A fully automatic filling and sealing method and system based on image recognition and analysis

Abstract

The present invention discloses a fully automatic filling and sealing method and system based on image recognition and analysis, which relates to the field of image analysis technology, including extracting fluid dynamic control parameters and bubble group dynamic characteristic parameters; controlling the annular microelectrode array to generate a directional electric field based on the fluid dynamic control parameters to regulate the liquid flow state; determining ultrasonic action parameters based on the bubble group dynamic characteristic parameters, and controlling the frequency-adjustable ultrasonic generator to output ultrasonic modulation pulses; detecting the temperature distribution in the container sealing area and applying differentiated heat sealing parameters to different temperature areas; establishing a multi-layer perception neural network optimization controller, and optimizing the control parameter combination with the filling and sealing process quality index as the optimization target. The present invention adopts a filling method that combines image recognition with electric field regulation to achieve precise control of the liquid flow state, solve the problems of liquid drawing and flow interruption in the traditional filling process, and improve filling accuracy and efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of image analysis, and in particular to a fully automatic filling and sealing method and system based on image recognition and analysis. Background Art

[0002] While automated filling equipment has been widely adopted in the filling and sealing industry, it still faces numerous technical challenges. Traditional filling equipment typically utilizes metering pumps or volumetric metering devices, relying on mechanical control systems. This limits filling accuracy and efficiency. During the production process, differences in liquid surface tension and viscosity, as well as environmental factors, can easily lead to problems such as uneven filling, overflow, and insufficient filling. Furthermore, existing filling systems have limited bubble handling capabilities, making bubbles formed during the filling process difficult to effectively eliminate, impacting product quality.

[0003] In terms of the tail-sealing process, existing technologies primarily rely on constant-temperature heating devices for sealing, making them difficult to adapt to the sealing requirements of containers of varying materials and shapes. Uneven temperature distribution in the sealing area of ​​the container is a common problem during the sealing process, which can easily lead to inconsistent seal strength. Currently, most filling and tail-sealing equipment on the market utilizes independent control systems, lacking integrated coordination mechanisms and intelligent analysis capabilities. Adjustment of production parameters relies primarily on manual experience, making efficient parameter optimization and adaptive adjustment difficult, thus limiting the further development of the filling and tail-sealing process. Summary of the Invention

[0004] The present invention provides a fully automatic filling and sealing method and system based on image recognition and analysis, which is used to solve the technical problems of low liquid filling accuracy, difficulty in eliminating bubbles and uneven sealing temperature.

[0005] In view of this, the first aspect of the present invention provides a fully automatic filling and sealing method based on image recognition and analysis, comprising:

[0006] Construct a three-dimensional dynamic model of the fluid and extract the fluid dynamic control parameters and bubble group dynamic characteristic parameters through image recognition;

[0007] According to the fluid dynamic control parameters, the annular microelectrode array set around the filling nozzle is controlled to generate a directional electric field to regulate the liquid flow state;

[0008] The ultrasonic action parameters are determined according to the dynamic characteristic parameters of the bubble group and the physical properties of the container liquid, and the frequency-adjustable ultrasonic generator is controlled to output ultrasonic modulated pulses to the bubble gathering area;

[0009] Detect the temperature distribution in the container sealing area, control the locally adjustable heat sealing head based on the detection results, and apply differentiated heat sealing parameters to different temperature areas;

[0010] A multi-layer perception neural network optimization controller is established to receive the system state parameter set. Taking the filling and sealing process quality index as the optimization target, the control parameter combination is optimized to complete the adaptive adjustment of the parameters of the entire filling and sealing process.

[0011] Optionally, the fluid dynamic control parameters include surface stress tensor field, three-dimensional flow velocity vector distribution, liquid surface motion state parameters, and the bubble group dynamic characteristic parameters include bubble volume concentration gradient, bubble equivalent particle size distribution and three-dimensional coordinates of the bubble aggregation area.

[0012] Optionally, controlling the annular microelectrode array disposed around the nozzle to generate a directional electric field includes:

[0013] Arrange multiple microelectrodes around the nozzle to form a circular microelectrode array;

[0014] Based on the fluid dynamic control parameters, a control parameter set for each microelectrode is generated;

[0015] A set of control parameters is applied to the annular microelectrode array to generate a directional electric field;

[0016] The electric stress distribution is formed at the liquid-gas interface through the directional electric field to regulate the liquid flow state;

[0017] At the end of the filling process, based on the liquid surface motion state parameters, the microelectrode voltage distribution is adjusted to form a convergent electric field gradient pointing to the center of the filling nozzle to control the liquid flow interruption.

[0018] Optionally, controlling the frequency-adjustable ultrasonic generator to output ultrasonic modulated pulses directed to the bubble accumulation area includes:

[0019] Calculate the ultrasonic transmission loss coefficient and liquid attenuation coefficient based on the physical properties of the container liquid;

[0020] Determine the bubble resonance frequency range based on the bubble equivalent particle size distribution;

[0021] Adjust the beam focusing parameters of the frequency-adjustable ultrasonic generator based on the bubble volume concentration gradient and the three-dimensional coordinates of the bubble aggregation area;

[0022] The frequency-adjustable ultrasonic generator is configured to a pulse output mode, and the pulse period and pulse intensity are determined according to the transmission loss coefficient, the liquid attenuation coefficient, and the bubble resonance frequency range;

[0023] Ultrasonic wave modulation pulses are output to the bubble gathering area through a frequency-adjustable ultrasonic wave generator.

[0024] Optionally, applying differentiated heat sealing parameters to different temperature zones includes:

[0025] Use thermal imaging technology to scan the container sealing area and obtain temperature distribution data of the container sealing area;

[0026] Perform temperature gradient analysis based on temperature distribution data to identify abnormal temperature areas in the container sealing area;

[0027] Based on the temperature anomaly area, the container sealing area is divided into multiple independent control blocks, and the temperature characteristic parameters of each control block are calculated;

[0028] Extract standard heat sealing parameters corresponding to container materials;

[0029] Based on the temperature characteristic parameters of each control block and the standard heat sealing parameters, differentiated heat sealing parameters are generated for each control block through parameter compensation calculation;

[0030] Through the locally adjustable heat sealing head, the corresponding differentiated heat sealing parameters are applied to each control block to implement the heat sealing operation.

