An electric control system for mixing feed with liquid level sensing

Abstract

The application discloses a kind of liquid level sensing mixing feed electric control system, it is related to control system field, including: material characteristic pre-detection module;Adaptive parameter configuration module;Multi-stage gradient stirring module;Real-time data twin processing module;Remote collaborative monitoring module;Fault linkage processing module;Energy-saving closed-loop control module.The application is combined water content-viscosity-chemical dosing quantity three-dimensional mapping table and viscosity-speed mapping table by material characteristic pre-detection module, realize the real-time scanning and parameter quantization matching of industrial waste granularity, water content, viscosity, compared with the system of traditional only relying on liquid level control, the scheme makes the stirring sufficiency of high-viscosity industrial sludge improve, low-viscosity material energy consumption reduces, and chemical dosing quantity is automatically matched according to three-dimensional mapping table, and solidifying agent utilization rate improves, fundamentally solves the core problem of stirring waste, stirring is not sufficient.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of control systems, in particular to a mixing feeding electric control system with liquid level sensing. BACKGROUND

[0002] In the field of industrial waste treatment, the traditional mixing feeding system usually only realizes feeding control through a liquid level sensor, lacking real-time detection capability of material characteristics (such as viscosity, moisture content, and granularity). Although the existing technology can adjust the stirring speed through the liquid level sensor, it cannot dynamically match the optimal parameters for high-viscosity industrial sludge and low-viscosity river sludge, often leading to problems such as "waste of high-viscosity material stirring energy" and "insufficient stirring of low-viscosity material". In addition, the dosage of reagents in the existing technology is mostly set by manual experience, without establishing a quantitative mapping relationship between material characteristics and dosage parameters, making it difficult to adapt to the solidification needs under different working conditions.

[0003] The stirring control of the traditional mixing system mostly adopts an open-loop control mode of "single speed + fixed liquid level threshold", lacking a dynamic adjustment mechanism according to the liquid level rising rate and material accumulation state. Typical defects include: liquid level fluctuation easily leads to frequent start and stop of the stirring motor, shortening the service life of the equipment; fixed phase stirring time cannot adapt to the change of material feeding rate, which may cause material blockage or insufficient stirring; without predictive control, it can only be adjusted passively after material accumulation, with a lagging response. Moreover, the fault diagnosis of the existing mixing system mostly relies on single sensor detection, lacking redundancy configuration and self-calibration mechanism, which easily leads to control failure when the sensor drifts or adheres. For example, the disclosure document only mentions the sensor fault self-diagnosis function, but does not form a three-level diagnosis and redundancy switching logic. In terms of remote monitoring, the traditional system is mostly locally controlled, which cannot realize 5G cloud data synchronization and real-time fault positioning, and is not suitable for industrial Internet of Things needs; in terms of energy-saving control, it generally lacks sleep mode and preheating strategy optimization, with large instantaneous power consumption at startup and residual material easily solidified and left. SUMMARY

[0004] The present application aims to provide a mixing feeding electric control system with liquid level sensing to solve the problems raised in the background.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical solution: a mixing feeding electric control system with liquid level sensing, comprising:

[0006] A material characteristic pre-detection module, through a multi-spectral sensor installed at the inlet of the feeding hopper, performs real-time scanning of the industrial waste entering the feeding hopper in terms of granularity, moisture content, density, and viscosity, and generates a material characteristic parameter data package;

[0007] An adaptive parameter configuration module, based on the material characteristic parameter data packet, uses a fuzzy logic algorithm built-in PLC to automatically match a stirring model, with viscosity as the primary basis for judgment, to set the initial speed of the variable frequency motor, the proportion of the medicine added, and the combination strategy of the stirring time simultaneously;

[0008] A multi-stage gradient stirring module is used to start low-speed stirring when the liquid level sensor detects that the material reaches the first liquid level threshold, switch to medium-speed stirring when it reaches the second liquid level threshold, and start high-speed stirring when it reaches the third liquid level threshold, with each stage maintaining a gradient stirring of 5-10 minutes;

[0009] A real-time data twin processing module is used to connect the real-time data of liquid level, motor current, and stirring torque to the edge computing unit to build a digital twin model of the mixing process, and to predict the material accumulation state and adjust the stirring strategy in advance by 0.5-1 minute through model predictive control algorithm;

