Adjustable seed cotton cleaning machine based on different moisture regain rates and machine parameter adjusting method thereof

Abstract

The invention relates to the technical field of seed cotton cleaning machines, and discloses an adjustable seed cotton cleaning machine based on different moisture regain rates and a machine parameter adjusting method thereof.The method comprises the following steps that S1, a seed cotton moisture regain rate online detection device installed at a seed cotton feeding opening is used for collecting moisture regain rate data of to-be-processed seed cotton in real time; the data is transmitted to a control system upper computer through a communication protocol; s2, preprocessing the collected moisture regain data based on a sliding window filtering algorithm, removing abnormal values, calculating a dynamic mean value, and improving the stability and reliability of control input; and S3, fusing the moisture regain change trend, detecting the moisture regain of the seed cotton in real time, and automatically adjusting key machining parameters of the cleaning machine according to the moisture regain, so that the seed cotton cleaning effect can be remarkably improved. Meanwhile, in the cleaning process, the distance between the adjacent lattice bars, the rotating speed of the barbed nail roller and the inclination angle of the cleaning part are accurately adjusted according to different moisture regain, and the relation between the impurity cleaning efficiency and fiber damage is effectively balanced.

Description

Technical Field

[0001] The invention relates to the technical field of seed cotton cleaning machines, in particular to a seed cotton cleaning machine adjustable based on different moisture regains and a machine parameter adjustment method thereof. Background Art

[0002] Seed cotton cleaning, a key step in cotton processing, plays a crucial role in the cotton's final quality. During the seed cotton harvest process, its moisture regain is affected by many factors, such as weather changes and differences in storage conditions. These factors cause the moisture regain of seed cotton to fluctuate significantly. Variations in moisture regain have a significant impact on the physical properties of seed cotton, particularly impurity adhesion and fiber strength. Specifically, as the moisture regain of cotton increases, the adhesion between fibers and impurities also increases. Generally speaking, a lower moisture regain improves cotton cleaning efficiency, but too low a moisture regain can lead to damage to cotton fibers, manifested as shortened cotton fiber length, reduced length uniformity index, and lower breaking strength, which in turn negatively impacts the cotton's final quality.

[0003] Given this, seed cotton generally needs to be pre-treated before cleaning to adjust its moisture regain to an appropriate range. However, the reality is that the initial moisture regain of seed cotton varies greatly due to the different cotton stacking conditions and harvest batches. Even after treatment in a drying tower, the moisture regain of the treated seed cotton cannot be consistent due to the use of the same drying parameters, and can only fluctuate within a wide range. In the seed cotton cleaning process of actual processing plants, there is often a lack of effective treatment methods for seed cotton with different moisture regains, and there is also a lack of corresponding theoretical guidance for the adjustment of key parameters of seed cotton cleaning machines, making it difficult to achieve a significant breakthrough in cleaning efficiency.

[0004] Currently, most cotton processing companies use a mixed loading method for seed cotton with different moisture regains. This method results in significant differences in cleaning performance even when using seed cotton cleaning machines with the same processing parameters. While cleaning performance improves for seed cotton with lower moisture regain, it fails to meet ideal cleaning requirements for seed cotton with higher moisture regain, making it difficult to effectively control impurity levels.

[0005] In addition, after searching Chinese patent CN116065242A, a distance adjustment device between adjacent lattices was proposed, which improved the mechanization level of the seed cotton cleaning stage. However, the machine parameter adjustment method of the seed cotton cleaning machine was not organically combined with the material properties of the seed cotton. As for how to accurately adjust the key parameters of the cleaning machine according to the changes in the moisture regain, such as the distance between adjacent lattices, the speed of the spike roller, the inclination angle of the cleaning part, etc., there is still a lack of a systematic solution.

[0006] Therefore, developing a seed cotton cleaning machine that can automatically adjust machine parameters according to the moisture regain of seed cotton is of great significance for improving cotton processing quality, reducing production costs, improving production efficiency and meeting increasingly stringent standard requirements. Summary of the Invention

[0007] In view of the deficiencies in the prior art, the present invention provides an adjustable seed cotton cleaning machine based on different moisture regains and a method for adjusting machine parameters thereof, which solves the problems raised in the above background technology.

