CN115983131B - Aquatic product growth oxygen content regulation and control method and device - Google Patents
Aquatic product growth oxygen content regulation and control method and device Download PDFInfo
- Publication number
- CN115983131B CN115983131B CN202310027610.XA CN202310027610A CN115983131B CN 115983131 B CN115983131 B CN 115983131B CN 202310027610 A CN202310027610 A CN 202310027610A CN 115983131 B CN115983131 B CN 115983131B
- Authority
- CN
- China
- Prior art keywords
- oxygen content
- representing
- prey
- algorithm
- follows
- 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
- 239000001301 oxygen Substances 0.000 title claims abstract description 111
- 229910052760 oxygen Inorganic materials 0.000 title claims abstract description 111
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 title claims abstract description 101
- 238000000034 method Methods 0.000 title claims abstract description 22
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 50
- 244000062645 predators Species 0.000 claims abstract description 40
- 230000001276 controlling effect Effects 0.000 claims abstract description 19
- 230000001105 regulatory effect Effects 0.000 claims abstract description 19
- 238000012544 monitoring process Methods 0.000 claims abstract description 15
- 230000008859 change Effects 0.000 claims abstract description 11
- 230000006870 function Effects 0.000 claims description 24
- 244000062662 apex predator Species 0.000 claims description 18
- 238000005457 optimization Methods 0.000 claims description 17
- 238000013178 mathematical model Methods 0.000 claims description 12
- 230000000694 effects Effects 0.000 claims description 11
- 208000015756 familial Alzheimer disease Diseases 0.000 claims description 9
- 235000019162 flavin adenine dinucleotide Nutrition 0.000 claims description 9
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000011161 development Methods 0.000 claims description 7
- 238000005276 aerator Methods 0.000 claims description 6
- 230000006399 behavior Effects 0.000 claims description 6
- 238000009826 distribution Methods 0.000 claims description 6
- 238000004519 manufacturing process Methods 0.000 claims description 6
- 230000006386 memory function Effects 0.000 claims description 6
- 238000003062 neural network model Methods 0.000 claims description 6
- 230000002441 reversible effect Effects 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 241000251468 Actinopterygii Species 0.000 claims description 3
- 230000005653 Brownian motion process Effects 0.000 claims description 3
- 230000004913 activation Effects 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 238000005537 brownian motion Methods 0.000 claims description 3
- 230000003247 decreasing effect Effects 0.000 claims description 3
- 230000006872 improvement Effects 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 241000894007 species Species 0.000 claims description 2
- 238000009360 aquaculture Methods 0.000 abstract description 11
- 244000144974 aquaculture Species 0.000 abstract description 8
- 230000008901 benefit Effects 0.000 abstract description 4
- 238000005265 energy consumption Methods 0.000 abstract description 4
- 230000007613 environmental effect Effects 0.000 abstract description 4
- 230000004083 survival effect Effects 0.000 abstract description 4
- QGZKDVFQNNGYKY-UHFFFAOYSA-N Ammonia Chemical compound N QGZKDVFQNNGYKY-UHFFFAOYSA-N 0.000 description 2
- RWSOTUBLDIXVET-UHFFFAOYSA-N Dihydrogen sulfide Chemical compound S RWSOTUBLDIXVET-UHFFFAOYSA-N 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 2
- 229910000037 hydrogen sulfide Inorganic materials 0.000 description 2
- 238000006213 oxygenation reaction Methods 0.000 description 2
- 231100000614 poison Toxicity 0.000 description 2
- 239000003440 toxic substance Substances 0.000 description 2
- 206010021143 Hypoxia Diseases 0.000 description 1
- 229910021529 ammonia Inorganic materials 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000007954 hypoxia Effects 0.000 description 1
- 244000005700 microbiome Species 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000002028 premature Effects 0.000 description 1
- 238000009827 uniform distribution Methods 0.000 description 1
Images
Classifications
-
- 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
- Feedback Control In General (AREA)
Abstract
The invention discloses a method and a device for regulating and controlling the oxygen content of aquatic product growth, comprising a monitoring module, a prediction module and a control module. The monitoring module comprises a plurality of temperature sensors, a water oxygen content detector and a humidity sensor, and is used for monitoring environmental parameters around the aquaculture farm; the prediction module comprises a long-short-term memory network model LSTM and a marine predator algorithm, the hidden layer node number and the learning rate of the LSTM prediction model are optimized through the marine predator algorithm, the change of the oxygen content of the culture pond in the future 24 hours is predicted, and a basis is provided for intelligent regulation; the control module regulates and controls the power of the oxygen increasing device of the cultivation pond according to the oxygen content prediction result, so that the oxygen content is kept in an optimal state suitable for the survival of aquatic products. The invention can make corresponding adjustment according to the environmental change of the aquiculture farm, reduce the energy consumption while guaranteeing the optimal state of the aquiculture pond, and ensure the reliable operation of the aquiculture farm and simultaneously optimize the economic benefit and the environmental benefit.
Description
Technical Field
The invention relates to the technical field of oxygen content monitoring and controlling of a culture pond, in particular to a method and a device for regulating and controlling the oxygen content of aquatic product growth.
Background
In order to ensure that the dissolved oxygen in the culture water is sufficient, the aquaculture generally adopts a production mode of starting the aerator with high power all the day, which causes great energy consumption. Industrial aquaculture is an important production model for future aquaculture, and its production efficiency will directly determine the total yield and income of aquaculture.
The aquaculture water environment is a habitat where aquatic products live, wherein the dissolved oxygen content of the water body is a key water quality factor for measuring the aquaculture water quality. In industrial cultivation, as the cultivation density is high, the cultivation environment is closed, the cultivation process is often subjected to risks caused by hypoxia, when the dissolved oxygen in water is insufficient, ammonia and hydrogen sulfide in water are difficult to decompose and convert, meanwhile anaerobic microorganisms in the water are increased, aquatic animals and plants are further caused to die and anaerobically decompose, more toxic substances such as hydrogen sulfide are released, the toxic substances are easy to accumulate to the extent of damaging the health of the aquatic animals and plants, and the water and the surrounding ecological environment are deteriorated. In order to improve the content of dissolved oxygen, a high-power aerator is usually started all the day in a culture factory, but the oxygenation operation has the problems of overlarge energy consumption and the like, and in order to solve the problems, equipment is needed for regulating and controlling the water oxygen content of the culture farm in real time, flexibly regulating the power of the aerator, increasing the yield of the culture farm and simultaneously reducing the cost to the minimum.
Disclosure of Invention
The invention aims to: aiming at the problem of water oxygen content regulation in the current aquaculture industry, the invention provides a method and a device for regulating and controlling the oxygen content of aquatic product growth, which can predict the water oxygen content change of a farm within 24 hours in the future according to the environmental change around the aquatic product farm and the water oxygen content most suitable for the survival of the aquatic product in the current stage, flexibly regulate and control the water oxygen content of the farm according to the predicted data, and ensure that the aquatic product is always in a culture pond with the most suitable water oxygen content, thereby improving the yield of the aquatic product and the economic benefit.
The technical scheme is as follows: the invention provides a method for regulating and controlling the oxygen content of aquatic product growth, which comprises the following steps:
step 1: acquiring historical data of a period of temperature, humidity and oxygen content of water in a water product growing environment in a culture pond;
step 2: constructing an LSTM network model, optimizing the number of initial hidden nodes and the learning rate of the LSTM network by adopting an improved marine predator algorithm MPA, constructing an MPA-LSTM prediction model, performing differential evolution on a high-speed-ratio-stage prey by adopting a staged improved marine predator algorithm MPA, introducing a sine-cosine algorithm to implement same-rhythm sine-cosine fluctuation in a peer-speed-ratio-stage parallel architecture, and constructing a cauchy mirror image predator in a low-speed-ratio stage;
step 3: based on the historical data collected in the step 1, predicting the oxygen content change in the culture pond within 24 hours in the future by adopting an MPA-LSTM prediction model;
step 4: and (3) setting the optimal oxygen content of the aquatic product growing environment of the culture pond through big data retrieval, comparing the oxygen content prediction data in the step (3) with the optimal oxygen content of the aquatic product growing environment to obtain a difference value, and when the absolute value of the difference value of the oxygen content prediction data and the oxygen content of the aquatic product growing environment is larger than a limiting value, adopting a northern hawk optimizing algorithm NGO to optimally control the operation duration of the water oxygen content regulator, and regulating and controlling the oxygen content of the culture pond in real time.
Further, the historical data in the step 1 are monitored through a temperature sensor, a humidity sensor and a water oxygen content detector, wherein the temperature sensor and the humidity sensor are installed at multiple points around the culture pond, and the water oxygen content detector is installed at multiple points in the water of the culture pond.
Further, the specific steps of constructing the MPA-LSTM prediction model in the step 2 are as follows:
21 An LSTM neural network model is constructed, and 3 modules with memory functions of an input gate, an output gate and a forget gate of the LSTM in the neural network model are constructed;
22 Forgetting door f t Is responsible for deciding which information to discard from the memory unit, and updates the formula as follows:
f t ＝σ(w fx x t +w fh h t-1 +w fc C t-1 +b f )
sigma (·) -sigmoid: x is x t An input representing a time t; h is a t Output at time t; h is a t-1 Output C representing time t-1 t A candidate vector representing time t; w (w) fx 、w fh 、w fc Weight coefficients representing forgetting gates; b f Representing a forget gate bias;
23 Output gate i) t Is responsible for deciding which information can be stored in the memory unit, and updating the formula as follows:
i t ＝σ(w ix x i +w ih h t-1 +w ic C t-1 +b i )
wherein: w (w) ix 、w ih 、w ic A weight coefficient representing an input gate; b i Representing input gate bias; w (w) cx 、w ch A weight coefficient representing the candidate vector; b 0 Representing candidate vector bias; tanh () represents a hyperbolic tangent activation function;
an updated value representing the candidate vector;
24 An output gate decides which information to output in the following manner:
o t ＝σ(w ox x t +w oh h t-1 +w oc C t-1 +b o )
h t ＝o t tanh(C t )
wherein: o (o) t Representing an output gate; w (w) ox 、w oh 、w oc A weight coefficient representing an output gate;
25 The number of initial hidden nodes and the learning rate of the LSTM network are optimized by adopting an improved marine predator algorithm MPA, and an MPA-LSTM prediction model is constructed.
Further, the specific steps of the improved marine predator algorithm MPA are as follows:
31 Selecting an optimal fitness prey individual to act as a top predator E, initializing a marine environment, and adopting the following initialization formula:
P 0 ＝P min +rand(0,1)·(P max -P min )
wherein P is max And P min Respectively representing upper and lower bounds of a solution space; rand (0, 1) is an own random factor, which is the presentThe quality is a uniform random vector in the range of (0, 1);
32 With the current iteration times t and the maximum iteration times t max Dividing the algorithm process into a high speed ratio stage, an equal speed ratio stage and a low speed ratio stage for criteria, designing FADs effect to disturb a local optimal solution, and re-evaluating individual fitness by using a marine memory function to update a top predator;
33 High speed ratio phase (t < t) max 3) the top predator gives up hunting because it is far lower in speed than the prey, which performs brownian motion, the mathematical model of this stage is as follows:
wherein p is a constant, 0.5 is taken; r is the value range [0,1 ]]Is a uniform random number vector of (a); p (P) i And E is connected with i Respectively representing the ith prey and the top predator individuals of the current iteration; r is R B Is a Brownian random vector in normal distribution;
34 Parallel architecture of MPA consists of constant velocity ratio phase (t) max /3＜t＜2t max Equally dividing the prey population of/3), wherein the first half of individuals form a Laiweighua population responsible for development, and the second half of individuals form a Brownian population responsible for exploration at the same time, and the parallel architecture mathematical model is as follows:
wherein p is a constant, 0.5 is taken; CF is an adaptive parameter; r is R L Is a random vector in a Lewye distribution;
35 Low speed ratio phase (t > 2 t) max The speed of the prey at/3) is far lower than that of the top predator, the Lewy flight trajectory of the top predator is referred to avoid predation when the prey moves, and the mathematical model of the stage is as follows:
36 Eddies or fish gathering devices FADs that affect marine predators' behavior and can be considered locally optimal, the FADs effect is determined by a perturbation probability factor p of 0.2 F MPA is introduced to reduce the algorithm local stagnation probability, and the mathematical model is as follows:
wherein U is a binary number group for randomly generating a binary vector; r is [0,1 ]]A uniform random number within the range; p (P) r1 And P r2 Representing two randomly drawn prey species in a prey population;
37 Stage-wise improvement of the MPA algorithm, comprising the following steps:
38 For the high speed ratio phase (t < t) max And 3) performing differential evolution on the prey, and setting a d-dimensional space t-th generation prey individual as follows:
39 Optionally 3 isotropically diverse individuals
Production of variant prey->
Wherein F is a differential variation factor;
310 Through the original individual
And variant individuals->
Cross-producing variant Cross prey individuals +.>
Wherein d rand The value range is [1, d ] for the random dimension coefficient]；
311 Selecting original individuals
Individuals crossing the mutation->
The better fitness of the middle is reserved as the evolutionary prey individual
The quality of the hunting body is optimized by traversing, which is indirectly sub-stage cultivationCultivating high-quality predators;
312 Constant velocity ratio stage parallel architecture implements the same-rhythm sine and cosine fluctuation:
after the high speed ratio stage is finished, a uniform random fluctuation probability pf and a fluctuation operator f simultaneous fluctuation parallel architecture is introduced to generate a fluctuation prey with higher flexibility
Screening out fitness optima as wave predators +.>
Wherein r is 1 An adjustment parameter r decreasing linearly 2 Is in the range of 0,2 pi]The uniform random angle in the inner part is matched with the fluctuation probability p f Equal probability selection fluctuation mechanism, same-rhythm optimization development capability and exploration capability, r 3 Is [0,2]The random numbers which are uniformly distributed are taken in the range;
313 Cauchy mirror variation of predators at low speed phase for fluctuating predators entering low speed ratio phase
Reverse learning is performed to generate mirror image fluctuation predators->
And constructing a cauchy mirror variant predator +.>
The formula is as follows:
wherein ub and lb respectively represent the upper and lower bounds of the corresponding solution space, and rand is a random number matrix uniformly distributed in compliance with the (0, 1) standard; b is a reverse learning pseudo information exchange coefficient;
for wave predators->
And mirror image wave predator->
Elite predators generated by b implementing reverse learning information exchange, < >>
Is a cauchy mirror image variant predator.
Further, in the step 4, the northern eagle optimization algorithm is adopted to perform the specific operation of optimizing and controlling the operation time of the aerator, wherein the specific operation is as follows:
41 Initializing population members, the initialization matrix is as follows:
wherein X represents a northern eagle population matrix, X i Representing the initial solution of the ith individual, x i,j A value representing the j-th dimension of the i-th individual, N representing the number of population members, m being the dimension of the problem space;
42 Using the objective function value vector to represent the objective function value of the northern eagle population, wherein the expression formula is as follows:
wherein F is a vector of the obtained objective function values, F i Is the objective function value obtained by the ith solution;
43 Identifying the optimal region, and the region search formula is as follows:
P i ＝X k ,i＝1,2,…,N,k＝1,2,…,i-1,i+1,…,N
wherein P is i Is the hunting position of the ith northern hawk, F Pi Is the objective function value, k is the interval [1, N ]]Is not a random integer of i,
is the new position of the ith solution, +.>
Is the value of the j-th dimension thereof, +.>
Is NThe objective function value of the first stage GO, r is the interval [0,1]I is a random number with a value of 1 or 2, and the parameters r and I are random numbers that generate random NGO behavior in search and update;
44 Improving the ability of the algorithm to search locally for the search space, assuming such hunting activity is close to an attack location with radius R, the location update formula is as follows:
where T is the current iteration number, T is the maximum iteration number,
new positions of i solutions, +.>
Is the value of the j-th dimension thereof, +.>
Is the objective function value of the second stage of NGO.
The invention also discloses a regulating and controlling device based on the aquatic product growth oxygen content regulating and controlling method, which comprises a data monitoring module, an oxygen content predicting module and a regulator control module;
the data monitoring module comprises a plurality of temperature sensors, a plurality of humidity sensors and a plurality of water oxygen content detectors, and is used for monitoring historical data of temperature, humidity and water oxygen content of different places and different time periods respectively;
the oxygen content prediction module comprises an MPA-LSTM prediction model and is used for predicting the oxygen content change in the culture pond within 24 hours in the future by combining historical data of temperature, humidity and water oxygen content;
the regulator control module comprises an optimization algorithm module and a water-oxygen content regulator, wherein the optimization algorithm module is used for combining oxygen content prediction data and an optimal oxygen content difference value of an aquatic product growth environment, and when the absolute value of the difference value of the oxygen content prediction data and the optimal oxygen content difference value of the aquatic product growth environment is larger than a limiting value, a northern hawk optimization algorithm is adopted to optimally control the operation time of the water-oxygen content regulator.
The beneficial effects are that:
1. compared with the traditional method for estimating the water oxygen content of the culture pond by artificial experience, the MPA-LSTM water oxygen content prediction model provided by the invention can accurately predict the water oxygen content change in the future 24 hours, and can obtain accurate water oxygen content data and has stronger reliability.
2. Compared with the traditional artificial oxygenation, the intelligent regulator provided by the invention can timely and flexibly regulate the water and oxygen content of the farm, so that the aquatic product is always in the environment most suitable for survival of the aquatic product.
3. According to the invention, the MPA optimization algorithm and the northern eagle optimization algorithm NGO are adopted to optimize the water oxygen content prediction model and the control model respectively, so that the prediction accuracy and the control effect are remarkably improved.
