WO2022170901A1 - 一种鱼类偏好栖息地的确定方法及终端设备 - Google Patents

一种鱼类偏好栖息地的确定方法及终端设备 Download PDF

Info

Publication number
WO2022170901A1
WO2022170901A1 PCT/CN2022/071244 CN2022071244W WO2022170901A1 WO 2022170901 A1 WO2022170901 A1 WO 2022170901A1 CN 2022071244 W CN2022071244 W CN 2022071244W WO 2022170901 A1 WO2022170901 A1 WO 2022170901A1
Authority
WO
WIPO (PCT)
Prior art keywords
target fish
fish
time
formula
target
Prior art date
Application number
PCT/CN2022/071244
Other languages
English (en)
French (fr)
Inventor
权全
高少泽
杨思敏
樊荣
Original Assignee
西安理工大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 西安理工大学 filed Critical 西安理工大学
Priority to US17/750,027 priority Critical patent/US11645354B2/en
Publication of WO2022170901A1 publication Critical patent/WO2022170901A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/10—Numerical modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/08—Fluids
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/08—Thermal analysis or thermal optimisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
    • Y02A40/80—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in fisheries management
    • Y02A40/81—Aquaculture, e.g. of fish

Definitions

  • the invention discloses a method and terminal equipment for determining the preferred habitat of fish, belonging to the technical field of hydraulic engineering.
  • the purpose of this application is to provide a method and terminal device for determining the preferred habitat of fish, which can carry out targeted restoration of the habitat, thereby solving the technical problem of high cost of restoration of the existing habitat.
  • a first aspect of the present invention provides a method for determining the preferred habitat of fish, comprising:
  • the preferred habitat of the target fish is determined from the potential habitats using a preference learning model based on the cumulative density method.
  • the ecological functions include growth functions, flocking functions and foraging functions.
  • the growth function is determined according to a first formula, and the first formula is:
  • l t is the average body length of the target fish at time t
  • W t is the average body weight of the target fish at time t
  • l ⁇ is the average asymptotic body length of the target fish
  • W ⁇ is the The average progressive body weight of the target fish is described
  • k is the growth coefficient
  • t 0 is the assumed theoretical growth starting age.
  • the cluster function is determined according to a second formula, and the second formula is:
  • D i,t+1 ⁇ 1 D i,t + ⁇ 2 D′ i,t + ⁇ 3 D′′ i,t + ⁇ 4 D′′′ i, t
  • D i,t+1 is the movement direction of the i-th target fish individual at time t+1
  • D i,t is the movement direction of the i-th target fish individual at time t
  • D′ i,t is t
  • the foraging function is determined according to a third formula, and the third formula is:
  • the movement trajectory of the target fish in the biological simulation model is obtained, specifically:
  • the movement trajectory of the target fish in the biological simulation model is obtained according to the fourth formula, where the fourth formula is:
  • the movement speed and movement direction of the target fish are determined according to the fifth formula, and the fifth formula is:
  • S i,t is the movement speed of the ith target fish at time t
  • D i,t is the movement direction of the ith target fish at time t
  • D_fav(t) and S_fav(t) are respectively
  • the preferred flow velocity of the i-th target fish individual within its sensing range is compared with the current position and the movement speed toward the direction of movement
  • D_flee(t) and S_flee(t) are respectively The movement direction of the i target fish individuals fleeing from the nearest neighboring individuals within their sensing range and the movement speed of escaping the movement direction.
  • the method for determining the sensing range is:
  • the perception range of the target fish is determined.
  • the preferred habitat of the target fish is determined from the potential habitat using a preference learning model based on the cumulative density method, specifically:
  • the preference learning model based on the cumulative density method uses the preference learning model based on the cumulative density method to determine the preference value of the ith target fish for each of the potential habitats; the preference learning model based on the cumulative density method is determined according to the sixth formula, and the sixth formula for:
  • P ij is the cumulative density of the i-th target fish in the j-th potential habitat, and is recorded as the preference value; is the total cumulative density of the target fish species present in all potential habitats;
  • the preferred habitat of the target fish is determined according to the preference value.
  • a second aspect of the present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the above when executing the computer program steps of the method.
  • the method for determining the preferred habitat of fish and the terminal device of the present invention have the following beneficial effects:
  • the method for determining the fish's preferred habitat of the present invention can determine the fish's preferred habitat, so as to carry out targeted restoration, while ensuring the living environment of the fish, the cost of restoring the habitat is reduced.
  • Fig. 1 is the flow chart of the determination method of fish preference habitat provided by the present invention
  • FIG. 2 is a schematic structural diagram of a subject-based model in a specific embodiment of the present invention.
  • FIG. 3 is a schematic diagram of the motion rule of the agent fish particle in a specific embodiment of the present invention.
  • a first aspect of the present invention provides a method for determining the preferred habitat of fish, as shown in Figure 1, comprising:
  • Step 1 Obtain the parameter information of the water environment where the target fish is located, and use the parameter information of the water environment to establish a three-dimensional water environment model, specifically:
  • Step 1.1 Obtain the parameter information of the water environment such as regional boundary conditions, underwater topographic data, measured water level, water temperature, water quality, and water velocity of the water environment where the target fish is located.
  • Step 1.2 using the parameter information of the water environment to establish a three-dimensional water environment model.
  • a three-dimensional water environment model is established using the MIKE 3FM module in the MIKE series software.
  • the simulation information corresponding to the parameter information of the water environment can be obtained from the model.
  • the above-mentioned simulation information and the three-dimensional water environment model provide a platform for building a biological simulation model.
  • Step 2 Determine the ecological function of the target fish, and build a biological simulation model on the three-dimensional water environment model in combination with the ecological function, specifically:
  • Step 2.1 Determine the growth function, cluster function and foraging function of the target fish, specifically:
  • Step 2.1.1 Obtain long-term domestication observation data and ecological research data of target fish.
  • Step 2.1.2 According to long-term domestication observation data and ecological research data, determine the ecological characteristics of the target fish, and establish a growth function, a colony function and a foraging function according to its ecological characteristics;
  • the growth function is determined according to the first formula, and the first formula is:
  • l t is the average body length of the target fish at time t
  • W t is the average body weight of the target fish at time t
  • l ⁇ is the average asymptotic body length of the target fish
  • W ⁇ is the average asymptotic body weight of the target fish
  • k is the growth coefficient
  • t 0 is the assumed theoretical growth starting age.
  • the cluster function is determined according to the second formula, and the second formula is:
  • D i,t+1 ⁇ 1 D i,t + ⁇ 2 D′ i,t + ⁇ 3 D′′ i,t + ⁇ 4 D′′′ i, t
  • D i,t+1 is the movement direction of the i-th target fish individual at time t+1
  • D i,t is the movement direction of the i-th target fish individual at time t
  • D′ i,t is t
  • the average value of the direction from the adjacent individuals of the safe distance to the i-th target fish individual to avoid obstacles).
  • the size of the weight can be determined according to preference.
  • the foraging function is determined according to the third formula, which is:
  • the selection function of the target fish body weight growth pattern is:
  • Step 2.2 combining the growth function, the cluster function and the foraging function, build a biological simulation model on the three-dimensional water environment model.
  • the built biological simulation model is an agent-based model (ABM).
  • the method for determining the preferred habitat of the target fish of the present invention is determined on the biological simulation model.
  • Step 3 Obtain the movement trajectory of the target fish in the biological simulation model, and determine the potential habitat of the target fish according to the movement trajectory, specifically:
  • Step 3.1 Obtain the motion trajectory of the target fish in the biological simulation model according to the fourth formula.
  • the fourth formula is:
  • the movement speed and movement direction of the target fish are determined according to the fifth formula, and the fifth formula is:
  • S i,t is the movement speed of the ith target fish at time t
  • D i,t is the movement direction of the ith target fish at time t
  • D_fav(t) and S_fav(t) are respectively
  • the preferred flow velocity of the i-th target fish individual within its sensing range is compared with the current position and the movement speed towards the direction of movement
  • D_flee(t) and S_flee(t) are respectively The movement direction of the i target fish individuals fleeing from the nearest neighboring individuals within their sensing range and the movement speed of escaping the movement direction.
  • the above perception range is determined according to the visual information, auditory information and olfactory information of the target fish, specifically:
  • the experimental fish were trained on conditioning reflexes first, and a striped plate was used as a stimulus signal. Displays the streaks within the fish's visual range, and then delivers the bait to the aquarium. Through the two events of streak plate and food, train fish to establish conditioned reflex to streak plate. After the conditioned reflex is established, if the fish shows the behavior of feeding, it will be used as the basis for seeing the streak, and the fish's visual distance measurement experiment will be carried out. Experiments have shown that the sight distance of fish is 10m on average. The experimental record is 15m, up to 30m. Different fish have different sight distances. Under the same conditions, human visual ability can reach more than 100m.
  • the stones and bait wrapped with algae were used as the experimental objects, and the reaction of the fish was observed. It was found that the fish did not respond to the stones, and the bait-coated experimental objects were found and swallowed within an average of 3 minutes. When the nostrils of the class are blocked, nothing happens. When the bait is thrown into the sea, the fish at 10m immediately swim to the bait.
  • the perception range of the target fish is determined.
  • the perceptual range of the target fish is 10 meters.
  • Different species of fish have different perception ranges, which need to be comprehensively considered according to the experimental results.
  • Step 3.2 Determine the potential habitat of the target fish according to the movement trajectory, specifically:
  • the determination rule adopted can be that the river shoals, ravines and bays, etc. whose linear distance from the motion trajectory is less than the set distance threshold, are potential habitats. In this step, many potential habitats are obtained, and if all of them are restored, the restoration cost will be high.
