An automatic water adding and dispensing device with accurate identification

Abstract

This invention discloses an automatic water and drug dispensing device with precise identification capabilities, belonging to the field of drug dispensing device technology. It includes a cabinet, a filter press tank, a tank cover, a pressing assembly, a heating plate, a water replenishment assembly, a liquid mixing assembly, a temperature sensor, and a control system integrated into the control panel. The control system includes data acquisition, edge computing processing, communication interaction, and execution control units. The water replenishment assembly is equipped with a liquid level sensor, and the filter press tank and tank cover have a sealing structure. The edge computing processing unit integrates a self-identification module for operating conditions and an adaptive tuning module for algorithm parameters. It can also be optimized using model predictive control algorithms combined with multi-source data fusion technology. The device eliminates data errors through synchronous acquisition and preprocessing of multi-source data, and achieves precise water replenishment through dynamic evaporation prediction and closed-loop control. This invention solves the problems of insufficient water addition accuracy, cumbersome water replenishment, and poor adaptability to operating conditions in existing devices, achieving precise, automated, and sealed drug dispensing and filtration.

Description

Technical Field

[0001] This invention relates to the field of medicine dispensing equipment technology, and in particular to an automatic water dispensing device that can accurately identify the contents. Background Technology

[0002] In the fields of traditional Chinese medicine dispensing, formulation production, and pharmaceutical filtration, automatic water-adding dispensing devices are core supporting equipment for ensuring dispensing accuracy and standardizing production processes. Their performance directly affects the quality of the finished pharmaceutical solution, production efficiency, and production compliance. With the continuous improvement of standardization and automation requirements in the traditional Chinese medicine pharmaceutical industry, the industry has placed higher demands on the water addition control accuracy, closed-loop filtration performance, and adaptability of dispensing devices.

[0003] Existing water-dosing and dispensing devices largely rely on manual judgment of water volume using graduation marks. The accuracy of water dosing depends entirely on the operator's experience, making them highly susceptible to level deviations. This can lead to excessive drug concentrations and compromise the stability of dosing quality. Furthermore, conventional devices use only a single fixed threshold filter for level acquisition, failing to optimize data by incorporating the correlation between temperature, pressure, and level during the filtration process. This fails to eliminate false level errors caused by boiling disturbances and steam condensation, further exacerbating the accuracy deviation in water dosing control.

[0004] The existing drug preparation equipment requires opening the lid to complete the operation when adding water during the filter press. This not only makes the operation cumbersome and the production efficiency low, but also causes a large amount of steam leakage in the filter press, resulting in unnecessary energy consumption. At the same time, external dust and impurities can easily enter the pot when the lid is opened, contaminating the drug solution and damaging the purity and hygiene standards of the drug solution. This makes it difficult to meet the actual production needs of closed filter press and aseptic drug preparation in the production of traditional Chinese medicine.

[0005] In addition, existing automatic water replenishment schemes mostly adopt a fixed threshold switching control mode and a fixed parameter control algorithm, which cannot adapt to the changes in working conditions at different stages of the entire filter press process, such as preheating, micro-boiling, boiling, and constant temperature filter press. They are also difficult to adapt to the filter press characteristics of different raw materials and different formulations, and are prone to problems such as water replenishment overshoot, large fluctuations in liquid level, and control response lag. Summary of the Invention

[0006] The purpose of this invention is to address the shortcomings of existing technologies by proposing an automatic water and medicine dispensing device that can accurately identify the problem.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: an automatic water and chemical dosing device with precise identification capability, comprising a cabinet, a slot on the top surface of the cabinet, a filter press tank inserted into the slot, a tank cover for sealing the filter press tank hinged to one side of the top surface of the cabinet, a safety valve on the top surface of the tank cover, a pressure gauge on one side of the safety valve, a locking handle installed on the outer wall of the end of the tank cover away from the hinge, a corresponding threaded hole seat fixed to the upper end of the outer wall of the filter press tank, a pressing component installed on the top surface of the tank cover, a heating plate abutting the bottom end of the filter press tank on the inner bottom surface of the cabinet, and a [missing information - likely a design feature] on one end of the top surface of the cabinet. The unit includes a control panel on a vertical plate, a water supply component for replenishing water to the filter press tank mounted on the upper end of one end face, a liquid mixing component extending from the outer wall of the filter press tank, and a temperature sensor mounted on a clamp at the lower end of the outer wall of the filter press tank. The top surface of the tank lid has a reagent filling port with a sealing cap for quantitative dispensing of pharmaceutical raw materials. Sealing grooves are formed on the opposing surfaces of the filter press tank and the tank lid, and a sealing ring is fixedly connected to the sealing groove of the tank lid, with the bottom shape of the sealing ring corresponding to the sealing groove of the filter press tank. The water supply component is connected to the filter press tank in a sealed manner via a flange connection structure.

[0008] The water replenishment component includes a first flange pipe fixed to the outer wall of the cabinet. One end of the first flange pipe is connected to a solenoid valve pipe. The upper end of the outer wall of the filter press tank is provided with a water replenishment pipe connected to the other end of the first flange pipe. One end of the water replenishment pipe is connected to the inside of the filter press tank. A liquid level sensor is provided on one side of the top surface of the tank cover. The detection end of the liquid level sensor extends into the inside of the filter press tank.

[0009] The control panel integrates a control system, which includes a data acquisition unit, an edge computing processing unit, a communication interaction unit, and an execution control unit. The signal output terminal of the data acquisition unit is communicatively connected to the signal input terminal of the edge computing processing unit. The edge computing processing unit establishes a bidirectional communication link with the execution control unit through the communication interaction unit. The data acquisition unit is used to synchronously acquire liquid level, temperature, and pressure data in the filter press tank, perform preprocessing, and then output the data to the edge computing processing unit. The edge computing processing unit is used to predict liquid level evaporation loss, calculate water replenishment control amount, and generate water replenishment control commands based on the acquired data. The execution control unit is used to control the operating status of the water replenishment component and the heating plate according to the water replenishment control commands.

[0010] Preferably, the extrusion assembly includes a guide sleeve that is fixedly connected to the center of the top surface of the can lid. A transmission box is fixedly connected to the top of the guide sleeve. A transmission pulley is provided inside the transmission box, and the transmission pulley includes a driving pulley and a driven pulley. A handwheel is rotatably provided on one side of the top surface of the transmission box via a bearing. The shaft of the handwheel is coaxially fixedly connected to the driven pulley. A drive motor is installed on one side of the bottom surface of the transmission box. The output shaft of the drive motor is coaxially fixedly connected to the driving pulley via an electromagnetic clutch. A screw is rotatably provided inside the guide sleeve. The top of the screw is coaxially fixedly connected to the shaft of the handwheel. An internally threaded cylinder is screwed onto the screw. An extrusion disc located inside the filter press can is fixedly connected to the bottom end of the internally threaded cylinder. A guide groove is opened on the outer wall of the internally threaded cylinder. A guide strip corresponding to the guide groove is fixedly connected to the inner wall of the guide sleeve. Multiple through holes are evenly opened on the extrusion disc.

[0011] Preferably, the liquid mixing assembly includes two symmetrically arranged connecting pipes fixed to the lower end of one side of the outer wall of the filter press tank. One end of each connecting pipe is connected to the middle of one side of the mixing tank. A first motor is installed in the middle of the top surface of the mixing tank. The output shaft of the first motor is coaxially fixed to a stirring paddle rotatably disposed inside the mixing tank. A liquid outlet pipe is connected to the lower end of one side of the mixing tank. One end of the liquid outlet pipe extends out of the outer wall of the cabinet. A valve is provided on the pipe outside the cabinet. The liquid outlet pipe is a Y-shaped pipe, and a first solenoid valve is installed at the bifurcation point. A second solenoid valve is provided on the connecting pipe.

[0012] Preferably, the data acquisition unit is equipped with a multi-channel signal acquisition interface and an analog-to-digital conversion module, which are respectively connected to the liquid level sensor, temperature sensor and pressure gauge signal, for synchronously acquiring liquid level, temperature and pressure data in the filter press tank, and outputting the acquired data to the edge computing processing unit after preprocessing.

[0013] Preferably, the edge computing processing unit receives data output from the data acquisition unit, extracts data features and establishes a correlation model, predicts liquid level evaporation loss, calculates water replenishment control amount based on the set target liquid level, and generates water replenishment control command.

[0014] Preferably, the communication interaction unit establishes a bidirectional redundant communication link, which is connected to the edge computing processing unit, the control panel, and the execution control unit respectively, to realize local data transmission and remote data interaction.

[0015] Preferably, the execution control unit is connected to the controlled end of the solenoid valve tube, the heating plate, and the drive motor respectively, and is used to adjust the on / off state and duration of the solenoid valve tube according to the water replenishment control command, and to adjust the output power of the heating plate in linkage, and has built-in safety interlock control logic.

[0016] Preferably, the edge computing processing unit integrates a working condition self-identification module and an algorithm parameter adaptive tuning module. The working condition self-identification module identifies the filter press working condition based on the collected temperature, pressure, and liquid level data, and the algorithm parameter adaptive tuning module dynamically adjusts the operating parameters of the control algorithm and prediction model according to the identified filter press working condition.

[0017] The working condition self-identification module is based on pre-processed multi-dimensional data of temperature, pressure and liquid level to build a filter press working condition feature library. Through feature matching, it realizes real-time intelligent identification of the preheating stage, micro-boiling stage, violent boiling stage and constant temperature filter press stage.

[0018] The algorithm parameter adaptive tuning module adjusts the proportional coefficient of the incremental closed-loop control algorithm based on the real-time operating condition results output by the operating condition self-identification module. Integral coefficient Differential coefficients Dynamic tuning is performed, and the correction coefficient K of the dynamic evaporation prediction model is corrected in real time to ensure that the control algorithm and prediction model are always adapted to the current filter press conditions.

[0019] Compared with the prior art, the beneficial effects of the present invention are:

[0020] 1. This solution uses a data acquisition unit to simultaneously collect and preprocess liquid level, temperature, and pressure data, eliminating time-series deviations and false data. Combined with the dynamic evaporation prediction and closed-loop control of the edge computing processing unit, it accurately controls the water addition, achieving dynamic sealed water replenishment during the pressure filtration process and stably controlling the concentration of the chemical solution without manual intervention.

[0021] 2. This solution connects directly to the filter press tank via a sealed water replenishment component, and works with the control system to achieve automatic water replenishment without opening the lid throughout the process. This significantly reduces steam leakage and energy consumption, while preventing external contaminants from entering the tank, ensuring the purity and hygiene standards of the chemical solution, simplifying the water replenishment process, and improving production efficiency.

[0022] 3. This solution achieves intelligent identification of all stages of the filter press process through the working condition self-identification module. Combined with the algorithm parameter adaptive tuning module, it dynamically adjusts the operating parameters to adapt to different filter press working conditions and raw material characteristics, suppresses liquid level fluctuations, avoids water replenishment overshoot and control lag, and further improves the accuracy of water addition control and the adaptability of working conditions.

[0023] In summary, this solution, through the coordinated operation of hardware actuators and intelligent control systems, systematically solves the core pain points of existing drug dispensing devices, such as insufficient water addition accuracy, cumbersome water replenishment operations, and poor adaptability to operating conditions. It achieves precise, automated, and closed-loop operation of the drug dispensing and filtration process, stably ensuring the quality and hygiene standards of the drug solution, improving production efficiency and operational safety, and adapting to the standardized production needs of multiple scenarios such as drug dispensing and formulation production. Attached Figure Description

[0024] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0025] Figure 1 This is a first-view schematic diagram of the overall structure proposed in this invention;

[0026] Figure 2 This is a second-view schematic diagram of the overall structure proposed in this invention;

[0027] Figure 3 This is a schematic diagram of the overall cross-sectional structure of the cabinet proposed in this invention;

[0028] Figure 4 This is a schematic diagram of the overall structure of the transmission pulley proposed in this invention;

[0029] Figure 5 This is a schematic diagram of the overall structure for removing the can lid proposed in this invention;

[0030] Figure 6 This is a partial cross-sectional view of the mixing tank proposed in this invention;

[0031] Figure 7 This is a block diagram of the overall architecture of the control system proposed in this invention.

[0032] The numbers in the diagram are: 1. Cabinet; 2. Filter press tank; 3. Tank cover; 4. Solenoid valve pipe; 5. Liquid level sensor; 6. Guide sleeve; 7. Drive pulley; 8. Squeezing disc; 9. Connecting pipe; 10. Discharge pipe; 11. Sealing ring; 12. Handwheel; 13. Mixing tank. Detailed Implementation

[0033] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0034] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of the components in a specific posture. If the specific posture changes, the directional indication will also change accordingly. Furthermore, descriptions involving "first," "second," etc., in the present invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. In the description of the present invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly defined.

[0035] In this invention, unless otherwise explicitly specified and limited, the terms "connection," "fixing," etc., should be interpreted broadly. Those skilled in the art can understand the specific meaning of these terms in this invention based on the specific circumstances. Furthermore, the technical solutions of the various embodiments of this invention can be combined with each other, but only on the basis of being achievable by those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0036] See Figures 1-7 The automatic water and medicine dispensing device with accurate identification provided in this embodiment includes a hardware actuator and a control system adapted to the hardware actuator to achieve the function of accurate identification and automatic water dispensing. The hardware actuator includes a cabinet 1, a filter press tank 2, a tank cover 3, a squeezing assembly, a heating plate, a control panel, a water replenishment assembly, a liquid mixing assembly, and a temperature sensor. The specific implementation methods of each component are as follows:

[0037] The cabinet 1 is made of cold-rolled steel plate and bent into one piece, providing a stable installation base for each component and isolating external dust and impurities. The top surface of the cabinet 1 has a slot that matches the shape of the filter press tank 2. The filter press tank 2 is made of food-grade 304 stainless steel and is stamped into one piece, and is inserted into the slot. A heating plate is fixedly installed on the bottom surface of the cabinet 1. The top surface of the heating plate is in close contact with the bottom surface of the filter press tank 2 for heating the material inside the filter press tank 2.

