A method for controlling the rotation speed of an automatically adjustable screw conveyor for multi-layer bullfrog feed dispensing in a breeding box.

Abstract

This invention provides a method for controlling the rotation speed of an automatically adjustable screw conveyor for multi-layer bullfrog feed dispensing in a breeding tank. By using sensors to collect the number and range of bullfrog jumps before feeding, and connecting the sensors to a computer to generate images and data, the activity level of the bullfrogs is analyzed from the images and data. Simultaneously, combined with the bullfrog weight obtained from the sensors, a comprehensive analysis is performed to determine the appropriate amount of bullfrog feed to be dispensed. Based on this quantitative analysis, precise feeding of the bullfrogs is achieved, solving the problem that existing technologies relying on breeding experience or traditional feeding machines often result in insufficient or excessive feeding.

Description

Technical Field

[0001] This invention relates to the field of bullfrog farming technology, and in particular to a method for controlling the rotational speed of an automatically adjustable screw conveyor for multi-layer bullfrog feed dispensing in a breeding box. Background Technology

[0002] First introduced to my country in the late 1950s, bullfrog farming has flourished nationwide after nearly 30 years of development, overcoming the challenges of dietary domestication. Bullfrogs have high utilization value due to their edible (fast growth, delicious taste, rich nutrition, high protein content), medicinal (internal organs can be used in medicine), and other uses. Today, farmed bullfrogs, with their high economic value, have become a major product of specialized aquaculture. However, observations have revealed that existing bullfrog farming methods rely on traditional high-density farming, facing numerous problems such as large land area requirements, high input but low output, and wasted labor. Based on this, the inventors have previously applied for an automatic three-dimensional bullfrog farming device and method (application number: 2025101977040). In this application, a spiral feed dispensing system is installed from top to bottom at the center of a multi-layered farming tank, automatically dispensing feed through rotation. Currently, when using this feeding system, farmers mainly rely on experience to determine the size of the bullfrogs and formulate feeding plans based on visual assessment. They use artificial feed, which cannot accurately and effectively feed the bullfrogs according to their specific growth status and activity level. Feeding based solely on visual judgment and farmer experience easily leads to overfeeding or underfeeding, both of which can negatively impact the bullfrogs' normal growth. Summary of the Invention

[0003] To address the existing technical problems, the main objective of this invention is to provide a method for controlling the rotational speed of an automatically adjustable spiral conveyor for multi-layer bullfrog feed distribution in a breeding box. This algorithm quantitatively analyzes the feeding amount based on the bullfrog's weight and activity level, evaluates and adjusts the feed amount, and then achieves precise feeding by accurately controlling the spiral conveyor of the spiral feed distribution system. This ensures the quality of bullfrog rearing while effectively improving feed utilization, thus achieving scientific feeding.

[0004] To achieve the above-mentioned technical features, the objective of this invention is as follows: 1. A method for controlling the rotational speed of an automatically adjustable screw conveyor for multi-layer bullfrog feed dispensing in a breeding box, characterized in that it includes:

[0005] At each stage of the entire bullfrog feeding process, the distribution of the number of bullfrog jumps and the weight data of each bullfrog were obtained through pressure sensors below the bullfrog breeding area.

[0006] The activity level of the bullfrogs per minute is calculated based on the actual number of jumps per minute, the maximum number of jumps, and the minimum number of jumps per minute.

[0007] Calculate the amount of bullfrog feed to be obtained based on the actual number of bullfrogs released, the weight and activity level of each bullfrog;

[0008] The location of the bullfrogs is estimated, and the feeding amount is controlled by an automatic adjustment screw conveyor to feed the bullfrogs.

[0009] Preferably, the entire process of feeding the bullfrog includes before and after feeding;

[0010] The steps for obtaining pressure sensor data below the bullfrog breeding area include: dividing the area into several small areas of a certain size and number according to the number of bullfrogs; installing pressure sensors in each small area; connecting the pressure sensors to a computer to obtain bullfrog activity data; and analyzing and processing the data. The bullfrog activity data includes the number of jumps per minute per bullfrog and the weight of each bullfrog.

