Method for predicting workload of controller approaching controlled sector

Abstract

The invention discloses a method for predicting the workload of a controller approaching a controlled sector. The method comprises: a sector complexity index for evaluating the workload of a controller approaching a controlled sector is established; sector complexity index data and a controller workload self-evaluation value are collected; and then a controller workload model based on a BP neuralnetwork algorithm of PA and PCA is established and a prediction result is obtained. With the method disclosed by the invention, the workload level of the controller can be evaluated timely; the powerful support is provided for the safe, high-efficiency and ordered operation of the aircraft in the sector; and the prediction accuracy is improved.

Description

technical field [0001] The invention relates to a method for predicting the workload of a controller in an approach control sector. Background technique [0002] With the vigorous development of China's civil aviation transportation industry, the flight volume continues to rise, resulting in saturation or even supersaturation of the control sector. The work of the controller is one of the important links to ensure the safe, efficient and orderly operation of aviation. As the sector traffic flow is getting closer and closer to the controller's load limit, and the duration of the approach to the load limit is getting longer and longer, the control Operators may not be able to accurately respond to the traffic situation of each aircraft and judge potential conflicts, and incidents of "mistakes, forgetting, and omissions" in control will occur, which will be extremely unfavorable to the safe flight of aircraft in the sector, and even cause great harm to aviation safety. Hidden ...

AI Technical Summary

Problems solved by technology

However, these two algorithms have shortcomings. The regression model based on the principal component analysis method will not fully reflect the workload of controllers after individual indicators are eliminated. Although the indicators are simplified, they also weaken the impact of indicators on the work of controllers. The comprehensiveness of the reflection; the support vector machine algorithm solves the support vector by means of quadratic programming, which is difficult to implement for large-scale training samples, and the amount of sample data is relatively small, which is contrary to the original intention of the experiment to reduce the controller's workload prediction error through a large number of data samples , so the small sample data prediction value obtained by the support vector machine algorithm will have a large error with the actual value

Images