[0010] A remote collaborative monitoring module is used to synchronize the mixing data to the cloud management platform through a 5G communication module, and when the liquid level fluctuates abnormally by more than ±5% threshold, it automatically triggers the audible and visual alarm of the remote terminal and pushes the fault positioning report containing real-time video stream;

[0011] A fault linkage processing module is used to automatically switch to a backup capacitive sensor and start a self-calibration program when the liquid level sensor detects a deviation of more than ±3% for three consecutive times, and to reduce the motor speed to a safe range, and to restore normal operation after the fault is repaired, and to trigger an emergency stop when the torque exceeds 120% of the rated value;

[0012] An energy-saving closed-loop control module is used to automatically enter a low-power standby sleep mode when the mixing is completed and the liquid level is below 10% of the volume threshold for 30 seconds, and to optimize the motor preheating strategy for the next start based on historical 24-hour operation data to avoid false triggering due to instantaneous fluctuations.

[0013] Preferably, in the adaptive parameter configuration module, the synchronous setting of the medicine adding proportion control logic is: a three-dimensional mapping table of moisture content-viscosity-medicine adding amount is established and stored in the PLC register, and the medicine adding port uses electromagnetic flowmeter closed-loop control with a control accuracy of ±1.5% of the rated flow rate, and the three-dimensional mapping table is constructed by the following method:

[0014] When the viscosity is ≤5000 cP, the Newtonian fluid medicine adding model is matched, and the medicine adding amount is set as: when the moisture content is ≤30%, the adding amount is 15-20 kg / m 3 ; when 30% < moisture content ≤50%, the adding amount is 20-25 kg / m 3 ; when the moisture content is >50%, the adding amount is increased by 10% of the base value;

[0015] When the viscosity is greater than 5000cP and the water content is less than 40%, the Bingham plastomer compensation model is matched and the dosing strategy is: when the viscosity is less than 5000cP and the water content is less than 8000cP, the dosage is set to 25-30kg / m 3 , and start the pulse dosing mode (10 seconds / time interval); when 8000cP<viscosity≤10000cP, the dosage is increased to 30-35kg / m 3 , and activate the medicine pre-mixing program at the same time;

[0016] When the viscosity is greater than 10000 cP, the high viscosity agent strengthening strategy is activated. Specifically, when the water content is ≥40%, the dosage is 1.5 times the upper limit of the Newtonian fluid model (i.e. 30-35 kg / m 3 ); Synchronously start the dual-agent dosing port collaborative working mode, the main dosing port adds 70% of the calculated value, and the auxiliary dosing port adds 30% compensation; when the moisture content of the material is greater than 80%, the dosing ratio of the agent will automatically increase by 10%-15%.

[0017] Preferably, in the adaptive parameter configuration module, the fuzzy logic algorithm has a built-in viscosity-speed mapping table, which is constructed in the following way:

[0018] When the viscosity is ≤5000cP, the Newtonian fluid model is matched and the speed is set to 50-80rpm;

[0019] When the viscosity is greater than 5000 cP and the water content is less than 40%, the Bingham plastic model is matched and the speed is set to 120-180 rpm;

[0020] When the viscosity is greater than 10,000 cP, the high torque compensation strategy is activated and the speed is increased to 200-300 rpm.

[0021] Preferably, in the multi-stage gradient stirring module, the liquid level threshold trigger adopts a hysteresis control method: after the first threshold triggers low-speed stirring, the liquid level must drop below 25% before it is allowed to stop; a ±2% buffer zone is set between the second and third thresholds; and the first threshold is Volume, the second threshold is Volume, the third threshold is volume;

[0022] The duration of each mixing stage is dynamically adjusted according to the material characteristics:

[0023]

[0024] Where μ is viscosity, ρ is density, k=0.08, and b=5.

[0025] Preferably, in the multi-stage gradient stirring module, the time parameter maintained by each stage is dynamically adjusted according to the liquid level rising rate, specifically:

[0026] When the liquid level rising rate is > 5 cm / min, the stage maintenance time is shortened to 5-7 minutes;

[0027] When the liquid level rising rate is ≤ 5 cm / min, the stage maintenance time is extended to 8-10 minutes;

[0028] The liquid level rising rate is calculated by the difference between the data of two adjacent liquid level detections and the detection interval time.

