Rotating speed control method of automatic adjusting spiral conveying device based on multi-layer bullfrog feed putting in breeding box

Abstract

The invention provides a rotating speed control method of an automatic adjusting spiral conveying device based on multi-layer bullfrog feed putting in a breeding box, which comprises the following steps of: acquiring the jumping times and the jumping range of bullfrogs before feeding by using a sensor, connecting the sensor with a computer to form images and data, analyzing the activeness of the bullfrogs from the images and the data, and determining the activeness of the bullfrogs. Meanwhile, the corresponding bullfrog feed feeding amount is obtained through comprehensive analysis in combination with the bullfrog body weight obtained through a sensor, accurate feeding of bullfrogs is conducted according to quantitative analysis of the feeding amount, and the problem that in the prior art, feeding is conducted depending on breeding experience or a traditional feeder, and feeding is insufficient or excessive easily can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of bullfrog farming, and in particular to a rotational speed control method for an automatic adjustment screw conveyor device for multi-layer bullfrog feed feeding in a breeding tank. Background Art

[0002] Since it was first introduced into China at the end of the 1950s, after nearly 30 years of development, after overcoming the technical difficulties of diet domestication, bullfrog farming has risen across the country. Bullfrogs have high utilization value because they can be used for food (fast growth, delicious taste, rich nutrition, high protein content), medicine (internal organs can be used to make medicine), and many other aspects. Now, the farmed bullfrogs with high economic value have become the main products of special aquatic farming. However, through observation, it is found that the existing bullfrog farming method adopts traditional high-density farming, facing many problems such as large floor area, high input and low output value, and waste of labor input. Based on this, the inventor of the present application previously applied for an automatic three-dimensional bullfrog breeding device and method, application number: 2025101977040. In this application, a screw feed feeding system is installed at the center of the circle of multiple breeding tanks from top to bottom, and feeds are automatically fed by rotation. Currently, when using this feeding system to feed, mainly farmers feed according to experience, judge the size of bullfrogs by the naked eye to formulate a feeding plan, and adopt manual feed feeding, which cannot accurately feed according to the specific growth situation and activity of bullfrogs. Just feeding according to the naked eye judgment and farmers' experience is likely to result in excessive or insufficient feed, which will affect the normal growth of bullfrogs to a certain extent. Summary of the Invention

[0003] To solve the current technical problems, the main purpose of the present invention is to provide a rotational speed control method for an automatic adjustment screw conveyor device for multi-layer bullfrog feed feeding in a breeding tank. This algorithm quantitatively analyzes the feeding amount according to the weight and activity of bullfrogs, evaluates and adjusts the feed feeding amount, and then realizes precise feeding of feed by precisely controlling the screw feeding of the screw feed feeding system. While ensuring the feeding quality of bullfrogs, it effectively improves the feed utilization rate to achieve scientific feeding.

[0004] To achieve the above technical features, the object of the present invention is realized as follows: 1. A rotational speed control method for an automatic adjustment screw conveyor device for multi-layer bullfrog feed feeding in a breeding tank, characterized by comprising: In each stage of the whole process of bullfrog feed feeding, respectively obtain the number distribution of bullfrog jumps and the weight data of each bullfrog through the pressure sensors under the bullfrog breeding area; Calculate the activity of bullfrogs per minute according to the actual number of jumps per bullfrog per minute, the maximum number of jumps, and the minimum number of jumps; Calculate the feeding amount of bullfrog feed based on the actual number of bullfrogs put in, the weight and activity of each bullfrog. Estimate the position of the bullfrogs and control the automatic adjustment of the screw conveyor to feed the bullfrogs according to the feeding amount.

[0005] Preferably, the whole process of bullfrog feed feeding includes before and after feeding. Obtaining the data of the pressure sensors under the bullfrog breeding area includes the following steps: According to the number of bullfrogs in breeding, divide the area where the bullfrogs are located into several small areas of a certain size and number, install pressure sensors in each small area, connect the pressure sensors to a computer, obtain the bullfrog activity data, and analyze and process the data. The bullfrog activity data includes the number of jumps per minute per bullfrog and the weight of each bullfrog.

[0006] Preferably, obtaining the bullfrog activity data and analyzing and processing the data specifically include the following steps: Conduct a preliminary analysis of the data to obtain the maximum and minimum values of the number of jumps per minute of the bullfrogs and the mass of each bullfrog, take the average value of the mass, and perform a fitting analysis with the normal growth state curve of bullfrogs of the same mass to preliminarily judge the growth state of the bullfrogs. The analysis and processing of the data: Use the normalization algorithm to map the number of jumps per minute of each bullfrog on each layer to an interval, count the data of the number of jumps of the bullfrogs in multiple time periods, and screen out the minimum and maximum number of jumps of the bullfrogs in the data of each layer, which are respectively and Then use linear mapping to map the number of jumps of the bullfrogs on each layer within a certain period of time Changing within the interval interval to the activity , and finally calculate the activity of the bullfrogs on each layer according to the normalization formula. The specific formula is as follows: , where in the formula represents the activity of the bullfrogs on each layer. For the data processing, add up the weights of all the bullfrogs and divide by the total number of bullfrogs to obtain the average weight of the bullfrogs: ; In the formula, i is the number of bullfrogs on each layer; j is the number of layers of bullfrogs; represents the mass of each bullfrog; represents the average weight of the bullfrogs on each layer.

[0007] Preferably, before obtaining the area of the region where each bullfrog is located, it includes the following steps: Transmit the data obtained by the pressure sensors to a computer for edge analysis to obtain the edge information of the position where each bullfrog is located. Dilate and optimize the edge information of the bullfrogs, and extract the contour of each bullfrog's area from the optimized edge information.

[0008] Preferably, obtaining the median value of the contour area of each bullfrog includes the following steps: Perform multiple clippings on different numerical regions according to the pressure sensor data to obtain multiple different numerical regions; Select the median value from the obtained multiple numerical regions as the median value of the regional position of each bullfrog.

[0009] Preferably, determining the feeding amount of bullfrog feed includes the following steps: Establish a quantitative model for bullfrog feed feeding. The quantitative model for bullfrog feed feeding is: ; Among them, represents the feeding amount of feed at this position for each bullfrog on each layer, represents the average weight of bullfrogs on each layer, represents the basic feeding amount of each bullfrog, represents the mass of each bullfrog, represents the activity correction coefficient, represents the activity; Substitute the average weight of bullfrogs on each layer, the basic feeding amount of each bullfrog, the mass of each bullfrog, the activity correction coefficient, and the activity into the quantitative model for bullfrog feed feeding, and calculate the feeding amount of each bullfrog.

[0010] Preferably, before determining the number of jumps and activity of bullfrogs on each layer, the following algorithm is used to estimate the position state of bullfrogs: The recursive Bayesian filtering algorithm based on the Monte Carlo method, that is, the particle filter algorithm, approximates the probability distribution of the system state through a particle set. Each particle represents a hypothesis in space, and the weight of the particle represents the probability of this hypothesis. By weighted updating the sampling points and combining the basic data obtained by the bottom sensor and the infrared sensor, an objective function is established: , estimate the state coordinates of the bullfrog; Among them: represents the energy consumption, represents the coefficient controlling the energy consumption weight, is the difference between the actual feed amount fed to the target position and the estimated value.

[0011] Preferably, the establishment of the particle set under each weight includes the following steps: Randomly initialize particles , and each particle has an initial weight ; Track the position and velocity of each bullfrog, and perform state prediction according to the dynamic model. Update the weight of each particle based on the measurement values provided by the sensors , and update the weight of each particle; Repeat the sampling step, resample according to the weights of the particles, and generate a new set of particles.

[0012] Preferably, the specific steps for establishing the objective function are as follows: Establish the state vector of the system; Establish the dynamic state transition function; Establish the update function; The dynamic adjustment mechanism controls the rotation speed and establishes the constraint conditions; The objective function is defined as the sum of the feeding errors of each layer and the comprehensive objective of factors such as energy consumption, and the objective function is obtained.

[0013] Preferably, the acquisition of the state vector, dynamic state transition function, and update function includes the following steps: Directly establish the state vector of the system according to the sensor data: ; where , represent the position of the bullfrog at the moment; , represent the velocity of the bullfrog along the axis at the and axes; Establish the state transition of the bullfrog through the classical motion equation: ; ; where: represents the time step; represents the external control input; represents the acceleration of the bullfrog in the direction; represents the acceleration of the bullfrog in the direction; represents the dynamic state transition function; represents the displacement caused by the velocity in the and direction; represents the displacement caused by the velocity in the direction; Assume a non-linear mapping function to describe the conversion from the state space to the observation space, and update the weight of each particle according to the measurement probability formula to perform real-time tracking and prediction of the bullfrog position: ; Wherein: represents the measurement probability formula; represents the measurement noise; represents a non - linear function, indicating the mapping from the state to the observed value mapping; The specific function for establishing the constraint conditions is as follows: ; Wherein: represents the height of the baffle; represents the updated rotational speed; represents the air resistance coefficient; represents the maximum height of the baffle; represents the rotational speed of the previous device; represents the distance of the bullfrog from the center of the circle; represents the existing rotational speed of the spiral device; represents the feeding range of the spiral feeder; represents the difference between the feeding amount and the demand; represents the amount of feed actually fed to this place by the feeder; represents a self - set tolerance error threshold to avoid over - adjustment.

[0014] The present invention has the following beneficial effects: 1. By collecting the weight - related coefficient of the bullfrog and the activity deviation index of the bullfrog before feeding, and combining with water temperature and feeding coefficient, the present invention comprehensively analyzes to obtain the feeding amount of bullfrog feed, and accurately feeds the bullfrog feed according to the quantitative analysis of the feeding amount, which can solve the problem that the existing technology is prone to under - feeding or over - feeding when feeding depending on breeding experience or traditional feeding machines. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The present invention will be further described below in conjunction with the drawings and embodiments.

[0016] Figure 1 is the flowchart for starting the spiral feeder.

[0017] Figure 2 is the mechanical structure diagram of the spiral feeder. SPECIFIC EMBODIMENTS

[0018] To make the purpose, innovative points, application fields and operation methods of the present invention clearer and more transparent, the present invention will be further described in detail below in conjunction with embodiments and additional flowcharts.

[0019] Compared with the traditional feeding method, the spiral conveyor feeding device can dynamically adjust the rotation speed according to the height of each baffle and the feed demand, and can more accurately control the feeding range and the amount of feed delivered to each layer, enabling the feed to be delivered near the bullfrogs and overcoming the problems of feed waste or uneven feeding of bullfrogs in the traditional method.

[0020] The algorithm of the present invention can be implemented through a PLC control system. The control system can use sensors to continuously monitor the weight of the bullfrogs on each layer, the activity level of the bullfrogs, and the approximate position coordinates of each bullfrog, and automatically adjust the rotation speed of the spiral conveyor device through the algorithm of the present invention. The specific operation process is as follows: Step1: Estimation of the weight of each bullfrog on each layer and the average weight of the bullfrogs: First, the area where the bullfrogs are located on each layer can be divided into a reasonable number of small areas according to the size of the range. Pressure sensors are installed in each area. If a bullfrog is exactly in this position, the sensor at this place will record data. On the contrary, if the bullfrog jumps to other places, the data displayed by the sensor at this place will be 0. In this way, the weight of each bullfrog on each layer in the breeding device can be obtained at any time , and by comparing the normal weight of the bullfrogs after eating, it can be judged whether feed needs to be fed during this period.

[0021] Second, according to the data statistically obtained by the above sensors, add up the weights of all the bullfrogs and divide by the total number of bullfrogs , which is the average weight of the bullfrogs . By comparing the average weight of the bullfrogs, the weight of each bullfrog, and the weight of the bullfrogs after normal eating, the breeding personnel can set a basic feeding amount based on these data .

