Algorithm Of Track Spatial Alignment Inspection Based On Intertial Measurement Platform

The track irregularity is the main source of train excitation, affecting the operationsafety and riding comfort. Track Inspection has a vital importance in track maintance andscientific research. And the track long-wave irregularity parameters are determined by thecalculation of track spatial alignment parameters. Based on the inertial measurementplatform, this paper used track state inspection method to acquire the measuringinformation from IMU (Inertial Measurement Unit) and denoised it by the EMD(Empirical Mode Decomposition) method. Also, by using digital integrator, this paperacquired the spatial alignment of the carrier, and finally, made a compensation for thealignment by getting the track angle information from a inclinometer. At last, the trackdeformation will be quantificationally acquired through the conparison of different linearresults from multi-times detection under the same datum point.The main research contentsare:1. On the base of strap-down inertial navigation technology, accelerometers were usedto measure the acceleration signals on three axises, gyroscopes were used to acquire theangular rate signals, through which attitude matrix was updated in real time, so that theacceleration signals would be transferred to the geographic coordinate system. After that,the spatial alignment would be acquired through twice continuous integration.2. According to the spatial sampling theorem, an equi-spaced sampling system wasdesigned, through which, the inconsistency of time-based information calculation and trackspatial alignment was eliminated. While the traditional time-based sampling method mightcause the change of excitation frequency along with the change of velocity, which wouldcause the difficulty in filter designing.3. By using the IMU signals, a digital integrator was designed to acquire the trackspatial alignment. As to the integrator saturation caused by low frequency in theacceleration signals, Hilbert Huang Transform (HHT) was used to filter it, and the verylow-frequency component would be eliminated based on the EMD method. 4. A inclinometer was introduced in and a data acquisition system was designed to getthe angel information, by which, a arithmetic was designed to compensate the track spatialalignment.Through a series of experiments on the inertia measurement platform, the paperdemostrated that it is workable to use strap-down inertial technology for track statedetection, the equi-spaced sampling system and the signal data acquisition system werefeasible, and it has emarkable effect in low frequency denoising by EMD method, and also,the angel information of three axises measured by the inclinometer are useful in thedisplacement compensation. The result showed that, the detection errer was within4mm,and it met the requirement.