An aquaculture feeding control method based on a two-type fuzzy model

Abstract

The application discloses a kind of based on two type fuzzy model's aquaculture feeding control method, comprising: division breeding area, initialize the density of aquaculture of each breeding area, and obtain the water quality parameter and water generation stage of each breeding area;According to the initialized density of aquaculture, the fuzzy language variable set of aquaculture density is counted, and is converted into aquaculture density fuzzy number by class gaussian fuzzy membership function;According to the water generation stage in breeding area and aquaculture density fuzzy number, determine the initial feeding amount fuzzy value in breeding area;On the basis of initial feeding amount fuzzy value in breeding area, two type fuzzy is carried out based on the evaluation result of water quality parameter, using mandani inference method, after reduction type and de-fuzzy, the feeding amount in breeding area is inferred;According to the feeding amount in breeding area, the length of waterway of breeding area and the running speed of feeding equipment are determined to determine feeding speed.The method realizes accurate feeding, reduces bait waste, saves cost.

Description

Technical Field

[0001] This invention relates to the field of aquaculture feeding control technology, specifically to an aquaculture feeding control method based on a type II fuzzy model. Background Technology

[0002] In aquaculture, feeding the water is a daily and crucial task. Traditional feeding operations rely heavily on manual labor or semi-automated equipment, which suffers from low efficiency, high labor intensity, uneven feed distribution, and difficulty adapting to adverse weather conditions. Currently, several automated feeding technologies and equipment have been proposed, such as timed and quantitative automatic feeders and remotely controlled or GPS-guided automated feeding vessels.

[0003] However, these existing technologies still have some shortcomings. For example, the feeding equipment is small in size and requires frequent feeding operations. However, the feeding operation requires a one-to-one correspondence between personnel and feeding equipment, and the amount of feed to be added is judged by human experience, which is quite difficult and requires a lot of manpower and resources. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention provides a feeding control method for aquaculture based on a type II fuzzy model, which achieves precise feeding, reduces feed waste, and saves costs.

[0005] To achieve the above technical objectives, the present invention adopts the following technical solution: a feeding control method for aquaculture based on a type-II fuzzy model, specifically including the following steps:

[0006] Step 1: Divide the aquaculture area according to the size of the aquaculture pond, initialize the aquaculture density of each aquaculture area, and obtain the water quality parameters and aquatic growth stage of each aquaculture area;

[0007] Step 2: Based on the initial aquaculture density, statistically analyze the set of fuzzy linguistic variables of aquaculture density, and convert them into fuzzy numbers of aquaculture density using a Gaussian-like fuzzy membership function.

[0008] Step 3: Determine the fuzzy value of the initial feeding amount in the aquaculture area based on the aquatic growth stage and the fuzzy number of aquaculture density in the aquaculture area;

[0009] Step 4: Based on the initial fuzzy value of the feeding amount in the breeding area, perform type II fuzzification based on the evaluation results of water quality parameters, and use the Mandani inference method to infer the feeding amount in the breeding area after type reduction and defuzzification.

[0010] Step 5: Determine the feeding speed based on the amount of feed given in the breeding area, the length of the waterway in the breeding area, and the operating speed of the feeding equipment.

[0011] Furthermore, the initialization process for the aquaculture density in each aquaculture area is as follows:

[0012]

[0013] Where, N total This represents the total number of aquatic seedlings introduced into the aquaculture pond at the start of aquaculture, and m represents the number of aquaculture zones.

[0014] Furthermore, the conversion process of the fuzzy number of aquaculture density in step 2 is as follows:

[0015]

[0016] in, N represents the fuzzy number of aquaculture density in the m-th aquaculture area after transformation by a Gaussian-like fuzzy membership function. m A fuzzy language set representing the aquaculture density of the m-th aquaculture area, N m ={N -lm N -(l-1)m ,…,N -1m N 0m N 1m …,N (l-1)m N lm}, where l represents the number of fuzzy linguistic subsets of aquaculture density based on the initial aquaculture density, and N -lm This represents the fuzzy linguistic language for aquaculture under the l-th fuzzy linguistic variable subset when the aquaculture density in the m-th aquaculture area is lower than the initial aquaculture density, N. lm This represents the fuzzy language for aquaculture under the l-th fuzzy language variable subset when the aquaculture density in the m-th aquaculture area is higher than the initial aquaculture density; Let γ represent the central value of the aquaculture fuzzy language under the j-th subset of fuzzy language variables, where γ and σ are both constants greater than 0.