[0022] Step2: Obtaining the activity level : First, through consulting relevant materials, it can be known that judging whether bullfrogs need to eat can be comprehensively analyzed from multiple aspects. First of all, the size and morphological changes of the bellies of bullfrogs are an important indicator. After eating, the bellies of bullfrogs will significantly expand and their weights will also increase. On the contrary, it means that the bullfrogs have not eaten. In addition, the feeding cycle is also a key factor. The feeding interval of bullfrogs is closely related to their physiological needs. Finally, the activity level of the bullfrogs on each layer can also provide a reference for judgment. If the activity frequency of the bullfrogs is relatively high, it usually means that they are in a state of looking for food, and this active behavior is likely to have a certain impact on the feeding amount. Therefore, by comprehensively observing the changes in the above aspects, it is possible to effectively evaluate whether the bullfrogs need to eat.

[0023] Second, according to the above, if a bullfrog moves from one place to another, the data of the pressure sensor will change once. Based on this, the number of times the pressure sensor data changes within a few minutes can be counted, and then the activity level of the bullfrogs on this layer per minute can be obtained by using relevant mathematical knowledge.

[0024] Third, in these examples, a normalization algorithm is used to map the number of jumps per minute of the bullfrogs on each layer to the interval , and the specific steps are as follows: First, count the data of the bullfrogs on each layer over a period of time to make the results more generalizable. Second, select the minimum and maximum number of jumps of the bullfrogs in the data of each layer, which are and respectively. Then, use linear mapping to map the number of jumps of the bullfrogs on each layer within a certain period of time varying within the interval to the activity level . Finally, calculate the activity level of the bullfrogs on each layer according to the normalization formula. The specific formula is as follows: , where represents the activity level of the bullfrogs on each layer.

[0025] Since the activity coefficient cannot fully regulate the required amount of feed, the activity correction coefficient can now be set by observing similar situations according to the empirical rule or the knowledge in the field of bullfrog farming. When the activity level changes significantly compared to before, it may be necessary to increase to make the change in the number of jumps more sensitive; while when the activity level changes smoothly, can be reduced to avoid over-adjustment.

[0026] Step3: Feed amount for each bullfrog's position on each layer: Based on the weight of each bullfrog obtained previously, the average weight of the bullfrogs, the activity level of the bullfrogs on each layer, and the self-set basic feed amount, the feed amount for each position can be determined through the following calculation formula. This method can accurately control the feeding amount of each bullfrog, thus effectively ensuring its health status, and at the same time can also monitor the activity of the bullfrogs in real time. The specific calculation formula is as follows: , where represents the feed feeding amount for each bullfrog's position on each layer, represents the activity level of the bullfrogs on each layer after being adjusted by the activity coefficient, that is, the final activity level value.

[0027] Step4: Estimate the position state of the bullfrogs using the particle filter algorithm: Since the movement state of the bullfrog cannot be kept stable at all times, it may suddenly jump to other positions when the machine determines its position and transmits the relevant data to the computer, thus causing certain data errors. In addition, since the particle filter algorithm is good at dealing with systems with non-linear and non-Gaussian noise and can effectively solve non-linear problems, therefore, our group chose the recursive Bayesian filtering algorithm based on the Monte Carlo method, that is, the particle filter algorithm. By weighted updating the sampling points and combining the basic data obtained by the bottom sensor and the infrared sensor, this algorithm can effectively estimate the state coordinates of the bullfrog, thereby reducing data errors and improving the accuracy and reliability of the system.

[0028] The basic idea of the particle filter is to approximately represent the probability distribution of the system state through a particle set. Each particle represents a hypothesis in space, and the weight of the particle represents the probability of that hypothesis. First, randomly initialize particles , and each particle has an initial weight . Next, clarify the state of the bullfrog system, track the position and speed of each bullfrog, and make state predictions according to the dynamic model. According to the measurement values provided by the sensor, update the weight of each particle, and then repeat the sampling step, resample according to the weight of the particles, and generate a new particle set. In this way, particles with larger weights will obtain more samplings, thereby improving the accuracy of the system's estimation of the bullfrog state and ensuring effective filtering of inaccurate data and noise. The detailed description is as follows: Definition of the state vector of the system: , where , represents the position of the bullfrog at time; , represents the speed of the bullfrog along and and axes at

[0029] Establishment of the dynamic state transition function : The state function describes the process of system evolution from one moment to the next. For the state transition of the bullfrog, it can be established through the classical motion equation, especially in the case of no external control (that is, the movement of the bullfrog is controlled by natural factors), and the specific formula is as follows: ; In the left formula, and respectively represent the displacements caused by the speed in the and directions, is the time step, is an external control input, such as the movement control of bullfrogs, but usually no additional control input is required under natural conditions; the right formula indicates that if the speed of the bullfrog is affected by some external forces (such as flow velocity, slope, etc.), the control input can be expressed as the right formula, where and represent the acceleration of the bullfrog in the and directions.

[0030] Establishment of the update function: In particle filtering, the measurement model can also be expressed as a conditional probability, and its general formula is usually non - linear. Assume a non - linear mapping function to describe the conversion from the state space to the observation space, and update 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: ; where, is a non - linear function, representing the mapping from the state to the observed value (for example, mapping the position of the bullfrog to the image coordinates obtained by the camera), is the measurement noise, and it is only necessary to assume that it follows a certain known probability distribution (such as Gaussian distribution), represents the measurement probability formula.

[0031] Resampling step: Resample according to the weights of the particles to generate a new set of particles. At this time, the weights of the particles are re - initialized to .

[0032] Step5: The dynamic adjustment mechanism controls the rotation speed Since the bullfrogs in the entire feeding device are in the breeding box, and the spiral feeder is located at the center of each layer and runs through the upper and lower layers, the feeding area must be a region centered on the center of the breeding box. Therefore, the present invention introduces a real - time feedback mechanism for dynamically adjusting the rotation speed to ensure that each bullfrog on each layer can obtain an appropriate amount of feed as much as possible, avoiding the situation of uneven nutrition. In summary, the present invention takes factors such as the feeding range must be greater than or equal to the distance of the bullfrog from the center of the circle, feed error distribution, and the influence of the baffle height as constraint conditions, 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.

[0033] Establishment of the constraint conditions: Since the device is generally in the air and is blocked by baffles, the feeding range of each layer of feed is affected by the rotation speed of the spiral feeder, the height of the baffle, and air resistance. At the same time, to ensure that each bullfrog can receive 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: ; Among them, represents the feeding range of the spiral feeder, represents the distance of the bullfrog from the center of the circle, represents the existing rotational speed of the spiral device, represents the height of the baffle, represents the maximum height of the baffle, represents the air resistance coefficient, represents the difference between the feeding amount and the demand, is the amount of feed actually fed by the feeder to this place, and respectively represent the updated rotational speed and the rotational speed of the previous device, is a self-set tolerance error threshold to avoid over-adjustment. For example, when the error is less than a certain set threshold, the rotational speed adjustment stops and the existing rotational speed is maintained.

[0034] Establishment of the objective function: Considering that the spiral feeding device has a certain power consumption and there may be some errors in the feeding orientation, the present invention defines the objective function as the comprehensive objective of the total feeding error of each layer and factors such as energy consumption. The specific formula is as follows: ; Among them, represents the energy consumption (such as the power consumption of the feeding device), which can be specified according to specific circumstances, represents the coefficient for controlling the energy consumption weight, which can be specified according to the importance of the energy consumption in the entire device.

[0035] Actual implementation and optimization: First, according to the activity level of the bullfrogs in each layer, the height of the baffle, and the feed demand of each bullfrog, initialize the rotational speed of the spiral feeder, and put a certain amount of feed into the system to start the operation of the spiral feeder. Then, after the feed is scattered into the breeding tank, the sensor will monitor the data changes in real time and import them into the computer. At each time step, the system calculates the feeding error through the feedback mechanism and dynamically adjusts the motor rotational speed according to this error. To improve the accuracy, the system can also adopt optimization methods such as the particle swarm optimization algorithm and the genetic algorithm to further precisely adjust the rotational speed. When the feeding error is less than the preset threshold, the motor stops operating and enters the stable operation stage. After the feeding amount reaches the predetermined value, the motor completely stops, thereby ensuring the precise control of the feed feeding amount.

