TWI646492B - Method for optimizing artificial reef deployment in a marine farm - Google Patents

Abstract

The invention provides an optimization method for arranging artificial reefs in marine pastures, comprising: step one, parameter and decision variable setting steps; step two, setting a restriction step of the reef design; step three, optimal purchase quantity ( The steps taken; and the fourth step, the reef configuration configuration steps. Accordingly, the present invention is based on a number of factors (eg, setting a reef pile (or patch) of different functional types, and a unit price of a reef reef ( ), the bottom area of a single pile of reef reefs ( ),total budget And the total area of the reef plan area ...etc.), to provide an artificial fish reef layout optimization method in the marine pasture, in order to increase the unit's investment in the effect of the fish, that is, the investment of each dollar can bring the greatest effect of the fish, to the real To achieve the purpose of improving the effect of poly fish.

Description

Optimizing methods for arranging artificial reefs in marine pastures

The invention relates to the artificial fish reef arrangement method, in particular to an optimized method for arranging artificial reefs in marine pastures to obtain the maximum poly-fish effect.

Artificial reefs are an artificial seabed deposit. It is the natural or artificial structure that is put into the sea. By these objects, it provides a place for marine organisms to inhabit, avoid enemies and bait. It also protects and proliferates fishery resources because of the ability of artificial reefs to gather fish.

At present, the design of artificial reefs does not have certain construction specifications to follow, only the size of the reef piles or the spacing of the reefs of the reefs; therefore, for the ecological engineers who design the artificial reef ecosystem, when the reefs are placed After the site and the reefs to be selected, the next problem is to arrange these artificial reefs in the selected site. Among them, the Artificial Reef Communities (ARCs), while the reefs contain several piles of artificial reefs, and each artificial reef contains artificial reefs (Unit/Modules) of artificial reefs.

According to past experience, most of the engineers' layout design relies on the subjective judgment of engineers. However, the arrangement is to consider the cost and produce the best poly fish effect but it is rarely discussed. Secondly, although the area of the reef layout should be as wide as possible. However, the use of fish reefs to fill an entire area not only consumes a large number of fish reefs, but also has a large increase in the layout area under limited reef resources. Therefore, the ideal way is to pile up and pile up to form a wide range. The fish reef area. Based on this, how to factor in (for example, setting the reef pile (or patch) of different functional types, and the unit price of the reef reef ( ), the bottom area of a single pile of reef reefs ( ),total budget And the total area of the reef plan area (etc.), to provide an artificial fish reef layout optimization method in the marine pasture, in order to increase the unit's investment in the effect of the fish, that is, the investment of each dollar can bring the biggest effect of the fish is the case The motivation of the invention.

It is an object of the present invention to provide an optimized method for arranging artificial reefs in marine pastures.

In order to achieve the foregoing objective, an optimization method for arranging artificial reefs in marine pastures according to the present invention includes:

Step 1. Parameter and decision variable setting steps: This parameter includes the unit price of the reef reef and the reef reef piled with different functional types ( ), the bottom area of a single pile of reef reefs ( ),total budget And the total area of the reef plan area The decision variable is the number of different reef reefs ( );

Step 2: Set the restrictive steps for the design of the reef: the total construction cost shall not exceed the budget, the total area of the reef shall not exceed the total area of the project area, and the location of the reef shall not exceed the range set by the reef plan area;

Step 3: Optimal purchase quantity of reef reefs of different functional types ( The steps taken: using a landscape diversity index as the target of the optimization model to determine the optimal number of purchases for the reef piles of the different functional types ( );as well as

Step 4: Reef configuration configuration steps: The optimal purchase quantity of the reef reef according to the different functional types ( The deployment is carried out in the reef plan area to obtain candidate layout patterns of a plurality of design reef groups, each of which adopts a landscape index and is computerized to obtain an optimal layout.

Preferably, the mathematical cost in which the total construction cost in the second step does not exceed the budget is expressed as: The mathematical formula of the total bottom area of the reef in step 2 shall not be greater than the total area of the project area as: .

Preferably, in the third step, the landscape diversity index is a Shannon-Weaver diversity index, and the Shannon-Weaver diversity index is SHDI, and this step is The objective function of the optimization mode, which is optimized, means that the third step can make the SHDI value have a maximum value. Mathematical formula Express For the decision variable; the Shannon-Weaver (SHDI) diversity index is: ,among them Is a reef pile of different functional types The probability of appearing in the landscape; A type of reef reef for different functional types.

Preferably, in the third step, the landscape diversity index is a Simpson diversity index, and the Simpson diversity index is SIDI, and is the target function of the optimization mode of the step, and the best The system means that step three can make the SIDI value have a maximum value. Mathematical formula Express For decision variables; the formula for the Simpson diversity index is: ,among them Is a reef pile of different functional types The probability of appearing in the landscape; A type of reef reef for different functional types.

Preferably, in the third step, the landscape diversity index is a modified Simpson diversity index, and the modified Simpson diversity index is MSIDI, and is the target function of the step optimization mode. The optimization means that step 3 can make the MSIDI value have a maximum value. Mathematical formula Express For decision variables; the formula for the modified Simpson diversity index is: ,among them Is a reef pile of different functional types The probability of appearing in the landscape; A type of reef reef for different functional types.

Preferably, in the step 4, the landscape index adopts a landscape relative concentration index (contagion index), so that the landscape relative concentration index (contagion index) is RC, and is the target function of the step optimization mode, Optimization means that step four can have a maximum value for the RC value. Mathematical formula Express For decision-making variables, it refers to the position of the reefs of each reef group; the formula for the relative concentration index of the landscape is: , Different types of functional reef reef stack type k i and adjacent type probability. Which should The method of seeking is ,among them Is a randomly selected cells in the grid (lattice cell), belonging to different functional types of reef fish reef probability i stack can usually be different types of functional stack reef fish reef i proportion to estimated area of the entire landscape; and Given in the case of different functional types of reef fish reef stack i, the different functional types of reef fish reef stack adjacent thereto conditions the probability of k, i.e. In the formula Is the common boundary length in the landscape, different reef types, type i and type k adjacent; It is the total number of sides of i- cells of fish reef piles of different functional types or estimated by the total length of the boundaries between different patches in the landscape.

Preferably, in the fourth step, the landscape indicator adopts an average patch block fractal dimension, so that the average patch block fractal dimension is MPFD, and is the target function of the step optimization mode, and the optimization refers to the step. Fourth, the MPFD value can be maximized. Mathematical formula Express For the decision variable, it refers to the position of the reef of each reef group; the formula of the average patch fractal dimension is: , For the reef piles of different functional types, the perimeter of type i and type k , For the reef piles of different functional types, the area of type i and type k , For the reef piles of different functional types, the number of types i and type k ; .

Preferably, in the fourth step, the landscape index adopts an area weighted average block fractal dimension, so that the area weighted average block fractal index is AWMPFD, and is the target function of the step optimization mode, and the best The system means that step 4 can make the AWMPFD value have a maximum value. Can be mathematically Express For decision-making variables, it refers to the location of the reefs of each reef group; the formula for the area-weighted average patch fractal dimension is: In the formula For the reef piles of different functional types, the ratio of the area of type i and type k to the total area of the landscape; .

Preferably, in the fourth step, the landscape index adopts a perimeter area fractal dimension, so that the perimeter area fractal dimension is PAFRAC, and is the target function of the step optimization mode, and the optimization refers to the step Four can make the PAFRAC value have a maximum value. Can be mathematically Express For decision-making variables, it refers to the location of the reefs of each reef group; the formula for the fractal dimension of the perimeter area is: , L is a reef pile of different functional types, the perimeter of type i and type k , For fish reef reefs of different functional types, the area of type i and type k , h is a constant; .

Accordingly, the present invention is based on a number of factors (eg, setting the unit price of reef piles and reef reefs of different functional types ( ), the bottom area of a single pile of reef reefs ( ),total budget And the total area of the reef plan area ...etc.), to provide an artificial fish reef layout optimization method in the marine pasture, in order to increase the unit's investment in the effect of the fish, that is, the investment of each dollar can bring the greatest effect of the fish, to the real To achieve the purpose of improving the effect of poly fish.

The present invention has been described in connection with the preferred embodiments of the present invention in accordance with the accompanying drawings.

It should be noted that although the area of the fish reef should be wider, the better. However, the use of fish reefs to fill an entire area not only consumes a large number of fish reefs, but also has a large increase in the layout area under limited reef resources. Therefore, the ideal way is to pile up and pile up to form a wide range. The fish reef area. Therefore, this embodiment is based on a fish reef reef. The design method can also be inferred for the layout design of the fish reef group. In the present embodiment, a pile is formed by several (ten) reef modules, and the distance between each pile should be 50 to 100 meters. If you want to promote the arrangement of the group, there are 10~20 piles of reefs per unit of reef, and the distance between the two groups is 300~500 meters. Its state is shown in Figure 1, which shows a schematic diagram of the composition of artificial reef habitats. Figure 1 shows an artificial reef habitat consisting of several artificial reef communities (sets); in addition, any reef group consists of several artificial reef blocks.

As shown in FIG. 2, an optimization method for arranging an artificial reef in an ocean pasture according to an embodiment of the present invention is mainly performed by Step 1: Parameter and Decision Variable Setting Step 11 and Step 2: Setting the Reef Design Restricted Step 12, Step 3: Optimal Purchase Quantity ( ) Step 13 taken, and Step 4: Reef configuration configuration step 14 is completed, where:

Step 1, parameter and decision variable setting step 11: Referring to FIG. 3, a schematic example of parameter setting is displayed. This parameter includes the unit price of a reef reef (or patch) and a reef reef set for different functional types ( ), the bottom area of a single pile of reef reefs ( ),total budget And the total area of the reef plan area The decision variable is the number of different reef reefs ( Where k is the reef pile number of different functional types, and in this embodiment, the reef pile of the different functional types is exemplified by four examples, and the reef pile refers to the appearance or nature of the surrounding area. Different spatial elements and spatial units with certain internal homogeneity. In this embodiment, it may be a giant artificial reef reef pile, a larva reef pile, or an induced reef pile. As for the unremoved seabed relative to other forms of reef, its position in space, besides being regarded as a matrix, can also be regarded as a “reef-free reef” reef pile, so in this implementation In the case, it is considered an "air reef" reef type. The base refers to the most widely distributed and most continuous background structure in the landscape. In this embodiment, it is a seabed landscape that is not reefed.

Step 2: Set the restrictive step 12 of the reef design: including the total construction cost shall not exceed the budget, the total area of the reef shall not exceed the total area of the project area, and the location of the reef shall not exceed the range set by the reef plan area; In this embodiment, the total cost of the construction shall not exceed the mathematical formula of the budget as: ; Referring to Figure 3, the m = 4. The mathematical formula of the total area of the reef must not exceed the total area of the project area as: ; Referring to Figure 3, the m = 4.

Step 3, the optimal number of purchases ( Step 13: Refer to Figure 3 and Figure 4, using a target of the Landscape Diversity Index to find the optimal number of purchases for the reef piles of the different functional types ( ). In this embodiment, the landscape diversity index is the Shannon-Weaver diversity index, and the Shannon-Weaver diversity index is SHDI, and the step optimization mode is adopted. The objective function, optimization means that step 3 can make the SHDI value have a maximum value. Formula can Express For the decision variable; the Shannon-Weaver (SHDI) diversity index is: ,among them Is a reef pile of different functional types The probability of appearing in the landscape; For the reef pile types of different functional types; see Figure 3, the m = 4; ln is the natural logarithm of e. Referring to Figure 4, this embodiment is based on three different reef designs, and the Shannon-Weaver diversity index is used as an illustration. Assuming a total budget of 30 units, in conjunction with the parameters of Figure 3, through the second step, at least three candidate design cases with design case numbers 1-1, 2-1 and 3-1 in Figure 4 can be obtained. Through the calculation of Step 3, the recommended recommended value of each reef purchase quantity can be obtained. The number of giant artificial reefs is 6 piles, 10 piles of larvae and 10 piles of induced reefs, that is, design case No. 1-1 is the preferred proposal for this design.

Step 4: Reef configuration configuration step 14: Refer to Figure 5, according to the optimal purchase quantity of the reef pile of the different function types ( The deployment is carried out in the reef plan area to obtain candidate layout patterns of a plurality of design reef groups, each of which adopts a landscape index and is computerized to obtain an optimal layout. In this embodiment, the recommended values of the reef purchase quantities obtained through the third step are arranged to obtain the three candidate reef designs as shown in FIG. 5, that is, the design case numbers 1-1, 2-1, and 3 in FIG. -1, and each has a corresponding layout diagram; at this time, the FRAGSTATS software can adopt the landscape relative aggregation index (contagion index, RC or CONTAG) and the average patch fractal index (MPFD) according to the selected landscape index. , area weighted average block fractal dimension index (abbreviated as AWMPFD), or perimeter area fractal dimension index (abbreviated as PAFRAC) to obtain the objective function value corresponding to each reef type in step four, which can be found from the calculation result, design Case No. 1-1 can be used as the recommended preferred arrangement when using MPFD as the target type; as for the design case number 1-2, it can be used as the recommended preferred arrangement when RC is the target; Design Case No. 1-3 It can be used as a suggested preferred arrangement when targeting AWMPFD or PAFRAC.

Accordingly, the present invention is based on a number of factors (eg, setting the unit price of reef piles and reef reefs of different functional types ( ), the bottom area of a single pile of reef reefs ( ),total budget And the total area of the reef plan area Under the consideration of ..., etc., it is possible to provide an artificial fish reef in the marine pasture optimization method to improve the effect of the unit to invest in the fish, that is, the investment of each dollar can bring the biggest fish effect In order to truly achieve the purpose of improving the effect of poly fish.

It is worth mentioning that in other embodiments, the landscape diversity index in step 3 may be the Simpson diversity index, and the Simpson diversity index is SIDI, and the step is optimized. The objective function of the mode, the optimization means that the third step can make the SIDI value have a maximum value, and the mathematical formula can be Express For decision variables; the formula for the Simpson diversity index is: ,among them Is a reef pile of different functional types The probability of appearing in the landscape; A type of reef reef for different functional types. Referring to Figure 4, when the Simpson diversity index is used in Step 3, after the steps 2 and 3, the recommended number of recommended reef purchases, the number of giant artificial reefs, can be obtained. 8 piles, 8 piles of larvae and 6 piles of induced reefs, ie design case number 3-1 is the preferred design proposal.

In other embodiments, the landscape diversity index in step 3 is a modified Simpson diversity index, and the modified Simpson diversity index is MSIDI, and is the target of the optimization mode of the step. The function, which is optimized, means that step 3 can make the MSIDI value have a maximum value. Mathematical formula Express For decision variables; the formula for the modified Simpson diversity index is: ,among them Is a reef pile of different functional types The probability of appearing in the landscape; A type of reef reef for different functional types. Referring to Figure 4, when the landscape diversity index is in the modified Simpson diversity index in step three, after the steps 2 and 3, the recommended number of recommended reef purchases can be obtained. There are 8 piles of reefs, 8 piles of larvae and 6 piles of induced reefs. Design case No. 3-1 is the preferred design proposal.

It is worth noting that in step 4, the landscape index adopts the landscape relative aggregation index (contagion index, RC for short), the average patch block fractal index (MPFD), and the area weighted average block index (AWMPFD). And the perimeter area fractal index (referred to as PAFRAC) are described as follows:

The landscape index adopts a landscape relative concentration index (contagion index), so that the relative concentration index of the landscape is RC, and is the target function of the optimization mode of the step, and the optimization means that step 4 can be This RC value is made to have a maximum value. Mathematical formula Express For decision-making variables, it refers to the position of the reefs of each reef group; the formula for the relative concentration index of the landscape is: , Different types of functional reef reef stack type k i and adjacent type probability. Which should The method of seeking is ,among them Is a randomly selected cells in the grid (lattice cell), belonging to different functional types of reef fish reef probability i stack can usually be different types of functional stack reef fish reef i proportion to estimated area of the entire landscape; and Given in the case of different functional types of reef fish reef stack i, the different functional types of reef fish reef stack adjacent thereto conditions the probability of k, i.e. In the formula Is the common boundary length of the different reef types in the landscape, type i and type k (or estimated by the length of the common boundary); It is the total number of sides of i- cells of fish reef piles of different functional types or estimated by the total length of the boundaries between different patches in the landscape.

The landscape index adopts an average patch block fractal dimension, so that the average patch block fractal dimension is MPFD, and is the target function of the step optimization mode, and the optimization means that step 4 can make the MPFD value have the largest value. . Mathematical formula Express For decision-making variables, it refers to the location of the reefs of each reef group; the formula for the fractal index of the reef piles of the average different functional types is: , For the reef piles of different functional types, the perimeter of type i and type k , For the reef piles of different functional types, the area of type i and type k , For the reef piles of different functional types, the number of types i and type k ; .

The landscape index adopts an area-weighted average patch block fractal dimension, so that the area weighted average patch block fractal dimension is AWMPFD, and is the target function of the step optimization mode, and the optimization means that step 4 can make the The AWMPFD value is the largest. Can be mathematically Express For decision-making variables, it refers to the location of the reefs of each reef group; the formula for the fractal dimension of the reef piles of the weighted average different functional types is: In the formula For the reef piles of different functional types, the ratio of the area of type i and type k to the total area of the landscape; .

The landscape index adopts the perimeter area fractal index, so that the perimeter area fractal dimension is PAFRAC, and is the target function of the step optimization mode, and the optimization means that step 4 can make the PAFRAC value the largest. . Can be mathematically Express For decision-making variables, it refers to the location of the reefs of each reef group; the formula for the fractal dimension of the perimeter area is: , L is a reef pile of different functional types, the perimeter of type i and type k , For fish reef reefs of different functional types, the area of type i and type k , h is a constant; .

Finally, all of the aforementioned landscape indicators can be obtained by FRAGSTATS software. The instructions are compared to the table below.

<TABLE border="1" borderColor="#000000" width="85%"><TBODY><tr><td><b>landscape index name</b></td><td><b>FRAGSTATS </b><b>Instruction Name</b></td></tr><tr><td> Shannon Diversity Index</td><td> SHDI </td></tr><tr> <td> Simpson Diversity Index</td><td> SIDI </td></tr><tr><td> Modified Simpson Diversity Index</td><td> MSIDI </td></tr ><tr><td> Relative Aggregation Index</td><td> RC(CONTAG) </td></tr><tr><td> Average Patch Fractal Dimension Index</td><td> MPFD (FRAC_MN) </td></tr><tr><td> Area Weighted Average Patch Fractal Dimension Index</td><td> AWMPFD (FRAC_AM) </td></tr><tr><td> Perimeter area fractal dimension <i>perimeter-area fractal dimension</i></td><td> PAFRAC </td></tr></TBODY></TABLE>

In the above, the above embodiments and the drawings are only the preferred embodiments of the present invention, and the scope of the present invention is not limited thereto, and the equivalent changes and modifications according to the scope of the present invention should be The scope of the invention is covered.

11‧‧‧Parameter and decision variable setting steps

12‧‧‧Setting restricted steps for reef design

13‧‧‧Best purchase quantity ( ) steps taken

14‧‧‧ Reef configuration configuration steps

‧‧‧Unit price of reef reef

‧‧‧Bottom area of single pile reef reefs

‧‧‧total budget

‧‧‧ Total area of the reef plan area

‧‧‧Decision variables for the number of different reef reefs

SHDI‧‧ Shannon-Weaver Diversity Index

‧‧‧Reef piles of different functional types Probability in the landscape

‧‧‧Fish reef pile types of different functional types

RC (or CONTAG) ‧ ‧ landscape relative aggregation index

MPFD‧‧‧Average block fractal index

AWMPFD‧‧‧ area weighted average block fractal index

PAFRAC‧‧‧ perimeter area fractal index

SIDI‧‧ Simpson Diversity Index

MSIDI‧‧‧Modified Simpson Diversity Index

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic illustration of an embodiment of the invention showing the composition of an artificial reef habitat. Figure 2 is a step diagram of an embodiment of the present invention. FIG. 3 is a diagram showing a specific example of parameters of an embodiment of the present invention. 4 is a schematic diagram of a design example of the third step of the embodiment of the present invention. FIG. 5 is a schematic diagram of a design example of the fourth step of the embodiment of the present invention.

Claims (9)

An optimization method for arranging artificial reefs in marine pastures, comprising: step one, parameter and decision variable setting steps: the parameters include setting the unit price ( C _k ) of the reef reef pile and the reef reef pile of different functional types, and the single pile The bottom area ( A _k ) of the reef reef stack, the total budget ( B ), and the total area of the reef plan area ( A ). The decision variable is the number of different reef piles ( n _k ); Step 2 Set the restrictive steps for the design of the reef: including the total construction cost should not exceed the budget, the total area of the reef should not exceed the total area of the project area, and the location of the reef should not exceed the range set by the reef plan area; Step 3, different functions The optimal number of purchases of the type of reef reef ( n _k ): a landscape diversity index is used as the target of the optimization model to determine the optimal purchase quantity of the reef pile of the different functional types ( n _k ); and step four, the reef configuration configuration steps: according to the optimal purchase quantity ( n _k ) of the reef pile of the different functional types, deployed in the reef plan area to obtain a plurality of design reefs Group waiting Selecting the layout type, each candidate arrangement pattern adopts a landscape index and is calculated by computer to obtain an optimal layout type; wherein the total construction cost in the second step cannot exceed the budget mathematical expression: B ; The mathematical formula of the total bottom area of the reef in step 2 shall not be greater than the total area of the project area: The method for optimizing the artificial reef disposed in the marine pasture described in claim 1, wherein the landscape diversity index in step 3 is the Shannon-Weaver diversity index, and the Shannon-Weaver (Shannon-Weaver) The diversity index is SHDI, and is the target function of the optimization mode of this step. The optimization means that step 3 can make the SHDI value have the maximum value; Representation; n _k is the decision variable; the formula for the Shannon-Weaver (SHDI) diversity index is: , Where P _k is a function of different types of reef fish reef stack probability k appears in the landscape; m type is a stack of different functional types reef fish reef. The optimization method for arranging artificial reefs in marine pastures as described in claim 1, wherein the landscape diversity index in step 3 is a Simpson diversity index, and the Simpson diversity index is SIDI. And for the objective function of the optimization mode of the step, the optimization means that the third step can make the SIDI value have a maximum value; Representation; n _k is the decision variable; the formula for the Simpson diversity index is: , Where P _k is a function of different types of reef fish reef stack probability k appears in the landscape; m type is a stack of different functional types reef fish reef. The optimization method for arranging artificial reefs in marine pastures as described in claim 1, wherein the landscape diversity index in step 3 is a modified Simpson diversity index, which makes the modified Simpson diversity The index is MSIDI, and is the target function of the optimization mode of the step, and the optimization means that step 3 can make the MSIDI value have a maximum value; Representation; n _k is the decision variable; the formula for the modified Simpson diversity index is: MSIDI= , Where P _k is a function of different types of reef fish reef stack probability k appears in the landscape; m type is a stack of different functional types reef fish reef. The optimization method for arranging artificial reefs in marine pastures as described in claim 1, wherein in step 4, the landscape index adopts a landscape relative concentration index (contagion index), so that the relative concentration index of the landscape is (contagion index) RC, and the target function of the optimization mode of the step, the optimization means that step 4 can make the RC value have a maximum value; Representation; X is the decision variable, which refers to the reef position of each reef group; the formula of the relative concentration index of the landscape is: , P _ik is a reef pile of different functional types, the probability of type i adjacent to type k . The requested item 5 of the artificial reefs is arranged in the preferred method of ocean ranch, wherein the method of seeking a P _{ik P ik = P i P k /} i, where P _i is a randomly selected cells in the grid (lattice Cell), belong to the probability of different functional types of reef reef stack i usually be different functional types of reef reef stack i proportion of the area of the whole landscape to estimate; and P _{k / i} at a given reefs reef different functional types of reactor i the case, the different types of functions k reef reef stack adjacent thereto conditions the probability, i.e. Where g _ik is the common boundary length in the landscape, different reef types, type i and type k adjacent; g _i is the total number of i- cells of the reef piles of different functional types or differently decorated in the landscape Estimated total length of the boundary. The method for optimizing the artificial reef in the marine pasture as described in claim 1, wherein in step 4, the landscape index adopts an average patch fractal index, so that the average patch fractal index is MPFD, and this step is The objective function of the optimization mode, the optimization means that step 4 can make the MPFD value have a maximum value; Representation; X is the decision variable, which refers to the reef position of each reef group; the formula of the average patch block fractal dimension is: , L _ik is a reef pile of different functional types, the perimeter of type i and type k , a _ik is a reef pile of different functional types, the area of type i and type k , and N is a reef pile of different functional types, Number of type i and type k ; again 1 MPFD 2. The optimization method for arranging the artificial reef in the marine pasture as described in claim 1, wherein in the fourth step, the landscape index adopts an area-weighted average patch fractal index, so that the area weighted average patch fractal index is AWMPFD. And for the objective function of the step optimization mode, the optimization means that step 4 can make the AWMPFD value have a maximum value; Representation; X is the decision variable, which refers to the reef position of each reef group; the formula of the area weighted average block fractal dimension index is: In the formula For fish reef reefs of different functional types, the ratio of the area of type i and type k to the total area of the landscape; AW MPFD 2. The optimization method for arranging the artificial reef in the marine pasture as described in claim 1, wherein in the fourth step, the landscape index adopts a perimeter area fractal index, so that the perimeter area fractal index is PAFRAC, and this step is The optimization function of the target function, the optimization means that step 4 can make the PAFRAC value have a maximum value; Representation; X is the decision variable, which refers to the reef position of each reef group; the formula of the perimeter area fractal dimension is: , L is a reef pile of different functional types, the perimeter of type i and type k , A is a reef pile of different functional types, the area of type i and type k , h is a constant; PAFRAC 2.