CN103810366A - Rapid grouping method for precise elements - Google Patents

Abstract

The invention belongs to the field of precise instruments for aviation and particularly relates to a rapid grouping method for precise elements. With the adoption of the rapid grouping method for the precise elements, provided by the invention, the automation of a grouping process can be realized. Data can be processed by using a strong computing capacity of a computer, so that the work, which needs to be finished for some hours originally, can be immediately finished; a result can be automatically stored. With the adoption of the rapid grouping method for the precise elements, the precise elements failing to meet the requirements can be randomly replaced, so that the grouping work can be finished in a maximum and optimal manner; an optimal scheme can be rapidly found out through random selection and random replacement by using the strong computing capacity of the computer, so that not only can the human resources can be released, but also the grouping quality and the grouping stability can be improved. Meanwhile, the working efficiency can be greatly improved.

Description

A kind of precision element fast grouping method

Technical field

The invention belongs to aviation field of precision instruments, particularly relate to a kind of precision element fast grouping method.

Background technology

Along with constantly improving and development of aeronautical product technology, be widely used as the redundance product sensor of system information feedback.Current redundance sensor be single lot sensor by artificial judgment performance quality, several be singlely divided into different combinations by what meet remaining coherence request, these are combined in and carry out general assembly and performance adjustment realizes remaining product function.Whether can realize the factors such as experience that maximized grouping, optimized grouping be limited by grouping personnel, method, working attitude, because artificial grouping process need to be put dozens or even hundreds of form together and be carried out corresponding data and carry out comparing calculation, selecting suitable one group puts aside, all the other forms are being processed equally, until complete.But can not meet grouping while requiring when having more than the single product of expection, the part combination being divided into group need to be upset and redistributes, work repeatedly, inefficiency, success ratio be low, also exists and miscopy, and misunderstands the possibility of miscalculation.

Summary of the invention

In the object of the invention is that all every batch products data are automatically kept at a electronic data file in the time testing, utilize program automatically to complete a kind of precision element fast grouping method of distinguishing combination by certain requirement.

The technical scheme that the present invention takes is: a kind of precision element fast grouping method, comprises the following steps:

Step 1, the precision element that each is processed carry out serial number, and the test parameter of each precision element is input to same form, form and treat group list;

Step 2, from treat group list, first take out from front to back the precision element of n, n is grouping radix;

Step 3, judge whether the test parameter of this n precision element meets allowable tolerance Δ: if the error of the test parameter of this n precision element between is mutually no more than allowable tolerance Δ, meet allowable tolerance Δ, the test parameter of this n precision element is deposited into group list, and at the test parameter for the treatment of to delete in group list this n precision element, enter step 4; If do not meet allowable tolerance Δ, enter step 5:

Step 4, from treat group list, again take out from front to back n precision element, repeating step three is to step 5, until treat the part precision element that all precision elements divide into groups successfully or only residue cannot be divided into groups in group list.

Step 5, from treating that group list chooses at random a precision element and replace at random this n and treat point set of pieces, return to step 3; Until the test parameter of this n precision element meets allowable tolerance Δ, enter step 4.

Described grouping radix n and allowable tolerance Δ are determined concrete numerical value according to the actual requirements.

The advantage that the present invention has and beneficial effect: traditional grouping pairing needs manually to calculate contrast and carries out, by the performance quality of the single sensor of artificial judgment lot, several be singlely divided into different combinations by what meet remaining coherence request.Whether can realize the factors such as experience that maximized grouping, optimized grouping be limited by grouping personnel, method, working attitude, workload is large, and error probability is large, and group result superiority-inferiority is not etc.This grouping process is realized robotization by the present invention.Utilize the powerful computing power of computing machine to process data, make the work that originally need to just can complete in a few hours, complete in moment, and saving result automatically.Undesirable precision element is replaced at random, can be maximized, optimization completes grouping work, utilizes the powerful arithmetic capability of computing machine, adopt and choose at random and replace and can find rapidly preferred plan at random.Not only liberating human resources improves grouping quality simultaneously, divides organizing, stability to improve greatly work efficiency simultaneously.

Accompanying drawing explanation

Fig. 1 is process flow diagram of the present invention.

Embodiment

Below in conjunction with instantiation, the present invention is described in further detail.

Step 1, the precision element that each is processed carry out serial number, and the test parameter of each precision element is input to same form, form and treat group list; The form of this form need to meet the set form that this method requires, and test parameter comprises the needed all parameters of this type product grouping

Step 2, from treat group list, first take out from front to back the precision element of n, " n " is the grouping radix of this type product requirement;

Step 3, judge this just the test parameter of this n precision element whether meet allowable tolerance Δ: all test parameters to this n precision element carry out across comparison, for product reaches best performance, allowable tolerance Δ is done to the suitable Δ that is reduced into according to actual conditions and experience ₁, be generally 40%～60% of allowable tolerance Δ.If the error between the test parameter of this n precision element is mutual is no more than allowable tolerance Δ ₁, meet allowable tolerance Δ ₁, the test parameter of this n precision element is deposited into group list, and at the test parameter for the treatment of to delete in group list this n precision element, enters step 4; If do not meet allowable tolerance Δ ₁, enter step 5:

Step 4: again take out from front to back n precision element from treat group list, repeating step three, until treat the part precision element that all precision elements divide into groups successfully or only residue cannot be divided into groups in group list.Cannot divide into groups to be divided into two kinds of situations: situation one, the test parameter between residue precision element differs too large, can not meet allowable tolerance requirement, situation two, the number of residue precision element is less than grouping radix n.

Step 5, from treating that group list chooses at random a precision element and replace at random this n and treat point set of pieces, return to step 3;

Described grouping radix n and require allowable tolerance Δ and actual allowable tolerance Δ ₁according to the actual requirements and actual conditions determine concrete numerical value.

Example:

Treat that grouping products has 46 for this batch, the data that each product comprises needs grouping have 6, require one group of 4 production sharing, and the difference of these 4 all corresponding data of product is not more than 30,, sequence number is 2012001～2012046.

Determine now grouping radix Δ n=4, require allowable tolerance Δ=30, actual allowable tolerance Δ ₁=15

the data file that selection comprises these 46 products, in table 1;

first take out 4 in turn and be numbered 2012001,2012002,2012003,2012004 sophisticated product;

2012001,2012002,2012003,2012004 numbers are judged, suppose that result is greater than Δ ₁, do not meet grouping requirement;

from treat group list, precision element of random taking-up also replaces at random one and repeats previous step, suppose until replace No. 2012001 time with No. 2012006, these combination lattice meet grouping and judge requirement, write group list 2 by 2012002,2012003,2012004, No. 2012006, and these four numberings are deleted from treat group list 1;

again from treat group list, take out 4 precision elements in turn, the judgement of dividing into groups, writes group list 2 by these four numberings if meet point set condition, and these four numberings is deleted from treat group list 1;

repeat above step until residue is numbered 2012001,2,012,042 two and cannot successfully divides into groups;

automatically preserve group list and residue list.

Table 1 is treated group list

A	B	C	D	Error
					2012002	2012003	2012004	2012006	15
2012019	2012028	2012032	2012033	12
					2012007	2012021	2012022	2012023	16
2012026	2012029	2012031	2012037	8
					2012013	2012014	2012015	2012025	14
2012010	2012012	2012016	2012017	10
					2012011	2012020	2012024	2012027	16
2012030	2012034	2012035	2012036	9
					2012038	2012040	2012041	2012044	14
2012039	2012043	2012045	2012046	16
					2012005	2012008	2012009	2012018	14

Table 2 group list

Table 3 remains list

Claims (2)

1. a precision element fast grouping method, is characterized in that, comprises the following steps:

Step 1, the precision element that each is processed carry out serial number, and the test parameter of each precision element is input to same form, form and treat group list;

Step 2, from treat group list, first take out from front to back the precision element of n, n is grouping radix;

Step 3, judge whether the test parameter of this n precision element meets allowable tolerance Δ: if the error of the test parameter of this n precision element between is mutually no more than allowable tolerance Δ, meet allowable tolerance Δ, the test parameter of this n precision element is deposited into group list, and at the test parameter for the treatment of to delete in group list this n precision element, enter step 4; If do not meet allowable tolerance Δ, enter step 5;

Step 4: again take out from front to back n precision element from treat group list, repeating step three is to step 5, until treat the part precision element that all precision elements divide into groups successfully or only residue cannot be divided into groups in group list;

Step 5, from treating that group list chooses at random a precision element and replace at random this n and treat point set of pieces, return to step 3; Until the test parameter of this n precision element meets allowable tolerance Δ, enter step 4.

2. a kind of precision element fast grouping method according to claim 1, is characterized in that, described grouping radix n and allowable tolerance Δ are determined concrete numerical value according to the actual requirements.

Images