Abstract:

A maximum correlation entropy Kalman filter （MCKF） error correction method is proposed to address the issue of test data errors caused by errors in the testing equipment itself and improper operation by testing personnel during the testing process of various equipment performance parameters. This method improves the accuracy of error correction by introducing maximum correlation entropy to process non Gaussian noise interference in test data. Applying it to the error correction of performance parameter test data of a certain device， the simulation results show that this method can effectively handle the impact of non Gaussian noise interference on the error correction accuracy in the test data.

Key words: Kalman filtering, maximum correlation entropy, test data, non Gaussian noise, error correction

摘要：

针对各类设备性能参数测试过程中，由于测试设备自身误差和测试人员操作不当带来的测试数据误差问题，提出一种最大相关熵卡尔曼滤波（maximum correlation entropy Kalman filter，MCKF）误差修正方法。该方法通过引入最大相关熵，对测试数据中的非高斯噪声干扰进行处理，提高误差修正精度。将其运用到某设备性能参数测试数据的误差修正中，仿真结果表明：该方法能够有效处理测试数据中的非高斯噪声干扰对误差修正精度带来的影响。

关键词: 卡尔曼滤波, 最大相关熵, 测试数据, 非高斯噪声, 误差修正