5 Experimental Fluid Mechanics
375
these factors affect the result always shifting in one direction, and the
size and symbol are exactly the same in the same set of test measurements.
When the test conditions are determined, the systematic error obtains an
objective constant value. When the test conditions are changed, the variation of the systematic error can be found. The cause of the system error:
the measuring instrument is not good, such as the scale is not accurate,
the instrument zero point is not corrected or the standard table itself
has deviation; Changes in the surrounding environment, such as temperature, pressure, humidity, etc., deviate from the calibration value; Test
personnel’s habits and biases, such as errors caused by high or low readings. For the shortcomings of the instrument, the influence of changes
in external conditions, and the individual’s bias, the system error can be
basically cleared after being corrected separately.
(2) Random error: In the observation of all the measured values that have
eliminated the system error, the measured data are still different in the
last digit or the last two digits, and their absolute value and symbol
change, sometimes large or small, sometimes positive or negative, with
no definite law. Such errors are called random errors or accidental errors.
The cause of the random error is unknown and cannot be controlled
and compensated. However, if a certain value is measured with enough
precision many times, it will be found that the random error completely
obeys the statistical law, and the size of the error or the probability of
positive and negative occurrence can be determined. Therefore, as the
number of measurements increases, the arithmetic mean of the random
errors approaches zero, so the arithmetic mean of multiple measurements
will be closer to the real value.
(3) Negligence error: Negligence error is an error that is obviously inconsistent with the facts. It is often caused by carelessness, excessive fatigue, and
incorrect operation of the tester. Such errors can be found without rules.
As long as the sense of responsibility, vigilance, and careful operation are
strengthened, the negligence error can be avoided.
2. Precision, Correctness, and Accuracy
The amount that reflects how close the measurement is to the true value is
called accuracy. It corresponds to the error size, and the higher the accuracy
of the measurement, the smaller the measurement error. “Accuracy” should
include two meanings of precision and correctness.
375
these factors affect the result always shifting in one direction, and the
size and symbol are exactly the same in the same set of test measurements.
When the test conditions are determined, the systematic error obtains an
objective constant value. When the test conditions are changed, the variation of the systematic error can be found. The cause of the system error:
the measuring instrument is not good, such as the scale is not accurate,
the instrument zero point is not corrected or the standard table itself
has deviation; Changes in the surrounding environment, such as temperature, pressure, humidity, etc., deviate from the calibration value; Test
personnel’s habits and biases, such as errors caused by high or low readings. For the shortcomings of the instrument, the influence of changes
in external conditions, and the individual’s bias, the system error can be
basically cleared after being corrected separately.
(2) Random error: In the observation of all the measured values that have
eliminated the system error, the measured data are still different in the
last digit or the last two digits, and their absolute value and symbol
change, sometimes large or small, sometimes positive or negative, with
no definite law. Such errors are called random errors or accidental errors.
The cause of the random error is unknown and cannot be controlled
and compensated. However, if a certain value is measured with enough
precision many times, it will be found that the random error completely
obeys the statistical law, and the size of the error or the probability of
positive and negative occurrence can be determined. Therefore, as the
number of measurements increases, the arithmetic mean of the random
errors approaches zero, so the arithmetic mean of multiple measurements
will be closer to the real value.
(3) Negligence error: Negligence error is an error that is obviously inconsistent with the facts. It is often caused by carelessness, excessive fatigue, and
incorrect operation of the tester. Such errors can be found without rules.
As long as the sense of responsibility, vigilance, and careful operation are
strengthened, the negligence error can be avoided.
2. Precision, Correctness, and Accuracy
The amount that reflects how close the measurement is to the true value is
called accuracy. It corresponds to the error size, and the higher the accuracy
of the measurement, the smaller the measurement error. “Accuracy” should
include two meanings of precision and correctness.