Production equipment and method of concrete additive
By introducing parallel vibration screening mechanism, intelligent batching module, reaction control module and energy-saving and emission reduction module into concrete additive production equipment, the problems of low level of equipment automation and unstable product quality are solved, and efficient and stable production is achieved Process and environmentally friendly production methods.
Patent Information
- Application Number
- CN202411184278.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-27
- Publication Date
- 2024-11-22
AI Technical Summary
The existing concrete additive production equipment has low degree of automation, resulting in unstable product quality and poor process control, which increases labor costs and low production efficiency.
A production equipment including raw material processing module, intelligent batching module, reaction control module, quality control module and energy-saving and emission reduction module was designed, and a parallel vibration screening mechanism, RFID identification system, modified PID control algorithm, PLC controller, Technology such as pH meter, conductivity meter, intelligent image recognition algorithm, hot air circulation dryer and automated packaging robotic arms to achieve accurate processing of raw materials, automated ingredients, precise reaction control, online quality inspection and energy conservation and emission reduction.
It improves the degree of automation of production equipment, ensures the consistency and stability of product quality, reduces energy consumption and environmental pollution, and improves production efficiency and market competitiveness.
Smart Images

Figure BDA0005013089130000031 
Figure BDA0005013089130000041 
Figure BDA0005013089130000051
Abstract
Description
Technical Field
[0001] The present invention relates to the field of chemical automation, and more specifically to a production device and method for concrete additives. Background Art
[0002] With the rapid development of the global construction industry, the demand for high-quality concrete is growing. Especially in the fields of infrastructure construction and real estate development, the application of high-performance concrete is becoming more and more extensive. This not only promotes the development of the concrete additive industry, but also promotes the progress of related production equipment and methods. Since the beginning of the 21st century, with the continuous improvement of scientific and technological levels, the requirements for the performance of building materials have become higher and higher, especially for the strength, durability and construction performance of concrete. Therefore, the research and development of efficient concrete additives has become one of the key factors in improving the performance of concrete. At the same time, the concept of green environmental protection has gradually taken root in the hearts of the people, prompting the concrete additive industry to develop in a more environmentally friendly and energy-saving direction.
[0003] At present, the production of concrete additives mainly relies on a series of mechanical equipment, including mixing equipment, reactors, metering systems, storage tanks and packaging equipment. These equipment can realize the whole process from raw material proportioning to final product packaging. For example, mixing equipment can evenly mix multiple raw materials to ensure that the effective ingredients of additives are evenly distributed; reactors are suitable for additives that need to be synthesized through chemical reactions. However, although these production equipment and technologies are relatively mature, there are still some problems that need to be solved. First, although many production lines have achieved a certain degree of automation, a large amount of manual operation is still required in key links such as raw material handling, intermediate testing, and quality control, which not only increases labor costs but also reduces production efficiency. Secondly, due to the lack of fine process control and the uncertainty caused by manual operation, different batches of products may have quality fluctuations, which is a problem that cannot be ignored for construction projects that require strict consistency.
[0004] Therefore, the present invention discloses a production device and method for concrete additives to solve the shortcomings of low automation and unstable product quality in existing concrete additive production devices. Summary of the invention
[0005] In view of the deficiencies in the prior art, the present invention discloses a production device and method for a concrete additive, which improves the automation level and product quality of the existing concrete additive production equipment.
[0006] The present invention adopts the following technical scheme: a production device for concrete additives,
[0007] The equipment includes a raw material processing module, an intelligent batching module, a reaction control module, a quality control module, a drying and packaging module and an energy-saving and emission reduction module;
[0008] The raw material processing module is used to remove impurities and lumps in the additive raw material through a parallel vibration screening mechanism, and to convey the processed additive raw material to the intelligent batching module through a screw conveyor;
[0009] The intelligent batching module is used to realize the automated batching process; the intelligent batching module is provided with a plurality of independent silos, each of which is equipped with an RFID identification system. Based on the identification information and flow monitoring data, the intelligent batching module automatically controls the opening of the batching valve according to the preset formula ratio through a modified PID control algorithm;
[0010] The reaction control module is used to automatically control the feeding speed, stirring intensity and operation of the heating / cooling equipment of the raw materials through the PLC controller according to the preset process parameters; the reaction control module is also used to monitor the reaction state in real time through the feedback monitoring system, and transmit the real-time monitoring data to the PLC controller to form a closed-loop control; the real-time monitoring data at least includes the temperature, pressure and liquid level data parameters of the production site, and the process parameters include but are not limited to temperature, pressure and reaction time;
[0011] The quality control module is used to detect the pH value and solid content of the semi-finished additive product after the reaction is completed by a pH meter and a conductivity meter, and to assist in analyzing the neutralization effect by an intelligent image recognition algorithm. If the neutralization effect does not meet the preset standard, the system triggers an alarm and sends an adjustment instruction to the reaction control module; the preset standard at least includes color, texture and bubble distribution indicators; the adjustment instruction includes adjusting the reaction temperature, stirring speed and pH regulator dosage parameters; if the neutralization effect meets the preset standard, the semi-finished additive product after the reaction is sent to the drying and packaging module through a conveyor belt;
[0012] The drying and packaging module is used to perform drying treatment by a hot air circulation dryer, and to perform rapid packaging and labeling by an automated packaging robot arm; the automated packaging robot arm realizes product positioning and packaging by a visual recognition system;
[0013] The energy-saving and emission-reduction module is used to reduce energy consumption and emissions during the production process; the energy-saving and emission-reduction module includes a waste heat recovery submodule and a waste gas treatment submodule, the waste heat recovery submodule is used to recover the waste heat generated during the mixing and reaction process and preheat the next wave of raw materials; the waste gas treatment submodule is used to filter the dust generated during the production process.
[0014] As a further technical solution of the present invention, the parallel vibration screening mechanism screens different types of raw materials through a multi-channel screening unit, and the multi-channel screening unit screens different raw materials through multiple screening grids; then, the processed raw materials are transported to several independent silos in the intelligent batching module through an intelligent conveying mechanism and an RFID identification system; the intelligent conveying mechanism conveys the processed raw materials to the designated silos in steps according to the grids through an adjustable speed screw conveying method and a real-time monitoring sensor; the adjustable speed screw conveying method controls the motor speed through a frequency converter, and the real-time monitoring sensor monitors the flow state of the raw materials during the conveying process in real time through a sensor network, and feeds back the data in real time to the embedded control system; the real-time monitoring sensor includes a photoelectric sensor and a proximity switch; the embedded control system adjusts the conveying speed through the real-time data of the differential to ensure the accurate conveying of the raw materials; the RFID identification system realizes automatic identification and classification of raw materials through RFID tags and readers; each screening grid is equipped with an RFID tag to store raw material information; the raw material information at least includes the name, batch number and expiration date information, the raw material information is read by the reader, and the raw material information data is transmitted to the embedded control system through the wireless communication protocol; based on the raw material information data, the embedded control system controls the screw conveyor through the frequency converter to transport the processed raw materials to the designated silo.
[0015] As a further technical solution of the present invention, the modified PID control algorithm first uses the proportional term to respond to the deviation change, and adjusts the proportional gain through the deviation calculation to reflect the deviation size and produce a corresponding control effect; the deviation calculation realizes the integral action through the forgetting factor to gradually weaken the memory of the past deviation over time, and simulates the periodic change of external interference through periodic modulation. The modified PID control algorithm then performs nonlinear adjustment according to the sign and size of the deviation change rate through the nonlinear differential term to suppress the overshoot and oscillation of the system; in the intelligent batching module, the modified PID control algorithm also identifies the type of raw materials through the RFID identification system, and calculates the deviation of each raw material in combination with the real-time flow data; finally, the modified PID control algorithm also estimates and compensates the effect of external interference on the system in real time through the external interference compensation mechanism; the formula expression of the modified PID control algorithm for the batching valve opening control is:
[0016]
[0017] In formula (1), u t Represents the output signal of the controller, which is used to adjust the control quantity of the controlled object, K p , K i and K d Respectively represent proportional gain, integral gain and differential gain; e tIt represents the control deviation, that is, the difference between the set value and the actual value, which is used to reflect the gap between the current state of the system and the expected state; δ The modulation function of the integral term is used to adjust the cumulative effect of the integral term and improve the stability and response speed of the system; sgn represents the nonlinear modulation function of the differential term, which is used to make nonlinear adjustments according to the characteristics of the deviation change rate and enhance the system's adaptability to dynamic changes; t represents the current time; i represents the type of raw material; γ represents the measurement error, h and x represent the adjustment parameters, which are used to control the intensity of the integral action and the forgetting speed; θ represents the sampling interval, x represents the number of iterations, and β i Represents the weight coefficient of RFID and flow feedback, which is used to determine the relative importance of RFID identification and flow monitoring data in the control function; RFID i (t) represents the RFID identification value of the i-th raw material, which is used to provide accurate information about the type of raw material; Q i (t) represents the real-time flow rate of the i-th raw material at time t, Q i Indicates the corresponding real-time traffic Q i (t) is the set value, which is used to calculate the flow deviation; ε represents the external disturbance compensation coefficient, which is used to determine the weight of the external disturbance compensation term in the control action; Disturbance(t) represents the estimated value of the external disturbance.
