What are the rules of digital revision? What are the rules for digital revision?

The numerical modification rule refers to the process of omitting the last few digits of the original value and adjusting the last digit of the retained value before performing specific number operations, so that the final value is closest to the original value.

The specific rules that guide numerical revisions are known as numerical revisionist rules. When making a numerical amendment, the "Revision Interval" and the "Rounding Rule" should be determined first.

Once determined, the revision value must be an integer multiple of the "revision interval". Then specify the expression, i.e., choose to reserve to the specified number of digits according to the Revision Interval.

Unless otherwise specified, the revision of various numerical values measured and calculated in scientific and technological work shall be carried out in accordance with the national standard document "Rules for Numerical Amendment".

In practice, sometimes the test and calculation department will first report the obtained value by one or more digits according to the specified number of correction digits, and then the other departments will make a decision. In order to avoid the error of successive amendments, the following steps should be followed. If the rightmost non-zero digit is 5, it should be added after the value"(+)"or"(-)"or without a sign to indicate that rounding, entering, or not rounding has been made, respectively.

For example, it means that the actual value is greater than that, and it becomes after the amendment and discarding it; It means that the actual value is less than and has been further reduced to one. If it is determined that the quoted value needs to be revised, when the leftmost digit of the number to be discarded is 5 and there is no number or all of them are zero, the value with a (+) sign after it is further one, and the value after the value with a (-) sign is rounded, and the rest is still carried out according to the rules of Chapter 3.

The rounding rule is a numerical reduction rule that people are accustomed to. The specific way to use the rounding rule is to go one place after the number that needs to be reserved, and then round it every five, and round it every four.

For example, if you keep the number to the thousandth place (the third decimal place), since the fourth decimal place is 5, the previous digit should be advanced by one according to this rule, so the result is. Similarly, if all of the following numbers are rounded to two decimal places, the result is:

When making a numerical revision according to the rounding rule, it should be revised to the specified number of digits at a time, and it is not possible to make several revisions, otherwise you may get the wrong result. For example, when reducing a number to a single digit, it should be done in one step: correct).

If you fix it step by step, you will get the wrong result: Error).

In order to avoid such a situation and minimize the error caused by the amendment, it is necessary to use the rounding and five to leave double amendment rules in some cases.

The rules are as follows:

1. When the leftmost digit of the number to be discarded is less than 5, it will be rounded, that is, the reserved digits remain unchanged.

2. The leftmost digit of the number to be discarded is greater than 5; or equal to 5, followed by a digit that is not all 0, then the last digit is retained plus 1. (If the "Revision Interval" is specified, the specified number of digits shall prevail.) ）

3. When the leftmost digit of the number to be discarded is equal to 5, and there is no number on the right or all are 0, if the last digit of the reserved number is an odd number, it will be further one, and if it is an even number (including 0), it will be discarded.

4. When the negative number is revised, the absolute value shall be revised in accordance with the above 1 3 provisions, plus a negative sign.

In accordance with the rules of amendment.

The general reference is to the rounding rule.

However, in the analysis chemistry and some of the more rigorous calculations, the DU uses the rule of rounding to 6. If zhi encounters 5, it should be determined according to whether the previous digit of the contract is odd or even. If it becomes an even number, it will move forward, and if it becomes an odd number, then choose to round.

Use the method of rounding up bai5 to 4.

du, if it is 5, it is necessary to repair the previous dao according to the odd number also.

It's an even number. When the mantissa is 5, and there is any number that is not 0 after the mantissa "5", no matter whether the previous digit is odd or even at this time, and no matter which digit is not 0 after "5", it should be moved forward by one place.

Significant Figures Revision Rules: Rounded to five in two.

If the number to be amended is less than 5, the number will be rounded; When the number to be revised is greater than 5, it is rounded; When the number to be revised is equal to 5, it is necessary to look at the number before 5, if it is an odd number, it will be rounded, if it is an even number, 5 will be rounded, that is, the last number after the amendment will become an even number; If there is any number after 5 that is not "0", then it should be carried regardless of whether the 5 is preceded by an odd or even number.

1 Addition and subtraction: In addition and subtraction, the number of digits of significant digits is retained, whichever has the fewest digits after the decimal point. That is, the absolute error is the largest.

2. Multiplication and division chain rolling method: In the multiplication and division operation, the number of significant digits is retained, and the number with the least number of digits shall prevail, that is, the number with the largest relative error shall prevail.

3.In quantitative constant analysis, four significant figures are generally retained, but in central control, only two or three significant digits are required.

The rules for the revision of significant figures are:

1. Round off to five pairs.

The number to be modified in the measured value is discarded when it is equal to or less than 4, and rounded when it is equal to or greater than 6. When it is equal to 5 and infinite after 5, if the last digit of the measured value after carry is an even number, then it is carried; After rounding, the last digit of the measured value is an odd number and is discarded. If there is a number after 5, it means that the number of revisions is larger than 5, and it is advisable to carry it forward.

For example, if the following measurements are reduced to three digits, the result of the reduction is: for, for, for, for, for.

2. The original measured value should be revised to the required number of digits at one time, and the contract cannot be revised in stages.

For example, if you want to fix a three-digit contract, you can't fix it first and then make it, you can only fix it once.

3. In the process of operation, in order to reduce the rounding error, one more significant digit can be retained, and after the result is calculated, the result will be revised to the due number of significant digits according to the revision rules.

In the case of a standard deviation or other value that represents uncertainty, the result of the correction should make the estimate of accuracy worse. For example, standard deviation (s) = , if two significant figures are taken, it is advisable to revise the approximation, and if one is taken, it is.

Significant number crunching rules

1. Addition and subtraction: first according to the data with the least number of digits after the decimal point, retain the number of digits of other digits, and then add and subtract the calculation, and the calculation result also retains the same number of digits after the decimal point.

2. Multiplication and division: first retain the other numbers according to the data with the least significant digits, and then multiply and divide the operation, and the calculation result still retains the same significant digits.

The numerical modification rule refers to the process of omitting the last digits of the original value and adjusting the last digit of the reserved value to make the final value closest to the original value before the specific number operation is carried out.

The specific rules that guide numerical revisions are known as numerical revisionist rules. When making a numerical amendment, the "Revision Interval" and the "Rounding Rule" should be determined first.

Once determined, the revision value must be an integer standby or multiple of the Revision Interval. Then specify the expression, i.e., choose to reserve to the specified number of digits according to the Revision Interval.

Unless otherwise specified, the various numerical values measured and calculated in scientific and technological work shall be revised in accordance with the national standard document "Rules for Numerical Amendment".

The rules for amendments are rounded to five and doubles:

1. When the mantissa is less than or equal to 4, the mantissa is directly rounded.

For example:; 2. When the mantissa is greater than or equal to 6, the mantissa will be rounded to the previous digit.

For example:; 3. When the mantissa is 5, and the numbers after the mantissa are 0, you should look at the first digit of the mantissa "5": if the previous digit is an odd number at this time, it should move forward one place; If the previous digit is even, the mantissa should be rounded. The number "0" should be considered even at this point.

For example:; 4. When the mantissa is 5, and there is any number that is not 0 after the mantissa "5", no matter whether the previous digit is odd or even at this time, and no matter which digit is not 0 after "5", it should be moved forward by one place. For example:;

The revision rule refers to the principle based on which the number of digits of the actual measurement data is enumerated according to the error of the measuring instrument and the experimental method and the requirements for the accuracy of the measurement data. It adopts the "rounding to five" criterion, i.e., if the first digit after the valid allocation number is 4, it is rounded; If it is 6, it will be 1 in the previous digit; If it is exactly 5, and the digits after 5 are not all 0, then the digit before 5 is 1; If the digit before 5 is an odd number, it will be rounded off.