[0038] A canister cover 3 for sealing the filter press tank 2 is hinged to one side of the top surface of the cabinet 1. A safety valve and a pressure gauge are fixedly installed on the top surface of the canister cover 3. The safety valve is used for pressure relief protection when the filter press tank 2 is overpressurized, and the pressure gauge is used to collect the pressure data inside the filter press tank 2 in real time. A locking handle is installed on the outer wall of the end of the canister cover 3 away from the hinge. An internal threaded hole seat that matches the locking handle is welded to the upper end of the outer wall of the filter press tank 2. The canister cover 3 and the filter press tank 2 are locked and fixed by the threaded engagement between the locking handle and the internal threaded hole seat.

[0039] Both the filter press 2 and the lid 3 have annular sealing grooves on their opposite surfaces. A sealing ring 11 is interference-fitted into the sealing groove of the lid 3. The sealing ring 11 is made of heat-resistant and oil-resistant food-grade silicone. The flange structure at the bottom of the sealing ring 11 fits into the sealing groove of the filter press 2 to achieve sealing protection during the filter pressing process.

[0040] The cabinet, filter press, tank cover, safety valve, pressure gauge, locking structure and sealing structure in this section are all standard supporting structures for filter press equipment. Their specific installation methods and selection and adaptation are existing conventional technologies, and will not be elaborated here.

[0041] Specifically, the water replenishment assembly includes a first flange pipe welded and fixed to the upper end of the outer wall of the cabinet 1, with a solenoid valve pipe 4 connected to the outer flange of the first flange pipe. The water inlet end of the solenoid valve pipe 4 is used to connect with an external water supply pipe. A water replenishment pipe is welded and fixed to the upper end of the outer wall of the filter press tank 2. The outer end of the water replenishment pipe is connected to the inner flange of the first flange pipe, and the inner end of the water replenishment pipe is connected to the inside of the filter press tank 2. A liquid level sensor 5 is fixedly installed on one side of the top surface of the tank cover 3. The detection end of the liquid level sensor 5 extends into the inside of the filter press tank 2 to collect the liquid level data in the filter press tank 2 in real time.

[0042] The flange connection structure and water supply pipeline structure in this part are conventional pipeline structures in the field of fluid transportation. The solenoid valve pipe 4 and the liquid level sensor 5 are conventional execution and detection devices in the field of industrial measurement and control. Their basic selection and installation methods are existing conventional technologies, and will not be described in detail here.

[0043] Specifically, the extrusion assembly includes a guide sleeve 6 that is fixedly connected to the center of the top surface of the can lid 3. A transmission box is fixedly connected to the top of the guide sleeve 6, and a transmission pulley 7 is provided inside the transmission box. The transmission pulley 7 includes a driving pulley and a driven pulley. A handwheel 12 is rotatably mounted on one side of the top surface of the transmission box via a bearing. The shaft of the handwheel 12 is coaxially fixedly connected to the driven pulley. A drive motor is installed on one side of the bottom surface of the transmission box. The output shaft of the drive motor is coaxially fixedly connected to the driving pulley via an electromagnetic clutch. A screw is rotatably mounted inside the guide sleeve 6, and the top of the screw is coaxially fixedly connected to the shaft of the handwheel 12. The screw is screwed with an internally threaded cylinder, and the bottom end of the internally threaded cylinder is fixed to an extrusion disc 8 located inside the filter press. The outer wall of the internally threaded cylinder has a guide groove, and the inner wall of the guide sleeve 6 is fixed with a guide strip corresponding to the guide groove. Multiple through holes are evenly opened on the extrusion disc 8. The outer wall of the internally threaded cylinder has a guide groove along the axial direction, and the inner wall of the guide sleeve 6 is welded and fixed with a guide strip that slides and adapts to the guide groove. Through the cooperation of the guide strip and the guide groove, the circumferential rotation of the internally threaded cylinder is restricted, so that the internally threaded cylinder drives the extrusion disc 8 to rise and fall vertically along the axial direction, thereby realizing the extrusion and solid-liquid separation of the medicine residue.

[0044] The working mechanism of the electromagnetic clutch is as follows:

[0045] In electric drive mode, the electromagnetic clutch is energized and closed, and the output shaft of the drive motor is connected to the drive pulley. The drive motor drives the driven pulley, handwheel shaft and screw to rotate through the transmission pulley 7, realizing the electric lifting and lowering of the extrusion disc 8. In manual drive mode, the electromagnetic clutch is de-energized and disconnected, and the output shaft of the drive motor is disengaged from the drive pulley. Turning the handwheel can drive the screw to rotate, realizing the manual lifting and lowering of the extrusion disc 8. During manual operation, the drive motor will not rotate synchronously with the screw, ensuring the safety and convenience of operation.

[0046] The pulley drive structure, screw lifting structure, and extrusion disc structure in this section are all conventional extrusion execution structures used in solid-liquid separation equipment. Their transmission principles and mechanical coordination methods are existing conventional technologies and will not be elaborated here.

[0047] The liquid mixing assembly includes two symmetrically arranged connecting pipes 9 fixed to the lower end of one side of the outer wall of the filter press 2. One end of the connecting pipe 9 is connected to the middle of one side of the mixing tank 13, which is fixed to the outer wall. A first motor is installed in the middle of the top surface of the mixing tank 13. The output shaft of the first motor is coaxially fixed to a stirring paddle that is rotatably installed inside the mixing tank 13. A liquid outlet pipe 10 is connected to the lower end of one side of the mixing tank 13. One end of the liquid outlet pipe 10 extends out of the outer wall of the cabinet 1. A valve is provided on the pipe of the liquid outlet pipe 10 outside the cabinet 1. The liquid outlet pipe 10 adopts a Y-shaped pipe and a first solenoid valve is installed at the bifurcation. A second solenoid valve is provided on the connecting pipe 9. The stirring paddle adopts an anchor-type stirrer. The mixing tank 13, the connecting pipe 9 and the liquid outlet pipe 10 are all made of 304 stainless steel, which is rust-proof and corrosion-resistant, and can avoid contamination of the liquid.

[0048] A temperature sensor is fixedly installed on the lower end of the outer wall of the filter press 2 by a clamp. The detection surface of the temperature sensor is in close contact with the outer wall of the filter press 2, and the contact surface is coated with thermally conductive silicone grease for real-time acquisition of the temperature data of the liquid in the filter press 2.

[0049] The liquid outlet pipeline structure, valves, and temperature sensors in this part are all standard supporting structures and devices for pharmaceutical hydraulic filtration equipment. Their installation methods and signal acquisition principles are existing conventional technologies and will not be described in detail here.

[0050] The top surface of the can lid 3 is equipped with a medicine filling port with a sealing cap for quantitative dispensing of raw materials. The medicine filling port on the top surface of the can lid is equipped with a metering dispensing component, which enables quantitative dispensing of raw materials and ensures the accuracy of the dispensing ratio. The dispensing function of this invention covers the entire process of traditional Chinese medicine dispensing and production: First, quantitative dispensing of raw materials, achieving precise dispensing of medicinal materials through the metering dispensing component; second, precise quantitative dispensing of solvent, achieving closed-loop precise control of water addition through a precise identification automatic water replenishment system, accurately regulating the solvent ratio of the medicine solution, and stably controlling the concentration of the medicine solution (a core indicator of traditional Chinese medicine dispensing); third, homogeneous mixing of the medicine solution, achieving uniform mixing of the medicine solution through the stirring structure of the dispensing mixing component, ensuring the uniformity of the concentration of the finished medicine solution; fourth, pressure filtration and separation of the medicine solution, achieving solid-liquid separation of the medicine residue and the medicine solution through the extrusion component, obtaining a finished medicine solution that meets the mixing ratio requirements.

[0051] In this embodiment, the control system, which adapts to the water replenishment component to achieve accurate identification and automatic water replenishment, is designed to realize the core functions of this invention. It includes a data acquisition unit, an edge computing processing unit, a communication interaction unit, and an execution control unit. These units are integrated on the main control circuit board of the control panel, enabling accurate identification of the liquid level during the filtration process, multi-dimensional correlation analysis, closed-loop control, and automatic sealing and water replenishment. This solves the problems of insufficient accuracy in manual water replenishment, steam leakage due to mid-process opening for water replenishment, and chemical contamination in existing technologies. The specific implementation methods of each unit are detailed below:

[0052] The data acquisition unit is equipped with a multi-channel analog signal acquisition interface and an analog-to-digital conversion module. Each signal acquisition interface is electrically connected to the signal output terminals of the liquid level sensor 5, the temperature sensor, and the pressure gauge, respectively. During the operation of the device, the data acquisition unit synchronously acquires real-time liquid level data, real-time temperature data, and real-time pressure data inside the filter press 2 at a preset acquisition cycle. The unit also performs preprocessing operations such as timestamp synchronization, sliding window filtering, and outlier removal on the acquired raw data in sequence to provide reliable and effective data for subsequent calculations.

[0053] (1) Timestamp synchronization: For the liquid level data, temperature data and pressure data acquired in the same acquisition cycle, a unified timestamp based on the same clock reference is assigned to eliminate the timing deviation caused by hardware transmission delay during multi-channel acquisition, and ensure that the three sets of data are in the same time dimension during subsequent correlation analysis, so as to avoid analysis errors caused by timing misalignment.

[0054] (2) Sliding window filtering: To address the issue of spike noise in the raw data collected by the liquid level sensor 5 due to liquid level fluctuations caused by boiling during the pressure filtration process, this embodiment performs a sliding average filtering operation on the continuously collected liquid level data sequence to filter out spike noise and obtain stable and effective liquid level data. The corresponding calculation formula is as follows: In the formula The effective liquid level data after filtering in the nth acquisition cycle; This represents the raw liquid level sampling data for the nith acquisition cycle; N is the length of the sliding window, which is adaptively adjusted according to the boiling degree of the pressure filtration condition.

[0055] (3) Outlier removal: Based on the temperature and pressure data after the timestamp synchronization, an associated feature interval is established with the filtered liquid level data. When the liquid level data of a certain collection period exceeds the corresponding associated feature interval, it is determined to be a false liquid level outlier. The outlier is removed and the effective liquid level data of the previous period is used for interpolation to supplement it, so as to avoid false data causing water replenishment control malfunction.

[0056] By setting up a data acquisition unit that differs from conventional technologies, conventional liquid level acquisition schemes only perform a single fixed threshold filter on the liquid level data, lacking consideration of the correlation characteristics between temperature, pressure and liquid level during the pressure filtration process, and failing to eliminate false liquid level errors caused by boiling disturbances and steam condensation. The data acquisition unit proposed in this embodiment achieves time-series alignment of multi-source data through timestamp synchronization, and combined with sliding window filtering and outlier removal based on multi-dimensional correlation, it improves the acquisition accuracy and anti-interference capability of liquid level data, providing a reliable data foundation for subsequent precise water replenishment control.

[0057] The edge computing processing unit uses an embedded microprocessor as its computing core. Its signal input terminal is communicatively connected to the signal output terminal of the data acquisition unit. It receives pre-processed effective liquid level data, temperature data, and pressure data, and sequentially performs feature extraction, correlation analysis, dynamic evaporation prediction, and closed-loop control quantity calculation, finally outputting the corresponding water replenishment control command.

[0058] (1) Feature extraction and correlation analysis: The edge computing processing unit performs feature dimension mapping on the preprocessed multi-dimensional data, extracts the correlation features of saturated vapor pressure corresponding to the liquid level change rate, temperature change rate, and real-time pressure, establishes a dynamic correlation model of temperature-pressure-liquid level during the filter press process, and clarifies the influence of temperature rise and pressure change on liquid level evaporation loss during the filter press process.

[0059] (2) Dynamic Evaporation Prediction Model: To address the issue that conventional water replenishment schemes do not consider the liquid level loss caused by continuous evaporation during filtration, thus failing to achieve dynamic water replenishment and leading to deviations in the drug concentration during filtration, this embodiment constructs a dynamic evaporation prediction model based on extracted correlation features. This model outputs the predicted liquid level evaporation loss per unit time in real time, and the corresponding calculation formula is: In the formula, t represents the predicted liquid level evaporation loss per unit time interval at time t; K is a correction coefficient related to the inner cavity volume and cross-sectional area of ​​filter press 2, which is specifically obtained through pre-experiment calibration. The real-time pressure data inside filter press 2 at time t; The preset standard atmospheric pressure; The enthalpy of saturated vapor corresponds to the real-time temperature at time t, which can be obtained by looking up the thermodynamic properties of steam. The unit time interval; based on the set dynamic evaporation prediction model, the amount of liquid level drop caused by evaporation during the filter press process can be predicted in real time, providing a basis for dynamic water replenishment and avoiding fluctuations in the filter press operation caused by replenishing water after the liquid level is below the set threshold.

