阅读说明：本技术 基于林分发育指数的次生林间伐方法 (Secondary forest cutting method based on forest stand development index ) 是由 胡淑萍 程桂霞 于 2021-08-10 设计创作，主要内容包括：本发明涉及次生林间伐技术领域,特别是一种基于林分发育指数的次生林间伐方法,包括以下步骤,A、对样地内胸径大于5cm的样木进行每木检尺；B、依据样地位置与树种组成,确定植物区系和所处的演替阶段,并按顶级树种、伴生树种、先锋树种和外来树种划分样地树种；C、构建基于林分发育指数的多目标模型；D、利用改进的遗传算法对多目标模型进行求解,获取次生林最优抚育间伐方案。本发明利用林分发育指数构建次生林抚育间伐多目标模型,利用改进的遗传算法进行全局优化,快速获得林分合理的间伐强度和间伐木空间位置,确保间伐方案的准确性和合理性,提高林分质量,促进次生林正向演替。(The invention relates to the technical field of secondary forest cutting, in particular to a secondary forest cutting method based on forest stand development indexes, which comprises the following steps of A, performing per-tree detection on sample trees with the breast height of more than 5cm in a sample plot; B. determining a plant district system and a succession stage according to the sample plot position and the tree species composition, and dividing sample plot tree species according to top tree species, associated tree species, pioneer tree species and foreign tree species; C. constructing a multi-target model based on forest stand development indexes; D. and solving the multi-target model by using an improved genetic algorithm to obtain an optimal secondary forest tending intermediate cutting scheme. According to the secondary forest intermediate cutting method, a secondary forest intermediate cutting multi-target model is constructed by using forest stand development indexes, global optimization is performed by using an improved genetic algorithm, reasonable intermediate cutting strength and intermediate cutting spatial position of the forest stand are obtained rapidly, the accuracy and the reasonability of an intermediate cutting scheme are ensured, the quality of the forest stand is improved, and forward succession of the secondary forest is promoted.)

1. A secondary forest cutting method based on forest stand development indexes is characterized by comprising the following steps of,

A. checking each sample wood with the breast diameter of more than 5cm in the sample plot, and recording the tree species name, the breast diameter, the tree height and the position coordinate of each sample wood;

B. determining a plant district system and a succession stage according to the sample plot position and the tree species composition, and dividing sample plot tree species according to top tree species, associated tree species, pioneer tree species and foreign tree species;

C. constructing a multi-target model based on forest stand development indexes, wherein the multi-target model comprises forest stand mixing degree, tree species dominance degree, forest stand competition indexes, forest stand angle scales and a plurality of constraint conditions;

D. and solving the multi-target model by using an improved genetic algorithm to obtain an optimal secondary forest tending intermediate cutting scheme.

2. A method of secondary forest cutting based on forest stand development index as claimed in claim 1 wherein: the specification of the same in the above step A is 100 m.times.100 m.

3. A method of secondary forest cutting based on forest stand development index as claimed in claim 1 wherein: the top tree species are the final direction of community succession, namely the target guidance of the intermediate cutting.

4. The method for secondary forest cutting based on the forest stand development index as claimed in claim 1, wherein the multi-objective model based on the forest stand development index in the step C is as follows:

wherein SDI is forest stand development index, g is reserved wood after thinning, and M_gIs the forest stand mixed degree after intermediate cutting, A_gFor the dominance degree, CI, of the tree species after intermediate cutting_gIs the post-thinning stand competition index, W_gThe angle scale of the forest stand after intermediate cutting.

5. The method for secondary forest cutting based on forest stand development index as claimed in claim 4, wherein M in the multi-objective model_gIs the single-wood mixing degree M_iMean value of M_iThe calculation formula is as follows:

wherein u is different from the adjacent tree of the j-th plant in the reference tree i_ijIs 1, otherwise is 0.