  • Step 4 Use the preference learning model based on the cumulative density method to determine the preferred habitat of the target fish from the potential habitats, specifically:
  • Step 4.1 Use the preference learning model based on the cumulative density method to obtain the preference value of the ith target fish for each potential habitat; the preference learning model based on the cumulative density method is determined according to the ninth formula, and the ninth formula is:
  • P ij is the cumulative density of the ith target fish in the jth potential habitat, is the total cumulative density of target fish species present in all potential habitats;
  • the value of P ij is set as the preference value of the ith target fish for each potential habitat.
  • Step 4.2 Determine the preferred habitat of the target fish according to the preference value.
  • the preference value is compared with the preset preference threshold, and the potential habitat corresponding to the preference value greater than the preset preference threshold is the preferred habitat of the target fish.
  • the preferred habitat of the target fish can be restored, so as to achieve targeted restoration, achieve the best restoration effect while reducing the cost, and ensure the living environment of the fish.
  • a second aspect of the present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the above method when executing the computer program.
  • the simulation is performed using an Agent-based Model (ABM).
  • ABM Agent-based Model
  • the structure diagram of the agent-based model is shown in Figure 2.
  • This model is often used to simulate the behavior and interactions of autonomous agents, assessing the impact of autonomous agents on the overall system.
  • the subject being simulated in the ABM framework can take active actions in a specific environment according to different information instead of passively listening to instructions.
  • the framework consists of two levels: individual and environment. One is based on the hydrodynamic (HD) and advective-diffusion (AD) modules in the commercial software system Mike 21/3 FM developed by Danhua Water conserveancy. It calculates the parameter information of the water environment, such as water depth, flow velocity, water level, temperature and salinity. , get the state variables, process and calculation formula of the particle, and assign the particle properties.
  • HD hydrodynamic
  • AD advective-diffusion
  • the environment in which the above agent-based model is located is the environment provided by ECO Lab, which is used by users to develop equations, implement and execute them on the 3D high-definition model Mike 21/3 FM.
  • ECO Lab the environment provided by ECO Lab
  • AD convective-dispersion
  • Table 1 is the ABM model template parameter setting and process expression description.
  • the Lagrangian algorithm is used to construct a model for determining the preferred area of the target fish under the framework of ABM, and the visual ability, hearing ability, olfactory ability and related behaviors, swimming and swimming of fish are studied in combination with fish domestication experiments.
  • Ability and method to get the preferred area of target fish In this model, the target fish is replaced by surrogate fish particles, and the surrogate fish particle tracking is described in Lagrangian form, and the ordinary differential equations are solved using Newton's law of motion.
  • the forces on particles can be divided into two categories, one is derived from external fields, and the other is derived from particle interactions.
  • Fish movement has a population aggregation effect, and most fish populations have staged clustering behaviors in their life cycles, and there are chase-feeding behaviors from fry to adults.
  • the food concentration of the agent fish particles is the first choice within the good bait area, which is shown as foraging movement; the basic movement is selected outside the good bait area, which is shown as the first choice of suitable temperature, then suitable water depth, and finally suitable flow rate.
  • the floating swimming of individual particles has the characteristics of active migration and movement. First, it is an unrestricted random walk; if there are obstacles on the forward route, it will detour, and if favorable conditions appear, it will be induced; through the judgment of environmental variables within the perception range, A restricted area search is performed and the heading is induced.
  • the particles can reflect the habits of different stages from juvenile to adult fish, and different calculation functions need to be set for individual fish.
  • the motion functions of the surrogate particle random rules, deterministic equations and mixing rules, as well as growth functions (size, weight), feeding functions (fixed), and swarming functions are given to obtain a Lagrangian-based dynamic model of individual fish.
  • l t is the average body length of the target fish at time t
  • W t is the average body weight of the target fish at time t
  • l ⁇ is the average asymptotic body length of the target fish
  • W ⁇ is the The average progressive body weight of the target fish is described
  • k is the growth coefficient
  • t 0 is the assumed theoretical growth starting age.
  • the movement characteristics of the target fish population are in the same direction, alignment, and obstacle avoidance.
  • the formula for determining the movement direction of the group movement is:
  • D i,t+1 ⁇ 1 D i,t + ⁇ 2 D′ i,t + ⁇ 3 D′′ i,t + ⁇ 4 D′′′ i,t
  • D i,t+1 is the movement direction of the i-th target fish individual at time t+1
  • D i,t is the movement direction of the i-th target fish individual at time t
  • D′ i,t is t
  • the average value of the direction from the adjacent individuals of the safe distance to the i-th target fish individual to avoid obstacles).
  • the size of the weight can be determined according to preference.
  • the foraging rate discriminant is:
  • the movement of the agent can be affected by a variety of environmental factors at the same time, including factors such as flow velocity, water depth, water temperature, water concentration and other factors based on the Euler method, as well as other agents based on the Lagrangian method and the current /Interaction between target agents. If the spatial displacement of the agent in the next time step at each time step depends only on its preferred flow rate and the position of its neighbors (such as escaping from its closest individuals), the current speed and direction of the agent's upcoming movement will be the above two The superposition result of the agent's motion velocity vector under the influence of each factor. As shown in Figure 3.
  • the five-pointed star represents the agent fish particle, that is, the biological individual; all grids contain flow velocity information; the dotted circle is the perception range of a certain agent fish particle (pentagram); Pulling the field environment factor, in this case the flow rate, the flow rate has the flow velocity that the agent fish particles like the most (the flow rate that the individual fish prefers generally has a certain value range); the five-pointed star outside the dotted circle indicates the current agent fish particle perception ability Additional proxy fish particles out of range.
  • the moving speed of the agent fish particle at the current moment is S t
  • the moving direction is D t ; the grid where the preferred flow velocity perceived by the agent fish particle at the current moment within the search radius (dotted circle) is compared with the position of the agent fish particle itself.
  • the direction D_fav(t) of and the speed S_fav(t) moving in that direction is the reverse D_flee(t) of its own position compared to its perceived nearest agent within the search radius (dotted circle) and fleeing in that direction
  • the velocity S_flee(t) of the movement is superimposed and calculated by the two groups of velocity offsets to obtain the movement speed S(t) and the movement direction D(t) of the agent at the current moment.
  • the position of the agent fish particle individual i at a specific time step t is where x and y are Cartesian coordinates, and the position at time t+1 is modeled as:
  • the particles take the characteristics of the life history of fish at different stages as the priority, search for the fish preference characteristic parameters such as the maximum feed concentration, suitable temperature, suitable flow rate, etc., and add another parameter to the search parameters. Particle features that extend the simulation of single individuals in fish behavioral models. Finally, through the Euler-Lagrange information interaction under the ABM framework. Realize the simulation technology of "exogenous environment-endogenous perception-fish movement decision".
  • the purpose of studying the change of target fish quantity is to understand the status quo of fish resources and predict its changing trend, which is the scientific basis for aquatic fishing, fish proliferation and fishery resources management.
  • the changes in the number of fish populations are caused by various factors such as changes in fishing intensity, changes in environmental conditions such as water temperature, and river hydrological conditions, and fluctuations in the amount of bait.
  • simulation calculation and analysis are carried out on the reproduction behavior, aggregation effect and swimming direction consistency of fry near the shore, which is convenient for the subsequent proliferation of fish in newly formed habitats and the planning of the protection scope of fish spawning grounds. Provide certain scientific basis and support.
  • P ij is the cumulative density of the i-th target fish in the j-th potential habitat, is the cumulative total density of the target fish species present in all potential habitats.
  • Fry flocking behavior is reflected after the fry have swimming ability. As the individual grows up, the fry flock form changes constantly. The fry flock has the advantage of ecological adaptation, which is beneficial to the growth and survival of the individual.
  • Fish swimming in groups can save the energy of individuals and reduce consumption. When swimming in groups, the leading fish exchange continuously, and the energy consumption of individuals is relatively uniform. Fry group life has a sense of security, and group rest at night is one of the manifestations.
  • the fish in the group can quickly respond to changes in the swimming speed of their neighbors and can escape synchronously.
  • the swarming behavior helps fish to feed and avoid obstacles. Plankton sometimes gathers into groups, and the single-tailed fish enter the grouping area in a panic. Two groups of fish enter this area due to the consistency of swimming speed and swimming direction. It is beneficial to prey on plankton, and it can flexibly and maneuver around the plankton, which is beneficial to improve the survival rate of fry.
  • the newly hatched fry have no flocking behavior, and the flocking behavior appears gradually after they have the ability to swim. From the survival time and swimming trajectory of the fry, it can be seen that the group living area of the fry is within a certain range, which is in line with the phenomenon of fry aggregation found in the shallow and slow-flow area on the shore. It also reflects the avoidance effect of side wall obstacles.
  • the method of the invention can determine the habitats preferred by the fishes, so as to carry out targeted restoration, and while ensuring the living environment of the fishes, the cost of restoring the habitats is reduced.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • Computer Hardware Design (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Algebra (AREA)
  • Artificial Intelligence (AREA)
  • Fluid Mechanics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computational Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Farming Of Fish And Shellfish (AREA)
  • Environmental Sciences (AREA)
  • Marine Sciences & Fisheries (AREA)
  • Animal Husbandry (AREA)
  • Biodiversity & Conservation Biology (AREA)