[0031] Optionally, establishing a multi-layer perceptron neural network optimization controller includes:

[0032] Construct a multi-layer perceptron neural network optimization controller, which includes an input layer, a hidden layer, and an output layer. The input layer receives the system state parameter set, and the output layer generates the optimal control parameter combination.

[0033] Taking the filling and sealing process quality index evaluation system as the optimization target, a multi-layer perceptron neural network optimization controller is trained;

[0034] Configure a dual-mode control mechanism for the multi-layer perceptron neural network optimization controller;

[0035] Through the parameter sensitivity analysis mechanism of the multi-layer perception neural network optimization controller, the control parameter combination is optimized to complete the adaptive adjustment of the parameters of the entire filling and sealing process.

[0036] Optionally, identifying an abnormal temperature region in the container sealing area includes:

[0037] Convert temperature distribution data into a two-dimensional temperature matrix;

[0038] A multi-scale thermal anomaly detection mechanism is introduced to perform multi-scale decomposition of the two-dimensional temperature matrix through wavelet transform, establish thermal feature subspace, and extract abnormal feature points;

[0039] Calculate the temperature gradient vector between adjacent points in the two-dimensional temperature matrix to obtain the temperature gradient field, and mark the temperature anomaly candidate points according to the preset temperature gradient threshold;

[0040] Merge the abnormal feature points with the temperature anomaly candidate points to obtain the enhanced temperature anomaly candidate point set;

[0041] Connectivity analysis is performed on the enhanced temperature anomaly candidate point set to obtain the temperature anomaly area.

[0042] The second aspect of the present invention provides a fully automatic filling and sealing system based on image recognition and analysis, comprising:

[0043] Parameter analysis module, used to build a three-dimensional dynamic model of the fluid and extract the fluid dynamic control parameters and bubble group dynamic characteristic parameters through image recognition;

[0044] The electric field control module is used to control the annular microelectrode array set around the filling nozzle to generate a directional electric field according to the fluid dynamic control parameters, thereby regulating the liquid flow state;

[0045] The ultrasonic modulation module is used to determine the ultrasonic action parameters according to the dynamic characteristic parameters of the bubble group and the physical properties of the container liquid, and control the frequency-adjustable ultrasonic generator to output ultrasonic modulation pulses to the bubble accumulation area;

[0046] The heat seal adjustment module is used to detect the temperature distribution of the container sealing area, control the local adjustable heat seal head based on the detection results, and apply differentiated heat seal parameters to different temperature areas;

[0047] The control optimization module is used to establish a multi-layer perception neural network optimization controller, receive the system state parameter set, take the filling and sealing process quality indicators as the optimization target, optimize the control parameter combination, and complete the adaptive adjustment of the parameters of the entire filling and sealing process.

[0048] The beneficial effects of the present invention are: the present invention adopts a filling method that combines image recognition and electric field regulation, which realizes precise control of the liquid flow state, effectively solves the problems of liquid drawing and flow interruption in the traditional filling process, and improves the filling accuracy and efficiency; the directional ultrasonic technology based on the dynamic characteristics of bubbles can eliminate bubbles in the container in a targeted manner, greatly reducing the bubble residue rate in the product, while avoiding damage to the effective ingredients of the liquid and ensuring product quality; the differentiated heat sealing parameter application system solves the problems such as poor sealing reliability caused by uneven container sealing temperature in the traditional sealing process, and improves the consistency of sealing strength and product shelf life; the multi-layer perception neural network optimization control system realizes the intelligent coordinated adjustment of parameters of the entire filling and sealing process, breaking through the limitations of independent control of each link in the traditional way, so that the system has the adaptive ability to cope with complex and changeable production scenarios, and greatly improves production stability and product consistency. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0050] Figure 1 The figure is a flow chart of a fully automatic filling and sealing method based on image recognition and analysis.

[0051] Figure 2 The figure shows the ultrasonic modulation pulse output flow chart of a fully automatic filling and sealing method based on image recognition and analysis.

[0052] Figure 3 A flow chart for determining abnormal temperature areas in a fully automated filling and sealing method based on image recognition and analysis. DETAILED DESCRIPTION

[0053] In order to make the purpose, features, and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0054] Example 1, with reference to Figures 1 to 3 , which is the first embodiment of the present invention, provides a fully automatic filling and sealing method based on image recognition analysis. The flow chart of the method is as follows Figure 1 As shown, the method includes:

[0055] S1: A dual-frequency flash stereo imaging system is used to construct a three-dimensional dynamic model of the fluid, and image recognition is used to extract the fluid dynamic control parameters and bubble group dynamic characteristic parameters.

[0056] Among them, the fluid dynamic control parameters include the surface stress tensor field, three-dimensional flow velocity vector distribution, and liquid surface motion state parameters; the bubble group dynamic characteristic parameters include bubble volume concentration gradient, bubble equivalent particle size distribution, and three-dimensional coordinates of the bubble aggregation area.

[0057] First, the dual-frequency flash stereo imaging system was activated to capture the fluid from multiple angles at different times, acquiring multiple sets of fluid image data. These sets of fluid image data were then preprocessed: histogram equalization was used to enhance the image to improve the recognition of fluid and bubbles, and Gaussian filtering was used to remove interference such as thermal noise to ensure image clarity.

[0058] Next, an image matching algorithm is used to process images from different perspectives and identify corresponding features. Incorporating the principles of stereo vision, the two-dimensional image information is converted into three-dimensional spatial information to construct a three-dimensional dynamic model of the fluid. Fluid physical property data is calibrated according to standard methods to reduce measurement errors. For multi-source spatial coordinate data, a conversion model is established to unify the coordinate system.

[0059] Subsequently, various parameters were extracted based on the constructed three-dimensional dynamic model of the fluid: the fluid surface information was analyzed, and the surface stress tensor field was calculated based on fluid mechanics theory; particle image velocimetry technology was used to track the motion trajectory of tracer particles to obtain the three-dimensional velocity vector distribution; the liquid level height change and inclination angle were monitored, and corrections were made considering the liquid surface reflection and fluctuation factors to obtain the liquid surface motion state parameters; the bubble volume proportions in different regions and their spatial changes were counted to obtain the bubble volume concentration gradient; the bubbles were segmented and identified, their sizes were measured and statistically analyzed to obtain the bubble equivalent particle size distribution; the relatively dense bubble areas were marked, and the three-dimensional coordinates of the bubble aggregation area were determined based on the three-dimensional information.