6. The method for secondary forest cutting based on forest stand development index as claimed in claim 4, wherein CI in the multi-objective model_gIs a single wood competition index CI_iMean value of (CI)_iThe calculation formula is as follows:

in the formula (d)_jThe breast diameter, L, of adjacent wood_DAverage of the distances between 4 adjacent trees and the reference tree in the structural unit, d_iFor reference to the diameter at breast height of the tree, L_ijRepresenting the distance between the adjacent wood j and the reference tree i.

7. The method for secondary forest cutting based on forest stand development index as claimed in claim 4, wherein W in the multi-objective model_gIs a single angle measure W_iMean value of (1), W_iThe calculation formula is as follows:

wherein, when the alpha angle between the reference tree i and the adjacent tree of the jth plant is less than 72 DEG, Z_ijIs 1, otherwise is 0.

8. The method for secondary forest cutting based on forest stand development index as claimed in claim 4, wherein the dominance degree A of tree species in the multi-objective model_gThe calculation formula is as follows:

in the formula, D_gThe relative significance is the ratio of the sectional area of the top-level tree species to the total sectional area of the forest stand;the ratio of the size of the top tree species to the size of the top tree species is the mean value_iThe calculation formula is as follows:

wherein k is a value obtained by subtracting the diameter at breast height of the reference tree i from the diameter of the adjacent tree of the j-th tree_ijIs 1, otherwise is 0.

9. The method for secondary forest cutting based on forest stand development index as claimed in claim 4, wherein the constraint conditions of the multi-objective model are as follows:

1)S_g＝S₀

2)d_g＝d₀

3)1.3≤q_g≤1.7

4)M_g≥M₀

5)A_g≥A₀

6)CI_g≤CI₀

7)|W_g-0.496|≤|W₀-0.496|

8)N_p≥0.7N₀

9)D_t≤D₀

in the formula, S₀、S_gThe number of the forest stand seeds before and after intermediate cutting; d₀、d_gThe number of the forest stand diameter grades before and after intermediate cutting; q. q.s_gThe q value of the forest stand after intermediate cutting is obtained; m₀、M_gThe forest stand mixed degree before and after intermediate cutting; a. the₀、A_gThe dominance degree of the tree species before and after thinning; CI₀、CI_gThe forest stand competition index before and after intermediate cutting; w₀、W_gFor intermediate cuttingThe angle scale of the front forest stand and the rear forest stand; n is a radical of₀、N_pThe number of the trees before and after thinning; d_tThe diameter of the intermediate cut wood; d₀The average diameter of the dominant trees of the forest stand before thinning.

10. The method for secondary forest cutting based on forest stand development index as claimed in claim 1, wherein the improved genetic algorithm in the step D is as follows:

(1) coding the forest stand sample trees by adopting an integer coding mode, namely respectively representing thinning and reserving of the sample trees by 0 and 1;

(2) initializing parameters, wherein the parameters comprise population quantity, genetic algebra and intersample wood thinning probability;

(3) calculating individual fitness (namely forest stand development index), and recording a thinning scheme with the maximum fitness value in each generation;

(4) selecting the thinning wood by adopting a roulette method to form a matched chromosome;

(5) crossing the paired chromosomes in a multipoint crossing mode;

(6) carrying out mutation on the chromosome by adopting a random mutation mode;

(7) the generations are alternated, and the steps (3) to (6) are repeated until the iteration times are reached;

(8) and outputting an optimal thinning scheme according to the size of the fitness value.

Technical Field

The invention relates to the technical field of secondary forest cutting, in particular to a secondary forest cutting method based on forest stand development indexes.

Background

The secondary forest is formed by a large number of germinating forest trees and partial growing forest trees by means of natural force, wherein most of primary vegetation disappears after the primary forest is damaged by artificial interference such as high-strength felling and burning or by serious natural disasters. In China, the area of the secondary forest accounts for about half of the area of the forest in China, and most secondary forests are middle-age and young forests. In order to improve the quality of forest stands, promote the forward succession of secondary forests and restore the zonal top community, tending intermediate cutting is needed for the forest stands with the canopy closure degree larger than 0.7.

