An industrial automatic dust-free production control system and method for konjak powder

Abstract

The application relates to the technical field of industrial automation control, and discloses a konjak powder industrial automation dust-free production control system and method, which comprises a fluidized bed airflow pulverizer, a pulse control module, a multi-dimensional online sensing module, an in-situ conditioning module and a cooperative control core module. The core of the system is that the cooperative control core makes the system run in an alternating working pulse period and a dormant conditioning period through the pulse control module. In the dormant conditioning period with low energy, the multi-dimensional online sensing module collects state parameters such as the resonant frequency, temperature and image of the material. The cooperative control core makes decisions based on the parameters, starts the in-situ conditioning module for cooling or brittleness increasing treatment as needed, and adaptively adjusts the length of the next working pulse period to form a closed-loop control. The application realizes the automation of the whole production process, effectively prevents the thermal damage of the material, and ensures the high consistency of the product quality and the robustness of the production process.

Description

Technical Field

[0001] This invention relates to the field of industrial automation control technology, specifically to an automated dust-free production control system and method for konjac flour. Background Technology

[0002] The quality of konjac flour is highly correlated with the activity of its core component, konjac glucomannan, and fluidized bed air jet milling is the mainstream technology for achieving its ultrafine grinding. However, existing technologies have revealed inherent limitations in application, which directly affect the quality, consistency, and automation level of the final product.

[0003] During continuous high-energy pulverization, the temperature of the material accumulates due to intense friction and collision, which is extremely detrimental to heat-sensitive substances like konjac glucomannan. A major challenge of existing technologies is that high-speed airflow and dispersed dust clouds make accurate real-time monitoring of the material's true temperature virtually impossible, resulting in temperature control measures that are often delayed and indirect. This leads to the system being unable to avoid localized overheating even during operation, causing irreversible thermal deterioration of the material and ultimately damaging key quality indicators such as viscosity and color of the product.

[0004] Meanwhile, existing production processes are typically "open-loop" or rely on manual experience for control, generally lacking real-time feedback on the macroscopic state of the crushing process. When batch characteristics of raw materials, such as moisture content or hardness, fluctuate, fixed process parameters cannot be adaptively adjusted, leading to significant deviations in crushing efficiency and finished particle size. This inability to compensate for changes in operating conditions in real time is the root cause of uneven particle size distribution and poor quality consistency among different batches of products.

[0005] Finally, from a system design perspective, traditional production processes are also quite fragmented. To address temperature rises or improve material properties, independent downstream cooling and conditioning equipment is often required, or tedious offline sampling and analysis must be performed before adjusting the main equipment parameters. This multi-device, multi-stage layout not only complicates the production line structure but also compromises the system's airtightness, increasing the risk of human intervention and cross-contamination, failing to construct a truly closed-loop, unattended automated production system. Therefore, there is an urgent need in this field for a technical solution that can integrate and address the aforementioned problems. Summary of the Invention

[0006] The technical problem to be solved by this invention is that in the existing konjac flour production process, continuous airflow milling, while pursuing high efficiency, is difficult to avoid material thermal damage caused by continuous temperature rise in the grinding zone, as well as over-grinding or under-grinding caused by fluctuations in material characteristics, thereby affecting the stability and consistency of the final product quality.

[0007] The first aspect of this invention provides an automated dust-free production control system for konjac flour industry, comprising:

[0008] A fluidized bed air jet mill, the fluidized bed air jet mill comprising a grinding chamber and a grinding nozzle communicating with the grinding chamber;

[0009] A pulse control module is connected to the air inlet pipe of the grinding nozzle and is used to control the interruption of the high-pressure air supply to the grinding nozzle.

[0010] A multi-dimensional online sensing module is installed on the fluidized bed air jet mill to acquire the state parameters of the konjac material in the grinding chamber in real time.

[0011] An in-situ conditioning module is connected to the grinding chamber and is used to inject conditioning media into the grinding chamber;

[0012] The collaborative control core module is electrically connected to the pulse control module, the multi-dimensional online sensing module, and the in-situ conditioning module.

[0013] The collaborative control core module is configured to: control the pulse control module to supply the high-pressure airflow in a pulse manner, thereby forming alternating working pulse periods and dormant conditioning periods in the grinding chamber; and control the in-situ conditioning module to perform quality conditioning based on the state parameters obtained by the multi-dimensional online sensing module during the dormant conditioning period, and adaptively adjust the parameters of the subsequent working pulse periods.

[0014] Preferably, the multidimensional online sensing module includes:

[0015] The machine vision unit is used to acquire the particle size distribution and color characteristics of konjac materials.

[0016] Infrared temperature measurement array is used to obtain the average temperature of the particle group and the instantaneous maximum temperature of konjac material;

[0017] An acoustic characteristic sensing unit is used to obtain the resonant frequency of the system composed of konjac material and the grinding chamber.

[0018] In one specific embodiment, the acoustic characteristic sensing unit includes a sound wave transmitter and an acoustic sensor. The sound wave transmitter is used to transmit a sweep frequency signal S into the grinding chamber during the dormant conditioning period. in (t), which can be in the form of:

[0019]

[0020] Among them, f start and f end These are the start and end frequencies of the sweep, T. sweepThe sweep frequency duration is t, where t is a time variable. The acoustic sensor is used to receive the echo signal S after passing through the konjac material. out (t), which is used by the collaborative control core module to calculate the acoustic transfer function H(f):

[0021]

[0022] in, Here, f is the Fourier transform operator, and f is the frequency variable. The collaborative control core module identifies the resonant frequency f from the amplitude-frequency response of the transfer function H(f) using a peak detection algorithm. res :

[0023]

[0024] Furthermore, the collaborative control core module is also configured to operate according to the resonant frequency f. res (t), adaptively adjusting the duration T of the subsequent working pulse period. work (t+1). Adjustment amount ΔT work (t) can be calculated using a proportional-integral controller:

[0025]

[0026] Among them, K P and K I Let be the proportional and integral coefficients, respectively, τ be the integral variable, and the error term e(t) be a weighted combination:

[0027]

[0028] Among them, w res and w Θ These are the weighting coefficients. Θ is the target resonant frequency. avg (t) represents the average temperature of the particle swarm. This is the average temperature safety threshold.

[0029] Preferably, the in-situ conditioning module includes:

[0030] A cooling gas circuit unit is used to inject cooling gas to control the temperature of konjac material;

[0031] A micro-atomization crisping unit is used to inject micron-level water mist to adjust the crispness of konjac materials.

[0032] In one specific embodiment, the collaborative control core module is further configured to: when the average temperature Θ of the particle swarm acquired by the infrared thermometer array... avg (t) or instantaneous highest temperature Θ max (t) exceeds the preset safety threshold or During the dormancy conditioning period, cooling gas is injected into the cooling gas path unit.

