Energy-saving and consumption-reducing intermittent negative pressure noil suction system control method of wool combing machine

Abstract

The present application relates to a kind of wool spinning comber energy-saving intermittent negative pressure wool suction system control method, belong to wool spinning combing machinery control technical field.The method includes: through the real-time acquisition equipment running angle signal of main shaft encoder, wind pressure is monitored in combination with differential pressure sensor, and time sequence-wind pressure benchmark coupling model is established;Based on the model, using micro negative pressure sensor monitors each area actual demand, dynamically controls electromagnetic valve and fan speed, realizes wind pressure on-demand distribution;Pulse width modulation technology is introduced, variable duty cycle control is implemented to each area electromagnetic valve, in combination with wool shedding load dynamic adjustment pulse frequency and duty cycle;Real-time monitoring energy consumption, in combination with historical data to establish LSTM wool shedding pattern recognition model, dynamically optimizes system parameters.The present application realizes the change from continuous operation to intermittent on-demand control of negative pressure supply, effectively reduces energy consumption, improves system response speed and control accuracy, and guarantees carding quality.

Description

Technical Field

[0001] This invention relates to the field of wool combing machinery control technology, specifically to a control method for an intermittent negative pressure wool removal system for energy saving and consumption reduction in wool combing machines. Background Technology

[0002] The wool combing machine is a key piece of equipment in wool textile production. Its function is to repeatedly comb the fiber bundles through mechanisms such as cylinders and top combs, removing short fibers, impurities, and lint to produce combed slivers with uniform fiber length and straight parallel lines. During this process, a large amount of lint is generated, which must be promptly removed using a negative pressure suction system to ensure combing quality and the normal operation of the equipment.

[0003] Currently, wool combing machines generally employ a continuously operating negative pressure wool removal system. This system typically consists of a single fan connected to each wool removal point via a duct network, operating continuously with a constant or manually adjustable airflow. While this constant negative pressure operation mode can guarantee basic wool removal performance, it has significant drawbacks: First, the combing machine's process is cyclical, with vastly different wool shedding amounts at different stages (such as cylinder combing, separation and bonding, and cleaning preparation). Continuous, constant negative pressure supply results in significant energy waste during low-shedding stages. Second, the clogging conditions and instantaneous loads vary across different wool removal areas of the combing machine (such as the cylinder area and top combing area). A uniform negative pressure supply cannot be distributed on demand, potentially leading to insufficient suction in critical areas and excessive supply in secondary areas, affecting product quality or increasing energy consumption. Third, even with simple start-stop control, its response speed and adjustment precision are difficult to match the high-speed, precise process rhythm of the combing machine, and frequent operation of the actuators can affect their lifespan.

[0004] Therefore, the main drawback of existing technologies is that the negative pressure supply is mismatched with the periodic and multi-regional dynamic demands of the combing machine, resulting in significant energy waste and making it difficult to achieve precise on-demand control while ensuring process effectiveness. To address these shortcomings, this invention aims to provide an intelligent control method capable of achieving on-demand negative pressure supply, precise allocation, and deep energy savings. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a control method for an intermittent negative pressure wool removal system for energy saving and consumption reduction in wool combing machines, which can realize intelligent control of negative pressure supply on demand, precise distribution and deep energy saving.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A control method for an intermittent negative pressure wool removal system in a wool combing machine for energy saving and consumption reduction includes: S100. The comber spindle encoder collects the equipment operating angle signal in real time, and combines it with the differential pressure sensor installed in the lint suction pipe to monitor the wind pressure change, and establishes a coupled model that includes time-series characteristics and wind pressure reference. S200. Based on the time-pressure reference coupling model, micro negative pressure sensors distributed in each hair-collecting area monitor the actual needs of each area in real time, control the opening and closing mode of the solenoid valves and the fan speed in each area, and realize the dynamic distribution of regional wind pressure. S300. Based on the realization of dynamic distribution of regional wind pressure, by introducing pulse width modulation technology, the solenoid valves of each region are controlled with variable duty cycle, and the pulse frequency and duty cycle parameters are dynamically adjusted in combination with the regional shedding characteristics. S400 monitors energy consumption in real time, establishes a feather shedding pattern recognition model based on historical operating data, and dynamically optimizes the angle-wind pressure mapping relationship in S100, the dynamic distribution of wind pressure in S200, and the pulse waveform parameters in S300.