4. The invention can flexibly adjust the water-oxygen content of the farm according to the growth cycle of the aquatic products, so that the aquatic products are always in the most suitable living environment, the quality and the yield of the aquatic products are improved to a great extent, and the economic benefit is also improved obviously.
Drawings
FIG. 1 is a general frame diagram of the apparatus of the present invention;
FIG. 2 is a flow chart of an MPA-LSTM prediction model employed in the present invention;
FIG. 3 is a flow chart of a northern eagle optimization algorithm adopted by the invention;
FIG. 4 is a graph comparing actual water oxygen content with predicted effect of a farm;
figure 5 is a graph comparing net annual revenue before and after use of the apparatus.
Detailed Description
The invention will be described in further detail with reference to the accompanying drawings
The invention discloses a method and a device for regulating and controlling the oxygen content of aquatic product growth, and the device comprises a data monitoring module, an oxygen content prediction module and a regulator control module, wherein the data monitoring module is used for monitoring the oxygen content of the aquatic product; the data monitoring module comprises a plurality of temperature sensors, a plurality of humidity sensors and a plurality of water oxygen content detectors. The data monitoring module collects external environment parameters, the temperature sensor and the humidity sensor are installed around the farm in a multipoint mode, and the water oxygen content detector is installed in water in a multipoint mode.
The oxygen content prediction module MPA-LSTM prediction model is used for predicting the oxygen content change in the culture pond within 24 hours in the future by combining historical data of temperature, humidity and water oxygen content.
The control module comprises an optimization algorithm module and a water-oxygen content regulator, wherein the optimization algorithm module is used for combining oxygen content prediction data and an optimal oxygen content difference value of an aquatic product growth environment, and when the absolute value of the difference value of the oxygen content prediction data and the optimal oxygen content difference value of the aquatic product growth environment is larger than a limiting value, a northern hawk optimization algorithm is adopted to optimally control the operation duration of the water-oxygen content regulator.
The control module compares the predicted data with the data acquired by the water oxygen content detector according to the information transmitted by the prediction module, and when the absolute value of the difference value of the predicted data and the data is larger than a limiting value, the control module intelligently regulates the water oxygen content;
the method for regulating and controlling the oxygen content of the aquatic product growth comprises the following steps:
step 1: and acquiring historical data of a period of temperature, humidity and oxygen content of water in the growth environment of the water product in the culture pond.
Step 2: and constructing an LSTM network model, optimizing the number of initial hidden layer nodes and the learning rate of the LSTM network by adopting a marine predator algorithm MPA, and constructing an MPA-LSTM prediction model.
As shown in figure 2, the MPA-LSTM prediction model disclosed by the invention is realized as follows:
21 An LSTM neural network model is constructed, and 3 modules with memory functions of an input gate, an output gate and a forget gate of the LSTM in the neural network model are constructed.
22 Forgetting door f t Is responsible for deciding which information to discard from the memory unit, and updates the formula as follows:
sigma (·) -sigmoid: x is x t An input representing a time t; h is a t Output at time t; h is a t-1 Output C representing time t-1 t A candidate vector representing time t; w (w) fx 、w fh 、w fc Weight coefficients representing forgetting gates; b f Indicating a forgetting gate bias.
23 Output gate i) t Is responsible for deciding which information can be stored in the memory unit, and updating the formula as follows:
i t ＝σ(w ix x i +w ih h t-1 +w ic C t-1 +b i )
wherein: w (w) ix 、w ih 、w ic A weight coefficient representing an input gate; b i Representing input gate bias; w (w) cx 、w ch A weight coefficient representing the candidate vector; b 0 Representing candidate vector bias; tanh () represents a hyperbolic tangent activation function;
representing the updated value of the candidate vector.
24 An output gate decides which information to output in the following manner:
o t ＝σ(w ox x t +w oh h t-1 +w oc C t-1 +b o )
wherein: o (o) t Representing an output gate; w (w) ox 、w oh 、w oc Representing the weight coefficient of the output gate.
25 The method comprises the following steps of) optimizing the number of initial hidden nodes and learning rate of an LSTM network by adopting an improved marine predator algorithm MPA, constructing an MPA-LSTM prediction model, wherein the specific steps of the improved marine predator algorithm MPA algorithm are as follows:
26 Selecting an optimal fitness prey individual to act as a top predator E, initializing a marine environment, and adopting the following initialization formula:
P 0 ＝P min +rand(0,1)·(P max -P min )
wherein P is max And P min Respectively representing upper and lower bounds of a solution space; rand (0, 1) is an self-contained random factor that is essentially a uniform random vector over the range of (0, 1).
27 With the current iteration times t and the maximum iteration times t max The algorithm process is divided into a high speed ratio stage, an equal speed ratio stage and a low speed ratio stage for criteria, FADs effect is designed to disturb a local optimal solution, and individual fitness is reevaluated by using a marine memory function to update top predators.
28 High speed ratio phase (t < t) max 3) the top predator gives up hunting because it is far lower in speed than the prey, which performs brownian motion, the mathematical model of this stage is as follows:
wherein p is a constant, 0.5 is taken; r is the value range [0,1 ]]Is a uniform random number vector of (a); p (P) i And E is connected with i Respectively representing the ith prey and the top predator individuals of the current iteration; r is R B Is a Brownian random vector in normal distribution;
29 Parallel architecture of MPA consists of constant velocity ratio phase (t) max /3＜t＜2t max Equally dividing the prey population of/3), wherein the first half of individuals form a Laiweighua population responsible for development, and the second half of individuals form a Brownian population responsible for exploration at the same time, and the parallel architecture mathematical model is as follows:
wherein p is a constant, 0.5 is taken; CF is an adaptive parameter; r is R L Is a random vector in a Lewye distribution.
210 Low speed ratio phase (t > 2 t) max The speed of the prey at/3) is far lower than that of the top predator, the Lewy flight trajectory of the top predator is referred to avoid predation when the prey moves, and the mathematical model of the stage is as follows:
211 Vortex or vortex flowThe FADs in the fish gathering device affect the behavior of marine predators and can be regarded as local optimum, and the FADs effect is represented by a disturbance probability factor p with the value of 0.2 F MPA is introduced to reduce the algorithm local stagnation probability, and the mathematical model is as follows:
wherein U is a binary number group for randomly generating a binary vector; r is [0,1 ]]A uniform random number within the range; p (P) r1 And P r2 Two randomly drawn prey from a prey population are represented.
212 Stage-wise improvement of the MPA algorithm, comprising the following steps:
213 For the high speed ratio phase (t < t) max And 3) performing differential evolution on the prey, and setting a d-dimensional space t-th generation prey individual as follows:
214 Optionally 3 isotropically diverse individuals
Production of variant prey->
Wherein F is a differential variation factor.
215 Through the original individual
And variant individuals->
Cross-producing variant Cross prey individuals +.>
Wherein d rand The value range is [1, d ] for the random dimension coefficient]；
216 Selecting original individuals
Individuals crossing the mutation->
The better fitness of the middle is reserved as the evolutionary prey individual
At this time, the individual quality of the prey is optimized through traversal, and the prey is cultivated in a sub-stage indirectly.
217 Constant velocity ratio stage parallel architecture implements the same-rhythm sine and cosine fluctuation:
after the high speed ratio stage is finished, a uniform random fluctuation probability pf and a fluctuation operator f simultaneous fluctuation parallel architecture is introduced to generate a fluctuation prey with higher flexibility
Screening out fitness optima as wave predators +.>
Wherein r is 1 An adjustment parameter r decreasing linearly 2 Is in the range of 0,2 pi]The uniform random angle in the inner part is matched with the fluctuation probability p f Equal probability selection fluctuation mechanism, same-rhythm optimization development capability and exploration capability, r 3 Is [0,2]The random numbers are taken from a uniform distribution.
218 Cauchy mirror variation of predators at low speed phase for fluctuating predators entering low speed ratio phase
Reverse learning is performed to generate mirror image fluctuation predators->
And constructing a cauchy mirror variant predator +.>
The formula is as follows:
b＝(1-t/tmax) t
wherein ub and lb respectively represent the upper and lower bounds, ran, of the corresponding solution spaced is a random number matrix uniformly distributed in compliance with the (0, 1) standard; b is a reverse learning pseudo information exchange coefficient;
for wave predators->
And mirror image wave predator->
Elite predators generated by b implementing reverse learning information exchange, < >>
Is a cauchy mirror image variant predator.
The invention adopts the improved MPA algorithm to optimize the hidden node number and the learning rate of the LSTM prediction model, and the prediction progress and the stability are obviously improved. The adopted staged improved marine predator algorithm performs differential evolution on the high-speed ratio stage prey, the equal-speed ratio stage parallel architecture is introduced into the sine-cosine algorithm to implement the same-rhythm sine-cosine fluctuation, the low-speed ratio stage is constructed to form the cauchy mirror predator, the influence of random factors can be reduced, the coordinated exploration and development can be realized, the algorithm is premature, and the optimization precision and convergence efficiency of the algorithm are remarkably improved.
Step 3: and (3) based on the historical data collected in the step (1), predicting the oxygen content change in the culture pond within 24 hours in the future by adopting an MPA-LSTM prediction model.
Step 4: and (3) setting the optimal oxygen content of the aquatic product growing environment of the culture pond through big data retrieval, comparing the oxygen content prediction data in the step (3) with the optimal oxygen content of the aquatic product growing environment to obtain a difference value, and when the absolute value of the difference value of the oxygen content prediction data and the oxygen content of the aquatic product growing environment is larger than a limiting value, adopting a northern hawk optimizing algorithm NGO to optimally control the operation duration of the water oxygen content regulator, and regulating and controlling the oxygen content of the culture pond in real time.
As shown in fig. 3, the implementation process of optimizing and controlling the operation time length of the water-oxygen content regulator by adopting the northern eagle optimizing algorithm NGO in the invention is as follows:
41 Initializing population members, the initialization matrix is as follows:
wherein X represents a northern eagle population matrix, X i Representing the initial solution of the ith individual, x i,j A value representing the j-th dimension of the i-th individual, N representing the number of population members, and m being the dimension of the problem space.
42 Using the objective function value vector to represent the objective function value of the northern eagle population, wherein the expression formula is as follows:
wherein F is a vector of the obtained objective function values, F i Is the objective function value obtained by the ith solution.
43 Identifying the optimal region, and the region search formula is as follows:
P i ＝X k ,i＝1,2,…,N,k＝1,2,…,i-1,i+1,…,N
wherein P is i Is the hunting position of the ith northern hawk, F Pi Is the objective function value, k is the interval [1, N ]]Is not a random integer of i,
is the new position of the ith solution, +.>
Is it isValue of j-th dimension->
Is the objective function value of the first stage of NGO, r is the interval [0,1 ]]I is a random number with a value of 1 or 2, and the parameters r and I are random numbers that generate random NGO behavior in searches and updates.
44 Improving the ability of the algorithm to search locally for the search space, assuming such hunting activity is close to an attack location with radius R, the location update formula is as follows:
where T is the current iteration number, T is the maximum iteration number,
new positions of i solutions, +.>
Is the value of the j-th dimension thereof, +.>
Is the objective function value of the second stage of NGO.
As shown in figure 4, the water-oxygen content prediction method adopted by the invention has high prediction accuracy, the average per-hour error is 0.154mg/L, the daily average accuracy reaches 97.38%, and a reliable basis can be provided for the advanced regulation and control of the water-oxygen content of the farm.
As shown in figure 5, the net income per quarter is obviously improved after the equipment is put into service, the average net income per quarter is increased by 2.625 ten thousand yuan, and the annual net income is increased by 10.5 ten thousand yuan. The increased net income mainly comes from the following two aspects, namely, the invention does not need to start the high-power aerator all the day, and the energy consumption is low; secondly, the invention can lead the aquatic products to be always in the most suitable state environment for survival, which greatly improves the quality and the yield of the aquatic products.
The foregoing embodiments are merely illustrative of the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the present invention and to implement the same, not to limit the scope of the present invention. All equivalent changes or modifications made according to the spirit of the present invention should be included in the scope of the present invention.
Claims (5)
1. The method for regulating and controlling the oxygen content of the aquatic product growth is characterized by comprising the following steps of:
step 1: acquiring historical data of a period of temperature, humidity and oxygen content of water in a water product growing environment in a culture pond;
step 2: constructing an LSTM network model, optimizing the number of initial hidden nodes and the learning rate of the LSTM network by adopting an improved marine predator algorithm MPA, constructing an MPA-LSTM prediction model, performing differential evolution on a high-speed-ratio-stage prey by adopting a staged improved marine predator algorithm MPA, introducing a sine-cosine algorithm to implement same-rhythm sine-cosine fluctuation in a peer-speed-ratio-stage parallel architecture, and constructing a cauchy mirror image predator in a low-speed-ratio stage; the specific steps of the improved marine predator algorithm are as follows:
31 Selecting an optimal fitness prey individual to act as a top predator E, initializing a marine environment, and adopting the following initialization formula:
P 0 ＝P min +rand(0,1)·(P max -P min )
wherein P is max And P min Respectively representing upper and lower bounds of a solution space; rand (0, 1) is an own random factor, which is essentially a uniform random vector within the range of (0, 1);
32 With the current iteration times t and the maximum iteration times t max Dividing the algorithm process into a high speed ratio stage, an equal speed ratio stage and a low speed ratio stage for criteria, designing FADs effect to disturb a local optimal solution, and re-evaluating individual fitness by using a marine memory function to update a top predator;
33 High speed ratio phase (t < t) max 3) the top predator gives up hunting because it is far lower in speed than the prey, which performs brownian motion, the mathematical model of this stage is as follows:
wherein p is a constant, 0.5 is taken; r is the value range [0,1 ]]Is a uniform random number vector of (a); p (P) i And E is connected with i Respectively representing the ith prey and the top predator individuals of the current iteration; r is R B Is a Brownian random vector in normal distribution;
34 Parallel architecture of MPA consists of constant velocity ratio phase (t) max /3＜t＜2t max Equally dividing the prey population of/3), wherein the first half of individuals form a Laiweighua population responsible for development, and the second half of individuals form a Brownian population responsible for exploration at the same time, and the parallel architecture mathematical model is as follows:
wherein p is a constant, 0.5 is taken; CF is an adaptive parameter; r is R L Is a random vector in a Lewye distribution;
35 Low speed ratio phase (t > 2 t) max The speed of the prey at/3) is far lower than that of the top predator, the Lewy flight trajectory of the top predator is referred to avoid predation when the prey moves, and the mathematical model of the stage is as follows:
36 Eddies or fish gathering devices FADs that affect marine predators' behavior and can be considered locally optimal, the FADs effect is determined by a perturbation probability factor p of 0.2 F MPA is introduced to reduce the algorithm local stagnation probability, and the mathematical model is as follows:
wherein U is a binary number group for randomly generating a binary vector; r is [0,1 ]]A uniform random number within the range; p (P) r1 And P r2 Representing two randomly drawn prey species in a prey population;
37 Stage-wise improvement of the MPA algorithm, comprising the following steps:
38 For the high speed ratio phase (t < t) max And 3) performing differential evolution on the prey, and setting a d-dimensional space t-th generation prey individual as follows:
39 Optionally 3 isotropically diverse individuals
Production of variant prey->
Wherein F is a differential variation factor;
310 Through the original individual P i t And individuals with variation
Cross-producing variant cross prey individuals V i t ：
Wherein d rand The value range is [1, d ] for the random dimension coefficient]；
311 Selecting the original individuals P i t Crossing variant individuals V i t The better fitness of the middle is reserved as the evolutionary prey individual
At this time, the quality of the prey individuals is traversed and optimized, and the prey individuals are cultivated to be high-quality predators in the secondary stage indirectly;
312 Constant velocity ratio stage parallel architecture implements the same-rhythm sine and cosine fluctuation:
after the high speed ratio stage is finished, a uniform random fluctuation probability pf and a fluctuation operator f simultaneous fluctuation parallel architecture is introduced, and the generation flexibility is higherHigh wave hunting object
Screening out fitness optima as wave predators +.>
Wherein r is 1 An adjustment parameter r decreasing linearly 2 Is in the range of 0,2 pi]The uniform random angle in the inner part is matched with the fluctuation probability p f Equal probability selection fluctuation mechanism, same-rhythm optimization development capability and exploration capability, r 3 Is [0,2]The random numbers which are uniformly distributed are taken in the range;
313 Cauchy mirror variation of predators at low speed phase for fluctuating predators entering low speed ratio phase
Reverse learning is performed to generate mirror image fluctuation predators->
And constructing cauchy mirror image variant predators by implementing cauchy variant strategy
The formula is as follows:
b＝(1-t/tmax) t
wherein ub and lb respectively represent the upper and lower bounds of the corresponding solution space, and rand is a random number matrix uniformly distributed in compliance with the (0, 1) standard; b is a reverse learning pseudo information exchange coefficient;
for wave predators->
And mirror image wave predator->
Elite predators generated by b implementing reverse learning information exchange, < >>
Is a cauchy mirror variant predator;
step 3: based on the historical data collected in the step 1, predicting the oxygen content change in the culture pond within 24 hours in the future by adopting an MPA-LSTM prediction model;
step 4: and (3) setting the optimal oxygen content of the aquatic product growing environment of the culture pond through big data retrieval, comparing the oxygen content prediction data in the step (3) with the optimal oxygen content of the aquatic product growing environment to obtain a difference value, and when the absolute value of the difference value of the oxygen content prediction data and the oxygen content of the aquatic product growing environment is larger than a limiting value, adopting a northern hawk optimizing algorithm NGO to optimally control the operation duration of the water oxygen content regulator, and regulating and controlling the oxygen content of the culture pond in real time.
2. The method for regulating and controlling the oxygen content of aquatic product growth according to claim 1, wherein the historical data in the step 1 are monitored by a temperature sensor, a humidity sensor and a water oxygen content detector, the temperature sensor and the humidity sensor are installed at multiple points around the cultivation pond, and the water oxygen content detector is installed at multiple points in the cultivation pond water.
3. The method for regulating and controlling the oxygen content of aquatic product growth according to claim 1, wherein the specific steps of constructing the MPA-LSTM prediction model in the step 2 are as follows:
21 An LSTM neural network model is constructed, and 3 modules with memory functions of an input gate, an output gate and a forget gate of the LSTM in the neural network model are constructed;
22 Forgetting door f t Is responsible for deciding which information to discard from the memory unit, and updates the formula as follows:
f t ＝σ(w fx x t +w fh h t-1 +w fc C t-1 +b f )
sigma (·) -sigmoid: x is x t An input representing a time t; h is a t Output at time t; h is a t-1 Output C representing time t-1 t A candidate vector representing time t; w (w) fx 、w fh 、w fc Weight coefficients representing forgetting gates; b f Representing a forget gate bias;
23 Output gate i) t Is responsible for deciding which information can be stored in the memory unit, and updating the formula as follows:
i t ＝σ(w ix x i +w ih h t-1 +w ic C t-1 +b i )
wherein: w (w) ix 、w ih 、w ic A weight coefficient representing an input gate; b i Representing input gate bias; w (w) cx 、w ch A weight coefficient representing the candidate vector; b 0 Representing candidate vector bias; tanh () represents a hyperbolic tangent activation function;
an updated value representing the candidate vector;
24 An output gate decides which information to output in the following manner:
o t ＝σ(w ox x t +w oh h t-1 +w oc C t-1 +b o )
h t ＝o t tanh(C t )
wherein: o (o) t Representing an output gate; w (w) ox 、w oh 、w oc A weight coefficient representing an output gate;
25 The number of initial hidden nodes and the learning rate of the LSTM network are optimized by adopting an improved marine predator algorithm MPA, and an MPA-LSTM prediction model is constructed.
4. The method for regulating and controlling the oxygen content in aquatic product growth according to claim 1, wherein the optimizing control of the aerator operation duration by adopting the northern hawk optimizing algorithm in the step 4 is specifically performed as follows:
41 Initializing population members, the initialization matrix is as follows:
wherein X represents a northern eagle population matrix, X i Representing the initial solution of the ith individual, x i,j A value representing the j-th dimension of the i-th individual, N representing the populationThe number of members, m, is the dimension of the problem space;
42 Using the objective function value vector to represent the objective function value of the northern eagle population, wherein the expression formula is as follows:
wherein F is a vector of the obtained objective function values, F i Is the objective function value obtained by the ith solution;
43 Identifying the optimal region, and the region search formula is as follows:
P i ＝X k ,i＝1,2,…,N,k＝1,2,…,i-1,i+1,…,N
wherein P is i Is the hunting position of the ith northern hawk, F Pi Is the objective function value, k is the interval [1, N ]]Is not a random integer of i,
is the new position of the ith solution, +.>
Is the value of the j-th dimension, F i new,P1 Is the objective function value of the first stage of NGO, r is the interval [0,1 ]]I is a random number with a value of 1 or 2, and the parameters r and I are random numbers that generate random NGO behavior in search and update;
44 Improving the ability of the algorithm to search locally for the search space, assuming such hunting activity is close to an attack location with radius R, the location update formula is as follows:
where T is the current iteration number, T is the maximum iteration number,
new positions of i solutions, +.>
Is the value of the j-th dimension, F i new,P2 Is the objective function value of the second stage of NGO.
5. A regulating device based on the method for regulating the oxygen content of the growth of the aquatic products according to any one of claims 1 to 4, which is characterized by comprising a data monitoring module, an oxygen content prediction module and a regulator control module;
the data monitoring module comprises a plurality of temperature sensors, a plurality of humidity sensors and a plurality of water oxygen content detectors, and is used for monitoring historical data of temperature, humidity and water oxygen content of different places and different time periods respectively;
the oxygen content prediction module comprises an MPA-LSTM prediction model and is used for predicting the oxygen content change in the culture pond within 24 hours in the future by combining historical data of temperature, humidity and water oxygen content;
the regulator control module comprises an optimization algorithm module and a water-oxygen content regulator, wherein the optimization algorithm module is used for combining oxygen content prediction data and an optimal oxygen content difference value of an aquatic product growth environment, and when the absolute value of the difference value of the oxygen content prediction data and the optimal oxygen content difference value of the aquatic product growth environment is larger than a limiting value, a northern hawk optimization algorithm is adopted to optimally control the operation time of the water-oxygen content regulator.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310027610.XA CN115983131B (en) | 2023-01-09 | 2023-01-09 | Aquatic product growth oxygen content regulation and control method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310027610.XA CN115983131B (en) | 2023-01-09 | 2023-01-09 | Aquatic product growth oxygen content regulation and control method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115983131A CN115983131A (en) | 2023-04-18 |
| CN115983131B true CN115983131B (en) | 2023-07-04 |
Family
ID=85974034
