论文部分内容阅读
文中提出了综合衰减记忆和限定记忆两种卡尔曼滤波技术的新的实用算法。把模型不准造成的滤波损失归结为噪声带来的影响和旧的观测点带来的影响，根据采样数据的随机统计模式来适应噪声模型和旧的观测点对当前滤波值的影响，弥补模型不准造成的损失。对这种算法进行了仿真研究，共同其它数字滤波进行了比较，解决了剧变和缓变的情况下不能同时有较好的滤波效果的问题，并应用到自然伽马测井仪器中，满足实时测井的要求．在平缓带和过渡带都能有较好的滤波精度。
In this paper, we propose a new practical algorithm that combines two kinds of Kalman filtering techniques, which are attenuated memory and finite memory. Attributed to the impact of noise and the influence of the old observation point, the filter loss caused by the inaccuracy of the model can be adjusted according to the random statistical model of the sampled data to adapt to the influence of the noise model and the old observation point on the current filtered value, Not allowed to cause damage. The algorithm is simulated and compared with other digital filtering to solve the problem of not having good filtering effect in the case of drastic changes and gradual change. The algorithm is applied to natural gamma logging tools to meet real-time measurement Well request. Smooth belt and transition zone can have better filtering accuracy.