[0018] As a further technical solution of the present invention, the working method of the reaction control module is:
[0019] Step 1: Capture the temperature, pressure and liquid level parameters of the production site through the ADC analog-to-digital converter, and pre-process the data through the filter and normalization method;
[0020] Step 2: Based on the preprocessed data, the reaction state is monitored in real time and comprehensively evaluated through the comprehensive strategy of multivariate statistical process control MSPC and deep learning model; the multivariate statistical process control MSPC is used to identify whether the production process deviates from the normal operating range, and the deep learning model is used to further classify and analyze the causes of abnormal states; the deep learning model is combined with the process knowledge base to analyze the impact of abnormal states on product quality;
[0021] Step 3: Based on the state evaluation data, construct the input data set F through the fuzzy control method sr ={e,e c , p, l} and the output dataset F sc = {v, s, g}; Input data set F sr In the equation, e represents the temperature deviation, e c represents the temperature deviation change rate, p represents pressure, and l represents liquid level; input data set F scIn the formula, v represents the feeding speed, s represents the stirring intensity, and g represents the heating or cooling power. Based on the expert experience and process knowledge, the fuzzy control method generates real-time control instructions through a dynamic adjustment function to improve the control accuracy. The formula of the dynamic adjustment function is:
[0022]
[0023] In formula (2), q ij represents the adjustment factor, z1, z2, z3 and z4 are weight coefficients, which are used to adjust the relative importance of temperature deviation, temperature deviation change rate, pressure deviation and liquid level deviation in the calculation of the adjustment factor; e represents the temperature deviation, e c Indicates the temperature deviation change rate, p and p target Respectively represent the current pressure value and the target pressure value to be achieved; l and l target Respectively represent the current liquid level value and the target liquid level value to be achieved; max|e|, max|e c |, max|p|, and max|l| represent the maximum absolute values of the deviations, which are used to normalize the deviation terms;
[0024] Step 4: The PLC controller adjusts the raw material feeding speed, stirring intensity, and the operating state of the heating or cooling equipment according to the received control instructions; at the same time, the PLC controller continuously collects new real-time data through the feedback monitoring system, and performs feedback evaluation through the control effect evaluation function, and the formula expression of the control effect evaluation function is:
[0025]
[0026] In formula (3), A(X) represents the control effect index, T c , P c and L c Respectively represent the temperature, pressure and liquid level values of the cth sampling point, which are used to evaluate the actual effect of the control strategy; T taeget , P taeget and L taeget Respectively represent the target values of temperature, pressure and liquid level; T range , P range and L range They represent the range of temperature, pressure and liquid level, respectively, and are used to normalize the deviation term; a represents the attenuation factor, which is used to control the speed of time attenuation and balance the needs between long-term stability and immediate response; represents the time of the cth sampling point; N represents the total number of sampling points;
[0027] Step 5. During the continuous production process, the reaction control module is also optimized and adjusted through a reinforcement learning framework and an adaptive control mechanism to cope with changes and uncertainties in production conditions; the reinforcement learning framework continuously optimizes the control strategy by defining a state space, an action space, and a reward function; the adaptive control mechanism automatically adjusts control parameters based on real-time feedback data and historical data by comparing the difference between the predicted state and the actual state to respond to changes in production conditions; the state space includes but is not limited to temperature, pressure, and liquid level parameters; the action space includes adjusting the feeding rate, stirring intensity, and heating or cooling power; the reward function includes but is not limited to a comprehensive evaluation based on product quality, production efficiency, and energy consumption indicators.
[0028] As a further technical solution of the present invention, the working method steps of the intelligent image recognition algorithm are as follows:
[0029] S1, image preprocessing and feature extraction;
[0030] The captured images are preprocessed by denoising, contrast enhancement and color correction, and then the color distribution, texture pattern, shape and distribution density of bubbles are extracted from the images by edge detection and texture segmentation methods;
[0031] S2, color space conversion and distribution quantization;
[0032] The deviation of the color in the image from the ideal neutralization effect is quantified by a color analysis function; the color analysis function evaluates the closeness of the overall color by calculating the weighted sum of the Euclidean distance between each color and the reference color; the formula expression of the color analysis function is:
[0033]
[0034] In formula (4), N c Indicates the number of color categories in the color space, which is used to measure the division interval of the color space during color quantization; C rj Represents the color vector of the jth pixel in the rth color class; C ref Indicates the ideal color when neutralization is complete; area rj A represents the area of the region to which the jth pixel in the rth color belongs; toe Represents the total area of the image, used to calculate the relative proportion of color distribution; weighet r The weight of the r-th color is used to reflect the importance of the r-th color in the evaluation;
[0035] S3, texture and bubble distribution analysis;
[0036] After extracting the color and texture features, the intelligent image recognition algorithm further analyzes the bubble distribution through the bubble distribution analysis function to evaluate the complexity and uniformity of the bubble distribution; the bubble distribution includes the size, shape, density and spatial distribution pattern of the bubbles; the formula expression of the bubble distribution analysis function is:
[0037]
[0038] In formula (5), N bubbles Indicates the number of bubbles detected in the image, which is used to evaluate the uniformity of bubble distribution; area k Represents the area of the kth bubble, which is used to evaluate the bubble distribution density and total proportion; area max represents the maximum bubble area in the image, which is used to normalize the bubble area; p k represents the perimeter of the kth bubble; DL k represents the difference of the kth bubble shape based on Hu moment, which is used to quantify the regularity of the bubble shape;
[0039] S4, comprehensive assessment and decision-making;
[0040] Based on the analysis results of the above color and texture features, the intelligent image recognition algorithm guides the neutralization effect of the additive semi-finished product through a comprehensive evaluation score, and compares the comprehensive evaluation score with a preset threshold through a rule engine to trigger an alarm or command adjustment operation; the calculation formula of the comprehensive evaluation score is:
[0041]
[0042] In formula (6), and σ represent the weight coefficients of color quantization and bubble complexity index, respectively, which are used to measure the importance of color quantization and bubble complexity index in the evaluation; CQ max and BCI max They represent the maximum values of color quantization and bubble complexity index respectively, and are used for normalization.
[0043] As a further technical solution of the present invention, the visual recognition system realizes accurate recognition and rapid packaging of dried products through image processing algorithms and robot motion control protocols; the image processing algorithm extracts contour information of products in complex backgrounds through edge detection and feature matching methods; the robot motion control protocol ensures the smoothness and continuity of the robot arm's movements through path planning algorithms and real-time scheduling mechanisms, and also improves packaging efficiency through multi-task parallel processing.
[0044] As a further technical solution of the present invention, the waste heat recovery submodule includes a high-efficiency heat exchange unit, a phase change energy storage unit and an intelligent control unit; the high-efficiency heat exchange unit transfers heat in the high-temperature exhaust gas through heat pipe heat exchange; the phase change energy storage unit stores high-temperature waste heat through phase change materials, and the phase change materials undergo phase change when absorbing heat, changing from solid to liquid, storing a large amount of latent heat, and when releasing heat, the phase change materials change from liquid to solid, releasing the stored heat; the intelligent control unit intelligently schedules and optimizes the stored waste heat through an embedded operating system and an optimization algorithm, and the embedded operating system monitors at least temperature and pressure parameters in the production process in real time through a real-time data processing method and a sensor network; the optimization algorithm uses a model predictive control method according to real-time data The working state of the heat exchanger is adjusted dynamically according to the working state of the heat exchanger; the exhaust gas treatment submodule includes a dust capture unit, an exhaust gas purification unit and an emission monitoring unit; the dust capture unit captures the dust generated in the production process through electrostatic dust removal and cyclone separation, and the electrostatic dust removal charges the dust through a high-voltage electrostatic field, and adsorbs the charged dust on the dust collecting plate through the electric field force, and then separates the dust from the airflow through the cyclone separation; the exhaust gas purification unit purifies the harmful substances in the exhaust gas through catalytic oxidation and activated carbon adsorption; the emission monitoring unit monitors the emission of the purified exhaust gas in real time through online monitoring and data transmission protocols, and transmits the monitoring data to the central control system for analysis and recording, and the online monitoring monitors the pollutant concentration in the exhaust gas in real time through spectral analysis and chemical sensors.
[0045] As a further technical solution of the present invention, the device also includes a data management and analysis module for automatically collecting, storing and analyzing production data; the data management and analysis module includes a data acquisition submodule, a data storage submodule, a data analysis submodule and a data visualization submodule; the data acquisition submodule includes a sensor interface unit, a data preprocessing unit and a data fusion unit; the sensor interface unit transmits sensor data to the embedded system through the serial communication protocol RS-485; the data preprocessing unit realizes preliminary processing of sensor data through Kalman filtering and outlier detection; the data fusion unit realizes fusion of different sensor data through Bayesian estimation; the data storage submodule includes a database management unit, a data backup unit and a data recovery unit; the database management unit The meta-module realizes the storage and management of production data through NoSQL database; the data backup unit realizes regular backup of database data through cron job scheduling and incremental backup; the data recovery unit ensures the integrity and availability of production data through disaster recovery plan DRP; the data analysis submodule includes real-time data analysis unit, historical data analysis unit and predictive maintenance unit; the real-time data analysis unit provides instant production status monitoring through stream processing framework; the historical data analysis unit discovers laws and problems in the production process through potential causal model and graph database; the predictive maintenance unit predicts the probability of equipment failure according to historical maintenance records through reinforcement learning framework; the data visualization submodule realizes intuitive display of production data through data visualization tool D3.js and document generation engine.
[0046] As a further technical solution of the present invention, a method for producing a concrete additive comprises the following steps:
[0047] Step 1: Start the raw material processing module, pour the additive raw material into the vibration screening machine, and remove impurities and lumps in the raw material through high-frequency vibration;
[0048] Step 2: The additive raw materials after screening are automatically transported to each independent silo of the intelligent batching module by a screw conveyor. Each silo is equipped with an RFID identification system to automatically record and identify the raw material information;
[0049] Step 3: The intelligent batching module confirms the type and batch of raw materials through the RFID identification system according to the preset formula ratio, and automatically adjusts the opening of the batching valve using the modified PID control algorithm in combination with the flow monitoring data to achieve accurate batching;
[0050] Step 4: The raw materials are put into the reactor in proportion, the reactor is heated and dissolved by automatically adjusting the heater through the PID controller, and the initiator solution is added dropwise at a uniform speed through the flow control valve to react;
[0051] Step 5: During the reaction process, the temperature, pressure and liquid level parameters of the production site are collected in real time through the feedback monitoring system and transmitted to the PLC controller to dynamically adjust and control the feeding speed, stirring intensity, and the operation of the heating and cooling equipment;
[0052] Step 6: After the reaction is completed, 50% NaOH solution is automatically added for neutralization to adjust the pH value to 7-8;
[0053] Step 7: After neutralization, the reactor is cleaned by an automatic cleaning system;
[0054] Step 8: The pH value and solid content of the semi-finished additive product are tested by a pH meter and a conductivity meter, and the neutralization effect is analyzed by an intelligent image recognition algorithm, including color, texture and bubble distribution index parameters. If the neutralization effect does not meet the preset standard, the system triggers an alarm and automatically sends an adjustment instruction to the reaction control module; the adjustment instruction includes adjusting the reaction temperature, stirring speed and pH regulator dosage parameters; if the neutralization effect meets the preset standard, the semi-finished additive product after the reaction is sent to the drying and packaging module through a conveyor belt;
[0055] Step nine: The dry packaging module is dried by a hot air circulation dryer, and is quickly packaged and labeled by an automated packaging robot.
[0056] As a further technical solution of the present invention, the automatic cleaning system includes a cleaning agent distribution module and a cleaning execution module; the cleaning agent distribution module includes a quantitative delivery unit and an automatic switching valve unit; the quantitative delivery unit delivers cleaning agent through a metering pump and flow control, and the metering pump is driven by a stepping motor; the flow control controls the opening and closing of the delivery valve through a solenoid valve; the cleaning execution module realizes strong spraying of the inner and outer surfaces of the reactor through a high-pressure pump and a nozzle, and the cleaning execution module also realizes the removal of fine particles on the surface of the reactor through ultrasonic vibration and cavitation effect; the ultrasonic vibration converts electrical energy into high-frequency vibration through a transducer, and the cavitation effect removes stubborn stains through the shock waves generated by the formation and rupture of bubbles.
[0057] Positive beneficial effects
[0058] The present invention discloses a production device and method for concrete additives. The present invention realizes efficient processing and automatic conveying of additive raw materials through a parallel vibration screening mechanism and a screw conveyor. The parallel vibration screening mechanism can quickly remove impurities and lumps in the raw materials, ensuring the purity of the raw materials; the screw conveyor automatically delivers the processed raw materials to the intelligent batching module through continuous conveying, reducing the need for manual operation and improving the automation level of raw material processing. Accurate identification and automatic batching of multiple raw materials are realized through an RFID identification system and a modified PID control algorithm. The RFID identification system ensures accurate reading of raw material information and avoids batching errors caused by human operation errors; the modified PID control algorithm realizes accurate addition of each raw material according to a preset formula ratio by dynamically adjusting the opening of the batching valve, improving the batching accuracy, thereby ensuring the consistency of product quality. Accurate control of the reaction process is realized through a PLC controller and a fuzzy control strategy. The PLC controller automatically controls the feeding speed, stirring intensity and operation of the heating / cooling equipment of the raw materials according to the preset process parameters, ensuring the consistency of the reaction conditions; the fuzzy control strategy ensures the precise control of the temperature in the reactor by real-time monitoring of the reaction state and dynamically adjusting the control strategy according to data such as temperature, pressure and liquid level, further improving the stability of product quality. The quality inspection and neutralization effect evaluation of the semi-finished additive products after the reaction are completed are realized through the pH meter, conductivity meter and intelligent image recognition algorithm. The pH meter and conductivity meter ensure that the pH value and solid content of the semi-finished additive products meet the standards; the intelligent image recognition algorithm further verifies whether the neutralization effect meets the standards by analyzing indicators such as color, texture and bubble distribution. If the standards are not met, the system will trigger an alarm and send an adjustment instruction to the reaction control module, so as to correct the deviation in the production process in time and ensure the consistency of product quality. The waste heat recovery submodule and the waste gas treatment submodule are used to effectively treat the waste heat and dust generated in the production process. The waste heat recovery submodule recovers the waste heat generated during the mixing and reaction process through heat pipe heat exchange technology and phase change material energy storage technology, and uses it to preheat the next wave of raw materials, significantly reducing energy consumption; the waste gas treatment submodule filters the dust generated during the production process through electrostatic dust removal technology and catalytic oxidation technology, reducing environmental pollution. The hot air circulation dryer and automated packaging robot arm achieve efficient drying and rapid packaging of additive semi-finished products. The hot air circulation dryer ensures the uniformity and efficiency of the drying process through efficient heat exchange technology; the automated packaging robot arm achieves product positioning and packaging through a visual recognition system, improving packaging speed and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0059] Figure 1 It is a schematic diagram of the overall structure of a production equipment for concrete additives of the present invention;
[0060] Figure 2 It is a schematic diagram of the principle framework of the modified PID control algorithm of the present invention;
[0061] Figure 3 Schematic diagram of the principle steps of the intelligent image recognition algorithm of the present invention;
[0062] Figure 4 A workflow framework diagram of a production device for a concrete additive according to the present invention;
[0063] Figure 5 This is a flow chart of the working framework of the intelligent image recognition algorithm of the present invention. DETAILED DESCRIPTION
[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0065] In the above embodiment, if Figure 1-Figure 5 As shown: a production equipment for concrete additives, the equipment includes a raw material processing module, an intelligent batching module, a reaction control module, a quality control module, a drying and packaging module and an energy-saving and emission reduction module;
[0066] The raw material processing module is used to remove impurities and lumps in the additive raw material through a parallel vibration screening mechanism, and to convey the processed additive raw material to the intelligent batching module through a screw conveyor;
[0067] In a specific embodiment, the raw material processing module equipment includes: a multi-channel screening unit, a vibration source and a screen; the multi-channel screening unit is composed of multiple independent screening grids, each of which performs customized screening for different types of raw materials, realizes parallel processing, and improves screening efficiency. The vibration source is a vibration motor, which provides the excitation force required for the screen surface, so that the raw material particles can jump, roll and other movements on the screen surface to achieve separation. The screen selects a suitable screen aperture according to the particle size requirements of different raw materials to ensure the effective separation of impurities and qualified raw materials. It also includes a screw conveyor, a real-time monitoring sensor, an embedded control system, an RFID identification system and an independent silo; the screw conveyor adopts an adjustable speed design, and the motor speed is controlled by a frequency converter to realize flexible adjustment of the raw material conveying speed. The real-time monitoring sensor includes a photoelectric sensor and a proximity switch, which are arranged along the conveying path to monitor the flow state of the raw materials in real time to ensure the safety and stability of the conveying process. As the control core of the entire conveying mechanism, the embedded control system receives the data fed back by the sensor and adjusts the conveying speed or issues an abnormal warning according to the preset algorithm. The RFID identification system is equipped with RFID tags in each screening grid and independent silo to store information such as the name, batch number and expiration date of the raw materials. The information in the RFID tag is read by the RFID reader, and the data is transmitted to the embedded control system through the wireless communication protocol to realize the automatic identification and classification of the raw materials. In addition, there are several independent silos for storing different types of processed raw materials. Each silo is equipped with a control valve and a metering device for the subsequent automated batching process.
[0068] The specific working method is: collect the operating data of the vibration screening mechanism and the screw conveyor in real time through the sensor network, including vibration frequency, amplitude, conveying speed, raw material flow, etc. And process and analyze the collected data, identify abnormal conditions (such as blockage, overspeed, etc.), and generate corresponding control instructions. By reading and storing the raw material information in the RFID tag. According to the read raw material information, the processed raw materials are distributed to the corresponding independent silos to ensure the accurate classification of the raw materials. Then, the intelligent control algorithm such as the modified PID control algorithm is introduced to automatically control the opening of the batching valve according to the preset formula ratio and real-time monitoring data to achieve accurate batching of raw materials. At the same time, according to the raw material flow state data during the conveying process, the speed of the screw conveyor is adjusted through the embedded control system to ensure continuous and uniform conveying of the raw materials. In addition, when an abnormal situation is detected (such as screening blockage, conveying failure, etc.), an alarm signal is immediately issued, and the working state is automatically adjusted or other emergency measures are taken. At the same time, a fault log recording function is provided to record the detailed information and treatment methods of the abnormal situation, providing a basis for subsequent troubleshooting and maintenance.
[0069] In the concrete additive production equipment, the raw material processing module adopts a parallel vibration screening mechanism. Its technical principle is based on the vibration screening theory, that is, the high-frequency vibration of the screen surface is used to make the raw material particles jump, roll and other movements on the screen surface, so as to achieve relative movement and stratification between particles, and then achieve the purpose of separation by particle size. The vibration screening mechanism optimizes the screening efficiency and accuracy by adjusting parameters such as vibration frequency, amplitude and screen surface inclination, ensuring that impurities of different particle sizes are effectively separated from qualified raw materials. The working mode of the raw material processing module is highly automated and parallel and efficient. First, the raw material enters the parallel vibration screening system through the feed port. The system consists of multiple independently operated multi-channel screening units, each of which performs customized screening for specific types of raw materials. During the screening process, the raw material is subjected to the exciting force on the screen surface, and large impurities are quickly separated onto the screen, while fine particles that meet the requirements fall into the collection device below through the screen holes. In this process, each screening unit works independently without interfering with each other, which greatly improves the processing efficiency.
[0070] Subsequently, the processed raw materials are further processed and distributed through the intelligent conveying mechanism. The intelligent conveying mechanism adopts an adjustable speed screw conveying method, and the motor speed is accurately controlled by the frequency converter to achieve flexible adjustment of the raw material conveying speed. At the same time, real-time monitoring sensors (such as photoelectric sensors and proximity switches) are arranged along the conveying path to monitor the flow status of the raw materials in real time, including flow, speed and blockage, etc., to ensure the safety and stability of the conveying process. Once an abnormal situation is detected, the sensor immediately feeds back the signal to the embedded control system, and the system immediately makes corresponding adjustments, such as adjusting the conveying speed or activating the alarm mechanism.
[0071] Compared with similar hardware, the combination of adjustable-speed spiral conveying method and real-time monitoring sensor in this raw material processing module realizes precise control and abnormal warning of raw material transportation, and improves the stability and safety of the production process. In addition, the application of RFID identification system realizes the automatic identification and classification of raw materials, reduces manual intervention, and improves production efficiency and accuracy. Secondly, the introduction of embedded control system puts the entire raw material processing process in a closed-loop control state, which can adjust the working state in real time and optimize the production process. At the same time, the modular design makes the system easy to expand and upgrade, and the screening unit and conveying path can be flexibly adjusted according to production needs to meet the production needs of different scales and types.
[0072] The intelligent batching module is used to realize the automated batching process; the intelligent batching module is provided with a plurality of independent silos, each of which is equipped with an RFID identification system. Based on the identification information and flow monitoring data, the intelligent batching module automatically controls the opening of the batching valve according to the preset formula ratio through a modified PID control algorithm;
[0073] Furthermore, the modified PID control algorithm first responds to the deviation change using the proportional term, and adjusts the proportional gain through deviation calculation to reflect the deviation size and produce a corresponding control effect; the deviation calculation realizes the integral action through the forgetting factor to gradually weaken the memory of the past deviation over time, and simulates the periodic change of external interference through periodic modulation. The modified PID control algorithm then performs nonlinear adjustment according to the sign and size of the deviation change rate through the nonlinear differential term to suppress the overshoot and oscillation of the system; in the intelligent batching module, the modified PID control algorithm also identifies the type of raw materials through the RFID identification system, and calculates the deviation of each raw material in combination with real-time flow data; finally, the modified PID control algorithm also estimates and compensates for the effect of external interference on the system in real time through the external interference compensation mechanism; the formula expression of the modified PID control algorithm for the batching valve opening control is:
[0074]
[0075] In formula (1), u t Represents the output signal of the controller, which is used to adjust the control quantity of the controlled object, K p , K i and K d Respectively represent proportional gain, integral gain and differential gain; e t It represents the control deviation, that is, the difference between the set value and the actual value, which is used to reflect the gap between the current state of the system and the expected state; δ The modulation function of the integral term is used to adjust the cumulative effect of the integral term and improve the stability and response speed of the system; sgn represents the nonlinear modulation function of the differential term, which is used to make nonlinear adjustments according to the characteristics of the deviation change rate and enhance the system's adaptability to dynamic changes; t represents the current time; i represents the type of raw material; γ represents the measurement error, h and x represent the adjustment parameters, which are used to control the intensity of the integral action and the forgetting speed; θ represents the sampling interval, x represents the number of iterations, and β i Represents the weight coefficient of RFID and flow feedback, which is used to determine the relative importance of RFID identification and flow monitoring data in the control function; RFID i (t) represents the RFID identification value of the i-th raw material, which is used to provide accurate information about the type of raw material; Q i (t) represents the real-time flow rate of the i-th raw material at time t, Q i Indicates the corresponding real-time traffic Q i (t) is the set value, which is used to calculate the flow deviation; ε represents the external disturbance compensation coefficient, which is used to determine the weight of the external disturbance compensation term in the control action; Disturbance(t) represents the estimated value of the external disturbance.
