Method and system for controlling uniformity of antimicrobial coating of silica gel cat litter

Abstract

The present application relates to the technical field of automatic control, in particular to a silica gel cat litter antibacterial coating uniformity control method and system, comprising the following steps: collecting spectrum screening characteristic waveband and setting film formation rate standard, performing logarithmic transformation on fluidized bed reflected light data to generate real-time absorbance, performing division operation to generate real-time coating distribution ratio, constructing cache sequence to analyze change slope to generate actual film formation rate index, and calculating pressure compensation amount according to deviation and generating regulation and control voltage instruction. In the present application, a standard is established by collecting spectrum signals, absorbance data is generated based on logarithmic transformation, a coverage index is obtained by performing division operation, a distribution ratio change slope is constructed by sequence analysis, the actual rate is compared with the standard to calculate the pressure compensation amount and generate the regulation and control instruction, the problem of uneven distribution caused by value control is solved by realizing closed-loop control, and the cumulative speed of the antibacterial agent is ensured to meet the expectation and the coating uniformity is improved.

Description

Technical Field

[0001] This invention relates to the field of automatic control technology, and in particular to a method and system for controlling the uniformity of silicone cat litter antibacterial coating. Background Technology

[0002] The field of automatic control technology encompasses the use of control devices to enable controlled objects or processes to operate automatically according to predetermined rules. Its core aspects include the acquisition of physical quantities by sensors and the transmission of signals to a controller. The controller calculates deviations based on preset control logic and then issues commands to the actuators to adjust the system state, thereby regulating parameters such as temperature, pressure, flow rate, or mechanical motion. This field systematically integrates detection technology, signal transmission technology, and computer control technology, and is widely used in the automation of industrial production processes. One example is the traditional method for controlling the uniformity of antibacterial coatings on silica gel cat litter. This method involves applying liquid antibacterial agents to the surface of solid silica gel granules. A mechanical stirring drum is used as a carrier, and a liquid pump delivers the antibacterial solution to a nozzle fixed above the granules. A programmable logic controller (PLC) is used to set a fixed spraying time and stirring speed. During operation, the controller outputs a constant voltage signal to the motor driver to maintain a constant speed of the stirring blades and simultaneously energizes the nozzle solenoid valve, continuously opening the valve for a preset time period to complete the coating operation.

[0003] Existing traditional processes rely on mechanical stirring drums and fixed nozzles. A controller is used to set a fixed spraying time and stirring speed, outputting a constant voltage to the motor driver to maintain the blade operation and open the valve at preset times. This open-loop fixed-value control mode ignores the dynamic changes in the particle motion state and surface adsorption capacity within the fluidized bed. It lacks a real-time monitoring and feedback mechanism for the actual coating effect, resulting in uneven coating distribution on the particle surface and difficulty in ensuring batch consistency. It is also impossible to compensate for rate deviations in a timely manner, leading to waste of antibacterial agents or incomplete coverage, thereby reducing the antibacterial performance of the product. Summary of the Invention

[0004] To address the technical problems existing in the prior art, embodiments of the present invention provide a method for controlling the uniformity of silicone cat litter antibacterial coating, comprising the following steps:

[0005] S1: Collect the fluidized bed reflectance spectrum of uncoated silicone particles and the transmission spectrum signal of pure antibacterial agent solution, screen the peak band of light intensity attenuation of silicone substrate and the peak band of antibacterial agent response, set the lower limit threshold of film formation rate of coating accumulation speed on particle surface per unit time, and generate characteristic wavelength and film formation rate standards of substrate and coating.

[0006] S2: Based on the characteristic wavelengths and film formation rate standards of the substrate and coating, collect the reflected light intensity data of the fluidized bed surface inside the rotating drum, perform logarithmic transformation on the reflected light intensity data, and generate real-time absorbance data of the substrate and coating;

[0007] S3: Call the real-time absorbance data of the substrate and coating, perform a division operation between the real-time absorbance data of the coating and the real-time absorbance data of the substrate, obtain the relative coverage index of the antibacterial agent, and generate the real-time coating distribution ratio;

[0008] S4: Construct a first-in-first-out data buffer sequence for the real-time coating distribution ratio, calculate the difference amplitude of the ratio at the beginning and end of the first-in-first-out data buffer sequence, analyze the slope of the change of the real-time coating distribution ratio over time, and generate the actual film formation rate index.

[0009] S5: Compare the actual film formation rate index with the lower limit threshold of the film formation rate, calculate the pressure compensation adjustment amount based on the deviation amplitude, superimpose it on the basic drive signal of the electric proportional valve, and generate the atomization kinetic energy control voltage command.

[0010] As a further aspect of the present invention, the characteristic wavelengths and film-forming rate standards of the substrate and coating include the center wavelength of light intensity attenuation of the silicone substrate, the center wavelength of the transmission response of the antibacterial agent, and the threshold of cumulative coating thickness per unit time. The real-time absorbance data of the substrate and coating include the isotropic light absorption intensity value of the silicone particles and the specific light absorption intensity value of the antibacterial agent coating. The real-time coating distribution ratio includes the absorbance ratio of the antibacterial agent layer to the silicone substrate and the instantaneous coating coverage on the particle surface. The actual film-forming rate index includes the time-domain variation gradient of the coating distribution ratio and the dynamic coating growth trend. The atomization kinetic energy control voltage command includes the basic drive voltage of the electric proportional valve, the pressure compensation voltage based on the rate deviation, and the air pressure regulation electrical signal.

[0011] As a further aspect of the present invention, the specific steps of S1 are as follows:

[0012] S101: Collect the reflectance spectrum of the fluidized bed without coated silica gel particles and the transmission spectrum of the pure antibacterial agent solution. Calculate the gradient change value of the light intensity attenuation amplitude for the reflectance spectrum of the fluidized bed without coated silica gel particles. Select wavelength points whose gradient change values ​​exceed the preset noise benchmark as substrate feature points. Identify the local maxima of absorbance values ​​for the transmission spectrum of the pure antibacterial agent solution as antibacterial agent feature points. Serialize and combine the wavelength values ​​of the substrate feature points and the antibacterial agent feature points to generate a set of photoresponse feature bands.

[0013] S102: Obtain the preset total specific surface area parameters of particles in the fluidized bed and the nozzle injection volume parameters per unit time. Divide the injection volume parameters by the total specific surface area parameters to obtain the theoretical liquid film thickening rate. Perform a difference calculation between the theoretical liquid film thickening rate and the preset critical rate benchmark for film integrity. Select the rate values ​​with a difference greater than zero as the allowable operating range to obtain the lower limit threshold of film formation rate.

[0014] S103: Call the set of light response characteristic bands and parse the wavelength numerical sequence, associate and map the wavelength numerical sequence with the lower limit threshold of film formation rate, construct a multi-dimensional constraint matrix including the spectral characteristic frequency range and the boundary value of film accumulation rate, and generate the characteristic wavelength and film formation rate standard of substrate and coating.

[0015] As a further aspect of the present invention, the specific steps of S2 are as follows:

[0016] S201: Based on the characteristic wavelengths of the substrate and coating and the film formation rate standard, read the defined key response wavelength values, activate the spectral sensing array installed inside the rotating drum and lock the detection frequency band to the key response frequency, perform continuous photon reflection signal capture on the surface of the particle layer in a fluidized tumbling state inside the rotating drum, process the captured photon reflection signals through analog-to-digital conversion, demodulate them into digital radiation intensity values, and vectorize and assemble the intensity values ​​at different frequencies to establish a multi-channel fluidized bed reflected light intensity sampling set;

[0017] S202: Call the multi-channel fluidized bed reflected light intensity sampling set, extract multiple radiation intensity values ​​in the vector and perform reciprocal calculation, convert the reciprocal calculation result into absorbance, and according to the frequency and material correspondence recorded in the substrate and coating characteristic wavelength and film formation rate standard, mark the converted absorbance values ​​as substrate background component and coating deposition response component respectively, and splice the marked component data in the time dimension to obtain real-time absorbance data of substrate and coating.

[0018] As a further aspect of the present invention, the specific steps of S3 are as follows:

[0019] S301: Call the real-time absorbance data of the substrate and coating, parse the index information about the spectral frequency band, deconstruct the mixed data stream into numerator values ​​characterizing the light absorption characteristics of the antibacterial agent film layer and denominator values ​​characterizing the light absorption characteristics of the silicone particle substrate, perform synchronous alignment processing based on sampling timestamps on these two sets of numerical sequences, remove isolated data frames that do not match on the time axis, and reorganize the aligned numerator and denominator into column vectors to generate a dual-channel absorbance numerical matrix;

[0020] S302: Call the dual-channel absorbance value matrix, extract the corresponding coating absorbance column vector and substrate absorbance column vector, perform element-wise division operation of coating value with respect to substrate value, calculate the ratio of coating signal intensity carried by unit substrate response at a single sampling point, map the coating signal intensity ratio to a preset linear value range, and generate a quantitative value of relative coverage of antibacterial agent.

[0021] S303: For the quantified value of the relative coverage of the antibacterial agent, establish a statistical analysis model based on a sliding time window, calculate the landing frequency of multiple discrete data points in the differential value interval within the sliding time window, divide the frequency of the differential value interval by the total number of samples within the sliding time window to obtain the local proportion value, construct the probability density sequence of the discrete distribution state of the coating thickness in the particle population, and update the probability density sequence in real time to generate the real-time coating distribution ratio.

[0022] As a further aspect of the present invention, the specific steps of S4 are as follows:

[0023] S401: Call the real-time coating distribution ratio, open a continuous storage space to establish a circular buffer, push the real-time input ratio value into the head of the circular buffer in time order, and release the earliest written data node at the tail of the circular buffer to maintain a constant queue length. Traverse the nodes in the circular buffer and mark the corresponding data acquisition time to generate a first-in-first-out data buffer sequence.

[0024] S402: Based on the first-in-first-out data cache sequence, locate and read the latest ratio sample value at the beginning of the sequence and the oldest ratio sample value at the end of the sequence through pointer addressing, establish subtraction operation logic, use the latest sample value as the minuend and the oldest sample value as the subtrahend to calculate, obtain the net ratio increment value within the real-time cache time window length, perform floating-point formatting on the net ratio increment value, and generate the ratio difference amplitude at the beginning and end times;

[0025] S403: For the difference amplitude of the ratio at the beginning and end of the time, extract the buffer sequence length parameter and multiply it by the single sampling period to obtain the total time span. Construct a division operation model, divide the difference amplitude value by the total time span value to calculate the slope value of the ratio evolution over time, and generate the actual film formation rate index.

[0026] As a further aspect of the present invention, the process of pushing the real-time input ratio value into the head of the circular buffer in chronological order and releasing the earliest written data node at the tail of the circular buffer specifically involves: initializing the write cursor and read cursor used to address the circular buffer, and setting the physical storage depth parameter of the circular buffer.

[0027] Calculate the offset address of the write cursor relative to the physical storage depth, and write the real-time coating distribution ratio to the corresponding storage cell;

[0028] After the write operation is completed, the read cursor is simultaneously pointed to a storage location that is a fixed time window width away from the write cursor, and the logical erase or overwrite operation of the old data is performed to ensure that the first-in-first-out data cache sequence contains the sampled data within the latest time period.

[0029] As a further aspect of the present invention, the specific steps of S5 are as follows:

[0030] S501: Call the actual film formation rate index, parse the data fields of the characteristic wavelength of the substrate and coating and the film formation rate standard, extract the defined lower limit threshold of film formation rate, construct a numerical comparison operation node, use the actual index value as the minuend and the lower limit threshold as the subtrahend to perform a subtraction operation, obtain the value reflecting the difference between the real-time coating accumulation speed and the standard bottom line, perform sign bit detection and absolute value amplitude extraction on the difference value, and generate the film formation rate deviation amplitude value.

[0031] S502: For the film formation rate deviation amplitude value, retrieve the preset pressure-rate response coefficient matrix, match the corresponding conversion gain parameter according to the magnitude range of the deviation amplitude, perform the multiplication operation of the deviation value and the gain parameter, calculate the fluid pressure correction value required to offset the rate deviation, input the fluid pressure correction value into the numerical limiting logic for boundary constraint, and generate the pressure compensation adjustment amount.

[0032] S503: Call the pressure compensation adjustment amount, synchronously collect the basic drive signal of the electric proportional valve to maintain the basic injection state in real time, establish a signal superposition calculation channel, add the compensation adjustment amount to the waveform amplitude of the basic drive signal in a linear addition manner, calculate the corrected synthetic drive amplitude, modulate the synthetic drive amplitude into the corresponding DC voltage signal through the digital-to-analog converter and load it to the controller output port to generate the atomization kinetic energy control voltage command.

[0033] As a further aspect of the present invention, the process of matching the corresponding conversion gain parameter according to the magnitude range of the deviation amplitude specifically involves: parsing a preset pressure-rate response coefficient matrix, wherein the preset pressure-rate response coefficient matrix is ​​specifically configured as a two-dimensional discrete lookup table containing a row index of the deviation magnitude and a column value of the gain coefficient.

[0034] Using the film formation rate deviation amplitude value as the search key, locate two preset deviation nodes adjacent to the film formation rate deviation amplitude value in the two-dimensional discrete lookup table.

[0035] Calculate the weighted average of the gain coefficients corresponding to the two preset deviation nodes, and confirm the weighted average as the conversion gain parameter.

[0036] A silicone cat litter antibacterial coating uniformity control system includes:

[0037] The coating benchmark calibration module collects the fluidized bed reflectance spectrum of uncoated silicone particles and the transmission spectrum signal of pure antibacterial agent solution. It analyzes the light intensity attenuation characteristics and transmission response characteristics in the spectral signals, screens the peak wavelength band of light intensity attenuation of silicone substrate and peak wavelength band of antibacterial agent response, sets the lower limit threshold of film formation rate of coating accumulation speed on particle surface per unit time, and generates characteristic wavelength and film formation rate standards for substrate and coating.

[0038] The spectral absorption conversion module collects reflected light intensity data from the surface of the fluidized bed inside the rotating drum based on the characteristic wavelengths of the substrate and coating, performs logarithmic transformation on the reflected light intensity data, establishes a light intensity attenuation model that conforms to the Lambert-Beer law, and generates real-time absorbance data of the substrate and coating.

[0039] The coating distribution analysis module calls the real-time absorbance data of the substrate and the coating, performs a division operation between the real-time absorbance data of the coating and the real-time absorbance data of the substrate, avoids the baseline drift caused by the fluctuation of fluidized bed particle density, and generates the real-time coating distribution ratio.

[0040] The film formation rate calculation module constructs a first-in-first-out data buffer sequence for the real-time coating distribution ratio, calculates the difference amplitude of the ratio at the beginning and end of the first-in-first-out data buffer sequence, analyzes the slope of the change of the real-time coating distribution ratio over time, and generates the actual film formation rate index.

[0041] The atomization pressure control module compares the actual film formation rate index with the lower limit threshold of the film formation rate, retrieves the preset pressure-rate response coefficient matrix, calculates the pressure compensation adjustment amount based on the deviation amplitude, and superimposes the pressure compensation adjustment amount onto the basic drive signal of the electric proportional valve to generate an atomization kinetic energy control voltage command.

[0042] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0043] In this invention, characteristic bands are selected by collecting reflection and transmission spectral signals to establish characteristic wavelengths and film formation rate standards for substrates and coatings. Real-time absorbance data is generated based on logarithmic transformation. Division operations are performed to obtain the relative coverage index of antibacterial agents and generate a real-time coating distribution ratio. A first-in-first-out data buffer sequence is constructed to calculate the difference amplitude of the ratio at the beginning and end of the time. The slope of the distribution ratio change is analyzed to generate an actual film formation rate index. This index is compared with the lower limit threshold of the film formation rate to calculate the pressure compensation adjustment amount based on the deviation amplitude. This is superimposed on the basic drive signal to generate control instructions, thereby achieving closed-loop precise control of the coating process. This effectively solves the problem of uneven coating distribution caused by fixed-value control, ensures that the accumulation rate of antibacterial agents on the particle surface meets expectations, and improves coating uniformity and raw material utilization. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a schematic diagram of the steps of the present invention;

[0046] Figure 2 This is a detailed schematic diagram of S1 of the present invention;

[0047] Figure 3 This is a detailed schematic diagram of S2 of the present invention;

[0048] Figure 4 This is a detailed schematic diagram of S3 of the present invention;

[0049] Figure 5 This is a detailed schematic diagram of S4 of the present invention;

[0050] Figure 6 This is a detailed schematic diagram of S5 of the present invention;

[0051] Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0052] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0053] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0054] Please see Figure 1 This invention provides a method for controlling the uniformity of silicone cat litter antibacterial coating, comprising the following steps:

[0055] S1: Collect the fluidized bed reflectance spectrum of uncoated silicone particles and the transmission spectrum signal of pure antibacterial agent solution, screen the peak band of light intensity attenuation of silicone substrate and the peak band of antibacterial agent response, set the lower limit threshold of film formation rate of coating accumulation speed on particle surface per unit time, and generate characteristic wavelength and film formation rate standards of substrate and coating.

[0056] S2: Based on the characteristic wavelengths of the substrate and coating and the film formation rate standard, drive the multi-band online spectral detection probe to collect the reflected light intensity data of the fluidized bed surface inside the rotating drum, perform logarithmic transformation on the reflected light intensity data, and generate real-time absorbance data of the substrate and coating;

[0057] S3: Call the real-time absorbance data of the substrate and coating, perform a division operation between the real-time absorbance data of the coating and the real-time absorbance data of the substrate to avoid geometric scattering interference caused by particle tumbling, obtain the relative coverage index of the antibacterial agent, and generate the real-time coating distribution ratio.

[0058] S4: Construct a first-in-first-out (FIFO) data buffer sequence for the real-time coating distribution ratio, calculate the difference amplitude of the ratio at the beginning and end of the FIFO data buffer sequence, analyze the slope of the change in the real-time coating distribution ratio over time, quantify the surface coating establishment speed, and generate actual film formation rate indicators.

[0059] S5: Compare the actual film formation rate index with the characteristic wavelengths of the substrate and coating and the lower limit threshold of the film formation rate included in the film formation rate standard, calculate the pressure compensation adjustment amount based on the deviation amplitude, superimpose it on the basic drive signal of the electric proportional valve, optimize the variable atomizing air pressure and droplet kinetic energy, and generate the atomizing kinetic energy control voltage command.

[0060] The substrate and coating characteristic wavelength and film formation rate standards include the center wavelength of light intensity attenuation of the silicone substrate, the center wavelength of the antibacterial agent transmission response, and the threshold of cumulative coating thickness per unit time. The real-time absorbance data of the substrate and coating include the isotropic light absorption intensity value of the silicone particles and the specific light absorption intensity value of the antibacterial agent coating. The real-time coating distribution ratio includes the absorbance ratio of the antibacterial agent layer to the silicone substrate and the instantaneous coating coverage on the particle surface. The actual film formation rate index includes the time-domain change gradient of the coating distribution ratio and the dynamic coating growth trend. The atomization kinetic energy control voltage command includes the basic drive voltage of the electric proportional valve, the pressure compensation voltage based on the rate deviation, and the air pressure regulation electrical signal.

[0061] Please see Figure 2 The specific steps of S1 are as follows:

[0062] S101: Collect the reflectance spectrum of the fluidized bed without coated silica gel particles and the transmission spectrum of the pure antibacterial agent solution. Calculate the gradient change value of the light intensity attenuation amplitude for the reflectance spectrum of the fluidized bed without coated silica gel particles. Select wavelength points whose gradient change values ​​exceed the preset noise benchmark as substrate feature points. Identify the local maxima of absorbance values ​​for the transmission spectrum of the pure antibacterial agent solution as antibacterial agent feature points. Serialize and combine the wavelength values ​​of the substrate feature points and the antibacterial agent feature points to generate a set of photoresponse feature bands.

[0063] A high-sensitivity fiber optic spectrometer probe was vertically positioned outside the transparent window of the fluidized bed. With an integration time of 10 milliseconds, diffuse reflectance spectral data of the uncoated silica gel particle fluidized bed were continuously acquired. Simultaneously, transmission spectral signals of a pure antibacterial agent solution (e.g., nano-silver solution) were acquired using a cuvette and a transmission spectroscopy accessory. To reduce the communication load on the central control device and improve the timeliness of data processing, an edge computing architecture was introduced. The initial spectral analysis task was offloaded to an intelligent edge gateway connected to the spectrometer. After acquiring full-spectrum data ranging from 400 nm to 2500 nm, the embedded processor inside the edge gateway performed gradient calculations on the reflectance spectrum of the uncoated silica gel particle fluidized bed. This operation extracted the light intensity values ​​of adjacent wavelengths, calculated the light intensity value of the next wavelength point minus the light intensity value of the previous wavelength point, and divided the difference by the wavelength interval to obtain the slope of the light intensity change with wavelength, i.e., the gradient value.

[0064] The process of obtaining the preset noise benchmark is as follows: Dark current noise data from the spectrometer is collected in a completely dark room environment with no light. This is repeated 100 times, and the standard deviation of these 100 data points is calculated. This standard deviation is then multiplied by 3 to obtain the noise benchmark. For example, when the standard deviation of the dark current noise is 2 counts, the edge computing unit calculates the preset noise benchmark as 6 counts through multiplication. Using the local computing power of the edge nodes, the calculated gradient change values ​​are compared point-by-point with this noise benchmark. All wavelength points with an absolute gradient change value greater than 6 are selected and defined as substrate feature points. For the transmission spectrum signal of the pure antibacterial agent solution, a peak search algorithm is executed to identify locations on the absorbance curve where the derivative is 0 and the second derivative is less than 0 (i.e., local maxima), and these are marked as antibacterial agent feature points. Finally, the wavelength values ​​corresponding to the substrate feature points and the wavelength values ​​corresponding to the antibacterial agent feature points are arranged in ascending order, and duplicate values ​​are removed to generate a set of photoresponse characteristic bands. For example, if the substrate feature point frequency is 500 terahertz and 600 terahertz, and the antibacterial agent feature point frequency is 550 terahertz, then the generated set of photoresponse feature bands is 500 terahertz, 550 terahertz, and 600 terahertz.

[0065] S102: Obtain the preset total specific surface area parameters of particles in the fluidized bed and the nozzle injection volume parameters per unit time. Divide the injection volume parameters by the total specific surface area parameters to obtain the theoretical liquid film thickening rate. Perform a difference calculation between the theoretical liquid film thickening rate and the preset critical rate benchmark for film integrity. Select the rate values ​​with a difference greater than zero as the allowable operating range to obtain the lower limit threshold of film formation rate.

[0066] The average particle size of the silica gel particles is pre-determined using a laser particle size analyzer. Combined with the total mass of particles packed in the fluidized bed and the density of the silica gel, the total specific surface area parameter of the particles in the fluidized bed is calculated. For example, when the particle packing weight is 5 kg, the calculated total specific surface area parameter is 200 square meters. Simultaneously, the flow rate setpoint of the precision metering pump is read as the nozzle's injection volume parameter per unit time. For example, setting the injection flow rate to 100 ml per minute results in an injection volume parameter of 0.0001 cubic meters per minute.

[0067] Edge computing nodes call locally stored algorithm libraries to perform division operations, dividing the injection volume parameter by the total specific surface area parameter to calculate the theoretical liquid film thickening rate. For example, dividing 0.0001 cubic meters per minute by 200 square meters yields 0.0000005 meters per minute, or 0.5 micrometers per minute.

[0068] A preset critical rate benchmark for film integrity is set at 0.2 micrometers per minute. This benchmark is the minimum rate determined through previous experiments to ensure the coating remains continuously intact. The calculated theoretical liquid film thickening rate is subtracted from the critical rate benchmark for film integrity, and the difference is calculated. For example, subtracting 0.2 micrometers per minute from 0.5 micrometers per minute yields a difference of 0.3 micrometers per minute. Since the difference is greater than 0, the current operating condition is determined to be within the allowable operating range, and this critical rate benchmark of 0.2 micrometers per minute is set as the lower limit threshold for film formation rate.

[0069] S103: Call the set of optical response characteristic bands and parse the wavelength numerical sequence, associate and map the wavelength numerical sequence with the lower limit threshold of film formation rate, construct a multi-dimensional constraint matrix including the spectral characteristic frequency range and the film layer accumulation rate boundary value, and generate the characteristic wavelength and film formation rate standard of substrate and coating.

[0070] The wavelength sequence, such as 500 terahertz, 550 terahertz, and 600 terahertz, is extracted. This frequency sequence is then mapped to the lower limit threshold of film formation rate (0.2 micrometers per minute) determined in step S102 to construct a multidimensional constraint matrix. The first dimension of this matrix stores the characteristic wavelength values, the second dimension stores the corresponding material type identifier (e.g., 0 represents the substrate, 1 represents the antibacterial agent), and the third dimension stores the lower limit threshold of film formation rate.

[0071] Following the pre-defined Modbus TCP industrial control communication protocol, the edge nodes convert the floating-point parameters in the processed multi-dimensional constraint matrix into 16-bit binary integer codes. For example, a wavelength value of 500 terahertz is mapped to data at register address 40001, and the lower limit threshold for film deposition rate is mapped to data at register address 40005. The generated coded data packets represent the characteristic wavelengths of the substrate and coating, as well as the film deposition rate standards, and are distributed to each distributed sensor front-end to complete the initial configuration.

[0072] Table 1 shows the implementation data for feature extraction and parameter setting in the above steps;

[0073] Table 1: Spectral Characteristics and Film Formation Rate Parameter Setting Table

[0074]

[0075] Please see Figure 3 The specific steps of S2 are as follows:

[0076] S201: Based on the characteristic wavelengths of the substrate and coating and the film formation rate standard, the key response wavelength value is read, the spectral sensing array installed inside the rotating drum is activated and the detection frequency band is locked to the key response frequency. Continuous photon reflection signal capture is performed on the surface of the particle layer in a fluidized tumbling state inside the rotating drum. The captured photon reflection signal is processed by analog-to-digital conversion, demodulated into digital form of radiation intensity value, and the intensity values ​​under different frequencies are vectorized and assembled to establish a multi-channel fluidized bed reflected light intensity sampling set.

[0077] The key response wavelength values, such as 500 terahertz, 550 terahertz, and 600 terahertz, are read via an industrial Ethernet bus. Configuration commands are sent to an online spectral sensing array installed inside a rotating drum to activate specific channels in the array, adjust the passband center wavelength of the tunable filter, and precisely lock the detection frequency band to the aforementioned key response frequencies, with a bandwidth set to 5 nanometers.

[0078] As the rotating drum spins at 15 revolutions per minute, causing the internal particle layer to fluidize and tumble, the spectral sensing array continuously captures photon reflection signals at a sampling rate of 1000 Hz. To cope with the massive data load brought about by high-frequency sampling, an FPGA preprocessing chip with edge computing capabilities is integrated at the back end of the photodetector. The photodetector converts the received photon energy into a weak current signal, which is amplified by a preamplifier and then input to a 16-bit high-precision analog-to-digital converter. The analog-to-digital converter quantizes the analog voltage signal into a digital form of radiation intensity value between 0 and 65535. For example, the radiation intensity value acquired at a frequency of 550 terahertz is 32000.

[0079] The FPGA chip directly arranges the radiation intensity values ​​collected from different frequency channels at the same time in ascending order of frequency on the edge side, assembles them into a one-dimensional vector, such as 45000, 32000, and 41000, and combines it with the current timestamp information to establish a multi-channel fluidized bed reflected light intensity sampling set.

[0080] S202: Call the multi-channel fluidized bed reflected light intensity sampling set, extract multiple radiation intensity values ​​in the vector and perform reciprocal operation, convert the reciprocal operation result into absorbance, and according to the frequency and material correspondence recorded in the substrate and coating characteristic wavelength and film formation rate standard, mark the converted absorbance values ​​as substrate background component and coating deposition response component respectively, and splice the marked component data in the time dimension to obtain the real-time absorbance data of substrate and coating;

[0081] The radiant intensity values ​​within a vector are extracted one by one using edge computing microservice units deployed at the acquisition end. A reciprocal operation is performed, dividing the numerical value by 1. The result of this reciprocal operation is then substituted into a logarithmic function to calculate the base-10 logarithm, thus completing the absorbance conversion. This calculation logic is based on a corollary of the Beer-Lambert law, reflecting the attenuation characteristics of light in a medium. For example, for a radiant intensity value of 32000, first dividing 1 by 32000 yields 0.00003125, then taking the logarithm and absolute value gives an absorbance value of 4.5.

[0082] Based on the correspondence recorded in the standards for characteristic wavelengths and film formation rates of substrates and coatings, the absorbance values ​​corresponding to 500 terahertz and 600 terahertz are determined to characterize the properties of the silicone substrate and are marked as the substrate background component; the absorbance value corresponding to 550 terahertz is determined to characterize the properties of the antibacterial agent coating and is marked as the coating deposition response component.

[0083] Edge computing units simply concatenate the labeled component data in chronological order. For example, the substrate component in the 1st millisecond is concatenated with the substrate component in the 2nd millisecond to form a continuous time-series data stream, thereby acquiring real-time absorbance data of the substrate and coating. This pre-calculation and filtering at the edge significantly reduces the data bandwidth requirements for transmission to the cloud.

[0084] Please see Figure 4 The specific steps of S3 are as follows:

[0085] S301: Call the real-time absorbance data of the substrate and coating, parse the index information of the spectral band, deconstruct the mixed data stream into numerator values ​​characterizing the absorbance properties of the antibacterial agent film layer and denominator values ​​characterizing the absorbance properties of the silicone particle substrate, perform synchronous alignment processing based on sampling timestamps on these two sets of numerical sequences, remove isolated data frames that do not match on the time axis, and reorganize the aligned numerator and denominator into column vectors to generate a dual-channel absorbance numerical matrix;

[0086] The index information in the packet header is parsed to identify which data belongs to the 550 terahertz band, and which belongs to the 500 terahertz and 600 terahertz bands. The data processing task is assigned to the data alignment unit in the distributed edge computing cluster for execution. The mixed data stream is decomposed into two independent numerical sequences: one is a sequence of numerator values ​​characterizing the light absorption properties of the antibacterial agent film, and the other is a sequence of denominator values ​​characterizing the light absorption properties of the silicone particle substrate.

[0087] Read the sampling timestamps attached to each data point in both sets of sequences and perform synchronization alignment processing based on the timestamps. Set the time tolerance window to 2 milliseconds. If there is a data point with timestamp T in the numerator sequence, but no data point with timestamps within the range of T plus or minus 2 milliseconds is found in the denominator sequence, then the data point is determined to be an isolated data frame and is discarded.

[0088] The aligned numerator values ​​are used as the first column, and the denominator values ​​are used as the second column. If there are multiple substrate frequencies, their arithmetic mean is taken. The column vectors are recombined to generate a dual-channel absorbance value matrix with the number of rows equal to the number of aligned sample points.

[0089] S302: Call the dual-channel absorbance numerical matrix, extract the corresponding coating absorbance column vector and substrate absorbance column vector, perform element-wise division operation of coating value with respect to substrate value, calculate the ratio of coating signal intensity carried by unit substrate response at a single sampling point, map the coating signal intensity ratio to a preset linear numerical range, and generate a quantitative value of the relative coverage of antibacterial agent.

[0090] The parallel computing core of the edge computing unit extracts the coating absorbance value from the first column and the substrate absorbance value from the second column, performing element-wise division. That is, it divides the coating absorbance value at each sampling moment by the substrate absorbance value at the same time. For example, if the coating absorbance is 0.5 and the substrate absorbance is 1.2 at a certain moment, the calculated ratio is 0.5 divided by 1.2, which is approximately 0.416. This calculation aims to eliminate overall light intensity fluctuations caused by changes in particle tumble angle and light source distance, extracting only the coating signal intensity ratio.

[0091] The calculated coating signal intensity ratio is mapped to a preset linear numerical range of 0 to 100. The mapping algorithm is as follows: the calculated ratio is multiplied by a preset calibration coefficient, which is derived by back-calculating the ratio of a full-coverage standard sample. For example, if the calibration coefficient is set to 240, multiplying the ratio 0.416 by 240 yields 99.84, which is the quantified value of the generated antibacterial agent relative coverage.

[0092] S303: For the quantified value of the relative coverage of antibacterial agents, a statistical analysis model based on a sliding time window is established. The frequency of multiple discrete data points in the differential value interval within the sliding time window is calculated. The frequency of the differential value interval is divided by the total number of samples within the sliding time window to obtain the local proportion value. The probability density sequence of the discrete distribution state of the coating thickness in the particle population is constructed, and the probability density sequence is updated and output in real time to generate the real-time coating distribution ratio.

[0093] A sliding time window with a length of 1000 data points is established in the local cache of the edge device. As new data is generated, the window slides backward. Within each window, the numerical range of 0 to 100 is divided into 10 level intervals, such as 0 to 10, 10 to 20, up to 90 to 100.

[0094] The edge statistics engine counts in real time how many times each of the 1000 discrete data points falls into each level interval within the sliding time window. For example, statistics show that 850 data points fall into the 90-100 interval. Performing a division operation, dividing the frequency of 850 by the total number of samples of 1000, yields a local proportion of 0.85.

[0095] The local proportion values ​​of each interval are arranged in interval order to construct a probability density sequence reflecting the discrete distribution of coating thickness in the particle population. This sequence is updated once a new data point is received and output in real time to generate the real-time coating distribution ratio. Due to the use of an edge computing architecture, the statistical latency is controlled at the millisecond level. Table 2 shows the statistical distribution implementation data within the sliding window;

[0096] Table 2: Statistical Table of Coating Coverage Distribution

[0097]

[0098] Please see Figure 5 The specific steps of S4 are as follows:

[0099] S401: Call the real-time coating distribution ratio, allocate a continuous storage space to establish a circular buffer, push the real-time input ratio value into the head of the circular buffer in time order, and release the earliest written data node at the tail of the circular buffer to maintain a constant queue length. Traverse the nodes in the circular buffer and mark the corresponding data acquisition time to generate a first-in-first-out data buffer sequence.

[0100] Select a high coverage range, such as 90 to 100, with a percentage value of 0.85. Allocate a contiguous storage space of 1024 bytes in the local high-speed SRAM (Static Random Access Memory) of the edge controller to create a circular buffer.

[0101] Utilizing the high-speed read / write capabilities of SRAM, the real-time input ratio value of 0.85 is written to the head position of the circular buffer, i.e., the position pointed to by the write pointer, in chronological order. If the buffer is full, the earliest written data node is released by moving the tail pointer, i.e., the read pointer, and new data is written to overwrite it, thus maintaining a constant queue length of 200 sampling points.

[0102] It iterates through each node in the circular buffer, binds the clock value of the time it was collected to each data point, generates a first-in-first-out data cache sequence containing numerical values ​​and time tags, and leverages the low latency advantage of edge storage to provide a stable data source for subsequent rate calculations.

[0103] S402: Based on the first-in-first-out data cache sequence, the latest ratio sample value at the beginning of the sequence and the oldest ratio sample value at the end of the sequence are located and read through pointer addressing. The subtraction operation logic is established, and the latest sample value is used as the minuend and the oldest sample value is used as the subtrahend to calculate the net ratio increment value within the real-time cache time window. The net ratio increment value is then formatted as a floating-point number to generate the ratio difference amplitude at the beginning and end of the time.

[0104] By using pointer addressing, the local memory address is directly accessed through edge computing to locate and read the latest ratio sample value at the sequence write pointer, assuming it is 0.88, and the oldest ratio sample value at the read pointer, assuming it is 0.78.

[0105] A subtraction operation logic is established, using the latest sampled value of 0.88 as the minuend and the oldest sampled value of 0.78 as the subtrahend. That is, 0.88 minus 0.78 equals 0.10. This value represents the net increase in the proportion of high coverage within the cache time window.

[0106] The net increment value of the ratio, 0.10, is formatted as a floating-point number, retaining four significant decimal places, to generate the ratio difference amplitude at the first and last time points.

[0107] S403: For the difference amplitude of the ratio at the beginning and end of the time, extract the buffer sequence length parameter and multiply it by the single sampling period to obtain the total time span. Construct a division operation model, divide the difference amplitude value by the total time span value to calculate the slope value of the ratio evolution over time, and generate the actual film formation rate index.

[0108] 200 parameters for the buffer sequence length are extracted and multiplied by a single sampling period of 0.1 seconds to calculate a total time span of 20 seconds. A division operation model is constructed to divide the difference amplitude value by the total time span value to calculate the slope value of the ratio evolving over time. For example, dividing the difference amplitude of 0.10 by 20 seconds yields a result of 0.005. This value represents the rate of increase in the high coverage percentage per second. This slope value of 0.005 is defined as the actual film formation rate index, in units of percentage per second. To correspond with the physical thickness rate in step S1, an empirical conversion factor can be multiplied, for example, 0.5 micrometers per percentage. Multiplying 0.005 by 0.5 yields 0.0025 micrometers per second, or 0.15 micrometers per minute, which serves as the final actual film formation rate index. This index can quantify the coating growth kinetics within the reactor in real time and broadcast it to downstream control nodes via an edge network.

[0109] Please see Figure 6 The specific steps of S5 are as follows:

[0110] S501: Call the actual film formation rate index, parse the data fields of the characteristic wavelength of the substrate and coating and the film formation rate standard, extract the defined lower limit threshold of the film formation rate, construct a numerical comparison operation node, use the actual index value as the minuend and the lower limit threshold as the subtrahend to perform a subtraction operation, obtain the value reflecting the difference between the real-time coating accumulation speed and the standard bottom line, perform sign bit detection and absolute value amplitude extraction on the difference value, and generate the film formation rate deviation amplitude value.

[0111] The real-time control process executed within the field edge control cabinet calls the actual film formation rate index generated in step S403, such as the converted 0.15 micrometers per minute, and analyzes the characteristic wavelengths of the substrate and coating, as well as the lower limit threshold of 0.2 micrometers per minute defined in the film formation rate standard. A numerical comparison operation node is constructed, using the actual index value as the minuend and the lower limit threshold as the subtrahend, to perform a subtraction operation. For example, subtracting 0.2 micrometers per minute from 0.15 micrometers per minute yields a difference of -0.05 micrometers per minute.

[0112] The sign of the drop value was checked, and it was confirmed to be negative, indicating insufficient rate. The absolute value amplitude of 0.05 was extracted to generate the film formation rate deviation amplitude. This deviation amplitude accurately quantifies the gap between the current production speed and the minimum standard. Because the calculation is completed at the edge, the lag caused by network transmission delay is eliminated, providing direct numerical basis for subsequent pressure compensation.

[0113] S502: For the film formation rate deviation amplitude, retrieve the preset pressure-rate response coefficient matrix, match the corresponding conversion gain parameter according to the magnitude range of the deviation amplitude, perform the multiplication operation of the deviation value and the gain parameter, calculate the fluid pressure correction value required to offset the rate deviation, input the fluid pressure correction value into the numerical limiting logic for boundary constraint, and generate the pressure compensation adjustment amount.

[0114] For a film formation rate deviation of 0.05, a preset pressure-rate response coefficient matrix is ​​retrieved. This matrix defines the gain parameters corresponding to different deviation ranges. As shown in Table 3, when the deviation is between 0.01 and 0.10, the matched conversion gain parameter is 50 kPa per micrometer per minute.

[0115] Performing a multiplication operation, multiplying the deviation value of 0.05 by the gain parameter of 50, yields a result of 2.5 kPa. This value is the fluid pressure correction value required to compensate for the rate deviation.

[0116] A fluid pressure correction value of 2.5 kPa is input into the numerical limiting logic. The maximum adjustment step is set to 5 kPa. Since 2.5 is less than 5, 2.5 kPa is directly output as the pressure compensation adjustment amount. This logic effectively prevents oscillations caused by excessive adjustment, ensuring the stability of pressure regulation.

[0117] Table 3: Pressure-rate response coefficient matrix

[0118]

[0119] S503: Call the pressure compensation adjustment amount, synchronously collect the basic drive signal of the electric proportional valve to maintain the basic injection state in real time, establish a signal superposition calculation channel, add the compensation adjustment amount to the waveform amplitude of the basic drive signal in a linear addition manner, calculate the corrected synthetic drive amplitude, and modulate the synthetic drive amplitude into the corresponding DC voltage signal through the digital-to-analog converter and load it to the controller output port to generate the atomization kinetic energy control voltage command.

[0120] The basic drive signal of the electro-proportional valve to maintain the basic injection state in real time is synchronously acquired. It is assumed that the basic pressure corresponding to this signal is 100 kPa and the corresponding voltage amplitude is 5 volts, that is, the sensitivity is 20 kPa per volt.

[0121] Establish a signal superposition operation channel. First, convert the pressure compensation adjustment of 2.5 kPa into a voltage increment: divide 2.5 by 20 to get 0.125 volts.

[0122] The voltage increment of 0.125 volts is linearly added to the waveform amplitude of the basic drive signal of 5 volts, and the corrected synthetic drive amplitude is calculated to be 5.125 volts.

[0123] The synthesized drive amplitude of 5.125 volts is modulated into a corresponding DC voltage signal by a 12-bit digital-to-analog converter and applied to the output port of the atomizing nozzle controller to generate an atomization kinetic energy regulation voltage command, thereby increasing the injection pressure and improving the film formation rate. The advantage of this operational logic is that by mapping the rate deviation to a pressure compensation amount and combining it with the ultra-low latency characteristics of edge computing, closed-loop feedback control is achieved. This automatically corrects for the decrease in film formation rate caused by nozzle blockage or gas source fluctuations, ensuring production efficiency.

[0124] Please see Figure 7 A silicone cat litter antibacterial coating uniformity control system, including:

[0125] The coating benchmark calibration module collects the fluidized bed reflectance spectrum of uncoated silicone particles and the transmission spectrum signal of pure antibacterial agent solution. It analyzes the light intensity attenuation characteristics and transmission response characteristics in the spectral signals, screens the peak wavelength band of light intensity attenuation of silicone substrate and peak wavelength band of antibacterial agent response, sets the lower limit threshold of film formation rate of coating accumulation speed on particle surface per unit time, and generates characteristic wavelength and film formation rate standards for substrate and coating.

[0126] The spectral absorption transformation module collects reflected light intensity data from the surface of the fluidized bed inside the rotating drum based on the characteristic wavelengths of the substrate and coating. It performs a logarithmic transformation on the reflected light intensity data, establishes a light intensity attenuation model that conforms to the Lambert-Beer law, and generates real-time absorbance data of the substrate and coating.

[0127] The coating distribution analysis module calls the real-time absorbance data of the substrate and the coating, performs a division operation between the real-time absorbance data of the coating and the real-time absorbance data of the substrate, avoids the baseline drift caused by the fluctuation of fluidized bed particle density, and generates the real-time coating distribution ratio.

[0128] The film formation rate calculation module constructs a first-in-first-out data buffer sequence for the real-time coating distribution ratio, calculates the difference amplitude of the ratio at the beginning and end of the first-in-first-out data buffer sequence, analyzes the slope of the change of the real-time coating distribution ratio over time, and generates the actual film formation rate index.

[0129] The atomization pressure control module compares the actual film formation rate index with the lower limit threshold of the film formation rate, retrieves the preset pressure-rate response coefficient matrix, calculates the pressure compensation adjustment amount based on the deviation amplitude, and superimposes the pressure compensation adjustment amount onto the basic drive signal of the electro-proportional valve to generate the atomization kinetic energy control voltage command.

[0130] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of protection of the described technical solutions.

Claims

1. A method for controlling the uniformity of antibacterial coating on silica gel cat litter, characterized in that, Includes the following steps: S1: Collect the fluidized bed reflectance spectrum of uncoated silicone particles and the transmission spectrum signal of pure antibacterial agent solution, screen the peak band of light intensity attenuation of silicone substrate and the peak band of antibacterial agent response, set the lower limit threshold of film formation rate of coating accumulation speed on particle surface per unit time, and generate characteristic wavelength and film formation rate standards of substrate and coating. S2: Based on the characteristic wavelengths and film formation rate standards of the substrate and coating, collect the reflected light intensity data of the fluidized bed surface inside the rotating drum, perform logarithmic transformation on the reflected light intensity data, and generate real-time absorbance data of the substrate and coating; S3: Call the real-time absorbance data of the substrate and coating, perform a division operation between the real-time absorbance data of the coating and the real-time absorbance data of the substrate, obtain the relative coverage index of the antibacterial agent, and generate the real-time coating distribution ratio; S4: Construct a first-in-first-out data buffer sequence for the real-time coating distribution ratio, calculate the difference amplitude of the ratio at the beginning and end of the first-in-first-out data buffer sequence, analyze the slope of the change of the real-time coating distribution ratio over time, and generate the actual film formation rate index. S5: Compare the actual film formation rate index with the lower limit threshold of the film formation rate, calculate the pressure compensation adjustment amount based on the deviation amplitude, superimpose it on the basic drive signal of the electric proportional valve, and generate the atomization kinetic energy control voltage command.

2. The method for controlling the uniformity of silicone cat litter antibacterial coating according to claim 1, characterized in that, The substrate and coating characteristic wavelength and film formation rate standards include the center wavelength of light intensity attenuation of the silicone substrate, the center wavelength of the antibacterial agent transmission response, and the threshold of cumulative coating thickness per unit time. The real-time absorbance data of the substrate and coating include the isotropic light absorption intensity value of the silicone particles and the specific light absorption intensity value of the antibacterial agent coating. The real-time coating distribution ratio includes the absorbance ratio of the antibacterial agent layer to the silicone substrate and the instantaneous coating coverage on the particle surface. The actual film formation rate index includes the time-domain change gradient of the coating distribution ratio and the dynamic coating growth trend. The atomization kinetic energy control voltage command includes the basic drive voltage of the electric proportional valve, the pressure compensation voltage based on the rate deviation, and the air pressure regulation electrical signal.

3. The method for controlling the uniformity of silicone cat litter antibacterial coating according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Collect the reflectance spectrum of the fluidized bed without coated silica gel particles and the transmission spectrum of the pure antibacterial agent solution. Calculate the gradient change value of the light intensity attenuation amplitude for the reflectance spectrum of the fluidized bed without coated silica gel particles. Select wavelength points whose gradient change values ​​exceed the preset noise benchmark as substrate feature points. Identify the local maxima of absorbance values ​​for the transmission spectrum of the pure antibacterial agent solution as antibacterial agent feature points. Serialize and combine the wavelength values ​​of the substrate feature points and the antibacterial agent feature points to generate a set of photoresponse feature bands. S102: Obtain the preset total specific surface area parameters of particles in the fluidized bed and the nozzle injection volume parameters per unit time. Divide the injection volume parameters by the total specific surface area parameters to obtain the theoretical liquid film thickening rate. Perform a difference calculation between the theoretical liquid film thickening rate and the preset critical rate benchmark for film integrity. Select the rate values ​​with a difference greater than zero as the allowable operating range to obtain the lower limit threshold of film formation rate. S103: Call the set of light response characteristic bands and parse the wavelength numerical sequence, associate and map the wavelength numerical sequence with the lower limit threshold of film formation rate, construct a multi-dimensional constraint matrix including the spectral characteristic frequency range and the boundary value of film accumulation rate, and generate the characteristic wavelength and film formation rate standard of substrate and coating.

4. The method for controlling the uniformity of silicone cat litter antibacterial coating according to claim 3, characterized in that, The specific steps of S2 are as follows: S201: Based on the characteristic wavelengths of the substrate and coating and the film formation rate standard, read the defined key response wavelength values, activate the spectral sensing array installed inside the rotating drum and lock the detection frequency band to the key response frequency, perform continuous photon reflection signal capture on the surface of the particle layer in a fluidized tumbling state inside the rotating drum, process the captured photon reflection signals through analog-to-digital conversion, demodulate them into digital radiation intensity values, and vectorize and assemble the intensity values ​​at different frequencies to establish a multi-channel fluidized bed reflected light intensity sampling set; S202: Call the multi-channel fluidized bed reflected light intensity sampling set, extract multiple radiation intensity values ​​in the vector and perform reciprocal calculation, convert the reciprocal calculation result into absorbance, and according to the frequency and material correspondence recorded in the substrate and coating characteristic wavelength and film formation rate standard, mark the converted absorbance values ​​as substrate background component and coating deposition response component respectively, and splice the marked component data in the time dimension to obtain real-time absorbance data of substrate and coating.

5. The method for controlling the uniformity of silicone cat litter antibacterial coating according to claim 4, characterized in that, The specific steps for S3 are as follows: S301: Call the real-time absorbance data of the substrate and coating, parse the index information about the spectral frequency band, deconstruct the mixed data stream into numerator values ​​characterizing the light absorption characteristics of the antibacterial agent film layer and denominator values ​​characterizing the light absorption characteristics of the silicone particle substrate, perform synchronous alignment processing based on sampling timestamps on these two sets of numerical sequences, remove isolated data frames that do not match on the time axis, and reorganize the aligned numerator and denominator into column vectors to generate a dual-channel absorbance numerical matrix; S302: Call the dual-channel absorbance value matrix, extract the corresponding coating absorbance column vector and substrate absorbance column vector, perform element-wise division operation of coating value with respect to substrate value, calculate the ratio of coating signal intensity carried by unit substrate response at a single sampling point, map the coating signal intensity ratio to a preset linear value range, and generate a quantitative value of relative coverage of antibacterial agent. S303: For the quantified value of the relative coverage of the antibacterial agent, establish a statistical analysis model based on a sliding time window, calculate the landing frequency of multiple discrete data points in the differential value interval within the sliding time window, divide the frequency of the differential value interval by the total number of samples within the sliding time window to obtain the local proportion value, construct the probability density sequence of the discrete distribution state of the coating thickness in the particle population, and update the probability density sequence in real time to generate the real-time coating distribution ratio.

6. The method for controlling the uniformity of silicone cat litter antibacterial coating according to claim 5, characterized in that, The specific steps of S4 are as follows: S401: Call the real-time coating distribution ratio, open a continuous storage space to establish a circular buffer, push the real-time input ratio value into the head of the circular buffer in time order, and release the earliest written data node at the tail of the circular buffer to maintain a constant queue length. Traverse the nodes in the circular buffer and mark the corresponding data acquisition time to generate a first-in-first-out data buffer sequence. S402: Based on the first-in-first-out data cache sequence, locate and read the latest ratio sample value at the beginning of the sequence and the oldest ratio sample value at the end of the sequence through pointer addressing, establish subtraction operation logic, use the latest sample value as the minuend and the oldest sample value as the subtrahend to calculate, obtain the net ratio increment value within the real-time cache time window length, perform floating-point formatting on the net ratio increment value, and generate the ratio difference amplitude at the beginning and end times; S403: For the difference amplitude of the ratio at the beginning and end of the time, extract the buffer sequence length parameter and multiply it by the single sampling period to obtain the total time span. Construct a division operation model, divide the difference amplitude value by the total time span value to calculate the slope value of the ratio evolution over time, and generate the actual film formation rate index.

7. The method for controlling the uniformity of silicone cat litter antibacterial coating according to claim 6, characterized in that, The process of pushing the real-time input ratio values ​​into the head of the circular buffer in chronological order and releasing the earliest written data node at the tail of the circular buffer specifically involves: initializing the write cursor and read cursor used to address the circular buffer, and setting the physical storage depth parameter of the circular buffer. Calculate the offset address of the write cursor relative to the physical storage depth, and write the real-time coating distribution ratio to the corresponding storage cell; After the write operation is completed, the read cursor is simultaneously pointed to a storage location that is a fixed time window width away from the write cursor, and the logical erase or overwrite operation of the old data is performed to ensure that the first-in-first-out data cache sequence contains the sampled data within the latest time period.

8. The method for controlling the uniformity of silicone cat litter antibacterial coating according to claim 6, characterized in that, The specific steps of S5 are as follows: S501: Call the actual film formation rate index, parse the data fields of the characteristic wavelength of the substrate and coating and the film formation rate standard, extract the defined lower limit threshold of film formation rate, construct a numerical comparison operation node, use the actual index value as the minuend and the lower limit threshold as the subtrahend to perform a subtraction operation, obtain the value reflecting the difference between the real-time coating accumulation speed and the standard bottom line, perform sign bit detection and absolute value amplitude extraction on the difference value, and generate the film formation rate deviation amplitude value. S502: For the film formation rate deviation amplitude value, retrieve the preset pressure-rate response coefficient matrix, match the corresponding conversion gain parameter according to the magnitude range of the deviation amplitude, perform the multiplication operation of the deviation value and the gain parameter, calculate the fluid pressure correction value required to offset the rate deviation, input the fluid pressure correction value into the numerical limiting logic for boundary constraint, and generate the pressure compensation adjustment amount. S503: Call the pressure compensation adjustment amount, synchronously collect the basic drive signal of the electric proportional valve to maintain the basic injection state in real time, establish a signal superposition calculation channel, add the compensation adjustment amount to the waveform amplitude of the basic drive signal in a linear addition manner, calculate the corrected synthetic drive amplitude, modulate the synthetic drive amplitude into the corresponding DC voltage signal through the digital-to-analog converter and load it to the controller output port to generate the atomization kinetic energy control voltage command.

9. The method for controlling the uniformity of silicone cat litter antibacterial coating according to claim 8, characterized in that, The process of matching the corresponding conversion gain parameter according to the magnitude range of the deviation is specifically as follows: parsing the preset pressure-rate response coefficient matrix, which is specifically configured as a two-dimensional discrete lookup table containing the deviation magnitude as the row index and the gain coefficient as the column value. Using the film formation rate deviation amplitude value as the search key, locate two preset deviation nodes adjacent to the film formation rate deviation amplitude value in the two-dimensional discrete lookup table. Calculate the weighted average of the gain coefficients corresponding to the two preset deviation nodes, and confirm the weighted average as the conversion gain parameter.

10. A control system for the uniformity of antibacterial coating on silicone cat litter, characterized in that, The system is used to implement the method for controlling the uniformity of silicone cat litter antibacterial coating as described in any one of claims 1-9, the system comprising: The coating benchmark calibration module collects the fluidized bed reflectance spectrum of uncoated silicone particles and the transmission spectrum signal of pure antibacterial agent solution. It analyzes the light intensity attenuation characteristics and transmission response characteristics in the spectral signals, screens the peak wavelength band of light intensity attenuation of silicone substrate and peak wavelength band of antibacterial agent response, sets the lower limit threshold of film formation rate of coating accumulation speed on particle surface per unit time, and generates characteristic wavelength and film formation rate standards for substrate and coating. The spectral absorption conversion module collects reflected light intensity data from the surface of the fluidized bed inside the rotating drum based on the characteristic wavelengths of the substrate and coating, performs logarithmic transformation on the reflected light intensity data, establishes a light intensity attenuation model that conforms to the Lambert-Beer law, and generates real-time absorbance data of the substrate and coating. The coating distribution analysis module calls the real-time absorbance data of the substrate and the coating, performs a division operation between the real-time absorbance data of the coating and the real-time absorbance data of the substrate, avoids the baseline drift caused by the fluctuation of fluidized bed particle density, and generates the real-time coating distribution ratio. The film formation rate calculation module constructs a first-in-first-out data buffer sequence for the real-time coating distribution ratio, calculates the difference amplitude of the ratio at the beginning and end of the first-in-first-out data buffer sequence, analyzes the slope of the change of the real-time coating distribution ratio over time, and generates the actual film formation rate index. The atomization pressure control module compares the actual film formation rate index with the lower limit threshold of the film formation rate, retrieves the preset pressure-rate response coefficient matrix, calculates the pressure compensation adjustment amount based on the deviation amplitude, and superimposes the pressure compensation adjustment amount onto the basic drive signal of the electric proportional valve to generate an atomization kinetic energy control voltage command.

Images