Outlier Detection by means of Monte Carlo Estimation including resistant Scale Estimation

2015 ◽  
Vol 9 (2) ◽  
Author(s):  
Christian Marx
Keyword(s):  
Monte Carlo ◽  
Measurement Data ◽  
Estimation Method ◽  
Bayes Estimation ◽  
Scale Estimation ◽  
Unknown Variance ◽  
Least Squares Adjustment ◽  
Monte Carlo Search ◽  
Monte Carlo Estimation ◽  
Variance Factor

AbstractThe identification of outliers in measurement data is hindered if they are present in leverage points as well as in rest of the data. A promising method for their identification is the Monte Carlo estimation (MCE), which is subject of the present investigation. In MCE the data are searched for data subsamples without leverage outliers and with few (or no) non-leverage outliers by a random generation of subsamples. The required number of subsamples by which several of such subsamples are generated with a given probability is derived. Each generated subsample is rated based on the residuals resulting from an adjustment. By means of a simulation it is shown that a least squares adjustment is suitable. For the rating of the subsamples, the sum of squared residuals is used as a measure of the fit. It is argued that this (unweighted) sum is also appropriate if data have unequal weights. An investigation of the robustness of a final Bayes estimation with the result of the Monte Carlo search as prior information reveals its inappropriateness. Furthermore, the case of an unknown variance factor is considered. A simulation for different scale estimators for the variance factor shows their impracticalness. A new resistant scale estimator is introduced which is based on a generalisation of the median absolut deviation. Taking into account the results of the investigations, a new procedure for MCE considering a scale estimation is proposed. Finally, this method is tested by simulation. MCE turns out to be more reliable in the identification of outliers than a conventional resistant estimation method.

Parameter Estimation of an Electrohydraulic Servo System Using a Markov Chain Monte Carlo Method

2012 ◽  
Vol 135 (1) ◽  
Author(s):  
Junhong Liu ◽  
Huapeng Wu ◽  
Heikki Handroos ◽  
Heikki Haario
Keyword(s):  
Monte Carlo ◽  
Parameter Estimation ◽  
Markov Chain ◽  
Markov Chain Monte Carlo ◽  
Measurement Data ◽  
Estimation Method ◽  
Operating Conditions ◽  
Point Estimate ◽  
Best Fitting ◽  
The One

A parameter estimation method is presented by an example of an electrohydraulic position servo. The method is based on the Markov chain Monte Carlo approach. The method allows utilization of noisy measurement data in identification process, making use of original physical data possible without the requirement of a filter. The method seeks for the best fitting point estimate of the unknown model parameter vector, but the solution to the parameter estimation problem is given as a statistical distribution that contains “all” the possible parameter combinations. The robustness of the model developed with the proposed method is further demonstrated by verification in operating conditions that are independent of each other and the one used in the identification step. Results show that the system model with the hybrid leakage formula for the studied valve describes the system dynamics more precisely and matches the real responses better.

Neutron dose evaluation of Elekta Linac at two energies (10 & 18 MV) by MCNP code and comparison with experimental measurements

2014 ◽  
Vol 6 (1) ◽  
pp. 1006-1015
Author(s):  
Negin Shagholi ◽  
Hassan Ali ◽  
Mahdi Sadeghi ◽  
Arjang Shahvar ◽  
Hoda Darestani ◽  
...  
Keyword(s):  
Monte Carlo ◽  
Measurement Data ◽  
Calculated Data ◽  
High Energy ◽  
Neutron Dose ◽  
Dose Assessment ◽  
Photon Flux ◽  
Dose Equivalent ◽  
Monte Carlo Techniques ◽  
Neutron Dose Equivalent