[0076] In a specific embodiment, a forgetting factor is introduced into the integral term, which decays exponentially with the increase of time, so that the integral term can gradually "forget" the past deviation, thereby avoiding the integral saturation phenomenon and improving the stability and response speed of the system. In addition, by adding a periodically changing modulation factor, the periodic interference that may be encountered in actual production is simulated, so that the integral term can adapt to these changes more flexibly and improve the robustness of the system. In addition, a nonlinear part is added on the basis of the traditional differential term. This part is nonlinearly adjusted according to the sign and size of the deviation change rate, which can provide a stronger damping effect when the deviation change rate is large, thereby suppressing the overshoot and oscillation of the system and improving the dynamic performance of the system. The intelligent batching module combines RFID identification technology with real-time flow monitoring through the deep integration of RFID and flow feedback, and realizes accurate monitoring of raw material types and flow rates. This deep integration not only improves the automation of the batching process, but also enables PID control to more accurately reflect the changes in the state of raw materials, thereby achieving more accurate batching control. External interference compensation mechanism. In the intelligent batching process, external interference may come from various factors such as changes in raw material properties, equipment wear, and ambient temperature. By introducing external disturbance compensation items, the algorithm can estimate and compensate for the impact of these disturbances on the system in real time, ensuring the stability and accuracy of the batching process. This modified PID control algorithm not only retains the advantages of the traditional modified PID control algorithm, but also significantly improves the stability, response speed, dynamic performance and robustness of the system by introducing innovative elements such as integral items with forgetting factors, periodically modulated integral items, nonlinear differential items, and deep integration of RFID and flow feedback. In the intelligent batching module, the algorithm can achieve precise control and compensation of raw material types, flow rates and external disturbances, providing an efficient and reliable solution for the field of industrial automation control.
[0077] In the specific implementation, the working mode of the modified PID control algorithm is as follows: the raw material type and flow data are collected in real time through the RFID identification system and the flow sensor. The RFID identification value provides accurate information about the raw material type, and the flow data is compared with the set value to calculate the flow deviation. Then, the deviation value is input into the PID controller, and the initial control effect is first generated through the proportional term; then, the integral cumulative effect is adjusted by the modulation function of the integral term and the forgetting factor; finally, further adjustment is made through the nonlinear differential term according to the rate of change of the deviation. The outputs of the three links are weighted and summed to obtain the final control signal. Then, the control signal is transmitted to the batching valve control system to adjust the opening of the batching valve to achieve precise control of the raw material flow. Finally, through the external interference compensation mechanism, the size of the external interference is estimated in real time, and it is included in the control calculation according to the external interference compensation coefficient to reduce the impact of the external interference on the system. In the production process of concrete additives, the working state of the modified PID control algorithm presents dynamic balance and continuous adjustment. As the production process proceeds, the system continuously receives new data input and calculates the control output in real time. When the system is in a stable state, the deviation value approaches zero, the opening of the batching valve remains in the optimal position, and the raw material flow rate accurately matches the set value, achieving an efficient and stable production process.
[0078] Specifically, the design experiment 1 verifies the control effect of the improved PID control algorithm in the concrete additive production method and equipment: Among them, the experimental environment selects the same concrete additive production equipment as the experimental platform to ensure the consistency of the experimental conditions. The traditional PID control algorithm and the improved PID control algorithm are deployed respectively to control the same production process. The test indicators include determining the control accuracy (deviation average), stability (standard deviation), response speed (time to reach steady state), overshoot, anti-interference ability (recovery time after external interference), RFID recognition accuracy and flow control accuracy as test indicators. The experimental method is: 1. Set the set value of the production process and calibrate the sensor. 2. Data acquisition: Start the equipment and record the test indicator data under the two control algorithms. 3. Interference simulation: artificially introduce external interference in the production process, observe and record the recovery time and control effect of the system. 4. Data analysis: process and analyze the collected data, and calculate the average value, standard deviation and other statistics of each indicator.
[0079] Table 1 Data comparison test table
[0080]
[0081] As shown in Table 1, the improved PID significantly reduces the average deviation of the control process through a more sophisticated adjustment mechanism, from 0.5% to 0.2%, reflecting higher control accuracy. Standard deviation is an important indicator for measuring system stability. The improved PID effectively reduces system fluctuations through forgetting factor and periodic modulation, and the standard deviation is reduced from 0.15% to 0.08%, making the system more stable. The nonlinear differential term of the improved PID enhances the system's adaptability to dynamic changes, allowing the system to reach a steady state faster, and the response speed is increased by 25%. Overshoot is an excessive deviation that may occur before the control system reaches a steady state. The improved PID effectively suppresses overshoot through nonlinear adjustment, and the overshoot is reduced from 1.2% to 0.5%, reducing system oscillation. The external interference compensation mechanism enables the improved PID to recover to a stable state faster when encountering external interference, and the recovery time is shortened from 15 seconds to 10 seconds, improving the system's anti-interference ability. Traditional PID does not involve RFID identification, while the improved PID integrates an RFID identification system to achieve accurate identification of raw material types with an accuracy rate of up to 99.9%, providing reliable protection for intelligent batching.
[0082] The reaction control module is used to automatically control the feeding speed, stirring intensity and operation of the heating / cooling equipment of the raw materials through the PLC controller according to the preset process parameters; the reaction control module is also used to monitor the reaction state in real time through the feedback monitoring system, and transmit the real-time monitoring data to the PLC controller to form a closed-loop control; the real-time monitoring data at least includes the temperature, pressure and liquid level data parameters of the production site, and the process parameters include but are not limited to temperature, pressure and reaction time;
[0083] In a specific embodiment, the working mode of the reaction control module follows the cycle of "setting-execution-monitoring-adjustment". First, according to the production process requirements, the process parameters such as temperature, pressure, reaction time, etc. are set in the PLC controller. Subsequently, the PLC automatically controls the feeding speed of the raw materials, the stirring intensity, and the operation of the heating / cooling equipment according to these parameters. At the same time, the feedback monitoring system collects data such as temperature, pressure, and liquid level at the production site in real time, and transmits these data to the PLC controller. The PLC controller uses fuzzy logic algorithms to analyze and process these data, and dynamically adjusts the control strategy according to preset control rules and expert experience to maintain the stability of the production process and optimize the production effect.
[0084] Specifically, multivariate statistical process control (MSPC) is a statistical method that detects abnormal conditions in the process by comprehensively analyzing multiple variables. In this process, MSPC can help identify whether the production process deviates from the normal operating range, so that timely measures can be taken to prevent quality problems. The deep learning model learns the complex relationship between input and output through a large amount of training data, which can then be used to classify and analyze the causes of abnormal conditions. On this basis, combined with the process knowledge base, the deep learning model can evaluate the impact of abnormal conditions on product quality, thereby providing a basis for subsequent control decisions. Fuzzy control theory is a nonlinear control method that imitates human thinking and handles uncertainty and complexity by defining fuzzy rules. Here, the fuzzy control method dynamically adjusts control parameters based on expert experience and process knowledge, improving the adaptability and robustness of the control system. Reinforcement learning is a machine learning method that learns the optimal behavior pattern through trial and error. In this framework, the system optimizes the control strategy through interaction with the environment, so that it can adapt to changing production conditions and ultimately achieve the best control effect.
[0085] Compared with traditional manual control or simple automatic control, the reaction control module can achieve precise control of complex production processes through fuzzy logic control and dynamic adjustment functions, ensuring that production parameters are always within the ideal range. In addition, through the close cooperation of the real-time monitoring system and the PLC controller, the system can respond to changes in the production process in real time and make adjustments through a closed-loop control mechanism to maintain the stability and continuity of the production process. Secondly, through the control effect evaluation function, the system can continuously optimize the control strategy, improve the control effect, reduce energy consumption and waste of raw materials, and thus reduce production costs. At the same time, combined with the process knowledge base for comprehensive evaluation, it is helpful to comprehensively consider the optimization goals of multiple aspects such as production efficiency, product quality and energy consumption, and ensure that the production process is both efficient and economical. In addition, by accurately controlling the reaction conditions, the system can ensure the consistency of product quality, reduce the production of unqualified products, and improve the market competitiveness of enterprises.
[0086] Specifically, formula (2) is developed based on the PID (proportional-integral-differential) control concept in classical control theory, but more complex factors are added on this basis, such as temperature deviation change rate, pressure deviation, liquid level deviation, etc., in order to better adapt to the complex production environment of multiple variables. This formula takes into account the simultaneous changes of multiple variables such as temperature, pressure, and liquid level, and normalizes the deviation term so that data of different dimensions can be compared with each other, thereby obtaining a comprehensive adjustment factor to guide the next control operation. By real-time monitoring of multiple key parameters in the production process and dynamically adjusting the control strategy according to the deviation of these parameters, the reaction conditions are always ensured to be in the best state. In specific implementation, formula 2 can adjust the reaction conditions more accurately by comprehensively considering multiple variables, avoiding the instability caused by single parameter adjustment. In addition, since the temperature deviation change rate is included, the system can respond to the temperature change trend more quickly. Secondly, by considering all relevant variables instead of focusing on only one or a few variables, global optimal control is achieved. Formula (3) is based on the concept of performance index, which is used to evaluate the performance of the control system. Formula (3) adds a time decay factor on this basis, so that the evaluation function can better reflect the change of control effect over time. By introducing the time decay factor, this formula makes the impact of early deviations on the control effect weaken over time, which can ensure both immediate response and long-term stability. Formula (3) is used to quantitatively evaluate the actual effect of the control strategy. In specific implementation, by incorporating the deviations of parameters such as temperature, pressure, and liquid level into the evaluation range, the effect of the control strategy can be fully understood. Through the time decay factor, the control system can strike a balance between immediate response and long-term stability. In addition, the evaluation function can be used as part of the feedback mechanism to help continuously optimize the control strategy to achieve better control results.
[0087] At the same time, Experiment 2 is designed to verify the effectiveness of the reaction control module in the production of concrete additives, and to compare the performance differences between the traditional PID control algorithm and the reaction control module. The experiment is divided into two stages: the baseline test uses the traditional PID control algorithm, and the advanced test uses this reaction control module; the experimental process is: 1. Capture the temperature, pressure and liquid level parameters of the production site through the ADC analog-to-digital converter. 2. Use filters and normalization methods to preprocess the data. 3. Combine MSPC and deep learning models to monitor and evaluate the reaction state in real time. 4. Use fuzzy control methods and dynamic adjustment functions to generate real-time control instructions. 5. The PLC controller adjusts the operating status of the production equipment according to the control instructions. 6. Feedback evaluation is performed through the control effect evaluation function to continuously optimize the control strategy. The data table is shown in Table 2:
[0088] Table 2 Experimental data record table
[0089]
[0090]
[0091] As can be seen from Table 2, the improved reaction control module significantly improves control accuracy and stability through more sophisticated data processing and advanced control strategies. In addition, through real-time monitoring and rapid decision-making, the improved module reduces the time required to reach steady state and significantly reduces overshoot. Secondly, the reinforcement learning framework and adaptive control mechanism enable the system to recover from external disturbances more quickly. At the same time, the combination of MSPC and deep learning models significantly improves the accuracy of abnormal state identification and cause analysis.
