Application of Adaptive UKF Algorithm in Multi-target Tracking and Positioning System

Adaptive filtering algorithm (FIR) is a design method of adaptive variable target tracking system based on probability density distribution model. The algorithm realizes the target movement in the global range by estimating the parameters of different regions in the image, which improves the real-time performance and effectiveness.

Clear-Air Turbulence (CAT) Identification with X-Band Dual Polarimetric Radar Based on Bayesian Approach

The echo of weather radar is seriously disturbed by clear-air turbulence echo (CAT) which needs identifying and eliminating to improve the data quality of weather radar. Using the data observed with the five X-band dual polarimetric radars in Changping, Fangshan, Miyun, Shunyi, and Tongzhou, Beijing in 2018, the probability density distribution (PDD) of the horizontal texture of four radar moments reflectively factor (ZH), differential reflectivity (ZDR), correlation coefficient (ρHV), differential propagation phase shift (ΦDP), and then the CAT is identified and removed using Bayesian method. The results show that the radar data can be effectively improved after the CAT has been eliminated, which include: (1) the removal rate of CAT is more than 98.2% in the analyzed cases. (2) In the area with high-frequency distribution of CAT, the CAT can be effectively suppressed; in the area with low-frequency distribution, some weather echo in the edge with SNR < 15 dB may be mistakenly identified as CAT, but the proportion of meteorological echoes to the total echoes is more than 85%, which indicate that the error rate is very low and does not affect the radar operation.

HAZARD IDENTIFICATION AND PROBABLISTIC SCENARIO SELECTION FOR SHIP- SHIP COLLISION ACCIDENTS

In collision risk-based design frameworks it is necessary to accurately define and select a set of credible scenarios to be used in the quantitative assessment and management of the collision risk between two ships. Prescriptive solutions and empirical knowledge are commonly used in current maritime industries, but are often insufficient for innovation because they can result in unfavourable design loads and may not address all circumstances of accidents involved. In this study, an innovative method using probabilistic approaches is proposed to identify relevant groups of ship-ship collision accident scenarios that collectively represent all possible scenarios. Ship-ship collision accidents and near-misses recently occurred worldwide are collated for the period of 21 years during 1991 to 2012. Collision scenarios are then described using a set of parameters that are treated individually as random variables and analysed by statistical methods to define the ranges and variability to formulate the probability density distribution for each scenario. As the consideration of all scenarios would not be practical, a sampling technique is applied to select a certain number of prospective collision scenarios. Applied examples for different types of vessels are presented to demonstrate the applicability of the method.

A Fault Diagnosis Method Based on EEMD and Statistical Distance Analysis

Serious vibration or wear with large friction usually appear when faults occur, which leads to more serious faults such as the destruction of the oil film, bringing great damages to both the society and economic sector. Therefore, the accurate diagnosis of a fault in the early stage is important for the safety operation of machinery. To effectively extract the fault features for diagnosis, EMD-based methods are widely used. However, these methods spend lots of efforts diagnosing faults and require plenty of professional knowledge of diagnosis. Although many intelligent classifiers can be used to automatically diagnose faults such as wear, a broken tooth and imbalance, the combing EMD-based method, the scarcity of samplings with labels hinder the application of these methods to engineering. It is because the model of the intelligent classifier must be constructed based on sufficient samplings with a label. To solve this problem, we propose a novel fault diagnosis method, which is performed based on the EEMD and statistical distance analysis. In this method, the EEMD is used to decompose one original signal into several IMFs and then the probability density distribution of each IMF is calculated. To diagnose the fault of the machinery, the Euclidean distance between the signal acquired under an unknown fault with the other referenced signals acquired previously under various fault types is calculated. At last, the fault of the signal is the same with the referenced signal when the distance is the smallest. To verify the effectiveness of our proposed method, a dataset of bearings with different faults, and a dataset of 2009 Prognostics and Health Management (PHM) data challenge, including gear, bearing and shaft faults are used. The result shows that the proposed method can not only automatically diagnose faults effectively, but also fewer samplings with a label are used compared with the intelligent methods.

Determination of Length of Individual Pellets and Pellets’ Lengths Distribution

A form and dimensions of fuel particles influence the intensity of their burning and approaches to the mathematic description of the process. Known methods do not allow correctly measuring all pellets’ lengths and describing pellets’ lengths distribution. The purpose of the study is to substantiate method for determining the individual pellet length and to specify statistical characteristics of pellets’ lengths distribution. The purpose was achieved by applying the proposed method of indirect determination of the length of each pellet by weighing it, followed by calculation of the equivalent length and modal cluster analysis of the distribution of pellets by length, based on the probability density distribution. The most noteworthy results are that the experimental calculation of the equivalent length gives results that coincide with direct measurements for pellets of the correct shape, but in contrast to direct measurements can also be used to determine the equivalent lengths of irregularly shaped pellets and their fragments. Clustering allowed grouping pellets around objectively existing local maxima in the probability density distribution, which can be identified at intervals of pellet lengths not exceeding 2 mm. The importance of the obtained results is that the indirect method of determining the length of pellets allows replacing the measurement of pellet lengths by their weighing, which eliminates subjective factors when measuring the length of irregularly shaped pellets and their fragments. Clustering characterised the granulometric composition of pellets with histograms of probability, mass fraction, and average length by clusters. Upon using proposed approaches, granulometric composition of industrially produced pellets was specified and increased probabilities were noted for 8 mm pellets in clusters of smaller lengths, compared to 6 mm pellets; while straw pellets are characterised by a higher probability in clusters with shorter lengths compared to wood pellets

Augmented Reality Assisted Assembly Training Oriented Dynamic Gesture Recognition and Prediction

Augmented reality assisted assembly training (ARAAT) is an effective and affordable technique for labor training in the automobile and electronic industry. In general, most tasks of ARAAT are conducted by real-time hand operations. In this paper, we propose an algorithm of dynamic gesture recognition and prediction that aims to evaluate the standard and achievement of the hand operations for a given task in ARAAT. We consider that the given task can be decomposed into a series of hand operations and furthermore each hand operation into several continuous actions. Then, each action is related with a standard gesture based on the practical assembly task such that the standard and achievement of the actions included in the operations can be identified and predicted by the sequences of gestures instead of the performance throughout the whole task. Based on the practical industrial assembly, we specified five typical tasks, three typical operations, and six standard actions. We used Zernike moments combined histogram of oriented gradient and linear interpolation motion trajectories to represent 2D static and 3D dynamic features of standard gestures, respectively, and chose the directional pulse-coupled neural network as the classifier to recognize the gestures. In addition, we defined an action unit to reduce the dimensions of features and computational cost. During gesture recognition, we optimized the gesture boundaries iteratively by calculating the score probability density distribution to reduce interferences of invalid gestures and improve precision. The proposed algorithm was evaluated on four datasets and proved to increase recognition accuracy and reduce the computational cost from the experimental results.

Characteristics of the Near-Ground Typhoon Morakot

Two WJ-3 anemometers placed at the same height on the top of an architectural engineering building in Wenzhou University are used to determine the wind speed of Typhoon Morakot during its landing in real time. This study aims to explore Typhoon Morakot’s wind field characteristics, including mean wind speed, probability density distribution of fluctuating wind speed, power spectral density, correlation analysis, and coherence, on the basis of data measured by the two anemometers. Results show that the probability density distribution of the fluctuating wind speed of the typhoon follows the Gaussian distribution, and the measured cross-power spectrum of fluctuating wind speed is in good agreement with the modified Karman spectrum. The autocorrelation decreases with the increase in time interval (τ). The longitudinal autocorrelation coefficient decays rapidly with the increase in τ, and the lateral autocorrelation coefficient decays at an unchanged rate. The exponential attenuation coefficients of the longitudinal and transverse fluctuating wind speeds increase with the increase in the mean wind speed, and their mean values are 10.86 and 15.33, respectively. The change trends of the coherence coefficients of the two wind speed components with the mean wind speed are the same. The measured coherence coefficients of the two wind speed components are in good agreement with the exponential function.

Experimental Study on the Influence Factors of Flow Velocity Structure and Turbulent Characteristics in Open Channel with Biomimetic Grass

The problem of suspension treatment of subsea oil-gas pipelines has been highly concerned by engineering construction units and researchers. The current research indicates that the bionic sea grass can effectively reduce the flow rate, promote sediment deposition, and control the development of the pipeline suspension area. The velocity distribution of open channel flow with bionic grass is very complex. The height and laying space of bionic grass will affect the flow velocity distribution. At present, the flow velocity in open channels with bionic grass is mainly studied by measuring the velocity variation at the front, middle, and back of bionic grass. Few effective measurements are made for the full velocity field. The velocity field distribution of bionic aquatic grass along the vertical plane is measured by using standard particle image velocimetry (PIV). The effects of height and laying space of bionic grass on probability density distribution, spatial correlation of pulsating velocity, turbulence intensity, Reynolds stress and turbulent kinetic energy in the open channel after the protection section of bionic grass are further analyzed.