Abstract

一种鱼类偏好栖息地的确定方法及终端设备，所述方法包括：获取目标鱼类所在水环境的参数信息，利用水环境的参数信息建立三维水环境模型（步骤1）；确定目标鱼类的生态学函数，结合生态学函数在三维水环境模型上搭建生物仿真模型（步骤2）；获取生物仿真模型中目标鱼类的运动轨迹，根据运动轨迹确定目标鱼类的潜在栖息地（步骤3）；利用基于累计密度法的偏好学习模型从潜在栖息地中确定目标鱼类的偏好栖息地（步骤4）。可以确定鱼类偏好的栖息地，从而进行有针对性的修复，在保证鱼类生存环境的同时，降低了修复栖息地的成本。

Description

一种鱼类偏好栖息地的确定方法及终端设备 技术领域
本发明公开了一种鱼类偏好栖息地的确定方法及终端设备，属于水利工程技术领域。
背景技术
河流生态系统中，鱼类作为主要的生物因素，其生长繁殖与栖息地水生生境关系密切。鱼类对生存环境中的温度变化反应强烈、快速，是检测和记录气候变化对淡水生态系统的理想指示物种。近年来，由于大规模进行河流的开发和利用，河流的水文情势、水动力、水环境的天然状态发生改变，鱼类栖息地环境遭到破坏，鱼类的生存受到威胁。鱼类栖息地的研究不仅对更好地了解水库建设对河流生态系统的影响具有重要意义，更是鱼类资源保护的重要手段。
现有技术在修复被破坏的栖息地时，是将各个栖息地全部进行修复。而鱼类对栖息地的选择是有偏好的，有些栖息地的利用率并不高，全部修复成本较高。
发明内容
本申请的目的在于，提供一种鱼类偏好栖息地的确定方法及终端设备，可对栖息地进行针对性的修复，从而解决现有栖息地修复成本高的技术问题。
本发明的第一方面提供了一种鱼类偏好栖息地的确定方法，包括：
获取目标鱼类所在水环境的参数信息，利用所述水环境的参数信息建立三维水环境模型；
确定所述目标鱼类的生态学函数，结合生态学函数在所述三维水环境模型上搭建生物仿真模型；
获取所述生物仿真模型中目标鱼类的运动轨迹，根据运动轨迹确定所 述目标鱼类的潜在栖息地；
利用基于累计密度法的偏好学习模型从所述潜在栖息地中确定所述目标鱼类的偏好栖息地。
优选地，所述生态学函数包括生长函数、集群函数和觅食函数。
优选地，所述生长函数根据第一公式确定，所述第一公式为：
Figure PCTCN2022071244-appb-000001
式中，l t为所述目标鱼类t时刻的平均体长，W t为所述目标鱼类t时刻的平均体重，l ∞为所述目标鱼类的平均渐进体长，W ∞为所述目标鱼类的平均渐进体重，k为生长系数，t 0为假设的理论生长起点年龄。
优选地，所述集群函数根据第二公式确定，所述第二公式为：
D i,t+1＝λ 1D i,t+λ 2D′ i,t+λ 3D″ i,t+λ 4D″′ i,t
式中，D i,t+1为t+1时刻第i个目标鱼类个体的运动方向，D i,t为t时刻第i个目标鱼类个体的运动方向，D′ i,t为t时刻第i个目标鱼类个体到临近个体平均位置的方向，D″ i,t为t时刻第i个目标鱼类个体的临近个体的平均方向，D″′ i,t为t时刻小于预设安全距离的临近个体到第i个目标鱼类个体方向的平均值，λ 1,λ 2,λ 3,λ 4为权重，且λ 1+λ 2+λ 3+λ 4＝1。
优选地，所述觅食函数根据第三公式确定，所述第三公式为：
eat＝if(fixed＜0.1,0,fixed::-(0.1*fixed))
式中，fixed为饵料浓度。
优选地，获取所述生物仿真模型中目标鱼类的运动轨迹，具体为：
根据第四公式获取所述生物仿真模型中目标鱼类的运动轨迹，所述第四公式为：
Figure PCTCN2022071244-appb-000002
式中，
Figure PCTCN2022071244-appb-000003
为t时刻的第i个所述目标鱼类所在位置，
Figure PCTCN2022071244-appb-000004
为t+1时刻第i个所述目标鱼类所在位置，x和y为笛卡尔坐标系中的横轴和纵轴，S i,t为t时刻第i个目标鱼类个体的运动速度，D i,t为t时刻第i个目标鱼类个体的运动方向，θ i,t为t时刻第i个目标鱼类个体运动方向为D i,t时的夹角，
Figure PCTCN2022071244-appb-000005
为位移偏角；S i,t∈(0,S max)，S max为所述目标鱼类从t时刻至t+1时刻的时间段内的最大移动速度。
优选地，目标鱼类个体的运动速度和运动方向根据第五公式确定，所述第五公式为：
Figure PCTCN2022071244-appb-000006
式中，S i,t为t时刻第i个目标鱼类个体的运动速度，D i,t为t时刻第i个目标鱼类个体的运动方向，D_fav(t)和S_fav(t)分别为t时刻第i个目标鱼类个体在其感知范围内的喜好流速所在位置相比于当前位置的运动方向及朝向该运动方向的运动速度；D_flee(t)和S_flee(t)分别为t时刻第i个目标鱼类个体逃离其感知范围内最近的临近个体的运动方向及逃离该运动方向的运动速度。
优选地，所述感知范围的确定方法为：
获取所述目标鱼类的视觉信息、听觉信息和嗅觉信息；
根据所述视觉信息、听觉信息和嗅觉信息，确定所述目标鱼类的感知范围。
优选地，利用基于累计密度法的偏好学习模型从所述潜在栖息地中确 定所述目标鱼类的偏好栖息地，具体为：
利用基于累计密度法的偏好学习模型，获取第i个目标鱼类对每个所述潜在栖息地的偏好值；所述基于累计密度法的偏好学习模型根据第六公式确定，所述第六公式为：
Figure PCTCN2022071244-appb-000007
式中，P ij为第i个目标鱼类在第j个所述潜在栖息地中出现的累计密度，记为偏好值；
Figure PCTCN2022071244-appb-000008
为所有潜在栖息地中出现所述目标鱼类的总累计密度；
根据所述偏好值确定所述目标鱼类的偏好栖息地。
本发明的第二方面提供了一种终端设备，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，所述处理器执行所述计算机程序时实现上述方法的步骤。
本发明的鱼类偏好栖息地的确定方法及终端设备，相较于现有技术，具有如下有益效果：