[0060] Ultimately, the obtained fluid dynamic control parameters and bubble group dynamic characteristic parameters are used to precisely control the annular microelectrode array to generate a directional electric field, so that the liquid flow state meets the filling requirements, and to determine the appropriate ultrasonic action parameters to eliminate bubbles and improve product quality and production efficiency.

[0061] S2: Based on the fluid dynamic control parameters, the annular microelectrode array set around the filling nozzle is controlled to generate a directional electric field, regulate the liquid flow state, and suppress the liquid drawing phenomenon at the end of filling.

[0062] In a specific embodiment of the present invention, step S2 specifically includes:

[0063] S2.1: Arrange multiple independently controllable microelectrodes in a circular direction around the nozzle to form a circular microelectrode array.

[0064] The microelectrodes range from 8 to 16 and are evenly distributed. Each microelectrode is made of a corrosion-resistant alloy, operates in a voltage range of 0-3000V, and measures 1.0mm x 3mm. The spacing between the electrodes and the filling channel is maintained at 3.0-5.0mm, ensuring sufficient field strength without interfering with fluid flow. By arranging a circular microelectrode array around the filling nozzle, the liquid flow state can be flexibly and precisely controlled, breaking through the limitations of traditional filling technology, which relies on a single method of liquid control.

[0065] S2.2: Based on the fluid dynamic control parameters, generate a control parameter set for each microelectrode.

[0066] First, the surface stress tensor field is characteristically decomposed to determine the maximum principal stress direction and stress gradient distribution. The three-dimensional velocity vector is then decomposed into radial and tangential components. The vorticity distribution is determined by calculating the velocity field curl, and the region where the vorticity exceeds a critical threshold is defined as the vortex core region. The critical threshold is 1.5 times the average vorticity or 30% of the maximum vorticity, whichever is greater. The microelectrode response timing is then optimized by combining the fluctuation frequency information in the liquid surface motion state parameters. Subsequently, the microelectrode voltage distribution is generated based on the stress gradient distribution, and electric field rotation compensation is superimposed in the vortex core region. Furthermore, adjacent microelectrodes are configured with a phase difference within the vortex core region, while maintaining phase synchronization in the non-vortex core region. Finally, the fluid dynamic response time is calculated based on the Reynolds and Weber numbers of the fluid, and the microelectrode triggering timing is generated accordingly. The voltage distribution, phase difference, and triggering timing are then integrated to output a microelectrode control parameter set. Generating a microelectrode control parameter set based on the fluid dynamic control parameters enables flexible adjustment of the electric field according to the real-time state of the fluid, effectively improving the adaptability to complex fluid flow states and the controllability of liquid flow during the filling process.

[0067] S2.3: Apply a set of control parameters to the annular microelectrode array to generate a directional electric field with spatially non-uniform and time-varying characteristics.

[0068] First, the voltage distribution parameters are converted into driving voltage signals for each microelectrode and output to the corresponding microelectrode via a high-voltage signal amplification module. A multi-channel signal generator is then configured based on the phase difference parameters to achieve phase difference control between adjacent microelectrodes. Next, the central control unit activates the driving circuits of each microelectrode in a time-sharing manner according to the trigger timing parameters. Simultaneously, the actual output voltage and phase signal of each microelectrode are monitored in real time to verify spatial non-uniformity and time-varying characteristics. When an electric field parameter deviation exceeds the limit, a hierarchical compensation control is implemented, including coordinated compensation of adjacent electrodes, trigger timing reallocation, and control parameter regeneration requests. Excessive deviations here specifically refer to: a driving voltage deviation of ±5% from the set value; an actual phase difference deviation of ±3° from the set value; or an electric field intensity distribution deviation exceeding 10% from the preset pattern.

[0069] Among them, spatial non-uniformity verification refers to confirming that the electric field forms a non-uniform field (including intensity gradient distribution and / or direction gradient distribution) that conforms to the preset distribution pattern within the array coverage area, and time-varying characteristic verification refers to confirming that the changes in electric field parameters over time conform to the preset dynamic change laws.

[0070] S2.4: Generate electric stress distribution at the liquid-gas interface through a directional electric field to regulate the liquid flow state.

[0071] First, the electric stress distribution at the liquid-gas interface is calculated based on the electric field intensity distribution generated by the annular microelectrode array. This distribution is proportional to the square of the electric field intensity. Next, the calculated electric stress distribution is converted into an equivalent surface tension correction and superimposed with the surface stress tensor field extracted from S1 to obtain the corrected surface force distribution. Then, based on the liquid surface motion state parameters and three-dimensional flow velocity vector distribution extracted from S1, the liquid surface stability index (defined as the ratio of the local curvature change rate of the liquid surface to the average curvature) is calculated. When the stability index exceeds the preset liquid surface fluctuation tolerance threshold of 0.25, the electric field adjustment is performed: The target electric field distribution is calculated based on the liquid surface state, and the voltage adjustment parameters and phase adjustment parameters of the microelectrode are generated. The preset allowable threshold of liquid surface fluctuation is determined through pre-experimental calibration according to the fluid characteristics and filling accuracy requirements. Subsequently, the voltage adjustment parameters and phase adjustment parameters are applied to the microelectrode array to reconstruct the electric field distribution. The reconstructed electric field generates a corrected electric stress at the liquid-gas interface, changes the local surface tension balance of the liquid surface, induces surface flow, and thus regulates the overall liquid flow state. Finally, the changes in the liquid surface motion state parameters are continuously monitored, and the electric field control strategy is updated in real time to form a closed-loop control to maintain the stability of the filling process.

[0072] It is worth noting that the rigorous verification of the spatial non-uniformity and time-varying characteristics of the electric field and the hierarchical compensation control mechanism ensure the stable operation of the electric field, reduce filling anomalies caused by electric field deviations, and improve the stability and reliability of the entire filling process.

[0073] S2.5: At the end of the filling process, based on the liquid surface motion parameters, the microelectrode voltage distribution is adjusted to form a convergent electric field gradient pointing to the center of the filling nozzle to control the liquid flow interruption.

