Research on Multi - temperature Sensor Data Fusion Method Based on Improved Kalman Filter

Multi-sensor data fusion has always been a difficulty in the field of power temperature measurement. Traditional data fusion methods often use Kalman filtering, and the performance of Kalman filtering is highly dependent on the selected filter parameters. If the filter parameters are not selected properly, it will affect the filter gain and greatly reduce the reliability and fault tolerance of the fusion results. This is because the filter parameters directly affect the weight allocation of the filter to the observed values and model predicted values, and improper allocation can lead to a large deviation between the fusion result and the true value, thus greatly reducing the reliability and fault tolerance of the fusion results. Based on the traditional Kalman filtering algorithm, this paragraph describes an improved data fusion method that dynamically estimates the temperature values of multiple temperature measurement points by updating the covariance matrix of the state variables to improve the Kalman filter gain. Specifically, this method performs data fusion for multiple temperature measurement points on the basis of the traditional Kalman filtering algorithm, and updates the covariance matrix of the state variables to enable the Kalman filter gain to more accurately reflect the data contributions of different measurement points. This enables dynamic tracking of multiple temperature measurement points and more accurate estimation of their temperature values. Through simulation and experimental analysis, this improved Kalman filtering data fusion method demonstrates high credibility and fault tolerance, which means it can work stably in complex environments and correctly handle issues such as data anomalies and measurement errors. Therefore, this method can be widely applied to data fusion processing of multiple temperature measurement points, thereby improving the accuracy and reliability of data processing.