[0011] Preferably, the acquisition and analysis of bullfrog activity data includes the following steps:

[0012] Preliminary analysis of the data was conducted to obtain the maximum and minimum number of jumps per minute of bullfrogs and the mass of each bullfrog. The average mass was then compared with the normal growth curve of bullfrogs of the same mass to preliminarily determine the growth status of the bullfrogs.

[0013] The data analysis and processing involves: using a normalization algorithm to map the number of jumps per minute for each bullfrog in each layer to an interval; statistically analyzing the bullfrog jump count data over multiple time periods; and selecting the minimum and maximum number of jumps for each layer of data. and Then, a linear mapping is used to calculate the number of jumps of the bullfrog at each level within a certain time period. In the interval Variations within a range are mapped to activity levels. Finally, the activity level of the bullfrogs at each level is calculated using a normalization formula, as shown below: In the formula The data processing represents the activity level of the bullfrogs at each level. The sum of the weights of all the bullfrogs is then divided by the total number of bullfrogs. To obtain the average weight of the bullfrogs: ;

[0014] In the formula, i The number of bullfrogs per layer; j The number of layers in the bullfrog enclosure; This represents the quality of each bullfrog. This represents the average weight of the bullfrogs in each layer.

[0015] Preferably, before obtaining the area of ​​each bullfrog's location, the following steps are included:

[0016] The data acquired by the pressure sensor is transmitted to a computer for edge analysis to obtain the edge information of the location of each bullfrog.

[0017] The edge information of the bullfrogs is dilated and optimized, and the contour of the region where each bullfrog is located is extracted from the optimized edge information.

[0018] Preferably, obtaining the median outline area of ​​each bullfrog includes the following steps:

[0019] Multiple cropping operations were performed on different numerical regions based on the pressure sensor data to obtain multiple different numerical regions.

[0020] The median value of the obtained numerical regions was selected as the median regional location value for each bullfrog.

[0021] Preferably, determining the amount of bullfrog feed includes the following steps:

[0022] A quantitative model for bullfrog feed administration is established. The quantitative model for bullfrog feed administration is as follows:

[0023] ;

[0024] in, This represents the amount of feed given to each bullfrog at that location on each floor. This represents the average weight of the bullfrogs in each layer. This represents the basic feeding amount for each bullfrog. This represents the quality of each bullfrog. This represents the activity level correction factor. Represents activity level;

[0025] The average weight of bullfrogs in each layer, the basic feed amount for each bullfrog, the mass of each bullfrog, the activity correction coefficient, and the activity level are input into the bullfrog feed quantification model to calculate the feed amount for each bullfrog.

[0026] Preferably, the following algorithm is used to estimate the bullfrog's positional state before determining the number of jumps and activity level of the bullfrogs at each level:

[0027] The recursive Bayesian filtering algorithm based on the Monte Carlo method, also known as the particle filter algorithm, approximates the probability distribution of the system state using a set of particles. Each particle represents a hypothesis in the space, and the weight of the particle represents the probability of that hypothesis. The objective function is established by weighting and updating the sampling points and combining basic data obtained from the bottom sensor and infrared sensor. Estimate the bullfrog's state coordinates;

[0028] in: Represents energy consumption. A coefficient representing the weighting of energy consumption control. This represents the difference between the actual amount of feed delivered to the target location and the expected amount.

[0029] Preferably, the establishment of the particle set under each weight includes the following steps:

[0030] Random initialization Particles Each particle has an initial weight. ;

[0031] The system tracks the position and speed of each bullfrog and predicts its state based on a dynamic model and sensor measurements. Update the weight of each particle;

[0032] Repeat the sampling step, resample according to the particle weights, and generate a new particle set.

[0033] Preferably, the specific steps for establishing the objective function are as follows:

[0034] Establish the system's state vector;

[0035] Establish a dynamic state transition function;

[0036] Create an update function;

[0037] A dynamic adjustment mechanism controls the rotational speed and establishes constraints.