[0007] As a further aspect of the present invention, the step of establishing a coupled model including time-series characteristics and wind pressure reference includes: S110. Equipment operating angle signal acquisition: The rotation angle signal of the main shaft is acquired in real time through an incremental encoder installed on the main shaft of the combing machine. The rotation angle signal is the timing reference for system operation and is used to accurately locate the current process stage of the combing machine. S120. System air pressure status monitoring: A differential pressure sensor is installed in the main pipeline of the lint suction system to monitor the actual air pressure value in the pipeline in real time. The actual air pressure value reflects the real-time working status of the entire negative pressure system. S130. Establishment of the time-pressure coupling model: Divide a complete working cycle of the combing machine into several consecutive process stages. At the same time, through data analysis and process optimization, determine an optimal benchmark air pressure setpoint for each process stage and construct a time-pressure benchmark model.

[0008] As a further aspect of the present invention, the time-series wind pressure reference model formula is as follows: ; in, This is an indicator function; the function value is 1 when the condition is true, and 0 otherwise. That is, the system at the current moment The corresponding reference wind pressure setting value; This is the optimal reference wind pressure setting. This refers to the number of consecutive process stages; For the current moment The rotation angle; for Any one of the continuous process stages.

[0009] As a further aspect of the present invention, the step of realizing dynamic distribution of regional wind pressure includes: S210. Multi-zone micro-negative pressure real-time monitoring: Micro-negative pressure sensors are installed in the key areas of the comber where the comber is designed to remove hair, and the local actual wind pressure in each area is monitored in real time. S220. Calculation of regional wind pressure demand deviation: Calculate the actual wind pressure for each region. Deviation from the reference wind pressure setting value The formula for calculating the deviation is: ; S230. Hierarchical priority dynamic scheduling, based on actual wind pressure. Deviation from the reference wind pressure setting value Based on the size and importance of the process in that region, a dynamic priority weight is calculated for each region. The formula for the dynamic priority weight is as follows: ; The number of key areas for picking up loose fibers in a combing machine; This is the sensitivity coefficient, used to adjust the response intensity of the region to wind pressure deviation; It serves as the basic priority constant, reflecting the importance of this region in the process; For the first Deviation values ​​of wind pressure demand in each region; S240. Global actuator coordinated control, which coordinates the control of fan speed and solenoid valves in each zone according to priority and total demand.

[0010] As a further aspect of the present invention, in the process of coordinating the control of the fan speed and the solenoid valves in each zone, The strategy adopted for controlling the fan speed is as follows: Target speed of the wind turbine Determined by both the reference wind pressure and the maximum regional demand deviation, the formula is as follows: ; , The proportional coefficient is obtained through system identification.

[0011] As a further aspect of the present invention, in the process of coordinating the control of the fan speed and the solenoid valves in each zone, The strategy adopted for solenoid valves is as follows: The continuous analog control commands for the solenoid valves in each zone are determined by their priority weights after normalization, to ensure that the total air pressure demand does not exceed the fan capacity. The formula is as follows: ; For continuous simulation control commands, It is a continuous variable between 0 and 1, representing the valve opening state under ideal continuous control, and providing an input reference for subsequent pulse modulation.

[0012] As a further aspect of the present invention, step S300 includes: S310. Pulse Width Modulation (PWM) technology is introduced to the switching signals of the solenoid valves controlling each zone. A fixed pulse period is set to convert the continuous analog control command calculated in S240 into a switching signal with a variable duty cycle within one cycle. S320. Opening-Duty Cycle Conversion and Variable Duty Cycle Control: This directly maps continuous analog control commands to the reference duty cycle of a PWM, ensuring that the opening time of the solenoid valve within one cycle is determined by the reference duty cycle; the formula is expressed as: ; ; For the solenoid valve in one cycle The on-time (pulse width) within the pulse; This is the reference duty cycle for the PWM. S330. Based on the dynamic optimization of duty cycle of the shedding load, a pulse control parameter optimization model is established so that the final optimized duty cycle is further correlated with the real-time shedding load of the area on the basis of the reference duty cycle, thereby achieving energy efficiency optimization. S340. Pulse frequency adaptive: The pulse frequency is dynamically adjusted according to the process stage. A higher frequency is used in the stage where lint shedding is severe to improve the response speed; a lower frequency (such as 2HZ) is used in the stage where lint shedding is sparse to further save energy and reduce the number of valve actions.

[0013] As a further aspect of the present invention, the pulse control parameter optimization model is expressed as follows: ; , These are the lower and upper limits of the duty cycle compensation coefficient, respectively. The maximum shedding load is designed for the system. As a regulating factor; Real-time shedding load in the area; As a regulatory factor .

[0014] As a further aspect of the present invention, step S400 includes: S410. System operation data monitoring and collection: Real-time monitoring and recording of all-dimensional data from the combing machine spindle encoder, negative pressure sensors in each area, and fan actuators to build a historical dataset; S420. A feather shedding pattern recognition model is established. Based on a historical dataset, a Long Short-Term Memory (LSTM) network model is trained. Leveraging the LSTM's ability to handle time-series data, the model learns the mapping relationship from historical sequences to future feather shedding load, outputting a prediction of future feather shedding load. The LSTM model is trained by minimizing the mean squared error (MSE) between the predicted and actual values. Its loss function is: ; The loss function value is a scalar used to measure the predictive performance of the LSTM model on the entire training dataset. The total number of training samples; No. The region is The actual shedding load at any given moment; No. In each region Predicted shedding load at any given time; S430. Multi-level parameter dynamic optimization: Utilizes future shedding load predictions and real-time energy consumption data to dynamically optimize key parameters. S440. Online Model Update and Validation: A validation mechanism is set up so that when the optimized parameters are proven to stably improve system performance in subsequent runs, the new parameter set and the corresponding running data are added to the historical dataset. The LSTM model is then incrementally learned or periodically retrained to ensure that it can continuously adapt to the latest state of the system, forming a complete intelligent closed loop of perception-decision-execution-optimization.

[0015] As a further aspect of the present invention, the historical dataset includes: combing machine spindle angle, time-series air pressure reference, actual air pressure in each region, air pressure demand deviation, dynamic priority weight, fan target speed, continuous simulation control commands for solenoid valves, real-time shedding load in each region, and total system energy consumption.

[0016] As a further aspect of the present invention, the key parameters include: optimizing different process stages. Optimal reference wind pressure setting value To achieve the lowest possible energy consumption while ensuring optimal process performance; optimize the sensitivity coefficient. and basic priority constant ; Optimize the proportional coefficient for fan speed control , ; Optimize the adjustment factor The lower and upper limits of the duty cycle compensation coefficient.

[0017] In summary, due to the adoption of the above technical solution, the beneficial technical effects of the invention are as follows: Compared with existing technologies, the control method provided by this invention produces significant comprehensive benefits by constructing a multi-level, adaptive intelligent control closed loop; This invention achieves a fundamental shift from continuous energy supply to intermittent, on-demand energy supply, resulting in significant direct energy savings. The core flaw of existing technologies lies in the disconnect between negative pressure supply and the time-varying process requirements of the combing machine. This invention, through a time-pressure coupling model established in step S100, strictly synchronizes negative pressure supply with the spindle angle (i.e., the process stage), fundamentally avoiding energy waste during periods of low shedding demand. This changes the traditional extensive mode of continuous high-volume operation, representing the primary and direct technical effect of energy saving and consumption reduction.

[0018] This invention solves the problem of competition for wind pressure resources across multiple regions, ensuring and improving process efficiency while saving energy. Addressing the issue that existing technologies cannot differentiate the priority of demand in different regions, this invention achieves dynamic spatial optimization of wind pressure resource allocation through step S200. This method not only considers the real-time wind pressure demand deviations in each region but also incorporates their process importance. Through dynamic priority weighting and a normalized allocation algorithm, it ensures that, when total wind pressure is limited, resources are prioritized for the most needed and critical regions (such as the Xilin area). This directly overcomes the shortcomings of traditional systems' one-size-fits-all supply methods, which may lead to localized poor hair removal or global oversupply, ensuring stable and improved combing quality while saving energy.

[0019] This invention performs in-depth energy efficiency optimization at the execution level, further tapping into energy-saving potential and improving system reliability. Building upon the on-demand allocation in S200, the S300 step introduces adaptive pulse width modulation (PWM) technology to convert continuous analog control commands into efficient switching signals. In particular, the strategy of dynamically optimizing the duty cycle and frequency based on load dynamics allows the system to automatically enter a deep energy-saving mode under low loads and respond rapidly under high loads. This refined control in the time dimension not only further reduces the average power consumption of the solenoid valves but also reduces the continuous stress and number of actuations on the valve components due to its switching mode, thereby extending equipment lifespan and improving the overall reliability and energy efficiency of the system. Attached Figure Description

[0020] Figure 1 A flowchart illustrating a control method for an intermittent negative pressure wool removal system in a wool combing machine for energy saving and consumption reduction; Figure 2 A flowchart of S100, a control method for an intermittent negative pressure wool removal system for energy saving and consumption reduction in a wool combing machine; Figure 3 A flowchart of S200, a control method for an intermittent negative pressure wool removal system for energy saving and consumption reduction in a wool combing machine; Figure 4 A flowchart of S300, a control method for an intermittent negative pressure wool removal system for energy saving and consumption reduction in a wool combing machine; Figure 5 This is a flowchart of S400, a control method for an intermittent negative pressure wool removal system for energy saving and consumption reduction in a wool combing machine. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0023] The core objective of this invention is to solve the problem of continuously high energy consumption in the negative pressure suction system of wool combing machines. Traditional systems typically operate with a constant high air volume, regardless of actual process requirements, resulting in significant energy waste.

[0024] The core idea of ​​this method is to transform the original extensive and continuous negative pressure supply into a refined, intermittent, and on-demand intelligent control mode. Through progressive control at four levels—time-sequential coordination, spatial allocation, time modulation, and intelligent learning—a complete closed loop of perception, decision-making, execution, and optimization is constructed.

[0025] like Figure 1 As shown, this application illustrates an exemplary control method for an intermittent negative pressure wool removal system in a wool combing machine, which aims to save energy and reduce consumption. The method specifically includes the following steps: A control method for an intermittent negative pressure wool removal system in a wool combing machine for energy saving and consumption reduction includes: S100. The comber spindle encoder collects the equipment operating angle signal in real time, and combines it with the differential pressure sensor installed in the lint suction pipe to monitor the wind pressure change, and establishes a coupled model that includes time-series characteristics and wind pressure reference. S200. Based on the time-pressure reference coupling model, micro negative pressure sensors distributed in each hair-collecting area monitor the actual needs of each area in real time, control the opening and closing mode of the solenoid valves and the fan speed in each area, and realize the dynamic distribution of regional wind pressure. S300. Based on the realization of dynamic distribution of regional wind pressure, by introducing pulse width modulation technology, the solenoid valves of each region are controlled with variable duty cycle, and the pulse frequency and duty cycle parameters are dynamically adjusted in combination with the regional shedding characteristics. S400 monitors energy consumption in real time, establishes a feather shedding pattern recognition model based on historical operating data, and dynamically optimizes the angle-wind pressure mapping relationship in S100, the dynamic distribution of wind pressure in S200, and the pulse waveform parameters in S300.

[0026] Please refer to Figure 2 The diagram shows an exemplary control method S100 for an intermittent negative pressure wool removal system for energy saving and consumption reduction in a wool combing machine. This step aims to establish a reference correspondence between the working cycle of the combing machine and the required negative pressure air pressure, and to provide a reference signal for the system to synchronize the cycle. Its contents include: S110. Equipment operating angle signal acquisition: The rotation angle signal of the main shaft is acquired in real time through an incremental encoder installed on the main shaft of the combing machine. The rotation angle signal is the timing reference for system operation and is used to accurately locate the current process stage of the combing machine. S120. System air pressure status monitoring: A differential pressure sensor is installed in the main pipeline of the lint suction system to monitor the actual air pressure value in the pipeline in real time. The actual air pressure value reflects the real-time working status of the entire negative pressure system. S130. Establishment of the time-pressure coupling model: Divide a complete working cycle (0° to 360°) of the combing machine into several consecutive process stages. Simultaneously, through data analysis and process optimization, determine an optimal benchmark air pressure setpoint for each process stage, and construct a time-pressure benchmark model: ; in, This is an indicator function; the function value is 1 when the condition is true, and 0 otherwise. That is, the system at the current moment The corresponding reference wind pressure setting value; This is the optimal reference wind pressure setting. This refers to the number of consecutive process stages; For the current moment The rotation angle; for Any one of the continuous process stages, for example: For Xilin's sorting stage, For the separation and joining stage, Cleaning preparation stage; Model output That is, the system at the current moment The corresponding reference wind pressure setpoint ensures that the negative pressure supply is synchronized with the process cycle of the combing machine, thus avoiding the continuous waste of wind energy from a macroscopic timing perspective.

[0027] Please refer to Figure 3 The diagram shows an exemplary control method S200 for an intermittent negative pressure wool suction system for energy saving and consumption reduction in a wool combing machine. The purpose of this step is to dynamically allocate air pressure resources based on the time-pressure reference provided in S100 and the actual local needs of each wool suction area. Its contents include: S210. Multi-zone micro negative pressure real-time monitoring: Micro negative pressure sensors are installed in key areas of the comber where hair is easily drawn (such as below the cylinder, near the top comb, and in the brush area) to monitor the local actual wind pressure in each area in real time. S220. Calculation of regional wind pressure demand deviation: Calculate the actual wind pressure for each region. Deviation from the reference wind pressure setting value The formula for calculating the deviation is: ; deviation The current wind pressure demand gap in the region has been quantified; when When the actual wind pressure in the area is lower than the benchmark setting, there is a wind pressure demand gap, and it is necessary to increase the negative pressure supply in the area. when This indicates that the actual wind pressure in the area is higher than the benchmark setting, indicating an excess of wind pressure supply. The negative pressure supply in the area can be appropriately reduced. When the wind pressure supply and demand in the region are balanced, it indicates that the region's wind pressure supply and demand are in equilibrium. S230. Hierarchical priority dynamic scheduling, based on actual wind pressure. Deviation from the reference wind pressure setting value Based on the size of the region and the importance of the process in that region (e.g., the Xilin region has the highest priority), a dynamic priority weight is calculated for each region. The dynamic priority weight formula is expressed as: ; The number of key areas for picking up loose fibers in a combing machine; This is the sensitivity coefficient, used to adjust the response intensity of the region to wind pressure deviation; It serves as the basic priority constant, reflecting the importance of this region in the process; For the first Deviation values ​​of wind pressure demand in each region; The dynamic priority weight formula ensures that the weight is always positive, and the normalization of the denominator keeps the weight calculation stable. This amplifies the direct impact of the deviation, while This ensures that even if the deviation is zero, the critical process area can still obtain basic resource allocation priority. S240. Global actuator coordinated control, based on priority and total demand, coordinates the control of fan speed and solenoid valves in each zone; The strategy adopted for controlling the fan speed is: target fan speed. Determined by both the reference wind pressure and the maximum regional demand deviation, the formula is as follows: ; , The proportionality coefficient obtained through system identification; The formula shows that the target speed of the wind turbine is... It consists of two parts and is a composite control strategy of feedforward + feedback; For feedforward control, follow the process cycle; The angle is obtained from the S100 time-series wind pressure reference model and is related to the current principal axis angle. The corresponding baseline wind pressure setting value; it represents the theoretically required basic wind pressure of the system under the current process stage; This ensures that the fan speed can proactively follow the changes in the combing machine's process rhythm in advance; when entering a process stage with high air pressure requirements, the reference air pressure increases, and the fan speed increases accordingly to provide sufficient air pressure reserve for the system; conversely, the speed is reduced to save energy, which is the basis for energy saving.

[0028] For feedback control and to respond to emergency needs, this part serves as a safety compensation; even if the baseline air pressure is set reasonably, there may be a serious local air pressure deficiency due to sudden blockage in a certain area; by responding to the maximum deviation, the system can quickly make up for the weakest link, ensure production quality, and prevent poor hair suction; this is the guarantee of the system's robustness (stability). For solenoid valves, the continuous analog control commands for each zone's solenoid valves are determined by their priority weights after normalization, to ensure that the total air pressure demand does not exceed the fan capacity. The formula is as follows: ; For continuous simulation control commands, It is a continuous variable between 0 and 1, representing the valve opening state under ideal continuous control, and providing an input reference for subsequent pulse modulation; The formula transforms priority weights into specific control quantities, enabling precise and on-demand allocation of wind pressure resources across space (each region), and providing an input reference for subsequent pulse modulation. The formula's functions include: Priority allocation compares the weight of a region to the total weight. It directly reflects the share of the total wind pressure resources that the region should occupy. The higher the weight, the greater the opening command and the more wind pressure supply the region receives. By ensuring total conservation and normalization, the sum of opening commands for all regions is always equal to 1. This mathematically guarantees that the sum of local demands for all regions will not exceed the total supply capacity of the system (i.e., the total wind pressure currently provided by the fan), thus avoiding the situation where all valves are opened to the maximum, which could lead to system collapse and achieving optimal allocation under resource constraints.

[0029] Please refer to Figure 4 The diagram shows a flowchart of an exemplary intermittent negative pressure wool removal system control method S300 for energy saving and consumption reduction in a wool combing machine. The purpose of this step is to further optimize energy consumption by converting continuous analog control commands into efficient switching signals through precise time-dimensional control, based on the dynamic allocation of wind pressure demand in each area (S200). S310. Pulse Width Modulation (PWM) technology is introduced to the switching signals of the solenoid valves controlling each zone. A fixed pulse period is set to convert the continuous analog control command calculated in S240 into a switching signal with a variable duty cycle within one cycle. S320. Opening-Duty Cycle Conversion and Variable Duty Cycle Control: This directly maps continuous analog control commands to the reference duty cycle of a PWM, ensuring that the opening time (pulse width) of the solenoid valve within one cycle is determined by the reference duty cycle; the formula is expressed as: ; ; For the solenoid valve in one cycle The on-time (pulse width) within the pulse; This is the reference duty cycle for the PWM. The above mapping relationship will abstract the duty cycle ratio. Converted into specific, executable time amounts Thus, through digital switching signals (in the period) Internal conduction The duration is equivalent to simulating the original continuous control command. The desired effect lays a direct control foundation for subsequent duty cycle optimization based on energy efficiency; it constitutes a complete conversion process from continuous analog control commands to discrete-time control signals, which is the theoretical basis for PWM technology to achieve precise control and energy efficiency optimization in this system. S330. Based on the dynamic optimization of the duty cycle of the shedding load, a pulse control parameter optimization model is established so that the final optimized duty cycle is further correlated with the real-time shedding load of the area on the basis of the baseline duty cycle, thereby achieving energy efficiency optimization; the pulse control parameter optimization model is expressed as: ; , These are the lower and upper limits of the duty cycle compensation coefficient, respectively. The maximum shedding load is designed for the system. As a regulating factor; Real-time shedding load in the area; As a regulatory factor ; In the formula: when When I was very young, Approaching 0 This means that under low load conditions, the duty cycle will be further reduced from the baseline demand to achieve deep energy savings; when near hour, Approaching 1, This ensures sufficient air pressure is provided under high load conditions. when When the function exhibits an upward convex shape, the duty cycle increases slowly in the low load region, resulting in more significant energy-saving effects. S340. Pulse frequency adaptive: The pulse frequency is dynamically adjusted according to the process stage. A higher frequency (e.g., 10Hz) is used in the stage where lint shedding is severe to improve the response speed; a lower frequency (e.g., 2Hz) is used in the stage where lint shedding is sparse to further save energy and reduce the number of valve actions.

[0030] Please refer to Figure 5 The diagram illustrates a flowchart of an exemplary intermittent negative pressure wool removal system control method S400 for energy saving and consumption reduction in a wool combing machine. The purpose of this step is to build a self-learning and optimization closed loop so that the system can adapt to changes. S410. System operation data monitoring and collection: Real-time monitoring and recording of all-dimensional data from the combing machine spindle encoder, negative pressure sensors in each area, and fan actuators to build a historical dataset; Historical datasets include: combing machine spindle angle, time-series air pressure baseline, actual air pressure in each region, air pressure demand deviation, dynamic priority weight, fan target speed, solenoid valve continuous simulation control commands, real-time shedding load in each region, and total system energy consumption. S420. A feather shedding pattern recognition model is established. Based on a historical dataset, a Long Short-Term Memory (LSTM) network model is trained. Leveraging the LSTM's ability to handle time-series data, the model learns the mapping relationship from historical sequences to future feather shedding load, outputting a prediction of future feather shedding load. The LSTM model is trained by minimizing the mean squared error (MSE) between the predicted and actual values. Its loss function is: ; The loss function value is a scalar used to measure the predictive performance of the LSTM model on the entire training dataset. The total number of training samples; No. The region is The actual shedding load at any given moment; No. In each region Predicted shedding load at any given time; S430. Multi-level parameter dynamic optimization: Utilizes future shedding load predictions and real-time energy consumption data to dynamically optimize key parameters. Key parameters include: optimizing different process stages Optimal reference wind pressure setting To achieve the lowest possible energy consumption while ensuring optimal process performance; optimize the sensitivity coefficient. and basic priority constant ; Optimize the proportional coefficient for fan speed control , ; Optimize the adjustment factor Duty cycle compensation coefficient: lower and upper limits; S440. Online Model Update and Validation: A validation mechanism is set up so that when the optimized parameters are proven to stably improve system performance in subsequent runs (such as a significant reduction in energy consumption per unit output), the new parameter set and the corresponding running data are added to the historical dataset. The LSTM model is then incrementally learned or periodically retrained to ensure that it can continuously adapt to the latest state of the system, forming a complete intelligent closed loop of perception-decision-execution-optimization.

