The limitations of the georeference tests can be overcome, to some degree, with multi-station analysis of reference field measurements as outlined below. Clearly, such tests can only be conducted when measurements are available from a number of survey stations from the same survey. Detailed mathematical descriptions of these tests are given in reference 1.
Multi-station accelerometer test.The basic gravity error test can be expanded to a multi-station accelerometer test when xyz accelerometer measurements are available from a number of survey stations from the same tool. The major difference between the two methods is that the multi-station tests the magnitudes of the individual accelerometer errors, and/or the average resultant sensor error for each accelerometer, while the single-station gravity test tests only the lumped effect of all accelerometer errors. The multi-station test is therefore the more powerful of the two, but has the disadvantage that it can only be used after an extended period of drilling.
Analysis has shown that the quality of the multi-station test in terms of the number of individual error terms that can be estimated increases with a larger number of survey stations, and with increased variation in inclination and toolface. The test is of little or no value for surveys with only small angular variations, such as tangent section and/or constant toolface surveys. In these cases, the test mathematics will be singular, and strange results can be obtained. In such situations, it is better to use a reduced multi-station test with fewer error parameters1.
The Multi-station Gyro Test. This test can be used to estimate the biases and gravity dependent errors for drytuned dual-axis gyro systems, errors which tend to be less stable over time and can therefore vary following factory calibration of the survey tool. The test is also applicable, in principle, for survey tools that use other gyro types. The multi-station gyro test, like the multi-station accelerometer test, is subject to some geometrical limitations. The test is less effective for tangent section and constant tool face surveys, and for surveys of wells that lie predominantly in an east/west direction1.
A general test quality number, the standard deviation of the individual Earth’s rate measurements, is generated as part of this test procedure. This figure becomes excessively large and causes the test to fail in the presence of gross errors in any gyro parameter including errors which are expected to be stable over time; errors which are not estimated individually by the multi-station procedure.
The Multi-station Magnetometer Test. This test can be used to estimate magnetometer biases and scale factor errors, axial and cross axial magnetization errors, and local magnetic field strength and dip errors. The total number of parameters is large, and some terms are closely coupled to each other. A reduction in the number of estimated parameters is frequently implemented to avoid mathematical correlation problems and to obtain results that can be trusted. For this reason, this test is less reliable than the equivalent accelerometer and gyro tests.
The multi-station magnetometer test can be slightly strengthened if merged with the multi-station accelerometer test into a full multi-station analysis4-8. The application of in-field referencing (IFR) further strengthens both this process and the single-station magnetometer test described above.
Feedback
Post your comment on this topic.