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions: a seed cotton cleaning machine adjustable based on different moisture regains, including a fixed frame, an inclination sensor and an electric push rod, an electric cylinder and a fixed platform for adjusting the inclination of the cleaning part are installed on the fixed frame, the electric cylinder is connected to the fixed platform, and the electric cylinder cooperates with the inclination sensor, a front support platform and a rear support platform are welded on the fixed frame, the front support platform is tightly fitted with the outer support of the seed cotton cleaning machine but not fixed to achieve the limitation of the maximum lifting position and improve its stability; the electric push rod is connected to the elliptical grid on the grid through the T-shaped connecting arm, and the electric push rod cooperates with the linear switch to realize the control of the extension and retraction distance of the electric push rod, which is used to adjust and detect the distance between adjacent grids; the thorn drum is connected to the three-phase asynchronous motor through a V-belt and rotates at the same speed, the drum speed is adjusted by changing the frequency of the inverter, and the pulse value returned by the drum shaft is collected by the encoder for the control system to detect and feedback the thorn drum speed.

[0009] Preferably, the rear support platform is fitted with the outer support part of the seed cotton cleaning machine, the fixed platform is welded to the seed cotton cleaning machine, the upper side of the electric cylinder is fixed to the fixed platform, and the inclination sensor is installed on the outer web, consistent with the inclination angle of the grid support frame.

[0010] Preferably, the electric push rod is arranged obliquely and has the same inclination angle as the lattice support frame. The electric push rod is connected to the T-shaped connecting arm through the push rod. The T-shaped connecting arm drives the arc-shaped slide on the lower side of the two adjacent lattices to slide along the slide rail, thereby driving the rocker arm connected to the upper elliptical lattice to perform circular motion around the center of the ellipse, thereby changing the gap between the elliptical lattice and the circular lattice.

[0011] Preferably, the telescopic length of the electric push rod is detected in real time by a magnetic switch fixed on the outside of the guide sleeve, and the adjustment position is determined by the signal value fed back by it. The angle of the elliptical grid rotating around the center point is realized by derivation of the mechanism motion diagram and the corresponding mathematical model to adjust the distance between the grids.

[0012] Preferably, it further comprises supporting wall panels and side wall panels, which are used to support and fix the lattice structure while improving the airtightness of the device.

[0013] A parameter adjustment method for a seed cotton cleaning machine with adjustable moisture regains comprises the following steps:

[0014] S1, using a seed cotton moisture regain online detection device installed at a seed cotton feeding port to collect moisture regain data of the seed cotton to be processed in real time, and transmitting the data to a control system host computer through a communication protocol;

[0015] S2. Preprocess the collected moisture regain data based on a sliding window filtering algorithm, remove outliers, and calculate the dynamic mean to improve the stability and reliability of the control input;

[0016] S3, integrating the moisture regain trend, using the incremental PID control strategy to perform feed-forward compensation on the moisture regain data, achieving a rapid response to fluctuations in seed cotton moisture content;

[0017] S4. Input the moisture regain data processed by feedforward compensation into the four-dimensional high-order non-homogeneous regression model (the modeling objects cover four indicators: cleaning efficiency, short fiber rate, fiber length and machine energy consumption, and the independent variables are moisture regain x, nail roller speed ω, distance between adjacent grids γ, cleaning part inclination θ). The moisture regain variable x is eliminated by dimensionality reduction and elimination method, and it is converted into a three-dimensional regression model (only machine parameters such as ω, γ, θ are retained as independent variables) to eliminate the interactive coupling effect between moisture regain and mechanical parameters and reduce model error. Based on this model, the NSGA-Ⅲ multi-objective optimization algorithm is used to solve the Pareto optimal solution set under the constraints, and output the optimal machine parameter combination (ω * ,θ * , γ * );

[0018] S5. Based on the output of the regression model, the corresponding machining parameters are converted into electrical control signals and transmitted in real time to the industrial control computer (host computer) via the RS-485 communication interface (baud rate 115200bps). Based on the received instructions, the control system dynamically adjusts the core operating parameters of the seed cotton cleaning machine, including: the distance between adjacent grid bars (adjustable range 7-13mm, positioning accuracy ±0.1mm); the speed of the spike roller (adjustable range 500-1200rpm, steady-state error ±5rpm); and the inclination angle of the cleaning section (adjustable threshold 22.5°-45°, resolution 0.5°). This achieves adaptive matching of seed cotton cleaning process parameters under different moisture regain conditions (5%-16%).

[0019] Preferably, in the step S5, when adjusting the distance between adjacent grid bars, the T-shaped connecting arm is driven by the electric push rod to drive the arc slide on the lower side of the two adjacent grid bars to slide along the slide rail, thereby making the rocker arm connected to the upper elliptical grid bar move in a circle with the center of the ellipse, thereby changing the gap between the elliptical grid bar and the circular grid bar; when adjusting the speed of the nail roller, the speed of the three-phase asynchronous motor is adjusted by changing the frequency converter frequency, and the encoder is used to collect data feedback signals; when adjusting the inclination angle of the cleaning part, the front side of the body is driven to tilt by the electric cylinder, and the fixing effect of the electric cylinder fixing platform and the feedback signal of the inclination sensor are combined to achieve precise adjustment of the inclination angle of the cleaning part.

[0020] Preferably, in the S4 step, the quaternary high-order non-homogeneous regression model is determined based on the previous full-factor experimental results, and the improved NSGA-III algorithm is used to perform multi-objective optimization and solve the model to achieve multiple optimization goals of maximizing cleaning efficiency, minimizing fiber damage, and minimizing machine energy consumption.

[0021] Preferably, the method further includes a step of real-time monitoring of the working effect of the cleaning machine after adjustment, wherein data is collected by sensors at various mechanical structural parts and fed back to the control system host computer so as to dynamically evaluate and adjust the operating status of the cleaning machine.

[0022] The present invention provides a seed cotton cleaning machine with adjustable moisture regain and a method for adjusting its parameters. It has the following beneficial effects:

[0023] 1. The present invention can significantly improve the seed cotton cleaning effect by detecting the moisture regain of the seed cotton in real time and automatically adjusting the key mechanical processing parameters of the cleaning machine accordingly. At the same time, during the cleaning process, the distance between adjacent grids, the speed of the spike roller, and the inclination angle of the cleaning section are precisely adjusted according to different moisture regains, effectively balancing the relationship between cleaning efficiency and fiber damage. For seed cotton with high moisture regain, the distance between adjacent grids and the speed of the spike roller are appropriately increased to reduce fiber entanglement and damage; for seed cotton with low moisture regain, the cleaning force is optimized to reduce energy consumption. At the same time, the moisture regain data is processed based on the sliding window filtering algorithm and the incremental PID control strategy to ensure the rapidity and accuracy of the parameter adjustment of the cleaning machine.

[0024] 2. The present invention constructs a quaternary high-order non-homogeneous regression model (including moisture regain, top roller speed, gap between adjacent gratings, and cleaning section inclination) based on cotton evaluation indicators such as cleaning efficiency, short fiber rate, and cotton fiber length and machine energy consumption machine evaluation indicators through full-factor experiments, and brings the processed moisture regain data into the above-mentioned quaternary regression model, converting it into a ternary high-order non-homogeneous regression model (including top roller speed, gap between adjacent gratings, and cleaning section inclination). While reducing parameters, it avoids the errors caused by the interaction between other machines and cotton moisture regain, and uses the improved NSGA-Ⅲ algorithm to perform multi-objective optimization on the constructed regression model's cleaning efficiency, short fiber rate, cotton fiber length and machine energy consumption machine evaluation indicators, achieving multiple optimization goals of maximizing cleaning efficiency, minimizing fiber damage and minimizing machine energy consumption. The algorithm ensures the diversity and balance of the optimization results through non-dominated sorting and reference point guidance mechanisms, providing a reliable theoretical basis for parameter adjustment of cleaning machines. At the same time, the "perception-decision-execution" closed-loop control system constructed by the present invention decouples and coordinates the perception layer, decision layer and execution layer through information entropy constraints, thereby improving the intelligence level and automation degree of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic diagram of the structure of the whole machine of the present invention;

[0026] Figure 2 Schematic diagram of the structure of the distance adjustment part between adjacent grid bars of the present invention;

[0027] Figure 3 This is a signal transmission flow chart of the control system of the present invention;

[0028] Figure 4 This is a flow chart of the multi-objective optimization algorithm for the decision layer of the control system of the present invention;

[0029] Figure 5 This is an overall flow chart of the seed cotton cleaning machine adjustment method of the present invention.

[0030] Among them, 1. Fixed frame; 2. Electric push rod; 3. T-shaped connecting arm; 4. Barbed roller shaft; 5. Magnetic switch; 6. Fixed platform; 7. Electric cylinder; 8. Front support platform; 9. Rear support platform; 10. Inclination sensor; 11. Circular grid; 12. Elliptical grid; 13. Rocker arm; 14. Arc slide; 15. Slide rail; 16. Support wall panel; 17. Side wall panel. DETAILED DESCRIPTION