本发明的鱼类偏好栖息地的确定方法可以确定鱼类偏好的栖息地，从而进行有针对性的修复，在保证鱼类生存环境的同时，降低了修复栖息地的成本。
附图说明
图1为本发明提供的鱼类偏好栖息地的确定方法的流程图；
图2为本发明具体实施例中基于主体的模型的结构示意图；
图3为本发明具体实施例中代理鱼粒子运动规则示意图。
具体实施方式
以下描述中，为了说明而不是为了限定，提出了诸如特定系统结构、技术之类的具体细节，以便透彻理解本发明实施例。然而，本领域的技术人员应当清楚，在没有这些具体细节的其它实施例中也可以实现本发明。在其它情况中，省略对众所周知的系统、装置、电路以及方法的详细说明，以免不必要的细节妨碍本发明的描述。
本发明的第一方面提供了一种鱼类偏好栖息地的确定方法，如图1所 示，包括：
步骤1、获取目标鱼类所在水环境的参数信息，利用水环境的参数信息建立三维水环境模型，具体为：
步骤1.1、获取目标鱼类所在水环境的区域边界条件、水下地形数据、实测水位、水温、水质、水流速等水环境的参数信息。
步骤1.2、利用水环境的参数信息建立三维水环境模型。示例性地，利用MIKE系列软件中的MIKE 3FM模块建立三维水环境模型。
建立三维水环境模型后，就可以从模型中获取水环境的参数信息对应的模拟信息，上述模拟信息及三维水环境模型为生物仿真模型搭建提供了平台。
步骤2、确定目标鱼类的生态学函数，结合生态学函数在三维水环境模型上搭建生物仿真模型，具体为：
步骤2.1、确定目标鱼类的生长函数、集群函数和觅食函数，具体为：
步骤2.1.1、获取目标鱼类长期驯养观测资料及生态学研究资料。
步骤2.1.2、根据长期驯养观测资料及生态学研究资料，确定目标鱼类的生态学特性，并依据其生态学特性建立生长函数、集群函数和觅食函数；
其中，生长函数根据第一公式确定，第一公式为：
Figure PCTCN2022071244-appb-000009
式中，l t为目标鱼类t时刻的平均体长，W t为目标鱼类t时刻的平均体重，l ∞为目标鱼类的平均渐进体长，W ∞为目标鱼类的平均渐进体重，k为生长系数，t 0为假设的理论生长起点年龄。
集群函数根据第二公式确定，第二公式为：
D i,t+1＝λ 1D i,t+λ 2D′ i,t+λ 3D″ i,t+λ 4D″′ i,t
式中，D i,t+1为t+1时刻第i个目标鱼类个体的运动方向，D i,t为t时刻第i个目标鱼类个体的运动方向，D′ i,t为t时刻第i个目标鱼类个体到临近个体平均位置的方向，D″ i,t为t时刻第i个目标鱼类个体的临近个体的平均方 向，D″′ i,t为t时刻小于预设安全距离的临近个体到第i个目标鱼类个体方向的平均值(规避障碍)，考虑到对鱼的影响力不同，还需要对各个方向加权，取加权平均值，权重的大小可以根据偏好确定，λ 1,λ 2,λ 3,λ 4为权重，且λ 1+λ 2+λ 3+λ 4＝1。
觅食函数根据第三公式确定，第三公式为：
eat＝if(fixed＜0.1,0,fixed::-(0.1*fixed))
式中，fixed为饵料浓度。
目标鱼类体重生长模式的选择函数为：
G＝Fixed*0.01*area
式中，fixed为饵料浓度，area为单位网格面积。
步骤2.2、结合生长函数、集群函数和觅食函数，在三维水环境模型上搭建生物仿真模型，示例性地，所搭建的生物仿真模型为基于主体的模型(Agent-basedModel,ABM)。
本发明确定目标鱼类的偏好栖息地的方法是在该生物仿真模型上确定的。
步骤3、获取生物仿真模型中目标鱼类的运动轨迹，根据运动轨迹确定目标鱼类的潜在栖息地，具体为：
步骤3.1、根据第四公式获取生物仿真模型中目标鱼类的运动轨迹，第四公式为：
Figure PCTCN2022071244-appb-000010
式中，
Figure PCTCN2022071244-appb-000011
为t时刻的第i个目标鱼类所在位置，
Figure PCTCN2022071244-appb-000012
为t+1时刻第i 个目标鱼类所在位置，x和y为笛卡尔坐标系中的横轴和纵轴，S i,t为t时刻第i个目标鱼类个体的运动速度，D i,t为t时刻第i个目标鱼类个体的运动方向，θ i,t为t时刻第i个目标鱼类个体运动方向为D i,t时的夹角，
Figure PCTCN2022071244-appb-000013
为位移偏角；S i,t∈(0,S max)，S max为目标鱼类从t时刻至t+1时刻的时间段内的最大移动速度。
其中，目标鱼类个体的运动速度和运动方向根据第五公式确定，第五公式为：
Figure PCTCN2022071244-appb-000014
式中，S i,t为t时刻第i个目标鱼类个体的运动速度，D i,t为t时刻第i个目标鱼类个体的运动方向，D_fav(t)和S_fav(t)分别为t时刻第i个目标鱼类个体在其感知范围内的喜好流速所在位置相比于当前位置的运动方向及朝向该运动方向的运动速度；D_flee(t)和S_flee(t)分别为t时刻第i个目标鱼类个体逃离其感知范围内最近的临近个体的运动方向及逃离该运动方向的运动速度。
上述感知范围是根据目标鱼类的视觉信息、听觉信息和嗅觉信息确定的，具体为：
获取目标鱼类的视觉信息、听觉信息和嗅觉信息。上述信息可以根据实验获得。
示例性的：
a、目标鱼类的视觉信息的测定实验为：
先对实验鱼进行条件反射的训练，用绘有条纹的平板作为刺激信号。 在鱼类视觉范围内显示条纹板，随即向水族箱投放饵料。通过条纹板与食物两个事件，训练鱼类建立对条纹板的条件反射。条件反射建立后，如鱼类显示摄食的行为，则作为看见条纹板的依据，由此进行鱼类的视距测定实验。实验表明鱼类的视距平均为10m。实验记录为15m，最高达30m。不同的鱼类的视距不同，在相同条件下，人的视觉能力达100m以上。
b、目标鱼类的听觉信息的测定实验为：
实验采用水下扬声器在水下50cm处向鱼群放声，平均声压90分贝，水面照度为01～1lx：实验鱼群的鱼体平均体长143cm，平均体重30g。船上设有荧光灯，灯距水面2m。先用荧光灯集鱼，然后向聚有30尾左右的鱼群处放声。为了防止产生适应性，避免采用频率接近的单音，即相邻两次的声波频率要具有较大的差异。放声后观察鱼群的反应，直到鱼群全部恢复稳定后，再进行下一次实验。实验表明鱼群对900Hz反应最强烈，其次是950Hz、1050Hz和750Hz。
c、目标鱼类的嗅觉信息的测定实验为：
用大叶藻包了石块和饵料作为实验物，观察鱼类的反应，发现鱼类对包石头的没有反应，而在平均3min内就发现和吞食包饵料的实验物，当用棉花将鱼类的鼻孔堵上时，没有任何反应。将饵料投入海中时，10m处的鱼类立刻游向饵料。
根据视觉信息、听觉信息和嗅觉信息，确定目标鱼类的感知范围。
在该确定过程中，可以综合考虑视觉信息、听觉信息和嗅觉信息，然后确定目标鱼类的感知范围。
例如，综合考虑上述测定实验获得的目标鱼类的视觉信息、听觉信息 和嗅觉信息，得到目标鱼类的感知范围为10米。不同种类的鱼其感知范围不同，需要根据实验测定结果进行综合考虑。
步骤3.2、根据运动轨迹确定目标鱼类的潜在栖息地，具体为：
所采取的确定规则可以为距运动轨迹直线距离小于设定距离阈值的河道浅滩、沟壑和水湾等地，即为潜在栖息地。在该步骤，所得到的潜在栖息地较多，如全部进行修复，则修复成本较高。
步骤4、利用基于累计密度法的偏好学习模型从潜在栖息地中确定目标鱼类的偏好栖息地，具体为：
步骤4.1、利用基于累计密度法的偏好学习模型，获取第i个目标鱼类对每个潜在栖息地的偏好值；基于累计密度法的偏好学习模型根据第九公式确定，第九公式为：
Figure PCTCN2022071244-appb-000015
式中，P ij为第i个目标鱼类在第j个潜在栖息地中出现的累计密度，
Figure PCTCN2022071244-appb-000016
为所有潜在栖息地中出现目标鱼类的总累计密度；
P ij的数值设定为第i个目标鱼类对每个潜在栖息地的偏好值。
步骤4.2、根据偏好值确定目标鱼类的偏好栖息地。
将偏好值与预设偏好阈值进行对比，大于预设偏好阈值的偏好值对应的潜在栖息地即为目标鱼类的偏好栖息地。
在获得偏好栖息地以后，可以对该目标鱼类的偏好栖息地进行修复，从而实现有针对性的修复，在降低成本的同时，达到最佳的修复效果，确保鱼类的生存环境。
本发明的第二方面提供了一种终端设备，包括存储器、处理器以及存储在存储器中并可在处理器上运行的计算机程序，处理器执行计算机程序时实现上述方法的步骤。
下面将以更为具体的实施例详述本发明的方法。
在该具体实施例中，使用基于主体的模型(Agent-based Model,ABM)进行仿真。其中，基于主体的模型的结构图如图2所示。该模型通常用于模拟自主主体的行为和交互，评估自主个体对整个系统的影响。ABM框 架中被模拟的主体能根据不同信息在特定环境中主动采取行动而不是被动地听取指令。在本实施例中该框架由个体和环境两个层次组成，一是以丹华水利开发的商用软件系统Mike 21/3 FM中的水动力(HD)和对流扩散(AD)模块作为基础，应用其计算得到水环境的参数信息，如水深、流速、水位、温度和盐度等参数，二是以鱼类感知半径范围内的环境变量、其他个体的位置和集群运动定义个体移动的方向和速度，得到粒子的状态变量、过程和计算式，并赋予粒子属性。
在模拟关键环境因子的变化，再将模拟结果以及鱼类生长、运动等函数关系作为输入，根据鱼对水环境因子的响应关系，在欧拉框架以网格为标尺建立的外部环境下，得到基于拉格朗日框架的代理粒子的运动轨迹。
然后根据模型计算出目标鱼类个体在下一时刻经过运动所到达的位置，模拟鱼的个体行为和生长状态，提出改进的鱼类个体行为仿真模型，并在此基础上建立粒子群，赋予每个粒子不同的属性，所有的个体通过上述的规则运动，从而获得整个河段鱼的空间分布随水环境条件的动态变化，实现研究区鱼群的生长、存活、生殖等行为的精确模拟，准确重现真实的鱼类个体行为和种群分布的信息交互。
上述基于主体的模型所处环境为ECO Lab提供的环境，该环境用于用户在三维高清模型Mike 21/3 FM上开发方程、实现和执行。在水动力(HD)和对流扩散(AD)模拟的基础上可以很好的进行水质各参数的模拟。表1为ABM模型模板参数设置及过程表达式描述。
表1 ABM模型模板(鱼类种群)参数设置
Figure PCTCN2022071244-appb-000017
Figure PCTCN2022071244-appb-000018
本实施例中，在ABM框架下利用拉格朗日算法构建目标鱼类的偏好区域确定模型，结合鱼类驯养实验中研究鱼类的视觉能力、听觉能力、嗅觉能力和相关产生的行为、游泳能力和方式，得到目标鱼类的偏好区域。在该模型中将目标鱼类以代理鱼粒子代替，代理鱼粒子追踪是以拉格朗日形式描述问题，使用牛顿运动定律求解常微分方程。一般情况下，粒子上的作用力可以分为两类，一类来源于外场，另一类来源于粒子相互作用。鱼类运动具有种群聚集效应，大多鱼类种群在其生活周期内存在阶段性的集群行为，从鱼苗开始直至成鱼均存在着追食行为。
代理鱼粒子在良好的饵料区以内食物浓度为第一选择，表现为觅食运 动；在良好的饵料区之外选择基本运动，表现为首先选择适宜温度，然后是适宜水深，最后是适宜流速。粒子个体的漂浮游动具有主动迁移和运动的特性，首先是不受限制的随机漫步；前进路线上出现障碍物则绕行，出现利好条件则被诱导；通过对感知范围内环境变量的判断，进行受限区域搜索，前进方向被诱导。
在鱼类个体的成长过程中，粒子能反映出幼鱼到成鱼不同阶段的习性，需要给鱼类个体设置不同的计算函数。赋予代理粒子随机规则、确定性方程和混合规则的运动函数，以及生长函数(size、weight)、索饵函数(fixed)、集群函数，以得到基于拉格朗日的鱼类个体动态模型。
本实施例中采用了基于新陈代谢理论的von Bertalanffy生长方程作为生长函数。
Figure PCTCN2022071244-appb-000019
式中，l t为所述目标鱼类t时刻的平均体长，W t为所述目标鱼类t时刻的平均体重，l ∞为所述目标鱼类的平均渐进体长，W ∞为所述目标鱼类的平均渐进体重，k为生长系数，t 0为假设的理论生长起点年龄。
采用下式作为所述目标鱼类体重生长模式的选择：
G＝Fixed*0.01*area
式中，fixed为饵料浓度，area为单位网格面积。
目标鱼类种群运动特征表现为同向、对齐、规避障碍等，集群运动的运动方向的确定公式为：