[0060] (3) Closed-loop control quantity calculation: This embodiment adopts an incremental closed-loop control algorithm. The difference between the target liquid level data set by the user through the control panel and the filtered real-time liquid level data is used as the control deviation. Combined with the predicted evaporation loss value, feedforward compensation is performed to calculate the incremental control quantity of the on / off time of solenoid valve tube 4. The corresponding calculation formula is: In the formula, This is the incremental control quantity for the kth control cycle, corresponding to the increment of the on / off duration of solenoid valve tube 4 within this control cycle; This is the proportionality coefficient. The integral coefficient is... These are the differential coefficients, which are specifically obtained through tuning using the critical proportionality method or engineering experience method. These are existing conventional techniques and will not be elaborated upon here. The control deviation is the control error in the k-th control cycle. Hset is the target liquid level data set by the user. This is the real-time liquid level data after filtering during the k-th control cycle. This is the predicted value of liquid level evaporation loss for the kth control cycle; These are the control deviations for the (k-1)th and (k-2)th control cycles, respectively. Based on the accumulation of incremental control quantities, the total on / off duration control quantity of solenoid valve tube 4 is obtained, and the corresponding water replenishment control command is generated and transmitted to the execution control unit.

[0061] This embodiment differs from conventional technologies by setting up an edge computing processing unit. Conventional automatic water replenishment schemes often use fixed threshold switching control, turning on water replenishment only when the liquid level is below the lower threshold and turning it off when it is above the upper threshold. This does not consider the impact of dynamic evaporation loss during the filtration process and does not achieve precise closed-loop control, which easily leads to problems such as water replenishment overshoot and large liquid level fluctuations. This embodiment achieves continuous and precise liquid level control by constructing a dynamic evaporation prediction model to achieve feedforward compensation, combined with an incremental closed-loop control algorithm. There is no integral accumulation problem, the control process is stable and there is no overshoot, and it can also achieve dynamic sealed water replenishment during the filtration process without opening the lid. This solves the problems of insufficient water addition accuracy, steam leakage, and chemical contamination in conventional technologies.

[0062] The communication interaction unit establishes bidirectional communication links with the edge computing processing unit, control panel, and execution control unit, including two redundant communication links: a wired communication link and a wireless communication link. The wired communication link uses an RS485 bus or Ethernet bus to achieve local real-time data transmission between the edge computing processing unit and the execution control unit and control panel, ensuring low-latency transmission of control commands. The wireless communication link uses WiFi, Bluetooth, or cellular communication modules to achieve remote data interaction between the edge computing processing unit and external terminal devices, supporting users to remotely configure control parameters and view device operating status through external terminal devices. The wired and wireless communication links and corresponding communication protocols in this section are all conventional communication technologies in the field of industrial measurement and control, and their specific implementation methods are existing conventional technologies, which will not be elaborated here.

[0063] The execution control unit uses a relay drive module and a power drive module. Its signal input terminal is connected to the output terminal of the communication interaction unit, and its signal output terminal is electrically connected to the controlled terminals of the solenoid valve tube 4, the heating plate, and the drive motor, respectively. According to the received water replenishment control command, the execution control unit adjusts the on / off state and on / off duration of the solenoid valve tube 4 to achieve precise water replenishment. At the same time, according to the temperature control command, it adjusts the output power of the heating plate in linkage to stabilize the filter press operation.

[0064] The built-in safety interlock control logic of the control unit is executed as follows:

[0065] When the collected real-time pressure data exceeds the set pressure threshold, or the real-time temperature data exceeds the set temperature threshold, the execution control unit prioritizes cutting off the power supply circuit of the heating plate, simultaneously closing the solenoid valve tube 4, and outputting alarm information to the control panel through the communication interaction unit; when the real-time liquid level data reaches the set target liquid level threshold during the water replenishment process, the execution control unit immediately closes the solenoid valve tube 4, terminating the water replenishment process; when the device experiences abnormal operating conditions such as power failure or communication interruption, the execution control unit defaults to closing the power supply circuit between the solenoid valve tube 4 and the heating plate to ensure the safe operation of the device;

[0066] The relay drive and power drive in this section are conventional drive technologies for industrial actuators, and the basic safety interlock logic is a conventional safety design for industrial measurement and control equipment. The specific implementation method is an existing conventional technology, which will not be elaborated here. In this embodiment, the linkage control logic between the execution control unit and the liquid level acquisition, evaporation prediction, and closed-loop control algorithm is a creative content, which has been described in detail in the foregoing.

[0067] The complete workflow of the device proposed in this embodiment is as follows:

[0068] Preliminary preparation: Connect the inlet end of the solenoid valve pipe 4 to the external water supply pipe, apply thermally conductive silicone grease to the contact surface between the temperature sensor and the filter press tank 2, put the raw materials to be filtered and prepared into the filter press tank 2, close the tank cover 3, and lock the tank cover 3 by engaging the locking handle with the internal threaded hole seat to complete the sealing and fixing.

[0069] Parameter configuration: Users set control parameters such as target liquid level data, heating temperature, and filtration time through the control panel. The parameters are transmitted to the edge computing processing unit through the communication interaction unit.

[0070] Automatic dosing and filter press operation: When the device starts up, the data acquisition unit synchronously collects real-time liquid level, temperature, and pressure data in filter press tank 2 and completes preprocessing; the edge computing processing unit performs evaporation prediction and control calculations based on the collected data and outputs water replenishment control commands; the execution control unit controls the opening and closing of the solenoid valve 4 according to the water replenishment control commands. When the liquid level is lower than the set value, the solenoid valve 4 automatically opens, and water enters the filter press tank 2 through the first flange pipe and the water replenishment pipe until the liquid level reaches the target level. At the same time, the solenoid valve 4 automatically closes, completing the automatic water replenishment; simultaneously, the heating plate heats the filter press tank 2, the temperature sensor collects temperature data in real time, and the edge computing processing unit adjusts the output power of the heating plate according to the temperature data to ensure stable filter press temperature.

[0071] Drug residue compression and liquid discharge: During or after the filtration process, the drive motor drives the transmission pulley 7 to rotate, which in turn drives the screw to rotate, causing the internal threaded cylinder to drive the compression disc 8 to descend and compress the drug residue in the filter press 2. The compressed liquid is collected at the bottom of the filter press 2. The valve on the liquid outlet pipe 10 is opened, and the liquid is discharged through the connecting pipe 9 and the liquid outlet pipe 10, completing the drug preparation and liquid discharge operation.

[0072] Safety Protection: Throughout the entire operation of the device, the execution control unit monitors the operating parameters in real time. When abnormal conditions such as over-temperature or over-pressure occur, it immediately executes safety interlock actions, cuts off the heating circuit, closes the water supply valve, and outputs alarm information to ensure the safe operation of the device.

[0073] This embodiment further optimizes the edge computing processing unit described above by adding a working condition self-identification module and an algorithm parameter adaptive tuning module. This enables intelligent identification of working conditions and dynamic adjustment of control algorithm parameters throughout the entire filter press process. Compared to conventional fixed-parameter closed-loop control, the adaptability and control accuracy are further improved. The specific implementation method is as follows:

[0074] The edge computing processing unit expands the storage unit and computing logic based on the original embedded microprocessor, integrating a self-identification module for operating conditions and an adaptive tuning module for algorithm parameters. It works collaboratively with the existing feature extraction, correlation analysis, dynamic evaporation prediction, and closed-loop control quantity calculation modules. The self-identification module constructs a filter press operating condition feature library based on pre-processed multi-dimensional data of temperature, pressure, and liquid level, achieving real-time intelligent identification of the preheating stage, simmering stage, vigorous boiling stage, and isothermal filter press stage through feature matching. The adaptive tuning module adjusts the proportional coefficient of the incremental closed-loop control algorithm based on the real-time operating condition results output by the self-identification module. Integral coefficient Differential coefficients Dynamic tuning is performed, and the correction coefficient K of the dynamic evaporation prediction model is corrected in real time, so that the control algorithm and the prediction model are always adapted to the current filter press conditions, avoiding the decrease in control accuracy due to changes in operating conditions.

