Belt filter combined with slip detection

Abstract

The invention discloses a belt filter combined with slip detection. The belt filter comprises a machine body; the driving roller and the driven roller are respectively mounted at the driving end and the driven end of the machine body; a magnetic encoder label and a film type pressure sensor are embedded in the filter cloth; the inner side of the upper end of each protective support is provided with an anti-diffusion baffle, and the outer side of the upper end of each protective support is provided with a Hall sensor and an infrared temperature sensor; the motor is connected with the driving roller through a coupler; the filter cloth brackets and the supporting legs are respectively arranged at the top and the bottom of the machine body and are arranged at equal intervals; and the PLC control system is electrically connected with the film type pressure sensor, the Hall sensor and the infrared temperature sensor. Various sensors and an intelligent PLC control system are integrated, efficient real-time monitoring and accurate grading response of the slipping state of the filter cloth are achieved, meanwhile, the service life of equipment is prolonged by adopting a high-durability material and an anti-corrosion coating, and the intelligent level and operation stability of the equipment are improved.

Description

Technical Field

[0001] The present invention relates to the field of belt filters, and more specifically to a belt filter combined with slip detection. Background Art

[0002] In the production process of organic peroxides, especially in the post-treatment process of benzoyl peroxide, belt filters are widely used for solid-liquid separation. It filters the mixed materials through a filter cloth to effectively separate solid particles from the liquid. However, due to the high viscosity, volatility, and potential hazards of organic peroxides, higher requirements are imposed on the design and operation of the equipment, especially in preventing slip phenomena.

[0003] In the prior art, a single speed sensor is usually used to detect slip phenomena, but this method has obvious limitations. Due to the complex and variable working environment, such as changes in humidity, temperature, and material properties, a single sensor is easily interfered with, resulting in inaccurate slip detection. This inaccurate detection not only affects the normal operation of the equipment but may also lead to safety accidents, seriously affecting production.

[0004] At the same time, the filter cloths of existing belt filters mostly adopt single-layer or simple composite structures, which are difficult to meet the multiple requirements of high wear resistance, corrosion resistance, and stable embedding of sensors. During long-term operation, the filter cloth is easily worn and corroded, resulting in a short service life and frequent replacement, increasing the maintenance cost. At the same time, due to the problems of the filter cloth material and structure, the embedding and stable operation of the sensor also face challenges, further affecting the accuracy and reliability of slip detection.

[0005] In addition, in terms of the control system, existing belt filters mostly adopt a simple threshold judgment method to handle slip problems. It lacks intelligence and cannot flexibly adjust the equipment operation state according to the slip degree. For example, in the case of slight slip, it may overreact, resulting in equipment shutdown or reduced speed operation, affecting the processing efficiency; while in the case of severe slip, it may be slow to react and unable to take timely measures to protect the equipment, increasing the safety risk. In addition, the existing control system also lacks self-learning and optimization functions and cannot continuously optimize the slip detection and response strategy based on the equipment operation data.

[0006] Therefore, how to design a belt filter combined with slip detection, which can accurately detect slip, has high durability, and intelligent control ability, is an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0007] In view of this, the present invention provides a belt filter combined with slip detection, which integrates a variety of sensors to improve the accuracy of slip detection, adopts high-durability materials and advanced structures to extend the service life of the filter cloth and stabilize the operation of the sensors. At the same time, it also needs to be equipped with an intelligent PLC control system to achieve accurate determination of slip levels and hierarchical response strategies, and can continuously optimize its own performance according to the equipment operation data, improving the operation stability and intelligent level of the equipment.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] A belt filter combined with slip detection, comprising:

[0010] A body;

[0011] A driving roller and a driven roller, respectively installed at the driving end and the driven end of the body;

[0012] A filter cloth, covering the surfaces of the driving roller and the driven roller, with magnetic encoder tags and thin-film pressure sensors embedded inside;

[0013] A protective bracket, with a main body in a circular ring shape, symmetrically arranged on both sides of the driving end of the body. An anti-diffusion baffle is provided on the inner side of the upper end of each protective bracket, and a Hall sensor and an infrared temperature sensor are installed on the outer side of the upper end;

[0014] A motor, installed on the outer side of the driving end of the body, and connected to the driving roller through a coupling;

[0015] A filter cloth bracket and support feet, respectively arranged at the top and bottom of the body, and both are arranged at equal intervals;

[0016] A PLC control system, electrically connected to the thin-film pressure sensor, the Hall sensor, and the infrared temperature sensor.

[0017] Preferably, the body, the filter cloth bracket, and the support feet are made of 316L stainless steel, and the surfaces of the body and the filter cloth bracket are sprayed with an epoxy resin anti-corrosion coating.

[0018] Preferably, the filter cloth adopts a three-layer composite structure, including:

[0019] A filtering layer, woven from polyester or polypropylene fibers, and the surface is treated with fluorination to form a hydrophobic coating;

[0020] A sensing layer, composed of a flexible polyimide substrate, and a thin-film pressure sensor is embedded;

[0021] A support layer, woven from a composite of aramid fiber and polyurethane, and a magnetic encoder tag is embedded.

[0022] Preferably, before the magnetic encoder tag is embedded in the support layer, it is wrapped in a polyether ether ketone film by a vacuum packaging process, and the thickness of the film is 0.2-0.5 mm.

[0023] Preferably, the protective bracket is made of 316L stainless steel and is surface-treated with hard chromium plating; and the clearance between the inner diameter of the circular main body of the protective bracket and the outer diameter of the driving roller is 5-10 mm, which is used to limit the lateral displacement of the filter cloth.

[0024] Preferably, the anti-diffusion baffle is an arc-shaped baffle made of polytetrafluoroethylene.

[0025] Preferably, the PLC control system includes:

[0026] Signal acquisition and preprocessing module: It is used to collect the original data of the Hall sensor, infrared temperature sensor and thin-film pressure sensor, and perform sliding window normalization processing to eliminate instantaneous noise interference;

[0027] Dynamic feature extraction module: It is used to calculate the actual linear velocity of the filter cloth through the adaptive Kalman filter algorithm, and extract the surface temperature gradient of the driving roller and the variance of pressure fluctuation as dynamic feature parameters;

[0028] Multi-modal fusion decision-making module: It is used to perform joint probability inference through an improved Bayesian network model to generate a multi-modal fusion decision result;

[0029] Hierarchical response execution module: It is used to determine the slip level and trigger a hierarchical response strategy according to the multi-modal fusion decision result;

[0030] Self-learning optimization module: It is used to update the conditional probability table of the improved Bayesian network through an online incremental learning algorithm to form a closed-loop feedback mechanism.

[0031] Preferably, in the multi-modal fusion decision-making module, the improved Bayesian network model is expressed as:

[0032]

[0033] Among them, S fail represents the slip state, |R V | represents the actual linear velocity of the filter cloth, represents the surface temperature gradient of the driving roller, σ 2 represents the variance of pressure fluctuation, w v 、w T 、w P represents the weight coefficient.

[0034] Preferably, the improved Bayesian network model dynamically allocates weights in combination with the entropy weight method, including:

[0035] When R VWhen it exceeds ±5% for three consecutive cycles, w v Increase by 0.1;

[0036] When then, w T Increase by 0.1;

[0037] When ≥20% of the initial pressure, w P Increase by 0.1.

[0038] Preferably, in the hierarchical response execution module, determining the slip level and triggering the hierarchical response strategy includes:

[0039] If S fail ≥0.8, it is determined as severe slip, and emergency shutdown is triggered;

[0040] If 0.5 ≤ S fail <0.8, it is determined as potential slip, and speed reduction operation is started and an alarm is given;

[0041] If S fail <0.5, it is determined as the normal state.

[0042] Through the above technical solutions, compared with the prior art, the technical solutions of the present invention have the following

[0043] Advantages:

[0044] 1. The belt filter integrates a thin-film pressure sensor, a Hall sensor, and an infrared temperature sensor, and realizes multi-modal data fusion determination through a PLC control system. It can monitor the actual linear speed of the filter cloth, the surface temperature gradient of the driving roller, and the pressure fluctuation variance in real time, accurately judge the slip state, and trigger the hierarchical response strategy according to the slip level, including emergency shutdown, speed reduction operation, and alarm, ensuring the safety and stability of the equipment operation.

[0045] 2. The machine body, the filter cloth support, and the support feet are all made of 316L stainless steel and sprayed with an epoxy resin anti-corrosion coating, improving the corrosion resistance and service life of the equipment. The filter cloth adopts a three-layer composite structure, including a filtering layer with hydrophobic treatment, a sensing layer embedded with sensors, and a support layer woven by composite of aramid fiber and polyurethane, enhancing the wear resistance and durability of the filter cloth, and at the same time ensuring the stable operation of the sensors.

[0046] 3. The PLC control system uses an adaptive Kalman filter algorithm to calculate the actual linear speed of the filter cloth, conducts joint probability inference through an improved Bayesian network model, and dynamically allocates weights in combination with the entropy weight method, realizing accurate judgment of the slip state. In addition, the system also has a self-learning and optimization function, updates and improves the Bayesian network conditional probability table through an online incremental learning algorithm, forms a closed-loop feedback mechanism, continuously optimizes the slip detection and response strategy, and improves the intelligent level of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0048] Figure 1 Structural schematic diagram of a belt filter combined with slip detection provided by an embodiment of the present invention;

[0049] Figure 2 Structural schematic diagram of a filter cloth provided by an embodiment of the present invention;

[0050] Figure 3 Structural schematic diagram of a PLC control system provided by an embodiment of the present invention;

[0051] Figure 4 Schematic diagram of the specific implementation process in the post-treatment process of benzoyl peroxide provided by an embodiment of the present invention;

[0052] In the figure, 1 - body, 2 - driving roller, 3 - driven roller, 4 - filter cloth, 5 - protective bracket, 6 - anti-diffusion baffle, 7 - motor, 8 - filter cloth support, 9 - support feet, 10 - magnetic encoder label, 11 - thin-film pressure sensor, 12 - Hall sensor, 13 - infrared temperature sensor, 14 - PLC control system, 4a - filter layer, 4b - sensing layer, 4c - support layer, 13a - signal acquisition and preprocessing module, 13b - dynamic feature extraction module, 13c - multi-modal fusion determination module, 13d - hierarchical response execution module, 13e - self-learning optimization module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0054] As Figure 1 shown, this embodiment provides a belt filter combined with slip detection, including: body 1;

[0055] Driving roller 2 and driven roller 3 are respectively installed at the driving end and the driven end of the body 1;

[0056] The filter cloth 4 is wrapped around the surfaces of the driving roller 2 and the driven roller 3, and a magnetic encoder tag 10 and a thin-film pressure sensor 11 are embedded inside;

[0057] The protective bracket 5 has a circular main body and is symmetrically arranged on both sides of the driving end of the machine body 1. An anti-diffusion baffle 6 is provided inside the upper end of each protective bracket 5, and a Hall sensor 12 and an infrared temperature sensor 13 are installed outside the upper end;

[0058] The motor 7 is installed outside the driving end of the machine body 1 and is connected to the driving roller 2 through a coupling;

[0059] The filter cloth brackets 8 and the support feet 9 are respectively arranged at the top and bottom of the machine body 1, and are both arranged at equal intervals;

[0060] The PLC control system 14 is electrically connected to the thin-film pressure sensor 11, the Hall sensor 12, and the infrared temperature sensor 13.

[0061] The components of this equipment work together to meet the relevant treatment requirements of highly corrosive, flammable and explosive organic peroxides. And through multi-modal sensing such as magnetic encoders and pressure sensors, efficient real-time monitoring and accurate grading response of the filter cloth slipping state are realized, improving the intelligent level and operation stability of the equipment.

[0062] The following further details each structure of the above belt filter:

[0063] In this embodiment, the materials of the machine body 1, the filter cloth brackets 8 and the support feet 9 are 316L stainless steel, and the surfaces of the machine body 1 and the filter cloth brackets 8 are sprayed with an epoxy resin anti-corrosion coating; the combination of 316L stainless steel and the epoxy resin coating can resist the strong oxidative corrosion of organic peroxides, extend the service life of the equipment, and the maintenance period can be extended from once a week to once a month.

[0064] Furthermore, each bracket bearing in the filter cloth bracket 8 can be embedded with a silver-based conductive slip ring, which can eliminate static electricity accumulation.

[0065] In this embodiment, for the driving roller 2 and the driven roller 3, the roller matrix is made of high-strength aluminum alloy, and the surface is sprayed with a tungsten carbide-cobalt alloy wear-resistant coating; furthermore, an annular heat dissipation channel can be arranged inside the roller, copper heat dissipation fins are embedded in the channel, and it is connected to an external circulating cooling system;

[0066] As Figure 2 shown, the filter cloth 4 adopts a three-layer composite structure, including:

[0067] The filtering layer 4a is woven from polyester or polypropylene fibers, and the surface is fluorinated to form a hydrophobic coating;

[0068] The sensing layer 4b is composed of a flexible polyimide substrate, and a thin-film pressure sensor 11 is embedded;

[0069] The support layer 4c is made of a composite weave of aramid fiber and polyurethane, and a magnetic encoder tag 10 is embedded therein.

[0070] Furthermore, before the magnetic encoder tag 10 is embedded in the support layer 4c, it is encapsulated in a polyetheretherketone film by a vacuum encapsulation process, and the thickness of the film is 0.2 - 0.5 mm.

[0071] In this embodiment, the filter cloth 4 adopts a three-layer composite structure. The fluorination-treated filter layer 4a provides hydrophobic performance. The sensing layer 4b composed of a flexible polyimide substrate embeds a thin-film pressure sensor to monitor the pressure change in real time. The support layer 4c woven from a composite of aramid fiber and polyurethane enhances the strength and wear resistance of the filter cloth, and embeds a magnetic encoder tag vacuum-encapsulated in a polyetheretherketone film to achieve precise positioning.

[0072] It not only improves the filtration efficiency and durability of the filter cloth, but also realizes the precise detection of the slipping phenomenon, significantly enhancing the operating stability and intelligent level of the belt filter.

[0073] In this embodiment, the protective bracket 5 is made of 316L stainless steel and is surface-treated with hard chromium plating; and the inner diameter of the circular main body of the protective bracket 5 has a clearance of 5 - 10 mm from the outer diameter of the driving roller 2, which is used to limit the lateral displacement of the filter cloth 4.

[0074] The anti-diffusion baffle 6 is an arc-shaped baffle made of polytetrafluoroethylene. In addition, a diversion groove can be provided on its inner side, and a negative pressure suction device is connected to its end for the directional recovery of peroxide overflow slurry.

[0075] The motor 7 is an explosion-proof permanent magnet synchronous motor with a rated power of 7.5 kW. The output shaft is equipped with a torque limiter, which automatically cuts off the power when the load torque exceeds 15% of the rated value; the motor 7 is connected to the PLC control system 14 through the Modbus RTU communication protocol to receive speed control instructions and start / stop signals;

[0076] In this embodiment, the magnetic encoder tag 10 uses an RFID chip to store a unique position code, and the temperature resistance range is -40°C to 150°C; the thin-film pressure sensor 11 has an IP68 protection level, a response time ≤ 1 ms, and an operating temperature of -20°C to 120°C; the Hall sensor 12 adopts a dual-probe redundant design, with a specific model of Honeywell SS49E, and automatically switches to the standby probe when the output signal deviation rate ≥ 10%; the infrared temperature sensor 13 uses non-contact temperature measurement, with a specific model of Optris CTlaser3M, a temperature measurement range of 0 - 200°C, an accuracy of ±1°C, and a focused spot diameter of 2 mm.

[0077] Here, the thin-film pressure sensor 11 monitors the pressure change on the filter cloth in real time, the Hall sensor 12 detects the moving speed of the filter cloth, and the infrared temperature sensor 13 measures the surface temperature of the driving roller; specifically, the Hall sensor 12 measures the moving speed of the filter cloth by detecting the magnetic field change. When the magnetic encoder tag 10 on the filter cloth passes by the Hall sensor, the magnetic field generated by the built-in magnet in it will change the current distribution in the Hall sensor, thereby generating a measurable voltage change.

[0078] As Figure 3 shown, the PLC control system 14 includes:

[0079] Signal acquisition and preprocessing module 13a: It is used to collect the original data of the Hall sensor, infrared temperature sensor and thin-film pressure sensor, and perform sliding window normalization processing to eliminate instantaneous noise interference; specifically, this module takes the sensor data within a certain time period as a sliding window, and by calculating the mean and standard deviation of the data within the window, normalizes each data point to the range of [-1,1], thereby effectively reducing the impact of noise on subsequent processing.

[0080] Dynamic feature extraction module 13b: It is used to calculate the actual linear speed of the filter cloth through the adaptive Kalman filtering algorithm, and extract the surface temperature gradient of the driving roller and the pressure fluctuation variance as dynamic feature parameters;

[0081] In this module, for the actual linear speed of the filter cloth, the adaptive Kalman filtering algorithm is used. By fusing the driving roller rotation data provided by the Hall sensor, the moving speed of the filter cloth is estimated in real time and accurately, effectively filtering out the noise and interference in the measurement process. At the same time, for the surface temperature gradient of the driving roller, this module continuously monitors the data of the infrared temperature sensor and calculates the rate of change of temperature over time, that is, the temperature gradient, to reflect the working state of the roller and potential heat accumulation problems.

[0082] In addition, for the pressure fluctuation variance, the module collects the continuous readings of the thin-film pressure sensor, calculates the variance of the pressure change, thereby quantifying the stability of the contact between the filter cloth and the roller and identifying possible slipping signs. The extraction of these dynamic feature parameters not only provides comprehensive and accurate data support for the slipping determination of the system, but also ensures that the system can maintain a high degree of sensitivity and reliability in a complex and changeable working environment.

[0083] Multi-modal fusion determination module 13c: It is used to perform joint probability inference through an improved Bayesian network model to generate a multi-modal fusion determination result; the improved Bayesian network model is expressed as:

[0084]

[0085] Among them, S failIndicates the slipping state, |R V |Indicates the actual linear velocity of the filter cloth, Indicates the surface temperature gradient of the driving roller, σ 2 Indicates the pressure fluctuation variance, w v 、w T 、w P Indicates the weight coefficient.

[0086] Furthermore, the improved Bayesian network model combines the entropy weight method to dynamically allocate weights, including:

[0087] When R V exceeds ±5% for 3 consecutive cycles, w v is increased by 0.1;

[0088] When occurs, w T is increased by 0.1;

[0089] When ≥20% of the initial pressure, w P is increased by 0.1.

[0090] Hierarchical response execution module 13d: Used to determine the slipping level and trigger the hierarchical response strategy according to the multi-modal fusion determination result; including:

[0091] If S fail ≥0.8, it is determined as severe slipping and an emergency stop is triggered;

[0092] If 0.5 ≤ S fail <0.8, it is determined as potential slipping, and the speed reduction operation is started and an alarm is given;

[0093] If S fail <0.5, it is determined as the normal state.

[0094] The multi-modal fusion determination module 13c performs joint probability reasoning through the improved Bayesian network model to generate the multi-modal fusion determination result. This model comprehensively considers the actual linear velocity of the filter cloth, the surface temperature gradient of the driving roller, and the pressure fluctuation variance, and combines the entropy weight method to dynamically allocate weights to ensure accurate assessment of the slipping state under different working conditions. Furthermore, the hierarchical response execution module 13d triggers the corresponding response strategy according to the determination result, effectively improving the fault detection accuracy and response efficiency of the system.

[0095] Self-learning optimization module 13e: Used to update the conditional probability table of the improved Bayesian network through the online incremental learning algorithm to form a closed-loop feedback mechanism. The self-learning optimization module collects the sensor data and slipping event records during the operation of the belt filter press in real time, and gradually updates the conditional probability relationship of the Bayesian network using the online incremental learning algorithm.

[0096] Upon detecting a new slippage state or a change in operating conditions each time, the module automatically analyzes the data differences and adjusts the probability weights of each node in the network to make it more conform to the actual operating rules. At the same time, combined with the feedback of historical operation effects, the closed-loop mechanism continuously verifies and optimizes the model parameters to ensure that the system adapts to the changes in material characteristics and environmental fluctuations over time, ultimately improving the accuracy of slippage determination and the adaptability of control strategies.

[0097] Furthermore, the PLC control system further includes: a spray program control module configured to automatically execute a spray process according to a preset time sequence and dynamically adjust the spray pressure and the start / stop logic of the vacuum pump based on the slippage level determined by the multi-modal fusion determination module.

[0098] Specifically, it can send control instructions to an external spray device through the Modbus communication protocol based on a preset process time sequence; and dynamically adjust the spray pressure and the start / stop logic of the vacuum pump according to the slippage level output by the multi-modal fusion determination module; for example: in the case of potential slippage, reduce the spray pressure to 0.3 MPa and extend the vacuum pumping time to 20 s; in the case of severe slippage, immediately close the spray valve and stop the vacuum pump.

[0099] It should be noted that although the spray program control module is a part of the PLC control system, the spray device itself can exist as an external independent device; it can be connected to the PLC control system through an electrical interface, and the specific interfaces of the spray device can include: 4-20 mA analog input for receiving the spray pressure set value; digital output for controlling the start / stop of the spray valve and the vacuum pump; and an RS-485 communication link for transmitting status feedback signals.

[0100] This not only ensures the flexibility and controllability of the spray operation but also makes the entire belt filter system more modular in structure and function, easier to maintain and upgrade.

[0101] The PLC control system in this embodiment integrates multi-source data, dynamically analyzes the speed deviation, temperature gradient, and pressure fluctuation, determines the slippage level through a Bayesian network model, and drives hierarchical responses such as spraying, speed reduction, or shutdown to achieve intelligent control and safe operation of the post-treatment of benzoyl peroxide.

[0102] Further, as Figure 4 shown: The specific implementation process of the belt filter combined with slippage detection in the post-treatment process of benzoyl peroxide includes:

[0103] 1) Equipment startup and process initialization;

[0104] After the belt filter starts, the PLC control system first completes the self-check of the sensors to confirm normal communication of each component. Subsequently, it loads the preset process parameters:

[0105] Spraying sequence: Spraying water for 5 seconds, evacuating for 18 seconds, emptying for 3 seconds, and filter cloth moving for 3 seconds;

[0106] Theoretical linear velocity: Set to 0.24 m / s according to the filter cloth length and process cycle;

[0107] Safety threshold: When the surface temperature of the driving roller exceeds 60°C, a warning is triggered, and when it reaches 80°C, an emergency stop is performed.

[0108] 2) Spraying and filtering process;

[0109] Spraying water stage: The high-pressure spraying valve is opened, and the filter cloth surface is flushed with a water pressure of 0.3 to 0.5 MPa to remove the residual benzoyl peroxide slurry.

[0110] Evacuating stage: The vacuum pump is started, and the vacuum degree in the filter cloth adsorption area is maintained at -0.08 to -0.1 MPa to accelerate the solid-liquid separation.

[0111] Emptying stage: The drain valve is opened, and the filtrate is discharged into the storage tank through the diversion trough to avoid the residue of oxidizing medium.

[0112] Filter cloth moving stage: The driving roller drives the filter cloth to move to the next working station, and the displacement is calibrated in real time by the magnetic encoder label to ensure the moving accuracy.

[0113] 3) Multimodal slip detection and dynamic response;

[0114] Data acquisition:

[0115] Hall sensor, generates a pulse signal by detecting the magnetic field change of the driving roller and calculates the theoretical linear velocity; magnetic encoder label, real-time feedback of the actual displacement of the filter cloth to calibrate the actual linear velocity; pressure sensor, monitors the contact pressure between the filter cloth and the roller and calculates the pressure fluctuation variance; infrared temperature sensor, non-contact measurement of the surface temperature of the driving roller.

[0116] Furthermore, the PLC dynamically allocates weights through an improved Bayesian network model, comprehensively considering the speed deviation rate, temperature gradient, and pressure fluctuation variance, to obtain the determination result;

[0117] 4) Safety protection and dynamic optimization;

[0118] Temperature control: When the temperature exceeds 60°C, the internal circulating cooling system of the roller is started; when the temperature reaches 80°C, an emergency stop is triggered and nitrogen is sprayed to prevent the thermal decomposition of benzoyl peroxide.

[0119] Self-learning optimization: After each slip event, the PLC records the operation data, optimizes the Bayesian network parameters through incremental learning, and improves the subsequent detection accuracy. For example, when the moisture content of the material increases by 5%, the theoretical linear velocity threshold is automatically reduced by 0.03 m / s.

[0120] Explosion-proof and anti-corrosion design: The motor adopts an explosion-proof permanent magnet synchronous motor, and the protection level complies with the chemical explosion-proof standard; the filter cloth surface is fluorinated, and the 316L stainless steel body and polytetrafluoroethylene baffle are used to resist strong oxidizing corrosion.

[0121] 5) Circulating operation and maintenance guarantee;

[0122] The filter cloth support is embedded with a conductive slip ring to continuously conduct the static electricity generated during operation; the anti-diffusion baffle is connected to a negative pressure suction device to directionally recover the overflow slurry and reduce material loss; the tungsten carbide coated rollers, aramid fiber filter cloth and closed-loop cooling system ensure the stable operation of the equipment in a corrosive environment.

[0123] This belt filter uses PLC control to achieve automatic operation in the post-treatment of benzoyl peroxide, and completes solid-liquid separation through precise steps of spraying, vacuuming, emptying and filter cloth movement. The system uses multi-sensor data acquisition and an improved Bayesian network model to dynamically monitor the slipping situation, and is equipped with a temperature safety threshold and a self-learning optimization mechanism to improve the operation stability and detection accuracy. The equipment is also specially designed with explosion-proof and anti-corrosion measures and static electricity conduction devices to ensure safe and stable operation in a corrosive environment, while reducing material loss and improving production efficiency.

[0124] In this specification, the various embodiments are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts among the various embodiments, reference can be made to each other. For the system disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and reference can be made to the description of the method part for the relevant parts.

[0125] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A belt filter combined with slip detection, characterized in that: include: Body (1); The driving roller (2) and the driven roller (3) are respectively mounted on the driving end and the driven end of the machine body (1); A filter cloth (4) is coated on the surface of the driving roller (2) and the driven roller (3), and has a magnetic encoder tag (10) and a thin film pressure sensor (11) embedded therein; The protective bracket (5) has a main body in the shape of a circular ring and is symmetrically arranged on both sides of the driving end of the machine body (1). An anti-diffusion baffle (6) is arranged on the inner side of the upper end of each protective bracket (5), and a Hall sensor (12) and an infrared temperature sensor (13) are installed on the outer side of the upper end; The motor (7) is mounted on the outside of the driving end of the machine body (1) and connected to the driving roller (2) via a coupling; The filter cloth support (8) and the support foot (9) are respectively arranged at the top and bottom of the machine body (1), and are both arranged in multiple equal intervals; The PLC control system (14) is electrically connected to the film pressure sensor (11), the Hall sensor (12), and the infrared temperature sensor (13).

2. A belt filter combined with slippage detection according to claim 1, characterized in that: The machine body (1), the filter cloth support (8) and the support foot (9) are made of 316L stainless steel, and the surfaces of the machine body (1) and the filter cloth support (8) are sprayed with an epoxy resin anti-corrosion coating.

3. A belt filter combined with slippage detection according to claim 1, characterized in that: The filter cloth (4) adopts a three-layer composite structure, comprising: The filter layer (4a) is woven from polyester or polypropylene fibers, and the surface of the filter layer is fluorinated to form a hydrophobic coating; A sensing layer (4b) is formed of a flexible polyimide substrate and is embedded with a thin film pressure sensor (11); The support layer (4c) is made of aramid fiber and polyurethane composite weaving and is embedded with a magnetic encoder tag (10).

4. A belt filter combined with slippage detection according to claim 3, characterized in that: Before the magnetic encoder tag (10) is embedded in the support layer (4c), it is wrapped in a polyetheretherketone film using a vacuum packaging process, and the film thickness is 0.2-0.5 mm.

5. A belt filter combined with slippage detection according to claim 1, characterized in that: The material of the protective bracket (5) is 316L stainless steel, and the surface is hard chrome plated; and the gap between the inner diameter of the annular main body of the protective bracket (5) and the outer diameter of the driving roller (2) is 5-10 mm, which is used to limit the lateral displacement of the filter cloth (4).

6. A belt filter combined with slippage detection according to claim 1, characterized in that: The anti-diffusion baffle (6) is a curved baffle made of polytetrafluoroethylene.

7. A belt filter combined with slippage detection according to claim 1, characterized in that: The PLC control system (14) comprises: Signal acquisition and preprocessing module (13a): used to collect raw data from the Hall sensor, infrared temperature sensor and thin-film pressure sensor, and perform sliding window normalization processing to eliminate instantaneous noise interference; Dynamic feature extraction module (13b): used to calculate the actual linear velocity of the filter cloth through an adaptive Kalman filter algorithm, and to extract the surface temperature gradient of the driving roller and the pressure fluctuation variance as dynamic feature parameters; Multimodal fusion judgment module (13c): used to generate multimodal fusion judgment results by performing joint probability reasoning through an improved Bayesian network model; A graded response execution module (13d): used to determine the level of slippage and trigger a graded response strategy based on the multi-modal fusion determination result; Self-learning optimization module (13e): used to update and improve the Bayesian network conditional probability table through an online incremental learning algorithm to form a closed-loop feedback mechanism.

8. A belt filter combined with slippage detection according to claim 7, characterized in that: In the multimodal fusion decision module, the improved Bayesian network model is expressed as: Among them, S fail Indicates slipping state, |R V | indicates the actual linear speed of the filter cloth, represents the temperature gradient on the surface of the driving roller, σ 2 represents the pressure fluctuation variance, w v 、w T 、w P Represents the weight coefficient.

9. A belt filter combined with slippage detection according to claim 8, characterized in that: The improved Bayesian network model combines the entropy weight method to dynamically allocate weights, including: When R V When the error exceeds ±5% for 3 consecutive cycles, w v Increase by 0.1; when When T Increase by 0.1; When ≥20% of the initial pressure, w P Increased by 0.

1.

10. A belt filter combined with slippage detection according to claim 7, characterized in that: In the graded response execution module, determining the slip level and triggering the graded response strategy includes: If S fail ≥0.8, it is judged as severe slippage and triggers an emergency stop; If 0.5≤S fail <0.8, it is judged as potential slip, the speed reduction operation is started and an alarm is issued; If S fail <0.5, it is considered to be in normal state.

Images