[0029] Preferably, in the real-time data twin processing module, the digital twin model is constructed in the following manner: at least 20 groups of liquid level-torque-current sample data under different material characteristics are collected to train an LSTM neural network; the rolling optimization time domain of the model predictive control (MPC) algorithm is set to 1 minute, and the control time domain is 0.5 minute; when the predicted material accumulation probability is > 70%, an automatic adjustment instruction of increasing the stirring speed by 10-20 revolutions per minute is generated; and the edge computing unit performs the following operations:

[0030] The material rheology state equation is established:

[0031]

[0032] wherein, is the shear stress, is the yield stress, is the consistency coefficient, is the shear rate, is the rheological index;

[0033] The current signal and the torque signal are fused through a Kalman filter, and the prediction accuracy is ±1.5%;

[0034] When the predicted accumulation risk index is > 0.7, the reverse pulse stirring (frequency 2 Hz, amplitude 10% rated speed) is increased in advance.

[0035] Preferably, in the remote collaborative monitoring module, the sound-light alarm threshold is specifically set as:

[0036] The liquid level abnormal fluctuation threshold of ±5% triggers when the actual liquid level deviation is > 10 cm;

[0037] The fault positioning report includes real-time video stream (resolution 1280×720, frame rate 25 fps), sensor waveform graph, and motor current curve;

[0038] The remote terminal (mobile phone / PC) supports parameter fine-tuning of the system through a VPN network, and the adjustment step is: speed ±1 revolution per minute, and medicament addition ratio ±0.5%.

[0039] Preferably, in the fault linkage processing module, the self-calibration program contains three levels of diagnosis:

[0040] Primary diagnosis: compare the difference between the capacitive sensor and ultrasonic sensor data, if |ΔH|>3%, start temperature drift compensation;

[0041] Intermediate diagnosis: inject test signals to verify the linearity of the sensor response (R 2 ≥0.98);

[0042] Advanced diagnosis: back-propagate the actual liquid level value through motor current spectrum analysis (FFT resolution 0.5Hz).

[0043] Preferably, in the energy-saving closed-loop control module, the residual material emptying program before entering the sleep mode is:

[0044] When the liquid level is below the 10% threshold, control the variable frequency motor to continuously stir at 150-200 rpm for 3-5 minutes;

[0045] After stirring, detect the material accumulation height in the feed hopper, if >5cm, repeat the emptying and stirring.

[0046] Preferably, in the energy-saving closed-loop control module, the motor preheating strategy includes:

[0047] Establish a start-up power consumption model based on historical data:

[0048]

[0049] Where, is the maximum start-up power peak, is the decay coefficient, is the steady-state power, t is the time, maintain a 0.5% rated power winding temperature maintenance current during sleep, and pre-rotate the speed according to an S-shaped curve 3 minutes before the next start-up.

[0050] Technical effects and advantages of the present application:

[0051] The present application realizes real-time scanning and parameter quantization matching of industrial waste particle size, moisture content, and viscosity through the material characteristic pre-detection module combined with the moisture content-viscosity-drug dosage three-dimensional mapping table and the viscosity-rotation speed mapping table. Compared with the traditional system which only relies on liquid level control, this scheme improves the stirring sufficiency of high-viscosity industrial sludge, reduces the energy consumption of low-viscosity materials, automatically matches the drug dosage according to the three-dimensional mapping table, and improves the utilization rate of solidifying agent, fundamentally solving the core problems of stirring waste and insufficient stirring in the background technology.

[0052] The application upgrades the traditional open-loop control to a forward-looking regulation mode by using the constructed hysteresis control + liquid level rate dynamic adjustment + digital twin prediction control system. The hysteresis control reduces the start-stop frequency of the stirring motor, prolongs the service life of the bearing; the liquid level rising rate dynamic adjustment reduces the risk of material blockage; the LSTM neural network + MPC algorithm predicts the material accumulation in advance 0.5-1 minutes, cooperates with the reverse pulse stirring, and the response speed is improved compared with the traditional lag control, the stirring uniformity standard deviation is reduced, and the defects of control lag and parameter rigidity in the background technology are solved;

[0053] The application fills the blank of the intelligent level of the traditional system through the combination of double-sensor redundancy + three-level diagnosis + 5G remote monitoring + energy-saving closed loop, reduces the downtime rate caused by sensor failure through the redundant configuration of ultrasonic and capacitive sensors and three-level diagnosis, shortens the self-calibration time; and shortens the fault positioning time through 5G cloud cooperation, supports remote parameter fine tuning; through the residual material emptying program cooperates with the start-up power consumption model, reduces the residual material solidification rate, reduces the start-up instantaneous power consumption, reduces the annual energy consumption cost, and solves the systematic problems of poor reliability and high energy consumption in the background technology. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 The application provides an electric control system schematic diagram. DETAILED DESCRIPTION

[0055] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the application.