[0092] The quality control module is used to detect the pH value and solid content of the semi-finished additive product after the reaction is completed by a pH meter and a conductivity meter, and to assist in analyzing the neutralization effect by an intelligent image recognition algorithm. If the neutralization effect does not meet the preset standard, the system triggers an alarm and sends an adjustment instruction to the reaction control module; the preset standard at least includes color, texture and bubble distribution indicators; the adjustment instruction includes adjusting the reaction temperature, stirring speed and pH regulator dosage parameters; if the neutralization effect meets the preset standard, the semi-finished additive product after the reaction is sent to the drying and packaging module through a conveyor belt;
[0093] The drying and packaging module is used to perform drying treatment by a hot air circulation dryer, and to perform rapid packaging and labeling by an automated packaging robot arm; the automated packaging robot arm realizes product positioning and packaging by a visual recognition system;
[0094] The energy-saving and emission-reduction module is used to reduce energy consumption and emissions during the production process; the energy-saving and emission-reduction module includes a waste heat recovery submodule and a waste gas treatment submodule, the waste heat recovery submodule is used to recover the waste heat generated during the mixing and reaction process and preheat the next wave of raw materials; the waste gas treatment submodule is used to filter the dust generated during the production process.
[0095] Furthermore, a method for producing a concrete additive is applied to a production device for a concrete additive, comprising the following steps:
[0096] Step 1: Start the raw material processing module, pour the additive raw material into the vibration screening machine, and remove impurities and lumps in the raw material through high-frequency vibration;
[0097] Step 2: The additive raw materials after screening are automatically transported to each independent silo of the intelligent batching module by a screw conveyor. Each silo is equipped with an RFID identification system to automatically record and identify the raw material information;
[0098] Step 3: The intelligent batching module confirms the type and batch of raw materials through the RFID identification system according to the preset formula ratio, and automatically adjusts the opening of the batching valve using the modified PID control algorithm in combination with the flow monitoring data to achieve accurate batching;
[0099] Step 4: The raw materials are put into the reactor in proportion, the reactor is heated and dissolved by automatically adjusting the heater through the PID controller, and the initiator solution is added dropwise at a uniform speed through the flow control valve to react;
[0100] Step 5: During the reaction process, the temperature, pressure and liquid level parameters of the production site are collected in real time through the feedback monitoring system and transmitted to the PLC controller to dynamically adjust and control the feeding speed, stirring intensity, and the operation of the heating and cooling equipment;
[0101] Step 6: After the reaction is completed, 50% NaOH solution is automatically added for neutralization to adjust the pH value to 7-8;
[0102] Step 7: After neutralization, the reactor is cleaned by an automatic cleaning system;
[0103] Step 8: The pH value and solid content of the semi-finished additive product are tested by a pH meter and a conductivity meter, and the neutralization effect is analyzed by an intelligent image recognition algorithm, including color, texture and bubble distribution index parameters. If the neutralization effect does not meet the preset standard, the system triggers an alarm and automatically sends an adjustment instruction to the reaction control module; the adjustment instruction includes adjusting the reaction temperature, stirring speed and pH regulator dosage parameters; if the neutralization effect meets the preset standard, the semi-finished additive product after the reaction is sent to the drying and packaging module through a conveyor belt;
[0104] Step nine: The dry packaging module is dried by a hot air circulation dryer, and is quickly packaged and labeled by an automated packaging robot.
[0105] Specifically, in the specific embodiment of the overall scheme, after the raw materials are received, they first enter the parallel vibration screening mechanism, which uses efficient vibration screening technology to automatically remove impurities and lumps in the raw materials, ensuring the purity of the raw materials in the subsequent production process. The cleaned raw materials are automatically sent to the intelligent batching module through a screw conveyor. The module uses RFID technology to automatically identify the raw materials in the silo, combines flow monitoring data, and uses an advanced modified PID control algorithm to accurately control the opening of the batching valve to ensure the accuracy and consistency of the formula ratio. The raw materials after intelligent batching enter the reaction control module, where a PLC controller is used to automatically control the feeding speed, stirring intensity, and operation of the heating / cooling equipment of the raw materials according to preset process parameters, thereby achieving precise reaction condition control. At the same time, the feedback monitoring system monitors the key parameters such as temperature, pressure, and liquid level of the production site in real time, and feeds back the data to the PLC controller to form a closed-loop control, which improves the stability and controllability of the reaction process. The PLC controller also uses a fuzzy control strategy to accurately control the temperature in the reactor and optimize the reaction conditions. In the quality control module, pH meters and conductivity meters are used to test the key parameters of the additive semi-finished products, and the neutralization effect of the products is comprehensively evaluated in combination with intelligent image recognition algorithms. If the test results do not meet the preset standards, the system automatically triggers an alarm and sends adjustment instructions to the reaction control module, such as adjusting the reaction temperature, stirring speed, and the amount of pH adjuster to ensure product quality. Qualified semi-finished products then enter the drying and packaging module, where a hot air circulation dryer is used for drying to ensure the uniformity of the product's drying, and an automated packaging robotic arm is used to achieve fast and accurate packaging and labeling. The robotic arm is equipped with a visual recognition system to achieve automatic positioning and precise packaging of the product. In order to reduce energy consumption and emissions during the production process, an energy-saving and emission reduction module is specially designed, including a waste heat recovery submodule and an exhaust gas treatment submodule. The waste heat recovery submodule recovers the waste heat generated during the mixing and reaction process and uses it to preheat the next wave of raw materials, while the exhaust gas treatment submodule filters the dust generated during the production process, reducing environmental pollution.
[0106] Furthermore, the working method steps of the intelligent image recognition algorithm are as follows:
[0107] S1. Image preprocessing and feature extraction: The captured images are preprocessed by denoising, contrast enhancement and color correction, and then the color distribution, texture pattern, shape and distribution density of bubbles are extracted from the images by edge detection and texture segmentation methods;
[0108] S2, color space conversion and distribution quantification; quantify the deviation of the color in the image from the ideal neutralization effect through the color analysis function; the color analysis function evaluates the closeness of the overall color by calculating the weighted sum of the Euclidean distance between each color and the reference color; the formula expression of the color analysis function is:
[0109]
[0110] In formula (4), N c Indicates the number of color categories in the color space, which is used to measure the division interval of the color space during color quantization; C rj Represents the color vector of the jth pixel in the rth color class; C ref Indicates the ideal color when neutralization is complete; area rj A represents the area of the region to which the jth pixel in the rth color belongs; toe Represents the total area of the image, used to calculate the relative proportion of color distribution; weighet r The weight of the r-th color is used to reflect the importance of the r-th color in the evaluation;
[0111] S3, texture and bubble distribution analysis; after extracting the color and texture features, the intelligent image recognition algorithm further analyzes the bubble distribution through the bubble distribution analysis function to evaluate the complexity and uniformity of the bubble distribution; the bubble distribution includes the size, shape, density and spatial distribution pattern of the bubbles; the formula expression of the bubble distribution analysis function is:
[0112]
[0113] In formula (5), N bubbles Indicates the number of bubbles detected in the image, which is used to evaluate the uniformity of bubble distribution; area k Represents the area of the kth bubble, which is used to evaluate the bubble distribution density and total proportion; area max represents the maximum bubble area in the image, which is used to normalize the bubble area; p k represents the perimeter of the kth bubble; DL k represents the difference of the kth bubble shape based on Hu moment, which is used to quantify the regularity of the bubble shape;
[0114] S4, comprehensive evaluation and decision-making; Based on the analysis results of the above color and texture features, the intelligent image recognition algorithm guides the neutralization effect of the additive semi-finished product through the comprehensive evaluation score, and compares the comprehensive evaluation score with the preset threshold through the rule engine to trigger an alarm or command adjustment operation; the calculation formula of the comprehensive evaluation score is:
[0115]
[0116] In formula (6), and σ represent the weight coefficients of color quantization and bubble complexity index, respectively, which are used to measure the importance of color quantization and bubble complexity index in the evaluation; CQ max and BCImax They represent the maximum values of color quantization and bubble complexity index respectively, and are used for normalization.
[0117] In a specific embodiment, the color analysis function of formula (4) is based on the image retrieval technology of color histogram, which characterizes the content of the image by counting the frequency of occurrence of different colors in the image. However, in this application scenario, we need to more accurately evaluate the deviation between the image color and the ideal neutralization effect, so the concept of color vector is introduced and the Euclidean distance is used to quantify this deviation. Formula (4) evaluates the color consistency of the concrete additive semi-finished product by quantifying the deviation between the color in the image and the ideal color. In a specific implementation, a quantitative index of color consistency can be obtained by calculating the color deviation of each pixel and weighted summing it. Through the color analysis function, samples with large color deviation can be found in time, so as to take measures to improve the production process. Formula (5) The bubble distribution analysis function is based on the image processing technology of morphological analysis, which is usually used to detect specific structural features in the image, such as edges, holes, etc. In this specific application scenario, we are concerned with the distribution of bubbles, including the size, shape, density and spatial distribution pattern of bubbles. Formula (5) is used to evaluate the complexity and uniformity of bubble distribution, thereby indirectly reflecting the quality characteristics of concrete additives. In the specific implementation, by calculating the area, perimeter and shape difference of the bubbles, a quantitative index of bubble distribution can be obtained. The quality characteristics of concrete additives, such as fluidity and stability, can be evaluated through the bubble distribution analysis function. Formula (6) is based on a multi-index comprehensive evaluation method, which is usually used to evaluate the overall performance of a system or product. Here, we combine color quantification with the bubble complexity index to guide the neutralization effect of the additive semi-finished product through a comprehensive evaluation score. The purpose of this formula is to provide a comprehensive indicator for guiding the neutralization effect of the additive semi-finished product, and to compare it with the preset threshold through the rule engine to decide whether an alarm or adjustment of production parameters is needed. Through the comprehensive evaluation score, an overall performance index of the additive semi-finished product can be obtained. In addition, by comparing with the preset threshold, an alarm or adjustment instruction can be automatically triggered, thereby realizing intelligent management of the production process.
[0118] In specific implementation, the working mode of the intelligent image recognition algorithm can be summarized as five steps: "image acquisition-preprocessing-feature extraction-comparison analysis-decision output". During the production process, when the additive semi-finished product completes the neutralization reaction, the quality control module starts the image acquisition device (such as a high-definition camera) to capture the image in the reaction container. Subsequently, the algorithm preprocesses the collected images, including denoising, contrast enhancement and other operations to improve image quality. Next, the algorithm uses a deep learning model to extract key features in the image and compares and analyzes them with feature templates in the preset standards. Finally, based on the comparison results, the algorithm outputs decision information: if the neutralization effect does not meet the preset standard, an alarm is triggered and an adjustment instruction is sent to the reaction control module; if the neutralization effect meets the standard, the additive semi-finished product is sent to the dry packaging module via a conveyor belt.
[0119] When in operation, the intelligent image recognition algorithm can continuously monitor the process of the neutralization reaction of the additive, collect and analyze image information in real time, and ensure the accuracy and stability of the neutralization effect. When it is detected that the neutralization effect is not up to standard, the algorithm can quickly trigger the alarm mechanism and send specific adjustment instructions to the reaction control module to correct the deviation in the production process in time. At the same time, the algorithm also has the ability of self-learning and optimization, and can continuously optimize the model parameters of feature extraction and comparison analysis based on historical data and experience, so as to improve the accuracy and efficiency of recognition.
[0120] Compared with similar hardware, the intelligent image recognition algorithm can accurately extract key features from additive images through the application of deep learning algorithms, and perform high-precision comparison and analysis with preset standards to ensure accurate evaluation of the neutralization effect. In addition, compared with traditional contact detection methods (such as sampling detection), the intelligent image recognition algorithm realizes non-contact online detection, avoiding contamination and errors that may be introduced during the sampling process. Secondly, the algorithm can collect and analyze additive image information in real time, and promptly feedback the evaluation results of the neutralization effect, providing strong support for timely adjustment of the production process. At the same time, the algorithm has the ability of self-learning and optimization, and can continuously optimize the recognition model parameters based on historical data and experience to improve the accuracy and efficiency of recognition. In addition, through automated and intelligent image recognition technology, the need for manual intervention is reduced, labor costs are reduced, and production efficiency is improved.
[0121] At the same time, Experiment 3 is designed to evaluate the application effect of the intelligent image recognition algorithm in the production of concrete additives. The experiment is divided into two stages: baseline test and advanced test. The baseline test uses traditional image processing methods, and the advanced test uses intelligent image recognition algorithms. Experimental equipment and materials include: concrete additive production equipment. High-resolution industrial camera. Image processing software and intelligent image recognition algorithm. Color analysis and texture analysis tools. The experimental process is: 1. Use an industrial camera to capture images of semi-finished concrete additives. 2. Denoise, contrast enhancement and color correction of the image. 3. Feature extraction: Extract color, texture and bubble features through edge detection and texture segmentation methods. 4. Color space conversion and distribution quantification: Use color analysis functions to quantify color deviations. 5. Evaluate the complexity and uniformity of bubble distribution. 6. Based on the analysis results, use the rule engine to evaluate the neutralization effect. The data table is shown in Table 3:
[0122] Table 3 Intelligent image recognition algorithm experimental evaluation record
[0123]
[0124] As shown in Table 3, the intelligent algorithm provides higher quality image preprocessing and more accurate feature extraction. At the same time, the algorithm significantly improves the accuracy of color quantification and texture analysis. In addition, the intelligent algorithm more effectively evaluates the complexity and uniformity of bubble distribution. Secondly, the intelligent algorithm provides a more comprehensive and accurate comprehensive evaluation score to guide production adjustments.
[0125] Furthermore, the visual recognition system realizes accurate recognition and rapid packaging of dried products through image processing algorithms and robot motion control protocols; the image processing algorithm extracts contour information of products in complex backgrounds through edge detection and feature matching methods; the robot motion control protocol ensures the smoothness and continuity of the robot arm's movements through path planning algorithms and real-time scheduling mechanisms, and also improves packaging efficiency through multi-task parallel processing.
[0126] In a specific embodiment, the visual recognition system adopts a method based on edge detection and feature matching. Edge detection uses the Sobel operator, Canny edge detection and other techniques to locate the position where the intensity changes significantly in the image, that is, the edge. This is because in the image, the boundary of the object often corresponds to a sharp change in the pixel value. Feature matching refers to the process of finding corresponding points between the same object photographed at different viewing angles or under different conditions. For example, algorithms such as scale-invariant feature transform (SIFT) or accelerated segment test (FAST) can be used to identify key points in the image, and the descriptors of these points can be calculated to find matching key point pairs. Through the path planning algorithm, the system can generate an optimal path from the starting point to the target point for the robot arm. This usually involves obstacle avoidance algorithms, dynamic programming and other technologies. The real-time scheduling mechanism ensures that in a multi-task environment, the robot arm can perform tasks efficiently with limited resources. In addition, the multi-task parallel processing method means that the system can process multiple packaging tasks at the same time, thereby improving overall production efficiency.
[0127] In a method and device for producing concrete additives, the visual recognition system first captures the image of the product on the production line, and then analyzes the grayscale changes in the image through an edge detection algorithm to extract the contour information of the product. Then, the system uses feature matching technology to compare these contour information with known templates to determine the type and location of the product. Once the product is correctly identified, the system sends instructions to the robot, and the robot arm moves to the top of the product according to the pre-set path planning, and then completes the grabbing and placing actions through fine control. The whole process is fast and accurate, reducing the possibility of human intervention. When the system is in normal working condition, it can continuously monitor every product on the conveyor belt and respond almost instantly. Even on a high-speed production line, the system can maintain a high-precision recognition rate. In addition, due to the use of a real-time scheduling mechanism, the system can flexibly respond to emergencies on the production line, such as a sudden increase in the number of products or temporarily adjusted packaging requirements.
[0128] Compared with traditional manual packaging or simple automated equipment, this production equipment integrated with a visual recognition system has obvious advantages. It not only improves production efficiency and reduces error rates, but also reduces labor costs. More importantly, through intelligent image processing technology and precise robot control, the system can adapt to the packaging needs of many different types of products, enhancing the flexibility and adaptability of the production line. In addition, multi-task parallel processing allows more products to be processed at the same time, which is crucial to improving the market competitiveness of enterprises. In summary, the application of this technology is an important advancement for modern manufacturing, marking a major step towards a more intelligent and automated production model.
[0129] Furthermore, the waste heat recovery submodule includes a high-efficiency heat exchange unit, a phase change energy storage unit and an intelligent control unit;
[0130] In a specific embodiment, the waste heat recovery submodule is based on heat exchanger technology. In this embodiment, paraffin C28 (octacosane) is used as a phase change material for high-temperature waste heat recovery in concrete additive production, and its phase change temperature is about 80°C, wherein the phase change temperature of paraffin C28 is about 80°C; the latent heat is about 200-250 J / g; the thermal conductivity is about 0.2 W / (m·K); the density is about 850 kg / m 3 (Solid State);
[0131] In actual application, paraffin C28 will begin to melt when the temperature rises above 80°C, absorbing a large amount of heat; when the temperature drops, the paraffin re-solidifies and releases the previously stored heat. In this way, the phase change material can effectively store the waste heat generated during the production process and release it when needed for preheating raw materials or other processes that require heat. The exhaust gas treatment submodule is mainly used to filter dust and other particulate matter generated during the production process. In this embodiment, we used a bag filter (model: DC-500), which is an efficient dry dust removal device. The bag filter captures particulate matter in the air through a bag. When the dust-containing gas enters the dust collector, the larger particles settle directly to the bottom collection box due to gravity, while the fine particles are intercepted by the bag. As the filtering process continues, a layer of filter cake will form on the surface of the bag, which further enhances the dust removal effect.
[0132] The energy-saving and emission-reduction module plays a key role in the production process of concrete additives. First, the high-temperature gas generated during the mixing and reaction process is directed to the heat exchanger in the waste heat recovery submodule. Here, the heat is transferred to the fresh raw materials that will be added to the reactor. The preheated raw materials help to speed up the chemical reaction rate and reduce the energy consumption required for heating. Then, the waste gas after heat exchange will be sent to the waste gas treatment submodule for filtration. The bags in the bag filter effectively capture the dust in the waste gas, and the purified air is eventually discharged into the atmosphere, reducing the impact on the environment. When the energy-saving and emission-reduction module is in normal working condition, it can continuously and stably recover waste heat and effectively filter dust in the waste gas. For example, in a typical production cycle, the waste heat recovery submodule can preheat the fresh raw materials from room temperature (about 20°C) to a temperature close to the reaction temperature (about 150°C), which greatly reduces the energy consumption of the heating equipment. At the same time, the waste gas treatment submodule can reduce the dust concentration in the exhaust gas to below the national standard, ensuring that the exhaust gas meets the environmental protection standards. In specific implementation, the energy-saving and emission-reduction module can significantly reduce energy costs through waste heat recovery, which is particularly important when energy prices fluctuate greatly. On the other hand, waste gas treatment measures effectively reduce pollution to the surrounding environment, meeting the country's increasingly stringent environmental protection requirements for industrial production. In addition, efficient heat exchange and dust filtration technology can extend the service life of production equipment and reduce maintenance costs. In short, this energy-saving and emission-reduction system not only helps companies achieve sustainable development goals, but also demonstrates a positive image of companies in fulfilling their social responsibilities.
[0133] Furthermore, the device also includes a data management and analysis module for automatically collecting, storing and analyzing production data;
[0134] In a specific embodiment, the data management and analysis module plays the role of the central nervous system in the concrete additive production process. First, the data acquisition submodule continuously collects data from various key points of the production line. After preprocessing and fusion, these data are securely stored in the NoSQL database. Subsequently, the data analysis submodule performs real-time and historical analysis on these data to identify factors that may affect product quality and predict the operating status of the equipment. Finally, the data visualization submodule converts these complex analysis results into an intuitive graphical interface, allowing managers to make decisions quickly.
[0135] Integrating such a data management and analysis module greatly improves the utilization rate of production data, making the production process more transparent and controllable. Secondly, through real-time monitoring and predictive maintenance, the equipment failure rate can be significantly reduced, the equipment service life can be extended, and the maintenance cost can be reduced. Furthermore, historical data analysis helps to discover potential problems in the production process, optimize the process flow, and improve product quality. Finally, data visualization tools enable managers to understand and communicate production conditions more easily, strengthen communication and collaboration between teams, and promote the digital transformation of enterprises. In summary, this module is not only the basis for modern factories to realize intelligent manufacturing, but also an important means for enterprises to enhance their core competitiveness.
[0136] Furthermore, the automatic cleaning system includes a cleaning agent distribution module and a cleaning execution module; the cleaning agent distribution module includes a quantitative delivery unit and an automatic switching valve unit; the quantitative delivery unit delivers cleaning agent through a metering pump and flow control, and the metering pump is driven by a stepping motor; the flow control controls the opening and closing of the delivery valve through a solenoid valve; the cleaning execution module realizes strong spraying of the inner and outer surfaces of the reactor through a high-pressure pump and a nozzle, and the cleaning execution module also realizes the removal of fine particles on the surface of the reactor through ultrasonic vibration and cavitation effect; the ultrasonic vibration converts electrical energy into high-frequency vibration through a transducer, and the cavitation effect removes stubborn stains through the shock waves generated by the formation and rupture of bubbles.
[0137] In a specific embodiment, the quantitative delivery unit of the automatic cleaning system delivers the cleaning agent through a high-precision metering pump, which is precisely driven by a stepper motor to ensure that the amount of cleaning agent delivered each time is accurate. The working principle of the metering pump is based on the volumetric principle. Through the precise control of the stepper motor, the reciprocating motion of the piston in the pump body is realized, thereby accurately metering and delivering the cleaning agent. The automatic switching valve unit uses a solenoid valve to control the opening and closing of the delivery valve to achieve accurate delivery and switching of the cleaning agent. The solenoid valve uses the electromagnet to attract or release to control the opening and closing of the valve, and has the characteristics of fast response speed and high control accuracy.
[0138] The high-pressure pump and nozzle system in the cleaning execution module, the high-pressure pump provides high-pressure water flow, and sprays the inner and outer surfaces of the reactor through special nozzles (such as fan-shaped nozzles or rotating nozzles). The working principle of the high-pressure pump is based on the multi-stage centrifugal or plunger principle, which pressurizes the water flow to hundreds or even thousands of bars, forming a strong impact force to effectively remove surface dirt. Ultrasonic vibration converts electrical energy into high-frequency vibration through the transducer, acts on the surface of the reactor, generates tiny bubbles and quickly breaks, forming shock waves. This process is the cavitation effect, which can effectively peel off and remove fine particles and stubborn stains on the surface. Ultrasonic vibration technology uses the energy generated by high-frequency vibration to form local high-pressure and low-pressure areas in the liquid, thereby generating bubbles and causing them to expand and break rapidly, and the energy released is enough to remove stains that are difficult to reach.
[0139] When the system is working, the control system first issues a command to start the cleaning agent delivery module. Driven by the stepper motor, the metering pump accurately delivers a certain amount of cleaning agent to the reactor according to the preset parameters. Subsequently, the automatic switching valve unit quickly switches to the cleaning state, closes the cleaning agent delivery channel, and turns on the cleaning execution module. The high-pressure pump starts, and the high-pressure water flow sprays the inner and outer surfaces of the reactor through the nozzle. At the same time, the ultrasonic vibration device starts working, and the transducer converts electrical energy into high-frequency vibration, which acts on the surface of the reactor to produce a cavitation effect, further enhancing the cleaning effect. The entire cleaning process is monitored in real time by the control system to ensure the precise execution of each step.
[0140] In the specific implementation, the cleaning process is highly efficient and energy-saving by precisely controlling the delivery volume of the cleaning agent and the output power of the high-pressure pump. Compared with the traditional manual cleaning method, it greatly improves the cleaning efficiency and reduces energy consumption. At the same time, the use of ultrasonic vibration and cavitation effect technology can deeply remove fine particles and stubborn stains on the surface of the reactor, ensuring the consistency and thoroughness of the cleaning effect. In addition, the entire cleaning process is automatically completed by the control system without manual intervention, which reduces labor intensity and improves production safety.
[0141] Although the specific embodiments of the present invention are described above, it should be understood by those skilled in the art that these specific embodiments are only illustrative, and those skilled in the art may omit, replace, and change the details of the above methods and systems in various ways without departing from the principles and essence of the present invention. For example, merging the above method steps so as to perform substantially the same functions in substantially the same manner to achieve substantially the same results is within the scope of the present invention. Therefore, the scope of the present invention is limited only by the appended claims.
Claims
1. A production equipment for concrete additives, characterized in that: The equipment includes a raw material processing module, an intelligent batching module, a reaction control module, a quality control module, a drying and packaging module and an energy-saving and emission reduction module; The raw material processing module is used to remove impurities and lumps in the additive raw material through a parallel vibration screening mechanism, and to convey the processed additive raw material to the intelligent batching module through a screw conveyor; The intelligent batching module is used to realize the automated batching process; the intelligent batching module is provided with a plurality of independent silos, each of which is equipped with an RFID identification system. Based on the identification information and flow monitoring data, the intelligent batching module automatically controls the opening of the batching valve according to the preset formula ratio through a modified PID control algorithm; The reaction control module is used to automatically control the feeding speed, stirring intensity and operation of the heating / cooling equipment of the raw materials through the PLC controller according to the preset process parameters; the reaction control module is also used to monitor the reaction state in real time through the feedback monitoring system, and transmit the real-time monitoring data to the PLC controller to form a closed-loop control; the real-time monitoring data at least includes the temperature, pressure and liquid level data parameters of the production site, and the process parameters include but are not limited to temperature, pressure and reaction time; The quality control module is used to detect the pH value and solid content of the semi-finished additive product after the reaction is completed by a pH meter and a conductivity meter, and to assist in analyzing the neutralization effect by an intelligent image recognition algorithm. If the neutralization effect does not meet the preset standard, the system triggers an alarm and sends an adjustment instruction to the reaction control module; the preset standard at least includes color, texture and bubble distribution indicators; the adjustment instruction includes adjusting the reaction temperature, stirring speed and pH regulator dosage parameters; if the neutralization effect meets the preset standard, the semi-finished additive product after the reaction is sent to the drying and packaging module through a conveyor belt; The drying and packaging module is used to perform drying treatment by a hot air circulation dryer, and to perform rapid packaging and labeling by an automated packaging robot arm; the automated packaging robot arm realizes product positioning and packaging by a visual recognition system; The energy-saving and emission-reduction module is used to reduce energy consumption and emissions during the production process; the energy-saving and emission-reduction module includes a waste heat recovery submodule and a waste gas treatment submodule, the waste heat recovery submodule is used to recover the waste heat generated during the mixing and reaction process and preheat the next wave of raw materials; the waste gas treatment submodule is used to filter the dust generated during the production process.
2. The production equipment of a concrete additive according to claim 1, characterized in that: The parallel vibration screening mechanism screens different types of raw materials through a multi-channel screening unit, and the multi-channel screening unit screens different raw materials through multiple screening grids; then, the processed raw materials are transported to several independent silos in the intelligent batching module through an intelligent conveying mechanism and an RFID identification system; the intelligent conveying mechanism conveys the processed raw materials to the designated silos in steps according to the grids through an adjustable speed screw conveying method and a real-time monitoring sensor; the adjustable speed screw conveying method controls the motor speed through a frequency converter, and the real-time monitoring sensor monitors the flow state of the raw materials in the conveying process in real time through a sensor network, and feeds back the data to the embedded control system in real time. system; the real-time monitoring sensor includes a photoelectric sensor and a proximity switch; the embedded control system adjusts the conveying speed through the real-time data of the differential to ensure the accurate conveying of the raw materials; the RFID identification system realizes automatic identification and classification of raw materials through RFID tags and readers; each screening grid is equipped with an RFID tag to store raw material information; the raw material information at least includes the name, batch number and expiration date information, the raw material information is read by the reader, and the raw material information data is transmitted to the embedded control system through the wireless communication protocol; based on the raw material information data, the embedded control system controls the screw conveyor through the frequency converter to transport the processed raw materials to the designated silo.
3. The production equipment of a concrete additive according to claim 1, characterized in that: The modified PID control algorithm first uses the proportional term to respond to the deviation change, and adjusts the proportional gain through deviation calculation to reflect the deviation size and produce corresponding control effect; the deviation calculation realizes the integral action through the forgetting factor to gradually weaken the memory of past deviations over time, and simulates the periodic change of external interference through periodic modulation. The modified PID control algorithm then performs nonlinear adjustment according to the sign and size of the deviation change rate through the nonlinear differential term to suppress the overshoot and oscillation of the system; in the intelligent batching module, the modified PID control algorithm also identifies the type of raw materials through the RFID identification system, and calculates the deviation of each raw material in combination with real-time flow data; finally, the modified PID control algorithm also estimates and compensates the effect of external interference on the system in real time through the external interference compensation mechanism; the formula expression of the modified PID control algorithm for the opening control of the batching valve is: In formula (1), u t Represents the output signal of the controller, which is used to adjust the control quantity of the controlled object, K p , K i and K d Respectively represent proportional gain, integral gain and differential gain; e t It represents the control deviation, that is, the difference between the set value and the actual value, which is used to reflect the gap between the current state of the system and the expected state; δ The modulation function representing the integral term is used to adjust the cumulative effect of the integral term and improve the stability and response speed of the system; represents the nonlinear modulation function of the differential term, which is used to make nonlinear adjustments according to the characteristics of the deviation change rate and enhance the system's adaptability to dynamic changes; t represents the current time; i represents the type of raw material; γ represents the measurement error, h and x represent the adjustment parameters, which are used to control the intensity of the integral action and the forgetting speed; θ represents the sampling interval, x represents the number of iterations, and β i Represents the weight coefficient of RFID and flow feedback, which is used to determine the relative importance of RFID identification and flow monitoring data in the control function; RFID i (t) represents the RFID identification value of the i-th raw material, which is used to provide accurate information about the type of raw material; Q i (t) represents the real-time flow rate of the i-th raw material at time t, Q i Indicates the corresponding real-time traffic Q i The setting value of (t) is used to calculate the flow deviation; ε represents the external disturbance compensation coefficient, which is used to determine the weight of the external disturbance compensation term in the control action; Disturbance(t) represents the estimated value of external disturbance.
4. The production equipment of a concrete additive according to claim 1, characterized in that: The working method of the reaction control module is: Step 1: Capture the temperature, pressure and liquid level parameters of the production site through the ADC analog-to-digital converter, and pre-process the data through filters and normalization methods; Step 2: Based on the preprocessed data, the reaction state is monitored in real time and comprehensively evaluated through the comprehensive strategy of multivariate statistical process control MSPC and deep learning model; the multivariate statistical process control MSPC is used to identify whether the production process deviates from the normal operating range, and the deep learning model is used to further classify and analyze the causes of abnormal states; the deep learning model is combined with the process knowledge base to analyze the impact of abnormal states on product quality; Step 3: Based on the state evaluation data, construct the input data set F through the fuzzy control method sr ={e,e c , p, l} and the output dataset F sc = {v, s, g}; Input data set F sr In the equation, e represents the temperature deviation, e c represents the temperature deviation change rate, p represents pressure, and l represents liquid level; input data set F sc In the formula, v represents the feeding speed, s represents the stirring intensity, and g represents the heating or cooling power. Based on the expert experience and process knowledge, the fuzzy control method generates real-time control instructions through a dynamic adjustment function to improve the control accuracy. The formula of the dynamic adjustment function is: In formula (2), q ij represents the adjustment factor, z1, z2, z3 and z4 are weight coefficients, which are used to adjust the relative importance of temperature deviation, temperature deviation change rate, pressure deviation and liquid level deviation in the calculation of the adjustment factor; e represents the temperature deviation, e c Indicates the temperature deviation change rate, p and p target Respectively represent the current pressure value and the target pressure value to be achieved; l and l target Respectively represent the current liquid level value and the target liquid level value to be achieved; max|e|, max|e c |, max|p|, and max|l| represent the maximum absolute values of the deviations, which are used to normalize the deviation terms; Step 4: The PLC controller adjusts the raw material feeding speed, stirring intensity, and the operating state of the heating or cooling equipment according to the received control instructions; at the same time, the PLC controller continuously collects new real-time data through the feedback monitoring system, and performs feedback evaluation through the control effect evaluation function, and the formula expression of the control effect evaluation function is: In formula (3), A(X) represents the control effect index, T c , P c and L c Respectively represent the temperature, pressure and liquid level values of the cth sampling point, which are used to evaluate the actual effect of the control strategy; T taeget , P taeget and L taeget Respectively represent the target values of temperature, pressure and liquid level; T range , P range and L range They represent the range of temperature, pressure and liquid level, respectively, and are used to normalize the deviation term; a represents the attenuation factor, which is used to control the speed of time attenuation and balance the needs between long-term stability and immediate response; represents the time of the cth sampling point; N represents the total number of sampling points; Step 5. During the continuous production process, the reaction control module is also optimized and adjusted through a reinforcement learning framework and an adaptive control mechanism to cope with changes and uncertainties in production conditions; the reinforcement learning framework continuously optimizes the control strategy by defining a state space, an action space, and a reward function; the adaptive control mechanism automatically adjusts control parameters based on real-time feedback data and historical data by comparing the difference between the predicted state and the actual state to respond to changes in production conditions; the state space includes but is not limited to temperature, pressure, and liquid level parameters; the action space includes adjusting the feeding rate, stirring intensity, and heating or cooling power; the reward function includes but is not limited to a comprehensive evaluation based on product quality, production efficiency, and energy consumption indicators.
5. The production equipment of a concrete additive according to claim 1, characterized in that: The working method steps of the intelligent image recognition algorithm are as follows: S1, image preprocessing and feature extraction; The captured images are preprocessed by denoising, contrast enhancement and color correction, and then the color distribution, texture pattern, shape and distribution density of bubbles are extracted from the images by edge detection and texture segmentation methods; S2, color space conversion and distribution quantization; The deviation of the color in the image from the ideal neutralization effect is quantified by a color analysis function; the color analysis function evaluates the closeness of the overall color by calculating the weighted sum of the Euclidean distance between each color and the reference color; the formula expression of the color analysis function is: In formula (4), N c Indicates the number of color categories in the color space, which is used to measure the division interval of the color space during color quantization; C rj Represents the color vector of the jth pixel in the rth color class; C ref Indicates the ideal color when neutralization is complete; area rj A represents the area of the region to which the jth pixel in the rth color belongs; toe Represents the total area of the image, used to calculate the relative proportion of color distribution; weighet r The weight of the r-th color is used to reflect the importance of the r-th color in the evaluation; S3, texture and bubble distribution analysis; After extracting the color and texture features, the intelligent image recognition algorithm further analyzes the bubble distribution through the bubble distribution analysis function to evaluate the complexity and uniformity of the bubble distribution; the bubble distribution includes the size, shape, density and spatial distribution pattern of the bubbles; the formula expression of the bubble distribution analysis function is: In formula (5), N bubbles Indicates the number of bubbles detected in the image, which is used to evaluate the uniformity of bubble distribution; area k Represents the area of the kth bubble, which is used to evaluate the bubble distribution density and total proportion; area max represents the maximum bubble area in the image, which is used to normalize the bubble area; p k represents the perimeter of the kth bubble; DL k represents the difference of the kth bubble shape based on Hu moment, which is used to quantify the regularity of the bubble shape; S4, comprehensive assessment and decision-making; Based on the analysis results of the above color and texture features, the intelligent image recognition algorithm guides the neutralization effect of the additive semi-finished product through a comprehensive evaluation score, and compares the comprehensive evaluation score with a preset threshold through a rule engine to trigger an alarm or command adjustment operation; the calculation formula of the comprehensive evaluation score is: In formula (6), and σ represent the weight coefficients of color quantization and bubble complexity index, respectively, which are used to measure the importance of color quantization and bubble complexity index in the evaluation; CQ max and BCI max They represent the maximum values of color quantization and bubble complexity index respectively, and are used for normalization.
6. The production equipment of a concrete additive according to claim 1, characterized in that: The visual recognition system realizes accurate recognition and rapid packaging of dried products through image processing algorithms and robot motion control protocols; the image processing algorithm extracts contour information of products in complex backgrounds through edge detection and feature matching methods; the robot motion control protocol ensures the smoothness and continuity of the robot arm's movements through path planning algorithms and real-time scheduling mechanisms, and also improves packaging efficiency through multi-task parallel processing.
7. The production equipment of a concrete additive according to claim 1, characterized in that: The waste heat recovery submodule includes a high-efficiency heat exchange unit, a phase change energy storage unit and an intelligent control unit; the high-efficiency heat exchange unit transfers heat in high-temperature exhaust gas through heat pipe heat exchange; the phase change energy storage unit stores high-temperature waste heat through phase change materials, and the phase change materials undergo phase change when absorbing heat, changing from solid to liquid, storing a large amount of latent heat, and when releasing heat, the phase change materials change from liquid to solid, releasing the stored heat; the intelligent control unit intelligently schedules and optimizes the stored waste heat through an embedded operating system and an optimization algorithm, and the embedded operating system monitors at least temperature and pressure parameters in the production process in real time through a real-time data processing method and a sensor network; the optimization algorithm dynamically adjusts the heat exchange according to real-time data through a model predictive control method The working status of the device; the exhaust gas treatment submodule includes a dust capture unit, an exhaust gas purification unit and an emission monitoring unit; the dust capture unit captures the dust generated in the production process through electrostatic dust removal and cyclone separation, the electrostatic dust removal charges the dust through a high-voltage electrostatic field, and adsorbs the charged dust on the dust collecting plate through the electric field force, and then separates the dust from the airflow through the cyclone separation; the exhaust gas purification unit purifies the harmful substances in the exhaust gas through catalytic oxidation and activated carbon adsorption; the emission monitoring unit monitors the emission of the purified exhaust gas in real time through online monitoring and data transmission protocols, and transmits the monitoring data to the central control system for analysis and recording, and the online monitoring monitors the pollutant concentration in the exhaust gas in real time through spectral analysis and chemical sensors.
8. The production equipment of a concrete additive according to claim 1, characterized in that: The device also includes a data management and analysis module for automatically collecting, storing and analyzing production data; the data management and analysis module includes a data collection submodule, a data storage submodule, a data analysis submodule and a data visualization submodule; the data collection submodule includes a sensor interface unit, a data preprocessing unit and a data fusion unit; the sensor interface unit transmits sensor data to the embedded system through the serial communication protocol RS-485; the data preprocessing unit performs preliminary processing of sensor data through Kalman filtering and outlier detection; the data fusion unit performs fusion of different sensor data through Bayesian estimation; the data storage submodule includes a database management unit, a data backup unit and a data recovery unit; the database management unit performs NoSQ L database realizes the storage and management of production data; the data backup unit realizes regular backup of database data through cron job scheduling and incremental backup; the data recovery unit ensures the integrity and availability of production data through the disaster recovery plan DRP; the data analysis submodule includes a real-time data analysis unit, a historical data analysis unit and a predictive maintenance unit; the real-time data analysis unit provides instant production status monitoring through a stream processing framework; the historical data analysis unit discovers the rules and problems in the production process through potential causal models and graph databases; the predictive maintenance unit predicts the probability of equipment failure based on historical maintenance records through a reinforcement learning framework; the data visualization submodule realizes intuitive display of production data through the data visualization tool D3.js and the document generation engine.
9. A method for producing a concrete additive, characterized in that: A production device for a concrete additive as claimed in any one of claims 1 to 8, comprising the following steps: Step 1: Start the raw material processing module, pour the additive raw material into the vibration screening machine, and remove impurities and lumps in the raw material through high-frequency vibration; Step 2: The additive raw materials after screening are automatically transported to each independent silo of the intelligent batching module by a screw conveyor. Each silo is equipped with an RFID identification system to automatically record and identify the raw material information; Step 3: The intelligent batching module confirms the type and batch of raw materials through the RFID identification system according to the preset formula ratio, and automatically adjusts the opening of the batching valve using the modified PID control algorithm in combination with the flow monitoring data to achieve accurate batching; Step 4: The raw materials are put into the reactor in proportion, the reactor is heated and dissolved by automatically adjusting the heater through the PID controller, and the initiator solution is added dropwise at a uniform speed through the flow control valve to react; Step 5: During the reaction process, the temperature, pressure and liquid level parameters of the production site are collected in real time through the feedback monitoring system and transmitted to the PLC controller to dynamically adjust and control the feeding speed, stirring intensity, and the operation of the heating and cooling equipment; Step 6: After the reaction is completed, 50% NaOH solution is automatically added for neutralization to adjust the pH value to 7-8; Step 7: After neutralization, the reactor is cleaned by an automatic cleaning system; Step 8: The pH value and solid content of the semi-finished additive product are tested by a pH meter and a conductivity meter, and the neutralization effect is analyzed by an intelligent image recognition algorithm, including color, texture and bubble distribution index parameters. If the neutralization effect does not meet the preset standard, the system triggers an alarm and automatically sends an adjustment instruction to the reaction control module; the adjustment instruction includes adjusting the reaction temperature, stirring speed and pH regulator dosage parameters; if the neutralization effect meets the preset standard, the semi-finished additive product after the reaction is sent to the drying and packaging module through a conveyor belt; Step nine: The dry packaging module is dried by a hot air circulation dryer, and is quickly packaged and labeled by an automated packaging robot.
10. The method for producing a concrete additive according to claim 9, characterized in that: The automatic cleaning system includes a cleaning agent distribution module and a cleaning execution module; the cleaning agent distribution module includes a quantitative delivery unit and an automatic switching valve unit; the quantitative delivery unit delivers cleaning agent through a metering pump and flow control, and the metering pump is driven by a stepping motor; the flow control controls the opening and closing of the delivery valve through a solenoid valve; the cleaning execution module realizes strong spraying on the inner and outer surfaces of the reactor through a high-pressure pump and a nozzle, and the cleaning execution module also realizes the removal of fine particles on the surface of the reactor through ultrasonic vibration and cavitation effect; the ultrasonic vibration converts electrical energy into high-frequency vibration through a transducer, and the cavitation effect removes stubborn stains through the shock waves generated by the formation and rupture of bubbles.
Citation Information
Cited By
Impurity removal and decolorization kettle for chemical production
CN117065686A
Plastic additive cooling equipment for PVC composite tile production and control system
CN119458665A