[0038] The objective function is defined as the sum of the feeding errors at each layer, as well as the comprehensive objective of factors such as energy consumption. The objective function is then calculated.

[0039] Preferably, obtaining the state vector, dynamic state transition function, and update function includes the following steps:

[0040] Establish the system's state vector directly from sensor data:

[0041] ;

[0042] in , Representing bullfrogs The position at that moment; , Representing bullfrogs Time along and The speed of the shaft;

[0043] Establish the state transition of the bullfrog using classical equations of motion:

[0044] ;

[0045] ;

[0046] in: Represents the time step; Represents external control input; Representing bullfrogs Acceleration in the direction of; Representing bullfrogs Acceleration in the direction of; Represents the dynamic state transition function; Representative at and displacement caused by velocity in the direction; Representative at Displacement caused by velocity in a certain direction;

[0047] Suppose a nonlinear mapping function This describes the transition from the state space to the observation space, and updates the weight of each particle according to the measurement probability formula to perform real-time tracking and prediction of the bullfrog's position:

[0048] ;

[0049] in: The formula representing the probability of measurement; Represents measurement noise; Represents a nonlinear function, indicating the state. to the observed value Mapping;

[0050] The specific functions for establishing the constraints are as follows:

[0051] ;

[0052] in: This represents the height of the baffle. Represents the updated rotational speed; Represents the air drag coefficient; This represents the maximum height of the baffle. This represents the rotational speed of the device in the previous operation; This represents the distance of the bullfrog from the center of the circle; This represents the current rotational speed of the screw mechanism; This represents the feeding range of the spiral feeder; This represents the difference between the amount of feed given and the amount required. This represents the actual amount of feed delivered to that location by the feeder. This represents a self-defined tolerance threshold to avoid over-adjustment.

[0053] The present invention has the following beneficial effects:

[0054] 1. This invention collects the correlation coefficient of bullfrog weight and the bullfrog activity deviation index before feeding, and combines them with water temperature and feeding coefficient to comprehensively analyze and determine the amount of bullfrog feed. Based on the quantitative analysis of the feeding amount, the bullfrog feed can be accurately fed, which can solve the problem that the existing technology of relying on breeding experience or traditional feeding machines can easily lead to insufficient or excessive feeding. Attached Figure Description

[0055] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0056] Figure 1 Flowchart for starting the spiral feeder.

[0057] Figure 2 This is a mechanical structure diagram of a screw feeder. Detailed Implementation

[0058] To make the objectives, innovations, application areas, and operation methods of this invention clearer and more transparent, the invention will be further described in detail below with reference to embodiments and additional flowcharts.

[0059] Compared with traditional feeding methods, the screw conveyor feeding device can dynamically adjust the rotation speed according to the height of each baffle and the feed demand, which can more accurately control the feeding range and amount of each layer of feed, so that the feed can be delivered to the vicinity of the bullfrog, overcoming the problems of feed waste or uneven feeding of bullfrogs in traditional methods.

[0060] The algorithm of this invention can be implemented by a PLC control system. The control system can monitor the weight of the bullfrogs, their activity level, and the approximate location coordinates of each bullfrog in real time using sensors, and automatically adjust the rotation speed of the screw conveyor using the algorithm of this invention. The specific operation process is as follows:

[0061] Step 1: Estimating the weight of each bullfrog on each floor and the average weight of the bullfrogs:

[0062] First, the area where each layer of bullfrogs is located can be divided into a reasonable number of smaller zones based on the overall size of the facility. A pressure sensor is installed in each zone. If a bullfrog is positioned precisely in that zone, the sensor at that location will record data; conversely, if the bullfrog jumps to another location, the sensor at that location will display zero data. Using this method, the weight of each bullfrog in each layer of the breeding system can be obtained at any time. By comparing the bullfrogs' normal weight after feeding, it can be determined whether feeding is necessary during this period.

[0063] Second, based on the data collected by the aforementioned sensors, add up the weights of all the bullfrogs and divide by the total number of bullfrogs. That is, the average weight of the bullfrog. By comparing the average weight of bullfrogs, the weight of each bullfrog, and the weight of bullfrogs after a normal feeding, farmers can set a basic feeding amount based on this data. .

