JPH11272727A - Wiring route design support method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a wiring route design support method which can design the most suitable arrangement of cables in a short time. SOLUTION: Plural candidate routes of each of cables 1 to 15 are extracted, and characters A to F are given to extracted candidate routes. One candidate route is selected from plural candidate routes A to F, to which discrimination codes are given, for each of cables 1 to 15, and N character strings are generated with discrimination codes of selected candidate routes. At this time, only candidate routes having the shortest route length (shown by oblique lines in the figure) out of candidate routes A to F of cables 1 to 15 are selected to generate a part of N character strings. Generated character strings are optimized by a genetic algorithm, and the combination of candidate routes indicated by an optimized character string is taken as the most suitable cable layout.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a wiring route design support method, and more particularly, to selecting an optimum wiring route for a cable tray network set in advance when designing electrical equipment of a plant such as a chemical industry. The present invention relates to a wiring route design support method that can perform the above.

[0002]

2. Description of the Related Art In general, when designing wiring in a construction plant, a cable tray for storing cables is designed after designing a wiring route for the entire plant in consideration of the position of the equipment and basic conditions. However, due to the design process,
I sometimes design a cable tray network first. In addition, the location of the equipment to be wired may be moved without changing the equipment of the cable tray for the convenience of the construction schedule or periodic inspection. As described above, as a technique for searching for an optimum wiring route for a predetermined cable tray, a network theory, an expert system, and the like are known, and a technique combining both of them is also known (particularly). JP-A-2-71373, JP-B-7-97379, etc.). Recently, a method using a genetic algorithm has been developed.

In the method using the network theory, a cable route is searched for one by one.
The results differ depending on the order in which the search is performed, and a cable cannot be laid on a truly optimal route, and the know-how such as the experience and intuition of the designer is required. In addition, a database that describes the know-how of experienced people, the balance between equipment and cables, and the dependencies (eg, high-cost cables, short-distance cables, and the priority order of cables with a fixed number of cable trays) are described. A system that is constructed and automatically determines the order of route search by an expert system, and a system that microscopically monitors around a cable tray that exceeds the allowable space and performs rearrangement, are used. However, even when these expert systems are used, know-how of the creator is required when creating a database, and there is a drawback that a truly optimum cable route cannot be extracted.

[0004] On the other hand, in a method applying a genetic algorithm, a plurality of route candidates are created in advance for all cables, and characters for identifying the route candidates are added. One route candidate is selected from a plurality of route candidates corresponding to each cable, a large number of character strings of combinations of route candidates are created, and the character strings are focused on the total extension length, the number of turns, and the cable storage status value. Then, the optimal combination of route candidates is found out by optimizing by a genetic algorithm. This optimal combination of route candidates is determined as an optimal wiring route of the plurality of cables.

[0005]

However, in this method, when the plant becomes large, the number of route candidates for each cable becomes large and the number of character strings becomes enormous, so that it is extremely difficult to find the optimum combination of route candidates. Requires a lot of processing time. The present invention has been made in view of such circumstances, and an object of the present invention is to provide a wiring route design support method capable of designing an optimal layout of cables in a short time.

[0006]

SUMMARY OF THE INVENTION In order to achieve the above object, the present invention provides a cable tray network in which cable trays for accommodating cables are stretched in a net-like manner. In a wiring route design support method for designing a plurality of cables connecting between them to be laid in an optimum wiring route, a plurality of route candidates on the cable tray network are obtained for each cable, and the route candidates are identified. And assigning a character string to each of the cables, evaluating the fitness of the plurality of route candidates for each cable, selecting one route candidate from the plurality of route candidates corresponding to each cable, and forming a character string of the combination of the route candidates. When creating a plurality of routes, only the route candidates with the same length as the shortest route length of each cable, or the route with the highest evaluation Include a character string composed only of complements, optimize the plurality of character strings by a genetic algorithm, and use a combination of route candidates indicated by the optimized character string as an optimal wiring route of the plurality of cables. It is characterized by.

According to the present invention, a plurality of route candidates on the cable tray network are obtained for each cable, and characters for identifying the route candidates are added. One route candidate is selected from a plurality of route candidates corresponding to each cable, and a plurality of character strings of combinations of route candidates are created. At this time, a character string composed of only a route candidate having the same length as the shortest route length of each cable or only a route candidate having the highest evaluation is included. The created character string is optimized by a genetic algorithm, and a combination of route candidates indicated by the optimized character string is set as an optimal wiring route of the plurality of cables. By including a string that is the same length as the shortest route length of each cable, or a string consisting of only the route candidates with the highest evaluation, the created character string can be optimized in a short time by a genetic algorithm. As a result, it is possible to efficiently derive an optimal wiring route for a plurality of cables.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of a wiring route design support method according to the present invention will be described below in detail with reference to the accompanying drawings. FIG. 1 is a block diagram showing the configuration of the wiring route design support device. As shown in FIG. 1, the wiring route design support device 10 mainly includes a network data storage device 12, a cable data storage device 14, a CPU 16, a storage device 18, an input device 20, and a display device 22.

The network data storage device 12 stores information relating to a cable tray network for which cable layout is to be designed, and the cable data storage device 14 stores information relating to the specification of the cable to be deployed and the start and end points of the cable. . The CPU 16 performs necessary calculations according to a predetermined processing procedure, and the storage device 18 stores data such as calculation results obtained by the calculation processing device 16.

A keyboard, a mouse, and the like are used as the input device 20. When an operator inputs a command necessary for the arithmetic processing device or changes data stored in the data storage devices 12, 14, 18, used. The data input from the input device 20 and the arithmetic processing device 16
Are displayed on the display of the display device 22, and can be confirmed by printing with the printer 24. An external recording device 26 such as a floppy disk drive is connected to the CPU 16.
Processing results and the like can be stored in a recording medium.

Next, the processing procedure of the wiring design support apparatus will be described with reference to an example in which cables are arranged in the cable tray network shown in FIG. Cable tray 32,
32 are designed in advance, and each cable tray 32 is assigned a series of tray numbers (1, 2, 3,..., 29). A node 34 to which the cable trays 32 are connected;
.. Are assigned a series of node numbers (1, 2, 3,..., 21). The tray number is displayed on each cable tray 32 by surrounding it with a square, and the node number is displayed on each node 34 by shading.

FIG. 3 shows data of each of the cable trays 32. The tray length of each cable tray 32 and the allowable space representing the capacity of cables that can be accommodated in each cable tray 32 are determined in advance due to space restrictions between devices and the like. For example, the tray length of cable tray 1 is 10m
And the allowable space is 5. The data of the cable tray shown in FIG. 3 and the data relating to the cable tray network shown in FIG. 2 are stored in the network data storage device 12 shown in FIG.

FIG. 4 shows data of cables laid in the cable tray network shown in FIG. Each cable is assigned a cable number (1 to 9), and the start node and end node of the cable route on which each cable is laid and the thickness of each cable are given as data. For example, the cable route in which the cable 1 is laid is a node
1 is a start node, node 21 is an end node, and the thickness of the cable is 5. Data relating to these cables is stored in the cable data storage device 14 shown in FIG.

Based on the network data stored in the network data storage device 12 and the cable data stored in the cable data storage device 14, a route candidate is extracted for each cable by a shortest path search method (Dijecstra method). . As a specific example, a method for extracting a route candidate of the cable 1 will be described. In the Dijecstra method,
First, all routes from the start node 1 to the end node 21 (for example, 1-5-10-11-16-25-29) are listed. The total length (route length) of the passing trays of all the listed routes is compared, and the route having the shortest route length is searched. As a result, 1-5-10-11-16-25-29, 1-5-14-1 of all routes
9-20-25-29,…, 4-9-10-11-16-25-29, 4-13-17-18-19-
24-28-29 is the shortest with a route length of 51m. In this case, since there are a plurality of shortest routes, 4-9-10-11-16-25-29, which has the least number of turns three times, among the plurality of routes, is selected as the representative route. In the present embodiment, 4-9-
10-11-16-25-29 is extracted as the representative route.

Next, all the detour routes when the passing trays of the representative route extracted in this way are disconnected one by one are listed. For example, if you disconnect tray 4,
Many detours starting from Tray 1 are obtained, and Tray 9
Disconnection, a number of bypass routes bypassing the tray 9 are obtained. In this way, by cutting each passing tray one by one, a large number of bypass routes can be obtained for the representative route. The thus obtained detour routes and the shortest route are set as route candidates for the cable 1, and a serial number (A, B, C, D, E, F) for identification is assigned to each route candidate.

FIG. 5 shows the route candidates A to F of the cable 1 thus obtained.
, The route length and the number of turns are shown. Similarly, a plurality of route candidates are extracted for the other cables 2 to 9 shown in FIG. 4 and characters (A, B, C,...) Are added. These route candidate data are stored in the storage device 18 shown in FIG.

Next, from the route candidates of the cables 1 to 9, those which are inappropriate as routes are excluded. For example,
Route candidates whose route length exceeds 120% of the shortest route will be excluded. For example, the cable 6 shown in FIG.
The shortest route is 7-16-25-29 with a route length of 25m,
The second shortest route, 2-6-15-24-28-29, has a route length of 34
m is 136% of the shortest route. Therefore, all the route candidates other than the shortest route are excluded, and the shortest route is adopted as the route of the cable 6.

In this way, invalid route candidates for all the cables 1 to 9 are eliminated. As a result, the shortest route is adopted for the cables 6 to 9. That is, cable 6 is 7-16-25-29, cable 7 is 17-18-19-20,
Cable 8 is laid on 4-13-21, and cable 9 is laid on 26-27. Next, the outer diameters of the cables 6 to 9 are subtracted from the allowable occupancy of the cable trays through which the cables 6 to 9 have been routed by the above operation, and the new allowable occupancy of each cable tray shown in FIG. Calculate the product. For example, the new allowable space of the cable tray 7 through which the cable 6 passes is 5 obtained by subtracting the thickness 5 of the cable 6 shown in FIG. 4 from the original allowable space 10 shown in FIG. In addition, cable 6
The new permissible occupancy of the cable tray that does not pass through any of -9 remains the original permissible occupancy.

After calculating a new allowable space, a route search is performed for cables 1 to 5 for which a route has not been determined. One route candidate is selected from a plurality of route candidates for each of the cables 1 to 5, and a plurality of character strings indicating combinations of route candidates as shown in FIG. 7 are created. These character strings are optimized by a genetic algorithm, and a combination of route candidates indicated by the optimized character strings is determined as an optimal wiring route of the cables 1 to 5.

Next, a method of selecting one route candidate from a plurality of route candidates for each of the cables 1 to 5 will be described. Conventionally, when searching for an optimal combination of route candidates using a genetic algorithm, one character string is created by randomly selecting one route candidate from a plurality of route candidates. A character string consisting of only the shortest route candidate is included. When determining the wiring route, the designer first considers the route distribution based on the shortest route at least roughly. Incorporate this idea into creating route combinations.

A method for creating a character string will be described. FIG.
Indicates route candidates A to F corresponding to the respective cables 1 to 15, and hatched portions in the figure indicate route candidates having the shortest route length among route candidates corresponding to the respective cables. When all the route candidates are used, a combination of 10 ²⁷ routes is obtained, and a maximum of 10 ⁷¹⁰⁰⁰ routes is set in an actual plant, which cannot be calculated by a current computer, and the following measures are required.

As shown in FIG. 9, when creating N character strings 1 to N, the first character string 1 is obtained by selecting only the route candidate A from the shortest route candidates of each cable. Created. The N 1-10% character strings from the character string 2 are created by successively selecting route candidates of the same number from short-length route candidates so that the characters are continuous. For example, in the case of the character string 2, the route candidate B is selected for the cables 1 to 4, the route candidate A is selected for the cables 5 to 7, the route candidate C is selected for the cables 8 and 9, and the route candidate is selected for the cables 10 to 12. A is selected,
The route candidate C is selected for the cables 13 and 14, and the route candidate B is selected for the cable 15. In the route candidates shown in FIG. 8, there are 96 (2 × 1 × 3 × 2 × 4 × 2) combinations of such route candidates.

The N 10 to 50% character strings from the character string 1 are randomly selected from the shortest route candidates of each cable and are created. Such a combination is 2 ⁸ × 3 ² × 4 ² = 3 in the route candidate shown in FIG.
6864 ≒ 4 × 10 ⁴ types exist, and 10
There are ¹⁵⁵⁰ ways. The N 50% character strings from the character string m are created by the method described below. For example,
When selecting one route candidate from the route candidates A to F of the cable 1, first, the fitness for each of the route candidates A to F is evaluated based on the route lengths of the route candidates A to F shown in FIG. A judgment reference value is calculated. The criterion value is a value obtained by subtracting each route length from the longest route length 71 and adding 1. For example, the criterion value of the route candidate A is 71-51.
+ 1 = 21, and the criterion value becomes larger as the route length becomes shorter. Next, the selection probabilities of the route candidates A to F are calculated based on the judgment reference values. The selection probability is multiplied by 100 by dividing each criterion value by a total of 86 criterion values. For example, the selection probability of the route candidate A is 21 ÷
86 × 100 = 24%, and the total of the selection probabilities of the route candidates A to F becomes 100%. As described above, since the selection probability of each route candidate is proportional to the criterion value, the shorter the route length, the higher the selection probability. Also,
By adding 1 when calculating the criterion value,
The selection probability of the route candidate F having the longest route length is prevented from becoming 0%.

After calculating the selection probabilities of the route candidates A to F, a roulette shown in FIG. 10 is created. This roulette is divided into six sectors corresponding to the route candidates A to F, and the central angle of each sector is determined by the route candidates A to F shown in FIG.
It is set to be proportional to the selection probability of F. A rotatable needle is rotatably provided at the center of the roulette, and the rotatable needle is rotated to select a route candidate (a route candidate C in FIG. 10) corresponding to the position where the rotatable needle stops.

Similarly, for the cables 2 to 5, the selection probabilities of the route candidates A, B, C,... Are calculated, a roulette is created based on the selection probabilities, and one route candidate is selected by the roulette. When one route candidate is selected for each of the cables 1 to 5, the selected route candidates for each of the cables 1 to 5 are arranged and a character string (FIG.
In the above, CADAE) is created. This character string indicates a combination of the selected route candidates of the cables 1 to 5, and the selected route candidate of each of the cables 1 to 5 (the route candidate C of the cable 1 in the figure) is shown in FIGS.
The route data is associated with the cable data and the route data shown in (1), and the passing tray, route length, number of turns, and outer diameter of the route candidate can be referred to.

A route candidate is repeatedly selected one by one for each cable, and character strings from character string m to character string N shown in FIG. 9 are created. When N character strings are created in the above-described procedure, a genetic algorithm (Genetic A
lgorithms) to obtain the optimal cable layout by optimizing the string.

When a genetic algorithm is applied to a character string, the character string is regarded as a chromosome of an organism, the characters (A, B, C,...) Constituting the character string are regarded as genes, and cables 1 to 5 are connected. Considered a locus. A character string with a high evaluation is selected from the character strings (selection), the selected character string is divided, and the other divided character strings are joined (crossover), or a part of the character string is replaced with another character. A new character string is created while performing an operation of replacing (mutating) with a character in the character string. The newly created character string and the original character string are evaluated, and a character string with a high evaluation is selected from the character strings. By repeating such an operation, an optimum character string is found recursively.

When a mutation process is performed, a roulette wheel as shown in FIG. 10 is used when selecting a route candidate to be replaced so that a route candidate having a higher evaluation has a higher probability of being selected. In the present embodiment, since a character string composed of a serial number corresponding to each route candidate of each cable is created, a genetic algorithm method can be applied, and the optimal combination of route candidates is derived inductively. I will

Next, an objective function for evaluating a chromosome is determined. As the evaluation items, the material cost of the cable and the wire drawing cost in the plant construction are considered. Regarding material costs, it is desirable that the cable route is short,
Equation (1) is given.

[0030]

[Number 1] _{N (l) = Σ i {} l i (j) / l imax} ... (1) l i (j) represents the root length of the route candidates for configuring the string, l _imax is each cable Indicates the longest route length among route candidates. Next, with regard to the wire extension cost, considering the burden of wiring bending work and the movement load of the wiring drum, it is desirable that the number of route bends and the number of trays used are smaller, so that Equations (2) and (3) are Given.

[0031]

[Number 2] _{N (t) = Σ i {} t i (j) / t imax} ... (2)

N (p) = Σ _i ｛p _i (j) / p _imax ｔ (3) t _i (j) represents the number of turns of the route candidate forming the character string, and t _imax is the cable i's Indicates the maximum number of turns among the route candidates. Further, p _i (j) represents the number of use trays of the route candidates constituting the character string, and p _imax represents the maximum number of use trays among the route candidates of the cable i.

Further, since it is a condition that the cables are stored without overflowing from the tray, and it is preferable to concentrate the wiring on one cable tray, the expression (4) is given.

[0033]

N _d (σ) = Σ _k (Q _k −σ _k ) / p _all (4) Q _k represents the allowable space of the tray k, and σ _k is outside the cable i passing through the tray k. Represents the sum of the diameters, p _all represents the total allowed space of the tray. _{However, σ k = Σ i d i} : If the cable i passes tray k sigma _k = 0: thus if the cable i does not pass the tray k, cables i from the allowable occupying Q _k of each tray k
Is calculated by subtracting the sum of the outer diameters of the trays.

If the tray margin value becomes negative, it means that the cable is arranged beyond the allowable space of the tray and overflows from the tray. If the tray margin becomes negative, equation (5) is given.

[0035]

N _o (σ) = Σ _k (| Q _k −σ _k |) / p _all (5) A tray overflow value is obtained by multiplying the absolute value of the tray margin value by a large value (for example, 10). According to the tray overflow value, the cables are excessively concentrated on a specific tray, and it is determined that the cable arrangement is extremely skewed. Further, a cable tray through which no cable passes is evaluated as a useless cable tray, and the tray margin value is set to 0.

A tray margin value (a tray overflow value in the case of a minus value) is calculated for all the cable trays constituting the cable tray network, and the calculated value is regarded as an allowable space error, and the cable arrangement state in the cable tray network of the entire plant is calculated. Was used as an index to indicate However, those with a tray margin value of 0 are excluded from multiplication targets. If the allowable occupation error thus obtained is small, it means that cables are laid without overflowing in the cable trays constituting the cable tray network, and that the wiring is concentrated on the used cable trays. A large error indicates that the cable has overflowed from the cable tray and that a biased cable arrangement has been made in which the cable is excessively concentrated on a particular cable tray.

FIGS. 11 to 13 show a part of a calculation example of the allowable occupation error. First, paying attention to the tray 10 and the tray 19, a total of five cables 1, 2, 3, 4, and 7 pass through these two trays. In this, it is determined that the cable 7 passes through the tray 19 as described above, and the allowable space of the tray 19 is also reduced as shown in FIG. 6 because the outer diameter of the cable 7 has already been reduced. Exclude from calculation.

As shown in FIG. 11, when the cables 1, 2, 3 pass through the tray 10 and the cable 4 passes through the tray 19, the tray margin value of the tray 10 is 40− (5 + 10).
+20) = 5, and the tray margin value of the tray 19 is 10−
5 = 5. Therefore, the allowable space error is 5 × 5 = 25 as the product of the tray margin values of the two cable trays.

Further, as another case, as shown in FIG. 12, when the cables 1 and 2 pass through the tray 10 and the cables 3 and 4 pass through the tray 19, the tray margin value of the tray 10 is 40. − (5 + 10) = 25, and the tray margin value of the tray 19 is 10− (20 + 5) = − 15, which is a minus value.
5 | × 10 = 150. Therefore, the allowable space error is 25 × 150 = 3750.

Further, as another case, as shown in FIG. 13, cables 1, 3, and 4
When cable 2 passes through 19, cable tray 10
Is 40− (5 + 20 + 5) = 10, and the tray allowance of the cable tray 19 is 10−10 = 0.
It is. If the tray margin value is zero, it is excluded from the multiplication target according to the above rule, so that the allowable space error is 10
Becomes

In FIGS. 11 to 13, only the cable tray 10 and the cable tray 19 have been described. However, the allowable space error is obtained for all the cable trays on the route.
The objective function for evaluating a character string is given by Expression (6) obtained by adding Expressions (1) to (5).

[0042]

F (l, t, p, σ _d , σ _o ) = αN (l) + βN (t) + γN (p) + δN _d (σ) + ηN _o (σ) (6) α, β, γ , Δ, η are parameters, and the chromosome (character string) having a smaller value of the objective function is evaluated as being better.

A character string is evaluated by the objective function, a plurality of character strings evaluated as good are extracted (selection and selection), and the character string is divided in the middle and replaced with another character string (crossover). Rewriting characters (genes) in a character string with other characters (mutation), creating a new character string, and repeatedly reevaluating the new character string By creating, a character string with the smallest evaluation value is derived recursively.

As for the GA operator used for the processing of the genetic algorithm, a general Selection
(Selection, selection), Crossover, Mutation were used. That is, in the selection (selection, selection) processing, one minimum value of the generation is stored, and the remaining one adopts a roulette rule. In the crossover processing, the number of genes is fixed and no discrimination is caused by loci, so that normal single-point crossover is performed. Also, Mutation
In the (mutation) process, a gene other than the gene selected in that generation is stochastically generated.

As a result of operating the genetic algorithm using this method, one character string having the minimum evaluation value can be obtained, and the character string is replaced with a cable route to design an optimal cable layout. be able to. As a result, the cable tray is designed (or installed) before the cable route, such as when the cable tray designer is different from the cable route designer, or when the arrangement of equipment is changed during periodic inspection work or the like. In this case, the optimal and theoretical cable layout can be automatically designed in a short time.

Further, when a plurality of character strings are created, by including a character string in which only a route candidate having the shortest route length of each cable is included, an optimum cable layout can be designed in a short time. . When N character strings are randomly selected from all the route candidates, calculation is required up to 10,000 generations. However, if all character strings other than the character string 1 shown in FIG. 9 are randomly selected from all the route candidates, , Up to 6000 generations. When a character string other than a character string having consecutive numbers as shown in character string 1 and character string 2 is randomly selected from all route candidates, 40
It is necessary to calculate up to 00 generations, character string 1 and character string 2
Calculated up to 1500 generations when a random number is selected from all the route candidates except for a character string randomly selected from the shortest route candidate as shown in the character string l and the shortest route candidate as shown in the character string l It is possible to derive the optimal route combination in a short time.

FIG. 15 shows a power cable of about 40 for a real plant.
It is a graph which shows the change of the evaluation value according to the number of generations when the route candidate is extracted at random when laying zero, and when it is extracted by the criterion value. As a result, the evaluation value when randomly extracted was about 14.5 in the 0 generation, gradually decreased with the number of generations, and was 11.9 in the 10,000 generation. The evaluation value when extracting from the judgment reference value was about 13.5 in the 0 generation and decreased to 11.9 in the 900 generation.
Thus, when extracting from the judgment criterion value, the evaluation value is lower at 0 generation than at random extraction, and when randomly extracted, the evaluation value at 10,000 generation is 900
You can see that this is achieved in generations.

In the above embodiment, when a route candidate is selected and a character string is created, a character string composed of only route candidates having the same length as the shortest route length is included.
The present invention is not limited to this, and may include a character string composed of only the route candidates having the highest fitness evaluation. Further, when creating a character string other than a character string composed of only a route candidate having the same length as the shortest route length or the highest fitness evaluation, without using the roulette shown in FIG. May be randomly selected from the route candidates.

In the above embodiment, the criterion value is calculated from the route length of each route candidate. However, the present invention is not limited to this. The criterion value may be calculated from the number of turns and the cost. In the above embodiment, the evaluation of the character string is based on the total cable extension length,
Although the evaluation is performed based on the total number of cable bending times and the allowable space error, other evaluation values such as a material cost value obtained by multiplying the cable length by the outer diameter may be considered.

In the above-described embodiment, when creating cable route candidates, all cases have been tried. However, the number of candidates may be reduced by creating cable route candidates based on constraints. In this case, the calculation speed is improved, but there is a demerit that it is difficult to verify whether or not the candidate includes a route that can be the optimum route. In the above embodiment, as shown in FIG. 1, a case where cables are arranged by using a network of cable trays arranged on a two-dimensional plane has been described as an example. The present invention is also applicable to a network of cable trays formed in a space.

[0051]

As described above, according to the wiring route design support method according to the present invention, when a plurality of character strings are created, the character string composed of only the best route candidate is evaluated for each cable. Is included, the created character string can be optimized in a short time by a genetic algorithm, and optimal wiring routes of a plurality of cables can be efficiently derived.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a wiring route design support device.

FIG. 2 is a diagram showing a cable tray network for laying cables.

FIG. 3 is a table showing data of a cable tray.

FIG. 4 is a table showing cable data.

FIG. 5 is a chart showing route candidates of the cable 1;

FIG. 6 is a chart showing a new tray allowable space without a cable whose arrangement is fixed;

FIG. 7 is an explanatory diagram showing a character string of a route candidate.

FIG. 8 is a diagram showing a route candidate having the shortest route length among route candidates of each cable;

FIG. 9 is an explanatory diagram showing the created N character strings.

FIG. 10 is a diagram showing a roulette for extracting a route candidate of the cable 1;

FIG. 11 is an explanatory diagram showing a calculation example of an allowable occupation error.

FIG. 12 is an explanatory diagram showing a calculation example of an allowable space error.

FIG. 13 is an explanatory diagram showing an example of calculating an allowable occupation error.

FIG. 14 is a graph showing a change in an evaluation value according to the number of generations when a route candidate is randomly extracted and when a route candidate is extracted based on a criterion value.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 ... Wiring route design support device 12 ... Network data storage device 14 ... Cable data storage device 16 ... CPU 18 ... Storage device 20 ... Input device 22 ... Display device 32 ... Cable tray 34 ... Node

Claims (3)

[Claims] 1. A plurality of cables connecting desired connection points of the cable tray network are laid in an optimum wiring route with respect to a cable tray network in which cable trays accommodating cables are stretched in a net shape. In the wiring route design support method, a plurality of route candidates on the cable tray network are obtained for each cable, characters for identifying the route candidates are added, and a plurality of route candidates are assigned to each cable. When evaluating the fitness of the route candidates and selecting one route candidate from a plurality of route candidates corresponding to each cable and creating a plurality of character strings of combinations of route candidates, the shortest route length of each cable and A character string composed of only the route candidates of the same length or only the route candidate with the highest evaluation is included, A wiring path design support method, wherein a character string is optimized by a genetic algorithm, and a combination of route candidates indicated by the optimized character string is used as an optimum wiring path of the plurality of cables. 2. When creating a character string other than a character string consisting of only a route candidate having the same length as the shortest route length of each cable or a highest-rated route candidate, a route candidate with a higher evaluation is selected. 2. The wiring route design support method according to claim 1, wherein a route candidate is selected such that the probability of the route is high. 3. The genetic algorithm includes a mutation process, wherein a certain route candidate in the character string is
When replacing and mutating a route candidate selected from a plurality of route candidates to which the route candidate belongs, the route candidate is selected and replaced so that a route candidate with a higher evaluation has a higher probability of being selected. 3. The wiring route design support method according to claim 1, wherein: