CN117495061B - An intelligent marine ranch feeding system based on the Internet of Things - Google Patents
An intelligent marine ranch feeding system based on the Internet of Things Download PDFInfo
- Publication number
- CN117495061B CN117495061B CN202410001445.5A CN202410001445A CN117495061B CN 117495061 B CN117495061 B CN 117495061B CN 202410001445 A CN202410001445 A CN 202410001445A CN 117495061 B CN117495061 B CN 117495061B
- Authority
- CN
- China
- Prior art keywords
- feeding
- data
- robot
- target
- feed
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 241000251468 Actinopterygii Species 0.000 claims abstract description 92
- 239000000446 fuel Substances 0.000 claims abstract description 60
- 238000004458 analytical method Methods 0.000 claims abstract description 21
- 239000002245 particle Substances 0.000 claims description 75
- 210000000349 chromosome Anatomy 0.000 claims description 55
- 230000002068 genetic effect Effects 0.000 claims description 25
- 230000009471 action Effects 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 33
- 230000008901 benefit Effects 0.000 abstract description 7
- 230000006870 function Effects 0.000 description 35
- 230000008569 process Effects 0.000 description 21
- 238000005457 optimization Methods 0.000 description 19
- 238000012544 monitoring process Methods 0.000 description 7
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 6
- 230000003044 adaptive effect Effects 0.000 description 4
- 239000002699 waste material Substances 0.000 description 4
- 238000009395 breeding Methods 0.000 description 3
- 230000001488 breeding effect Effects 0.000 description 3
- 238000009313 farming Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000002592 echocardiography Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000009396 hybridization Methods 0.000 description 2
- 230000013016 learning Effects 0.000 description 2
- 230000003278 mimic effect Effects 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 230000002028 premature Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000000750 progressive effect Effects 0.000 description 2
- 108090000623 proteins and genes Proteins 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000009360 aquaculture Methods 0.000 description 1
- 244000144974 aquaculture Species 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006353 environmental stress Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000003031 feeding effect Effects 0.000 description 1
- 239000012634 fragment Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000011065 in-situ storage Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000005541 medical transmission Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 239000013618 particulate matter Substances 0.000 description 1
- 102000004169 proteins and genes Human genes 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 230000011273 social behavior Effects 0.000 description 1
- 230000009326 social learning Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01K—ANIMAL HUSBANDRY; AVICULTURE; APICULTURE; PISCICULTURE; FISHING; REARING OR BREEDING ANIMALS, NOT OTHERWISE PROVIDED FOR; NEW BREEDS OF ANIMALS
- A01K61/00—Culture of aquatic animals
- A01K61/80—Feeding devices
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Y—INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT]
- G16Y10/00—Economic sectors
- G16Y10/15—Fishing
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Marine Sciences & Fisheries (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- General Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Environmental Sciences (AREA)
- Animal Husbandry (AREA)
- Computing Systems (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Educational Administration (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Mining & Mineral Resources (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Primary Health Care (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Accounting & Taxation (AREA)
- Zoology (AREA)
- Biodiversity & Conservation Biology (AREA)
- Agronomy & Crop Science (AREA)
Abstract
The invention relates to an intelligent marine pasture feeding system based on the Internet of things, which comprises a plurality of feeding robots and a central control system, wherein the central control system acquires position data, feed allowance data, fuel allowance data and fish school density data through a data acquisition module, obtains a target feeding position through a position analysis module, obtains a dispatching plan through a dispatching analysis module, and controls the plurality of feeding robots to move and feed through a dispatching control module. Compared with the prior art, the method and the device have the advantages that the position analysis module analyzes the movement direction of the fish shoal by utilizing the distribution condition of the position data and the fish shoal density data so as to obtain the optimal target feeding position, maximize the utilization rate of feed, and the scheduling analysis module is used for scheduling a plurality of feeding robots to feed to the target feeding position by combining the residual quantity of the feed and the residual quantity of fuel so as to maximize the utilization rate of fuel, and the two points are combined so that the economic efficiency is maximized when feeding work is performed.
Description
Technical Field
The invention relates to the technical field of intelligent ocean pastures, in particular to an intelligent ocean pasture feeding system based on the Internet of things.
Background
Marine ranches are an innovative agricultural way to grow fish in the ocean, aimed at meeting the ever-increasing market demand for fish. With the increasing population and progressive depletion of fishery resources, marine rangelands are considered as one of the important solutions for sustainable development. And the application of the intelligent technology is imperative to realize the efficient operation and management of the marine pasture. The feeding robot is used as key equipment of an intelligent marine pasture and has an automatic feeding function. However, although presently existing automated feeding robotics exist, they generally consider only a single influencing factor during feeding and do not take into account economic maximization as a priority.
On the one hand, unlike other scheduling problems, shoals of fish in the ocean exhibit dynamic characteristics. Once the feeding starts, the shoal of fish often gathers to the feeding point and robs food, and this action can lead to traditional fixed-point feeding mode to cause the waste of fodder, probably even causes the pollution to marine environment. On the other hand, if the robot can freely move, the fuel utilization rate of the robot itself is also a problem to be considered, and it is necessary to minimize the amount of fuel used and to improve the feeding efficiency.
Therefore, there is a need for a marine ranch feeding system that takes into account the above factors, maximizing the economic benefits of the feeding process.
Disclosure of Invention
Therefore, the invention provides an intelligent marine pasture feeding system based on the Internet of things, which is used for solving the problem of how to maximize economic benefits in the feeding process in the prior art.
The invention provides an intelligent marine pasture feeding system based on the Internet of things, which comprises a plurality of feeding robots and a central control system, wherein the feeding robots are in communication connection with the central control system, the plurality of feeding robots are respectively used for cruising in a plurality of preset areas in the marine pasture, and carry feed for feeding, fuel for driving and a detection device for detecting the density of fish shoals; the central control system includes:
the data acquisition module is used for acquiring the position data, the feed allowance data and the fuel allowance data of each feeding robot based on the feeding robots and the fish school density data obtained by detecting each feeding robot under the corresponding position data;
the position analysis module is used for obtaining a target feeding position according to the distribution condition of the position data and the shoal density data;
The scheduling analysis module is used for obtaining scheduling plans of a plurality of feeding robots based on the target feeding positions and according to the position data, the feed allowance data, the fuel allowance data and the fish school density data of each feeding robot;
and the dispatching control module is used for controlling the plurality of feeding robots to move and feed according to the dispatching plan.
In a preferred embodiment, the obtaining the target feeding position according to the distribution of the position data and the fish school density data includes:
obtaining the number of feeding points according to the distribution condition of the position data and the fish school density data;
establishing an initial particle swarm based on the number of feeding points, wherein the initial particle swarm comprises a plurality of initial particles, each initial particle comprises initial coordinates in a plurality of marine pastures, and the number of the initial coordinates in each initial particle is equal to the number of the feeding points;
and constructing a first fitness function based on the difficulty degree of gathering the fish shoal to a plurality of feeding points, optimizing an initial particle swarm through a particle swarm algorithm based on the first fitness function to obtain optimal particles comprising a plurality of optimal coordinates, and taking each optimal coordinate as a target feeding position.
In a preferred embodiment, the obtaining the number of feeding points according to the distribution of the position data and the fish school density data includes:
Establishing a plurality of three-dimensional data according to each position data and the corresponding fish school density data;
fitting a plurality of three-dimensional data to obtain a fitted surface function, wherein the total independent variable of the fitted surface function is the coordinates in the two-dimensional ocean pasture, and the independent variable is the fish swarm density;
solving a maximum value in the fitting curved surface function, and comparing the maximum value with a preset threshold value;
if the maximum value number exceeding the preset threshold exceeds the preset number, taking the maximum value number exceeding the preset threshold as the feeding point number;
if the maximum number exceeding the preset threshold value does not exceed the preset number, taking a preset value as the number of feeding points.
In a preferred embodiment, the establishing the initial particle swarm based on the number of feeding points comprises:
acquiring coordinates in the ocean pasture corresponding to each maximum value, wherein the coordinates are used as reference coordinates;
randomly selecting one coordinate from a preset adjacent area of each reference coordinate to obtain a plurality of initial coordinates and establishing an initial particle;
and repeatedly establishing a plurality of initial particles to obtain an initial particle group.
In a preferred embodiment, the first fitness function comprises:
；
wherein,for the particle to be analyzed currently in the particle swarm algorithm +. >Is adaptive to->Is particle->Middle->Coordinates of->For the number of feeding points, +.>For being located in coordinates->Is the center of a circle>Is the sum of the fish school density data corresponding to the position data within the radius range, +.>The value of (2) is a preset value, +.>As an exponential function +.>And->Respectively different adjustment coefficients.
In a preferred embodiment, the obtaining a scheduling plan of a plurality of feeding robots based on the target feeding position according to the position data, the feed allowance data, the fuel allowance data and the fish school density data of each feeding robot includes:
obtaining estimated feeding quantity of each target feeding position based on the fish swarm density data corresponding to each position data according to the position relation between the position data and the target feeding position;
establishing a plurality of initial chromosomes based on the corresponding relation from the feeding robot to the target feeding position to obtain an initial population;
according to the feed allowance data, the fuel allowance data and the estimated feeding amount, a second fitness function is constructed by taking the fuel consumption cost of the feeding robot running to the corresponding target feeding position and the utilization rate of the feeding robot feeding feed as well as optimizing the initial population through a genetic algorithm based on the second fitness function, so that an optimal chromosome representing the optimal corresponding relation between the feeding robot and the target feeding position is obtained, and a dispatching plan is obtained.
In a preferred embodiment, the estimated feed amount for each target feed location is obtained according to the following equation:
；
wherein,is->Individual target feeding position +.>Is a predicted feeding amount of->For being located at the targeted feeding position +.>Is the center of a circle, ">Is the sum of the fish school density data corresponding to the position data within the radius range, +.>The coefficients are adjusted for the first unit.
In a preferred embodiment, the chromosome in the genetic algorithm is a one-dimensional vector, each element position in the chromosome represents a feeding robot respectively, each element value is used for representing that the feeding robot corresponding to the element position where the element value is located needs to move to a target feeding position corresponding to the element value to perform feeding or the feeding robot does not participate in feeding, the value of each element value comprises a plurality of first numbers and second numbers, the plurality of first numbers represent the plurality of target feeding positions respectively, and the second number represents that the feeding robot does not participate in feeding;
the genetic algorithm comprises: obtaining a offspring chromosome according to the parent chromosome based on a preset rule;
wherein, the preset rule comprises:
parent chromosomes do not cross to obtain offspring chromosomes;
The probability that the element value in the parent chromosome is changed to the second number is related to the feed allowance data and the fuel allowance data of the feeding robot corresponding to the element value.
In a preferred embodiment, the probability of an element value variation in the parent chromosome being the second number is obtained by:
；
wherein,is chromosome->Middle->Probability of variation of the element value to the second number,/->Chromosome->Middle->Feed allowance data of feeding robot corresponding to individual element values, < >>Is chromosome->Middle->Fuel balance data of feeding robot corresponding to individual element values, < >>The coefficients are adjusted for the second unit.
In a preferred embodiment, the second fitness function comprises:
；
；
wherein,is chromosome +.>Is adaptive to->Is chromosome->Corresponding target feeding position +.>The sum of the feed margins of the feeding robot corresponding to the element values, < >>For the preset difference, ++>Is chromosome->Middle->Element value->For characterizing element values +.>The corresponding feeding robot can reach the target feeding position +.>The Boolean value of 1 indicating that it is possible to achieve, the Boolean value of 0 indicating that it is not possible to achieve, the Boolean value being based on the element value +. >Feed allowance data of corresponding feeding robot +.>And element value->Fuel balance data +.>Obtained (I)>Is chromosome->Corresponding target feeding position +.>Feeding robot corresponding to element value of (2) and target feeding position +.>Distance sum of>Is chromosome->Corresponding target feeding position +.>The utilization rate of all the feeds fed by the feeding robot corresponding to the element values, < + >>、/>、/>Respectively different weight coefficients, +.>Is a slope adjustment coefficient.
The beneficial effects of adopting the embodiment are as follows:
the invention provides an intelligent marine pasture feeding system based on the Internet of things, which comprises a plurality of feeding robots and a central control system, wherein the plurality of feeding robots respectively cruises in the marine pasture, the central control system acquires position data, feed allowance data, fuel allowance data and fish school density data based on the feeding robots through a data acquisition module, obtains a target feeding position through a position analysis module according to the distribution condition of the position data and the fish school density data, obtains a scheduling plan of the plurality of feeding robots based on the target feeding position, the position data, the feed allowance data, the fuel allowance data and the fish school density data through a scheduling analysis module, and controls the plurality of feeding robots to move and feed through a scheduling control module. Compared with the prior art, the invention analyzes the direction of the fish shoal by utilizing the distribution condition of the position data and the fish shoal density data through the position analysis module so as to obtain the optimal target feeding position and maximize the utilization rate of the feed, and on the other hand, the scheduling analysis module is used for scheduling a plurality of feeding robots to feed to the target feeding position by combining the residual quantity of the feed and the residual quantity of the fuel so as to maximize the utilization rate of the fuel.
Drawings
FIG. 1 is a system architecture diagram of an embodiment of an intelligent marine ranch feeding system based on the Internet of things provided by the invention;
FIG. 2 is a flow chart of a method for operating a location analysis module according to an embodiment of the present invention;
FIG. 3 is a flow chart of a method for operating a dispatch analysis module according to one embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, in one specific embodiment of the invention, an intelligent marine pasture feeding system based on the internet of things is disclosed, which comprises a plurality of feeding robots 110 and a central control system 120, wherein the feeding robots are in communication connection with the central control system and are respectively used for cruising in a plurality of preset areas in the marine pasture, and each feeding robot carries feed for feeding, fuel for driving and a detection device for detecting the density of fish shoals;
The central control system comprises:
a data acquisition module 121, configured to acquire, based on the feeding robots, position data, feed allowance data, and fuel allowance data of each feeding robot, and fish school density data detected by each feeding robot under the corresponding position data;
the position analysis module 122 is configured to obtain a target feeding position according to the distribution situation of the position data and the shoal density data;
the scheduling analysis module 123 is configured to obtain a scheduling plan of a plurality of feeding robots based on the target feeding position and according to the position data, the feed allowance data, the fuel allowance data and the fish school density data of each feeding robot;
the dispatch control module 124 is configured to control the plurality of feeding robots to move and feed according to a dispatch plan.
Compared with the prior art, the invention analyzes the direction of the fish shoal by utilizing the distribution condition of the position data and the fish shoal density data through the position analysis module so as to obtain the optimal target feeding position and maximize the utilization rate of the feed, and on the other hand, the scheduling analysis module is used for scheduling a plurality of feeding robots to feed to the target feeding position by combining the residual quantity of the feed and the residual quantity of the fuel so as to maximize the utilization rate of the fuel.
In this embodiment, the data acquisition module 121 is only used as a module for receiving data, in fact, the position data, the feed allowance data, the fuel allowance data and the fish school density data are all collected by a feeding robot, the feeding robot can be implemented by existing equipment such as an unmanned ship, the position data can be obtained by a positioning module (such as a GPS) on the feeding robot, the feed allowance data and the fuel allowance data can be obtained by sensors in corresponding containers on the feeding robot, and the fish school density data can be obtained by a camera, a sonar, etc., for example:
visual monitoring: visual monitoring is performed using equipment such as an underwater camera mounted to the feeding robot. By observing the number and distribution of fish shoals in a particular spatial region, the density of the fish shoals can be estimated initially.
Sonar monitoring: and (3) sending sound waves into water by using sonar equipment installed on the feeding robot and receiving echoes, and estimating the density of the fish shoal by analyzing the characteristics of the echoes and the signal intensity. The sonar device can measure acoustic reflection and scattering of the fish shoal to estimate the fish shoal's distribution and density information.
The water quality monitoring method comprises the following steps: the water quality monitoring instrument arranged on the feeding robot is used for measuring indexes such as suspended particulate matter concentration, oxygen content and the like in the water sample, so that the density of the fish shoal is indirectly estimated. An increase in the concentration of particulates in the water may be indicative of the presence of a fish school.
Of course, the above manners may also be combined, and the manner of specifically obtaining the fish school density is the prior art, so that the description thereof will not be repeated here.
It should be noted that different monitoring methods may be suitable for different shoal density ranges and water environments. In order to obtain more accurate and reliable fish school density data, it is often necessary to combine multiple monitoring methods and to analyze and process the data according to specific farming environments and research requirements.
Further, as shown in connection with FIG. 2, in a preferred embodiment, the location analysis module 122 performs the functions of: according to the distribution condition of the position data and the fish school density data, a target feeding position is obtained, and the method specifically comprises the following steps:
s201, obtaining the number of feeding points according to the distribution situation of the position data and the shoal density data;
s202, establishing an initial particle swarm based on the number of feeding points, wherein the initial particle swarm comprises a plurality of initial particles, each initial particle comprises initial coordinates in a plurality of ocean pastures, and the number of the initial coordinates in each initial particle is equal to the number of the feeding points;
s203, constructing a first fitness function based on the difficulty degree of gathering the fish shoal to a plurality of feeding points, optimizing an initial particle swarm through a particle swarm algorithm based on the first fitness function to obtain optimal particles comprising a plurality of optimal coordinates, and taking each optimal coordinate as a target feeding position.
The process obtains the most suitable feeding point quantity considering the direction of the fish shoal through the position data and the fish shoal density, and further obtains a plurality of optimal target feeding positions through a particle swarm optimization algorithm so as to avoid the condition that the fish shoals are excessively gathered to a certain feeding point to cause excessive feeding of another feeding point. This strategy allows the fish farm manager to reasonably allocate resources, avoid wastage, and ensure that the fish shoal gets food evenly, thereby improving overall farming efficiency. By constructing a proper first fitness function and utilizing a particle swarm optimization algorithm, the position of the feeding point can be finely adjusted effectively, and the optimal position can be found. Such algorithm optimization not only improves feeding efficiency, but also reduces resource competition and environmental stress that may be caused by excessive concentration of fish shoals. In addition, the scheme can effectively prevent the dense gathering of the fish shoals at specific feeding points. This is beneficial to reducing the risk of disease transmission and also to the balanced growth of the fish population, thereby improving the health level of the entire fish population.
It is understood that the initial particles, the optimal particles, etc. in the above process are one of the particles in the particle swarm optimization algorithm, and the particles have the same structure, and for these different prefixes or suffixes, those skilled in the art will understand the meaning, and therefore, they will not be described in any great detail herein.
Further, in a preferred embodiment, the step S201 obtains the number of feeding points according to the distribution of the position data and the fish school density data, which specifically includes:
establishing a plurality of three-dimensional data according to each position data and the corresponding fish school density data;
fitting a plurality of three-dimensional data to obtain a fitted surface function, wherein the total independent variable of the fitted surface function is the coordinates in the two-dimensional ocean pasture, and the independent variable is the fish swarm density;
solving a maximum value in the fitting curved surface function, and comparing the maximum value with a preset threshold value;
if the maximum value number exceeding the preset threshold exceeds the preset number, taking the maximum value number exceeding the preset threshold as the feeding point number;
if the maximum number exceeding the preset threshold value does not exceed the preset number, taking a preset value as the number of feeding points.
The density distribution of the fish shoal is intuitively obtained in a curved surface fitting mode in the process, namely, the position data can correspond to coordinates of an x axis and a y axis in a three-dimensional coordinate system, the fish shoal density can be regarded as data of a z axis, and in a graph of a fitting curved surface function obtained in the fitting process, the distribution of the fish shoal density can be intuitively seen through the height in the z axis direction. In this embodiment, the movement of the fish farm is indirectly analyzed by the extremum, so that feeding is reasonably performed according to the movement capability of the fish farm.
For example, if the number of maxima exceeding the preset threshold exceeds the preset number, the current uneven distribution of the fish shoals in the ocean pasture can be represented, and at the moment, the number of feeding points is designated by taking a plurality of maxima as the standard, so that the fish shoals can be fed in places where the fish shoals are dense and are easy to gather, and the feed utilization rate is improved. Conversely, if the maximum number exceeding the preset threshold value does not exceed the preset number, the current fish shoals are uniformly distributed in the ocean pasture, and then feeding can be performed based on a preset general scheme, for example, average feeding is performed at a preset number of feeding points in the ocean pasture, so that the fish shoals are prevented from being excessively gathered in a certain place due to feeding, and the feed waste in other positions is avoided.
After the number of feeding points is obtained, in this embodiment, a particle swarm optimization algorithm is used to further obtain each specific target feeding position. Particle Swarm Optimization (PSO) is an optimization tool that simulates social behavior. In aquaculture, particularly when determining a target feeding position, the particle swarm algorithm regards each scheme as an individual, and the optimal scheme is obtained by simulating the behavior of birds or fish to find food through the individual. Specifically, each particle represents a potential feeding position plan. Each particle has its own velocity and direction, which will move within the search space while updating its own position with reference to its own historical optimal position (individual learning) and the historical optimal position of the whole particle population (social learning). After the most suitable number of feeding points is obtained, the particle swarm algorithm optimizes the specific position of each point by evaluating each particle. The evaluation criteria (fitness function) are typically based on maximizing resource utilization efficiency and minimizing feeding costs. The particle swarm algorithm pursues that the whole system is optimized, in the embodiment, the particle swarm algorithm enables all feeding points to be located at the most favorable positions for the shoal, so that the shoal is convenient to gather, and meanwhile waste or deficiency of feed after gathering is avoided.
The particle swarm algorithm is relatively simple to realize and has high convergence speed. Meanwhile, different specific conditions, such as inertia weight or learning factors of particles, can be adapted by adjusting algorithm parameters. In addition, the particle swarm optimization algorithm can maintain a plurality of potential solutions in the global search process, so that the probability of finding a globally optimal solution is increased. The particle swarm optimization algorithm is utilized to carry out strategy optimization, so that the overall optimization of the strategy can be ensured, the complex and changeable environment is adapted, and the method is important to improving the feeding accuracy and the resource utilization efficiency. Through continuous optimization iteration, the particle swarm algorithm identifies and locks the optimal feeding position, and the efficiency and productivity of the breeding industry are improved.
However, the particle swarm algorithm may have a locally optimal problem, and to solve the problem, the present invention further provides a preferred embodiment, wherein the establishing the initial particle swarm based on the number of feeding points specifically includes:
acquiring coordinates in the ocean pasture corresponding to each maximum value, wherein the coordinates are used as reference coordinates;
randomly selecting one coordinate from a preset adjacent area of each reference coordinate to obtain a plurality of initial coordinates and establishing an initial particle;
and repeatedly establishing a plurality of initial particles to obtain an initial particle group.
The maximum value of the number of the feeding points is analyzed before, and is obtained through fitting a curved surface, although the maximum value obtained through the maximum value is still capable of reflecting the distribution condition of the fish shoal to a certain extent when the fitted curved surface function is compared with the actual condition, the initial coordinates are selected near the maximum value to establish the initial particle swarm, so that the initial data of the particle swarm algorithm is enabled to be more approximate to an optimal solution, the particle swarm optimization algorithm starts to optimize near the optimal solution, optimization time is shortened, optimization accuracy is improved, and the phenomenon of overfitting possibly caused by the traditional random generation of the initial particle swarm is avoided.
It is to be understood that other processes and details in the particle swarm algorithm, other than those mentioned above, are of the prior art that those skilled in the art will understand and implement, and thus, are not described in great detail herein.
Further, in a preferred embodiment, the first fitness function includes:
；
wherein,for the particle to be analyzed currently in the particle swarm algorithm +.>Is adaptive to->Is particle->Middle->Coordinates of->For the number of feeding points, +.>For being located in coordinates- >Is the center of a circle>Is the sum of the fish school density data corresponding to the position data within the radius range, +.>The value of (2) is a preset value, +.>As an exponential function +.>And->Respectively different adjustment coefficients.
Above-mentionedThe function is essentially a summation of the fish school density data reflecting the coordinates +.>The number of fish shoals in the vicinity, however, in practice the sea area is continuousIn the region, the number of fish shoal in the sea can be accurately represented by not stacking the numerical values of a plurality of discrete coordinate points, and fish exist in the region between the plurality of coordinate points, so that the number of fish shoal is far more than the level of the number of fish shoal reflected by summing the density data of the fish shoal in practice. Therefore, the exponential function is further utilized in this embodiment>Optimization->The existence of the errors is compensated through the nonlinear steep increase characteristic of the exponential function, so that the adaptability of the particle swarm algorithm is more scientific.
Further, as shown in connection with FIG. 3, in a preferred embodiment, the scheduling analysis module 123 performs the functions of: based on the target feeding position, according to the position data, the feed allowance data, the fuel allowance data and the fish school density data of each feeding robot, a scheduling plan of a plurality of feeding robots is obtained, and the method specifically comprises the following steps:
S301, obtaining estimated feeding quantity of each target feeding position based on the fish swarm density data corresponding to each position data according to the position relation between the position data and the target feeding position;
s302, establishing a plurality of initial chromosomes based on the corresponding relation from the feeding robot to the target feeding position, so as to obtain an initial population;
s303, constructing a second fitness function by taking the fuel consumption cost of the feeding robot running to the corresponding target feeding position and the utilization rate of the feeding robot feeding the feed as well as optimizing the initial population through a genetic algorithm based on the second fitness function according to the feed remaining data, the fuel remaining data and the estimated feeding amount, obtaining an optimal chromosome representing the optimal corresponding relation between the feeding robot and the target feeding position, and obtaining a scheduling plan.
According to the method, firstly, the estimated feeding amount of each target feeding position is estimated through the shoal density data near the target feeding position, and then an optimal scheduling plan of a plurality of feeding robots is obtained through a genetic algorithm, wherein the scheduling plan is the target feeding positions to which the feeding robots go for feeding, and the feeding robots and the target feeding positions can be in a many-to-one relationship. The process can enable the robot to reasonably go to the target feeding point position most suitable for the situation according to the feed allowance and the fuel allowance of the robot, so that the situation that the fuel consumption of the individual feeding robot is large, but the feeding amount of the individual feeding robot is small is avoided, and the utilization of the fuel is maximized.
Such an optimization process provides powerful support for automated and intelligent precision farming. Through the steps, the maximization of the utilization of fuel and feed can be realized, and the balance and the high efficiency of the feeding process are ensured. The feeding robot intelligently adjusts the action plan according to the real-time data, so that the waste of resources and the scheduling conflict among robots can be avoided. The optimal scheduling plan of the plurality of feeding robots is finally formed, so that the feeding work of the whole farm is ensured to be orderly carried out, and the environment and the cultivation condition can be flexibly responded. In the whole, the particle swarm optimization algorithm and the genetic algorithm are comprehensively utilized to optimize the feeding points and the robot scheduling, so that the resource allocation intelligence level of the whole breeding system is improved, and the breeding efficiency and economic benefit are remarkably improved.
Among these, genetic algorithms (Genetic Algorithm, GA) are search heuristics that mimic the process of natural selection to solve optimization and search problems. The term used in genetic algorithms is generally borrowed from evolutionary biology, such as populations, individuals, inheritance, mutations, environmental adaptations, etc.
The genetic algorithm is generally described as follows:
1. Initializing: a population comprising a plurality of individuals is randomly generated. Each individual (called a chromosome) represents one possible solution to the problem, which is typically in the form of a string, such as a binary code.
2. Evaluation: the fitness of each individual in the population is calculated. Fitness is a numerical value of an individual that measures the ability to solve a problem, and is typically defined by a fitness function.
3. Selecting: the selection is made according to the fitness of the individual. Individuals with high fitness have a higher chance to be selected for the next generation. Common selection methods are roulette selection, tournament selection, etc.
4. Crossover (hybridization): pairing selected pairs of individuals and exchanging fragments results in new individuals. Crossover operations are a major source of diversity for genetic algorithms that mimic biological hybridization processes, creating new combinations on genetic material.
5. Variation: part of the genes on the chromosome are randomly changed with a small probability. The mutation operation may introduce new genetic structures during the search process to avoid premature convergence of the algorithm to a locally optimal solution.
6. Iteration: the process of evaluating, selecting, crossing and mutating is repeated. Each iteration is referred to as a generation. After many successive generations, the population will evolve towards a more environmentally friendly direction.
7. And (3) terminating: when a certain termination condition is met (e.g., a preset number of iterations is reached, fitness exceeds a certain threshold, individuals in the population are sufficiently similar, etc.), the algorithm stops.
The genetic algorithm can effectively process various types of objective functions and constraint conditions, including nonlinear, discontinuous and non-smooth problems, has good robustness, and is less prone to be trapped in a locally optimal solution due to the characteristic of random search, but tends to find a globally optimal solution. Genetic algorithms evaluate each individual in a population independently, naturally supporting parallel processing, and are suitable for operation in modern multiprocessor computing environments. In addition, the genetic algorithm does not need gradient information of the solution, and only needs to evaluate the advantages and disadvantages of the solution through a fitness function, and the algorithm enhances global searching capability and avoids premature convergence by maintaining individual diversity in the population.
Further, in a preferred embodiment, the estimated feeding amount of each target feeding position is obtained according to the following formula according to the positional relationship between the position data and the target feeding position:
；
wherein,is->Individual target feeding position +.>Is a predicted feeding amount of->For being located at the targeted feeding position +. >Is the center of a circle, ">Is the sum of the fish school density data corresponding to the position data within the radius range, +.>The coefficients are adjusted for the first unit.
It can be seen that the above formula and the previous calculation of the particles to be analyzed currently in the particle swarm algorithmIs adapted to (a)The formula of the formula (C) is similar, the feeding amount is estimated through the density conditions around the European and American target feeding positions, so that in actual implementation, the system can save the data obtained when the first fitness function is calculated before, the data can be directly used for estimating the feeding amount in the step, the operation process is reduced, and the speed of the system for processing the analysis problem is improved.
Further, in an preferred embodiment, the chromosome in the genetic algorithm is a one-dimensional vector, each element position in the chromosome represents a feeding robot, each element value is used for representing that the feeding robot corresponding to the element position where the element value is located needs to move to a target feeding position corresponding to the element value to perform feeding or the feeding robot does not participate in feeding, the value of each element value includes a plurality of first numbers and a plurality of second numbers, the plurality of first numbers represent the plurality of target feeding positions, and the second numbers represent that the feeding robot does not participate in feeding.
For example, in a preferred embodiment, if five feeding robots are provided and three target feeding positions are analyzed, then the three target feeding positions are respectively numbered as position 1, position 2, position 3 and are used as a first number, and 0 is used as a second number, and the following is the example of coding for one chromosome:
；
the meaning of the code is: the first feeding robot goes to the position 1 for feeding, the second feeding robot goes to the position 2 for feeding, the third feeding robot goes to the position 3 for feeding, the fourth feeding robot goes to the position 1 for feeding, and the fifth feeding robot stands by in situ or continues to cruise and does not participate in feeding work.
The coding has the advantages that the condition that the feeding robot does not participate in feeding is considered, the coding is one-dimensional data, and subsequent chromosome iteration is facilitated.
Based on the above coding, the iterative rules in the genetic algorithm need to be somewhat restricted, and therefore, in a preferred embodiment, the genetic algorithm comprises:
obtaining a offspring chromosome according to the parent chromosome based on a preset rule;
wherein, the preset rule comprises:
parent chromosomes do not cross to obtain offspring chromosomes;
The probability that the element value in the parent chromosome is changed to the second number is related to the feed allowance data and the fuel allowance data of the feeding robot corresponding to the element value.
Because of the encoding method of this embodiment, if a new chromosome is generated by the crossover method, there may be a case where a feeding robot that is farther from the target feeding position is assigned to work to the target feeding position in the new chromosome. So to avoid such inverse optimization, the chromosomes in the present application are iterated only through variations other than crossover, inheritance, etc., to obtain new chromosomes.
In addition, in the variation process, the probability of the element value variation to the second number is combined with the feed allowance data and the fuel allowance data, so that some feeding robots with lower economic benefits are excluded from the feeding task.
It will be appreciated that the parent chromosome, the offspring chromosome, etc. in the above process are one of the chromosomes in the genetic algorithm, and these chromosomes have the same structure, and for these different prefixes or suffixes, those skilled in the art will understand the meaning, and therefore will not be described in any great detail herein.
Further, in a preferred embodiment, the probability of the element value variation in the parent chromosome being the second number is obtained by:
；
wherein,is chromosome->Middle->Probability of variation of the element value to the second number,/->Chromosome->Middle->Individual elementFeed allowance data of feeding robot corresponding to prime value, < ->Is chromosome->Middle->Fuel balance data of feeding robot corresponding to individual element values, < >>The coefficients are adjusted for the second unit.
The meaning of the above formula is that when the feed allowance and the fuel allowance of the feeding robot are more, the weight of the feeding robot is larger at the moment, the cost of driving the feeding robot to move is higher, the fuel utilization rate is lower, and then the feeding robot is not suitable for moving, or can be regarded as being unsuitable for participating in the feeding activity and waiting for the next feeding condition of fully utilizing the fuel, and the probability of obtaining the second code is improved at the moment. When the feed allowance and the fuel allowance of the feeding robot are smaller, the weight of the feeding robot is smaller, the cost of driving the feeding robot to move is lower, the fuel utilization rate is higher, the feeding robot is suitable for moving and participating in feeding, and the feed and the fuel of the feeding robot can be emptied more rapidly so as to be timely supplied, and the probability of obtaining the second code is reduced. And when the feed allowance is more and the fuel allowance is smaller or the feed allowance is larger and the fuel allowance is smaller, the fuel utilization rate of the mobile robot is moderate, and the probability of obtaining the second code should also be moderate. This enables a higher fuel utilization of the overall system.
Further, in a preferred embodiment, the second fitness function includes:
；
；
wherein,is chromosome +.>Is adaptive to->Is chromosome->Corresponding target feeding position +.>The sum of the feed margins of the feeding robot corresponding to the element values, < >>For the preset difference, ++>Is chromosome->Middle->Element value->For characterizing element values +.>The corresponding feeding robot can reach the target feeding position +.>The Boolean value of 1 indicating that it is possible to achieve, the Boolean value of 0 indicating that it is not possible to achieve, the Boolean value being based on the element value +.>Feed allowance data of corresponding feeding robot +.>And element value->Fuel balance data +.>Obtained (I)>Is chromosome->Corresponding target feeding position +.>Feeding robot corresponding to element value of (2) and target feeding position +.>Distance sum of>Is chromosome->Corresponding target feeding position +.>The utilization rate of all the feeds fed by the feeding robot corresponding to the element values, < + >>、/>、/>Respectively different weight coefficients, +.>Is a slope adjustment coefficient.
The meaning of the above formula is that when the chromosomeCorresponding target feeding position +.>When the difference between the sum of the fuel allowance of the feeding robot corresponding to the element values and the estimated feeding amount is larger, the chromosome +. >The corresponding schedule cannot meet the current feeding demand, and its fitness should be set to 0. At the same time, if the chromosome->The chromosome +.A situation where the fuel of the feeding robot is insufficient to support the feeding robot to carry its feed to the target feeding position is also indicated>The corresponding schedule cannot meet the current feeding demand, and the adaptability should be set to 0.
And to other circumstances that can satisfy the feeding demand, through combining fodder surplus, fuel surplus and fodder utilization ratio in this embodiment to obtain scientific and reasonable fitness from a plurality of angles. Wherein,the cost of carrying the feed by the feeding robot is represented, and obviously, the more the feed is carried, the higher the cost of running the robot is, the lower the fitness is.The distance between the feeding robot and the corresponding target feeding position is represented, the fuel utilization condition of the feeding robot is indirectly represented, the farther the distance is, the more fuel is consumed, the lower the utilization rate is, and the lower the fitness is。The corresponding feed utilization rate of the chromosome is characterized, and the higher the feed utilization rate is, the higher the fitness is. Furthermore, the->According to->And->The ratio of the two is obtained by the ratio of the two to be closer to one, which represents the higher utilization rate of the feed. Through the second fitness function in the embodiment, a genetic algorithm can be enabled to obtain a more scientific and reasonable scheduling plan.
Likewise, it is to be understood that other processes and details in genetic algorithms, in addition to those mentioned above, are well known and practiced by those skilled in the art and, therefore, are not described in great detail herein.
It should be noted that, in the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described as different from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (9)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410001445.5A CN117495061B (en) | 2024-01-02 | 2024-01-02 | An intelligent marine ranch feeding system based on the Internet of Things |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410001445.5A CN117495061B (en) | 2024-01-02 | 2024-01-02 | An intelligent marine ranch feeding system based on the Internet of Things |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117495061A CN117495061A (en) | 2024-02-02 |
| CN117495061B true CN117495061B (en) | 2024-04-12 |
Family
ID=89680454
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410001445.5A Active CN117495061B (en) | 2024-01-02 | 2024-01-02 | An intelligent marine ranch feeding system based on the Internet of Things |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117495061B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116473009A (en) * | 2023-05-23 | 2023-07-25 | 西安电子科技大学杭州研究院 | A sea intelligent fishery feeding system |
-
2024
- 2024-01-02 CN CN202410001445.5A patent/CN117495061B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116473009A (en) * | 2023-05-23 | 2023-07-25 | 西安电子科技大学杭州研究院 | A sea intelligent fishery feeding system |
Non-Patent Citations (2)
| Title |
|---|
| 基于遗传算法的计量泵投料系统优化;丁言丰等;机电工程;20150331;第32卷(第3期);第343-347、357页 * |
| 基于鱼群摄食规律的投饵系统研究;贾成功等;机械工程师;20170810(第8期);第22-25页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117495061A (en) | 2024-02-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wu et al. | Application of intelligent and unmanned equipment in aquaculture: A review | |
| Zhou et al. | Near infrared computer vision and neuro-fuzzy model-based feeding decision system for fish in aquaculture | |
| CN113807017B (en) | Method for determining fish preference habitat and terminal equipment | |
| CN114994691B (en) | Fishery resource cluster distribution analysis method based on visualization | |
| CN101867869A (en) | Network water monitoring and early warning system of mobile robot on water and its realization method | |
| Robert et al. | Does social behavior influence the dynamics of aggregations formed by tropical tunas around floating objects? An experimental approach | |
| CN110045771B (en) | An intelligent monitoring system for fish pond water quality | |
| CN118355874B (en) | Oyster reef proliferation repair regional control method | |
| CN114493007B (en) | A method, device and system for evaluating the biocapacity of marine ranches | |
| CN103760834A (en) | Fishery aquaculture growth process internet-of-things monitoring system | |
| CN112001641B (en) | A remote sensing assessment system for the suitability of scallop culture areas | |
| CN116739817B (en) | Marine organism diversity monitoring system and data processing method | |
| TWM578511U (en) | Feed supplying system of marine cultivation | |
| CN117592662A (en) | Ecological fishery resource evaluation system based on data analysis | |
| CN117243145A (en) | Accurate feeding method and device for meat poultry based on feed intake | |
| CN117495061B (en) | An intelligent marine ranch feeding system based on the Internet of Things | |
| CN118446419A (en) | Intermittent culture feasibility analysis and dynamic optimization method applied to estuary | |
| CN118296717A (en) | A method and system for evaluating ecological fishway | |
| CN116754738A (en) | Water area carbon dynamic monitoring method, equipment, robot and computer storage medium | |
| Brunner et al. | Bertalanffy-Pütter models for avian growth | |
| Cristea et al. | The use of artificial intelligence in sturgeon aquaculture | |
| CN116579504A (en) | Prediction method for pond mariculture water temperature based on machine learning | |
| CN116965362A (en) | Intelligent feeding system for micropterus salmoides based on machine vision | |
| Bhattacharya et al. | To join or not to join: collective foraging strategies | |
| Latief et al. | Innovative technologies in fisheries sector |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |