A multi-stage cooperative potassium removal activated carbon continuous treatment device and a control method thereof

Abstract

The application discloses a kind of multi-stage synergic de-potassium activated carbon continuous processing device and control method thereof, belong to activated carbon purification technical field.The device includes the water washing unit, acid washing unit, ion exchange unit and high-temperature calcination unit connected in sequence, and material conveying assembly, soft measurement prediction system and central control system.Through intelligent anti-blocking material conveying system, blockage risk is sensed and intervened in real time;Based on soft measurement prediction system, the real-time online estimation of activated carbon potassium content is realized;Using dynamic pressure balance control system, the safety pressure gradient between each unit is maintained.Central control system cooperates with the three, realizes the closed-loop control of material flow, information flow and atmosphere flow, solves the engineering problems such as material interruption, control lag and atmosphere imbalance in continuous production, significantly improves production efficiency, product quality and operation safety.

Description

Technical Field

[0001] This invention relates to the field of high-efficiency purification technology of activated carbon materials, specifically to a multi-stage synergistic potassium removal continuous activated carbon treatment device and its control method, which is particularly suitable for solving specific engineering and control problems encountered in the continuous and intelligent production process of alkali activated carbon. Background Technology

[0002] High-performance activated carbon, especially products used in supercapacitors, pharmaceuticals, and food, has extremely stringent requirements regarding impurity content, particularly potassium ion residue (typically below 50 ppm). To address this, the industry has developed multi-stage synergistic potassium removal processes, such as the four-step method of "water washing-acid washing-ion exchange-nitrogen calcination" disclosed in patent application CN120483155A, which effectively reduces potassium residue to below 20 ppm, demonstrating excellent laboratory results. To further achieve industrial-scale production, continuous processing devices have become an inevitable direction. For example, patent application CN120398057A proposes a multi-stage synergistic potassium removal continuous processing device and control method, connecting each process unit through material conveying components and introducing online detection and central control to form intelligent process control.

[0003] However, through in-depth research and practice, the inventors discovered that transitioning from intermittent processes to truly stable and efficient continuous production presents a series of "last mile" challenges at the equipment operation level that existing technologies have not fully addressed. These problems are specifically reflected in the details of research and development and pilot production:

[0004] 1. The Challenges of Continuous and Stable Material Conveying: The core of continuous production is the stability of material flow. Activated carbon, especially wet carbon after washing and acid washing, undergoes significant changes in its flowability and adhesiveness. When conveying such materials, screw conveyors and vibrating feeders, as described in Comparative Document 2, are prone to internal bridging or outlet blockage due to uneven material moisture content and slight agglomeration. While existing technologies mention equipment, they do not address how to detect blockage trends in real time and intervene automatically. Once a blockage occurs, manual intervention is necessary, interrupting continuous production and negating any economies of scale.

[0005] 2. The Challenge of Lag in Online Detection and Closed-Loop Control of Key Quality Parameters: Real-time sensing of key indicators is a prerequisite for intelligent control. Reference document 2 mentions online monitoring of potassium ion concentration, but in practical engineering, directly, quickly, and accurately detecting the potassium content within solid activated carbon particles online is extremely difficult. It typically relies on offline sampling and laboratory analysis, with results lagging by several hours. This causes the central control system to constantly adjust based on outdated data (such as adjusting acid washing pH or calcination temperature), failing to respond promptly to batch fluctuations in raw carbon, resulting in incomplete potassium removal from some products or excessive consumption of resources (acid, energy). The process control is essentially an "open-loop" or "large-lag closed-loop," significantly reducing accuracy and efficiency.

[0006] 3. The Challenge of Pressure and Atmosphere Balance in Multi-Unit Integrated Systems: Continuous processing units physically connect multiple units with vastly different physical and chemical environments, such as water washing (atmospheric pressure, water environment), acid washing (atmospheric pressure, acidic liquid environment), ion exchange (slight positive pressure), and high-temperature calcination (high temperature, inert atmosphere). Material transfer between units requires the use of components such as gas locks and valves. Existing technologies do not adequately address the sealing reliability of these connection points and the dynamic pressure balance between units. In practice, pressure imbalance can easily lead to: a) leakage of inert protective gas (such as N2) from the high-temperature calcination section to the upstream section, resulting in resource waste; b) more dangerously, air or water vapor from the upstream section can enter the high-temperature calcination section, causing activated carbon oxidation or even combustion accidents; c) turbulent airflow within the system, exacerbating dust dispersion and heat loss.

[0007] In summary, while existing technologies point towards multi-level collaboration and continuous processing, their devices and controls still have significant shortcomings when dealing with the aforementioned specific and detailed engineering problems, hindering the technology from moving from blueprints to efficient, safe, and reliable large-scale industrial production. Therefore, there is an urgent need for a new type of device and control method capable of systematically solving these practical node problems. Summary of the Invention

[0008] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a multi-stage synergistic potassium removal continuous activated carbon processing device and its control method. This solution not only inherits the process advantages of multi-stage synergistic potassium removal, but also focuses on providing innovative integrated solutions to three major detailed challenges exposed in actual continuous and intelligent production—material flow interruption, control signal lag, and system pressure imbalance—thereby achieving truly stable, efficient, intelligent, and safe large-scale continuous production.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] A multi-stage synergistic potassium removal continuous activated carbon treatment device includes a water washing unit, an acid washing unit, an ion exchange unit, and a high-temperature calcination unit connected in sequence, as well as a material conveying assembly, a soft measurement and prediction system, and a central control system. The core improvement of this invention lies in three synergistic technical points:

[0011] Intelligent anti-blocking material conveying system

[0012] To address the problem of production interruptions caused by clogging in the conveying of wet activated carbon, this invention features an intelligent upgrade to the material conveying components. Key conveying units (such as the screw conveyor connecting the washing and acid washing processes) integrate highly sensitive vibration sensors and drive shaft torque sensors. These sensors transmit real-time data to the central control system. The central control system has a pre-set "vibration frequency-drive torque" correlation analysis model. This model can identify early signs of clogging (such as a gradual increase in torque and a decrease in vibration). Once a risk is identified, the system can automatically trigger intervention procedures, such as instantaneously increasing the rotation speed for impact feeding, briefly reversing the direction to disrupt the bridging structure, or adjusting the associated pre-drying unit (such as using a hollow screw shaft with a heating medium) to improve material flowability, thereby eliminating clogging before it occurs and ensuring absolute continuity of material flow.

[0013] Online Real-Time Potassium Content Prediction System Based on Soft Measurement

[0014] To address the issue of control lag caused by the inability to directly detect the potassium content of activated carbon in real time, this invention creatively introduces "soft measurement" technology. Sensor arrays are deployed at key locations such as the pickling solution outlet and ion exchange column to collect easily measurable secondary variables in real time, such as the pH, conductivity, and temperature of the pickling solution, as well as the bed pressure drop and inlet / outlet concentration difference of the ion exchange column. The central control system incorporates a machine learning model (such as a neural network) trained on a large amount of historical production data. This model uses these easily measurable secondary variables as input to infer and output the estimated potassium content of the activated carbon currently being processed in real time. This prediction is updated in seconds, providing the central control system with real-time, continuous feedback of key quality parameters, enabling precise and proactive adjustment of downstream ion exchange regeneration and calcination temperatures.

[0015] Dynamic pressure and atmosphere balance control system

[0016] To address the pressure imbalance and atmosphere cross-contamination issues caused by multi-unit series connection, this invention constructs an active dynamic pressure balance control system. High-precision micro-differential pressure sensors are installed in the cavities of the main process units to monitor the pressure differences within and between each unit in real time. Pressure regulating valve groups and atmosphere isolation valves, precisely modulated by the central control system, are installed in the material conveying channels between units. The central control system sets a safe pressure gradient between different units according to process requirements (e.g., maintaining a safe pressure gradient of "calcination unit > ion exchange unit > front-end unit" to prevent gas cross-contamination). The system dynamically adjusts the opening of each valve by comparing the pressure sensor data with the set gradient in real time, acting like a precise "pressure control center" to maintain a stable pressure distribution and a pure atmosphere environment within the system, fundamentally eliminating gas cross-contamination, ensuring safety, and reducing protective gas consumption.

[0017] The three technological improvements mentioned above are not isolated, but are deeply coupled through the central control system to form an enhanced intelligent closed loop of "perception-decision-execution":

[0018] Intelligent conveying systems ensure the physical continuity of the "material flow";

[0019] Soft measurement systems provide a real-time, continuous "quality information flow";

[0020] The pressure balancing system ensures the stable and controlled flow of atmosphere and environment.

[0021] The central control system integrates these three pieces of information to make coordinated decisions. For example, when soft sensing predicts that the potassium content of a batch of activated carbon is too high, it can accelerate the switching rhythm of the ion exchange column in advance and fine-tune the calcination temperature curve, while ensuring that the conveying system and pressure balancing system adapt to this adjustment. The synergistic effect of the intelligent closed loop of the three technical improvements enables this invention to achieve efficient, stable, and continuous production.

[0022] Compared with the prior art, the beneficial effects of the present invention are:

[0023] 1. The intelligent anti-blockage conveying system fundamentally solves the biggest bottleneck in continuous production—the material interruption problem, significantly improving the annual operating time of the unit (effective production time is expected to increase by more than 15%) and reducing maintenance intensity.

[0024] 2. By using real-time prediction of potassium content based on soft measurement, the feedback time of key quality control points is reduced from several hours to several seconds, enabling the entire process system to achieve true real-time closed-loop intelligent control, improving product consistency (potassium content fluctuations are reduced by more than 30%), and optimizing chemical and energy consumption.

[0025] 3. The dynamic pressure balance control system ensures the safety and stability of the complex integrated device, avoids product deterioration or safety risks caused by atmospheric contamination, and reduces unnecessary consumption of inert protective gas, thereby lowering operating costs.

[0026] 4. The innovation and synergy of the above three subsystems have jointly overcome the core engineering challenges in moving from advanced technology to large-scale continuous industrial production, making the stable, efficient, and economical production of high-performance, low-potassium activated carbon a reality. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This is a flowchart illustrating the overall device structure and control method according to an embodiment of the present invention.

[0029] Figure 2 This is a schematic diagram of the working process of an intelligent anti-blocking spiral conveyor system in one embodiment of the present invention.

[0030] Figure 3 This is a schematic diagram of the workflow of a soft measurement prediction system in one embodiment of the present invention.

[0031] Figure 4 This is a schematic diagram of the working process of the dynamic pressure balance control system in one embodiment of the present invention. Detailed Implementation

[0032] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention. It should be noted that relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0033] This invention provides a multi-stage synergistic potassium removal continuous activated carbon treatment device and its control method, aiming to solve three specific and interrelated engineering bottlenecks exposed in the process of transitioning existing activated carbon potassium removal technology from efficient laboratory processes to industrial continuous, intelligent, and large-scale production:

[0034] Physical flow interruption problem: When wet, viscous activated carbon is transported between units, it is very easy for "bridging" and "blockage" to occur, which will force the interruption of the promised "continuous production" and prevent the realization of the advantages of scale.

[0035] Information lag problem: The core parameter that determines product quality (potassium content of activated carbon) cannot be obtained in real time. The central control system can only make adjustments based on analysis data or fixed formulas that are several hours behind, which cannot cope with raw material fluctuations, resulting in low control accuracy, unstable quality and waste of resources.

[0036] Energy flow and atmosphere imbalance problem: When multiple units and multiple atmosphere environments are integrated into a closed system, the balance of pressure and atmosphere is extremely fragile, and gas cross-mixing is prone to occur, which can lead to safety hazards (such as air entering the high-temperature calcination section), product deterioration, and energy waste.

[0037] In this invention, it is particularly important to note that:

[0038] Multi-stage synergy: In this application, it specifically refers to integrating four originally independent potassium removal process units—"water washing," "acid washing," "ion exchange," and "high-temperature calcination"—into a continuous, functionally complementary organic whole through material conveying and intelligent control systems. The "synergy" means that the various treatment stages are not simply connected in series; rather, each stage compensates for the limitations of the previous one (for example, acid washing removes most of the bound potassium, ion exchange adsorbs residual free potassium, and calcination decomposes the most stable bound potassium), ultimately achieving a deep and efficient potassium removal effect that cannot be achieved by a single method.

[0039] Continuous processing unit: Unlike traditional intermittent or batch processing equipment, this unit refers to a process in which the transfer and processing of materials (activated carbon) between various process units is continuous from infeed to discharge. The core challenge in achieving "continuous" processing lies in solving the engineering problem of ensuring stable and seamless connection of material flow, information flow, and energy flow between heterogeneous units.

[0040] Potassium residue in activated carbon: refers to the presence of potassium in activated carbon in ionic form (e.g., potassium hydroxide) in activated carbon prepared using alkaline activating agents such as potassium hydroxide. Residual potassium elements exist in the pore surface and framework in the form of potassium carbonate (K2CO3), oxide (K2O), or more complex forms. Their content is usually measured in ppm (parts per million) and is a key indicator for measuring the purity of activated carbon and affecting its performance in applications such as supercapacitors and medicine.

[0041] Soft sensing is an indirect measurement technique that uses readily measurable auxiliary variables (such as pH, temperature, and flow rate) to estimate in real time the dominant variable (such as the potassium content of activated carbon), which is difficult or impossible to measure directly online. The core of this application lies in using a machine learning model to establish a high-precision mapping relationship from multi-source process data to potassium content, thereby solving the problem of quality control signal lag.

[0042] Dynamic pressure balance: In a continuous system integrating atmospheric pressure (water washing), slight negative pressure (acid washing), slight positive pressure (ion exchange), and positive pressure inert atmosphere (high-temperature calcination), a stable and controllable pressure gradient is maintained between each processing unit through active monitoring and adjustment. The purpose is to prevent harmful gases from interfering with each other, ensure process safety, and reduce protective gas consumption, which is crucial for the safe and continuous operation of this unit.

[0043] The following will detail the specific technical solution of this invention for solving the aforementioned engineering-level bottleneck problem:

[0044] Example 1

[0045] Please see Figure 1-4 This embodiment provides a multi-stage synergistic potassium removal continuous activated carbon treatment device, including a water washing unit, an acid washing unit, an ion exchange unit and a high-temperature calcination unit connected in sequence, as well as an intelligent anti-clogging conveying system, a soft measurement prediction system, a dynamic pressure balance system and a central control system.

[0046] The water washing unit, as a pretreatment section, is primarily responsible for countercurrent washing of the alkaline activated carbon raw material using hot water at 60-80℃. Its core function is to physically dissolve and remove soluble potassium salts (such as KOH and KCl) adhering to the surface and macropores of the activated carbon, creating favorable conditions for subsequent deep chemical depotassium removal, while simultaneously reducing the impurity load entering the acid washing unit. Located at the very upstream of the entire continuous processing flow, the water washing unit is the inlet unit for the activated carbon material. Its outlet is directly connected to the inlet of the downstream acid washing unit via a material conveying assembly.

[0047] The acid washing unit performs the main chemical depotassium removal task. It reacts with chemically bound potassium compounds (such as K₂CO₃) in activated carbon by spraying or soaking it with a mixed acid of a specific concentration (such as HCl / HNO₃), converting them into soluble potassium ions. The acid washing unit is located immediately after the water washing unit. It receives the wet activated carbon from the water washing unit, and after acid washing, the activated carbon slurry or wet activated carbon is output to the next stage of treatment or solid-liquid separation stage. Finally, the solid activated carbon is sent to the ion exchange unit.

[0048] The ion exchange unit performs deep purification and potassium removal. Utilizing the selective adsorption capacity of sodium-type cation exchange resins (such as D001 type), it captures trace amounts of free potassium ions remaining in the liquid phase (washing liquid) or on the surface of the micropores of solid activated carbon after acid washing. ), through ion exchange ( and (Replacement) to deeply remove it, reducing potassium residue to extremely low levels. The ion exchange unit is located after the acid washing unit and before the high-temperature calcination unit. This unit typically exists in the form of a resin column or fixed bed, treating activated carbon that has undergone acid washing and preliminary dehydration.

[0049] The high-temperature calcination unit serves as the final shaping and thorough purification unit. Under an inert atmosphere (such as N2), activated carbon is calcined at a medium temperature (380-420℃). Its functions are: first, to thermally decompose the most stable residual potassium compounds (such as K2O); second, to repair the pore structure of activated carbon that may have been damaged during acid washing, restoring and stabilizing its specific surface area and adsorption performance; and third, to thoroughly dry the product. Located at the very bottom of the entire process, the high-temperature calcination unit is the final outlet unit for the activated carbon material. It receives activated carbon from the ion exchange unit, and after processing, yields qualified low-potassium activated carbon product.

[0050] The intelligent anti-blockage conveying system includes material conveying components, which act as the "arteries" connecting various process units. They are responsible for the continuous, stable, and controllable physical transfer of activated carbon between the four units of washing, acid washing, ion exchange, and calcination, ensuring continuous material flow. It needs to overcome conveying challenges such as wet carbon adhesion, agglomeration, and dust dispersion, forming the foundation for continuous production. Its drive motors, valves, and other actuators are directly controlled by the central control system, adjusting the conveying speed and starting / stopping. Vibration and torque sensors integrated into key conveying equipment (such as screw conveyors) of the material conveying components upload real-time operating data (such as vibration frequency and current / torque values) to the central control system for monitoring blockage risks and achieving intelligent anti-blockage.

[0051] The soft sensor prediction system, acting as the "sensory system" of the entire unit, is responsible for real-time and continuous acquisition of key parameters reflecting process status and quality, and for real-time estimation of potassium content. Typical monitoring points include: water temperature and flow rate in the water washing unit; pH value, acid concentration (or conductivity), and temperature in the acid washing unit; potassium ion concentration (or pressure difference) before and after the column in the ion exchange unit; and temperature, atmospheric oxygen content, and pressure in each section of the high-temperature calcination unit. The soft sensor prediction system comprises a sensor network distributed throughout each process unit, which transmits all detection signals to the central control system in real-time and unidirectionally via cables or industrial buses, forming the system's "sensory nerves." The soft sensor prediction system is not merely a simple data collector; its data stream (especially the process data from acid washing and ion exchange) is the input source for the "soft sensor prediction model" in the central control system, used to infer the potassium content of the activated carbon in real time.

[0052] The central control system serves as the "brain" and "decision-making center" of the entire system. Specifically, its core functions include:

[0053] Data Aggregation and Processing: The system receives all real-time data from the soft measurement prediction system and the intelligent anti-blocking conveyor system. This includes: Simultaneously with physical processing, a parallel information sensing network is activated. Vibration sensors (mounted externally) and torque sensors (integrated within the drive motor) on the intelligent anti-blocking screw conveyor begin collecting data several times per second and uploading it to the central control system in real time via signal cables. Simultaneously, pH electrodes and conductivity meters installed in the pickling unit's circulation tank, as well as temperature sensors installed on the pipelines, continuously monitor the chemical state of the pickling solution; in the ion exchange unit, pressure sensors monitor the differential pressure of the resin bed. These process parameters are also transmitted to the central control system via data lines. Furthermore, micro-differential pressure sensors installed in the chambers of the washing unit, ion exchange unit, and high-temperature calcination unit continuously send pressure data from each unit to the dynamic pressure balance control system. This system, as a subsystem, then summarizes and reports the integrated pressure status information to the central control system. At this moment, the central control system's screen begins to scroll through real-time data streams from every corner of the production line, as if giving the device complete "senses."

[0054] Intelligent analysis, decision-making, and collaborative control scheduling: Executing core algorithms (such as soft measurement prediction models, blockage early warning models, and pressure balance models) makes judgments based on real-time data. Control commands are sent to the actuators of downstream units (such as acid metering pumps, heaters, nitrogen valves, conveyor motors, and pressure regulating valves) to dynamically adjust process parameters (pickling pH, calcination temperature curves, etc.), conveying rhythm, and system pressure. This includes: After receiving massive amounts of real-time data, the central control system's internally pre-built intelligent models begin synchronous operation. First, it invokes the "vibration-torque correlation analysis model" to determine the conveyor's operating status. Once the model algorithm identifies a characteristic pattern of abnormally high torque and decreased vibration, the system makes a decision within milliseconds and sends a sequence of commands to the inverter of the intelligent anti-blockage screw conveyor via control cables: such as "instantaneous frequency increase impact - brief reverse - recovery," thus mitigating blockage risks before the operator notices. Second, it simultaneously inputs data such as pH, conductivity, and differential pressure from the pickling and ion exchange units into the "potassium content soft measurement model based on machine learning." Within seconds, the model calculates the predicted potassium content of the current batch of activated carbon. If the predicted value is too high, the central control system immediately sends a command to the acid metering pump in the acid washing unit to fine-tune the acid addition; sends a command to the valve controller in the ion exchange unit to prepare for early switching or regeneration of the resin column; and sends a command to the temperature controller in the high-temperature calcination unit to appropriately adjust the calcination temperature curve. Finally, based on the data reported by the dynamic pressure balance system, the central control system runs a pressure balance algorithm to calculate the valve opening adjustment required to maintain the safe gradient of "calcination unit pressure > ion exchange unit pressure > front-end unit pressure," and precisely adjusts the opening of the inlet valves, exhaust valves, and atmosphere isolation valves associated with the ion exchange unit and the high-temperature calcination unit through control signals. This is a multi-threaded, concurrent intelligent decision-making and control process, where the three core technologies are simultaneously effective.

[0055] The formation of synergistic effects and the achievement of stable operation do not rely on the isolated execution of control commands across the three dimensions mentioned above. For example, when a soft-sensor model predicts fluctuations in the potassium content of the raw material, the central control system, while instructing downstream processes to adjust parameters, may simultaneously fine-tune the conveyor speed to change the material's residence time in the upstream section and adjust pressure settings to accommodate potentially altered process venting requirements. The execution status of all units (such as valve opening, motor speed, and actual temperature) is again fed back to the central control system as feedback signals, compared with setpoints, and forms a closed loop. Through this continuous "perception-decision-execution-feedback" cycle, the continuity of material flow, the real-time nature of quality information flow, and the stability of the atmospheric environment flow are unified and coordinated.

[0056] Human-computer interaction and monitoring: Provide operators with a visual interface to display the status of the entire process and record historical data.

[0057] The central control system is at the core of the continuous processing unit and communicates bidirectionally with all other modules. Downstream control: It sets and adjusts parameters for the four process units: water washing, acid washing, ion exchange, and high-temperature calcination; it directs the operation of the material conveying components; and it precisely controls the pressure regulating valve group integrated into the system. Upstream sensing: It continuously receives process quality data from the soft measurement and prediction system, as well as equipment status data from the material conveying components.

[0058] The continuous processing device described above in this invention is not a simple series of independent devices, but rather constitutes a tightly coupled intelligent system of "physical flow - information flow - control flow":

[0059] Physical flow (material flow direction): Activated carbon → Water washing → [Conveying] → Acid washing → [Conveying] → Ion exchange → [Conveying] → High-temperature calcination → Product. This is a unidirectional, continuous, linear process.

[0060] Information flow: Status information of each process unit and conveying equipment is aggregated to the central control system through soft measurement prediction system and built-in sensors.

[0061] Control flow: After processing information, the central control system issues control commands that are then distributed in reverse to each process unit and conveying component, forming a closed loop.

[0062] The central control system predicts quality based on real-time data (information flow) from the pickling unit, adjusts the process parameters (control flow) of the ion exchange and high-temperature calcination units in advance, instructs the material conveying components to adjust their rhythm, and directs the pressure balancing system to maintain a stable atmosphere. This achieves precise coordination of multiple units in time and space, ensuring efficient, stable, and safe continuous production. The central control system transforms the activated carbon continuous processing unit from a static collection of mechanical equipment into a dynamic, intelligent system capable of self-regulation, interference resistance, and continuous optimization. Under the unified scheduling of the central control system, it achieves and maintains a state of efficient, stable, and safe continuous production.

[0063] In one specific embodiment of the present invention, please refer to Figure 2 This demonstrates the core workflow and sensing principles of an intelligent anti-congestion conveyor system. The intelligent anti-congestion conveyor system, including the conveyor, is an engineering-level solution for the "physical flow interruption problem." Its specific implementation process includes, for example, the following steps:

[0064] Status perception stage: After the device is started, the conveyor begins to operate. A high-precision torque sensor installed on the output shaft of the drive motor measures and outputs an analog signal of the driving torque in real time. At the same time, a vibration sensor (usually a piezoelectric accelerometer) installed on the conveyor housing continuously collects the vibration spectrum signal of the housing. The data of the above two sensors are uploaded to the central control system in real time through the fieldbus (such as PROFIBUS DP) at a frequency of 10Hz.

[0065] Data modeling and warning stage: There is a "vibration-torque correlation analysis model" preset in the central control system. This model has learned the operating characteristics of the conveyor under unobstructed and simulated blocked states during the commissioning stage. Specifically, when the material in the conveyor flows smoothly, the torque is stable and the amplitude of the main vibration frequency (usually related to the blade rotation frequency) is stable; once material accumulation or "bridging" occurs, the driving torque will increase significantly and continuously, and due to the obstruction of the material to the effective transmission of energy, the amplitude of the main vibration frequency representing the material fluidity will decrease or disappear synchronously. The central control system calculates the average value of the torque (T_avg) and the average value of the amplitude of the main vibration frequency (A_avg) in the most recent time period (such as 10 seconds) in real time, and compares them with the preset threshold (such as ).

[0066] Intelligent decision-making and execution stage: Once it is detected that the average value is greater than the preset threshold (such as satisfying T_avg > T_threshold and A_avg < A_threshold), the system immediately determines it as a "precursor to blockage". The central control system triggers a predetermined dredging strategy within milliseconds and sends a sequence of commands to the frequency converter driver and the hot medium valve of the conveyor, including:

[0067] Instantaneous frequency increase impact: First, command the conveyor to accelerate to 130% of the rated speed within 0.5 seconds, and use the high-speed impact force to break the initially formed material arch.

[0068] Short-term reverse dredging: Immediately afterwards, command the motor to reverse 1 - 2 turns (about 0.5 - 1 second) to loosen the stuck material in the reverse direction.

[0069] Resume operation and cooperate with drying: Finally, resume normal forward conveying. At the same time, the central control system will联动调节the flow rate of the circulating hot medium (such as 80°C hot water) in the hollow spiral shaft of the conveyor, temporarily increase the flow rate by 20%, strengthen the drying of the wet carbon, reduce the viscosity from the source, and prevent blockage again.

[0070] After the dredging procedure is completed, the system continues to monitor the torque and vibration data. If the data returns to the normal range within 5 seconds, record a warning event and the process ends; if it does not return to normal, upgrade the alarm level and may execute a more aggressive dredging procedure or notify manual intervention. Note: There seems to be a Chinese term "联动调节" in ID=17 that might need further clarification for a more accurate translation. I've translated it as "联动调节" for now as it's not clear what the exact English equivalent should be without more context. You may adjust it according to the actual meaning.

[0071] This invention upgrades the traditional "downtime for maintenance after a failure" model to a predictive maintenance and proactive intervention model based on multi-sensor data fusion and real-time model analysis. The conveyor transforms from a passive actuator into an intelligent terminal with self-sensing, self-diagnostic, and self-adjusting capabilities. This completely solves the blockage problem in wet carbon conveying. In a 72-hour continuous operation test, the system successfully warned of and automatically resolved six blockage risks, increasing the effective production time ratio of the unit from approximately 85% in the traditional design to over 98.5%, providing a fundamental guarantee for the stability of continuous production.

[0072] In one specific embodiment of the present invention, please refer to Figure 3 This demonstrates the workflow of the soft measurement prediction system. The system's core innovation addresses the "information flow lag problem," and the specific implementation process is as follows:

[0073] Multi-source data acquisition: such as Figure 3 As shown on the left, the system synchronously collects easily measurable "auxiliary variables" through sensor arrays deployed in the pickling and ion exchange units. Key variables include: the real-time pH value, conductivity value (reflecting the total ion concentration), and temperature of the pickling solution; and the inlet and outlet pressure difference of the ion exchange resin column (ΔP, reflecting the resin bed resistance and adsorption saturation state). These data are collected and transmitted to the central control system at a frequency of 1 Hz.

[0074] Feature Engineering and Model Input: The central control system preprocesses the raw data (e.g., filtering, normalization) and performs feature engineering. It not only uses instantaneous values ​​but also generates derived features, such as "the rate of pH decrease over the past 5 minutes" and "the correlation trend between conductivity and temperature." These features collectively constitute a multidimensional feature vector X(t) characterizing the current process state.

[0075] Real-time prediction by machine learning model: A multidimensional feature vector X(t) is fed into a soft-sensor prediction model running within the central control system. This model could be, for example, a gradient boosting tree (GBRT) regression model trained offline on a large amount of historical data. During model training, several months of production data were used, with the same sensor data as the input variables and the output label (target variable) being the true value of activated carbon potassium content accurately measured in an offline laboratory using ICP-OES (inductively coupled plasma optical emission spectrometry) during the same period. The model has learned a complex nonlinear mapping from the multidimensional feature vector X(t) to the potassium content Y. In production, the model outputs a real-time potassium content prediction value Y_pred(t) in ppm, updated at a frequency of 1 Hz.

[0076] Predictive-driven closed-loop control: such as Figure 3As shown on the right, the predicted value Y_pred(t) is used in real time for production control. The central control system sets control rules: for example, when Y_pred(t) exceeds 150 ppm (the set warning threshold) for 10 consecutive seconds, the system automatically performs the following coordinated dynamic adjustments:

[0077] Adjust pickling parameters: Send a command to the acid metering pump of the pickling unit to fine-tune the pH setting from 2.2 to 2.0 to increase the pickling intensity.

[0078] Optimize ion exchange cycle: Send instructions to the controller of the ion exchange unit to switch or start regenerating the current resin column 15% in advance.

[0079] Pre-set calcination temperature: Send a command to the temperature controller of the high-temperature calcination unit to temporarily raise the calcination temperature setpoint from 400℃ to 410℃ and maintain it for 20 minutes.

[0080] This invention creatively introduces machine learning soft measurement technology into the field of activated carbon depotassium removal, constructing a "virtual online potassium content analyzer" that reduces the feedback time of key quality parameters from several hours (laboratory analysis) to seconds. This achieves true real-time quality closed-loop control. The system can detect raw material fluctuations in advance and adaptively adjust, reducing the batch-to-batch fluctuation (standard deviation) of the final product potassium content (verified offline) by more than 35%, while avoiding waste of chemicals and energy due to "over-processing" or "under-processing," thus improving resource utilization efficiency.

[0081] In one specific embodiment of the present invention, please refer to Figure 4 It demonstrates a dynamic pressure balance control system (corresponding to) Figure 1The "Dynamic Pressure Balance System" consists of three parts: a micro-differential pressure sensor group (installed in each unit cavity and connecting duct), a pressure regulating and atmosphere isolation valve group (installed on the connecting pipes between units), and a pressure control algorithm in the central control system. The system establishes a safe pressure gradient: high-temperature calcination unit > ion exchange unit > acid washing unit ≈ water washing unit. Each micro-differential pressure sensor continuously monitors the pressure. For example, when the pressure in the calcination unit drops slightly due to temperature fluctuations, causing the pressure difference between it and the ion exchange unit to fall below the set value, the central control system immediately calculates and issues a command to slightly close the exhaust valve of the calcination unit or slightly open its inlet valve (N2), while simultaneously coordinating the adjustment of the outlet valve of the ion exchange unit to restore the pressure difference to the set range within seconds. Similarly, within the sealing cover of the chain conveyor between the ion exchange unit and the acid washing unit, a small amount of N2 is injected and exhaust is controlled to maintain a stable micro-positive pressure, ensuring that external air cannot infiltrate. This system acts like the device's "autonomic nervous system," constantly maintaining the stability of the internal environment. It prevents both the risk of explosion due to air backflow and the unnecessary leakage of valuable inert gases. This system is a safeguard against "energy flow and atmosphere imbalance," and its specific implementation process is as follows:

[0082] Construction of pressure monitoring network: such as Figure 4 As shown, a group of micro-differential pressure sensors is deployed at key nodes of the system to continuously monitor the pressure at each point. Specifically, this includes: P1 (monitoring the pressure inside the water washing unit), P2 (monitoring the pressure inside the acid washing unit), P3 (the pressure at the top of the ion exchange unit), P4 (the pressure at the feed end of the high-temperature calcination unit), and the ΔP34 sensor, which directly measures the pressure difference between the calcination and ion exchange units. All sensors have an accuracy of ±0.5 Pa.

[0083] Safety pressure gradient setting and algorithm operation: A key safety pressure gradient is set and maintained within the central control system: high-temperature calcination unit > ion exchange unit > acid washing unit ≈ water washing unit, i.e., P4 > P3 > P2 ≈ P1. In an optional implementation, this is specifically quantified as: P4 - P3 = +25 Pa (ensuring inert gas flow is forward), P3 - P2 = +10 Pa. The pressure balance control algorithm within the system (such as a proportional-integral controller) operates at a frequency of 10 Hz, continuously calculating the deviation e(t) between the actual pressure difference and the set value.

[0084] Dynamic regulation and valve coordinated control: such as Figure 4 As shown, the pressure regulating and atmosphere isolation valve groups are distributed at the unit connections and on the gas pipelines. When the system detects a deviation (e.g., P4 drops instantaneously due to the action of the calciner feed valve, e34 = (P4-P3) - 25 < 0), the central control system immediately calculates and issues a synchronization adjustment command:

[0085] The N2 intake regulating valve (V4_in) of the high-temperature calcination unit is instructed to increase its opening by 3%, while its exhaust valve (V4_out) is instructed to decrease its opening by 2% to quickly boost P4.

[0086] The opening of the outlet pressure regulating valve (V3_out) of the instruction ion exchange unit is reduced by 1% synchronously to maintain the relative stability of P3.

[0087] This series of coordinated actions can restore the pressure difference P4-P3 to the set value ±2 Pa within 2-3 seconds.

[0088] Atmosphere isolation and safety assurance: Inside the sealed cover of the chain conveyor between the ion exchange unit and the pickling unit, the internal pressure P_chute is always maintained 5-8 Pa higher than P2 through independent micro-flow N2 injection and pressure relief valve control, forming a stable air curtain that effectively isolates the different atmospheres of the upstream and downstream sections.

[0089] The embodiments of this invention establish an active, dynamic, system-wide pressure and atmosphere balance control scheme, replacing the traditional passive approach that relies on mechanical seals and static pressure settings. Through real-time feedback and precise coordinated valve control, it treats the system pressure as a dynamically stable whole, fundamentally eliminating dangerous gas cross-mixing. During long-term operation, the oxygen concentration at the inlet of the high-temperature calcination unit is stably controlled below 5 ppm, far below the safe threshold of 10 ppm, completely eliminating the risk of combustion. Simultaneously, by precisely controlling the N2 flow direction and pressure, the overall nitrogen consumption of the system is reduced by approximately 15% compared to systems without this dynamic balance, achieving the dual goals of safety and energy saving.

[0090] Example 2

[0091] This embodiment provides another preferred implementation scheme based on Embodiment 1. For the conveying of wet carbon from the pickling unit to the ion exchange unit, an intelligent vibrating feeder can be used instead of a screw conveyor. The vibrating motor of this feeder also integrates torque sensing, and a non-contact microwave moisture meter is installed below its discharge port. The central control system not only monitors torque but also acquires real-time surface moisture data of the material. When the system predicts a risk of blockage (increased torque) and detects excessively high material moisture, in addition to executing the vibration frequency adjustment program, it can also temporarily activate the infrared auxiliary drying lamp installed upstream to locally and rapidly dry the material in the outlet area, improving flowability from the source. This multi-sensor fusion anti-blockage strategy is suitable for material conveying processes that are more sensitive to temperature.

[0092] Example 3

[0093] This embodiment describes the specific application of Embodiments 1 and 2 of the present invention in the production of low-potassium activated carbon for supercapacitors. The raw material is alkaline activated carbon with a potassium residue of approximately 3000 ppm. After the device is started, the central control system operates in fully automatic mode.

[0094] In the initial batch, the soft sensing model predicted a high potassium content in the raw materials based on the initial pickling data. The system automatically lowered the pH control setpoint of pickling unit 2 by 0.2 and shortened the switching cycle of the first resin column of the ion exchange unit by 15%.

[0095] During operation, the intelligent conveying system issued multiple warnings and automatically eliminated the potential risk of wet carbon bridging, without requiring a single shutdown for clearing blockages.

[0096] The dynamic pressure balance system stabilizes the nitrogen consumption of the calcination unit at the set value, and the oxygen analyzer shows that the oxygen content at the calcination furnace inlet is always below the safety standard of 10 ppm.

[0097] After 72 hours of continuous operation, sampling and testing of the final product showed that potassium residue remained stable between 8-18 ppm (average 12 ppm), exceeding the design requirement of ≤20 ppm, with minimal batch-to-batch variation. The effective production time of the device reached 98.5%, far exceeding the 85% expected by traditional designs, fully demonstrating the comprehensive advantages of this invention in improving quality, efficiency, and stability.

[0098] In the description of this specification, the references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0099] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A multi-stage synergistic potassium removal continuous activated carbon treatment device, comprising a water washing unit, an acid washing unit, an ion exchange unit, and a high-temperature calcination unit connected in sequence, characterized in that, The device further includes: The central control system, and three subsystems that are communicatively connected to and coordinated by the central control system: The intelligent anti-blocking material conveying subsystem includes a material conveying component, as well as a vibration sensor and a drive shaft torque sensor integrated thereon, for real-time collection of conveying status data and uploading it to the central control system; The potassium content soft measurement and prediction subsystem includes sensor arrays deployed in the pickling unit and ion exchange unit to collect process parameters including the pH value, conductivity, temperature and pressure difference of the pickling solution and input them to the central control system. The multi-unit gradient pressure dynamic balance subsystem includes micro differential pressure sensors installed in each process unit, as well as pressure regulating valve groups and atmosphere isolation valves set in the inter-unit connection channels. The central control system integrates: The intelligent anti-blockage model is used to identify blockage risks and generate anti-blockage intervention commands based on data from the vibration sensor and torque sensor. A soft measurement prediction model is used to predict the potassium content of activated carbon in real time based on the process parameters. The soft measurement prediction model is a machine learning model trained based on historical data. The input is the process parameters of the acid washing and ion exchange unit, and the output is the predicted value of potassium content of activated carbon. The pressure balance control algorithm is used to dynamically adjust the opening of the pressure regulating valve group and the atmosphere isolation valve based on the monitoring data of the micro differential pressure sensor, so as to maintain a safe pressure gradient of "calcination unit pressure > ion exchange unit pressure > front-end unit pressure"; the front-end unit pressure includes pickling unit pressure and water washing unit pressure. The central control system is configured to receive data from three subsystems, run the model and algorithm, and collaboratively generate and issue control commands based on the predicted potassium content to adjust process parameters, material conveying rhythm and system pressure, thereby realizing closed-loop collaborative control of material flow, quality information flow and system atmosphere flow.

2. The apparatus according to claim 1, characterized in that, The intelligent anti-blockage model is based on the correlation analysis between vibration frequency and driving torque to identify blockage precursors and trigger intervention procedures, which include instantaneous acceleration, brief reversal, or adjustment of heat medium delivery.

3. The apparatus according to claim 1, characterized in that, The material conveying assembly includes a screw conveyor or a vibrating feeder, and its drive device and sensor data communicate with the central control system via an industrial bus.

4. The apparatus according to claim 1, characterized in that, The device also includes an infrared drying device or a hollow spiral shaft hot medium circulation system, which is linked with the intelligent anti-clogging material conveying subsystem to improve material flowability based on the risk of clogging.

5. A method for controlling the continuous treatment of activated carbon for multi-stage synergistic potassium removal, characterized in that, The method, when applied to the apparatus as described in any one of claims 1-4, includes the following steps: The intelligent anti-blocking material conveying subsystem collects vibration and torque data in real time, uses the intelligent anti-blocking model to determine the risk of blockage and automatically executes anti-blockage intervention. The potassium content soft measurement and prediction subsystem collects process parameters in real time and outputs potassium content prediction values ​​in real time using the soft measurement and prediction model. The multi-unit gradient pressure dynamic balance subsystem monitors the pressure of each unit in real time and uses the pressure balance control algorithm to dynamically adjust the valve opening to maintain the system's safe pressure gradient. Based on the predicted potassium content, the central control system coordinates the acid washing pH, ion exchange cycle, and calcination temperature, and also adjusts the material conveying rhythm and system pressure setting.

6. The method according to claim 5, characterized in that, The intelligent anti-blockage intervention includes a combination of instantaneous frequency boosting, brief reversal, and thermal medium flow regulation.

7. An electronic device comprising a processor, a memory, and a computer program stored in the memory, characterized in that, When the processor executes the program, it implements the method as described in claim 5 or 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in claim 5 or 6.

Images