[0033] In another specific embodiment, the collaborative control core module is further configured to: when the crushing efficiency index η is calculated based on the particle size distribution obtained by the machine vision unit... D (t) is lower than the preset efficiency threshold η min During the dormancy conditioning period, the micro-atomization embrittlement unit is controlled to inject micron-level water mist.

[0034] Preferably, the fluidized bed air jet mill further comprises:

[0035] An integrated grading wheel is used to dynamically grade the material leaving the grinding chamber, allowing only qualified products that reach a preset particle size to pass through, and returning unqualified coarse particles to the grinding chamber;

[0036] Cyclone collector for separating and collecting qualified products screened by the integrated grading wheel from the airflow;

[0037] The integrated grading wheel and the cyclone collector together form a fully enclosed automated production loop.

[0038] Preferably, the pulse control module includes at least one high-frequency solenoid valve connected in series with the air intake pipe of the grinding nozzle.

[0039] A second aspect of this invention provides an automated dust-free production control method for konjac flour industry, the method comprising the following steps:

[0040] Step 1: Control the high-pressure airflow supplied to the grinding nozzle of a fluidized bed air jet mill to periodically switch on and off, so that the fluidized bed air jet mill operates under alternating working pulse periods and dormant conditioning periods;

[0041] Step 2: During the dormant conditioning period, the state parameters of the konjac material are sensed online in real time. The state parameters include at least temperature, particle size distribution and resonant frequency.

[0042] Step 3: Based on the state parameters sensed in Step 2, execute the following cooperative control steps:

[0043] Based on the temperature and particle size distribution, a conditioning medium is injected into the grinding chamber of the fluidized bed air jet mill to perform in-situ quality conditioning of the konjac material.

[0044] Based on the resonant frequency, the duration of one or more subsequent working pulse periods or the frequency of high-pressure gas flow pulses are adaptively adjusted.

[0045] This invention provides an automated dust-free production control system and method for konjac flour production. It has the following beneficial effects:

[0046] 1. This invention uses a pulse control module to periodically switch high-pressure airflow on and off, forcing a working pulse period and a dormant conditioning period within the grinding chamber. During the dormant conditioning period, when there is no high-speed airflow impact, the infrared temperature measurement array in the multi-dimensional online sensing module can accurately obtain the true temperature of the material. Once the temperature exceeds the safety threshold, the collaborative control core module can immediately drive the in-situ conditioning module to inject cooling medium for intervention. This alternating "grinding-sensing-conditioning" operation mode effectively avoids the thermal deterioration of materials caused by continuous heat accumulation during traditional continuous grinding processes, thereby ensuring the activity of heat-sensitive components such as konjac glucomannan and improving the quality of the final product.

[0047] 2. This invention utilizes the quiet sensing window provided by the dormant conditioning period to acquire key parameters such as the resonant frequency and particle size distribution that reflect the current crushing state of the material through an acoustic characteristic sensing unit and a machine vision unit. Based on these real-time parameters, the collaborative control core module adaptively adjusts the duration or frequency of one or more subsequent working pulse periods. This closed-loop feedback control mechanism can compensate for the impact of fluctuations in raw material characteristics or changes in equipment operating conditions in real time, ensuring that the crushing process is always maintained within the preset operating range, thereby significantly improving the stability of the finished product particle size distribution and the consistency of products between different production batches.

[0048] 3. This invention integrates three core processes—material crushing, multi-dimensional state sensing, and online quality conditioning—into a single fluidized bed jet mill through time-division processing. The system requires no separate cooling or conditioning equipment, nor does it require offline sampling and analysis. All operations are automatically cyclically executed within a sealed cavity under the management of the collaborative control core module. This design not only simplifies the physical layout and equipment composition of the production line but also constructs a complete, unmanned automated production loop, reducing the potential risk of cross-contamination. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the structure of an automated dust-free production control system for konjac flour industry according to an embodiment of the present invention;

[0050] Figure 2 This is a single-cycle control flowchart of the collaborative control core module according to an embodiment of the present invention;

[0051] Figure 3 This is a graph showing the changes of key system parameters over time in Embodiment 1 of the present invention.

[0052] Figure 4 This is a graph showing the changes in key parameters when the system responds to abnormal operating conditions in Embodiment 2 of the present invention. Detailed Implementation

[0053] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] Please refer to the appendix. Figure 1 , Figure 1 This is a schematic diagram of the structure of an automated cleanroom production control system for konjac flour industry according to an embodiment of the present invention. The automated cleanroom production control system for konjac flour industry provided by the present invention may include: a fluidized bed airflow pulverizer 100, a pulse control module 200, a multi-dimensional online sensing module 300, an in-situ conditioning module 400, and a collaborative control core module 500.

[0055] The fluidized bed air jet mill 100 is the main equipment for pulverizing konjac materials. It includes a grinding chamber 110, a grinding nozzle 120 communicating with the grinding chamber 110, an integrated classifying wheel 130, and a cyclone collector 140. The material is pulverized in the grinding chamber 110 and classified by the integrated classifying wheel 130. Qualified finished products are collected by the cyclone collector 140, while unqualified coarse particles are returned to the grinding chamber 110, forming a fully enclosed material circulation path.

[0056] The pulse control module 200 is physically connected in series in the air inlet pipe that supplies high-pressure gas to the grinding nozzle 120, and is electrically connected to the collaborative control core module 500. Based on commands issued by the collaborative control core module 500, it periodically controls the on / off flow of the high-pressure gas.

[0057] The multidimensional online sensing module 300 is installed on the grinding chamber 110 of the fluidized bed air jet mill 100 to collect the state parameters of the konjac material in the grinding chamber 110 and send the collected parameter signals to the collaborative control core module 500.

[0058] The in-situ conditioning module 400 is connected to the grinding chamber 110 via a pipeline and is electrically connected to the collaborative control core module 500. Based on instructions from the collaborative control core module 500, it injects a preset conditioning medium into the grinding chamber 110.

[0059] The collaborative control core module 500 is the system's control unit, and its hardware form can be a programmable logic controller (PLC) or an industrial control computer. The collaborative control core module 500 establishes electrical or communication connections with the pulse control module 200, the multi-dimensional online sensing module 300, and the in-situ conditioning module 400. It receives status parameters from the multi-dimensional online sensing module 300 and, according to its internal control logic, sends control commands to the pulse control module 200 and the in-situ conditioning module 400, forming a closed-loop control system.

[0060] Please see the appendix Figure 2 The collaborative control core module 500 is configured to execute a periodic control method. In each control cycle, the pulse control module 200 controls the on / off state of the high-pressure airflow, forming alternating working pulse periods and dormant conditioning periods within the grinding chamber 110. During the dormant conditioning period, the collaborative control core module 500 performs decision calculations based on the state parameters acquired by the multi-dimensional online sensing module 300.

[0061] In one specific embodiment, the collaborative control core module 500 processes signals acquired by the acoustic characteristic sensing unit of the multi-dimensional online sensing module 300 to calculate the system's resonant frequency. This calculation process includes: based on the received echo signal S... out (t) and the transmitted sweep frequency signal S in (t), calculate the acoustic transfer function H(f):

[0062] Where t is the time variable, f is the frequency variable, and S in (t) represents the swept frequency signal emitted by the acoustic transmitter, S out (t) represents the echo signal received by the acoustic sensor. This is the Fourier transform operator.

[0063] The resonant frequency f of the system is determined by detecting the peak value of the amplitude-frequency response |H(f)| of the transfer function H(f). res :

[0064] in, This is the operator that takes the value of the independent variable f when the function value is maximized.

[0065] The collaborative control core module 500 further uses the calculated resonant frequency f res (t) and the obtained temperature parameter Θ avg (t), calculate the duration T of the subsequent working pulse period. work The adjustment amount ΔT of (t) work (t). This calculation can be performed using a proportional-integral controller:

[0066]

[0067] Where e(t) is the system error term at the current time, and it is calculated as follows:

[0068]

[0069] Here, K P K is the proportionality coefficient. I τ is the integral coefficient; w is the integral variable; res This is the weighting coefficient for the resonant frequency error; The preset target resonant frequency; f res (t) is the resonant frequency measured at time t; w Θ Θ is the weighting coefficient for temperature error; avg (t) represents the average temperature of the particle swarm measured at time t; This is the preset safe threshold for the average temperature of the particle swarm.

[0070] In this invention, the feeding and discharging of the fluidized bed jet mill 100 are configured in a quasi-continuous or batch mode coordinated with a pulse control process. Specifically, its feed inlet is fed through a screw feeder or rotary valve controlled by a collaborative control core module 500, and the feeding rate can be determined according to the system resonant frequency f. res The amount of material in the grinding chamber 110 is dynamically adjusted to maintain a stable range. The discharge port (i.e., the discharge port of the cyclone collector 140) discharges material through a double-layer ash discharge valve to ensure that the pressure balance in the chamber is not disrupted or external gas is introduced when material is taken out in the entire closed system.

[0071] Please refer to the appendix. Figure 1 The figure also illustrates the deployment and integration of various hardware modules in one embodiment of the present invention.

[0072] In this embodiment, the collaborative control core module 500 is deployed in a separate electrical control cabinet, which has an industrial protection rating and is isolated from the on-site operating environment of the fluidized bed jet mill 100. The collaborative control core module 500 is electrically connected and communicates with the various modules on-site via shielded cables.

[0073] The pulse control module 200, specifically a high-frequency solenoid valve with a response time of less than 50 milliseconds, is installed on the main air intake line supplying all grinding nozzles 120 of the fluidized bed jet mill 100, located between the air compressor and the airflow distribution manifold of the grinding nozzles 120. This deployment method allows for synchronous on / off control of the grinding energy source within the entire grinding chamber 110 using a single control point.

[0074] The components of the multi-dimensional online sensing module 300 are installed in the upper middle part of the grinding chamber 110. This area has a moderate particle density and is far from the high-speed airflow impact zone of the grinding nozzle 120 during the dormant conditioning period. Specifically, the machine vision unit includes an industrial camera and an LED surface light source, which are respectively installed on two opposite flange interfaces with wear-resistant quartz glass windows on the grinding chamber 110. The optical axis of the camera is orthogonal to the optical axis of the light source or forms a preset angle to obtain a high-contrast image of the material.

[0075] The infrared temperature measurement array is a non-contact thermal imager, mounted on another observation flange on the side wall of the grinding chamber 110. Its field of view is configured to completely cover the main area of ​​material settling during the dormant conditioning period, thereby acquiring a two-dimensional thermal map containing the overall temperature distribution of the material particle swarm. The collaborative control core module 500 processes this thermal map data to calculate the average temperature Θ of the particle swarm. avg (t) and identify the instantaneous highest temperature Θ max (t).

[0076] The acoustic characteristic sensing unit includes a sound wave transmitter and an acoustic sensor. In this embodiment, both are directly and rigidly fixed to the outer wall of the grinding chamber 110, rather than inside the chamber. This deployment method utilizes the chamber wall as the sound wave conduction medium, avoiding direct exposure of the sensor to the high-wear environment inside the chamber. The sound wave transmitter and the acoustic sensor are installed at a distance and are not symmetrically arranged to collect vibration response signals that contain more information about the material inside the chamber and are coupled with the structure.

[0077] The execution components of the in-situ conditioning module 400 are deployed on the top or upper side wall of the grinding chamber 110. Specifically, at least one wide-angle conical nozzle is installed by opening a standard threaded interface in the chamber wall. The cooling gas supply unit delivers gas from a nitrogen source or cooling air source to the nozzle via a mass flow controller controlled by the co-control core module 500. The micro-atomization embrittlement unit delivers liquid from a deionized water supply device to another dedicated ultrasonic atomizing nozzle via a precision metering pump to generate water mist with a particle size between 1 and 10 micrometers.

[0078] In this embodiment, the pulse control module 200 receives instructions from the collaborative control core module 500 and precisely executes the on / off control of the high-pressure airflow, thereby generating alternating working pulse periods T within the grinding chamber 110 of the fluidized bed air jet mill 100. work (t) and dormancy conditioning period T rest .

[0079] In this embodiment, the pulse control module 200 consists of one or more parallel high-speed switching solenoid valves. These solenoid valves are connected in series on the main air supply line from the air compressor to the grinding nozzle 120 of the fluidized bed jet mill 100. To ensure instantaneous control, a two-position two-way direct-acting solenoid valve is selected, with a stainless steel body to withstand high-pressure airflow and polytetrafluoroethylene (PTFE) seals to meet clean production requirements. The electrical characteristics of this solenoid valve are configured such that when it receives a 24V DC drive signal from the collaborative control core module 500, the opening and closing times of its valve core are both less than or equal to 30 milliseconds.

[0080] The collaborative control core module 500 sends a pulse width modulation (PWM) signal or a periodic high-low level signal to the pulse control module 200 through its digital output (DO) channel.

[0081] When the collaborative control core module 500 outputs a high-level signal, the solenoid valve coil of the pulse control module 200 is energized, the valve core actuates to open the valve, and the high-pressure airflow (e.g., clean compressed air with a pressure of 0.6 to 0.8 MPa) passes through unimpeded and is supplied to the grinding nozzle 120, and the fluidized bed air jet mill 100 enters the working pulse period T. work (t) is used to perform high-speed collision crushing of konjac material in grinding chamber 110.