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310027610.XA Active CN115983131B (en) | 2023-01-09 | 2023-01-09 | Aquatic product growth oxygen content regulation and control method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115983131B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116523122B (en) * | 2023-04-19 | 2023-12-01 | 淮阴工学院 | Intelligent crab culture environment monitoring and optimizing method and system |
| CN117095188B (en) * | 2023-10-19 | 2023-12-29 | 中国南方电网有限责任公司 | Electric power safety strengthening method and system based on image processing |
| CN117537826B (en) * | 2024-01-09 | 2024-03-22 | 中国民航大学 | Track planning method capable of sensing thunderstorm situation |
| CN117970784B (en) * | 2024-04-01 | 2024-06-21 | 广东海洋大学 | Oxygenation equipment control method for aquaculture |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017189337A1 (en) * | 2016-04-27 | 2017-11-02 | The Climate Corporation | Improving a digital nutrient model by assimilating a soil sample |
| CN108665106A (en) * | 2018-05-15 | 2018-10-16 | 中国农业大学 | A kind of aquaculture dissolved oxygen prediction method and device |
| CN114036850A (en) * | 2021-11-16 | 2022-02-11 | 南昌工程学院 | Runoff prediction method based on VECGM |
| CN114077931A (en) * | 2021-11-30 | 2022-02-22 | 中国水产科学研究院渔业机械仪器研究所 | Method for predicting content of dissolved oxygen in aquaculture pond |
| CN115511177A (en) * | 2022-09-26 | 2022-12-23 | 三峡大学 | Ultra-short-term wind speed prediction method based on INGO-SWGMN hybrid model |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10402919B2 (en) * | 2016-06-06 | 2019-09-03 | The Climate Corporation | Data assimilation for calculating computer-based models of crop growth |
| JP7109123B2 (en) * | 2019-04-15 | 2022-07-29 | 国立研究開発法人理化学研究所 | Environmental factor prediction device, method, program, learned model and storage medium |
-
2023
- 2023-01-09 CN CN202310027610.XA patent/CN115983131B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017189337A1 (en) * | 2016-04-27 | 2017-11-02 | The Climate Corporation | Improving a digital nutrient model by assimilating a soil sample |
| CN108665106A (en) * | 2018-05-15 | 2018-10-16 | 中国农业大学 | A kind of aquaculture dissolved oxygen prediction method and device |
| CN114036850A (en) * | 2021-11-16 | 2022-02-11 | 南昌工程学院 | Runoff prediction method based on VECGM |
| CN114077931A (en) * | 2021-11-30 | 2022-02-22 | 中国水产科学研究院渔业机械仪器研究所 | Method for predicting content of dissolved oxygen in aquaculture pond |
| CN115511177A (en) * | 2022-09-26 | 2022-12-23 | 三峡大学 | Ultra-short-term wind speed prediction method based on INGO-SWGMN hybrid model |
Non-Patent Citations (3)
| Title |
|---|
| 主成分分析和长短时记忆神经网络预测水产养殖水体溶解氧;陈英义;程倩倩;方晓敏;于辉辉;李道亮;;农业工程学报(第17期);全文 * |
| 基于PCA-Attention-LSTM网络的土壤氮含量监测;刘慧敏;甄佳奇;刘勇;解洪富;许文超;;中国农机化学报(第09期);全文 * |
| 基于局部化双向LSTM和状态转移约束的养殖水质分类预测;商艳红;张静;;渔业现代化(第02期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115983131A (en) | 2023-04-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115983131B (en) | Aquatic product growth oxygen content regulation and control method and device | |
| CN106292802B (en) | A kind of intelligent Prediction Control System and method for fish and vegetable symbiotic system | |
| CN113557890B (en) | Intelligent water precise irrigation control system and method for fruit and vegetable cultivation in sunlight greenhouse | |
| CN101796928B (en) | Method for predicting effect of water quality parameters of aquaculture water on growth conditions of aquaculture living beings | |
| CN103218669B (en) | A kind of live fish cultivation water quality comprehensive forecasting method of intelligence | |
| CN108614422B (en) | Method, device and system for optimally controlling dissolved oxygen in land-based factory circulating water aquaculture | |
| Zhou et al. | Modelling and controlling dissolved oxygen in recirculating aquaculture systems based on mechanism analysis and an adaptive PID controller | |
| CN107728477A (en) | A kind of industrialized aquiculture water quality dissolved oxygen prediction control method and system | |
| CN107169621A (en) | A kind of Dissolved Oxygen in Water Forecasting Methodology and device | |
| Mustafa et al. | A review of smart fish farming systems | |
| US20230144498A1 (en) | Simulation and automated control of physical systems | |
| CN116629550B (en) | Water environment supervision method and scheduling operation system based on cloud computing | |
| CN110119086A (en) | A kind of tomato greenhouse environmental parameter intelligent monitoring device based on ANFIS neural network | |
| Chahid et al. | Fish growth trajectory tracking using Q-learning in precision aquaculture | |
| Bracino et al. | Optimization of biofilter size for aquaponics using genetic algorithm | |
| An et al. | A simulator-based planning framework for optimizing autonomous greenhouse control strategy | |
| Lork et al. | Minimizing electricity cost through smart lighting control for indoor plant factories | |
| CN113962819A (en) | Method for predicting dissolved oxygen in industrial aquaculture based on extreme learning machine | |
| CN113545280A (en) | System and method for carrying out accurate irrigation based on plant wilting degree | |
| Nugroho et al. | Predictive control on lettuce NFT-based hydroponic IoT using deep neural network | |
| Xia et al. | Environmental factor assisted chlorophyll-a prediction and water quality eutrophication grade classification: A comparative analysis of multiple hybrid models based on a SVM | |
| Chen et al. | A water-saving irrigation decision-making model for greenhouse tomatoes based on genetic optimization TS fuzzy neural network | |
| Esmaeili et al. | Prediction of shrimp growth using an artificial neural network and regression models | |
| Wahjuni et al. | The fuzzy inference system for intelligent water quality monitoring system to optimize eel fish farming | |
| CN201830751U (en) | System for forecasting influence of water quality parameters of aquaculture water body over growth state of aquaculture organisms |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240119 Address after: 230000 floor 1, building 2, phase I, e-commerce Park, Jinggang Road, Shushan Economic Development Zone, Hefei City, Anhui Province Patentee after: Dragon totem Technology (Hefei) Co.,Ltd. Address before: 223005 Jiangsu Huaian economic and Technological Development Zone, 1 East Road. Patentee before: HUAIYIN INSTITUTE OF TECHNOLOGY |
|
| TR01 | Transfer of patent right |