equipment layer comprises real-time state attributes and static attributes of equipment, the attribute information of the network layer comprises network topology, bandwidth utilization rate and delay, and the attribute information of the service layer comprises service rules associated with the equipment.
3. The decision scheduling method based on the Internet of things according to claim 2 is characterized in that the construction process of the Internet of things knowledge graph is that the Internet of things knowledge graph is built according to static properties of equipment, network topology and business rules related to the equipment.
4. The decision scheduling method based on the internet of things according to claim 1, wherein the specific process of health assessment on the real-time state of the internet of things is as follows: Wherein, the Represents the health degree score of the internet of things, Representing the dimension of the real-time status attribute of the device, Representation of Is used for the indexing of (a), The number of devices in the internet of things is represented, Representation of Is used for the indexing of (a), Represent the first Device No. Health score for each real-time status attribute, Representing device layer weights; representing a dimension of a network attribute, Representation of Is used for the indexing of (a), Represent the first Health score for an individual network attribute, Representing network layer weights; representing the dimension of the business attribute, Representation of Is used for the indexing of (a), Represent the first Health score for individual business attributes, Representing traffic layer weights.
5. The decision scheduling method based on the internet of things according to claim 1, wherein the step S3 comprises the following sub-steps: step S3.1, collecting health scores of all equipment, networks and services of the Internet of things in a non-healthy state at historical moments, constructing a fault vector set by combining corresponding fault nodes, and storing the fault vector set in a fault vector database; s3.2, constructing query vectors by health scores of all equipment, networks and businesses in a non-health state estimated in real time, searching fault vectors with highest similarity through a fault vector database, and determining fault nodes; And S3.3, searching a corresponding fault node in the knowledge graph of the Internet of things, taking the space-time weight as the weight of the edge in the Internet of things, calculating the fault propagation probability, and realizing the tracing of the fault equipment.
6. The decision scheduling method based on the internet of things according to claim 5, wherein the calculating process of the fault propagation probability is as follows: Extracting neighbor node equipment of a fault node from an internet of things knowledge graph; calculating fault propagation probability according to the weight of the edge between the neighbor node equipment and the fault node: Wherein, the Representing a failed node Is a set of neighboring node devices of the (c), Representation of Is used for the indexing of (a), Represent the first Probability of failure propagation for the individual neighboring node devices, Representing a failed node Is used for the failure probability of the (c) in the (c), Represents the damping coefficient of the damping device, Represent the first Individual neighbor nodes and equipment failure nodes The weight of the edge between the two, , Represent the first Individual neighbor nodes and equipment failure nodes The distance between the two plates is set to be equal, The distance weight is represented as a function of the distance, Represent the first Individual neighbor nodes and equipment failure nodes The time of flight of the fault between them, Representing a time decay coefficient; And step iii, continuing to repeat the steps i-ii by using the neighbor node equipment with the maximum fault propagation probability until no neighbor node equipment exists, obtaining a fault propagation path and realizing the tracing of the fault equipment.
7. The decision scheduling method based on the internet of things according to claim 1, wherein the step S4 comprises the following sub-steps: Step S4.1, randomly combining spare equipment of fault equipment, setting maximum iteration times and a fitness function based on total shutdown cost as an individual of a genetic algorithm; s4.2, calculating the fitness of each individual by using a fitness function based on the total shutdown cost, and reserving the individual with the minimum fitness; In the next iteration, the rest individuals are updated based on selection, crossing and variation, the fitness of each individual is calculated by using a fitness function based on total shutdown cost, and compared with the fitness of the reserved individuals in the previous iteration, the individuals with the minimum fitness are reserved; And step S4.4, repeating the step S4.3 until the maximum iteration times are reached, taking the individual with the minimum reserved adaptability as the optimal standby equipment, and updating the decision scheduling strategy.
8. The decision scheduling method based on the internet of things according to claim 7, wherein the calculation process of the fitness function based on the total shutdown cost is as follows: Wherein, the The fitness function is represented as a function of the fitness, A normalized value representing the start-up delay of the standby device combination, Representation of Is used for the weight of the (c), A normalized value representing production loss during a standby equipment combination switch, Representation of Is a weight of (2).
9. A computer readable storage medium storing a computer program, wherein the computer program causes a computer to execute the decision scheduling method based on the internet of things according to any one of claims 1-8.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the internet of things-based decision scheduling method according to any one of claims 1-8 when the computer program is executed.

Description

Decision scheduling method, storage medium and equipment based on Internet of things Technical Field The invention relates to the technical field of the Internet of things, in particular to a decision scheduling method, a storage medium and equipment based on the Internet of things. Background In an application scene of the Internet of things, the types of equipment are various, the transmission protocol, the data format and the communication mode are large in difference, cooperation is needed to be realized through unified scheduling, hundreds of pieces of data can be generated by a single equipment per second, key information can be screened through unified decision scheduling, information overload is avoided, in addition, the equipment is possibly offline due to faults, movement or network fluctuation, and a decision scheduling strategy needs to be adjusted in real time to maintain the running stability of the Internet of things. Therefore, decision scheduling is a core link of normal operation of the Internet of things, and can optimize resource allocation and improve equipment operation efficiency in the Internet of things through an intelligent means under the dynamic change of a complex scene. However, the existing decision scheduling method depends on threshold value alarm of a single sensor, so that multi-source heterogeneous data in an Internet of things scene is difficult to integrate, thereby causing fault discovery hysteresis, and meanwhile, in the fault discovery process, due to the fact that manual analysis logs or topological structures are relied on, context awareness and cross-domain knowledge integration are lacked, long troubleshooting time is required to be consumed to find fault equipment and even errors. And then, based on the fault equipment, the decision scheduling strategy of the fixed rule is adjusted, so that the complexity of the evolution of the equipment state along with time cannot be dealt with, and the resource waste or the decision scheduling failure is caused. Disclosure of Invention Aiming at the problems in the prior art, the invention provides a decision scheduling method, a storage medium and equipment based on the Internet of things, which can rapidly and accurately trace the source of fault equipment under the Internet of things and improve the efficiency and the accuracy of decision scheduling. In order to achieve the technical purpose, the invention adopts the following technical scheme: A decision scheduling method based on the Internet of things comprises the following steps: step S1, acquiring attribute information of a device layer, a network layer and a service layer in the Internet of things, determining the relationship among the device, the network and the service, and constructing an Internet of things knowledge graph; Step S2, carrying out health assessment on the real-time state of the Internet of things according to the attribute information acquired in the step S1, if the Internet of things is healthy, operating the Internet of things according to an original decision scheduling strategy, and if the Internet of things is healthy, executing the step S3; S3, performing fault equipment tracing by utilizing a retrieval enhancement technology and combining the constructed Internet of things knowledge graph; And S4, determining the spare equipment with the optimal performance by the spare equipment of the fault equipment based on a genetic algorithm, switching the fault equipment into the optimal spare fault equipment, and updating a decision scheduling strategy. Further, the attribute information of the equipment layer comprises real-time state attributes and static attributes of equipment, the attribute information of the network layer comprises network topology, bandwidth utilization rate and delay, and the attribute information of the service layer comprises service rules associated with the equipment. Further, the construction process of the Internet of things knowledge graph comprises the step of building the Internet of things knowledge graph according to the static attribute of the equipment, the network topology and the business rule related to the equipment. Further, the specific process of health assessment on the real-time state of the internet of things is as follows: Wherein, the Represents the health degree score of the internet of things,Representing the dimension of the real-time status attribute of the device,Representation ofIs used for the indexing of (a),The number of devices in the internet of things is represented,Representation ofIs used for the indexing of (a),Represent the firstDevice No.Health score for each real-time status attribute,Representing device layer weights; representing a dimension of a network attribute, Representation ofIs used for the indexing of (a),Represent the firstHealth score for an individual network attribute,Representing network layer weights; representing the dimension of the business attribute, Represent