[0082] When the collaborative control core module 500 outputs a low-level signal, the solenoid valve coil of the pulse control module 200 is de-energized. Under the action of the reset spring, the valve core instantly closes the valve, and the high-pressure airflow supply is cut off. The grinding nozzle 120 stops spraying, the grinding energy in the grinding chamber 110 rapidly decays, and the system enters the dormant conditioning period T. rest .

[0083] Working pulse period T work Duration of (t) and dormancy conditioning period T rest The duration is dynamically set by the collaborative control core module 500 based on preset process parameters and real-time feedback data. A complete control cycle T... cycle It equals the sum of the working pulse duration and the dormancy conditioning period duration, i.e., T cycle =T work +T rest This precise on / off control of the high-pressure airflow source is the technical prerequisite for realizing the subsequent sensing, conditioning, and adaptive adjustment functions of the entire system.

[0084] See attached document Figure 1 The multidimensional online sensing module 300 is used during the dormancy and conditioning period T restThe system non-contactly acquires the physical state parameters of the konjac material within the grinding chamber 110, quantifies these parameters, and transmits them to the collaborative control core module 500. In this embodiment, the multi-dimensional online sensing module 300 includes a machine vision unit, an infrared temperature measurement array, and an acoustic characteristic sensing unit 330.

[0085] The machine vision unit is used to acquire the particle size distribution and color of the material. In this embodiment, it includes an industrial camera with a resolution of 1920x1080 pixels, a matching telecentric lens, and a strobe LED surface light source. The camera and light source are mounted on the upper part of the grinding chamber 110 via flanges, with their optical axes perpendicular to each other. During each dormant conditioning period T... rest Inside, when some material particles settle due to gravity and form a relatively sparse particle cloud, the collaborative control core module 500 triggers the LED surface light source to perform a short strobe illumination with a duration of 1 millisecond, and at the same time triggers the industrial camera to synchronously acquire a frame of image.

[0086] The acquired images are transmitted to the collaborative control core module 500 or a dedicated image processing module. Using image processing algorithms such as edge detection, contour recognition, and Hough transform, individual material particles in the image are identified, and their equivalent diameters are calculated. By statistically analyzing the equivalent diameters of all identified particles within the field of view, a particle size distribution histogram of the current material is obtained, and key particle size parameters, such as D50 (median diameter) and D90 (90% of particles are smaller than this diameter), are calculated. Simultaneously, by analyzing the RGB or HSV color space values ​​of the material particle regions in the image, the chromaticity of the material is quantified for monitoring indicators such as product whiteness.

[0087] An infrared thermometer array is used to acquire the temperature information of the material. In this embodiment, it is a long-wave infrared thermal imager with an array size of 160x120 pixels, a spectral response range of 8 to 14 micrometers, and a temperature measurement range of 0 to 150 degrees Celsius. It is mounted on the side wall of the grinding chamber 110 via a flange with a built-in germanium glass window, and its lens field of view is set to cover the main spatial area of ​​material settling during the dormant conditioning period.

[0088] During the dormancy conditioning period T rest Inside, the collaborative control core module 500 triggers the infrared temperature measurement array to acquire a frame of thermal image. After receiving the thermal image, the collaborative control core module 500 processes the temperature value of each pixel in the image: calculating the arithmetic mean of the temperature values ​​of all pixels to obtain the particle swarm average temperature Θ. avg (t); simultaneously, the highest temperature value among all pixels is retrieved and recorded to obtain the instantaneous highest temperature Θ. max (t). These two parameters are used for subsequent temperature control and conditioning decisions and adaptive adjustment calculations.

[0089] The acoustic characteristic sensing unit 330 is used to obtain the resonant frequency f of the coupled system formed by the current material and the grinding chamber 110. rest The frequency is related to the filling amount, density, and particle state of the material inside the cavity. In this embodiment, it includes a piezoelectric ceramic acoustic wave transmitter and a high-sensitivity accelerometer.

[0090] During the dormancy conditioning period T rest At the beginning, and at a specific point in time after the material has completed its initial settling, the collaborative control core module 500 drives the acoustic transmitter to emit a sweeping sinusoidal signal S with a duration of 100 milliseconds towards the wall of the grinding chamber 110. in (t), with a frequency range linearly scanning from 1 kHz to 10 kHz. Simultaneously, an accelerometer installed on another part of the cavity wall begins acquiring the vibration response signal S of the cavity wall. out (t).

[0091] After data collection is completed, the collaborative control core module will control 500 pairs of S... in (t) and S out (t) Perform a Fast Fourier Transform (FFT) to obtain their respective frequency domain representations, and determine the system's resonant frequency f using the aforementioned transfer function calculation formula. res This resonant frequency, as a comprehensive indicator, reflects the macroscopic progress of the current crushing operation and serves as a factor for adaptively adjusting the working pulse period T. work One of the core input parameters for duration.

[0092] In another embodiment, the acoustic wave emitter of the acoustic characteristic sensing unit 330 can be replaced by a small electromagnetic vibratory hammer mounted on the outer wall of the grinding chamber 110. During the dormant conditioning period, the co-control core 500 drives the vibratory hammer to deliver a momentary, energy-controlled mechanical impact to the chamber wall, thereby exciting a broadband vibration response within the chamber. An accelerometer mounted elsewhere collects the damped vibration signal caused by this impact, and by performing spectral analysis on this signal, the resonant frequency f of the system can also be determined. res This method provides richer information about low-frequency response.

[0093] Please refer to the appendix. Figure 1 The in-situ conditioning module 400 is used during the dormant conditioning period T. rest Within the grinding chamber 110, based on the decision instructions of the collaborative control core module 500, a conditioning medium is precisely injected to actively intervene in the physical properties of the material. In this embodiment, the in-situ conditioning module 400 includes a cooling air path unit and a micro-atomization embrittlement unit.

[0094] The function of the cooling gas circuit unit is to inject inert cooling gas into the grinding chamber 110 when the material temperature exceeds the preset safety threshold, so as to absorb heat, control the temperature rise, and prevent the konjac material from being thermally damaged.

[0095] In this embodiment, the unit includes a high-purity nitrogen source (such as a liquid nitrogen Dewar flask or a nitrogen cylinder assembly), a stainless steel delivery pipeline, a precision digital mass flow controller located on the pipeline, and a wide-angle nozzle mounted on top of the grinding chamber 110. The digital mass flow controller is connected to the analog output (AO) channel or fieldbus interface of the collaborative control core module 500.