The tending intermediate cutting is a forest culture and management measure for properly cutting off part of the forest in due time according to forest stand development, forest competition, natural sparse rules and forest culture targets, adjusting tree species composition and forest stand density, optimizing forest stand structures, improving forest growth environmental conditions, promoting reserved tree growth and shortening culture periods. At present, the research aiming at the tending intermediate felling has achieved fruitful results, but the obvious defects exist: firstly, secondary forest tending intermediate felling is equal to forest stand structure regulation, the tending target of a recovery zonal top-level community is not considered, and an optimization model aiming at secondary forest tending intermediate felling is lacked; secondly, the intermediate cutting model adopts a single-tree-adjusted non-global optimization algorithm or an exhaustive method and other algorithms with low optimization efficiency, and the effective support of an intelligent optimization algorithm is lacked; and thirdly, only paying attention to the accuracy of the thinning scheme and neglecting the determination of the optimal thinning intensity of the forest stand.

1. Secondary forest features and direction of business

The secondary forest is formed by a large number of germinating forest trees and partial seedling forest trees by means of natural force, and the tree species composition and structure of the secondary forest are complex. As the secondary forest in China is mainly young and middle-aged forests, in order to improve the stand quality, promote the forward succession of the secondary forest and restore the zonal top community, artificial auxiliary measures are required to be taken to perform intermediate cutting on the secondary forest.

2. Intermediate valve for tending

The intermediate cutting method is a forest culture and cultivation measure which takes young and middle-aged forests as culture objects, optimizes forest stand structures, improves forest growing environmental conditions, promotes the growth of reserved trees and shortens the cultivation period by adjusting tree species composition and forest stand density.

At present, the tending management conditions are as follows:

the theory is guiding. In the actual operation, GB/T15871-2015 forest tending rule is taken as guidance, the type, application condition, operation flow and the like of tending thinning are provided by the rule, but the position of thinning wood cannot be determined, and forestry workers need to perform field operation according to own working experience. The method is easily influenced by subjective factors, and the forest stand structure is difficult to optimize to the maximum extent.

Step-by-step judgment. And constructing a forest stand multi-target function according to the size ratio, the angular scale, the competition index and the like, calculating a function value of each tree, sequencing the function values, taking the forest with the minimum function value as a thinning object, judging according to index priority or constraint conditions, outputting the thinning object if the thinning object is met, and searching for the next thinning object. The method does not adopt a global optimization method, and the global rationality of the thinning scheme cannot be ensured.

And ③ randomly sampling. The method comprises the steps of constructing a forest stand multi-target function according to a size ratio, an angle scale, a competition index and the like, calculating an initial forest stand function value, sampling according to certain thinning intensity by adopting a Monte Carlo method, calculating a thinning scheme when the thinning function value is maximum after the sampling times are met, and taking the thinning scheme when the thinning function value is maximum as a final result. The method has low calculation efficiency, and the optimal thinning scheme under a certain thinning strength can be obtained by 5000-.

Disclosure of Invention

The technical problem to be solved by the invention is how to perform intercropping on the secondary forest with the canopy closure degree larger than 0.7.

In order to solve the technical problem, the invention comprises a secondary forest cutting method based on forest stand development index, which comprises the following steps,

A. checking each sample wood with the breast diameter of more than 5cm in the sample plot, and recording the tree species name, the breast diameter, the tree height and the position coordinate of each sample wood;

B. determining a plant district system and a succession stage according to the sample plot position and the tree species composition, and dividing sample plot tree species according to top tree species, associated tree species, pioneer tree species and foreign tree species;

C. constructing a multi-target model based on forest stand development indexes, wherein the multi-target model comprises forest stand mixing degree, tree species dominance degree, forest stand competition indexes, forest stand angle scales and a plurality of constraint conditions;

D. and solving the multi-target model by using an improved genetic algorithm to obtain an optimal secondary forest tending intermediate cutting scheme.

Preferably, the specification of the same in the step a is 100m × 100 m.

Preferably, the top tree species is the final direction of community succession, namely the target guide of the felling between nurturations.