[0031] The sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0032] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A control method for an intermittent negative pressure wool removal system in a wool combing machine for energy saving and consumption reduction, characterized in that, include: S100. The comber spindle encoder collects the equipment operating angle signal in real time, and combines it with the differential pressure sensor installed in the lint suction pipe to monitor the wind pressure change, and establishes a coupled model that includes time-series characteristics and wind pressure reference. S200. Based on the time-pressure reference coupling model, micro negative pressure sensors distributed in each hair-collecting area monitor the actual needs of each area in real time, control the opening and closing mode of the solenoid valves and the fan speed in each area, and realize the dynamic distribution of regional wind pressure. S300. Based on the realization of dynamic distribution of regional wind pressure, by introducing pulse width modulation technology, the solenoid valves of each region are controlled with variable duty cycle, and the pulse frequency and duty cycle parameters are dynamically adjusted in combination with the regional shedding characteristics. S400 monitors energy consumption in real time, establishes a feather shedding pattern recognition model based on historical operating data, and dynamically optimizes the angle-wind pressure mapping relationship in S100, the dynamic distribution of wind pressure in S200, and the pulse waveform parameters in S300.

2. The control method for an intermittent negative pressure wool suction system for energy saving and consumption reduction in a wool combing machine according to claim 1, characterized in that, The steps for establishing a coupled model that includes time-series features and a wind pressure reference include: S110. Equipment operating angle signal acquisition: The rotation angle signal of the main shaft is acquired in real time through an incremental encoder installed on the main shaft of the combing machine. The rotation angle signal is the timing reference for system operation and is used to accurately locate the current process stage of the combing machine. S120. System air pressure status monitoring: A differential pressure sensor is installed in the main pipeline of the lint suction system to monitor the actual air pressure value in the pipeline in real time. The actual air pressure value reflects the real-time working status of the entire negative pressure system. S130. Establishment of the time-pressure coupling model: Divide a complete working cycle of the combing machine into several consecutive process stages. At the same time, through data analysis and process optimization, determine an optimal benchmark air pressure setpoint for each process stage and construct a time-pressure benchmark model.

3. The control method for an intermittent negative pressure wool suction system for energy saving and consumption reduction in a wool combing machine according to claim 1, characterized in that, The steps for achieving dynamic distribution of regional wind pressure include: S210. Multi-zone micro-negative pressure real-time monitoring: Micro-negative pressure sensors are installed in the key areas of the comber where the wool is drawn off, and the local actual wind pressure in each area is monitored in real time. S220. Calculation of regional wind pressure demand deviation: Calculate the actual wind pressure for each region. Deviation from the reference wind pressure setting value The formula for calculating the deviation is: ； S230. Hierarchical priority dynamic scheduling, based on actual wind pressure. Deviation from the reference wind pressure setting value Based on the size and importance of the process in that region, a dynamic priority weight is calculated for each region. The formula for the dynamic priority weight is as follows: ； The number of key areas for picking up loose fibers in a combing machine; This is the sensitivity coefficient, used to adjust the response intensity of the region to wind pressure deviation; It serves as the basic priority constant, reflecting the importance of this region in the process; For the first Deviation values ​​of wind pressure demand in each region; S240. Global actuator coordinated control, which coordinates the control of fan speed and solenoid valves in each zone according to priority and total demand.

4. The control method for an intermittent negative pressure wool suction system for energy saving and consumption reduction in a wool combing machine according to claim 1, characterized in that, The S300 includes: S310. Pulse Width Modulation (PWM) technology is introduced to the switching signals of the solenoid valves controlling each zone. A fixed pulse period is set to convert the continuous analog control command calculated in S240 into a switching signal with a variable duty cycle within one cycle. S320. Opening-Duty Cycle Conversion and Variable Duty Cycle Control: This directly maps continuous analog control commands to the reference duty cycle of a PWM, ensuring that the opening time of the solenoid valve within one cycle is determined by the reference duty cycle; the formula is expressed as: ； ； For the solenoid valve in one cycle Opening time within; This is the reference duty cycle for the PWM. S330. Based on the dynamic optimization of duty cycle of the shedding load, a pulse control parameter optimization model is established so that the final optimized duty cycle is further correlated with the real-time shedding load of the area on the basis of the reference duty cycle, thereby achieving energy efficiency optimization. S340. Pulse frequency adaptive: The pulse frequency is dynamically adjusted according to the process stage. A higher frequency is used in the stage where lint shedding is severe to improve the response speed; a lower frequency is used in the stage where lint shedding is sparse to further save energy and reduce the number of valve actions.

5. The control method for an intermittent negative pressure wool suction system for energy saving and consumption reduction in a wool combing machine according to claim 1, characterized in that, The S400 includes: S410. System operation data monitoring and collection: Real-time monitoring and recording of all-dimensional data from the combing machine spindle encoder, negative pressure sensors in each area, and fan actuators to build a historical dataset; S420. A feather shedding pattern recognition model is established. Based on a historical dataset, a Long Short-Term Memory (LSTM) network model is trained. Leveraging the LSTM's ability to handle time-series data, the model learns the mapping relationship from historical sequences to future feather shedding load, outputting a prediction of future feather shedding load. The LSTM model is trained by minimizing the mean squared error (MSE) between the predicted and actual values. Its loss function is: ； The loss function value is a scalar used to measure the predictive performance of the LSTM model on the entire training dataset. The total number of training samples; No. In each region The actual shedding load at any given moment; No. In each region Predicted shedding load at any given time; S430. Multi-level parameter dynamic optimization: Utilizes future shedding load predictions and real-time energy consumption data to dynamically optimize key parameters. S440. Online Model Update and Validation: A validation mechanism is set up so that when the optimized parameters are proven to stably improve system performance in subsequent runs, the new parameter set and the corresponding running data are added to the historical dataset. The LSTM model is then incrementally learned or periodically retrained to ensure that it can continuously adapt to the latest state of the system, forming a complete intelligent closed loop of perception-decision-execution-optimization.

6. The control method for an intermittent negative pressure wool suction system for energy saving and consumption reduction in a wool combing machine according to claim 2, characterized in that, The formula for the time-series wind pressure reference model is: ； in, This is an indicator function; the function value is 1 when the condition is true, and 0 otherwise. That is, the system at the current moment The corresponding reference wind pressure setting value; This is the optimal reference wind pressure setting. This refers to the number of consecutive process stages; For the current moment The rotation angle; for Any one of the continuous process stages.

7. The control method for an intermittent negative pressure wool suction system for energy saving and consumption reduction in a wool combing machine according to claim 3, characterized in that, During the coordinated control of the fan speed and the solenoid valves in each zone... The strategy used to control the fan speed is as follows: Target speed of the wind turbine Determined by both the reference wind pressure and the maximum regional demand deviation, the formula is as follows: ； , The proportional coefficient is obtained through system identification.

8. The control method for an intermittent negative pressure wool suction system for energy saving and consumption reduction in a wool combing machine according to claim 3, characterized in that, In the process of coordinating the control of the fan speed and the solenoid valves in each zone, the strategy adopted by the solenoid valves is as follows: The continuous analog control commands for the solenoid valves in each zone are determined by their priority weights after normalization, to ensure that the total air pressure demand does not exceed the fan capacity. The formula is expressed as: ； For continuous simulation control commands, It is a continuous variable between 0 and 1, representing the valve opening state under ideal continuous control, and providing an input reference for subsequent pulse modulation.

9. The control method for an intermittent negative pressure wool suction system for energy saving and consumption reduction in a wool combing machine according to claim 4, characterized in that, The formula for the pulse control parameter optimization model is expressed as follows: ； , These are the lower and upper limits of the duty cycle compensation coefficient, respectively. The maximum shedding load is designed for the system. As a regulating factor; Real-time shedding load in the area; It is a regulating factor.

10. The control method for an intermittent negative pressure wool suction system for energy saving and consumption reduction in a wool combing machine according to claim 5, characterized in that, The historical dataset includes: combing machine spindle angle, time-series air pressure reference, actual air pressure in each region, air pressure demand deviation, dynamic priority weight, fan target speed, solenoid valve continuous simulation control commands, real-time shedding load in each region, and total system energy consumption. Key parameters include: optimizing different process stages Optimal reference wind pressure setting value To achieve the lowest possible energy consumption while ensuring optimal process performance; optimize the sensitivity coefficient. and basic priority constant ; Optimize the proportional coefficient for fan speed control , ; Optimize the adjustment factor The lower and upper limits of the duty cycle compensation coefficient.

Images