[0031] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the present specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0032] Please see the attached Figure 1 -Attached Figure 5 The embodiment of the present invention provides a seed cotton cleaning machine that is adjustable based on different moisture regains, including a fixed frame 1, an inclination sensor 10, and an electric push rod 2. An electric cylinder 7 and a fixed platform 6 for adjusting the inclination of the cleaning part are installed on the fixed frame 1. The electric cylinder 7 is connected to the fixed platform 6 and is used to lift the outlet end of the cleaning machine to achieve changes in the inclination of the cleaning part. The electric cylinder 7 cooperates with the inclination sensor 10, and the inclination sensor detects the change in inclination in real time to achieve corresponding adjustments. The electric push rod 2 is connected to the lattice bar grid through a T-shaped connecting arm 3 to adjust the distance between adjacent lattices. A front support platform 8 and a rear support platform 9 are welded to the fixed frame 1. The front support platform 8 is tightly fitted with the outer support of the seed cotton cleaning machine but is not fixed. The rear support platform 9 is partially fitted with the outer support of the seed cotton cleaning machine. The fixed platform 6 is welded to the seed cotton cleaning machine. The upper side of the electric cylinder 7 is fixed to the fixed platform 6. The inclination sensor 10 is installed on the outer web and is consistent with the inclination angle of the lattice bar support frame.

[0033] Specifically, the fixed frame 1 serves as the basic support structure of the entire seed cotton cleaning machine, and is used to install and fix the various components of the cleaning machine, provide stability and mechanical strength for the equipment, and ensure the reliability and safety of the equipment during operation. The inclination sensor 10 is used to monitor the inclination of the cleaning section in real time. During the operation of the seed cotton cleaning machine, the inclination sensor 10 can feed back the actual inclination information of the cleaning section to the control system for precise inclination adjustment and control. The electric cylinder 7 is used to adjust the inclination of the cleaning section, which is connected to the fixed platform 6. The fixed platform 6 is installed on the seed cotton cleaning machine and provides a stable installation base for the electric cylinder 7. The electric cylinder 7 drives the cleaning section to rotate around the fixed axis through telescopic movement, thereby changing the inclination of the cleaning section. The inclination sensor 10 is installed on the outer web, consistent with the inclination angle of the lattice support frame, and can monitor the inclination changes of the cleaning section in real time, and feed back the signal to the control system to ensure the accuracy and stability of the inclination adjustment.

[0034] When the cleaning machine is working, the moisture regain of the seed cotton is first monitored in real time through the online moisture regain detection device, and the data is transmitted to the control system. Based on the moisture regain data, the control system adjusts the distance between adjacent grid bars through the electric push rod and T-shaped connecting arm, and detects the position of the push rod in real time through the linear switch, and calculates the distance between adjacent grid bars based on the logic of the host computer. At the same time, the inclination angle of the cleaning section is adjusted through the electric cylinder 7 and the fixed platform 6. The inclination sensor 10 monitors the inclination angle of the cleaning section in real time and feeds the signal back to the control system to ensure the accuracy of the inclination adjustment. The front support platform 8 and the rear support platform 9 provide stable support for the cleaning section, ensuring the stability and reliability of the equipment during operation. At the same time, the inverter frequency is changed to change the speed of the spike roller. The encoder installed on the roller detects and feeds back to the system host computer for adjustment to meet the cleaning needs of seed cotton with different moisture regain rates.

[0035] The electric push rod 2 is arranged obliquely and has the same inclination angle as the lattice support frame. The electric push rod 2 is connected to the T-shaped connecting arm 3 through the push rod. The T-shaped connecting arm 3 drives the arc slide 14 on the lower side of the two adjacent lattices to slide along the slide rail 15, thereby driving the rocker arm 13 connected to the upper elliptical lattice 12 to perform circular motion around the center of the ellipse, thereby changing the gap between the elliptical lattice 12 and the circular lattice 11.

[0036] Specifically, the electric push rod 2 is arranged diagonally, its inclination angle consistent with that of the lattice support frame, and is connected to a T-shaped connecting arm via a push rod. When the electric push rod 2 extends or retracts, it drives the T-shaped connecting arm, which in turn drives the curved slide 14 on the lower side of two adjacent lattices to slide along the slide rail 15. This sliding motion causes the rocker arm 13 connected to the upper elliptical lattice to move in a circular motion about the center of the ellipse, ultimately changing the gap between the elliptical and circular lattices, thereby adjusting the distance between adjacent lattices to meet the cleaning needs of seed cotton with different moisture regains.