Preferably, the multi-target model based on forest stand development index in the step C is as follows:

wherein SDI is forest stand development index, g is reserved wood after thinning, and M_gIs the forest stand mixed degree after intermediate cutting, A_gFor the dominance degree, CI, of the tree species after intermediate cutting_gIs the post-thinning stand competition index, W_gThe angle scale of the forest stand after intermediate cutting.

Preferably, M in the multi-objective model_gIs the single-wood mixing degree M_iMean value of M_iThe calculation formula is as follows:

wherein u is different from the adjacent tree of the j-th plant in the reference tree i_ijIs 1, otherwise is 0.

Preferably, CI in the multi-objective model_gIs a single wood competition index CI_iMean value of (CI)_iThe calculation formula is as follows:

in the formula (d)_jThe breast diameter, L, of adjacent wood_DAverage of the distances between 4 adjacent trees and the reference tree in the structural unit, d_iFor reference to the diameter at breast height of the tree, L_ijRepresenting the distance between the adjacent wood j and the reference tree i.

Preferably, W in the multi-objective model_gIs a single angle measure W_iMean value of (1), W_iThe calculation formula is as follows:

wherein, when the alpha angle between the reference tree i and the adjacent tree of the jth plant is less than 72 DEG, Z_ijIs 1, otherwise is 0.

Preferably, the dominance degree A of the tree species in the multi-target model_gThe calculation formula is as follows:

in the formula, D_gThe relative significance is the ratio of the sectional area of the top-level tree species to the total sectional area of the forest stand;the ratio of the size of the top tree species to the size of the top tree species is the mean value_iThe calculation formula is as follows:

wherein k is a value obtained by subtracting the diameter at breast height of the reference tree i from the diameter of the adjacent tree of the j-th tree_ijIs 1, otherwise is 0.

Preferably, the constraint conditions of the multi-objective model are as follows:

1)S_g＝S₀

2)d_g＝d₀

3)1.3≤q_g≤1.7

4)M_g≥M₀

5)A_g≥A₀

6)CI_g≤CI₀

7)|W_g-0.496|≤|W₀-0.496|

8)N_p≥0.7N₀

9)D_t≤D₀

in the formula, S₀、S_gThe number of the forest stand seeds before and after intermediate cutting; d₀、d_gThe number of the forest stand diameter grades before and after intermediate cutting; q. q.s_gThe q value of the forest stand after intermediate cutting is obtained; m₀、M_gThe forest stand mixed degree before and after intermediate cutting; a. the₀、A_gThe dominance degree of the tree species before and after thinning; CI₀、CI_gThe forest stand competition index before and after intermediate cutting; w₀、W_gThe angle scale of the forest stand before and after intermediate cutting; n is a radical of₀、N_pThe number of the trees before and after thinning; d_tThe diameter of the intermediate cut wood; d₀The average diameter of the dominant trees of the forest stand before thinning.

Preferably, the genetic algorithm improved in the step D is:

(1) coding the forest stand sample trees by adopting an integer coding mode, namely respectively representing thinning and reserving of the sample trees by 0 and 1;

(2) initializing parameters, wherein the parameters comprise population quantity, genetic algebra and intersample wood thinning probability;

(3) calculating individual fitness (namely forest stand development index), and recording a thinning scheme with the maximum fitness value in each generation;

(4) selecting the thinning wood by adopting a roulette method to form a matched chromosome;

(5) crossing the paired chromosomes in a multipoint crossing mode;

(6) carrying out mutation on the chromosome by adopting a random mutation mode;

(7) the generations are alternated, and the steps (3) to (6) are repeated until the iteration times are reached;

(8) and outputting an optimal thinning scheme according to the size of the fitness value.

The method considers the forest stand horizontal structure, forest stand species diversity, forest stand competition and top tree species dominance degree which influence forest stand functions and succession, utilizes forest stand development indexes to construct a secondary forest tending intermediate cutting multi-target model, utilizes an improved genetic algorithm to carry out global optimization, quickly obtains reasonable intermediate cutting strength and intermediate cutting spatial position of the forest stand, ensures the accuracy and the reasonability of an intermediate cutting scheme, improves forest stand quality and promotes forward succession of the secondary forest.

Drawings

The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.

FIG. 1 is a graph showing the spatial position of each sample wood having an internal diameter at breast height of more than 5cm in example 1 of the present application;

FIG. 2 is a spatial position diagram of an inner fell as in example 1 of the present application.

Detailed Description

The forest stand tending and intermediate cutting regulation and control technology in the prior art comprises the following steps:

1. nursing intervalling regulation and control technology research aiming at artificial forest

Selecting a wood forest for 1-5-year-old short-rotation-period eucalyptus pulp, combining 5 sub-targets such as a size ratio, an angular scale, an openness degree, a competition index and a volume to provide an evaluation index, and establishing a forest stand intermediate cutting model by a Monte Carlo method to obtain an optimal cutting scheme. The method comprises the following specific steps:

(1) selecting 6 sample plots of 1-5 year old eucalyptus pure forest, measuring the tree height, the breast diameter and the space coordinate value of each sample plot, and arranging a buffer area of 3m at the edge of each sample plot, wherein the sample plot is 20m multiplied by 20m in size.

(2) Constructing a forest stand evaluation index:

in the formula: g is reserved wood obtained after intermediate cutting; w (g), U (g), UCI (g), K (g) and V (g) are the stand angle scale, the size ratio number, the competition index, the openness and the average individual volume after thinning respectively; sigma_W、σ_u、σ_UCI、σ_K、σ_VThe forest stand angle scale, the size ratio, the competition index, the openness and the standard deviation of the volume are respectively.

(3) Researching the relation between the thinning strength and the improvement amplitude of the evaluation index, and taking the inflection point value as a reasonable thinning strength interval value;

(4) calculating a forest stand evaluation index by adopting a Monte Carlo method, inputting a thinning strength value, setting a model termination condition as no more optimal solution after 10000 times of continuous operation, and taking a scheme with the maximum forest stand evaluation index value as an optimal cutting scheme.

2. Intermediate cutting adjustment technology for secondary forest

The method comprises the steps of taking a first-class checking sample plot of Hunan province as basic data, using spatial structure indexes such as full mixing degree, size ratio, angle scale, competition index and the like as modeling variables, adopting a multiplication method basic idea to construct a structural target function of the Nanmu secondary forest, determining the optimal fault area of the forest at each age stage according to a fitting result of a forest compatibility harvest estimation model, using the optimal fault area as a constraint index for determining thinning amount, constructing a structured management model of the Nanmu secondary forest of Hunan province, and providing a structured management technology of the Nanmu secondary forest through thinning adjustment to promote forest quality improvement increment. The method comprises the following specific steps:

(1) 55 nanmu sample plots in the censored sample plots of 1989-2014 in Hunan province are selected. The sample plot has a side length of 25.82m and an area of 0.067hm²Square of (2). Checking each tree for sample trees with the breast diameter greater than 5cm in the sample plot to obtain factors such as tree species, breast diameter, cutting type, azimuth angle, horizontal distance and the like;

(2) constructing a compatible forest stand growth harvesting model by utilizing survey data of the front stage and the rear stage, and testing to obtain the model with correlation coefficients of 0.97, 0.97 and 0.99 respectively, wherein the model comprises the following steps:

in the formula: m₁To achieve forest stand harvest, M₂Representing the future forest stand harvest, SI representing the status index, t₁To be realistic forest stand age, t₂For the age of the future forest stand, G₁To realize the area of forest stand, G₂The area of the forest stand is the future area of the forest stand.

(3) And determining the optimal sectional area of the forest stand at each age stage according to the fitting result of the forest stand compatibility harvest prediction model, and taking the optimal sectional area as a constraint index for determining the thinning amount.

(4) Adopting the spatial structure indexes of the full mixing degree, the size ratio, the angular scale and the competition index to construct a forest stand structure objective function:

in the formula: q (g) is a forest stand structure target function, M (g), CI (g), U (g) and W (g) respectively represent the total mixing degree, the competition index, the size ratio and the angular scale of a target tree, sigma_M、σ_CI、σ_U、σ_|W-0.375|The standard deviation of the corresponding index.