[0064] Step 2: Acquiring Activity Points :

[0065] First, according to relevant research, determining whether a bullfrog needs to feed can be done through a comprehensive analysis of several aspects. Firstly, changes in the size and shape of the bullfrog's abdomen are an important indicator; after feeding, the bullfrog's abdomen will noticeably swell, and its weight will increase; conversely, a shriveled abdomen indicates that the bullfrog has not eaten. Secondly, the feeding cycle is also a key factor; the bullfrog's feeding interval is closely related to its physiological needs. Finally, the activity level of the bullfrogs in each layer... This can also provide a reference for judgment. If the bullfrog's activity frequency is high, it usually means that it is in a state of searching for food, and this active behavior is likely to have a certain impact on the amount of food given. Therefore, by comprehensively observing the changes in the above aspects, it is possible to effectively assess whether the bullfrog needs to be fed.

[0066] Secondly, based on the above, if a bullfrog moves from one place to another, the pressure sensor data will change once. Based on this, the number of times the pressure sensor data changes within a few minutes can be counted. Then, using relevant mathematical knowledge, the activity level of the bullfrogs in that layer per minute can be obtained.

[0067] Third, in these examples, a normalization algorithm is used to map the number of jumps per minute of each layer of bullfrogs to intervals. The specific steps are as follows: First, statistically analyze the data of the bullfrog in each layer over multiple time periods to make the results more universal. Second, select the minimum and maximum number of jumps of the bullfrog in each layer of data, which are respectively... and Then, a linear mapping is used to calculate the number of jumps of the bullfrog at each level within a certain time period. In the interval Variations within a range are mapped to activity levels. Finally, the activity level of the bullfrogs at each level is calculated using a normalization formula, as shown below:

[0068] In the formula This represents the activity level of the bullfrogs on each floor.

[0069] Since the activity coefficient cannot completely control the required feed amount, an activity correction coefficient can be set based on empirical rules or knowledge in bullfrog farming by observing similar situations. When its activity level changes significantly compared to before, it may need to be increased. This makes the number of jumps more sensitive to changes; while when the activity level changes steadily, it can reduce... To avoid over-adjustment.

[0070] Step 3: Feeding amount for each bullfrog at each level:

[0071] Based on the aforementioned data on the weight of each bullfrog, the average weight of the bullfrogs, the activity level of the bullfrogs at each level, and the pre-set basic feeding amount, the feeding amount for each location can be determined using the following formula. This method can precisely control the food intake of each bullfrog, thereby effectively ensuring their health, and also allows for real-time monitoring of the bullfrogs' activity. The specific calculation formula is as follows:

[0072] In the formula This represents the amount of feed given to each bullfrog at that location on each floor. This represents the activity level of each layer of bullfrogs after the activity coefficient adjustment, which is the final activity value.

[0073] Step 4: Estimating the bullfrog's position and state using the particle filter algorithm:

[0074] Because the bullfrog's movement cannot remain stable at all times, it may suddenly jump to another location while the machine is determining its position and transmitting the relevant data to the computer, thus causing data errors. Furthermore, since the particle filter algorithm excels at handling nonlinear and non-Gaussian noise systems and can effectively solve nonlinear problems, our group chose a recursive Bayesian filtering algorithm based on the Monte Carlo method, namely the particle filter algorithm. This algorithm, by weighting and updating the sampling points and combining basic data obtained from the bottom sensor and infrared sensor, can effectively estimate the bullfrog's state coordinates, thereby reducing data errors and improving the accuracy and reliability of the system.