[0075] Specifically, the operating condition self-identification module pre-builds a four-stage operating condition feature library for the entire pressure filtration process. The core feature parameters for each stage are the temperature change rate threshold, pressure change rate threshold, and liquid level fluctuation amplitude threshold.

[0076] Preheating stage: The rate of temperature change is within the first threshold range, the pressure does not change significantly, and the liquid level does not fluctuate;

[0077] Micro-boiling stage: The rate of temperature change drops to the second threshold range, the pressure begins to rise slightly, and the liquid level fluctuates slightly;

[0078] Vigorous boiling stage: The temperature remains constant, the rate of pressure change is in the third threshold range, and the amplitude of liquid level fluctuation reaches its maximum value.

[0079] Constant temperature pressure filtration stage: Temperature and pressure are kept constant, liquid level shows a uniform downward trend, and fluctuation amplitude drops to the fourth threshold range;

[0080] Among them, the temperature remains constant, which means that the rate of change and the amplitude of fluctuation of the temperature fall within the preset constant threshold range during the continuous sampling period, without a continuous upward or downward trend; the temperature and pressure remain constant, which means that both the temperature and pressure parameters are within the constant threshold range according to the above rules; the liquid level shows a uniform downward trend, which means that the liquid level drops per unit time during the continuous period is calculated, and when the deviation of this value is within the preset range, it is determined to be a uniform downward trend. At the same time, combined with the liquid level fluctuation amplitude falling within the fourth threshold range, the determination of the constant temperature pressure filtration stage is finally completed.

[0081] During device operation, the operating condition self-identification module extracts the rate of change and fluctuation amplitude characteristics of preprocessed temperature, pressure, and liquid level data at a preset identification cycle. It then performs similarity matching between the real-time feature parameters and the feature parameters for each stage in the feature library. The similarity calculation formula is: [Formula omitted for brevity]. In the formula, S represents the similarity between the real-time feature and the feature in the database, and its value ranges from [0,1]. This is the normalized value of the m-th real-time feature parameter; is the normalized value of the m-th feature parameter in the library; M is the total number of feature parameters; when the similarity S is greater than the preset similarity threshold, it is determined to be a successful match, and the recognition result of the current filter press condition is output; if the match is unsuccessful, it is determined to be a transitional stage of the working condition, and the parameter configuration of the previous stable working condition is used until a stable working condition feature is matched.

[0082] The feature library construction and feature matching algorithm in this embodiment are conventional technologies in the field of pattern recognition. However, by combining them with the traditional Chinese medicine pressure filtration process, intelligent recognition of the pressure filtration stage is achieved, and these technologies serve as the basis for adjusting the control algorithm parameters.

[0083] The algorithm parameter adaptive tuning module pre-configures basic parameter sets and correction rules for each of the four stages of pressure filtration. The basic parameter sets are the optimal sets calibrated experimentally for each operating condition. The initial value of the correction coefficient K is used; the correction rule is a parameter fine-tuning logic based on the real-time liquid level deviation and the predicted deviation of evaporation loss.

[0084] During device operation, this module first retrieves the basic parameter set corresponding to the operating condition based on the real-time operating condition results from the self-identification module, using it as the initial parameters for the current control algorithm and prediction model. Then, it calculates the actual liquid level deviation and the predicted evaporation loss deviation in control cycles, and dynamically fine-tunes the parameters according to preset correction rules.

[0085] Actual liquid level deviation: The difference between the set liquid level and the filtered actual liquid level in the current control cycle. When the deviation continues to exceed the preset deviation threshold, it increases proportionally. , reduce To accelerate system response speed; when the deviation remains below the preset deviation threshold, reduce proportionally. Increase Improve system control stability;

[0086] Evaporation loss prediction deviation: The difference between the actual liquid level drop in the current control cycle and the evaporation loss predicted by the model. When the deviation is consistently positive, it indicates that the model prediction is too small, and the correction coefficient K is increased proportionally. When the deviation is consistently negative, it indicates that the model prediction is too large, and the correction coefficient K is decreased proportionally.

[0087] The formula for parameter fine-tuning is:

[0088] In the formula, These are the parameter values ​​after fine-tuning; These are the parameter values ​​before fine-tuning; α is the parameter fine-tuning coefficient, which is pre-calibrated according to the working conditions. This represents the deviation value for the current period. This is the preset deviation benchmark value.

[0089] The dynamically fine-tuned parameters are fed back to the closed-loop control quantity calculation module and the dynamic evaporation prediction model in real time, so as to realize the real-time adaptation of algorithm parameters and filter press conditions.

[0090] It should be noted that conventional closed-loop control algorithms use a fixed... The parameters cannot adapt to the changing operating conditions at different stages of the filtration process, and are prone to liquid level overshoot during the violent boiling stage and control lag during the isothermal filtration stage. Conventional evaporation prediction models use a fixed correction coefficient K, which cannot compensate for the differences in evaporation characteristics caused by different raw material types and dosages during the filtration process. This embodiment achieves intelligent identification and control of filtration conditions and dynamic adjustment of prediction parameters by adding an operating condition self-identification module and an algorithm parameter adaptive tuning module. This ensures that the calculation results of the edge computing processing unit are always highly matched with the actual filtration conditions, further improving the accuracy of automatic water addition and solving the problem of insufficient control accuracy of conventional fixed parameter schemes under complex operating conditions.

[0091] This embodiment also proposes another implementation method for the edge computing processing unit. It abandons the single incremental PID closed-loop control algorithm and replaces the original algorithm with a model predictive control (MPC) algorithm. Furthermore, it combines multi-source data fusion technology to optimize the dynamic evaporation prediction model. Compared to the original implementation method, it has stronger anti-interference capabilities and better look-ahead control performance, effectively suppressing liquid level fluctuations caused by multiple factors during the pressure filtration process. The specific implementation method is as follows:

[0092] The edge computing processing unit in this embodiment uses an embedded microprocessor as its computing core, integrating a multi-source data fusion module and a model prediction control module to replace the original feature extraction, correlation analysis and incremental closed-loop control modules. The dynamic evaporation prediction module is optimized based on the multi-source data fusion results, and the whole system realizes the integrated operation of "data fusion - accurate prediction - forward control" and outputs water replenishment control commands.

