Research on MEMS Anomaly Data Detection Model Based on If Algorithm

For high arch displacement dam monitoring during the construction, the demands of the MEMS inclinometer analyzed the measuring principle and measurement method resolution of each measuring position. Aiming at the outliers in the measurement process of MEMS inclinometer, this paper proposes an anomaly detection model based on wavelet analysis and improved isolated forest algorithm. By setting the test in creep state and mutation state, according to the result of eliminating outliers after model testing and the actual test adjustment displacement, the recognition rate of the traditional isolated forest method is 75 %, and the anomaly recognition law of the proposed model reaches <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 0 \%}$</tex>. The outlier detection effect is obviously better than that of the isolated forest model, which can effectively identify the outliers in displacement monitoring of the MEMS clinometer.