[0036] So far, the technical method of the present invention has been described in detail through the accompanying drawings and specific flowcharts. However, the protection scope of the present invention is not limited to the specific implementation manners described above. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and these technical solutions after the changes or substitutions still fall within the protection scope of the present invention.

Claims

1. A speed control method for an automatic adjustment spiral conveying device based on multi-layer bullfrog feed feeding in a breeding box, characterized in that, Including: At each stage of the whole process of bullfrog feed feeding, the number distribution of bullfrog jumps and the weight data of each bullfrog are obtained respectively through the pressure sensors under the bullfrog breeding area; According to the actual number of jumps per minute per bullfrog, the maximum number of jumps, and the minimum number of jumps, calculate the activity level of the bullfrog per minute; According to the actual number of bullfrogs put in, the weight and activity level of each bullfrog, calculate and obtain the feeding amount of bullfrog feed; Estimate the position of the bullfrog and control the automatic adjustment of the screw conveyor to feed the bullfrog according to the feeding amount.

2. The rotational speed control method of an automatic adjustment screw conveyor device based on multi-layer bullfrog feed feeding in a breeding box according to claim 1, characterized in that: The whole process of the bullfrog feed feeding includes before feeding and after feeding; Obtaining the data of the pressure sensors under the bullfrog breeding area includes the following steps: according to the number of bullfrogs in breeding, divide the area where the bullfrogs are located into several small areas of a certain size and quantity, install pressure sensors in each small area, connect the pressure sensors to a computer, obtain the bullfrog activity data, and analyze and process the data. The bullfrog activity data includes the number of jumps per minute per bullfrog and the weight of each bullfrog.

3. The rotational speed control method of an automatic adjustment screw conveyor device for multi-layer bullfrog feed feeding in a breeding box according to claim 2, characterized in that, Obtaining the bullfrog activity data and analyzing and processing the data specifically include the following steps: Conduct a preliminary analysis of the data to obtain the maximum and minimum values of the number of jumps per minute of the bullfrogs and the mass of each bullfrog, take the average value of the mass, fit and analyze it with the normal growth state curve of bullfrogs of the same mass, and preliminarily judge the growth state of the bullfrogs; The analysis and processing of the data: The normalization algorithm is used to map the number of jumps per minute of each bullfrog on each layer to an interval. The data of the number of jumps of bullfrogs in multiple time periods are statistically analyzed, and the minimum and maximum numbers of jumps of bullfrogs in the data of each layer are selected, which are respectively and Then, the number of jumps of bullfrogs on each layer within a certain period of time is Varied within the interval Interval, and mapped to the activity level , and finally, the activity level of bullfrogs on each layer is calculated according to the normalization formula. The specific formula is as follows: , where represents the activity level of bullfrogs on each layer. For the data processing, the weights of all bullfrogs are added up and divided by the total number of bullfrogs to obtain the average weight of bullfrogs: ; In the formula, i is the number of bullfrogs per layer; j is the number of layers of bullfrogs; represents the mass of each bullfrog; represents the average weight of bullfrogs per layer.

4. The rotational speed control method of an automatic adjustment screw conveyor device based on multi-layer bullfrog feed feeding in a breeding box according to claim 3, characterized in that Before obtaining the area of each bullfrog's location, it includes the following steps: Transmit the data obtained by the pressure sensors to the computer for edge analysis to obtain the edge information of the location of each bullfrog; Perform dilation optimization on the edge information of the bullfrogs, and extract the contour of each bullfrog's location from the optimized edge information.

5. The rotational speed control method of an automatic adjustment screw conveyor device based on multi-layer bullfrog feed feeding in a breeding box according to claim 4, characterized in that, Obtaining the median value of the contour area of each bullfrog includes the following steps: Perform multiple clippings on different numerical regions according to the pressure sensor data to obtain multiple different numerical regions; Screen out the median value from the obtained multiple numerical regions as the median value of the regional location of each bullfrog.