Constraint conditions are as follows:

1)S＝S₀

2)d＝d₀

3)M≥M₀

4)CI≤CI₀

5)|W-0.375|≤|W₀-0.375|

6)N_p≤N_P-0(1-20%) and N_c＞N_c-0(1-15％)

In the formula: s₀S is the number of species before and after operation, d₀D represents the number of front and back diameter of the operation, M₀M is the forest stand mixing angle before and after intermediate cutting, CI₀CI is competition index before and after thinning, W₀W is the angular dimension before and after intermediate cutting, N_P-0、N_pThe number of the trees before and after thinning, N_c-0、N_cI.e. the number of the tree species of the group in the forest stand before and after the management.

(5) And sequencing the values Q (g) of the objective functions in each space structure unit by calculating the values Q (g). Determining the minimum value of Q (g) as a thinning object, then judging whether constraint conditions 1) -6) are met, if so, outputting the thinning object, and searching the next thinning object. If the constraint condition is not met, the assumption is not true, and the thinning is required to be determined again. Repeating the operations, stopping thinning after the constraint conditions are met, and finally determining thinning.

The disadvantages of the prior art are represented by:

1. the target guidance for secondary interstation is unclear. Secondary forest raising intermediate cutting improves forest stand quality and promotes forward succession of forest stands, a multi-target model constructed at present mainly takes forest stand horizontal structure and vertical structure indexes as main indexes, and consideration on top-level tree species in forest stand succession is insufficient, so that the secondary forest cutting method based on forest stand development indexes is provided.

2. And the secondary intercropping yield based on the intelligent optimization algorithm is less. Secondary forest tending intermediate cutting is systematic work, and any one forest tree in intermediate cutting affects peripheral trees, so that a global optimization method is adopted, and judgment is not carried out according to indexes of fitness values. The Monte Carlo method adopted in forest stand intermediate cutting at present is a method for obtaining an optimal intermediate cutting scheme by sampling in a large quantity under the condition of determining intermediate cutting strength, and the method is large in calculation amount and high in time consumption, so that the method for solving the multi-target model based on the improved genetic algorithm can effectively reduce the running time of the model.

3. The optimal intermediate cutting strength of the secondary forest cannot be determined. At present, the secondary forest intermediate cutting strength is determined by adopting a manual setting mode, namely, an intermediate cutting strength value is manually set at the initial operation stage of a model, or the intermediate cutting strength is set according to the growth quantity of forest stands, the section area of the forest stands and the like, so that the optimal intermediate cutting strength of the forest stands cannot be obtained. The research is based on an improved genetic algorithm, and the optimal intermediate cutting strength and the optimal intermediate cutting position of the forest stand can be obtained simultaneously.

The invention discloses a secondary forest cutting method based on forest stand development index, which comprises the following steps,

A. checking each sample wood with the breast diameter of more than 5cm in the sample plot, and recording the tree species name, the breast diameter, the tree height and the position coordinate of each sample wood;

B. determining a plant district system and a succession stage according to the sample plot position and the tree species composition, and dividing sample plot tree species according to top tree species, associated tree species, pioneer tree species and foreign tree species;

C. constructing a multi-target model based on forest stand development indexes, wherein the multi-target model comprises forest stand mixing degree, tree species dominance degree, forest stand competition indexes, forest stand angle scales and a plurality of constraint conditions;

D. and solving the multi-target model by using an improved genetic algorithm to obtain an optimal secondary forest tending intermediate cutting scheme.

Preferably, the specification of the same in the step a is 100m × 100 m.

Preferably, the top tree species is the final direction of community succession, namely the target guide of the felling between nurturations.

Preferably, the multi-target model based on forest stand development index in the step C is as follows:

wherein SDI is forest stand development index, g is reserved wood after thinning, and M_gIs the forest stand mixed degree after intermediate cutting, A_gFor the dominance degree, CI, of the tree species after intermediate cutting_gIs the post-thinning stand competition index, W_gThe angle scale of the forest stand after intermediate cutting.