D i,t+1＝λ 1D i,t+λ 2D′ i,t+λ 3D″ i,t+λ 4D″′ i，t
式中，D i,t+1为t+1时刻第i个目标鱼类个体的运动方向，D i,t为t时刻第i个目标鱼类个体的运动方向，D′ i,t为t时刻第i个目标鱼类个体到临近个体平均位置的方向，D″ i,t为t时刻第i个目标鱼类个体的临近个体的平均方向，D″′ i,t为t时刻小于预设安全距离的临近个体到第i个目标鱼类个体方向的平均值(规避障碍)，考虑到对鱼的影响力不同，还需要对各个方向加权，取加权平均值，权重的大小可以根据偏好确定，λ 1,λ 2,λ 3,λ 4为权重， 且λ 1+λ 2+λ 3+λ 4＝1。
觅食速率判别式为：
eat＝if(fixed＜0.1,0,fixed::-(0.1*fixed))
式中，fixed为饵料浓度。
ABM模型计算模式下，代理的运动可以同时受到多种环境因子的影响，其中包括基于欧拉方法下流速、水深、水温、水质浓度等因子，也包括基于拉格朗日方法下其他代理与当前/目标代理之间的相互影响。每个时间步进行代理下一时间步空间位移的如果仅取决于其喜好流速及其临近的个体位置(如逃离其最近的个体)，则当前代理即将发生的运动速度和方向将是上述两个因子各自影响下的代理运动速度矢量的叠加结果。如图3所示。
在图3中五角星表示代理鱼粒子，即生物个体；所有网格含有流速信息；虚线圆圈为某一个代理鱼粒子(五角星)的感知范围；网格表示代理鱼粒子所能感知到的欧拉场环境因子，本例为流速，该流速中有代理鱼粒子最喜欢的流速(鱼类个体喜好流速一般有一定的取值范围)；虚线圈外的五角星表示处于当前代理鱼粒子感知能力范围外的其他代理鱼粒子。
当前时刻代理鱼粒子的运动速度为S t，运动方向为D t；两者通过当前时刻代理鱼粒子在搜索半径(虚线圈)内感知到的喜好流速所在网格相比于代理鱼粒子自身位置的方向D_fav(t)及朝向该方向运动的速度S_fav(t)与其在搜索半径(虚线圈)内感知到的离其最近的代理相比于自身位置的反向D_flee(t)及逃离该方向运动的速度S_flee(t)两组速度失量叠加计算得到当前时刻代理运动速度S(t)和运动方向D(t)。
假设代理鱼粒子个体i在特定时间步骤t的位置为
Figure PCTCN2022071244-appb-000020
其中x和y是笛卡尔坐标，时刻t+1的位置建模为：
Figure PCTCN2022071244-appb-000021
式中，
Figure PCTCN2022071244-appb-000022
为t时刻的第i个目标鱼类所在位置，
Figure PCTCN2022071244-appb-000023
为t+1时刻第i个目标鱼类所在位置，x和y为笛卡尔坐标系中的横轴和纵轴，S i,t为t时刻第i个目标鱼类个体的运动速度，D i,t为t时刻第i个目标鱼类个体的运动方向，θ i,t为t时刻第i个目标鱼类个体运动方向为D i,t时的夹角，
Figure PCTCN2022071244-appb-000024
为位移偏角；S i,t∈(0,S max)，S max为目标鱼类从t时刻至t+1时刻的时间段内的最大移动速度。
粒子在欧拉法计算的环境场中以鱼类不同阶段生活史特点为优先级，搜索鱼类偏好特征参数如最大饵料浓度、适宜温度、适宜流速等等，并在搜索参数中增加了另一粒子特征，拓展了鱼类行为模型中单一个体的模拟。最终通过ABM框架下欧拉-拉格朗日的信息交互。实现“外源环境-内源感知-鱼群运动决策”的模拟技术。
接下来进行目标鱼类偏好效应的模拟。
研究目标鱼类数量变动的目的是为了解鱼类资源的现状和预测其变动趋势，为水产捕捞、鱼类增殖及渔业资源管理理工科学依据。鱼类种群的数量变动是由捕捞强度变化、水温、河流水文条件等环境条件变化，饵料数量波动等多种因素造成的。模型的选用，参数的设计，各类生物学特征值，例如年龄、重量和生长函数的了解估算，在一定渔场的鱼苗资源量，以及环境因素对种群聚集效应、洄游与数量的影响等等。因此，本实施例对近岸的鱼苗繁衍行为、聚集效应及游泳游向一致性进行了模拟计算分析，便于后续新形成的栖息地的鱼类增殖性放流、鱼类产卵场保护范围划设提供一定的科学依据和支撑。
依靠大量的观察和资料积累，通过数字化和建模技术，利用计算机的 大容量和高速计算能力，模拟鱼群的动态，从相似度来探索现象的本质和主要因素，对各种条件下的趋势进行预测。代理粒子对于潜在栖息地j的偏好Pop j为：
Figure PCTCN2022071244-appb-000025
其中，P ij为第i个所述目标鱼类在第j个所述潜在栖息地中出现的累计密度，
Figure PCTCN2022071244-appb-000026
为所有潜在栖息地中出现的所述目标鱼类的累计总密度。
鱼苗集群行为是在鱼苗具有游泳能力后体现的，随着个体的长大，鱼苗集群形式不断变化，鱼苗集群使鱼苗具有生态学适应上的优越性，对其个体的生长和生存是有利的。鱼类成群游动可保存个体的能量，降低消耗，集群游泳时，带头鱼不断地交换，个体能量消耗比较均匀。鱼苗集群生活有安全感，夜晚集群休息则是表现之一，群体中的的鱼对邻伴游泳速度变化可迅速做出反应，能机动的同步逃逸。集群行为有助于鱼类的摄食及避开障碍物，浮游生物有时聚集成团，单尾鱼进入成团区域很慌乱，二集群鱼类进入这一区域由于游泳速度和游向具有一致性，利于捕食浮游生物，而且可以灵活的机动的包围浮游生物，有利于提高鱼苗的存活率。
刚孵化出的鱼苗无集群行为，集群行为是在具有游泳能力后逐渐显现出来的。从鱼苗的生存时间及游泳轨迹线可知，鱼苗群体生活区域在一定范围内，符合在岸边浅水缓流区域发现鱼苗聚集的现象，同时，鱼苗具有岸边边壁捕食苔藓等附着植物、浮游生物的特性，并体现出边壁障碍物的避开效应。
本发明的方法可以确定鱼类偏好的栖息地，从而进行有针对性的修复，在保证鱼类生存环境的同时，降低了修复栖息地的成本。
以上所述，仅是本申请的几个实施例，并非对本申请做任何形式的限制，虽然本申请以较佳实施例揭示如上，然而并非用以限制本申请，任何熟悉本专业的技术人员，在不脱离本申请技术方案的范围内，利用上述揭示的技术内容做出些许的变动或修饰均等同于等效实施案例，均属于技术方案范围内。