[0096] When the collaborative control core module 500 determines the average temperature Θ of the particle swarm based on the data from the infrared temperature measurement array... avg (t) or instantaneous highest temperature Θ max (t) exceeds its corresponding security threshold (e.g., When the flow rate is set, it sends a control signal to the digital mass flow controller. This signal specifies a precise flow rate value (e.g., 50 standard liters per minute) and an injection duration (e.g., 200 milliseconds). The controller then activates and injects nitrogen into the grinding chamber 110 according to the set value. The injected cryogenic nitrogen expands rapidly and exchanges heat with the suspended material particles, achieving rapid and uniform cooling of the material.

[0097] The function of the micro-atomization embrittlement unit is to inject micron-level water mist into the grinding chamber 110 when a decrease in grinding efficiency is detected, so as to slightly increase the embrittlement of the konjac material particles and thus enhance their embrittlement during the next working pulse T. work It internally improves the efficiency of its crushing.

[0098] In this embodiment, the unit includes a deionized water storage tank, a delivery pipeline, a micro-injection pump controlled by the collaborative control core module 500, and an ultrasonic atomizing nozzle mounted on the upper side wall of the grinding chamber 110. The ultrasonic atomizing nozzle is capable of atomizing liquid into fine droplets with an average particle size of less than 10 micrometers.

[0099] When the collaborative control core module 500 calculates the crushing efficiency index η based on the data from the machine vision unit... D (t) (e.g., determined by comparing the rate of change of the material's D90 particle size over several consecutive periods) is below a preset efficiency threshold η minAt this time, it drives a micro-injection pump to perform a precise injection. The injection pump delivers a tiny volume (e.g., 0.05 ml) of deionized water through tubing to the ultrasonic atomizing nozzle. The nozzle atomizes the water and sprays it into the grinding chamber 110. These micron-sized water droplets uniformly adhere to the surface of the material particles, altering their physical and mechanical properties, making them more susceptible to brittle fracture during subsequent airflow impact. The injected amount is strictly controlled to ensure that it does not cause significant humidification or adhesion of the material.

[0100] In one embodiment of the present invention, the industrial automated dust-free production control system for konjac flour undergoes an initialization and calibration phase before being put into formal production operation. The purpose of this phase is to set basic parameters for the control algorithm of the collaborative control core module 500 and to establish a benchmark for sensor measurements, ensuring the accuracy and stability of subsequent automated production.

[0101] Initialization parameter settings:

[0102] After the system starts, the operator inputs or loads a set of process parameter formulas for a specific batch of konjac raw materials and target finished product specifications through a human-machine interface (HMI) connected to the collaborative control core module 500. These parameters are stored in the non-volatile memory of the collaborative control core module 500, including but not limited to:

[0103] Control cycle parameter: initial working pulse duration (e.g., 2.0 seconds) and the duration of the dormancy conditioning period (For example, 1.5 seconds).

[0104] Temperature safety threshold: Particle swarm average temperature safety threshold (e.g., 55°C) and instantaneous maximum temperature safety threshold (For example, 65°C).

[0105] Efficiency and quality target parameters: Median diameter D50 of the target finished product target (e.g., 100 micrometers), pulverization efficiency threshold η min And the color range of the target product.

[0106] Adaptive adjustment algorithm parameters: including the proportional coefficient K in the proportional-integral (PI) controller used to calculate the adjustment amount during the working pulse period. P (e.g., 0.05) and the integral coefficient K I (e.g., 0.01); and the weighting coefficient w used to calculate the systematic error term e(t). res (e.g., 0.7) and w Θ (For example, 0.3).

[0107] proportionality coefficient K Pand integral coefficient K I The specific value can be determined through the Ziegler-Nichols tuning method, well-known to those skilled in the art, or through trial and error during actual system operation. The goal is to ensure the system exhibits fast, stable, and low overshoot characteristics in response to disturbances. Weighting coefficient w res and w Θ The setting depends on the emphasis placed on particle size consistency or temperature sensitivity in the production process for that batch. If temperature control requirements are extremely stringent, w can be appropriately increased. Θ The value of .

[0108] System no-load calibration:

[0109] After the parameters are set, the system performs an empty-load calibration process. With no material in the grinding chamber 110, the collaborative control core module 500 drives the system to run for several control cycles.

[0110] During this period, the acoustic characteristic sensing unit 330 performs frequency sweeping and signal acquisition, and coordinates with the core control module 500 to calculate and record the system resonant frequency under no-load conditions. This value will serve as a benchmark for subsequent calculations of material-related resonant frequency offsets.

[0111] Meanwhile, the machine vision unit and infrared temperature measurement array also collect data to obtain background noise and ambient temperature, and perform dark current correction or background subtraction to provide an accurate zero-point reference for subsequent material measurement.

[0112] Reference material calibration:

[0113] In a preferred embodiment, a baseline material calibration is further performed. A small amount (e.g., 5% of the volume of grinding chamber 110) of standard konjac flour raw material with known properties is added to grinding chamber 110. The system is run for several cycles at initial parameters until the material reaches a relatively stable fluidized state.

[0114] At this point, the collaborative control core module 500 records various parameters of the benchmark material under stable conditions. The most crucial parameter is the recording of the resonant frequency at this point, which is then set as the target resonant frequency in the adaptive adjustment algorithm. This provides the closed-loop control system with a clear control objective physically associated with the ideal crushing state, giving subsequent adaptive adjustments a physical anchor. After this step, the initialization and calibration phase ends, the system is ready to receive large quantities of material and enter fully automated production operation mode.

[0115] After initialization and calibration, the system enters a continuous automated production operation state, the core of which is a continuously looping "work-sleep" control process. This process is led by the collaborative control core module 500, with each complete control cycle T... cycle Including a working pulse period T work and a dormant conditioning period T rest The following will describe in detail the specific process within the nth control cycle.

[0116] Step A: Working pulse period (T) work (n))

[0117] At the start time t0 of cycle n, the collaborative control core module 500 sends a high-level command to the pulse control module 200. The high-frequency solenoid valve opens, and high-pressure airflow is instantaneously injected into the grinding chamber 110, marking the start of the system's working pulse period. During this period, the konjac material forms a fluidized bed under the action of the high-speed airflow, resulting in violent collisions between particles and between particles and the chamber wall, achieving pulverization. Simultaneously, the integrated classifying wheel 130 operates at a preset speed, separating the fine powder that meets the particle size requirements and sending it to the cyclone collector 140. The duration T of this stage is... work (n) is calculated using the adaptive adjustment algorithm of the previous cycle (n-1). At t0+T work At time (n), the working pulse period ends.

[0118] Step B: Dormant Conditioning Period (T) rest )

[0119] At time t0+T work (n), the collaborative control core module 500 sends a low-level command to the pulse control module 200. The high-frequency solenoid valve closes, and the high-pressure airflow supply is cut off. The airflow in the grinding chamber 110 rapidly weakens, and the material begins to settle under gravity, entering a dormant conditioning period. The duration of this stage is T. rest This is typically a fixed value (e.g., 1.5 seconds) to ensure sufficient time for sensing and conditioning operations. During this phase, the collaborative control core module 500 executes the following sub-steps sequentially or in parallel:

[0120] Sub-step B1: After a short delay Δt1 (e.g., 200 milliseconds) after the start of the dormancy conditioning period, the state perception core module 500 triggers the multi-dimensional online sensing module 300 to collect synchronous data once the dust turbulence in the cavity has initially subsided.

[0121] The strobe light source and camera of the machine vision unit are triggered to capture an image containing material particles.

[0122] An infrared temperature measurement array acquires a frame of thermal image of a material.

[0123] The acoustic wave transmitter of the acoustic characteristic sensing unit 330 is driven to emit a sweep frequency signal, while its accelerometer begins to collect the vibration response signal of the cavity wall.

[0124] Sub-step B2: The decision analysis and collaborative control core module 500 receives and processes the raw data collected in sub-step B1:

[0125] The machine vision image is processed to calculate the particle size distribution D90 value of the current material, and compared with the D90 value of the previous period to obtain the grinding efficiency index η. D (n).

[0126] The thermal image was processed to calculate the average temperature Θ of the particle swarm. avg (n) and instantaneous highest temperature Θ max (n).

[0127] The acoustic signal is processed to calculate the resonant frequency f of the current system. res (n).

[0128] Subsequently, the collaborative control core module 500 executes the decision-making logic:

[0129] Temperature Decision: Comparison Θ avg (n) and Θ max (n) and their respective security thresholds and If any temperature exceeds the limit, a cooling and conditioning command will be generated.

[0130] Efficiency Decision: Comparing Grinding Efficiency Indicators η D (n) and efficiency threshold η min If the value is below the threshold, a tempering instruction for increased brittleness is generated.

[0131] Adjustment decision: f res (n) and Θ avg (n) Substitute the input into the preset PI control algorithm formula to calculate the duration adjustment ΔT for the next working pulse period. work (n), and update the working pulse duration: T work (n+1)=T work (n)+ΔT work (n). Simultaneously for T work The value of (n+1) is limited to ensure that it is within a reasonable range (e.g., 0.5 to 5.0 seconds).

[0132] To prevent excessive accumulation of the integral term during large step changes in system state, which could lead to prolonged system overshoot, the PI control algorithm also includes anti-windup logic. When the calculated T... workWhen (n+1) reaches the preset upper or lower limit, the collaborative control core module 500 will temporarily stop accumulating the integral term until the sign of the system error e(t) changes or returns to the predetermined range.

[0133] Sub-step B3: Quality Conditioning. Based on the decision result of sub-step B2, the collaborative control core module 500 issues specific instructions to the in-situ quality conditioning module 400:

[0134] If a cooling conditioning command is generated, the cooling gas path unit is driven to inject cooling nitrogen according to the preset flow rate and duration.

[0135] If a tempering and embrittlement command is generated, the micro-atomization tempering unit is driven to inject micron-level water mist according to a preset volume.

[0136] If no conditioning instruction is generated, this sub-step is skipped.

[0137] At time t0+T work (n)+T rest The dormancy conditioning period ends, completing a full control cycle. The system then enters the (n+1)th control cycle, with the updated working pulse duration T. work (n+1) Start a new loop.

[0138] Example 1:

[0139] The following specific embodiment will illustrate the automated operation process of the present invention under standard operating conditions.

[0140] In this embodiment, the production objective is to process a batch of dried konjac slices with an initial particle size of 40-60 mesh to obtain finished konjac powder with a D90 particle size of 120 micrometers. After the system initialization and calibration phases are completed, the key process parameters set within the collaborative control core module 500 are as follows: initial working pulse duration. Seconds; Duration of dormancy conditioning period T rest = 1.5 seconds; Particle swarm average temperature safety threshold Instantaneous maximum temperature safety threshold Target resonant frequency kilohertz. The parameters of the adaptive adjustment algorithm are set as follows: K P =0.05, K I =0.01, w res =0.7, w Θ =0.3.

[0141] After the system starts continuous feeding, it enters a "work-sleep" cycle control process. In the 10th control cycle, the system's operating status is as follows:

[0142] During the working pulse period, the pulse control module 200 is activated for 2.15 seconds. It then enters a rest-dormant conditioning period. The multi-dimensional online sensing module 300 is triggered, and the collected state parameters are: particle swarm average temperature Θ. avg (10) = 51℃; instantaneous maximum temperature Θ max (10) = 59℃; System resonant frequency f res (10) = 5.35 kHz.

[0143] The collaborative control core module 500 performs decision analysis. First, temperature and efficiency checks are performed. Since 51℃ < 55℃ and 59℃ < 65℃, and the pulverization efficiency index is higher than the threshold, no operation of the in-situ conditioning module 400 is triggered. Next, the system error term e(10) for adjusting the next working pulse period is calculated:

[0144] e(10)=0.7·(5.50-5.35)-0.3·(51-55)=0.7·(0.15)-0.3·(-4)=0.105+1.2=1.305;

[0145] Based on this error, calculate the adjustment amount ΔT during the working pulse period. work (10) (For the sake of simplification, the cumulative effect of the integral terms is not considered here):

[0146] ΔT work (10)≈K P ·e(10)=0.05·1.305=0.065 seconds;

[0147] The collaborative control core module 500 updates the duration of the next working pulse period as follows:

[0148] T work (11)=T work (10)+ΔT work (10) = 2.15 + 0.065 = 2.215 seconds;

[0149] The system ran continuously, and after approximately 50 control cycles, it reached a dynamically stable state. In the 60th control cycle, the acquired parameter was: Θ avg (60) = 54℃, f res (60) = 5.49 kHz. The system error term e(60) is calculated as follows:

[0150] e(60)=0.7·(5.50-5.49)-0.3·(54-55)=0.7·(0.01)-0.3·(-1)=0.007+0.3=0.307;

[0151] The working pulse period adjustment ΔT calculated at this time work(60) becomes very small, working time T work The fluctuations remained stable at approximately 2.3 seconds. This indicates that the system, through adaptive adjustment, achieved a balance between the crushing intensity and the material state, ensuring that the system's resonant frequency closely followed the target value, while maintaining the material temperature at the upper edge of the safe range, thus achieving stable operation.

[0152] After 30 minutes of continuous production under this stable operating condition, a sample was taken from point 140 of the cyclone collector for inspection. The D90 particle size of the finished powder, measured using a laser particle size analyzer, was 121.5 micrometers, deviating from the target value of 120 micrometers by 1.25%. The whiteness value of the product, measured using a colorimeter, was 86, meeting the specifications for high-quality konjac flour. The entire process was fully automated, requiring no manual intervention.

[0153] Please see the appendix Figure 3 The graph accurately illustrates the dynamic operation process of the system in Embodiment 1 of the present invention, presented in the form of a data curve. The graph clearly reproduces the entire process from system startup to the "adaptive adjustment phase" (approximately 50 cycles), and finally to the "stable operation phase."

[0154] Specifically, the upper subplot shows the duration T of the working pulse period. work The curve changes. Starting from an initial value of approximately 2.0 seconds, it gradually increases under the PI regulation of the collaborative control core module 500, and converges and maintains near the optimal value of approximately 2.3 seconds during the stable operation phase. This is consistent with the implementation example where error calculation drives T. work The growth process matches perfectly.

[0155] The middle subplot shows the variation of the system's resonant frequency in kilohertz (kHz). The curve shows that the resonant frequency steadily increases from a low initial value (approximately 5.3 kHz), indicating that as the pulverization progresses, the material morphology improves and the system efficiency increases. During the stable operation phase, the actual resonant frequency successfully converged to the target value of 5.50 kHz (shown by the red dashed line in the figure), and its value at the 60th cycle (approximately 5.49 kHz) is highly consistent with the calculated value in the example.

[0156] The lower subplot shows the average temperature Θ of the particle swarm. avg The curve shows that the average temperature gradually increases from the initial room temperature, but under the control strategy of this invention, it is stabilized at about 54°C in the 60th cycle, and is always effectively controlled below the temperature safety threshold of 55°C (shown by the red dotted line in the figure). This strongly proves that this invention successfully avoids overheating and quality deterioration of konjac materials while pursuing high efficiency.

[0157] In summary, attached Figure 3The simulation curves are highly consistent with the description and calculation data of Example 1, which together and strongly verify that the control method of the present invention can guide the system to achieve rapid and accurate adaptive adjustment, and finally achieve a highly efficient and stable operating state with high crushing efficiency (resonant frequency meets the standard), excellent product quality (temperature is controlled) and no manual intervention required.

[0158] Example 2:

[0159] The following specific embodiment will illustrate the intervention and response capabilities of the present invention in the event of abnormal operating conditions.

[0160] The initial state of this embodiment follows the stable operating state of Embodiment 1, where the system operates stably with a working pulse duration of approximately 2.3 seconds, the average material temperature is maintained at 54°C, and the system resonant frequency is stable at approximately 5.49 kHz. The set crushing efficiency threshold η... min -5% (meaning the reduction rate of D90 particle size in a single cycle must not be less than 5%).

[0161] When the system reached its 100th control cycle, the raw materials in the upstream feeding system changed. A batch of dried konjac slices with a higher moisture content (moisture content increased from 8% to 11%) entered the fluidized bed air jet mill 100. This change resulted in increased material toughness and decreased grindability.

[0162] During the dormancy conditioning period of the 101st control cycle, the state parameters collected by the multi-dimensional online sensing module 300 showed significant deviations:

[0163] The collaborative control core module 500 calculates, based on data from the machine vision unit, that the D90 particle size has decreased by only 3% compared to the previous cycle, i.e., the crushing efficiency index η... D (101) = -3%, which is lower than the preset efficiency threshold η. min .

[0164] The average temperature Θ of the particle swarm was measured by an infrared thermometer array. avg (101) The temperature rose to 56°C, exceeding the safety threshold of 55°C.

[0165] The acoustic characteristic sensing unit 330 measures the system resonant frequency f. res (101) decreased to 5.25 kHz, which is significantly different from the target value of 5.50 kHz, indicating that the proportion of coarse particles in the cavity increased.

[0166] The collaborative control core module 500 immediately executes decisions and interventions. During the remainder of the dormancy and conditioning period of the 101st cycle, the following operations are performed:

[0167] Due to η D (101)<ηmin The collaborative control core module 500 sends a command to the micro-atomization and embrittlement unit of the in-situ conditioning module 400, driving the micro-injection pump to inject 0.05 ml of deionized water mist into the grinding chamber 110.

[0168] because The collaborative control core module 500 sends a command to the cooling gas circuit unit to inject cooling nitrogen gas at a flow rate of 50 standard liters / minute for a duration of 200 milliseconds.

[0169] Meanwhile, the collaborative control core module 500, based on f which contains a large deviation... res (101) and Θ avg (101) Calculate the duration T of the next working pulse period. work (102), which resulted in a significant increase to 2.65 seconds, providing stronger crushing energy to handle materials with increased toughness.

[0170] In the following cycles, the system continuously intervened and adaptively adjusted. During the 102nd working pulse, due to the embrittlement of the material surface by a trace amount of water mist, the pulverization efficiency recovered under the enhanced pulverization action of 2.65 seconds. The injection of cooling gas during the dormant conditioning period effectively controlled the temperature. By the 110th control cycle, the system reached a new dynamic equilibrium state. The parameter measured at this time was: average particle swarm temperature Θ. avg (110) = 53℃ (back to safe range); crushing efficiency index η D (110) Recovered to -6%; System resonant frequency f res (110) Rebounded to 5.48 kHz. Operating pulse duration T work It then stabilized at a new, higher plateau value of around 2.80 seconds.

[0171] This embodiment demonstrates that the system of the present invention can automatically detect abnormal operating conditions caused by raw material fluctuations, and actively intervene through the synergistic effect of the in-situ conditioning module and the adaptive adjustment algorithm. Without human intervention, it can bring the key parameters of the production process back to a controlled state, ensuring the stability of the final product quality and enabling the production system to adapt to changes in raw materials.

[0172] Please see the appendix Figure 4 The graph, presented in curve form, vividly and accurately illustrates the dynamic response and self-recovery capability of the system in Embodiment 2 of the present invention when encountering abnormal operating conditions caused by raw material fluctuations. The entire process can be broken down into the following key stages:

[0173] I. Stable Operation Phase (approximately the first 100 cycles in the diagram): The system maintains the stable state of Implementation Example 1, with a working pulse period T. workThe resonant frequency and average temperature both operate stably within their respective target ranges, and their values ​​are consistent with the initial state description of Example 2.

[0174] II. Occurrence of Abnormal Operating Conditions (Cycle 101 in the diagram): At this moment, the system encounters an anomaly caused by the increased moisture content of the raw materials. This event is immediately reflected in the monitoring parameters: the average temperature of the particle swarm (lower subplot) curve shows a sharp peak, rising to 56℃, instantly exceeding the temperature safety threshold of 55℃ (red dashed line); at the same time, the system resonant frequency (middle subplot) curve shows a significant steep drop, decreasing from approximately 5.5kHz to 5.25kHz, indicating a sharp deterioration in pulverization efficiency.

[0175] III. System Rapid Response and Intervention: The collaborative control core module 500 detected this abnormal state in real time. Its response measures are clearly visible in the diagram: First, to address the issues of increased material toughness and reduced efficiency, the controller immediately adjusted the working pulse period T... work The duration of (upper sub-diagram) has been significantly increased from approximately 2.3 seconds to 2.65 seconds to provide stronger crushing energy; secondly, the collaborative control core module 500 will immediately issue instructions to the in-situ conditioning module 400 (such as starting cooling and micro-atomization embrittlement), which is marked as "conditioning module intervention" in the figure.

[0176] IV. Recovery to a New Stable Phase (After the 102nd cycle in the diagram): Under the intelligent intervention of the control core, the system begins to recover. The temperature curve quickly drops below the safe threshold and stabilizes at approximately 53°C. The resonant frequency also begins to gradually rise, eventually converging again to near the target value around the 110th cycle. It is worth noting that the operating pulse period T... work It eventually stabilized at a new, higher plateau (approximately 2.8 seconds), indicating that the system successfully adapted to the new feedstock properties.

[0177] Appendix Figure 4 The results are highly consistent with the description and data of Example 2, convincingly demonstrating that the present invention can not only diagnose faults, but also actively intervene and adapt to changes in production conditions through the synergistic effect of in-situ conditioning and adaptive control, bringing the system back to a controlled state without human intervention, demonstrating a very high level of intelligence and production robustness.

Claims

1. A konjac flour industrial automation dust-free production control system, characterized in that, The application relates to a fluidized bed jet mill, which comprises a grinding cavity and a grinding nozzle in communication with the grinding cavity. A pulse control module is connected with an air inlet pipeline of the grinding nozzle and used for controlling the on-off of high-pressure airflow supplied to the grinding nozzle. A multi-dimensional online sensing module is installed on the fluidized bed jet mill and used for acquiring state parameters of konjac materials in the grinding cavity in real time. The multi-dimensional online sensing module comprises: a machine vision unit used for acquiring particle size distribution and chroma characteristics of the konjac materials; an infrared temperature measurement array used for acquiring particle group average temperature and instantaneous maximum temperature of the konjac materials; an acoustic characteristic sensing unit used for acquiring a resonance frequency of a system composed of the konjac materials and the grinding cavity; an in-situ conditioning module in communication with the grinding cavity and used for injecting a conditioning medium into the grinding cavity; The in-situ conditioning module comprises: a cooling gas pipeline unit used for injecting cooling gas to control the temperature of the konjac materials; a trace atomization and brittleness increasing unit used for injecting micron-sized water mist to adjust the brittleness of the konjac materials; a cooperative control core module electrically connected with the pulse control module, the multi-dimensional online sensing module and the in-situ conditioning module; The cooperative control core module is configured to control the pulse control module to supply the high-pressure airflow in a pulse mode, so that alternating working pulse periods and dormant conditioning periods are formed in the grinding cavity, and based on the state parameters acquired by the multi-dimensional online sensing module in the dormant conditioning periods, the in-situ conditioning module is controlled to perform quality conditioning, and the running time of the subsequent working pulse period or the frequency of the high-pressure airflow pulse is adaptively adjusted. The acoustic characteristic sensing unit comprises: a sound wave transmitter used for emitting a sweep frequency signal into the grinding cavity in the dormant conditioning period; an acoustic sensor used for receiving echo signals after the action of the konjac materials, so that the cooperative control core module calculates the resonance frequency. The cooperative control core module is further configured to:

2. The industrial automatic dust-free production control system of konjak flour according to claim 1, characterized in that, adaptively adjust the length of the working pulse period or the frequency of the high-pressure airflow pulse according to the resonance frequency measured by the acoustic sensor. The cooperative control core module is further configured to:

3. The industrial automatic dust-free production control system of konjak flour according to claim 1, characterized in that, when the particle group average temperature or the instantaneous maximum temperature acquired by the infrared temperature measurement array exceeds a preset safety threshold, in the dormant conditioning period, the cooling gas pipeline unit is controlled to inject cooling gas. The cooperative control core module is further configured to:

4. The industrial automatic dust-free production control system of konjak flour according to claim 1, characterized in that, when a crushing efficiency index calculated according to the particle size distribution acquired by the machine vision unit is lower than a preset efficiency threshold, in the dormant conditioning period, the trace atomization and brittleness increasing unit is controlled to inject micron-sized water mist. The fluidized bed jet mill further comprises an integrated classification wheel and a cyclone collector, which jointly constitute a fully-closed automatic production loop.

5. The industrial automatic dust-free production control system of konjak flour according to claim 1, characterized in that, The integrated classification wheel is used for dynamically classifying materials leaving the grinding cavity, only allowing qualified products reaching a preset particle size to pass through, and returning unqualified coarse particles to the grinding cavity. The cyclone collector is used for separating and collecting qualified products screened by the integrated classification wheel from airflow. ​ 6. The industrial automatic dust-free production control system of konjak flour according to claim 1, characterized in that, The pulse control module comprises at least one high-frequency electromagnetic valve connected in series with the gas inlet pipeline of the grinding nozzle.

7. A method for the automatic control of the production of konjac flour in an industrial, dust-free environment, characterized by the fact that The application is applied to the automatic dust-free production control system of the konjak powder industry according to any one of claims 1-6, and comprises the following steps: The high-pressure gas flow supplied to the grinding nozzle of the fluidized bed jet mill is periodically turned on and off, so that the fluidized bed jet mill operates in the alternating working pulse period and dormant conditioning period; During the dormant conditioning period, the state parameters of the konjak material are sensed in real time and online, and the sensing step of the state parameters comprises: The average temperature and the instantaneous maximum temperature of the konjak material are obtained by an infrared temperature measurement array; The particle size distribution and the colorimetric characteristics of the konjak material are obtained by a machine vision unit; A sweep signal is emitted into the grinding cavity by a sound wave emitter, and an acoustic sensor receives the echo signal, and the resonance frequency of the system composed of the konjak material and the grinding cavity is calculated; Based on the state parameters, the following cooperative control steps are performed: The quality of the konjak material is conditioned in situ by injecting cooling gas through a cooling gas path unit or injecting micron-sized water mist through a trace atomization and embrittlement unit; According to the resonance frequency, the running time of the subsequent one or more working pulse periods or the frequency of the high-pressure gas flow pulse is adaptively adjusted.

Images