Preferably, M in the multi-objective model_gIs the single-wood mixing degree M_iMean value of M_iThe calculation formula is as follows:

wherein u is different from the adjacent tree of the j-th plant in the reference tree i_ijIs 1, otherwise is 0.

Preferably, CI in the multi-objective model_gIs a single wood competition index CI_iMean value of (CI)_iThe calculation formula is as follows:

in the formula (d)_jOf adjacent woodDiameter at breast height, L_DAverage of the distances between 4 adjacent trees and the reference tree in the structural unit, d_iFor reference to the diameter at breast height of the tree, L_ijRepresenting the distance between the adjacent wood j and the reference tree i.

Preferably, W in the multi-objective model_gIs a single angle measure W_iMean value of (1), W_iThe calculation formula is as follows:

wherein, when the alpha angle between the reference tree i and the adjacent tree of the jth plant is less than 72 DEG, Z_ijIs 1, otherwise is 0.

Preferably, the dominance degree A of the tree species in the multi-target model_gThe calculation formula is as follows:

in the formula, D_gThe relative significance is the ratio of the sectional area of the top-level tree species to the total sectional area of the forest stand;the ratio of the size of the top tree species to the size of the top tree species is the mean value_iThe calculation formula is as follows:

wherein k is a value obtained by subtracting the diameter at breast height of the reference tree i from the diameter of the adjacent tree of the j-th tree_ijIs 1, otherwise is 0.

Preferably, the constraint conditions of the multi-objective model are as follows:

1)S_g＝S₀

2)d_g＝d₀

3)1.3≤q_g≤1.7

4)M_g≥M₀

5)A_g≥A₀

6)CI_g≤CI₀

7)|W_g-0.496|≤|W₀-0.496|

8)N_p≥0.7N₀

9)D_t≤D₀

in the formula, S₀、S_gThe number of the forest stand seeds before and after intermediate cutting; d₀、d_gThe number of the forest stand diameter grades before and after intermediate cutting; q. q.s_gThe q value of the forest stand after intermediate cutting is obtained; m₀、M_gThe forest stand mixed degree before and after intermediate cutting; a. the₀、A_gThe dominance degree of the tree species before and after thinning; CI₀、CI_gThe forest stand competition index before and after intermediate cutting; w₀、W_gThe angle scale of the forest stand before and after intermediate cutting; n is a radical of₀、N_pThe number of the trees before and after thinning; d_tThe diameter of the intermediate cut wood; d₀The average diameter of the dominant trees of the forest stand before thinning.

Preferably, the genetic algorithm improved in the step D is:

(1) coding the forest stand sample trees by adopting an integer coding mode, namely respectively representing thinning and reserving of the sample trees by 0 and 1;

(2) initializing parameters, wherein the parameters comprise population quantity, genetic algebra and intersample wood thinning probability;

(3) calculating individual fitness (namely forest stand development index), and recording a thinning scheme with the maximum fitness value in each generation;

(4) selecting the thinning wood by adopting a roulette method to form a matched chromosome;

(5) crossing the paired chromosomes in a multipoint crossing mode;

(6) carrying out mutation on the chromosome by adopting a random mutation mode;

(7) the generations are alternated, and the steps (3) to (6) are repeated until the iteration times are reached;

(8) and outputting an optimal thinning scheme according to the size of the fitness value.

Example 1:

the method takes 1 natural secondary stand of a Jingou ridge forest farm in Wanqing county of Jilin province as a research area.

Selecting one secondary forest stand with the area of 100m multiplied by 100m, dividing a fixed standard into 100 investigation units with the area of 10m multiplied by 10m by adopting an adjacent grid investigation method, listing sample trees with the breast diameter larger than 5cm in each investigation unit, and recording the tree species names, breast diameters, tree heights and position coordinates of the sample trees.