Claims (10)

  1. 一种鱼类偏好栖息地的确定方法，其特征在于，包括：
    获取目标鱼类所在水环境的参数信息，利用所述水环境的参数信息建立三维水环境模型；
    确定所述目标鱼类的生态学函数，结合所述生态学函数在所述三维水环境模型上搭建生物仿真模型；
    获取所述生物仿真模型中目标鱼类的运动轨迹，根据运动轨迹确定所述目标鱼类的潜在栖息地；
    利用基于累计密度法的偏好学习模型从所述潜在栖息地中确定所述目标鱼类的偏好栖息地。
  2. 根据权利要求1所述的方法，其特征在于，所述生态学函数包括生长函数、集群函数和觅食函数。
  3. 根据权利要求2所述的方法，其特征在于，所述生长函数根据第一公式确定，所述第一公式为：
    Figure PCTCN2022071244-appb-100001
    式中，l t为所述目标鱼类t时刻的平均体长，W t为所述目标鱼类t时刻的平均体重，l ∞为所述目标鱼类的平均渐进体长，W ∞为所述目标鱼类的平均渐进体重，k为生长系数，t 0为假设的理论生长起点年龄。
  4. 根据权利要求3所述的方法，其特征在于，所述集群函数根据第二公式确定，所述第二公式为：
    D i,t+1＝λ 1D i,t+λ 2D′ i,t+λ 3D″ i,t+λ 4D″′ i,t
    式中，D i,t+1为t+1时刻第i个目标鱼类个体的运动方向，D i,t为t时刻第i个目标鱼类个体的运动方向，D′ i,t为t时刻第i个目标鱼类个体到临近个体平均位置的方向，D″ i,t为t时刻第i个目标鱼类个体的临近个体的平均方向，D″′ i,t为t时刻小于预设安全距离的临近个体到第i个目标鱼类个体方向的平均值，λ 1,λ 2,λ 3,λ 4为权重，且λ 1+λ 2+λ 3+λ 4＝1。
  5. 根据权利要求4所述的方法，其特征在于，所述觅食函数根据第三公式确定，所述第三公式为：
    eat＝if(fixed＜0.1,0,fixed::-(0.1*fixed))
    式中，fixed为饵料浓度。
  6. 根据权利要求5所述的方法，其特征在于，获取所述生物仿真模型中目标鱼类的运动轨迹，具体为：
    根据第四公式获取所述生物仿真模型中目标鱼类的运动轨迹，所述第四公式为：
    Figure PCTCN2022071244-appb-100002
    式中，
    Figure PCTCN2022071244-appb-100003
    为t时刻的第i个所述目标鱼类所在位置，
    Figure PCTCN2022071244-appb-100004
    为t+1时刻第i个所述目标鱼类所在位置，x和y为笛卡尔坐标系中的横轴和纵轴，S i,t为t时刻第i个目标鱼类个体的运动速度，D i,t为t时刻第i个目标鱼类个体 的运动方向，θ i,t为t时刻第i个目标鱼类个体运动方向为D i,t时的夹角，
    Figure PCTCN2022071244-appb-100005
    为位移偏角；S i,t∈(0,S max)，S max为所述目标鱼类从t时刻至t+1时刻的时间段内的最大移动速度。
  7. 根据权利要求6所述的方法，其特征在于，目标鱼类个体的运动速度和运动方向根据第五公式确定，所述第五公式为：
    Figure PCTCN2022071244-appb-100006
    式中，S i,t为t时刻第i个目标鱼类个体的运动速度，D i,t为t时刻第i个目标鱼类个体的运动方向，D_fav(t)和S_fav(t)分别为t时刻第i个目标鱼类个体在其感知范围内的喜好流速所在位置相比于当前位置的运动方向及朝向该运动方向的运动速度；D_flee(t)和S_flee(t)分别为t时刻第i个目标鱼类个体逃离其感知范围内最近的临近个体的运动方向及逃离该运动方向的运动速度。
  8. 根据权利要求7所述的方法，其特征在于，感知范围的确定方法为：
    获取所述目标鱼类的视觉信息、听觉信息和嗅觉信息；
    根据所述视觉信息、听觉信息和嗅觉信息，确定所述目标鱼类的感知范围。
  9. 根据权利要求1-8任一项所述的方法，其特征在于，利用基于累计密度法的偏好学习模型从所述潜在栖息地中确定所述目标鱼类的偏好栖息地，具体为：
    利用基于累计密度法的偏好学习模型，获取第i个目标鱼类对每个所述潜在栖息地的偏好值；所述基于累计密度法的偏好学习模型根据第六公式确定，所述第六公式为：
    Figure PCTCN2022071244-appb-100007
    式中，P ij为第i个目标鱼类在第j个所述潜在栖息地中出现的累计密度，记为偏好值，
    Figure PCTCN2022071244-appb-100008
    为所有潜在栖息地中出现所述目标鱼类的总累计密度；
    根据所述偏好值确定所述目标鱼类的偏好栖息地。
  10. 一种终端设备，包括存储器、处理器以及存储在所述存储器中并可在所述处理器上运行的计算机程序，其特征是，所述处理器执行所述计算机程序时实现如权利要求1至9任一项所述方法的步骤。
PCT/CN2022/071244 2021-09-23 2022-01-11 一种鱼类偏好栖息地的确定方法及终端设备 WO2022170901A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/750,027 US11645354B2 (en) 2021-09-23 2022-05-20 Determination method for preferred habitat of fish and terminal device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111110237.1A CN113807017B (zh) 2021-09-23 2021-09-23 一种鱼类偏好栖息地的确定方法及终端设备
CN202111110237.1 2021-09-23

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/750,027 Continuation US11645354B2 (en) 2021-09-23 2022-05-20 Determination method for preferred habitat of fish and terminal device

Publications (1)

Publication Number Publication Date
WO2022170901A1 true WO2022170901A1 (zh) 2022-08-18

Family

ID=78896172

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/071244 WO2022170901A1 (zh) 2021-09-23 2022-01-11 一种鱼类偏好栖息地的确定方法及终端设备

Country Status (3)

Country Link
US (1) US11645354B2 (zh)
CN (1) CN113807017B (zh)
WO (1) WO2022170901A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115114873A (zh) * 2022-08-31 2022-09-27 中国海洋大学 海洋环境重现期设计标准推算方法与系统
CN117892980A (zh) * 2024-03-14 2024-04-16 长江水资源保护科学研究所 一种针对圆口铜鱼的生态调度方法和装置

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113807017B (zh) * 2021-09-23 2022-05-24 西安理工大学 一种鱼类偏好栖息地的确定方法及终端设备
US20230135897A1 (en) * 2021-10-28 2023-05-04 Softbank Corp. Information processing method, non-transitory computer-readable recording medium, and information processor
US20240037917A1 (en) * 2022-07-28 2024-02-01 Softbank Corp. Information processing method, non-transitory computer-readable storage medium, and information processing device
CN116432902B (zh) * 2023-03-31 2023-09-12 中国水利水电科学研究院 一种考虑水环境因子变异的物种长期生存能力评估方法
CN116720755B (zh) * 2023-08-10 2024-02-20 中国水产科学研究院南海水产研究所 一种基于大数据的渔业资源评价方法及系统
CN117612355B (zh) * 2023-11-23 2024-05-31 国家海洋局南海调查技术中心(国家海洋局南海浮标中心) 一种基于物联网的浮潜标防碰撞破坏系统及方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109271694A (zh) * 2018-09-06 2019-01-25 西安理工大学 基于鱼类个体动态模拟技术的栖息地识别方法
AU2021100058A4 (en) * 2021-01-06 2021-03-25 Shanghai Ocean University Technical Method for Evaluating and Predicting Habitat Suitability for Scomber Japonicus in East China Sea Based on Water Temperature at Different Depths
CN113807017A (zh) * 2021-09-23 2021-12-17 西安理工大学 一种鱼类偏好栖息地的确定方法及终端设备

Family Cites Families (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS5788021A (en) 1980-11-21 1982-06-01 Fuji:Kk Method and apparatus for manufacturing sodium silicate
US6160759A (en) * 1999-04-19 2000-12-12 Nestler; John Michael Method for determining probable response of aquatic species to selected components of water flow fields
US7681531B2 (en) * 2006-02-21 2010-03-23 O'neil Thomas System for assessing habitat value
JP2008185455A (ja) * 2007-01-30 2008-08-14 Institute Of Freshwater Biology 水域環境の評価方法
US8731748B2 (en) * 2010-02-26 2014-05-20 Strategic Fishing Systems, Llc Predictive mapping system for anglers
JP2013116092A (ja) * 2011-12-05 2013-06-13 Japan Aerospace Exploration Agency 魚類観測装置および方法
WO2017156443A1 (en) * 2016-03-10 2017-09-14 Rutgers, The State University Of New Jersey Global optimization-based method for improving human crowd trajectory estimation and tracking
CN107480884B (zh) * 2017-08-14 2019-10-22 水利部交通运输部国家能源局南京水利科学研究院 一种多闸坝平原河流冬季生态流量推求方法
JP6401411B1 (ja) * 2018-02-13 2018-10-10 株式会社Ａｉハヤブサ 人工知能による漁獲物識別システム、管理システム及び物流システム
CN109657940A (zh) * 2018-12-05 2019-04-19 北京师范大学 基于鱼类响应确定栖息地生态系统恢复因子优先级的方法
CN109615076A (zh) * 2018-12-13 2019-04-12 水利部交通运输部国家能源局南京水利科学研究院 一种面向鱼类生境保护的河流生态流量过程推求方法
WO2021014339A1 (en) * 2019-07-25 2021-01-28 Vasudeva Keshav A system for removal of solid waste from water bodies
CN112215116B (zh) * 2020-09-30 2022-06-24 江苏大学 一种移动式面向2d图像的3d河蟹实时检测方法
AU2020103130A4 (en) * 2020-10-30 2021-01-07 Xi’an University of Technology Habitat Identification Method Based on Fish Individual Dynamic Simulation Technology

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109271694A (zh) * 2018-09-06 2019-01-25 西安理工大学 基于鱼类个体动态模拟技术的栖息地识别方法
AU2021100058A4 (en) * 2021-01-06 2021-03-25 Shanghai Ocean University Technical Method for Evaluating and Predicting Habitat Suitability for Scomber Japonicus in East China Sea Based on Water Temperature at Different Depths
CN113807017A (zh) * 2021-09-23 2021-12-17 西安理工大学 一种鱼类偏好栖息地的确定方法及终端设备

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
DAI QIANG, GU HAI-JUN, WANG YUE-ZHAO: "Theories and Models for Habitat Selection", DONGWUXUE YANJIU - ZOOLOGICAL RESEARCH, YUNNAN RENMIN CHUBANSHE, KUNMING, CN, vol. 28, no. 6, 1 December 2007 (2007-12-01), CN , pages 681 - 688, XP055958827, ISSN: 0254-5853 *
GOODWIN, R.A. ; NESTLER, J.M. ; ANDERSON, J.J. ; WEBER, L.J. ; LOUCKS, D.P.: "Forecasting 3-D fish movement behavior using a Eulerian-Lagrangian-agent method (ELAM)", ECOLOGICAL MODELLING, ELSEVIER, AMSTERDAM, NL, vol. 192, no. 1-2, 15 February 2006 (2006-02-15), AMSTERDAM, NL, pages 197 - 223, XP027883130, ISSN: 0304-3800 *
QUAN QUAN, GAO SHAOZE, SHANG YANWU, WANG BOXING: "Assessment of the sustainability of Gymnocypris eckloni habitat under river damming in the source region of the Yellow River", SCIENCE OF THE TOTAL ENVIRONMENT, ELSEVIER, AMSTERDAM, NL, vol. 778, 1 July 2021 (2021-07-01), AMSTERDAM, NL , pages 146312, XP055958825, ISSN: 0048-9697, DOI: 10.1016/j.scitotenv.2021.146312 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115114873A (zh) * 2022-08-31 2022-09-27 中国海洋大学 海洋环境重现期设计标准推算方法与系统
CN117892980A (zh) * 2024-03-14 2024-04-16 长江水资源保护科学研究所 一种针对圆口铜鱼的生态调度方法和装置
CN117892980B (zh) * 2024-03-14 2024-05-24 长江水资源保护科学研究所 一种针对圆口铜鱼的生态调度方法和装置

Also Published As

Publication number Publication date
CN113807017B (zh) 2022-05-24
US20230102365A1 (en) 2023-03-30
US11645354B2 (en) 2023-05-09
CN113807017A (zh) 2021-12-17

Similar Documents

Publication Publication Date Title
WO2022170901A1 (zh) 一种鱼类偏好栖息地的确定方法及终端设备
AU2020103130A4 (en) Habitat Identification Method Based on Fish Individual Dynamic Simulation Technology
CN109271694B (zh) 基于鱼类个体动态模拟技术的栖息地识别方法
North et al. Manual of recommended practices for modelling physical–biological interactions during fish early life.
Claereboudt Fertilization success in spatially distributed populations of benthic free-spawners: a simulation model
Faillettaz et al. Swimming speeds of Mediterranean settlement‐stage fish larvae nuance Hjort's aberrant drift hypothesis
Scott et al. Seabirds and marine renewables: Are we asking the right questions?
Henry et al. Multi-scale interactions between local hydrography, seabed topography, and community assembly on cold-water coral reefs
Donahue et al. Location isn’t everything: timing of spawning aggregations optimizes larval replenishment
CN110933681B (zh) 一种鬣狗群捕食算法及其在节点部署优化中应用的方法
Kerr et al. Assessing hydrodynamic space use of brown trout, Salmo trutta, in a complex flow environment: a return to first principles
CN111802287A (zh) 基于漂流性鱼卵胚体发育时间的产卵场定位反演模拟方法
Dueri et al. Modelling the effect of marine protected areas on the population of skipjack tuna in the Indian Ocean
Mortensen et al. Agent‐based models to investigate sound impact on marine animals: bridging the gap between effects on individual behaviour and population level consequences
Broekhuizen et al. Dispersal of oyster (Ostrea chilensis) larvae in Tasman Bay inferred using a verified particle tracking model that incorporates larval behavior
Pedersen et al. Larval dispersal and mother populations of Pandalus borealis investigated by a Lagrangian particle-tracking model
Ayllón et al. Intercohort density dependence drives brown trout habitat selection
Ghezzo et al. Natural resources and climate change: A study of the potential impact on Manila clam in the Venice lagoon
Watkins et al. Simulating individual-based movement in dynamic environments
McDonald et al. Multiple-use management strategy evaluation for coastal marine ecosystems using InVitro
Letschert et al. Socio-ecological drivers of demersal fishing activity in the North Sea: the case of three German fleets
Ishiwaka et al. DeepFoids: Adaptive Bio-Inspired Fish Simulation with Deep Reinforcement Learning.
Kim et al. Basic modelling of fish behaviour in a towed trawl based on chaos in decision-making
Moláček et al. The fine art of surfacing: its efficacy in broadcast spawning
Lett et al. Modelling the dynamics of animal groups in motion: biological modelling

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22752063

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE