A Fault Tolerant CPS Simulation Test Method Based on Hierarchical Adaptive Strategy

Abstract

The invention discloses a fault-tolerant CPS simulation test method based on a hierarchical self-adaptive strategy, comprising the following steps: establishing a fault-tolerant CPS simulation resource model and allocating test resources, including establishing a fault-tolerant CPS simulation resource layered model and testing resource allocation Adaptive allocation; fault-tolerant CPS simulation test, including inter-layer message passing and co-simulation timing advancement; node liveness detection based on sliding window. The invention is used for the simulation test of the fault-tolerant CPS system, which can not only realize the effective allocation of test resources, improve the test efficiency, save time and cost, but also perform dynamic self-adaptation for CPS systems of different scales, build a scalable test platform, and According to the characteristics of multi-mode redundancy of the fault-tolerant CPS system, effective test execution can be carried out, and the consistency of the test master and slave nodes can be judged efficiently to ensure that the redundancy strategy is effective.

Description

technical field [0001] The invention relates to a fault-tolerant CPS simulation test method in the computer field, in particular to a fault-tolerant CPS simulation test method based on a hierarchical self-adaptive strategy. Background technique [0002] CPS is the abbreviation of Cyber ​​Physical System. The Chinese meaning is cyber-physical fusion system, which is a networked embedded system that integrates computing process and physical environment; CPS is usually deployed in some safety-critical key infrastructure and safety and reliability Highly demanding fields, such as power systems, traffic management systems, military, life monitoring, and aerospace systems, have extremely high requirements for reliability and fault tolerance. [0003] Fault-tolerant strategies are widely used in Safety Critical systems to improve the reliability of the system. The basis of fault tolerance is redundancy, which usually includes hardware redundancy and software redundancy. Hardware r...

AI Technical Summary

Problems solved by technology

There are many types of protocols used when the CPS node under test communicates with external devices, such as: RS232 / 422 / 485, I / O, CAN, 1553B, Ethernet, AD, DA, etc. Interface characteristics and possible failure modes to allocate test resources. If test resources are not properly allocated, the test method will be incorrect or the efficiency of the test environment will be low, resulting in waste of time and cost.

[0007] (2) Adaptive problem of different scale CPS systems

Due to the inherent heterogeneity of CPS, the nodes vary with the function and scale of the system, ranging from a few nodes to dozens or hundreds of nodes. It is a huge challenge to build a matching test platform according to the scale of the system : For small-scale systems, using a large-scale test platform will undoubtedly bring huge platform construction time costs and waste of idle resources. For larger-scale systems, use a small-scale test platform to test one by one on the one hand. Efficiency On the other hand, because of the small scale, it is difficult to completely cover, the system-level global functions cannot be tested, and the performance of the test platform is also difficult

[0008] (3) The problem of easy implementation of fault-tolerant CPS system simulation

The traditional simulation testing methods used require testers to have a full understanding of the multi-mode redundancy backup, interface correspondence, and node type determination of each CPS node. It is difficult for testers to stand at the top and design test cases for system-level business logic, resulting in The designed test cases lack globality and overallness, which not only increases the workload of testers, but also makes it difficult to ensure the adequacy of the test; at the same time, it is difficult to test the master and slave nodes at the same time, which seriously reduces the test efficiency

These all add challenges to the implementation of the test

Images