Medical linear accelerators, besides the clinically high energy electron and photon beams, produce other secondary particles such as neutrons which escalate the delivered dose. In this study the neutron dose at 10 and 18MV Elekta linac was obtained by using TLD600 and TLD700 as well as Monte Carlo simulation. For neutron dose assessment in 2020 cm2 field, TLDs were calibrated at first. Gamma calibration was performed with 10 and 18 MV linac and neutron calibration was done with 241Am-Be neutron source. For simulation, MCNPX code was used then calculated neutron dose equivalent was compared with measurement data. Neutron dose equivalent at 18 MV was measured by using TLDs on the phantom surface and depths of 1, 2, 3.3, 4, 5 and 6 cm. Neutron dose at depths of less than 3.3cm was zero and maximized at the depth of 4 cm (44.39 mSvGy-1), whereas calculation resulted  in the maximum of 2.32 mSvGy-1 at the same depth. Neutron dose at 10 MV was measured by using TLDs on the phantom surface and depths of 1, 2, 2.5, 3.3, 4 and 5 cm. No photoneutron dose was observed at depths of less than 3.3cm and the maximum was at 4cm equal to 5.44mSvGy-1, however, the calculated data showed the maximum of 0.077mSvGy-1 at the same depth. The comparison between measured photo neutron dose and calculated data along the beam axis in different depths, shows that the measurement data were much more than the calculated data, so it seems that TLD600 and TLD700 pairs are not suitable dosimeters for neutron dosimetry in linac central axis due to high photon flux, whereas MCNPX Monte Carlo techniques still remain a valuable tool for photonuclear dose studies.

Estimation of Track Irregularity Based on Genetic Algorithm and Unscented Kalman Filtering

2012 ◽  
Author(s):  
Hongmei Shi ◽  
Zujun Yu
Keyword(s):  
Genetic Algorithm ◽  
Kalman Filtering ◽  
High Speed ◽  
Measurement Data ◽  
Estimation Method ◽  
Coupling Model ◽  
Dynamic Responses ◽  
The Real ◽  
True Value ◽  
Track Irregularity

Track irregularity is the main excitation source of wheel-track interaction. Due to the difference of speed, axle load and suspension parameters between track inspection train and the operating trains, the data acquired from the inspection car cannot completely reflect the real status of track irregularity when the operating trains go through the rail. In this paper, an estimation method of track irregularity is proposed using genetic algorithm and Unscented Kalman Filtering. Firstly, a vehicle-track vertical coupling model is established, in which the high-speed vehicle is assumed as a rigid body with two layers of spring and damping system and the track is viewed as an elastic system with three layers. Then, the static track irregularity is estimated by genetic algorithm using the vibration data of vehicle and dynamic track irregularity which are acquired from the inspection car. And the dynamic responses of vehicle and track can be solved if the static track irregularity is known. So combining with vehicle track coupling model of different operating train, the potential dynamic track irregularity is solved by simulation, which the operating train could goes through. To get a better estimation result, Unscented Kalman Filtering (UKF) algorithm is employed to optimize the dynamic responses of rail using measurement data of vehicle vibration. The simulation results show that the estimated static track irregularity and the vibration responses of vehicle track system can go well with the true value. It can be realized to estimate the real rail status when different trains go through the rail by this method.

scholarly journals Revised Estimation Method for Emissions from Automated Plunger Lift Liquid Unloadings

Environments ◽  
2020 ◽  
Vol 7 (4) ◽  
pp. 25
Author(s):  
Adam Pacsi ◽  
David W. Sullivan ◽  
David T. Allen
Keyword(s):  
Direct Measurement ◽  
Oil And Gas ◽  
Measurement Data ◽  
Estimation Method ◽  
The United States ◽  
Gas Production ◽  
Small Subset ◽  
Emission Estimate ◽  
National Measurement ◽  
The Individual

A variety of liquid unloading techniques are used to clear accumulated liquids from the wellbore to increase production rates for oil and gas wells. Data from national measurement studies indicate that a small subset of wells with plunger lift assist, that vent with high frequency and short event duration, contribute a significant fraction of methane emissions from liquid unloading activities in the United States. Compared to direct measurement of emissions at 24 wells in a field campaign, the most commonly used engineering emission estimate for this source category, which is based on the volume of gas in the wellbore, does not accurately predict emissions at the individual well (R2 = 0.06). An alternative emission estimate is proposed that relies on the duration of the venting activity and the gas production rate of the well, which has promising statistical performance characteristics when compared to direct measurement data. This work recommends well parameters that should be collected from future field measurement campaigns that are focused on this emission source.

Sign in / Sign up

Export Citation Format

Share Document