[0037] The system also includes a spindle 4 for connecting the spike roller and a magnetic switch 5 mounted on the spindle 4. The spindle 4 is used to connect the spike roller, and the magnetic switch 5 is used to detect the speed of the spike roller. It also includes a support wall panel 16, a side wall panel 17, and an encoder. The support wall panel 16 and the side wall panel 17 support and secure the grating structure, and the encoder collects the speed data of the spike roller and provides feedback.

[0038] Specifically, the supporting wall panels 16 and the side wall panels 17 are used to support and fix the lattice structure to ensure the stability of the lattice during adjustment and operation; the encoder is responsible for collecting the speed data of the thorn roller and feeding back the signal, assisting the control system to achieve precise control of the speed of the thorn roller, thereby ensuring the stable operation and cleaning effect of the cleaning machine.

[0039] A parameter adjustment method for a seed cotton cleaning machine with adjustable moisture regains comprises the following steps:

[0040] S1, using the seed cotton moisture regain online detection device installed on the fixed frame 1 to collect the moisture regain data of the seed cotton to be processed in real time, and transmit the data to the control system host computer through the communication protocol;

[0041] S2. Preprocess the collected moisture regain data based on a sliding window filtering algorithm, remove outliers, and calculate the dynamic mean to improve the stability and reliability of the control input;

[0042] S3, integrating the moisture regain trend, using the incremental PID control strategy to perform feed-forward compensation on the moisture regain data, achieving a rapid response to fluctuations in seed cotton moisture content;

[0043] S4. Input the moisture regain data processed by feedforward compensation into a four-dimensional high-order non-homogeneous regression model (the modeling objects include four indicators: cleaning efficiency, short fiber rate, fiber length and machine energy consumption, and the independent variables are moisture regain x, nail roller speed ω, distance between adjacent grids γ, and cleaning part inclination θ). The moisture regain variable x is eliminated by dimensionality reduction and elimination method, and it is converted into a three-dimensional regression model (the independent variables only retain machine parameters such as ω, γ, and θ), eliminating the interactive coupling effect between moisture regain and mechanical parameters and reducing model errors. Based on this model, the NSGA-Ⅲ multi-objective optimization algorithm is used. For example, under the constraints:

[0044] Solve the Pareto optimal solution set and output the optimal machine parameter combination (ω * ,θ * , γ * )

[0045] S5. Based on the output of the regression model, the corresponding machining parameters are converted into electrical control signals and transmitted in real time to the industrial control computer (host computer) via the RS-485 communication interface (baud rate 115200bps). Based on the received instructions, the control system dynamically adjusts the core operating parameters of the seed cotton cleaning machine, including: the distance between adjacent grid bars (adjustable range 7-13mm, positioning accuracy ±0.1mm); the speed of the spike roller (adjustable range 500-1200rpm, steady-state error ±5rpm); and the inclination angle of the cleaning section (adjustable threshold 22.5°-45°, resolution 0.5°). This achieves adaptive matching of seed cotton cleaning process parameters under different moisture regain conditions (5%-16%).

[0046] To adjust the distance between adjacent bars, the electric push rod 2 drives the T-shaped connecting arm 3, driving the curved slide 14 on the lower sides of the adjacent bars to slide along the slide rail 15. This in turn causes the rocker arm 13 connected to the upper elliptical bar 12 to move in a circular motion about the center of the ellipse, changing the gap between the elliptical bar 12 and the circular bar 11. To adjust the speed of the spike roller, the three-phase asynchronous motor speed is adjusted by varying the frequency converter frequency, and an encoder is used to collect data feedback signals. To adjust the inclination angle of the cleaning section, the electric cylinder 7 is driven, combined with the fixed action of the electric cylinder 7 mounting platform 6 and the feedback signal from the inclination sensor 10, to achieve precise adjustment of the cleaning section inclination angle. When the moisture regain is high, the distance between adjacent bars is increased to reduce fiber damage, while the spike roller speed and the cleaning section inclination angle are adjusted to optimize cleaning efficiency. Operating parameters are adjusted based on the moisture regain, reducing energy consumption at low moisture regains and increasing cleaning power at high moisture regains. This ensures both cleaning quality and efficiency, achieving a dual optimization of energy conservation and efficiency.

[0047] It also includes the step of real-time monitoring of the working effect of the cleaning machine after adjustment, collecting data through sensors at various mechanical structure parts and feeding back to the control system host computer, so as to dynamically evaluate and adjust the operating status of the cleaning machine.

[0048] By integrating sensor parameters, algorithms, and industrial control technologies, a complete "perception-decision-execution" closed-loop control system has been constructed. This architecture breaks down the overall control problem into three levels: perception, decision-making, and execution. Information entropy constraints enable decoupling and coordination between these levels, facilitating real-time adjustment of machining parameters based on the moisture regain of seed cotton.

[0049] At the perception layer, the host computer collects feedback signals from each sensor to establish a dynamic feedback adjustment mechanism for real-time detection data. The specific steps are as follows:

[0050] The system uses an online moisture regain detection device to collect moisture regain data from processed seed cotton. This data is transmitted in real time to the control system host computer via the Modbus-RTU protocol and the RS-485 industrial bus. The communication cycle is extremely short, ensuring high real-time data acquisition. The control system preprocesses the moisture regain data using a sliding window filtering algorithm, removing outliers and calculating a dynamic mean to improve the stability and reliability of the control input. An incremental PID control strategy is employed during the moisture regain measurement process, incorporating the moisture regain trend (ΔRH / Δt) for feedforward compensation. This enables the system to quickly respond to fluctuations in seed cotton moisture content, maintaining an adjustment lag time of ≤1.5s and a steady-state error within ±0.5%, facilitating improved consistency in the seed cotton cleaning process. Parameters of sensors in various mechanical structural components are also collected.

[0051] In order to ensure the stability of the perception layer data and reduce the fluctuation caused by noise, the Figure 1 Some data in the perception layer must satisfy the corresponding relationship The symbols in the mathematical model are represented as follows:

[0052] is the derivative of the state variable of the perception layer (such as the dynamic response of the sensor). s A is the state variable of the perception layer (such as sensor measurement values: position, speed, temperature, etc.). s B is the system dynamic matrix, describing the sensor's own characteristics (such as damping, inertia). s is the control input matrix, which relates the effects of external inputs on the sensor. U is the external control input (e.g., control instructions, environmental stimuli). W is the noise or disturbance in the sensor layer (e.g., measurement noise, environmental interference).

[0053] At the decision layer, the moisture regain data x is transmitted to the control system host computer. After the internal processing unit completes the logical judgment, it is input into the four-variable high-order non-homogeneous regression model of cotton evaluation such as cleaning efficiency, short fiber rate, cotton fiber length and machine energy consumption in the corresponding moisture regain range, such as Figure 5 The dual control mode is designed as shown: it can achieve fast response by calling the optimal parameter database of the model results trained on the PC based on the improved NSGA-III algorithm model in the early history, greatly shortening the parameter adjustment response time; it can also dynamically correct the model parameters based on the online collected data and transmit the corresponding machine parameter signals to the lower computer;

[0054] like Figure 4 As shown in the figure, the optimal mechanical parameter set is calculated based on the improved NSGA-Ⅲ model. The specific core processing steps are as follows:

[0055] First, complete the initialization of the population, realize the standardization of the target, eliminate the difference in target dimensions, and ensure the balanced optimization of each target. Generate an initial feasible solution set. Randomly generate P that meets the conditions t (solution set), and x i ∈(L i ,U i ), where x i is the value corresponding to each mechanical parameter, L i ,U i are the upper and lower limits of each machining parameter respectively; perform non-dominated sorting F1>F2>...>F i Divide the solution set level, F1 is the optimal frontier, generate corresponding reference points and evenly distribute them to maintain the diversity of reference points. The reference point generation formula is: (Where H is the segmentation parameter, which determines the number of reference points m is the objective function dimension of the optimization problem (such as cleaning efficiency, short fiber rate, cotton fiber length, and machine energy consumption in seed cotton cleaning); 1m is an m-dimensional unit vector. For example, when m=2, 12=[1,1]°.

[0056] Associate reference points to map the solution to the nearest reference point Calculate the correlation distance to measure the closeness of the solution to the reference point for environment selection ( is the standardized target vector of the i-th solution, which realizes target standardization, eliminates target dimension differences, and ensures balanced optimization of each target z j is the coordinate of the jth reference point; is the Euclidean norm, representing the spatial distance between the solution and the reference point).

[0057] Niching selection, to avoid excessive aggregation of frontier solutions, press p j The number of reference point association solutions {p j =∑ i∈S δ ij ,δ ij =1 if association, p j is the number of solutions associated with the jth reference point; S is the current set of solutions to be selected; δ ij is an indicator function that is 1 when solution i is associated with reference point j and 0 otherwise to select scarce solutions.

[0058] Genetic operation performs local search to generate offspring population and complete SBX crossover (η c =20)+polynomial variation (η m =20){p c Crossover probability (SBX crossover), which indicates the probability of crossover in an individual; p m is the mutation probability (polynomial mutation), which is inversely proportional to the variable dimension n}.

[0059] When the conditions are met, the optimization of the results is stopped, the judgment is terminated and the Pareto solution set is output, and a signal is transmitted to the control system lower computer.

[0060] The system avoids the theoretical limitations of single-objective optimization and constructs a multi-objective collaborative optimization system based on the Pareto frontier. By introducing non-dominated sorting and reference point guidance mechanisms, a multi-dimensional balance of evaluation indicators such as cleaning efficiency and fiber damage control is achieved. Based on the previous full-factor experimental data, a polynomial regression model (p<0.001) of seed cotton regain and machine processing parameters on cleaning effect and processing energy consumption is constructed. The system uses an improved NSGA-III algorithm for multi-objective optimization and solution. Under the process constraints such as trash content (2±0.5%) and fiber length (21±1mm), the parameter combination with the highest comprehensive score is selected from the Pareto optimal solution set to achieve multiple optimization goals of maximizing cleaning efficiency, minimizing fiber damage, and minimizing machine energy consumption. In order to ensure the stability of decision-making layer data and reduce the fluctuations caused by its noise, the system Figure 1 Some data in the decision-making layer must satisfy the corresponding relationship The symbols in the mathematical model are represented as follows:

[0061] The derivative of the decision layer state variable (decision dynamic process), X d is the decision layer state variable (such as target trajectory, optimization parameters), f(X s ,θ) is a nonlinear decision function that depends on the state of the perception layer X s and parameter θ, θ is the decision parameter (such as weight coefficient, threshold), Γ(X d ) is the decision-making layer feedback mechanism (such as adaptive adjustment items and stability constraints).

[0062] At the control execution level, the system utilizes a Xinjie PLC and servo drive system as the core control unit, enabling rapid distribution of optimized parameters via PROFINET real-time Ethernet (cycle synchronization accuracy ≤ 1μs). Key parameters of the seed cotton cleaning machine, such as the distance between adjacent lattices, the speed of the spike roller, and the inclination angle of the cleaning section, can be automatically adjusted based on changes in moisture regain. Compared to traditional cleaning devices, this system offers greater adjustment flexibility. The modular design of the present invention utilizes an electric cylinder to adjust the inclination angle of the cleaning section, and an electric push rod to adjust the distance between adjacent lattices, ensuring precise and rapid adjustment.

[0063] To ensure the stability of the execution layer data and reduce the fluctuation caused by noise, some data in the execution layer must satisfy the corresponding relationship: The symbols in the mathematical model are represented as follows:

[0064] is the derivative of the state variable of the execution layer (dynamic response of the actuator) X e is the state variable of the execution layer (such as actuator output: force, displacement, current), K pis the proportional gain coefficient, which adjusts the fast response of the tracking error, K i is the integral gain coefficient, eliminating the steady-state error, ∫(X d -X e )dt is the cumulative error integral term, which is used for long-term error compensation.

[0065] At the control theory level, the system builds a dynamic adaptive mechanism with time-varying characteristics. The system introduces a feedforward-feedback composite control structure, which collects feedback signals from its sensors to effectively suppress the main disturbance frequencies in frequency domain analysis.

[0066] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A seed cotton cleaning machine with adjustable moisture regain, comprising a fixed frame (1), an inclination sensor (10) and an electric push rod (2) (expandable, which may make the frame appear to be the main body, and the method is weakened, mainly reflecting the adjustment of the corresponding mechanical modules by controlling multiple sensors), characterized in that: An electric cylinder (7) and a fixed platform (6) for adjusting the inclination angle of the cleaning portion are installed on the fixed frame (1); the electric cylinder (7) is connected to the fixed platform (6), and the electric cylinder (7) cooperates with the inclination sensor (10); the electric push rod (2) is connected to the lattice grid through the T-shaped connecting arm (3) and is used to adjust the distance between adjacent lattices; a front support platform (8) and a rear support platform (9) are welded to the fixed frame (1); the front support platform (8) is tightly fitted with the outer support of the seed cotton cleaning machine but is not fixed.

2. The seed cotton cleaning machine according to claim 1, wherein: The rear support platform (9) is fitted with the outer support portion of the seed cotton cleaning machine, the fixed platform (6) is welded to the seed cotton cleaning machine, the upper side of the electric cylinder (7) is fixed to the fixed platform (6), and the inclination sensor (10) is installed on the outer web and has the same inclination angle as the lattice support frame.

3. The seed cotton cleaning machine according to claim 1, wherein: The electric push rod (2) is arranged obliquely and has the same inclination angle as the lattice support frame. The electric push rod (2) is connected to the T-shaped connecting arm (3) through the push rod. The T-shaped connecting arm (3) drives the arc-shaped slideway (14) on the lower side of two adjacent lattices to slide along the slide rail (15), thereby driving the rocker arm (13) connected to the upper elliptical lattice (12) to perform a circular motion around the center of the ellipse, thereby achieving the change of the gap between the elliptical lattice (12) and the circular lattice (11).

4. The seed cotton cleaning machine according to claim 1, wherein: It also includes a thorn top roller rotating shaft (4) and an encoder (5) installed on the thorn top roller rotating shaft (4). The thorn top roller rotating shaft (4) is used to connect the thorn top roller, and the encoder (5) is used to detect the rotation speed of the thorn top roller.

5. The seed cotton cleaning machine according to claim 1, wherein: It also includes a supporting wall panel (16) and a side wall panel (17), wherein the supporting wall panel (16) and the side wall panel (17) are used to support and fix the lattice structure.

6. A parameter adjustment method for a seed cotton cleaning machine with adjustable moisture regain, characterized in that: The seed cotton cleaning machine adjustable based on different moisture regains according to any one of claims 1 to 5 comprises the following steps: S1, using a seed cotton moisture regain online detection device installed on a fixed frame (1) to collect moisture regain data of the seed cotton to be processed, and transmitting the data to a control system host computer through a communication protocol; S2. Preprocess the collected moisture regain data based on a sliding window filtering algorithm, remove outliers, and calculate the dynamic mean to improve the stability and reliability of the control input; S3, integrating the moisture regain trend, using the incremental PID control strategy to perform feed-forward compensation on the moisture regain data, achieving a rapid response to fluctuations in seed cotton moisture content; S4. Input the moisture regain data after feedforward compensation into a pre-constructed four-variable high-order non-homogeneous regression model covering cleaning efficiency, short fiber rate, cotton fiber length, and machine energy consumption. This model avoids the interaction between seed cotton moisture regain and other mechanical parameters, thereby converting it into a three-variable high-order non-homogeneous regression model. The model is then solved using multi-objective optimization to achieve the multiple optimization goals of maximizing cleaning efficiency, minimizing fiber damage, and minimizing machine energy consumption. S5. Based on the output results of the target optimization, the corresponding signals of the control system are converted to adjust the key mechanical processing parameters of the cleaning machine through the modular mechanical structure, including the distance between adjacent grid bars, the speed of the spike roller and the inclination angle of the cleaning part, so as to adapt to the cleaning needs of seed cotton with different moisture regain rates.

7. The parameter adjustment method of the seed cotton cleaning machine based on adjustable moisture regain according to claim 6, characterized in that: In the step S5, when adjusting the distance between adjacent grid bars, the electric push rod (2) drives the T-shaped connecting arm (3), driving the arc-shaped slideway (14) on the lower side of the two adjacent grid bars to slide along the slide rail (15), thereby causing the rocker arm (13) connected to the upper elliptical grid bar (12) to perform a circular motion with the center of the ellipse, changing the gap between the elliptical grid bar (12) and the circular grid bar (11), and detecting the position of the push rod in real time through the linear switch, and calculating the distance between the adjacent grid bars according to the logic of the host computer; when adjusting the speed of the nail roller, the speed of the three-phase asynchronous motor is adjusted by changing the frequency converter frequency, and the encoder is used to collect data and feedback signals to the system host computer to complete the adjustment of the corresponding frequency converter frequency; when adjusting the inclination angle of the cleaning part, the electric cylinder (7) is driven, combined with the fixing effect of the electric cylinder fixing platform (6) and the feedback signal of the inclination sensor (10), to achieve precise adjustment of the inclination angle of the cleaning part.

8. The parameter adjustment method of the seed cotton cleaning machine based on adjustable moisture regain according to claim 6, characterized in that: In the S4 step, a quaternary high-order non-homogeneous regression model is determined based on the results of the previous full-factor experiment, and the improved NSGA-III algorithm is used to perform multi-objective optimization on the model. By restricting the range of multiple processing parameters, multiple optimization goals of maximizing cleaning efficiency, minimizing fiber damage, and minimizing machine energy consumption are achieved.

9. The parameter adjustment method of the seed cotton cleaning machine based on adjustable moisture regain according to claim 6, characterized in that: It also includes the step of real-time monitoring of the working effect of the cleaning machine after adjustment, collecting data through sensors at various mechanical structure parts and feeding back to the control system host computer, so as to dynamically evaluate and adjust the operating status of the cleaning machine.

Images