Genetic algorithm optimization based polymorphic detector generating method

Abstract

A multi-morphology detector generation method based on genetic algorithm optimization, the specific steps are as follows: 1: Divide the normalized search space layer by layer, and set the division fineness to 3 iterations, and enter the next step when the division accuracy is reached Step 1; Step 2: Match the divided super-rectangular area with the training samples. If the super-rectangular area does not match the training samples, use it as a mature super-rectangular detector; 3: Apply the genetic evolution mechanism to optimize the traditional negative selection The algorithm generates the optimal spherical detector one by one; four: use the super-rectangular detector and the optimal spherical detector as a mature detector set to complete the detector generation; through the above steps, the introduction of the genetic algorithm improves the detection accuracy and efficiency, The rectangle detector is generated by layer-by-layer space division, and the data-based anomaly detector generation method is obtained. It reduces the search space of the evolutionary algorithm and improves the average coverage of mature detector sets.

Description

technical field [0001] The invention relates to a method for generating a multi-morphology detector based on genetic algorithm optimization, and belongs to the technical fields of artificial intelligence immune classification technology and data-based abnormality detection technology. Background technique [0002] Due to the increase in the number of satellites and the increasing complexity of the satellite system, its on-orbit failure rate has increased significantly, and it has been difficult to ensure the healthy operation of satellites only by relying on traditional signal processing and analysis methods based on expert experience. During the operation of the satellite, the measurement and control system periodically transmits on-orbit telemetry data to the ground measurement and control center. These data are the only reference basis for the monitoring center to judge the status of satellites in orbit. Whether it is abnormal or not is closely related to whether the sate...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the traditional artificial immune detector generation algorithm does not control the coverage of detectors, resulting in a large number of redundant detectors in the detection accuracy and low efficiency problems

Images