[0056] The application provides a method for controlling a stirring system, comprising the steps of Figure 1The illustrated mixing feed electric control system with liquid level sensing includes: a material characteristic pre-detection module, which performs real-time scanning of the particle size, moisture content, density, and viscosity of industrial waste entering the feed hopper through a multi-spectral sensor installed at the inlet of the feed hopper, generating a material characteristic parameter data package; an adaptive parameter configuration module, which automatically matches the stirring model based on the material characteristic parameter data package using the built-in fuzzy logic algorithm in the PLC, with viscosity as the primary basis for judgment, and simultaneously sets the initial speed of the variable frequency motor, the dosage ratio of the reagent, and the combined strategy of the stirring time; a multi-stage gradient stirring module, which is used to start low-speed stirring when the liquid level sensor detects that the material reaches the first liquid level threshold, switch to medium-speed stirring when it reaches the second liquid level threshold, and start high-speed stirring when it reaches the third liquid level threshold, with each stage maintaining a gradient stirring of 5-10 minutes; a real-time data twin processing module, which is used to connect the real-time data of liquid level, motor current, and stirring torque to the edge computing unit to build a digital twin model of the mixing process, and to predict the material accumulation state and adjust the stirring strategy in advance by 0.5-1 minute through model predictive control algorithm; a remote collaborative monitoring module, which synchronizes the mixing data to the cloud management platform through a 5G communication module, and automatically triggers the audible and visual alarm of the remote terminal when the liquid level abnormally fluctuates beyond the ±5% threshold, and pushes a fault location report containing real-time video stream; a fault linkage processing module, which automatically switches to a backup capacitive sensor and starts a self-calibration program when the liquid level sensor detects a deviation of more than ±3% for three consecutive times, and reduces the motor speed to a safe range, and resumes normal operation after the fault is repaired, and triggers an emergency stop when the torque exceeds the rated value of 120%; an energy-saving closed-loop control module, which automatically enters a low-power standby sleep mode when the mixing is complete and the liquid level remains below the 10% volume threshold for 30 seconds, and optimizes the motor preheating strategy for the next start based on historical 24-hour operation data to avoid instantaneous fluctuation false triggering, and realizes intelligent matching of stirring speed, reagent dosage ratio, and stirring time through real-time scanning of the particle size, moisture content, density, and viscosity of industrial waste by the material characteristic pre-detection module, combined with the fuzzy logic algorithm of the adaptive parameter configuration module. This integrated solution breaks through the limitations of traditional single liquid level control, and builds a full-chain intelligent control system of "material sensing-parameter configuration-stirring execution-digital optimization-remote monitoring-fault handling-energy-saving closed loop", which improves the adaptability of the system to industrial waste and realizes the leap from "experience control" to "data-driven control".

[0057] Among them, in the adaptive parameter configuration module, the synchronously set reagent dosing ratio control logic is: establish a three-dimensional mapping table of moisture content-viscosity-reagent dosing amount, store it in the PLC register, and the reagent dosing port adopts electromagnetic flowmeter closed-loop control with a control accuracy of ±1.5% rated flow. The three-dimensional mapping table is constructed in the following way: when the viscosity is ≤5000cP, the Newtonian fluid reagent dosing model is matched, and the reagent dosing amount is set as follows: when the moisture content is ≤30%, the dosage is 15-20kg / m 3 When the moisture content is 30%<≤50%, the dosage is 20-25kg / m 3 When the moisture content is greater than 50%, the dosage is increased by 10% according to the basic value; when the viscosity is greater than 5000cP and the moisture content is less than 40%, the Bingham plastomer compensation model is matched, and the dosage strategy is: when the viscosity is less than 5000cP and the viscosity is less than or equal to 8000cP, the dosage is set to 25-30kg / m 3 , and start the pulse dosing mode (10 seconds / time interval); when 8000cP<viscosity≤10000cP, the dosage is increased to 30-35kg / m 3 , and activate the agent pre-mixing program at the same time; when the viscosity is greater than 10000cP, activate the high viscosity agent strengthening strategy, specifically: when the water content is ≥40%, the dosage is 1.5 times the upper limit of the Newtonian fluid model (i.e. 30-35kg / m 3 ); The dual-dosage port collaborative working mode is simultaneously activated, with the main dosing port adding 70% of the calculated value, and the auxiliary dosing port adding 30% compensation. When the material moisture content exceeds 80%, the dosing ratio automatically increases by 10%-15%. Based on a three-dimensional mapping table of moisture content, viscosity, and dosing amount, and closed-loop control using an electromagnetic flowmeter, a quantitative relationship model between material properties and dosing is established. When the viscosity is ≤5000cP, the dosing amount is precisely matched according to the Newtonian fluid model. When the viscosity is greater than 5000cP, the Bingham plastic body compensation model or high viscosity enhancement strategy is activated. Combined with pulse dosing, pre-mixing procedures, and the dual-dosage port collaborative working mode, this solves the dosage deviation problem of traditional manual dosing, achieving dynamic and precise control of dosing and optimizing dispersion efficiency.

[0058] In the adaptive parameter configuration module, the fuzzy logic algorithm has a built-in viscosity-rotation speed mapping table, which is constructed by the following methods: when the viscosity is ≤5000 cP, the Newtonian fluid model is matched, and the rotation speed is set to 50-80 rpm; when the viscosity is >5000 cP and the water content is <40%, the Bingham plastic model is matched, and the rotation speed is set to 120-180 rpm; when the viscosity is >10000 cP, the high-torque compensation strategy is activated, and the rotation speed is increased to 200-300 rpm. The built-in viscosity-rotation speed mapping table automatically switches between the Newtonian fluid, Bingham plastic model and high-torque compensation strategy according to the viscosity threshold: when the viscosity is ≤5000 cP, the low-speed energy-saving mode is matched; when the viscosity is >5000 cP and the water content is <40%, the medium-high speed stirring is started; when the viscosity is >10000 cP, the high-speed compensation is activated. This strategy deeply couples the stirring rotation speed with the material rheological properties, avoids over-stirring or insufficient stirring caused by traditional "fixed rotation speed", and realizes dynamic adaptation of mechanical parameters and material properties.

[0059] In the multi-stage gradient stirring module, the liquid level threshold triggers a hysteresis control method: when the first threshold triggers low-speed stirring, the liquid level needs to fall below 25% before stopping is allowed; a ±2% buffer band is set between the second and third thresholds; the first threshold is volume, the second threshold is volume, and the third threshold is volume; the duration of each stirring stage is dynamically adjusted according to the material properties:

[0060]

[0061] where μ is the viscosity, ρ is the density, k=0.08, b=5, and a hysteresis control method is used to set the liquid level threshold trigger logic. After the first threshold is triggered, the liquid level needs to fall below 25% before stopping is allowed, and a ±2% buffer band is set between the second / third thresholds to prevent frequent start / stop caused by liquid level fluctuations. Combined with the stirring stage time calculation formula (T=k⋅ρμ+b), the stage maintenance time is dynamically adjusted according to the material viscosity and density, which improves the stability of the stirring process, prolongs the service life of the equipment, and solves the parameter rigidity problem of traditional fixed threshold and time control.

[0062] In the multi-stage gradient stirring module, the time parameters for each stage are dynamically adjusted according to the liquid level rise rate. Specifically, when the liquid level rise rate is greater than 5cm / min, the stage maintenance time is shortened to 5-7 minutes; when the liquid level rise rate is ≤5cm / min, the stage maintenance time is extended to 8-10 minutes. The liquid level rise rate is calculated by the difference between two adjacent liquid level detection data and the detection interval time. The stage stirring time is dynamically adjusted according to the liquid level rise rate: when the rise rate is greater than 5cm / min, it is shortened to 5-7 minutes, and when it is ≤5cm / min, it is extended to 8-10 minutes. This mechanism links the stirring stage time with the material feed rate in real time, avoiding the risk of blockage caused by excessively fast feeding or energy waste caused by excessively slow feeding in a fixed time mode, thereby improving the system's processing efficiency and adaptability to working conditions.