6. The rotational speed control method of an automatic adjustment screw conveyor device based on multi-layer bullfrog feed feeding in a breeding box according to claim 5, characterized in that, Determining the bullfrog feed feeding amount includes the following steps: Establish a quantitative model for bullfrog feed feeding. The quantitative model for bullfrog feed feeding is: ； Among them, represents the feed feeding amount at this position for each bullfrog on each layer, represents the average weight of bullfrogs on each layer, represents the basic feeding amount for each bullfrog, represents the mass of each bullfrog, represents the activity correction coefficient, represents the activity; Substitute the average weight of each layer of bullfrogs, the basic feeding amount of each bullfrog, the mass of each bullfrog, the activity correction coefficient, and the activity into the quantitative model for bullfrog feed feeding, and calculate the feed feeding amount of each bullfrog.

7. The rotational speed control method of an automatic adjustment screw conveyor device based on multi-layer bullfrog feed feeding in a breeding box according to claim 6, characterized in that, Before determining the number of jumps and activity level of each layer of bullfrogs, the following algorithm is used to estimate the position state of the bullfrogs: The recursive Bayesian filtering algorithm based on the Monte Carlo method, namely the particle filter algorithm, approximates the probability distribution of the system state through a particle set. Each particle represents a hypothesis in space, and the weight of the particle represents the probability of that hypothesis. By weighted updating the sampling points and combining the basic data obtained by the bottom sensor and the infrared sensor, an objective function is established: , estimate the state coordinates of the bullfrog; Wherein: represents energy consumption, represents the coefficient for controlling the weight of energy consumption, is the difference between the actual feed amount fed to the target position and the predicted value.

8. The rotational speed control method of an automatic adjustment screw conveyor device based on multi-layer bullfrog feed feeding in a breeding box according to claim 7, characterized in that, The establishment of the particle set under each weight includes the following steps: Random initialization particles , and each particle has an initial weight ; Track the position and speed of each bullfrog, and perform state prediction according to the dynamic model. Based on the measurement values provided by the sensors , update the weight of each particle; Repeat the sampling step, resample according to the weight of the particles, and generate a new particle set.

9. The rotational speed control method of an automatic adjustment screw conveyor device based on multi-layer bullfrog feed feeding in a breeding box according to claim 8, characterized in that The specific steps for establishing the objective function are as follows: Establish the state vector of the system; Establish the dynamic state transition function; Establish the update function; The dynamic adjustment mechanism controls the rotation speed and establishes the constraint conditions; The objective function is defined as the sum of the feeding errors of each layer and the comprehensive objective of factors such as energy consumption, and the objective function is obtained.

10. The rotational speed control method of an automatic adjustment screw conveyor device based on multi-layer bullfrog feed feeding in a breeding box according to claim 9, characterized in that, The acquisition of the state vector, dynamic state transition function, and update function includes the following steps: Directly establish the state vector of the system according to the sensor data: ； Among them , represent the position of the bullfrog at moment; , represent the velocity of the bullfrog along the axis at and axis; Establish the state transition of the bullfrogs through the classical motion equation: ； ； Wherein: represents the time step; represents the external control input; represents the acceleration of the bullfrog in direction; represents the acceleration of the bullfrog in direction; represents the dynamic state transition function; represents the displacement caused by velocity in and direction; represents the displacement caused by velocity in direction; Assume a non-linear mapping function to describe the transformation from the state space to the observation space, and update the weight of each particle according to the measurement probability formula to perform real-time tracking and prediction of the bullfrog's position: ； Wherein: represents the measurement probability formula; represents the measurement noise; represents a non-linear function, indicating the state to the observed value mapping; The specific functions for establishing the constraint conditions are as follows: ； Wherein: represents the height of the baffle; represents the updated rotational speed; represents the air resistance coefficient; represents the maximum height of the baffle; represents the rotational speed of the previous device; represents the distance of the bullfrog from the center of the circle; represents the existing rotational speed of the spiral device; represents the feeding range of the spiral feeder; represents the difference between the feeding amount and the demand; represents the amount of feed actually fed to this place by the feeder; represents a self - set tolerance error threshold to avoid over - adjustment.

Images