[0017] Furthermore, step 3 includes the following sub-steps:

[0018] Step 3.1: For each aquaculture area, establish the correspondence between the fuzzy linguistic variable of aquaculture density and the fuzzy linguistic variable of initial feed amount at each stage of aquaculture growth;

[0019] Step 3.2: Before feeding, measure the aquaculture density in the aquaculture area and convert it into a fuzzy linguistic variable of aquaculture density based on the aquaculture density domain transformation ratio factor.

[0020] Step 3.3: Determine the corresponding fuzzy number of aquaculture density from the fuzzy linguistic variable of aquaculture density, and determine the fuzzy range of initial feeding amount in the established correspondence between the fuzzy linguistic variable of aquaculture density and the fuzzy linguistic variable of initial feeding amount.

[0021] Step 3.4: Find the centroid of the fuzzy range of the initial feed amount using the centroid method, and obtain the fuzzy value of the initial feed amount corresponding to the centroid of the fuzzy range of the initial feed amount.

[0022] Furthermore, the calculation process for the aquaculture density universe of discourse transformation scaling factor k1 in step 3.2 is as follows:

[0023]

[0024] Where, N max N represents the maximum value of the fuzzy language related to aquaculture density in the set of fuzzy language corresponding to aquaculture density. min The minimum value of the fuzzy language for aquaculture density in the set of fuzzy language representing the corresponding aquaculture density, n. max n represents the maximum aquaculture density within an aquaculture area. min This represents the minimum aquaculture density within an aquaculture area.

[0025] Furthermore, step 4 includes the following sub-steps:

[0026] Step 4.1: Establish a fuzzy language set for water quality within the aquaculture area, establish a principal membership curve based on the initial feeding amount within the aquaculture area, and convert the fuzzy language set for water quality into fuzzy numbers for water quality using a Gaussian-like fuzzy membership function;

[0027] Step 4.2: Use the water quality fuzzy numbers to perform secondary fuzzification on the principal membership curve to obtain a type II fuzzy set, and adjust the principal membership curve through the type II fuzzy set;

[0028] Step 4.3: Determine the evaluation results based on the water quality parameters collected before feeding, and convert the evaluation results into water quality fuzzy language through the evaluation result universe of discourse transformation ratio factor to determine the corresponding water quality fuzzy number.

[0029] Step 4.4: Fine-tune the adjusted principal membership curve based on the water quality fuzzy number, and re-determine the fuzzy range of feeding amount. After the model is reduced, find the centroid of the fuzzy range of feeding amount using the centroid method, and obtain the feeding fuzzy value corresponding to the centroid of the fuzzy range of feeding amount.

[0030] Step 4.5: Convert the adjusted fuzzy feeding value into the feeding amount by using the feeding amount universe transformation scaling factor.

[0031] Furthermore, the process of establishing the principal membership curve in step 4.1 is as follows:

[0032]

[0033] in, w represents the principal membership degree corresponding to the fuzzy value of the initial feed amount in the m-th aquaculture area. m wq represents the fuzzy value of the initial feed amount for the m-th aquaculture area. m This represents the fuzzy value of the assessment result of the water quality parameter in the m-th aquaculture area. Indicates w m with wq m The established membership function relationship;

[0034] Step 4.1 Water quality fuzzy data The transformation process is as follows:

[0035]

[0036] in, γ represents the central value of the subset of water quality fuzzy language, and γ2 and σ2 both represent constants greater than 0.

[0037] Furthermore, the type II fuzzy set in step 4.2 specifically refers to:

[0038]

[0039] in, Let W represent the type-2 fuzzy set of the m-th aquaculture region. m This indicates the amount of feed given to aquatic animals within the aquaculture area. This indicates that within the divided aquaculture areas, wq is implied. m w m The corresponding fuzzy subset of primary membership.

[0040] Furthermore, the calculation process for the feed quantity conversion ratio factor k2 in step 4.5 is as follows:

[0041]

[0042] Among them, w max w represents the fuzzy maximum value of feed amount in the principal membership curve. min W represents the fuzzy minimum value of feed amount in the principal membership curve. max W represents the maximum actual feed input within the aquaculture area. min This represents the minimum actual amount of feed provided within the aquaculture area.

[0043] Furthermore, the calculation process for the feeding rate in step 5 is as follows:

[0044]

[0045] Among them, v fm v represents the feeding rate in the m-th breeding area. bm L represents the operating speed of the feeding equipment in the m-th aquaculture area. m W represents the length of the waterway within the m-th aquaculture area. m This represents the actual amount of feed provided in the m-th breeding area.

[0046] Compared with existing technologies, the present invention has the following beneficial effects: The aquaculture feeding control method based on the type II fuzzy model of the present invention fuzzifies the different stocking densities and water quality parameter evaluation results of different aquaculture areas for different aquaculture growth stages. It generates an inference rule base based on expert language, which can describe the initial fuzzy value of feeding according to different stocking densities. At the same time, the water quality parameter evaluation results are used as secondary variables for precise feeding amount inference in the aquaculture area and secondary membership degrees are established. Furthermore, the relationship between water quality and feeding amount in expert language is transformed into the type II fuzzy model inference rule base, which better replicates the influence of water quality on feeding amount in expert experience. It effectively integrates expert experience into precise feeding control decision-making through the type II fuzzy model. At the same time, it performs variable-speed coupled feeding according to real-time feeding amount, which has the advantages of reducing feed waste and saving costs. Attached Figure Description

[0047] Figure 1 This is a flowchart of the feeding control method for aquaculture based on a type II fuzzy model according to the present invention;

[0048] Figure 2 This is a schematic diagram of the unmanned surface vessel.

[0049] Figure 3 This is a schematic diagram of the material throwing disc. Detailed Implementation

[0050] The technical solution of the present invention will be further explained and described below with reference to the accompanying drawings.

[0051] like Figure 1 This is a flowchart of the aquaculture feeding control method based on a type II fuzzy model according to the present invention. The aquaculture feeding control method specifically includes the following steps:

[0052] Step 1: Divide the aquaculture area according to the size of the aquaculture pond, initialize the aquaculture density of each aquaculture area, and obtain the water quality parameters and aquatic growth stage of each aquaculture area;

[0053] The initialization process for the aquaculture density in each aquaculture area in this invention is as follows:

[0054]

[0055] Where, Ntotal This represents the total number of aquatic seedlings introduced into the aquaculture pond at the start of aquaculture, and m represents the number of aquaculture zones.

[0056] Step 2: Based on the initial aquaculture density, statistically analyze the fuzzy linguistic variable set of aquaculture density. Convert it into a fuzzy number of aquaculture density using a Gaussian-like fuzzy membership function. Based on the expert linguistic variable set, the fuzzy linguistic variable set of aquaculture density can be directly divided, which can accurately replicate expert experience without conducting a preliminary experiment to confirm the relationship between feed amount and aquaculture effect.

[0057] The transformation process of the fuzzy number of aquaculture density in this invention is as follows:

[0058]

[0059] in, N represents the fuzzy number of aquaculture density in the m-th aquaculture area after transformation by a Gaussian-like fuzzy membership function. m A fuzzy language set representing the aquaculture density of the m-th aquaculture area, N m ={N -lm N -(l-1)m ,…,N -1m N 0m N 1m …,N (l-1)m N lm}, where l represents the number of fuzzy linguistic subsets of aquaculture density based on the initial aquaculture density, and N -lm This represents the fuzzy linguistic language for aquaculture under the l-th fuzzy linguistic variable subset when the aquaculture density in the m-th aquaculture area is lower than the initial aquaculture density, N. lm This represents the fuzzy language for aquaculture under the l-th fuzzy language variable subset when the aquaculture density in the m-th aquaculture area is higher than the initial aquaculture density; Let γ represent the central value of the aquaculture fuzzy language under the j-th subset of fuzzy language variables, where γ and σ are both constants greater than 0.

[0060] Step 3: Determine the initial fuzzy value of the feeding amount within the aquaculture area based on the aquatic growth stage and the fuzzy number of aquaculture density. Although aquatic seedlings are uniformly stocked, due to some unknown reasons, the aquatic organisms often do not aggregate at a uniform density in the pond, but rather at different densities in different areas. For some territorial aquatic products, applying different feeding amounts based on different aggregation densities can effectively prevent water quality deterioration caused by excessive feeding in some areas, or limb damage caused by insufficient feeding due to competition for food. By extracting fuzzy inference engines from expert experience, it is possible to quickly achieve precise feeding without prior research into the causes of aquaculture density formation in different aquaculture areas of the same pond. This specifically includes the following sub-steps:

[0061] Step 3.1: For each aquaculture area, establish the correspondence between the fuzzy linguistic variable of aquaculture density and the fuzzy linguistic variable of initial feed amount at each stage of aquaculture growth;

[0062] Step 3.2: Before feeding, the aquaculture density in the aquaculture area is measured, and the density is converted into a fuzzy linguistic variable based on the aquaculture density universe of discourse transformation proportionality factor; the calculation process of the aquaculture density universe of discourse transformation proportionality factor k1 in this invention is as follows:

[0063]

[0064] Where, N max N represents the maximum value of the fuzzy language related to aquaculture density in the set of fuzzy language corresponding to aquaculture density. min The minimum value of the fuzzy language for aquaculture density in the set of fuzzy language representing the corresponding aquaculture density, n. max n represents the maximum aquaculture density within an aquaculture area. min This represents the minimum aquaculture density within an aquaculture area.

[0065] Step 3.3: Determine the corresponding fuzzy number of aquaculture density from the fuzzy linguistic variable of aquaculture density, and determine the fuzzy range of initial feeding amount in the established correspondence between the fuzzy linguistic variable of aquaculture density and the fuzzy linguistic variable of initial feeding amount.

[0066] Step 3.4: Find the centroid of the fuzzy range of the initial feed amount using the centroid method, and obtain the fuzzy value of the initial feed amount corresponding to the centroid of the fuzzy range of the initial feed amount.

[0067] Step 4: Based on the initial fuzzy value of the feeding amount in the aquaculture area and the evaluation results of water quality parameters, perform type II fuzzification. Using the Mandani inference method, after type reduction and defuzzification, infer the feeding amount in the aquaculture area. Feeding behavior usually leads to water quality deterioration, and this deterioration usually has a significant delay, ranging from 1 to 6 hours. If water quality conditions are not considered when feeding, it may lead to reduced production or even death of aquatic products due to water quality deterioration. Therefore, based on the inference engine formed by extracting expert experience, fuzzification is performed through the comprehensive evaluation of current data of multiple water quality parameters, and this is used for precise control of feeding amount. This characterizes the predictive mechanism of the impact of feeding on water quality in the future without the need for pre-experimentation for actual model building. Specifically, it includes the following sub-steps:

[0068] Step 4.1: Establish a fuzzy language set for water quality within the aquaculture area, establish a principal membership curve based on the initial feeding amount within the aquaculture area, and convert the fuzzy language set for water quality into fuzzy numbers for water quality using a Gaussian-like fuzzy membership function;

[0069] The process of establishing the principal membership curve is as follows:

[0070]

[0071] in, w represents the principal membership degree corresponding to the fuzzy value of the initial feed amount in the m-th aquaculture area. m wq represents the fuzzy value of the initial feed amount for the m-th aquaculture area. m This represents the fuzzy value of the assessment result of the water quality parameter in the m-th aquaculture area. Indicates w m with wq m The established membership function relationship;

[0072] Water quality fuzzy data The transformation process is as follows:

[0073]

[0074] in, γ represents the central value of the subset of water quality fuzzy language, and γ2 and σ2 both represent constants greater than 0.

[0075] Step 4.2: Use the water quality fuzzy numbers to perform secondary fuzzification on the principal membership curve to obtain a type II fuzzy set, and adjust the principal membership curve through the type II fuzzy set;

[0076] Type II fuzzy sets are specifically:

[0077]

[0078] in, Let W represent the type-2 fuzzy set of the m-th aquaculture region. m This indicates the amount of feed given to aquatic animals within the aquaculture area. This indicates that within the divided aquaculture areas, wq is implied. m w m The corresponding fuzzy set of primary membership degrees.

[0079] Step 4.3: Determine the evaluation results based on the water quality parameters collected before feeding, and convert the evaluation results into water quality fuzzy language through the evaluation result universe of discourse transformation ratio factor to determine the corresponding water quality fuzzy number.

[0080] The calculation process of the evaluation result universe transformation scaling factor k3 in this invention is as follows:

[0081]

[0082] Among them, wq max wq represents the maximum fuzzy value of the water quality parameter assessment results. min WQ represents the minimum fuzzy value of the water quality parameter assessment results. max WQ represents the maximum value of the water quality parameter assessment results. min This represents the minimum value of the assessment result for water quality parameters.

[0083] Step 4.4: Fine-tune the adjusted principal membership curve based on the water quality fuzzy number, and re-determine the fuzzy range of feeding amount. After the model is reduced, find the centroid of the fuzzy range of feeding amount using the centroid method, and obtain the feeding fuzzy value corresponding to the centroid of the fuzzy range of feeding amount.

[0084] Step 4.5: Convert the adjusted fuzzy feeding value into the feeding amount by using the feeding amount universe of discourse transformation scaling factor;

[0085] The calculation process for the feed quantity conversion ratio factor k2 is as follows:

[0086]

[0087] Among them, w max w represents the fuzzy maximum value of feed amount in the principal membership curve. min W represents the fuzzy minimum value of feed amount in the principal membership curve. max W represents the maximum actual feed input within the aquaculture area. min This represents the minimum actual amount of feed provided within the aquaculture area.

[0088] Step 5: Determine the feeding rate based on the amount of feed given in the aquaculture area, the length of the waterway in the aquaculture area, and the operating speed of the feeding equipment.

[0089]

[0090] Among them, v fm v represents the feeding rate in the m-th breeding area. bm L represents the operating speed of the feeding equipment in the m-th aquaculture area. m W represents the length of the waterway within the m-th aquaculture area. m This represents the actual amount of feed provided in the m-th breeding area.

[0091] This invention provides a feeding control method for aquaculture based on a type II fuzzy model. It can fuzzify different stocking densities in different aquaculture areas and the evaluation results of water quality parameters for different aquaculture growth stages. It can better simulate the expert experience of farmers in feeding control, thereby obtaining more accurate feeding amounts. At the same time, it can reduce feed waste and save costs by using variable speed coupling feeding based on real-time feeding amounts.

[0092] The feeding equipment used in this invention can be an unmanned surface vessel (USV) or an unmanned aerial vehicle (UAV). The following description uses an USV as an example, illustrating its specific structure. Figure 2-3 As shown, the unmanned vessel 100 is equipped with a mounting frame 6. The storage hopper 1 is fixedly installed on the upper layer of the mounting frame 6, the feeding box 2 is fixedly installed on the middle layer of the mounting frame 6, and the throwing disc 3 and the throwing motor 5 are fixedly installed on the lower layer of the mounting frame 6. The throwing motor 5 is fixedly installed on the unmanned vessel 100. By setting the mounting frame 6, the storage hopper 1, the feeding box 2, the throwing disc 3, and the throwing motor 5 are securely set on the unmanned vessel 100 to avoid shaking. At the same time, the relative position relationship of the four can be maintained. The edge of the throwing disc 3 is flush with the edge of the unmanned vessel 100, which can prevent the throwing disc 3 from throwing feed onto the unmanned vessel 100 when throwing feed. The discharge port at the bottom of the storage hopper 1 is connected to the feeding box 2. A control valve is provided at the connection between the storage hopper 1 and the feeding box 2. The throwing plate 3 is connected to the feeding box 2 through the feed extractor 4. Specifically, the throwing plate 3 is provided with a throwing cover 7. The throwing cover 7 has a through hole in the middle. The throwing cover 7 is fixedly connected to the discharge port of the feed extractor 4 through the through hole. The throwing cover 7 is provided with a throwing port 8. The diameter of the throwing cover 7 is slightly larger than the diameter of the throwing plate 3. When the throwing plate 3 rotates, the feed is thrown out from the throwing port 8. The feed in the feeding box 2 enters the throwing plate 3 through the feed extractor 4. The throwing cover 7 with the throwing port 8 is set above the throwing plate 3 to ensure that the feed is only thrown out from the throwing port 8, which plays a limiting role on other feed in the throwing plate 3. The feeding disc 3 is equipped with several baffles 9 evenly spaced within it. The baffles 9 are arranged in a circular pattern, and the interval between the baffles 9 is the same as the length of the feeding port 8. By setting the baffles 9 within the feeding disc 3, the feed can be distributed in the spaces separated by the baffles 9, preventing excessive feed accumulation in the feeding disc 3 and clogging of the feeding port 8. A feeding motor 5 is installed at the bottom of the feeding disc 3. The output shaft of the feeding motor 5 is fixedly connected to the feeding disc 3, and the feeding motor 5 is used to drive the feeding disc 3 to rotate.

[0093] Meanwhile, the unmanned vessel also features an automatic loading control system, including:

[0094] The detection module can obtain the position of the unmanned vessel 100 in real time when it travels along the preset route, and obtain the aquaculture density, water quality parameters and aquatic growth stage of the aquaculture area where the vessel is located from the monitoring equipment in the pond area.

[0095] The analysis module is used to calculate the feeding amount using the type-two fuzzy aquaculture feeding control method proposed in this invention;

[0096] The feed determination module is used to obtain the feed storage amount in the feeding box 2 and determine the feed storage difference between the feeding amount and the feed storage amount.

[0097] The feeding control module is used to determine the working status of the control valve based on the difference in feed storage, so as to enable the feeding hopper 1 to feed the feeding box 2.

[0098] In the embodiments disclosed in this application, a computer storage medium may be a tangible medium that may contain or store programs for use by or in conjunction with an instruction execution system, apparatus, or device. The computer storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of computer storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0099] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this application can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0100] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principle of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A feeding control method for aquaculture based on a type-two fuzzy model, characterized in that, Specifically, the steps include the following: Step 1: Divide the aquaculture area according to the size of the aquaculture pond, initialize the aquaculture density of each aquaculture area, and obtain the water quality parameters and aquatic growth stage of each aquaculture area; The initialization process for aquaculture density in each aquaculture area is as follows: in, This indicates the total number of aquatic seedlings introduced into the aquaculture pond at the start of aquaculture. m Indicates the number of aquaculture zones; Step 2: Based on the initial aquaculture density, statistically analyze the set of fuzzy linguistic variables of aquaculture density, and convert them into fuzzy numbers of aquaculture density using a Gaussian-like fuzzy membership function. The conversion process of the fuzzy number of aquaculture density is as follows: in, This indicates the result after transformation using a Gaussian-like fuzzy membership function. m Fuzzy data on aquaculture density in aquaculture areas Indicates the first m A fuzzy set of linguistic terms representing the aquaculture density of a particular aquaculture area. , l This represents the number of fuzzy linguistic subsets of aquaculture density based on the initial aquaculture density. Indicates the first m When the aquaculture density in a certain aquaculture area is lower than the initial aquaculture density, the first... l Fuzzy language for aquaculture under a subset of fuzzy linguistic variables Indicates the first m When the aquaculture density in a certain aquaculture area is higher than the initial aquaculture density, the first... l Fuzzy language for aquaculture under a subset of fuzzy language variables; Indicates the first j The central value of the fuzzy language of aquaculture under a subset of fuzzy linguistic variables. Both represent constants greater than 0; Step 3: Determine the fuzzy value of the initial feeding amount in the aquaculture area based on the aquatic growth stage and the fuzzy number of aquaculture density in the aquaculture area; Step 4: Based on the initial fuzzy value of the feed amount in the aquaculture area and the evaluation results of water quality parameters, perform type II fuzzification. Using the Mandani inference method, after type reduction and defuzzification, infer the feed amount in the aquaculture area; including the following sub-steps: Step 4.1: Establish a fuzzy language set for water quality within the aquaculture area, establish a principal membership curve based on the initial feeding amount within the aquaculture area, and convert the fuzzy language set for water quality into fuzzy numbers for water quality using a Gaussian-like fuzzy membership function; The process of establishing the principal membership curve is as follows: in, Indicates the first m The principal membership degree corresponding to the fuzzy value of the initial feed amount in each breeding area. Indicates the first m Fuzzy values ​​for initial feed amounts in each aquaculture area Indicates the first m The assessment results of water quality parameters for each aquaculture area are fuzzy values. express and The established membership function relationship; The water quality fuzzy number The transformation process is as follows: in, The central value represents a subset of the fuzzy language representation of water quality. Both represent constants greater than 0; Step 4.2: Use the water quality fuzzy numbers to perform secondary fuzzification on the principal membership curve to obtain a type II fuzzy set, and adjust the principal membership curve through the type II fuzzy set; Step 4.3: Determine the evaluation results based on the water quality parameters collected before feeding, and convert the evaluation results into water quality fuzzy language through the evaluation result universe of discourse transformation ratio factor to determine the corresponding water quality fuzzy number. Step 4.4: Fine-tune the adjusted principal membership curve based on the water quality fuzzy number, and re-determine the fuzzy range of feeding amount. After the model is reduced, find the centroid of the fuzzy range of feeding amount using the centroid method, and obtain the feeding fuzzy value corresponding to the centroid of the fuzzy range of feeding amount. Step 4.5: Convert the adjusted fuzzy feeding value into the feeding amount by using the feeding amount universe of discourse transformation scaling factor; Step 5: Determine the feeding speed based on the amount of feed given in the breeding area, the length of the waterway in the breeding area, and the operating speed of the feeding equipment.

2. The aquaculture feeding control method based on a type-two fuzzy model according to claim 1, characterized in that, Step 3 includes the following sub-steps: Step 3.1: For each aquaculture area, establish the correspondence between the fuzzy linguistic variable of aquaculture density and the fuzzy linguistic variable of initial feed amount at each stage of aquaculture growth; Step 3.2: Before feeding, measure the aquaculture density in the aquaculture area and convert it into a fuzzy linguistic variable of aquaculture density based on the aquaculture density domain transformation ratio factor. Step 3.3: Determine the corresponding fuzzy number of aquaculture density from the fuzzy linguistic variable of aquaculture density, and determine the fuzzy range of initial feeding amount in the established correspondence between the fuzzy linguistic variable of aquaculture density and the fuzzy linguistic variable of initial feeding amount. Step 3.4: Find the centroid of the fuzzy range of the initial feed amount using the centroid method, and obtain the fuzzy value of the initial feed amount corresponding to the centroid of the fuzzy range of the initial feed amount.

3. The aquaculture feeding control method based on a type-two fuzzy model according to claim 2, characterized in that, Step 3.2 Aquaculture Density Universe Transformation Ratio Factor The calculation process is as follows: in, The maximum value of the fuzzy language representing aquaculture density in the corresponding fuzzy language set. The minimum value of the fuzzy language representing aquaculture density in the set of fuzzy language representing the corresponding aquaculture density. This indicates the maximum aquaculture density within the aquaculture area. This represents the minimum aquaculture density within an aquaculture area.

4. The aquaculture feeding control method based on a type-two fuzzy model according to claim 1, characterized in that, In step 4.2, the type II fuzzy set specifically refers to: in, Indicates the first m Type II fuzzy sets of aquaculture areas This indicates the amount of feed given to aquatic animals within the aquaculture area. This indicates that within the divided aquaculture areas, there is a hidden potential. of The corresponding fuzzy subset of primary membership.

5. The aquaculture feeding control method based on a type-two fuzzy model according to claim 1, characterized in that, Step 4.5 Feeding Amount Conversion Ratio Factor The calculation process is as follows: in, This represents the fuzzy maximum value of the feeding amount in the principal membership curve. This represents the fuzzy minimum value of feed intake in the principal membership curve. This represents the maximum actual amount of feed fed within the aquaculture area. This represents the minimum actual amount of feed provided within the aquaculture area.

6. The aquaculture feeding control method based on a type-two fuzzy model according to claim 1, characterized in that, The calculation process for the feeding rate in step 5 is as follows: in, Indicates the first m Feeding rate in each breeding area Indicates the first m The operating speed of feeding equipment in each breeding area Indicates the first m The length of the waterway within each aquaculture area Indicates the first m The actual amount of feed fed in each breeding area.

Images