[0063] In the real-time data twin processing module, the digital twin model is constructed by collecting at least 20 sets of liquid level, torque, and current sample data under different material characteristics to train the LSTM neural network. The rolling optimization time domain of the model predictive control (MPC) algorithm is set to 1 minute, and the control time domain is set to 0.5 minutes. When the predicted probability of material accumulation is greater than 70%, an adjustment instruction to increase the stirring speed by 10-20 rpm is automatically generated. The edge computing unit performs the following operations: Establish the material rheological state equation:

[0064]

[0065] in, is the shear stress, is the yield stress, is the consistency coefficient, is the shear rate, The system uses a Kalman filter to fuse current and torque signals, achieving a prediction accuracy of ±1.5%. When the predicted accumulation risk index exceeds 0.7, reverse pulse stirring (2 Hz frequency, 10% of rated speed) is added in advance. A digital twin model is constructed using LSTM neural network training and the MPC algorithm to collect level, torque, and current data for real-time predictive control, enabling the prediction of material accumulation 0.5-1 minute in advance. By combining the material rheological equation of state with the Kalman filter fusion algorithm, reverse pulse stirring is initiated when the predicted accumulation risk index exceeds 0.7, achieving a transition from "delayed response" to "forward-looking regulation," improving mixing uniformity and system reliability.

[0066] In the remote cooperative monitoring module, the sound and light alarm threshold is specifically set as: the liquid level abnormal fluctuation threshold ± 5% triggers when the actual liquid level deviation is greater than 10 cm; the fault positioning report includes real-time video stream (resolution 1280x720, frame rate 25 fps), sensor waveform graph and motor current curve; the remote terminal (mobile phone / PC) supports parameter fine adjustment of the system through the VPN network, the adjustment step is: speed ± 1 revolution / minute, medicament addition ratio ± 0.5%, 5G communication module is used to realize cloud synchronization of mixing data, sound and light alarm threshold of liquid level abnormal fluctuation ± 5% (corresponding to actual deviation > 10 cm) is set, and fault positioning report including real-time video stream, sensor waveform graph and motor current curve is pushed. The remote terminal supports fine adjustment of speed ± 1 revolution / minute and medicament ratio ± 0.5% through the VPN network, breaking the limitation of traditional local control, realizing remote operation and maintenance and rapid fault positioning of industrial Internet of Things level.

[0067] In the fault linkage processing module, the self-calibration program includes three levels of diagnosis: primary diagnosis: comparing the data difference between the capacitive sensor and the ultrasonic sensor, if |AH|>3%, starting temperature drift compensation; middle-level diagnosis: injecting test signal to verify the sensor response linearity (R 2 ≥0.98); advanced diagnosis: actual liquid level value is deduced by motor current spectrum analysis (FFT resolution 0.5 Hz); the three-level diagnosis mechanism of the fault linkage processing module: primary diagnosis compares the data of the double sensors and starts temperature drift compensation; middle-level diagnosis injects test signal to verify the sensor linearity; advanced diagnosis deduces the liquid level value by motor current spectrum analysis. When the main sensor deviates by >±3% for 3 times in succession, switch to the standby capacitive sensor within 50 ms and start self-calibration, and reduce the speed to the safety interval, to build a multi-level fault protection system and improve the system operation reliability.

[0068] In the energy-saving closed-loop control module, the residual material emptying program before entering the sleep mode is: when the liquid level is lower than the 10% threshold, control the variable frequency motor to continuously stir at 150-200 revolutions / minute for 3-5 minutes; after stirring, detect the material accumulation height in the feed hopper, if >5 cm, repeat the emptying and stirring once; the residual material emptying program of the energy-saving closed-loop control module: when the liquid level is lower than the 10% threshold, stir at 150-200 revolutions / minute for 3-5 minutes, and repeat emptying when the accumulation height is >5 cm. This mechanism solves the problem of residual material solidification in traditional systems, avoids equipment corrosion and blockage, reduces maintenance cost, and at the same time provides a clean equipment state for the sleep mode, ensuring the stability of the next start.

[0069] Further, in the energy-saving closed-loop control module, the motor preheating strategy includes: establishing a start-up power consumption model based on historical data:

[0070]

[0071] wherein, is the maximum start-up power peak, is the decay coefficient, is the steady state power, t is the time, the winding temperature maintenance current of 0.5% rated power is maintained during the hibernation, the speed is pre-boosted according to the S-shaped curve 3 minutes before the next start, the start-up power consumption model is established based on historical data, the winding temperature maintenance current of 0.5% rated power is maintained during the hibernation, and the speed is pre-boosted according to the S-shaped curve 3 minutes before the start. The strategy optimizes the start-up current waveform, reduces the instantaneous power consumption, shortens the time to reach the rated working condition, and realizes the energy consumption optimization and start-up performance improvement of the mixing system in the whole life cycle by combining the low-power standby of the hibernation mode.

[0072] Finally, it should be noted that: the above is only the preferred embodiment of the present application, and is not used to limit the present application, although the present application has been described in detail with reference to the foregoing embodiments, for those skilled in the art, the technical solutions recorded in the foregoing embodiments can still be modified, or some technical features can be replaced, any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. An electronic control system for mixing and feeding with liquid level sensing, characterized in that: include: The material property pre-detection module uses a multispectral sensor installed at the entrance of the feed hopper to perform real-time scanning of the particle size, moisture content, density and viscosity of the industrial waste entering the feed hopper, generating a data package of material property parameters; The adaptive parameter configuration module automatically matches the stirring model based on the material characteristic parameter data package using the fuzzy logic algorithm built into the PLC. It uses viscosity as the primary judgment basis and simultaneously sets the variable frequency motor initial speed, reagent dosage ratio, and stirring time combination strategy. The multi-stage gradient stirring module is used to start low-speed stirring when the liquid level sensor detects that the material has reached the first liquid level threshold, switch to medium-speed stirring when it reaches the second liquid level threshold, and start high-speed stirring when it reaches the third liquid level threshold. Each stage maintains gradient stirring for 5-10 minutes; A real-time data twin processing module connects real-time data on liquid level, motor current, and stirring torque to the edge computing unit to build a digital twin model of the mixing process. This module uses a model predictive control algorithm to predict material accumulation status 0.5-1 minute in advance and fine-tune the mixing strategy. The remote collaborative monitoring module synchronizes mixing data to the cloud management platform via the 5G communication module. When the abnormal liquid level fluctuation exceeds the ±5% threshold, it automatically triggers the sound and light alarm on the remote terminal and pushes a fault location report including real-time video streaming; Fault linkage processing module: When the liquid level sensor detects data deviation exceeding ±3% for three consecutive times, the system automatically switches to the backup capacitive sensor and starts the self-calibration process. At the same time, it reduces the motor speed to a safe range and resumes normal operation after the fault is repaired. When the torque exceeds 120% of the rated value, an emergency stop is triggered. Energy-saving closed-loop control module: When mixing is completed and the liquid level remains below the 10% volume threshold for 30 seconds, the system automatically enters a low-power standby sleep mode. It also optimizes the motor preheating strategy for the next startup based on historical 24-hour operating data. In the adaptive parameter configuration module, the synchronously set reagent dosing ratio control logic is to establish a three-dimensional mapping table of moisture content-viscosity-reagent dosage, which is stored in the PLC register. The reagent dosing port adopts electromagnetic flowmeter closed-loop control with a control accuracy of ±1.5% of the rated flow rate. The three-dimensional mapping table is constructed in the following way: When the viscosity is ≤5000cP, the Newtonian fluid agent dosing model is matched, and the agent dosage is set as follows: when the water content is ≤30%, the dosage is 15-20kg / m 3 When the moisture content is 30%<≤50%, the dosage is 20-25kg / m 3 ; When the moisture content is greater than 50%, the dosage shall be increased by 10% based on the basic value; When the viscosity is greater than 5000cP and the water content is less than 40%, the Bingham plastomer compensation model is matched and the dosing strategy is: when the viscosity is less than 5000cP and the water content is less than 8000cP, the dosage is set to 25-30kg / m 3 , and start the pulse dosing mode; when 8000cP<viscosity≤10000cP, the dosage is increased to 30-35kg / m 3 , and activate the medicine pre-mixing program at the same time; When the viscosity is greater than 10000 cP, the high viscosity agent enhancement strategy is activated. Specifically, when the moisture content is ≥40%, the dosage is 1.5 times the upper limit of the Newtonian fluid model; the dual agent dosing port collaborative working mode is simultaneously started, with the main dosing port adding 70% of the calculated value and the auxiliary dosing port adding 30% compensation; when the material moisture content is greater than 80%, the agent dosage ratio is automatically increased by 10%-15%; In the multi-stage gradient stirring module, the liquid level threshold trigger adopts the hysteresis control method: after the first threshold triggers low-speed stirring, the liquid level must drop below 25% before it is allowed to stop; a ±2% buffer zone is set between the second and third thresholds; and the first threshold is Volume, the second threshold is Volume, the third threshold is volume; The duration of each mixing stage is dynamically adjusted according to the material characteristics: Where μ is viscosity, ρ is density, k=0.08, and b=5.

2. The mixing and feeding electronic control system with liquid level sensing according to claim 1, characterized in that: In the adaptive parameter configuration module, the fuzzy logic algorithm has a built-in viscosity-speed mapping table, which is constructed in the following way: When the viscosity is ≤5000cP, match the Newtonian fluid model and set the speed to 50-80rpm; When the viscosity is greater than 5000 cP and the water content is less than 40%, the Bingham plastomer model is matched and the rotation speed is set to 120-180 rpm; When the viscosity is greater than 10,000 cP, the high torque compensation strategy is activated and the speed is increased to 200-300 rpm.

3. The mixing and feeding electronic control system with liquid level sensing according to claim 1, characterized in that: In the multi-stage gradient stirring module, the time parameters for each stage are dynamically adjusted according to the liquid level rising rate, specifically: When the liquid level rising rate is greater than 5 cm / min, the stage maintenance time is shortened to 5-7 minutes; When the liquid level rising rate is ≤5cm / min, the stage maintenance time is extended to 8-10 minutes; The liquid level rising rate is calculated by calculating the difference between two adjacent liquid level detection data and the detection interval time.

4. The mixing and feeding electronic control system with liquid level sensing according to claim 1, characterized in that: In the real-time data twin processing module, the digital twin model is constructed by collecting at least 20 sets of liquid level, torque, and current sample data under different material characteristics to train the LSTM neural network. The rolling optimization time domain of the model predictive control algorithm is set to 1 minute, and the control time domain is set to 0.5 minutes. When the predicted probability of material accumulation is greater than 70%, an adjustment instruction to increase the stirring speed by 10-20 rpm is automatically generated. The edge computing unit also performs the following operations: Establish the material rheological state equation: in, is the shear stress, is the yield stress, is the consistency coefficient, is the shear rate, is the rheological index; The current signal and torque signal are fused by the Kalman filter, and the prediction accuracy reaches ±1.5%; When the predicted accumulation risk index is greater than 0.7, reverse pulse stirring is increased in advance.

5. The mixing and feeding electronic control system with liquid level sensing according to claim 1, characterized in that: In the remote collaborative monitoring module, the sound and light alarm thresholds are specifically set as follows: The abnormal liquid level fluctuation threshold of ±5% corresponds to a trigger when the actual liquid level deviation is greater than 10cm; Fault location report includes real-time video stream, sensor waveform and motor current curve; The remote terminal supports fine-tuning of system parameters through the VPN network, with adjustment steps of: speed ±1 rpm, agent addition ratio ±0.5%.

6. The mixing and feeding electronic control system with liquid level sensing according to claim 1, characterized in that: In the fault linkage processing module, the self-calibration program includes three levels of diagnosis: Primary diagnosis: Compare the data difference between the capacitive sensor and the ultrasonic sensor. If |ΔH|>3%, start temperature drift compensation. Intermediate diagnosis: inject test signal to verify sensor response linearity; Advanced diagnosis: The actual liquid level value can be inferred through motor current spectrum analysis.

7. The mixing and feeding electronic control system with liquid level sensing according to claim 1, characterized in that: In the energy-saving closed-loop control module, the residual material clearing procedure before entering the sleep mode is as follows: When the liquid level is lower than the 10% threshold, the variable frequency motor is controlled to continuously stir at a speed of 150-200 rpm for 3-5 minutes; After stirring, check the material accumulation height in the feed hopper. If it is greater than 5cm, empty and stir again.

8. The mixing and feeding electronic control system with liquid level sensing according to claim 1, characterized in that: In the energy-saving closed-loop control module, the motor preheating strategy includes: Build a startup power consumption model based on historical data: in, is the maximum startup power consumption peak, is the attenuation coefficient, is the steady-state power, t is the time, and the winding insulation current of 0.5% of the rated power is maintained during the dormant period. The speed is pre-increased according to the S-shaped curve 3 minutes before the next start.

Images