[0074] First, by monitoring the liquid volume and flow rate, the final filling stage is determined when preset conditions are met, including the liquid volume reaching 95% of the target value and the flow rate decreasing to below 0.1 m / s. Then, based on the liquid surface motion parameters, the microelectrode voltage distribution is reconfigured: the voltage of the peripheral electrodes is set higher than that of the electrodes in the central region, forming an electric field gradient pointing toward the center of the nozzle. Next, the microelectrode phases are adjusted to be in phase to ensure consistency in the electric field direction. Subsequently, the change in the liquid neck diameter is monitored. When the liquid neck diameter drops to the flow-off trigger diameter (typically 10% of the nozzle inner diameter), an electric field pulse sequence is applied to accelerate the flow-off. Furthermore, a weakened electric field is briefly maintained after the flow is shut off (the intensity is reduced to 30% of the peak value) to eliminate residual droplets and prevent liquid stringing. Finally, the electric field is smoothly turned off according to an exponential decay curve, completing the filling process. This special electric field adjustment method, designed for the final stage of filling, effectively solves the problem of liquid stringing, a long-standing problem in the filling process, improving filling accuracy, reducing material loss, and enhancing product consistency.

[0075] S3: Determine the ultrasonic action parameters according to the dynamic characteristic parameters of the bubble group and the physical properties of the container liquid, control the frequency-adjustable ultrasonic generator to output ultrasonic modulated pulses for the bubble gathering area, and eliminate the bubbles generated during the filling process.

[0076] In a specific embodiment of the present invention, the ultrasonic modulation pulse output flow chart is as follows: Figure 2 , specifically including:

[0077] S3.1: Calculate the ultrasonic transmission loss coefficient and liquid attenuation coefficient based on the physical properties of the container liquid.

[0078] Among them, the container liquid physical properties include container structural characteristics and liquid physical properties. The container structural characteristics include the acoustic impedance of the container material and the container wall thickness. The liquid physical properties include liquid viscosity, liquid density and sound speed.

[0079] First, the acoustic impedance of the liquid is calculated based on the liquid density and sound velocity. Then, combined with the acoustic impedance of the container material, the interface sound energy transmission coefficient and reflection coefficient are calculated according to acoustic theory. Next, based on the container wall thickness, ultrasonic frequency, and sound velocity in the container material, the phase difference of the ultrasonic wave propagating in the container wall is calculated according to wave theory, and the correction factor is obtained through transformation. In addition, considering the multiple reflections of the ultrasonic wave at the interface between the liquid and the container wall, the total transmitted energy is obtained by summing the reflection coefficient and transmission coefficient through an infinite geometric series. Finally, the transmission loss value is calculated by combining the total transmitted energy and the correction factor.

[0080] At the same time, based on the classical viscous attenuation theory, the liquid attenuation coefficient is calculated using liquid viscosity, liquid density, sound speed, and ultrasonic frequency.

[0081] S3.2: Determine the bubble resonance frequency range based on the bubble equivalent particle size distribution.

[0082] First, the bubble equivalent size distribution data is divided into multiple size intervals at preset intervals, and the proportion of bubbles in each interval is calculated. Then, the Minnaert resonance formula is applied to the bubbles in each size interval to calculate the theoretical resonant frequency. Next, the theoretical resonant frequency is corrected by determining the viscous damping correction factor based on the liquid viscosity, taking into account the effect of liquid viscosity on bubble resonance. Subsequently, a frequency-weighted average is calculated based on the bubble population proportion to determine the frequency interval covering the main bubble population (with a cumulative proportion exceeding 85%). Finally, the frequency interval boundaries are adjusted based on the acoustic response characteristics (including vibration amplitude, phase response, and damping characteristics) of bubbles of different equivalent sizes in the acoustic field to obtain the bubble resonance frequency range. For example, if the vibration amplitude of bubbles of a certain equivalent size varies dramatically near a certain frequency, the interval boundaries near that frequency can be appropriately widened. If there is a significant abrupt change in the phase response, the boundaries can also be adjusted accordingly.

[0083] The preset interval is a bubble equivalent particle size division unit determined based on the resolution accuracy of the image recognition system and the bubble size distribution range.

[0084] It is noteworthy that by carefully classifying the bubble sizes, considering the modification of the resonant frequency by liquid viscosity, and adjusting the frequency interval boundaries in combination with the acoustic response characteristics, the determined frequency range can effectively cover the main bubble groups. This method allows ultrasound to more effectively stimulate bubble resonance, which to a certain extent improves the efficiency of bubble elimination and improves the difficulty of traditional methods in accurately targeting bubbles of different particle sizes.

[0085] S3.3: Based on the bubble volume concentration gradient and the three-dimensional coordinates of the bubble aggregation area, adjust the beam focusing parameters of the frequency-adjustable ultrasonic generator so that the acoustic energy is directionally focused on the bubble aggregation area.

[0086] First, based on the bubble volume concentration gradient data, the area in the container where the bubble concentration is higher than the preset concentration threshold is identified, and the boundary of the bubble aggregation area that needs to be processed is determined; then, based on the three-dimensional coordinates of the bubble aggregation area, a ray tracing algorithm is used to calculate the propagation path of the sound wave from the generator to the discrete sampling points in the bubble aggregation area; then, the propagation path is corrected in combination with the container structure and the liquid sound velocity to determine the spatial pointing angle of the generator; then, according to the geometric characteristics of the bubble aggregation area, the phased array element parameters of the generator are adjusted to control the phase difference between the array elements and the shape of the sound wave front; in addition, by adjusting the phase difference and amplitude of each array element, a focused sound field that effectively covers the bubble aggregation area is formed; finally, the focusing effect is evaluated, and automatic calibration is triggered when the focusing effect is lower than the set standard; the changes in the bubble aggregation state are monitored in real time, and the beam focusing parameters are dynamically adjusted according to the preset frequency to ensure that the sound energy is always directionally focused on the current bubble aggregation area.

[0087] Among them, the beam focusing parameters include phased array element parameters, phase difference between elements, and element amplitude; the preset concentration threshold is determined according to the filling liquid type and filling rate; the set standard refers to the uniformity of the sound energy distribution of the focused sound field in the target area is not less than 80%; the preset frequency refers to the time interval for updating the beam focusing parameters, which depends on the rate of change of the bubble aggregation state.

[0088] Crucially, the beam focusing parameters are adjusted based on the bubble volume concentration gradient and the three-dimensional coordinates of the bubble cluster area, enabling the acoustic energy to be focused directly on the bubble cluster. This focusing method reduces the ineffective dispersion of ultrasonic energy and enhances its effect on bubbles in specific areas. Furthermore, real-time monitoring and dynamic adjustment ensure continuous bubble elimination even when the bubble state changes during the filling process, improving the stability of the bubble elimination process and addressing the energy dispersion and poor localized bubble elimination effects of traditional methods.

[0089] S3.4: Configure the frequency-adjustable ultrasonic generator to a pulse output mode, and determine the pulse period and pulse intensity based on the transmission loss coefficient, the liquid attenuation coefficient, and the bubble resonance frequency range.

[0090] First, based on the transmission loss coefficient and the liquid attenuation coefficient, the average attenuation value of the ultrasonic energy from the generator to each discrete sampling point in the bubble aggregation area is calculated; then, according to the bubble resonance frequency range, the ultrasonic frequency and modulation bandwidth of the generator are set to ensure that the output ultrasonic frequency can effectively excite bubble resonance; then, based on the average attenuation value and the structural characteristics of the container, the initial pulse intensity is determined so that the acoustic energy is sufficient to induce bubble resonance when it reaches the bubble aggregation area; in addition, by analyzing the distribution of acoustic energy in the bubble aggregation area under different pulse duty ratios, the pulse duty ratio is adjusted to ensure efficient utilization of acoustic energy; then, combined with the dynamic characteristic parameters of the bubble group, a pulse repetition frequency that matches the bubble resonance characteristics is designed; at the same time, based on the energy distribution of the ultrasonic wave in the bubble aggregation area and the bubble resonance characteristics, the pulse duration that can ensure the optimal resonance effect is determined; finally, different pulse parameter combinations and corresponding bubble elimination efficiency data are collected and compiled into a parameter optimization table; the real-time bubble elimination situation is compared with the data in the parameter optimization table, and the pulse period and pulse intensity are regulated using an adaptive control algorithm, laying the foundation for the subsequent realization of efficient bubble elimination.

[0091] S3.5: Output ultrasonic modulated pulses to the bubble gathering area through the frequency-adjustable ultrasonic generator, so that the bubbles reach a resonance state, prompting the bubbles to float up and be eliminated.

[0092] Specifically, a shielded cable is used to connect the generator to the power supply and control system. After the equipment is started, a self-test program is executed, and the results are fed back to the control system after the self-test is completed. Subsequently, the control system sends the pulse parameters determined by S3.4 to the generator, and the generator receives and stores the parameters. Next, the generator uses digital signal processing technology to generate and modulate the electric pulse signal according to the stored pulse parameters, including frequency and duty cycle modulation, so that the generated pulse characteristics match the bubble resonance frequency range. The modulated electric pulse signal is then power amplified and the amplified electric signal is converted into ultrasonic waves. In addition, according to the beam focusing parameters determined by S3.3, the ultrasonic waves are directionally focused and emitted to the bubble aggregation area to improve the effect of exciting bubble resonance. Finally, the focused ultrasonic waves act on the bubbles, prompting the bubbles to overcome resistance and float to the liquid surface and disappear.

[0093] S3.6: Adaptively adjust the ultrasonic frequency, pulse period, and pulse intensity based on the bubble elimination situation, while controlling the pulse intensity to be lower than the liquid component protection threshold.

[0094] The liquid component protection threshold is the maximum pulse intensity that does not decompose or denature the active ingredients in the liquid due to heat or mechanical stress, thereby affecting product quality. This threshold is determined through a comprehensive chemical analysis of the liquid's composition, combined with thermal stability testing, mechanical sensitivity testing, and reaction experiments simulated under ultrasonic waves.

[0095] S4: Use thermal imaging technology to detect the temperature distribution in the container sealing area, control the local adjustable heat sealing head based on the detection results, and apply differentiated heat sealing parameters to different temperature areas.

[0096] In a specific embodiment of the present invention, step S4 specifically includes:

[0097] S4.1: Scan the container sealing area using thermal imaging technology to obtain temperature distribution data in the container sealing area.

[0098] S4.2: Perform temperature gradient analysis based on the temperature distribution data to identify abnormal temperature areas in the container sealing area.

[0099] Specifically, the flow chart for determining the abnormal temperature area is as follows: Figure 3 As shown, the method includes: pre-processing the temperature distribution data and converting it into a two-dimensional temperature matrix, where each element of the two-dimensional temperature matrix represents the temperature value at the corresponding position of the container sealing area; introducing a multi-scale thermal anomaly detection mechanism, performing multi-scale decomposition on the two-dimensional temperature matrix through wavelet transform, establishing thermal feature subspaces in different frequency domains, and extracting abnormal feature points respectively, and the steps are as follows: selecting an adaptive wavelet basis function according to the data characteristics and analysis requirements of the two-dimensional temperature matrix, and determining the number of decomposition layers of the wavelet transform; performing multi-layer wavelet decomposition on the two-dimensional temperature matrix using the selected wavelet basis function to obtain low-frequency submatrices, horizontal high-frequency submatrices, vertical high-frequency submatrices, and diagonal high-frequency submatrices in different frequency domains, and constructing thermal feature subspaces; for each thermal feature subspace, calculating statistical features, setting a threshold according to the extracted eigenvector, and marking points whose eigenvalues ​​exceed the threshold as abnormal feature candidate points; comprehensively considering the correlation and consistency of different thermal feature subspaces, determining points that present abnormal features in multiple subspaces as final abnormal feature points, and removing isolated abnormal points caused by noise.

[0100] Subsequently, the temperature gradient vectors between adjacent points in the two-dimensional temperature matrix are calculated to obtain the temperature gradient field of the container sealing area; a preset temperature gradient threshold is set, and points with gradient values ​​exceeding the threshold are marked as temperature anomaly candidate points; the abnormal feature points are merged with the temperature gradient anomaly point set through a feature fusion algorithm to form an enhanced temperature anomaly candidate point set; a connectivity analysis is performed on the enhanced temperature anomaly candidate point set, and points with adjacent spatial positions and similar temperature characteristics are classified into the same temperature anomaly area; the area size, average temperature value and temperature standard deviation of each temperature anomaly area are calculated, and areas with an area smaller than the preset area threshold or a temperature standard deviation lower than the preset standard deviation threshold are eliminated to obtain the final temperature anomaly area.

[0101] Among them, the preset temperature gradient threshold, preset area threshold and preset standard deviation threshold are determined based on statistical analysis of historical data, characteristics of the container sealing area and production practice, normal temperature fluctuation range and expert experience.

[0102] S4.3: Based on the temperature anomaly area, the container sealing area is divided into multiple independent control blocks, and the temperature characteristic parameters of each control block are calculated.

[0103] Specifically, the container sealing area is divided into an abnormal temperature area and a normal temperature area based on the boundary contour of the abnormal temperature area; the minimum control unit size of the container sealing area is determined according to the preset control accuracy requirements; the abnormal temperature area is further subdivided according to the minimum control unit size to generate an abnormal temperature control block; the normal temperature area is divided into several normal temperature control blocks according to the preset area division rules, and the area division rules include the maximum size limit of the control unit and the boundary continuity requirement; a unique identification code is assigned to each control block, and a control block index table is established; the temperature characteristic parameters of each control block are calculated, including the average temperature value, maximum temperature value, minimum temperature value, temperature standard deviation and temperature uniformity index of the block; the temperature characteristic parameters of all control blocks are associated with the control block index table to form a block temperature characteristic data structure of the container sealing area. This step breaks through the limitation of the traditional unified treatment of the entire sealing area, and can formulate personalized processing strategies for areas with different temperature characteristics, thereby improving the pertinence and effectiveness of heat sealing parameter adjustment.

[0104] S4.4: Extract standard heat sealing parameters corresponding to the container material from a preset material-temperature parameter database.

[0105] Among them, the standard heat sealing parameters include reference heat sealing temperature, pressure and time parameters.

[0106] S4.5: Generate differentiated heat sealing parameters for each control block through parameter compensation calculation based on the temperature characteristic parameters of each control block and the standard heat sealing parameters.

[0107] Specifically, a mapping function between temperature characteristic parameters and heat sealing parameter adjustments is established, which describes the impact of temperature differences on heat sealing temperature, pressure, and time parameters. For each control block, the difference between the block average temperature value and the baseline heat sealing temperature is calculated, and the parameter adjustment direction and amplitude are determined based on the block temperature standard deviation and temperature uniformity index. Based on the calculation results, the adjustment amounts of the heat sealing temperature, pressure, and time parameters are determined respectively. When the temperature difference is positive, the heat sealing temperature is lowered. When the temperature standard deviation is large, the heat sealing pressure is increased. When the temperature uniformity is low, the heat sealing time is extended. The baseline heat sealing parameters are added to the corresponding parameter adjustments to generate the differentiated heat sealing temperature, pressure, and time parameters of the control block. The differentiated heat sealing parameters of adjacent control blocks are smoothed to ensure that the parameters change continuously in space and avoid uneven sealing quality. This step can better adapt to temperature changes at different locations in the container sealing area. Compared with the traditional method of fixing heat sealing parameters, it improves the adaptability of the heat sealing process to the actual conditions of the container and effectively ensures the consistency of sealing quality.

[0108] S4.6: Heat sealing operations are performed by applying corresponding differentiated heat sealing parameters to each control zone through a locally adjustable heat sealing head with multi-zone independent control function.

[0109] Specifically, the spatial position information of the control block is mapped to the execution unit of the local adjustable heat sealing head to establish a one-to-one correspondence between the control block and the execution unit; the differentiated heat sealing parameters of each control block are transmitted to the control system of the local adjustable heat sealing head; each execution unit of the local adjustable heat sealing head independently adjusts its own heat sealing temperature, pressure and action time according to the received differentiated heat sealing parameters; the local adjustable heat sealing head is started to synchronously perform heat sealing operations on each control block in the container sealing area to form a complete seal.

[0110] S4.7: During the heat sealing process, the temperature changes of each control block are monitored in real time through thermal imaging technology. The monitoring data is fed back to the control system to dynamically adjust the differentiated heat sealing parameters to ensure the sealing quality.

[0111] S5: Establish a multi-layer perception neural network optimization controller, receive the system state parameter set, take the filling and sealing process quality index as the optimization target, optimize the control parameter combination, and complete the adaptive adjustment of the parameters of the entire filling and sealing process.

[0112] In a specific embodiment of the present invention, step S5 specifically includes:

[0113] S5.1: Construct a multi-layer perceptron neural network optimization controller. The multi-layer perceptron neural network optimization controller includes an input layer, a hidden layer, and an output layer. The input layer receives a set of system state parameters, and the output layer generates an optimal control parameter combination.

[0114] Specifically, a deep multi-layer perceptual neural network structure with memory units is constructed, including 1 input layer, an appropriate number of hidden layers (such as 4) determined experimentally, and 1 output layer, wherein at least one hidden layer is an LSTM structure to process sequence data; a multi-channel feature mapping module is configured in the input layer to receive fluid dynamic control parameters, bubble group dynamic feature parameters, container temperature distribution data, and equipment state parameters to form a system state parameter set; a residual connection structure is set between hidden layers to alleviate the gradient vanishing problem caused by the increase in network depth; an attention mechanism module is embedded in the hidden layer to dynamically assign weights to key features; a multi-task learning structure is designed in the output layer to simultaneously generate a control parameter combination, namely, the optimal combination of annular microelectrode array control parameters, ultrasonic action parameters, and heat sealing parameters; batch normalization processing technology is used to accelerate network training and improve the generalization ability of the model; a dropout mechanism is introduced to prevent overfitting, and an appropriate dropout rate is set (such as 0.3).

[0115] S5.2: Based on historical filling and sealing process data, and with the filling and sealing process quality index evaluation system as the optimization target, a multi-layer perceptron neural network optimization controller is trained.

[0116] Specifically, historical filling and sealing process data were collected, covering the system status parameter set and the corresponding control parameter combination and quality index data, which were cleaned and labeled, and divided into training set, validation set and test set according to the ratio of 70%, 15% and 15% respectively; a multi-dimensional filling and sealing process quality index evaluation system was constructed, including filling accuracy index, liquid level stability index, bubble residual rate index and sealing strength index; the multi-dimensional filling and sealing process quality index evaluation system was converted into a weighted combination loss function, and the weight coefficient of each index was determined according to product characteristics, production process requirements and expert experience.

[0117] In order to enrich the training data and enhance the adaptability of the model, a generative adversarial network (GAN) is used to expand the training data. The generator generates a set of system state parameters similar to the real data distribution by learning the characteristics of the real data. At the same time, a filling and sealing knowledge graph is constructed, and the relationship between parameters and quality indicators is extracted and graphed. With the help of graph embedding technology, the model can quickly grasp the key data connections and accelerate convergence.

[0118] In the model training phase, the training set is input into the multi-layer perceptron neural network optimization controller, and the stochastic gradient descent method is used for iterative training. The predicted value is obtained and the loss is calculated through forward propagation. The weights and biases are updated through back propagation. The validation set is used to regularly evaluate the model performance to dynamically adjust the hyperparameters.

[0119] After the model training is completed, the test set data is input into the model to calculate the prediction accuracy of various quality indicators, verify the generalization ability of the model on unknown data, and ensure that an effective control parameter combination can be generated.

[0120] S5.3: Configure a dual-mode control mechanism for the multi-layer perceptron neural network optimization controller, where the dual-mode control mechanism includes an offline learning mode and an online adaptive mode.

[0121] Specifically, in offline learning mode, the multi-layer perception neural network optimization controller learns and trains based on historical filling and sealing process data, and uses algorithms such as stochastic gradient descent to adjust model weights and biases until the loss function converges to a satisfactory range; in online adaptive mode, the system state parameters of the filling and sealing process are monitored in real time. Once it is detected that the parameters deviate from the normal range, a new control parameter combination is immediately generated according to the trained model, and the filling and sealing process is adjusted in real time.

[0122] At the same time, a mode switching mechanism is designed to switch between offline learning mode and online adaptive mode based on production needs and model status. For example, when a new product or process is introduced, the model switches to offline learning mode. If the model's prediction error exceeds a preset threshold multiple times in a row in online adaptive mode, it switches to offline learning mode for retraining.

[0123] S5.4: Through the parameter sensitivity analysis mechanism of the multi-layer perception neural network optimization controller, the control parameter combination is optimized to complete the adaptive adjustment of the parameters of the entire filling and sealing process.

[0124] Since actual production conditions are complex and changeable and contain many uncertainties, it is difficult to always maintain the optimal state by relying solely on the control parameter combination generated by the controller trained with S5.2. Therefore, a parameter sensitivity analysis mechanism is needed to deeply analyze the impact of each parameter on the filling and sealing process and optimize the control parameter combination in a targeted manner.

[0125] Specifically, the parameters to be analyzed are first determined, including control parameters such as filling speed, sealing temperature, and pressure, as well as influencing factors such as material properties and environmental conditions. A global sensitivity analysis method is used to evaluate the degree of influence of each parameter on the quality index. A multi-objective optimization function that comprehensively considers quality indexes, production efficiency, and cost factors is set, and an intelligent optimization algorithm is used to search for the optimal control parameter combination in the parameter space. Various sensors are installed on the production line to collect system status parameters and quality index data in real time and transmit them to the controller for processing. The controller dynamically adjusts the control parameters based on the parameter sensitivity analysis results and real-time monitoring data, adopts different adjustment strategies in different production stages, and continuously optimizes the parameter analysis and control parameter combination based on the deviation between the actual production results and the optimization target to achieve adaptive adjustment of the parameters of the entire filling and sealing process.

[0126] By constructing a multi-layer perceptual neural network optimization controller, the system state parameters of each filling and sealing process are integrated and collaboratively optimized. By augmenting data with a generative adversarial network and assisting with knowledge graph training, the model's adaptability to complex working conditions and the accuracy of parameter optimization are enhanced. A dual-mode control mechanism and mode switching strategy enhance the automation level and product quality stability of the filling and sealing process. A parameter sensitivity analysis mechanism enables adaptive adjustment of parameters throughout the filling and sealing process, strengthening the system's ability to cope with complex and changing production scenarios.

[0127] Furthermore, this embodiment also provides a fully automatic filling and sealing system based on image recognition and analysis, including: a parameter analysis module, which is used to construct a three-dimensional dynamic model of the fluid and extract the fluid dynamic control parameters and bubble group dynamic characteristic parameters through image recognition; an electric field control module, which is used to control the annular microelectrode array arranged around the filling nozzle to generate a directional electric field according to the fluid dynamic control parameters, so as to regulate the liquid flow state; an ultrasonic modulation module, which is used to determine the ultrasonic action parameters according to the dynamic characteristic parameters of the bubble group and the physical properties of the container liquid, and control the adjustable frequency ultrasonic generator to output ultrasonic modulated pulses for the bubble aggregation area; a heat sealing adjustment module, which is used to detect the temperature distribution of the container sealing area, control the local adjustable heat sealing head according to the detection results, and apply differentiated heat sealing parameters to different temperature areas; a control optimization module, which is used to establish a multi-layer perception neural network optimization controller, receive a system state parameter set, optimize the control parameter combination with the filling and sealing process quality index as the optimization target, and complete the adaptive adjustment of the parameters of the entire filling and sealing process.

[0128] In summary, the present invention adopts a filling method that combines image recognition and electric field regulation, which realizes precise control of the liquid flow state, solves the problems of liquid drawing and inaccurate flow interruption in the traditional filling process, and improves the filling accuracy and efficiency; the directional ultrasonic technology based on the dynamic characteristics of bubbles can eliminate bubbles in the container in a targeted manner, reduce the residual bubble rate of the product, and avoid damage to the effective ingredients of the liquid, thereby ensuring product quality; the differentiated heat sealing parameter application system solves the problems such as poor sealing reliability caused by uneven container sealing temperature in the traditional sealing process, and improves the consistency of sealing strength and product shelf life; the multi-layer perception neural network optimization control system realizes the intelligent coordinated adjustment of parameters in the entire filling and sealing process, breaking through the limitations of independent control of each link in the traditional way, so that the system has the adaptive ability to cope with complex and changeable production scenarios, and improves production stability and product consistency.

[0129] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A fully automatic filling and sealing method based on image recognition and analysis, characterized in that: include: Construct a three-dimensional dynamic model of the fluid and extract the fluid dynamic control parameters and bubble group dynamic characteristic parameters through image recognition; According to the fluid dynamic control parameters, the annular microelectrode array arranged around the filling nozzle is controlled to generate a directional electric field to regulate the liquid flow state; Determine ultrasonic action parameters according to the dynamic characteristic parameters of the bubble group and the physical properties of the container liquid, and control the frequency-adjustable ultrasonic generator to output ultrasonic modulated pulses for the bubble gathering area; Detect the temperature distribution in the container sealing area, control the locally adjustable heat sealing head based on the detection results, and apply differentiated heat sealing parameters to different temperature areas; A multi-layer perception neural network optimization controller is established to receive the system state parameter set, take the filling and sealing process quality index as the optimization target, optimize the control parameter combination, and complete the adaptive adjustment of the parameters of the entire filling and sealing process; The control of the annular microelectrode array arranged around the nozzle to generate a directional electric field comprises: Arrange multiple microelectrodes around the nozzle to form a circular microelectrode array; Based on the fluid dynamic control parameters, a control parameter set for each microelectrode is generated; applying the control parameter set to the annular microelectrode array to generate a directional electric field; The directional electric field forms an electric stress distribution at the liquid-gas interface to regulate the liquid flow state; At the end of the filling process, based on the liquid surface motion parameters, the microelectrode voltage distribution is adjusted to form a convergent electric field gradient pointing to the center of the filling nozzle, thus controlling the liquid flow interruption. The controlling of the frequency-adjustable ultrasonic generator to output ultrasonic modulation pulses in the bubble accumulation area comprises: Calculate the ultrasonic transmission loss coefficient and liquid attenuation coefficient based on the physical properties of the container liquid; Determine the bubble resonance frequency range based on the bubble equivalent particle size distribution; Adjust the beam focusing parameters of the frequency-adjustable ultrasonic generator based on the bubble volume concentration gradient and the three-dimensional coordinates of the bubble aggregation area; The frequency-adjustable ultrasonic generator is configured to a pulse output mode, and the pulse period and pulse intensity are determined according to the transmission loss coefficient, the liquid attenuation coefficient, and the bubble resonance frequency range; Ultrasonic wave modulation pulses are output to the bubble gathering area through a frequency-adjustable ultrasonic wave generator.

2. The fully automatic filling and sealing method based on image recognition analysis according to claim 1 is characterized in that: The fluid dynamic control parameters include surface stress tensor field, three-dimensional velocity vector distribution, and liquid surface motion state parameters; the bubble group dynamic characteristic parameters include bubble volume concentration gradient, bubble equivalent particle size distribution, and three-dimensional coordinates of bubble aggregation area.

3. The fully automatic filling and sealing method based on image recognition analysis according to claim 1 is characterized in that: The differential heat sealing parameters applied to different temperature zones include: Use thermal imaging technology to scan the container sealing area and obtain temperature distribution data of the container sealing area; Performing temperature gradient analysis based on the temperature distribution data to identify abnormal temperature areas in the container sealing area; Based on the temperature abnormality area, the container sealing area is divided into multiple independent control blocks, and the temperature characteristic parameters of each control block are calculated; Extract standard heat sealing parameters corresponding to container materials; Based on the temperature characteristic parameters of each control block and the standard heat sealing parameters, differentiated heat sealing parameters are generated for each control block through parameter compensation calculation; Through the locally adjustable heat sealing head, the corresponding differentiated heat sealing parameters are applied to each control block to implement the heat sealing operation.

4. The fully automatic filling and sealing method based on image recognition analysis according to claim 1 is characterized in that: The establishment of a multi-layer perceptron neural network optimization controller includes: Constructing a multi-layer perceptual neural network optimization controller, the multi-layer perceptual neural network optimization controller comprising an input layer, a hidden layer, and an output layer, the input layer receiving a system state parameter set, and the output layer generating an optimal control parameter combination; Taking the filling and sealing process quality index evaluation system as the optimization target, the multi-layer perception neural network optimization controller is trained; Configuring a dual-mode control mechanism for the multi-layer perceptron neural network optimization controller; The control parameter combination is optimized through the parameter sensitivity analysis mechanism of the multi-layer perception neural network optimization controller, and the adaptive adjustment of the parameters of the entire filling and sealing process is completed.

5. The fully automatic filling and sealing method based on image recognition analysis according to claim 3 is characterized in that: The method of identifying the abnormal temperature area in the container sealing area includes: Convert temperature distribution data into a two-dimensional temperature matrix; A multi-scale thermal anomaly detection mechanism is introduced to perform multi-scale decomposition of the two-dimensional temperature matrix through wavelet transform, establish a thermal feature subspace, and extract abnormal feature points; Calculating the temperature gradient vectors between adjacent points in the two-dimensional temperature matrix to obtain a temperature gradient field, and marking temperature anomaly candidate points according to a preset temperature gradient threshold; Merge the abnormal feature points with the temperature anomaly candidate points to obtain the enhanced temperature anomaly candidate point set; Connectivity analysis is performed on the enhanced temperature anomaly candidate point set to obtain a temperature anomaly area.

6. A fully automatic filling and sealing system based on image recognition and analysis, used to implement the fully automatic filling and sealing method based on image recognition and analysis according to any one of claims 1 to 5, characterized in that: include: Parameter analysis module, used to build a three-dimensional dynamic model of the fluid and extract the fluid dynamic control parameters and bubble group dynamic characteristic parameters through image recognition; An electric field control module is used to control the annular microelectrode array provided around the filling nozzle to generate a directional electric field according to the fluid dynamic control parameters, thereby regulating the liquid flow state; An ultrasonic modulation module is used to determine ultrasonic action parameters based on the dynamic characteristic parameters of the bubble group and the physical properties of the container liquid, and control the frequency-adjustable ultrasonic generator to output ultrasonic modulation pulses to the bubble accumulation area; The heat seal adjustment module is used to detect the temperature distribution of the container sealing area, control the local adjustable heat seal head based on the detection results, and apply differentiated heat seal parameters to different temperature areas; The control optimization module is used to establish a multi-layer perception neural network optimization controller, receive the system state parameter set, take the filling and sealing process quality indicators as the optimization target, optimize the control parameter combination, and complete the adaptive adjustment of the parameters of the entire filling and sealing process.

Images