[0075] The basic idea of ​​a particle filter is to approximate the probability distribution of the system state using a set of particles. Each particle represents a hypothesis in the space, and the weight of the particle represents the probability of that hypothesis. First, random initialization... Particles Each particle has an initial weight. Next, the state of the bullfrog system is determined by tracking the position and speed of each bullfrog and predicting its state based on a dynamic model. This is done based on measurements provided by the sensors. The system updates the weights of each particle, then repeats the sampling step, resampling according to the particle weights to generate a new particle set. This way, particles with larger weights receive more samples, improving the accuracy of the system's bullfrog state estimation and ensuring effective filtering of inaccurate data and noise. Detailed explanation follows:

[0076] Definition of the system's state vector: ,in , Indicates bullfrogs in The position at that moment; , Indicates bullfrogs in Time along and The speed of the shaft.

[0077] Dynamic state transition function Establishment: The state function describes the evolution of the system from one time step to the next. The state transition of the bullfrog can be established using classical equations of motion, especially in the absence of external control (i.e., the bullfrog's movement is controlled by natural factors). The specific formula is as follows:

[0078] ;

[0079] Left side and They represent in and Displacement caused by velocity in the direction It is the time step. This refers to external control inputs, such as the movement control of a bullfrog, but under natural conditions, additional control inputs are usually not needed. The right-hand side indicates that if the bullfrog's speed is affected by some external force (such as flow velocity, slope, etc.), the control input can be represented as shown in the right-hand side, where... and Indicates bullfrogs in and Acceleration in the direction of.

[0080] Establishing the update function: In particle filtering, the measurement model can also be expressed as a conditional probability, the usual form of which is nonlinear. Assume a nonlinear mapping function... This describes the transition from the state space to the observation space, and updates the weight of each particle according to the measurement probability formula to perform real-time tracking and prediction of the bullfrog's position. The specific calculation formula is as follows: ;

[0081] in, It is a nonlinear function that represents the state. to the observed value Mapping (e.g., mapping the bullfrog's location to image coordinates captured by the camera). It is a measurement of noise, which can be assumed to follow a certain known probability distribution (such as a Gaussian distribution). This represents the formula for the probability of measurement.

[0082] Repeat sampling step: Resample according to the particle weights to generate a new particle set. At this time, the particle weights are reinitialized. .

[0083] Step 5: Dynamic adjustment mechanism controls speed

[0084] Since the bullfrogs in the entire feeding device are located inside the rearing box, and the spiral feeder is located at the center of each layer and runs through both layers, the feeding area must be the area centered on the center of the rearing box. Therefore, this invention introduces a real-time feedback mechanism that dynamically adjusts the rotation speed to ensure that each bullfrog in each layer can obtain the appropriate amount of feed, avoiding uneven nutrition. In summary, this invention uses constraints such as the feeding range being greater than or equal to the distance of the bullfrog from the center, feed error distribution, and the influence of baffle height as constraints, and minimizes the total error as the optimization objective function to establish a real-time feedback model for bullfrog feeding. The detailed steps are as follows.

[0085] Establishing constraints: Since the device is generally located in the air and obstructed by baffles, the feeding range of each layer of feed is affected by the rotation speed of the screw feeder, the height of the baffles, and air resistance. Furthermore, to ensure that each bullfrog can receive the feed, the feeding range must be greater than or equal to the distance of each bullfrog from the center of the circle. The specific formula is as follows:

[0086] ;

[0087] in, This indicates the feeding range of the screw feeder. This indicates the distance of the bullfrog from the center of the circle. This indicates the current rotational speed of the screw mechanism. Indicates the height of the baffle. Indicates the maximum height of the baffle. Indicates the air drag coefficient. This represents the difference between the amount of feed given and the amount required. This refers to the actual amount of feed delivered to that location by the feeder. and These represent the updated rotational speed and the previous rotational speed of the device, respectively. A self-defined tolerance threshold is used to avoid over-adjustment. For example, when the error is less than a certain set threshold, the speed adjustment is stopped and the existing speed is maintained.

[0088] Establishment of the objective function: Considering that the spiral feeding device has a certain power consumption and the feeding orientation may have some errors, this invention defines the objective function as the sum of the feeding errors of each layer, as well as the comprehensive objective of factors such as energy consumption. The specific formula is as follows:

[0089] ;

[0090] in, This indicates energy consumption (such as the electrical energy consumption of the feeding device), which can be specified according to specific circumstances. The coefficient representing the weight of energy consumption control can be specified based on the importance of energy consumption in the overall device.

[0091] Implementation and Optimization: First, the rotation speed of the spiral feeder was initialized based on the activity level of the bullfrogs in each layer, the height of the baffle, and the feed requirement of each bullfrog. The system then feeds a certain amount of feed to activate the screw feeder. After the feed is placed into the rearing box, sensors monitor data changes in real time and import the data into a computer. At each time step, the system calculates the feeding error through a feedback mechanism and dynamically adjusts the motor speed based on this error. To improve accuracy, the system can also employ optimization methods such as particle swarm optimization and genetic algorithms to further refine the speed adjustment. When the feeding error is less than a preset threshold, the motor stops operating, entering a stable operation phase. Once the feed amount reaches the predetermined value, the motor stops completely, thus ensuring precise control of the feed amount.

[0092] Thus, the technical method of this invention has been described in detail with reference to the accompanying drawings and specific flowcharts. However, the scope of protection of this invention is not limited to the specific embodiments described. Without departing from the principles of this invention, those skilled in the art can make equivalent modifications or substitutions to the relevant technical features, and these modified or substituted technical solutions still fall within the scope of protection of this invention.

Claims

1. A method for controlling the rotational speed of an automatically adjustable screw conveyor for multi-layer bullfrog feed dispensing in a breeding box, characterized in that, include: At each stage of the entire bullfrog feeding process, the distribution of the number of bullfrog jumps and the weight data of each bullfrog were obtained through pressure sensors below the bullfrog breeding area. The activity level of the bullfrogs per minute is calculated based on the actual number of jumps per minute, the maximum number of jumps, and the minimum number of jumps per minute. Calculate the amount of bullfrog feed to be obtained based on the actual number of bullfrogs released, the weight and activity level of each bullfrog; The location of the bullfrogs is estimated, and the feeding amount is controlled to automatically adjust the screw conveyor to feed the bullfrogs. The entire process of feeding bullfrogs includes before and after feeding; The steps for obtaining pressure sensor data below the bullfrog breeding area include: dividing the bullfrog area into several small areas of a certain size and number according to the number of bullfrogs; installing pressure sensors in each small area; connecting the pressure sensors to a computer to obtain bullfrog activity data; and analyzing and processing the data. The bullfrog activity data includes the number of jumps per minute per bullfrog and the weight of each bullfrog. Obtaining and analyzing bullfrog activity data includes the following steps: Preliminary analysis of the data was conducted to obtain the maximum and minimum number of jumps per minute of bullfrogs and the mass of each bullfrog. The average mass was then compared with the normal growth curve of bullfrogs of the same mass to preliminarily determine the growth status of the bullfrogs. The data analysis and processing involves: using a normalization algorithm to map the number of jumps per minute for each bullfrog in each layer to an interval; statistically analyzing the bullfrog jump count data over multiple time periods; and selecting the minimum and maximum number of jumps for each layer of data. and Then, a linear mapping is used to calculate the number of jumps of the bullfrog at each level within a certain time period. In the interval Variations within a range are mapped to activity levels. Finally, the activity level of the bullfrogs at each level is calculated using a normalization formula, as shown below: In the formula The sum of the weights of all the bullfrogs on each floor represents their activity level; this is calculated by adding up the weights of all the bullfrogs and dividing by the total number of bullfrogs. To obtain the average weight of the bullfrogs: ; In the formula, i The number of bullfrogs per layer; j This represents the number of layers in the bullfrog enclosure; This represents the quality of each bullfrog. This represents the average weight of the bullfrogs in each layer; Before obtaining the area of ​​each bullfrog's location, the following steps are included: The data acquired by the pressure sensor is transmitted to a computer for edge analysis to obtain the edge information of the location of each bullfrog. Dilation optimization is performed on the edge information of the bullfrogs, and the contour of the region where each bullfrog is located is extracted from the optimized edge information; Obtaining the median outline area of ​​each bullfrog involves the following steps: Multiple cropping operations were performed on different numerical regions based on the pressure sensor data to obtain multiple different numerical regions. The median value of the obtained numerical regions was selected as the median regional location for each bullfrog. Determining the amount of bullfrog feed to be given includes the following steps: A quantitative model for bullfrog feed administration is established. The quantitative model for bullfrog feed administration is as follows: ； in, This represents the amount of feed given to each bullfrog at that location on each floor. This represents the average weight of the bullfrogs in each layer. This represents the basic feeding amount for each bullfrog. This represents the quality of each bullfrog. This represents the activity level correction factor. Represents activity level; The average weight of bullfrogs in each layer, the basic feeding amount for each bullfrog, the mass of each bullfrog, the activity correction coefficient, and the activity level are input into the bullfrog feed quantification model to calculate the feed amount for each bullfrog. Before determining the number of jumps and activity level of the bullfrogs at each level, the following algorithm is used to estimate the bullfrogs' positional state: The recursive Bayesian filtering algorithm based on the Monte Carlo method, also known as the particle filter algorithm, approximates the probability distribution of the system state using a set of particles. Each particle represents a hypothesis in the space, and the weight of the particle represents the probability of that hypothesis. The objective function is established by weighting and updating the sampling points and combining basic data obtained from the bottom sensor and infrared sensor. Estimate the bullfrog's state coordinates; in: Represents energy consumption. A coefficient representing the weighting of energy consumption control. This represents the difference between the actual amount of feed delivered to the target location and the expected amount.

2. The method for controlling the rotational speed of an automatically adjustable screw conveyor for multi-layer bullfrog feed dispensing in a breeding box, as described in claim 1, is characterized in that... The establishment of particle sets under each weight includes the following steps: Random initialization Particles Each particle has an initial weight. ; The system tracks the position and speed of each bullfrog and predicts its state based on a dynamic model and sensor measurements. Update the weight of each particle; Repeat the sampling step, resample according to the particle weights, and generate a new particle set.

3. The method for controlling the rotational speed of an automatically adjustable screw conveyor for multi-layer bullfrog feed dispensing in a breeding box, as described in claim 2, is characterized in that... The specific steps for establishing the objective function are as follows: Establish the system's state vector; Establish a dynamic state transition function; Create an update function; A dynamic adjustment mechanism controls the rotational speed and establishes constraints. The objective function is defined as the sum of the feeding errors at each layer and the combined objective of energy consumption factors. The objective function is then calculated.

4. The method for controlling the rotational speed of an automatically adjustable screw conveyor for multi-layer bullfrog feed dispensing in a breeding box, as described in claim 3, is characterized in that... Obtaining the state vector, dynamic state transition function, and update function involves the following steps: Establish the system's state vector directly from sensor data: ； in , Representing bullfrogs The position at that moment; , Representing bullfrogs Time along and The speed of the shaft; Establish the state transition of the bullfrog using classical equations of motion: ； ； in: Represents the time step; Represents external control input; Representing bullfrogs Acceleration in the direction of; Representing bullfrogs Acceleration in the direction of; Represents the dynamic state transition function; Representative at Displacement caused by velocity in the direction; Representative at Displacement caused by velocity in a certain direction; Suppose a nonlinear mapping function This describes the transition from the state space to the observation space, and updates the weight of each particle according to the measurement probability formula to perform real-time tracking and prediction of the bullfrog's position: ； in: The formula representing the probability of measurement; Represents measurement noise; Represents a nonlinear function, indicating the state. to the observed value Mapping; The specific functions for establishing the constraints are as follows: ； in: This represents the height of the baffle. Represents the updated rotational speed; Represents the air drag coefficient; This represents the maximum height of the baffle. This represents the rotational speed of the device in the previous operation; This represents the distance of the bullfrog from the center of the circle; This represents the current rotational speed of the screw mechanism; This represents the feeding range of the spiral feeder; This represents the difference between the amount of feed given and the amount required. This represents the actual amount of feed delivered to that location by the feeder. This represents a self-defined tolerance threshold to avoid over-adjustment.

Images