According to the Chinese forest, the area belongs to the northeast temperate zone coniferous forest and coniferous and broadleaf mixed forest area, and the classification result of the tree species is shown in the following table:

taking forest stand development index as a multi-target model of secondary forest intermediate cutting, and calculating a model initial value:

the forest stand sample tree space coordinate graph is shown in fig. 1, and the initial parameters of the forest stand are shown in the following table:

(1) coding the forest stand sample trees by adopting an integer coding mode, namely respectively representing thinning and reserving of the sample trees by 0 and 1;

(2) initializing population parameters, wherein the population number n is 50, the genetic algebra k is 300, and the population space S is np.

(3) Calculating individual fitness value SS by using forest stand development index_iThen, each individual is judged by using the constraint condition of the multi-target model, a penalty function coefficient t is set, and when the constraint condition is S_gWhen t is t-0.2 when 13 is not satisfied, the constraint of c_gWhen 20 is not satisfied, t is t-0.3, and when the constraint condition is that q is not less than 1.3_gT is t-0.1 when not more than 1.7 is satisfied, and when the constraint condition is M_gT is t-0.1 when not more than 0.801 is satisfied, and A is a constraint condition_gWhen not less than 0.443 is not true, t is t-0.1, and when constraint Condition Is (CI)_gWhen 3.297 is not satisfied, t is t-0.1, and when constraint condition is (W)_gWhen-0.496 ≦ 0.136 is not true, t is t-0.1, and when the constraint condition is (N)_pWhen not more than 428 stands, t is t-0.1, when the constraint condition is ninthly_tWhen no more than 33.48 is not satisfied, t is t-0.2;

(4) passing the maximum fitness value and population space value in each generation to maxfittness_[k,:]；

(5) Selecting individuals by using a roulette method, wherein the selection probability of each individual in a population is the ratio of an individual fitness value to a population fitness value, and generating selection probabilities by using random functions of a (np).

(6) The chromosome crossing adopts a random mode and carries out random function m₁Randint (10000) results are segmented when m is₁When the number is less than 10, 4 cross points are generated, and when the number is more than or equal to 10 and less than or equal to m₁When the cross point is less than 100, 3 cross points are generated, and when m is more than or equal to 100₁< 1000, 2 intersections are generated, when m₁When the number is more than or equal to 1000, generating 1 intersection point, wherein the position of each intersection point is determined by t₁Random. randint (left, right), left being the chromosome start position, right being the chromosome end position, paired chromosomes crossing according to the number and position of the crossing points;

(7) the chromosome mutation is carried out in a random mode when a random function m₂Randint (10000) result is less than 3000, a mutation point is generated, otherwise, no mutation occurs, and the position of the mutation point is t₂Randint (1, right), mutating the chromosome according to the number and position of mutation points;

(8) the generations are alternated, and the steps (3) to (7) are repeated for the new filial generation population;

(9) for maxfittness_[k,:]The maximum fitness value in the intermediate tree is sorted, the optimal thinning scheme shows that the optimal thinning intensity of the sample plot is 15.4%, and thinning information is shown in the following table:

(10) the space coordinate diagram of thinning is shown in fig. 2, thinning is marked as a solid circle, and the effect of the plot thinning scheme is shown in the following table:

the invention has the advantages that:

1. and constructing an intermediate cutting model based on forest stand development indexes. The forest stand development index is constructed by adopting the forest stand angular dimension, the forest stand mixed degree, the forest stand competition index and the tree species dominance degree, the horizontal structure of the forest stand, the variety of the forest stand species, the forest stand competition and the top-level tree species dominance degree are comprehensively considered by the index, and a reference basis can be provided for secondary forest tending thinning.

2. The optimal intermediate cutting strength and the intermediate cutting position of the forest stand can be synchronously determined. According to the research, an improved genetic algorithm is utilized, the optimal thinning intensity is intelligently searched within the thinning allowable intensity, thinning is accurately marked, and quantification and precision of secondary forest tending can be realized.

Although specific embodiments of the present invention have been described above, it will be appreciated by those skilled in the art that these are merely examples and that many variations or modifications may be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims.