We explore the effect of dimensionality on the nearest neighbor problem. We show that under a broad set of conditions (much broader than independent and identically distributed dimensions), as dimensionality increases, the distance to the nearest data point approaches the distance to the farthest data point. To provide a practical perspective, we present empirical results on both real and synthetic data sets that demonstrate that this effect can occur for as few as 10-15 dimensions. These results should not be interpreted to mean that high-dimensional indexing is never meaningful; we illustrate this point by identifying some high-dimensional workloads for which this effect does not occur. However, our results do emphasize that the methodology used almost universally in the database literature to evaluate high-dimensional indexing techniques is flawed, and should be modified. In particular, most such techniques proposed in the literature are not evaluated versus simple linear scan, and are evaluated over workloads for which nearest neighbor is not meaningful. Often, even the reported experiments, when analyzed carefully, show that linear scan would outperform the techniques being proposed on the workloads studied in high (10-15) dimensionality!.

When is nearest neighbor meaningful

random data analysis and measurement procedures is available in our digital library an online access to it is set as public so you can get it instantly. Our book servers spans in multiple countries, allowing you to get the most less latency time to download any of our books like this one. Merely said, the random data analysis and measurement procedures is universally compatible with any devices to read.

Random Data Analysis And Measurement Procedures

Vibration based condition monitoring and fault diagnosis of wind turbine planetary gearbox: A review

https://hal.archives-ouvertes.fr/hal-01714325/document

Fast computation of the spectral correlation

The objective of this paper is to provide a review of recent advances in automatic vibration- and audio-based fault diagnosis in machinery using condition monitoring strategies. It presents the most valuable techniques and results in this field and highlights the most profitable directions of research to present. Automatic fault diagnosis systems provide greater security in surveillance of strategic infrastructures, such as electrical substations and industrial scenarios, reduce downtime of machines, decrease maintenance costs, and avoid accidents which may have devastating consequences. Automatic fault diagnosis systems include signal acquisition, signal processing, decision support, and fault diagnosis. The paper includes a comprehensive bibliography of more than 100 selected references which can be used by researchers working in this field.

Review of Automatic Fault Diagnosis Systems Using Audio and Vibration Signals

Blind deconvolution based on cyclostationarity maximization and its application to fault identification

On the wear and lubrication regime in variable displacement vane pumps

Before marketing external gear pumps are subjected to a running in process to increase their efficiency. However, this is one of the most time-consuming tasks of the entire manufacturing process. Therefore, a mathematical model for optimizing the running in process can be a useful tool for time-to-market reduction. In particular, in this paper a model for the analysis of the dynamic behaviour of external gear pumps, developed by the authors in previous works, is modified and used for simulating the running in process. The modified model is presented and validated via experimental data. A good correlation between simulation and test results guarantees the effectiveness of the model in determining the amount and the distribution of the removed material during the running in process. A meaningful reduction (16%) of the global running in time has been achieved with the introduction of a modified running in process drawn from simulation results.

Simulation of the running in process in external gear pumps and experimental verification

In the field of condition monitoring the availability of a real test-bench is not so common. Furthermore, the early validation of a new diagnostic technique on a proper simulated signal is crucial and a fundamental step in order to provide a feedback to the researcher and to increase the chances of getting a positive result in the real case. In this context, the aim of this paper is to detail a step-by-step analytical model of faulted bearing that the reader could freely and immediately use to simulate different faults and different operating conditions. The vision of the project is a set of tools accepted by the community of researchers on condition monitoring, for the preliminary validation of new diagnostics techniques. The tool proposed in this paper is focused on ball bearing, and it is based on the well-known model published by Antoni in 2007. The features available are the following: selection of the location of the fault, stage of the fault, cyclostationarity of the signal, random contributions, deterministic contributions, effects of resonances in the machine and working conditions (stationary and non-stationary). The script is provided for the open-source Octave environment. The output signal is finally analysed to prove the expected features.

An algorithm for the simulation of faulted bearings in non-stationary conditions

This paper addresses the use of several signal processing tools for monitoring and diagnosis of assembly faults in diesel engines through the cold test technology One specific fault is considered here as an example: connecting rod with incorrectly tightened screws First, the experimental apparatus concerning the vibration tests is introduced Subsequently, the dynamic analysis of the engine has been carried out in order to calculate the connecting rod forces against the crankpin for predicting the position where mechanical impacts are expected Then, a vibration signal model for this type of faults is introduced It deals with the cyclostationary model in which the signal is subdivided into two main parts: deterministic and nondeterministic Finally, the acceleration signals acquired from the engine block during a cold test cycle at the end of the assembly line are analyzed For quality control purposes in order to obtain reliable thresholds for the pass/fail decision, a method based on the image correlation of symmetrized dot patterns is proposed This method visualizes vibration signals in a diagrammatic representation in order to quickly detect the faulty engines in cold tests Moreover, the fault identification is discussed on the basis of the cyclostationary model of the signals The first-order cyclostationarity is exploited by the analysis of the time synchronous average (TSA) In addition, the residual signal is evaluated by subtracting the TSA from the raw synchronized signal, and thus, the second-order cyclostationarity analysis is developed by means of the Wigner–Ville distribution (WVD), Wigner–Ville spectrum (WVS), and mean instantaneous power Moreover, continuous wavelet transform is presented and compared with the WVD and WVS

Gianluca D’Elia

Papers

Blind deconvolution based on cyclostationarity maximization and its application to fault identification

On the wear and lubrication regime in variable displacement vane pumps

Simulation of the running in process in external gear pumps and experimental verification

An algorithm for the simulation of faulted bearings in non-stationary conditions

Advanced Signal Processing Tools for the Vibratory Surveillance of Assembly Faults in Diesel Engine Cold Tests