[0093] The multi-source data fusion module uses a weighted fusion algorithm to perform multi-dimensional fusion of pre-processed liquid level, temperature, and pressure data, as well as historical filter press process data and raw material type correlation data, eliminating the limitations of a single data dimension and improving the reliability and comprehensiveness of feature data.

[0094] First, the data of each dimension are normalized, and the feature parameters of different dimensions are mapped to the [0,1] interval. The normalization process is a conventional technique and will not be elaborated here.

[0095] Subsequently, based on the influence weights of each dimension of data on evaporation prediction and liquid level control, weighting coefficients were assigned to each normalized eigenvalue. These weighting coefficients were pre-calibrated using the Analytic Hierarchy Process (AHP). The final formula for merging the eigenvalues ​​is as follows: ;

[0096] In the formula, F is the integrated eigenvalue after fusion; Let be the weighting coefficients of the nth dimension data, and satisfy . ; is the normalized feature value of the nth dimension data; N is the total number of terms in the data dimension; the fused comprehensive feature value is transmitted to the dynamic evaporation prediction module in real time as the core input parameter for model calculation.

[0097] The dynamic evaporation prediction module optimizes the original prediction model based on the comprehensive feature values ​​obtained from multi-source data fusion. It introduces raw material type correction factors and historical process correction factors to eliminate the influence of individual factors such as raw material type and dosage on evaporation characteristics. The optimized prediction formula is as follows: In the formula, λ is the raw material type correction factor, which is pre-calibrated based on the water absorption and air permeability of different Chinese herbal raw materials, and its value range is [0.8, 1.2]; μ is the historical process correction factor, which is corrected in real time based on the historical pressure filtration and evaporation data of the same batch of raw materials, and its value range is [0.9, 1.1]; the meanings of the other parameters are consistent with the original prediction model mentioned above.

[0098] By introducing a double correction factor This enables the prediction model to not only adapt to the dynamic changes in filter press conditions, but also to compensate for the influence of individual raw material factors, thus significantly improving the accuracy of evaporation loss prediction.

[0099] The model predictive control module, based on the optimized dynamic evaporation prediction model, constructs a predictive model for liquid level control. It employs a model predictive control algorithm to achieve look-ahead control of solenoid valve tube 4. Compared to the original incremental PID algorithm, this algorithm can predict the liquid level change trend over several control cycles in advance and optimize the control quantity according to constraints, avoiding liquid level overshoot and lag. The specific implementation process is as follows:

[0100] Predictive step size setting: Based on the rate of change of the filter press operating conditions, set the future step size. Prediction step size for each control cycle, It is a positive integer and can be dynamically adjusted by the working condition self-identification module;

[0101] Liquid level trend prediction: Based on the current fused feature values ​​and the optimized evaporation loss prediction model, predict future liquid levels. The liquid level change trend of each control cycle is used to obtain the predicted liquid level sequence. ;

[0102] Control quantity optimization: with "future" The objective function is to minimize the sum of the deviations between the predicted liquid level and the set liquid level for each control cycle. The on / off time of solenoid valve 4 is used as the control variable, and the upper and lower limits of the liquid level and the upper and lower limits of the solenoid valve on / off time are used as constraints. An optimization problem is constructed and solved to obtain the future... The optimal control quantity sequence for each control cycle ( );

[0103] Rolling optimization execution: Only the control quantity in the first control cycle of the optimal control quantity sequence is executed to control the on / off time of solenoid valve tube 4; in the next control cycle, data is collected again, liquid level is predicted, and optimization problem is solved to achieve rolling optimization of control quantity;

[0104] The formula for calculating the objective function is as follows: In the formula, J is the objective function value; To set the target liquid level; The predicted liquid level for the j-th prediction cycle; This is the penalty coefficient for the control quantity, used to suppress drastic changes in the control quantity; This is the control increment for the kth control cycle; through the above model predictive control algorithm, we achieve forward-looking, precise, and stable control of the liquid level, and the output optimal control quantity is transmitted to the execution control unit to control the action of the water replenishment component;

[0105] It should be noted that existing liquid level control methods mostly employ PID series algorithms, which are "feedback control" systems. These only adjust the control quantity based on the current liquid level deviation, resulting in control lag and an inability to handle multi-factor disturbances during the pressure filtration process. Conventional evaporation prediction is based solely on single-dimensional data such as temperature and pressure, neglecting personalized factors like raw material type and historical process history, thus limiting prediction accuracy. This implementation method uses Model Predictive Control (MPC) algorithms, which are "look-ahead control" systems. They can predict future liquid level changes and optimize the control quantity in advance, fundamentally solving the lag problem of conventional feedback control. Furthermore, by combining multi-source data fusion technology and introducing dual correction factors to optimize the evaporation prediction model, it achieves multi-dimensional and personalized evaporation loss prediction with significantly higher accuracy than conventional single-dimensional prediction models. Compared to conventional control schemes, this implementation method exhibits significant technical differences, and its control accuracy, anti-interference capability, and adaptability to operating conditions are all effectively improved.

[0106] Both of the above-mentioned edge computing processing unit implementation methods can be applied independently to the control system of the present invention, or they can be combined. Both methods can further optimize the original functions and improve the core performance of the device in accurately identifying automatic water addition.

[0107] It should be further noted that in this specific implementation, some letter symbols in the algorithm formulas may be identical but have different meanings. The meaning of each letter symbol shall be based on the text description of the corresponding formula paragraph. Specifically, the letter H in the sliding window filtering formula, the original and optimized formulas of the dynamic evaporation prediction model, and the control deviation calculation formula only corresponds to the filtered effective liquid level data, the original liquid level sampling data, the predicted value of liquid level evaporation loss, the real-time liquid level data, the set target liquid level data, and the predicted liquid level data, respectively, in the formula. The letter K in the dynamic evaporation prediction model refers to the filter press related modification... Positive coefficients, in incremental closed-loop control algorithms, are proportional, integral, and derivative coefficients, respectively. They are not used in parameter fine-tuning formulas; they are only used to distinguish parameters in different formulas. The letter N represents the window length in sliding window filtering formulas, and in similarity matching formulas, multi-source data fusion formulas, and model predictive control algorithms, it represents the total number of feature parameters, the total number of data dimensions, and the prediction step size / control step size, respectively. The use of the above letter symbols is an independent parameter identifier within each algorithm formula. Identical letter symbols in different formulas are unrelated, and their specific meanings are limited to the application scenario of the corresponding formula and the textual explanation of the paragraph to which they belong.

[0108] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An automatic water and medicine dispensing device with precise identification capability, comprising a cabinet (1), a slot on the top surface of the cabinet (1), a filter press tank (2) inserted into the slot, a tank cover (3) for sealing the filter press tank (2) hinged to one side of the top surface of the cabinet (1), a safety valve provided on the top surface of the tank cover (3), a pressure gauge provided on one side of the safety valve, a locking handle installed on the outer wall of the end of the tank cover (3) away from the hinge, and an internal threaded hole seat corresponding to the locking handle fixed to the upper end of the outer wall of the filter press tank (2), characterized in that: The top surface of the can lid (3) is equipped with a squeezing assembly. The bottom surface of the cabinet (1) is provided with a heating plate that abuts against the bottom of the filter press (2). One end of the top surface of the cabinet (1) is provided with a vertical plate. The vertical plate is provided with a control panel. The upper end of one end of the cabinet (1) is equipped with a water replenishment assembly for replenishing water to the filter press (2). The lower end of the outer wall of the filter press (2) is provided with a liquid mixing assembly that extends out of the outer wall of the cabinet (1). The lower end of the outer wall of the filter press (2) is provided with a temperature sensor. The top surface of the can lid (3) is provided with a drug filling port with a sealing cap for quantitative dispensing of drug raw materials. Sealing grooves are opened on the opposite surfaces of the filter press (2) and the can lid (3). A sealing ring (11) is fixedly connected in the sealing groove of the can lid (3). The bottom shape of the sealing ring (11) corresponds to the sealing groove of the filter press (2). The water replenishment assembly is connected to the filter press (2) in a sealed manner through a flange connection structure. The water replenishment assembly includes a first flange pipe fixed to the outer wall of the cabinet (1), a solenoid valve pipe (4) is connected to one end of the first flange pipe, a water replenishment pipe is provided on the upper end of the outer wall of the filter press (2) and connected to the other end of the first flange pipe, one end of the water replenishment pipe is connected to the inside of the filter press (2), and a liquid level sensor (5) is provided on one side of the top surface of the tank cover (3), the detection end of the liquid level sensor (5) extends into the inside of the filter press (2); The control panel integrates a control system, which includes a data acquisition unit, an edge computing processing unit, a communication interaction unit, and an execution control unit. The signal output terminal of the data acquisition unit is communicatively connected to the signal input terminal of the edge computing processing unit. The edge computing processing unit establishes a bidirectional communication link with the execution control unit through the communication interaction unit. The data acquisition unit is used to synchronously acquire liquid level, temperature, and pressure data in the filter press tank, perform preprocessing, and then output the data to the edge computing processing unit. The edge computing processing unit is used to predict liquid level evaporation loss, calculate water replenishment control amount, and generate water replenishment control commands based on the acquired data. The execution control unit is used to control the operating status of the water replenishment component and the heating plate according to the water replenishment control commands.

2. The automatic water and medicine dispensing device with precise identification capability according to claim 1, characterized in that: The extrusion assembly includes a guide sleeve (6) that is fixedly connected to the middle of the top surface of the can lid (3). A transmission box is fixedly connected to the top of the guide sleeve (6). A transmission pulley (7) is provided inside the transmission box. The transmission pulley (7) includes a driving pulley and a driven pulley. A handwheel (12) is rotatably provided on one side of the top surface of the transmission box through a bearing. The shaft of the handwheel (12) is coaxially fixedly connected to the driven pulley. A drive motor is installed on one side of the bottom surface of the transmission box. The output shaft of the drive motor is coaxially fixedly connected to the driving pulley through an electromagnetic clutch. A screw is rotatably provided inside the guide sleeve (6). The top of the screw is coaxially fixedly connected to the shaft of the handwheel (12). An internal threaded cylinder is screwed onto the screw. An extrusion disc (8) located inside the filter press is fixedly connected to the bottom end of the internal threaded cylinder. A guide groove is opened on the outer wall of the internal threaded cylinder. A guide strip corresponding to the guide groove is fixedly connected to the inner wall of the guide sleeve (6). Multiple through holes are evenly opened on the extrusion disc (8).

3. The automatic water and medicine dispensing device with precise identification capability according to claim 1, characterized in that: The liquid mixing assembly includes two symmetrically arranged connecting pipes (9) fixed to the lower end of one side of the outer wall of the filter press (2). One end of the connecting pipe (9) is connected to the middle of one side of the mixing tank (13). A first motor is installed in the middle of the top surface of the mixing tank (13). The output shaft of the first motor is coaxially fixed to a stirring paddle that is rotatably installed inside the mixing tank (13). A liquid outlet pipe (10) is connected to the lower end of one side of the mixing tank (13). One end of the liquid outlet pipe (10) extends out of the outer wall of the cabinet (1). A valve is provided on the pipe of the liquid outlet pipe (10) located outside the cabinet (1). The liquid outlet pipe (10) adopts a Y-shaped pipe and a first solenoid valve is installed at the bifurcation. A second solenoid valve is provided on the connecting pipe (9).

4. The automatic water and medicine dispensing device with precise identification capability according to claim 1, characterized in that: The data acquisition unit is equipped with a multi-channel signal acquisition interface and an analog-to-digital conversion module, which are respectively connected to the liquid level sensor (5), the temperature sensor, and the pressure gauge. It is used to synchronously acquire the liquid level, temperature, and pressure data in the filter press (2), and output the acquired data to the edge computing processing unit after preprocessing.

5. The automatic water and medicine dispensing device with precise identification capability according to claim 1, characterized in that: The edge computing processing unit receives data output from the data acquisition unit, extracts data features and establishes a correlation model, predicts liquid level evaporation loss, calculates water replenishment control amount based on the set target liquid level, and generates water replenishment control command.

6. The automatic water and medicine dispensing device with precise identification capability according to claim 1, characterized in that: The communication interaction unit establishes a bidirectional redundant communication link, which is connected to the edge computing processing unit, the control panel, and the execution control unit respectively, to realize local data transmission and remote data interaction.

7. The automatic water and medicine dispensing device with precise identification capability according to claim 1, characterized in that: The execution control unit is connected to the controlled end of the solenoid valve tube (4), the heating plate, and the drive motor respectively. It is used to adjust the on / off state and duration of the solenoid valve tube (4) according to the water replenishment control command, and to adjust the output power of the heating plate in linkage. It also has built-in safety interlock control logic.

8. The automatic water and medicine dispensing device with precise identification capability according to claim 5, characterized in that: The edge computing processing unit integrates a working condition self-identification module and an algorithm parameter adaptive tuning module. The working condition self-identification module identifies the filter press working condition based on the collected temperature, pressure, and liquid level data. The algorithm parameter adaptive tuning module dynamically adjusts the operating parameters of the control algorithm and prediction model according to the identified filter press working condition. The working condition self-identification module is based on pre-processed multi-dimensional data of temperature, pressure and liquid level to build a filter press working condition feature library. Through feature matching, it realizes real-time intelligent identification of the preheating stage, micro-boiling stage, violent boiling stage and constant temperature filter press stage. The algorithm parameter adaptive tuning module adjusts the proportional coefficient of the incremental closed-loop control algorithm based on the real-time operating condition results output by the operating condition self-identification module. Integral coefficient Differential coefficients Dynamic tuning is performed, and the correction coefficient K of the dynamic evaporation prediction model is corrected in real time to ensure that the control algorithm and prediction model are always adapted to the current filter press conditions.

Images