Browse Prior Art Database

Data Collection Using Trends Disclosure Number: IPCOM000225855D
Publication Date: 2013-Mar-08
Document File: 2 page(s) / 86K

Publishing Venue

The Prior Art Database


The objective of this idea is to recreate a signal using samples that are multi-dimensional, derivate based samples instead of the traditional two dimensional X-Y representation using date/time stamps and scalar sample values, particularly for down-hole (wellbore) applications. The method outlined in the following will use inflection points and derivatives to describe a curve instead of a collection of data points. It is our intention to increase the apparent bandwidth of a system by sending more information with each data point and fewer samples. We assert that, by using the inflection points and derivative information we can get a broader picture of the behavior of samples with minimal transmission of data. There are various theorems for determining appropriate sample rate to reconstruct values, but in data acquisition, sample rates are determined by customer need. The acoustic communication system currently used to get real time data from down hole environments to the surface are bandwidth limited and only allow a small portion of the collected data to be transmitted to the surface. This method will use more information to reconstruct the shape of the curve to give a more accurate picture of the data prior to retrieval of the full sample set. In calculus, any two-dimensional function can be described graphically as a point represented by coordinates along the X and Y axis. The current data acquisition model grabs as many of these samples as possible and then transmits a packet of information to the surface as a time stamp for the X axis and Y value representing the sampled value. This method is accurate in real time, but is limited by the bandwidth of the system; however, by increasing the information about the shape of the signal, we can send more detail about the signal shape. The second derivative of the sample provides the rate of change. With this, we can accurately predict the samples that are adjacent to the collected value. This alone will increase the estimation of the data while sending smaller packets of information. Sending the first derivate with any given point allows us to derive other points, but we can get more accurate prediction by sending the second derivative. The second derivative shows us the rate of change of the first derivative. This in turn will allow us to improve our predictive modeling by making the first derivate used to calculate adjacent samples a more accurate prediction. By sending the sample point and first and second derivatives, we assert that reasonable real-time predictions can be made to provide the shape of a down hole sample curve with greater accuracy and fewer samples sent. At a minimum, the samples must be sent at inflection points, but with both derivatives at each inflection point, the system can accurately provide curve information to the next sample. Generally, inflection points are defined when the first or second derivative become zero. This is an overly simplistic explanation of inflection, but is a sufficient condition for the purposes of this sample.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2


Data Collection Using Trends 


The objective of this idea is to recreate a signal using samples that are multi‐dimensional, derivate based  samples instead of the traditional two dimensional X‐Y representation using date/time stamps and  scalar sample values, particularly for down‐hole (wellbore) applications.  The method outlined in the  following will use inflection points and derivatives to describe a curve instead of a collection of data  points. 

It is our intention to increase the apparent bandwidth of a system by sending more information with  each data point and fewer samples.  We assert that, by using the inflection points and derivative  information we can get a broader picture of the behavior of samples with minimal transmission of data. 


There are various theorems for determining appropriate sample rate to reconstruct values, but in data  acquisition, sample rates are determined by customer need.  The acoustic communication system  currently used to get real time data from down hole environments to the surface are bandwidth limited  and only allow a small portion of the collected data to be transmitted to the surface. 

This method will use more information to reconstruct the shape of the curve to give a more accurate  picture of the data prior to retrieval of the full sample set.  In calculus, any two‐dimensional function can  be described graphically as a point represented by coordinates along the X and Y axis.  The current data  acquisition model grabs as many of these samples as possible and then transmits a packet of  information to the surface as a time stamp for the X axis and Y value representing the sampled value.   This method is accurate in real time, but is limited by the bandwidth of the system; however, by  increasing the information about the shape of the signal, we can send more detail about the signal  shape.  The second derivative of the sample provides the rate of change. With this, we can accurately  predict the samples that are adjacent to the collected value.  This alone will increase the estimation of  the data while send...